maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/lldb/source/DataFormatters/FormatManager.cpp

1478 lines
80 KiB
C++
Raw Normal View History

Committing type format code for Enrico Granata. This commit adds a new top level command named "type". Currently this command implements three commands: type format add <format> <typename1> [<typename2> ...] type format delete <typename1> [<typename2> ...] type format list [<typename1> [<typename2>] ...] This allows you to specify the default format that will be used to display types when you use "frame variable" or "expression", or the SBValue classes. Examples: // Format uint*_t as hex type format add x uint16_t uint32_t uint64_t // Format intptr_t as a pointer type format add p intptr_t The format characters are the same as "printf" for the most part with many additions. These format character specifiers are also used in many other commands ("frame variable" for one). The current list of format characters include: a - char buffer b - binary B - boolean c - char C - printable char d - signed decimal e - float f - float g - float i - signed decimal I - complex integer o - octal O - OSType p - pointer s - c-string u - unsigned decimal x - hex X - complex float y - bytes Y - bytes with ASCII llvm-svn: 133728
2011-06-24 01:59:56 +08:00
//===-- FormatManager.cpp -------------------------------------------*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
Fix Linux build warnings due to redefinition of macros: - add new header lldb-python.h to be included before other system headers - short term fix (eventually python dependencies must be cleaned up) Patch by Matt Kopec! llvm-svn: 169341
2012-12-05 08:20:57 +08:00
#include "lldb/lldb-python.h"
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
#include "lldb/DataFormatters/FormatManager.h"
Committing type format code for Enrico Granata. This commit adds a new top level command named "type". Currently this command implements three commands: type format add <format> <typename1> [<typename2> ...] type format delete <typename1> [<typename2> ...] type format list [<typename1> [<typename2>] ...] This allows you to specify the default format that will be used to display types when you use "frame variable" or "expression", or the SBValue classes. Examples: // Format uint*_t as hex type format add x uint16_t uint32_t uint64_t // Format intptr_t as a pointer type format add p intptr_t The format characters are the same as "printf" for the most part with many additions. These format character specifiers are also used in many other commands ("frame variable" for one). The current list of format characters include: a - char buffer b - binary B - boolean c - char C - printable char d - signed decimal e - float f - float g - float i - signed decimal I - complex integer o - octal O - OSType p - pointer s - c-string u - unsigned decimal x - hex X - complex float y - bytes Y - bytes with ASCII llvm-svn: 133728
2011-06-24 01:59:56 +08:00
// C Includes
// C++ Includes
// Other libraries and framework includes
// Project includes
Python summary strings: - you can use a Python script to write a summary string for data-types, in one of three ways: -P option and typing the script a line at a time -s option and passing a one-line Python script -F option and passing the name of a Python function these options all work for the "type summary add" command your Python code (if provided through -P or -s) is wrapped in a function that accepts two parameters: valobj (a ValueObject) and dict (an LLDB internal dictionary object). if you use -F and give a function name, you're expected to define the function on your own and with the right prototype. your function, however defined, must return a Python string - test case for the Python summary feature - a few quirks: Python summaries cannot have names, and cannot use regex as type names both issues will be fixed ASAP major redesign of type summary code: - type summary working with strings and type summary working with Python code are two classes, with a common base class SummaryFormat - SummaryFormat classes now are able to actively format objects rather than just aggregating data - cleaner code to print descriptions for summaries the public API now exports a method to easily navigate a ValueObject hierarchy New InputReaderEZ and PriorityPointerPair classes Several minor fixes and improvements llvm-svn: 135238
2011-07-15 10:26:42 +08:00
#include "lldb/Core/Debugger.h"
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
#include "lldb/DataFormatters/CXXFormatterFunctions.h"
#include "lldb/Interpreter/ScriptInterpreterPython.h"
#include "lldb/Target/ExecutionContext.h"
#include "lldb/Target/Platform.h"
lldb: remove adhoc implementation of array_sizeof Replace adhoc inline implementation of llvm::array_lengthof in favour of the implementation in LLVM. This is simply a cleanup change, no functional change intended. llvm-svn: 211868
2014-06-27 13:17:41 +08:00
#include "llvm/ADT/STLExtras.h"
Python summary strings: - you can use a Python script to write a summary string for data-types, in one of three ways: -P option and typing the script a line at a time -s option and passing a one-line Python script -F option and passing the name of a Python function these options all work for the "type summary add" command your Python code (if provided through -P or -s) is wrapped in a function that accepts two parameters: valobj (a ValueObject) and dict (an LLDB internal dictionary object). if you use -F and give a function name, you're expected to define the function on your own and with the right prototype. your function, however defined, must return a Python string - test case for the Python summary feature - a few quirks: Python summaries cannot have names, and cannot use regex as type names both issues will be fixed ASAP major redesign of type summary code: - type summary working with strings and type summary working with Python code are two classes, with a common base class SummaryFormat - SummaryFormat classes now are able to actively format objects rather than just aggregating data - cleaner code to print descriptions for summaries the public API now exports a method to easily navigate a ValueObject hierarchy New InputReaderEZ and PriorityPointerPair classes Several minor fixes and improvements llvm-svn: 135238
2011-07-15 10:26:42 +08:00
Committing type format code for Enrico Granata. This commit adds a new top level command named "type". Currently this command implements three commands: type format add <format> <typename1> [<typename2> ...] type format delete <typename1> [<typename2> ...] type format list [<typename1> [<typename2>] ...] This allows you to specify the default format that will be used to display types when you use "frame variable" or "expression", or the SBValue classes. Examples: // Format uint*_t as hex type format add x uint16_t uint32_t uint64_t // Format intptr_t as a pointer type format add p intptr_t The format characters are the same as "printf" for the most part with many additions. These format character specifiers are also used in many other commands ("frame variable" for one). The current list of format characters include: a - char buffer b - binary B - boolean c - char C - printable char d - signed decimal e - float f - float g - float i - signed decimal I - complex integer o - octal O - OSType p - pointer s - c-string u - unsigned decimal x - hex X - complex float y - bytes Y - bytes with ASCII llvm-svn: 133728
2011-06-24 01:59:56 +08:00
using namespace lldb;
using namespace lldb_private;
Centralized all of the format to c-string and to format character code inside the FormatManager class. Modified the format arguments in any commands to be able to use a single character format, or a full format name, or a partial format name if no full format names match. Modified any code that was displaying formats to use the new FormatManager calls so that our help text and errors never get out of date. Modified the display of the "type format list" command to be a bit more human readable by showing the format as a format string rather than the single character format char. llvm-svn: 133765
2011-06-24 05:22:24 +08:00
struct FormatInfo
{
Format format;
const char format_char; // One or more format characters that can be used for this format.
const char *format_name; // Long format name that can be used to specify the current format
};
static FormatInfo
g_format_infos[] =
{
{ eFormatDefault , '\0' , "default" },
{ eFormatBoolean , 'B' , "boolean" },
{ eFormatBinary , 'b' , "binary" },
{ eFormatBytes , 'y' , "bytes" },
{ eFormatBytesWithASCII , 'Y' , "bytes with ASCII" },
{ eFormatChar , 'c' , "character" },
{ eFormatCharPrintable , 'C' , "printable character" },
{ eFormatComplexFloat , 'F' , "complex float" },
{ eFormatCString , 's' , "c-string" },
Added support for the new ".apple_objc" accelerator tables. These tables are in the same hashed format as the ".apple_names", but they map objective C class names to all of the methods and class functions. We need to do this because in the DWARF the methods for Objective C are never contained in the class definition, they are scattered about at the translation unit level and they don't even have attributes that say the are contained within the class itself. Added 3 new formats which can be used to display data: eFormatAddressInfo eFormatHexFloat eFormatInstruction eFormatAddressInfo describes an address such as function+offset and file+line, or symbol + offset, or constant data (c string, 2, 4, 8, or 16 byte constants). The format character for this is "A", the long format is "address". eFormatHexFloat will print out the hex float format that compilers tend to use. The format character for this is "X", the long format is "hex float". eFormatInstruction will print out disassembly with bytes and it will use the current target's architecture. The format character for this is "i" (which used to be being used for the integer format, but the integer format also has "d", so we gave the "i" format to disassembly), the long format is "instruction". Mate the lldb::FormatterChoiceCriterion enumeration private as it should have been from the start. It is very specialized and doesn't belong in the public API. llvm-svn: 143114
2011-10-28 01:55:14 +08:00
{ eFormatDecimal , 'd' , "decimal" },
Centralized all of the format to c-string and to format character code inside the FormatManager class. Modified the format arguments in any commands to be able to use a single character format, or a full format name, or a partial format name if no full format names match. Modified any code that was displaying formats to use the new FormatManager calls so that our help text and errors never get out of date. Modified the display of the "type format list" command to be a bit more human readable by showing the format as a format string rather than the single character format char. llvm-svn: 133765
2011-06-24 05:22:24 +08:00
{ eFormatEnum , 'E' , "enumeration" },
{ eFormatHex , 'x' , "hex" },
<rdar://problem/10449092> Adding a new uppercase hex format specifier. This commit also changes the short names for formats so that uppercase hex can be 'X', which was previously assigned to hex float. hex float now has no short name. llvm-svn: 161606
2012-08-10 03:33:34 +08:00
{ eFormatHexUppercase , 'X' , "uppercase hex" },
Centralized all of the format to c-string and to format character code inside the FormatManager class. Modified the format arguments in any commands to be able to use a single character format, or a full format name, or a partial format name if no full format names match. Modified any code that was displaying formats to use the new FormatManager calls so that our help text and errors never get out of date. Modified the display of the "type format list" command to be a bit more human readable by showing the format as a format string rather than the single character format char. llvm-svn: 133765
2011-06-24 05:22:24 +08:00
{ eFormatFloat , 'f' , "float" },
{ eFormatOctal , 'o' , "octal" },
{ eFormatOSType , 'O' , "OSType" },
{ eFormatUnicode16 , 'U' , "unicode16" },
{ eFormatUnicode32 , '\0' , "unicode32" },
{ eFormatUnsigned , 'u' , "unsigned decimal" },
{ eFormatPointer , 'p' , "pointer" },
{ eFormatVectorOfChar , '\0' , "char[]" },
{ eFormatVectorOfSInt8 , '\0' , "int8_t[]" },
{ eFormatVectorOfUInt8 , '\0' , "uint8_t[]" },
{ eFormatVectorOfSInt16 , '\0' , "int16_t[]" },
{ eFormatVectorOfUInt16 , '\0' , "uint16_t[]" },
Fixed some format names llvm-svn: 134492
2011-07-06 23:56:06 +08:00
{ eFormatVectorOfSInt32 , '\0' , "int32_t[]" },
{ eFormatVectorOfUInt32 , '\0' , "uint32_t[]" },
{ eFormatVectorOfSInt64 , '\0' , "int64_t[]" },
{ eFormatVectorOfUInt64 , '\0' , "uint64_t[]" },
Centralized all of the format to c-string and to format character code inside the FormatManager class. Modified the format arguments in any commands to be able to use a single character format, or a full format name, or a partial format name if no full format names match. Modified any code that was displaying formats to use the new FormatManager calls so that our help text and errors never get out of date. Modified the display of the "type format list" command to be a bit more human readable by showing the format as a format string rather than the single character format char. llvm-svn: 133765
2011-06-24 05:22:24 +08:00
{ eFormatVectorOfFloat32, '\0' , "float32[]" },
{ eFormatVectorOfFloat64, '\0' , "float64[]" },
{ eFormatVectorOfUInt128, '\0' , "uint128_t[]" },
{ eFormatComplexInteger , 'I' , "complex integer" },
Added support for the new ".apple_objc" accelerator tables. These tables are in the same hashed format as the ".apple_names", but they map objective C class names to all of the methods and class functions. We need to do this because in the DWARF the methods for Objective C are never contained in the class definition, they are scattered about at the translation unit level and they don't even have attributes that say the are contained within the class itself. Added 3 new formats which can be used to display data: eFormatAddressInfo eFormatHexFloat eFormatInstruction eFormatAddressInfo describes an address such as function+offset and file+line, or symbol + offset, or constant data (c string, 2, 4, 8, or 16 byte constants). The format character for this is "A", the long format is "address". eFormatHexFloat will print out the hex float format that compilers tend to use. The format character for this is "X", the long format is "hex float". eFormatInstruction will print out disassembly with bytes and it will use the current target's architecture. The format character for this is "i" (which used to be being used for the integer format, but the integer format also has "d", so we gave the "i" format to disassembly), the long format is "instruction". Mate the lldb::FormatterChoiceCriterion enumeration private as it should have been from the start. It is very specialized and doesn't belong in the public API. llvm-svn: 143114
2011-10-28 01:55:14 +08:00
{ eFormatCharArray , 'a' , "character array" },
{ eFormatAddressInfo , 'A' , "address" },
<rdar://problem/10449092> Adding a new uppercase hex format specifier. This commit also changes the short names for formats so that uppercase hex can be 'X', which was previously assigned to hex float. hex float now has no short name. llvm-svn: 161606
2012-08-10 03:33:34 +08:00
{ eFormatHexFloat , '\0' , "hex float" },
Added a 'void' format so that the user can manually suppress all non-error output from the "expression" command. <rdar://problem/11225150> llvm-svn: 161502
2012-08-09 01:35:10 +08:00
{ eFormatInstruction , 'i' , "instruction" },
{ eFormatVoid , 'v' , "void" }
Centralized all of the format to c-string and to format character code inside the FormatManager class. Modified the format arguments in any commands to be able to use a single character format, or a full format name, or a partial format name if no full format names match. Modified any code that was displaying formats to use the new FormatManager calls so that our help text and errors never get out of date. Modified the display of the "type format list" command to be a bit more human readable by showing the format as a format string rather than the single character format char. llvm-svn: 133765
2011-06-24 05:22:24 +08:00
};
lldb: remove adhoc implementation of array_sizeof Replace adhoc inline implementation of llvm::array_lengthof in favour of the implementation in LLVM. This is simply a cleanup change, no functional change intended. llvm-svn: 211868
2014-06-27 13:17:41 +08:00
static uint32_t g_num_format_infos = llvm::array_lengthof(g_format_infos);
Centralized all of the format to c-string and to format character code inside the FormatManager class. Modified the format arguments in any commands to be able to use a single character format, or a full format name, or a partial format name if no full format names match. Modified any code that was displaying formats to use the new FormatManager calls so that our help text and errors never get out of date. Modified the display of the "type format list" command to be a bit more human readable by showing the format as a format string rather than the single character format char. llvm-svn: 133765
2011-06-24 05:22:24 +08:00
static bool
GetFormatFromFormatChar (char format_char, Format &format)
{
for (uint32_t i=0; i<g_num_format_infos; ++i)
{
if (g_format_infos[i].format_char == format_char)
{
format = g_format_infos[i].format;
return true;
}
}
format = eFormatInvalid;
return false;
}
static bool
GetFormatFromFormatName (const char *format_name, bool partial_match_ok, Format &format)
{
uint32_t i;
for (i=0; i<g_num_format_infos; ++i)
{
if (strcasecmp (g_format_infos[i].format_name, format_name) == 0)
{
format = g_format_infos[i].format;
return true;
}
}
if (partial_match_ok)
{
for (i=0; i<g_num_format_infos; ++i)
{
if (strcasestr (g_format_infos[i].format_name, format_name) == g_format_infos[i].format_name)
{
format = g_format_infos[i].format;
return true;
}
}
}
format = eFormatInvalid;
return false;
}
bool
FormatManager::GetFormatFromCString (const char *format_cstr,
bool partial_match_ok,
lldb::Format &format)
{
bool success = false;
if (format_cstr && format_cstr[0])
{
if (format_cstr[1] == '\0')
{
success = GetFormatFromFormatChar (format_cstr[0], format);
if (success)
return true;
}
success = GetFormatFromFormatName (format_cstr, partial_match_ok, format);
}
if (!success)
format = eFormatInvalid;
return success;
}
char
FormatManager::GetFormatAsFormatChar (lldb::Format format)
{
for (uint32_t i=0; i<g_num_format_infos; ++i)
{
if (g_format_infos[i].format == format)
return g_format_infos[i].format_char;
}
return '\0';
}
const char *
FormatManager::GetFormatAsCString (Format format)
{
if (format >= eFormatDefault && format < kNumFormats)
return g_format_infos[format].format_name;
return NULL;
}
several improvements to "type summary": - type names can now be regular expressions (exact matching is done first, and is faster) - integral (and floating) types can be printed as bitfields, i.e. ${var[low-high]} will extract bits low thru high of the value and print them - array subscripts are supported, both for arrays and for pointers. the syntax is ${*var[low-high]}, or ${*var[]} to print the whole array (the latter only works for statically sized arrays) - summary is now printed by default when a summary string references a variable. if that variable's type has no summary, value is printed instead. to force value, you can use %V as a format specifier - basic support for ObjectiveC: - ObjectiveC inheritance chains are now walked through - %@ can be specified as a summary format, to print the ObjectiveC runtime description for an object - some bug fixes llvm-svn: 134293
2011-07-02 08:25:22 +08:00
<rdar://problem/15530080> Rework data formatters matching algorithm What happens now is that, for each category, the FormatNavigator generates all possible matches, and checks them one by one Since the possible matches do not actually depend on the category (whether a match is accepted or not does, but that check can be shifted at a more convenient time), it is actually feasible to generate every possible match upfront and then let individual categories just scan through those This commit changes things by introducing a notion of formatters match candidate, and shifting responsibility for generating all of them given a (ValueObject,DynamicValueType) pair from the FormatNavigator back to the FormatManager A list of these candidates is then passed down to each category for matching Candidates also need to remember whether they were generated by stripping pointers, references, typedefs, since this is something that individual formatters can choose to reject This check, however, is conveniently only done once a "textual" match has been found, so that the list of candidates is truly category-independent While the performance benefit is small (mostly, due to caching), this is much cleaner from a design perspective llvm-svn: 195395
2013-11-22 08:02:13 +08:00
void
FormatManager::GetPossibleMatches (ValueObject& valobj,
ClangASTType clang_type,
uint32_t reason,
lldb::DynamicValueType use_dynamic,
FormattersMatchVector& entries,
bool did_strip_ptr,
bool did_strip_ref,
bool did_strip_typedef,
bool root_level)
{
clang_type = clang_type.RemoveFastQualifiers();
ConstString type_name(clang_type.GetConstTypeName());
if (valobj.GetBitfieldBitSize() > 0)
{
StreamString sstring;
sstring.Printf("%s:%d",type_name.AsCString(),valobj.GetBitfieldBitSize());
ConstString bitfieldname = ConstString(sstring.GetData());
entries.push_back({bitfieldname,0,did_strip_ptr,did_strip_ref,did_strip_typedef});
reason |= lldb_private::eFormatterChoiceCriterionStrippedBitField;
}
entries.push_back({type_name,reason,did_strip_ptr,did_strip_ref,did_strip_typedef});
Introduce the concept of a "display name" for types Rationale: Pretty simply, the idea is that sometimes type names are way too long and contain way too many details for the average developer to care about. For instance, a plain ol' vector of int might be shown as std::__1::vector<int, std::__1::allocator<.... rather than the much simpler std::vector<int> form, which is what most developers would actually type in their code Proposed solution: Introduce a notion of "display name" and a corresponding API GetDisplayTypeName() to return such a crafted for visual representation type name Obviously, the display name and the fully qualified (or "true") name are not necessarily the same - that's the whole point LLDB could choose to pick the "display name" as its one true notion of a type name, and if somebody really needs the fully qualified version of it, let them deal with the problem Or, LLDB could rename what it currently calls the "type name" to be the "display name", and add new APIs for the fully qualified name, making the display name the default choice The choice that I am making here is that the type name will keep meaning the same, and people who want a type name suited for display will explicitly ask for one It is the less risky/disruptive choice - and it should eventually make it fairly obvious when someone is asking for the wrong type Caveats: - for now, GetDisplayTypeName() == GetTypeName(), there is no logic to produce customized display type names yet. - while the fully-qualified type name is still the main key to the kingdom of data formatters, if we start showing custom names to people, those should match formatters llvm-svn: 209072
2014-05-18 03:14:17 +08:00
ConstString display_type_name(clang_type.GetDisplayTypeName());
if (display_type_name != type_name)
entries.push_back({display_type_name,reason,did_strip_ptr,did_strip_ref,did_strip_typedef});
<rdar://problem/12055586> Enable data formatters to see-through pointers/references to typedefs For instance, if Foo is a typedef to Bar, and there is a formatter for any/all of Bar*, Bar&, Bar&&, then Foo*, Foo&, and Foo&& should pick these up if Foo-specific formatters don't exist llvm-svn: 205939
2014-04-10 08:14:07 +08:00
for (bool is_rvalue_ref = true, j = true; j && clang_type.IsReferenceType(nullptr, &is_rvalue_ref); j = false)
<rdar://problem/15530080> Rework data formatters matching algorithm What happens now is that, for each category, the FormatNavigator generates all possible matches, and checks them one by one Since the possible matches do not actually depend on the category (whether a match is accepted or not does, but that check can be shifted at a more convenient time), it is actually feasible to generate every possible match upfront and then let individual categories just scan through those This commit changes things by introducing a notion of formatters match candidate, and shifting responsibility for generating all of them given a (ValueObject,DynamicValueType) pair from the FormatNavigator back to the FormatManager A list of these candidates is then passed down to each category for matching Candidates also need to remember whether they were generated by stripping pointers, references, typedefs, since this is something that individual formatters can choose to reject This check, however, is conveniently only done once a "textual" match has been found, so that the list of candidates is truly category-independent While the performance benefit is small (mostly, due to caching), this is much cleaner from a design perspective llvm-svn: 195395
2013-11-22 08:02:13 +08:00
{
ClangASTType non_ref_type = clang_type.GetNonReferenceType();
GetPossibleMatches(valobj,
non_ref_type,
reason | lldb_private::eFormatterChoiceCriterionStrippedPointerReference,
use_dynamic,
entries,
did_strip_ptr,
true,
did_strip_typedef);
<rdar://problem/12055586> Enable data formatters to see-through pointers/references to typedefs For instance, if Foo is a typedef to Bar, and there is a formatter for any/all of Bar*, Bar&, Bar&&, then Foo*, Foo&, and Foo&& should pick these up if Foo-specific formatters don't exist llvm-svn: 205939
2014-04-10 08:14:07 +08:00
if (non_ref_type.IsTypedefType())
{
ClangASTType deffed_referenced_type = non_ref_type.GetTypedefedType();
deffed_referenced_type = is_rvalue_ref ? deffed_referenced_type.GetRValueReferenceType() : deffed_referenced_type.GetLValueReferenceType();
GetPossibleMatches(valobj,
deffed_referenced_type,
reason | lldb_private::eFormatterChoiceCriterionNavigatedTypedefs,
use_dynamic,
entries,
did_strip_ptr,
did_strip_ref,
true); // this is not exactly the usual meaning of stripping typedefs
}
<rdar://problem/15530080> Rework data formatters matching algorithm What happens now is that, for each category, the FormatNavigator generates all possible matches, and checks them one by one Since the possible matches do not actually depend on the category (whether a match is accepted or not does, but that check can be shifted at a more convenient time), it is actually feasible to generate every possible match upfront and then let individual categories just scan through those This commit changes things by introducing a notion of formatters match candidate, and shifting responsibility for generating all of them given a (ValueObject,DynamicValueType) pair from the FormatNavigator back to the FormatManager A list of these candidates is then passed down to each category for matching Candidates also need to remember whether they were generated by stripping pointers, references, typedefs, since this is something that individual formatters can choose to reject This check, however, is conveniently only done once a "textual" match has been found, so that the list of candidates is truly category-independent While the performance benefit is small (mostly, due to caching), this is much cleaner from a design perspective llvm-svn: 195395
2013-11-22 08:02:13 +08:00
}
<rdar://problem/12055586> Enable data formatters to see-through pointers/references to typedefs For instance, if Foo is a typedef to Bar, and there is a formatter for any/all of Bar*, Bar&, Bar&&, then Foo*, Foo&, and Foo&& should pick these up if Foo-specific formatters don't exist llvm-svn: 205939
2014-04-10 08:14:07 +08:00
if (clang_type.IsPointerType())
<rdar://problem/15530080> Rework data formatters matching algorithm What happens now is that, for each category, the FormatNavigator generates all possible matches, and checks them one by one Since the possible matches do not actually depend on the category (whether a match is accepted or not does, but that check can be shifted at a more convenient time), it is actually feasible to generate every possible match upfront and then let individual categories just scan through those This commit changes things by introducing a notion of formatters match candidate, and shifting responsibility for generating all of them given a (ValueObject,DynamicValueType) pair from the FormatNavigator back to the FormatManager A list of these candidates is then passed down to each category for matching Candidates also need to remember whether they were generated by stripping pointers, references, typedefs, since this is something that individual formatters can choose to reject This check, however, is conveniently only done once a "textual" match has been found, so that the list of candidates is truly category-independent While the performance benefit is small (mostly, due to caching), this is much cleaner from a design perspective llvm-svn: 195395
2013-11-22 08:02:13 +08:00
{
ClangASTType non_ptr_type = clang_type.GetPointeeType();
GetPossibleMatches(valobj,
non_ptr_type,
reason | lldb_private::eFormatterChoiceCriterionStrippedPointerReference,
use_dynamic,
entries,
true,
did_strip_ref,
did_strip_typedef);
<rdar://problem/12055586> Enable data formatters to see-through pointers/references to typedefs For instance, if Foo is a typedef to Bar, and there is a formatter for any/all of Bar*, Bar&, Bar&&, then Foo*, Foo&, and Foo&& should pick these up if Foo-specific formatters don't exist llvm-svn: 205939
2014-04-10 08:14:07 +08:00
if (non_ptr_type.IsTypedefType())
{
ClangASTType deffed_pointed_type = non_ptr_type.GetTypedefedType().GetPointerType();
GetPossibleMatches(valobj,
deffed_pointed_type,
reason | lldb_private::eFormatterChoiceCriterionNavigatedTypedefs,
use_dynamic,
entries,
did_strip_ptr,
did_strip_ref,
true); // this is not exactly the usual meaning of stripping typedefs
}
<rdar://problem/15530080> Rework data formatters matching algorithm What happens now is that, for each category, the FormatNavigator generates all possible matches, and checks them one by one Since the possible matches do not actually depend on the category (whether a match is accepted or not does, but that check can be shifted at a more convenient time), it is actually feasible to generate every possible match upfront and then let individual categories just scan through those This commit changes things by introducing a notion of formatters match candidate, and shifting responsibility for generating all of them given a (ValueObject,DynamicValueType) pair from the FormatNavigator back to the FormatManager A list of these candidates is then passed down to each category for matching Candidates also need to remember whether they were generated by stripping pointers, references, typedefs, since this is something that individual formatters can choose to reject This check, however, is conveniently only done once a "textual" match has been found, so that the list of candidates is truly category-independent While the performance benefit is small (mostly, due to caching), this is much cleaner from a design perspective llvm-svn: 195395
2013-11-22 08:02:13 +08:00
}
bool canBeObjCDynamic = clang_type.IsPossibleDynamicType (NULL,
false, // no C
true); // yes ObjC
if (canBeObjCDynamic)
{
if (use_dynamic != lldb::eNoDynamicValues)
{
do
{
lldb::ProcessSP process_sp = valobj.GetProcessSP();
ObjCLanguageRuntime* runtime = process_sp->GetObjCLanguageRuntime();
if (runtime == nullptr)
break;
ObjCLanguageRuntime::ClassDescriptorSP objc_class_sp (runtime->GetClassDescriptor(valobj));
if (!objc_class_sp)
break;
ConstString name (objc_class_sp->GetClassName());
entries.push_back({name,reason | lldb_private::eFormatterChoiceCriterionDynamicObjCDiscovery,did_strip_ptr,did_strip_ref,did_strip_typedef});
} while (false);
}
ClangASTType non_ptr_type = clang_type.GetPointeeType();
GetPossibleMatches(valobj,
non_ptr_type,
reason | lldb_private::eFormatterChoiceCriterionStrippedPointerReference,
use_dynamic,
entries,
true,
did_strip_ref,
did_strip_typedef);
}
// try to strip typedef chains
if (clang_type.IsTypedefType())
{
ClangASTType deffed_type = clang_type.GetTypedefedType();
GetPossibleMatches(valobj,
deffed_type,
reason | lldb_private::eFormatterChoiceCriterionNavigatedTypedefs,
use_dynamic,
entries,
did_strip_ptr,
did_strip_ref,
true);
}
if (root_level)
{
do {
if (!clang_type.IsValid())
break;
ClangASTType unqual_clang_ast_type = clang_type.GetFullyUnqualifiedType();
if (!unqual_clang_ast_type.IsValid())
break;
if (unqual_clang_ast_type.GetOpaqueQualType() != clang_type.GetOpaqueQualType())
GetPossibleMatches (valobj,
unqual_clang_ast_type,
reason,
use_dynamic,
entries,
did_strip_ptr,
did_strip_ref,
did_strip_typedef);
} while(false);
// if all else fails, go to static type
if (valobj.IsDynamic())
{
lldb::ValueObjectSP static_value_sp(valobj.GetStaticValue());
if (static_value_sp)
GetPossibleMatches(*static_value_sp.get(),
static_value_sp->GetClangType(),
reason | lldb_private::eFormatterChoiceCriterionWentToStaticValue,
use_dynamic,
entries,
did_strip_ptr,
did_strip_ref,
did_strip_typedef,
true);
}
}
}
<rdar://problem/11778815> Formats (as in "type format") are now included in categories The only bit missing is caching formats along with synthetic children and summaries, which might be now desirable llvm-svn: 192217
2013-10-09 03:03:22 +08:00
lldb::TypeFormatImplSP
FormatManager::GetFormatForType (lldb::TypeNameSpecifierImplSP type_sp)
{
if (!type_sp)
return lldb::TypeFormatImplSP();
lldb::TypeFormatImplSP format_chosen_sp;
uint32_t num_categories = m_categories_map.GetCount();
lldb::TypeCategoryImplSP category_sp;
uint32_t prio_category = UINT32_MAX;
for (uint32_t category_id = 0;
category_id < num_categories;
category_id++)
{
category_sp = GetCategoryAtIndex(category_id);
if (category_sp->IsEnabled() == false)
continue;
lldb::TypeFormatImplSP format_current_sp = category_sp->GetFormatForType(type_sp);
if (format_current_sp && (format_chosen_sp.get() == NULL || (prio_category > category_sp->GetEnabledPosition())))
{
prio_category = category_sp->GetEnabledPosition();
format_chosen_sp = format_current_sp;
}
}
return format_chosen_sp;
}
<rdar://problem/11338654> Fixing a bug where having a summary for a bitfield without a format specified would in certain cases crash LLDB - This has also led to refactoring the by-type accessors for the data formatter subsystem. These now belong in our internal layer, and are just invoked by the public API stratum llvm-svn: 156429
2012-05-09 05:49:57 +08:00
lldb::TypeSummaryImplSP
FormatManager::GetSummaryForType (lldb::TypeNameSpecifierImplSP type_sp)
{
if (!type_sp)
return lldb::TypeSummaryImplSP();
lldb::TypeSummaryImplSP summary_chosen_sp;
uint32_t num_categories = m_categories_map.GetCount();
lldb::TypeCategoryImplSP category_sp;
uint32_t prio_category = UINT32_MAX;
for (uint32_t category_id = 0;
category_id < num_categories;
category_id++)
{
category_sp = GetCategoryAtIndex(category_id);
if (category_sp->IsEnabled() == false)
continue;
lldb::TypeSummaryImplSP summary_current_sp = category_sp->GetSummaryForType(type_sp);
if (summary_current_sp && (summary_chosen_sp.get() == NULL || (prio_category > category_sp->GetEnabledPosition())))
{
prio_category = category_sp->GetEnabledPosition();
summary_chosen_sp = summary_current_sp;
}
}
return summary_chosen_sp;
}
lldb::TypeFilterImplSP
FormatManager::GetFilterForType (lldb::TypeNameSpecifierImplSP type_sp)
{
if (!type_sp)
return lldb::TypeFilterImplSP();
lldb::TypeFilterImplSP filter_chosen_sp;
uint32_t num_categories = m_categories_map.GetCount();
lldb::TypeCategoryImplSP category_sp;
uint32_t prio_category = UINT32_MAX;
for (uint32_t category_id = 0;
category_id < num_categories;
category_id++)
{
category_sp = GetCategoryAtIndex(category_id);
if (category_sp->IsEnabled() == false)
continue;
lldb::TypeFilterImplSP filter_current_sp((TypeFilterImpl*)category_sp->GetFilterForType(type_sp).get());
if (filter_current_sp && (filter_chosen_sp.get() == NULL || (prio_category > category_sp->GetEnabledPosition())))
{
prio_category = category_sp->GetEnabledPosition();
filter_chosen_sp = filter_current_sp;
}
}
return filter_chosen_sp;
}
Add LLDB_DISABLE_PYTHON around newly added methods in DataVisualization.h / DataVisualization.cpp / ValueObject.cpp and FormatManager.h / FormatManager.cpp llvm-svn: 156886
2012-05-16 08:38:08 +08:00
#ifndef LLDB_DISABLE_PYTHON
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
lldb::ScriptedSyntheticChildrenSP
<rdar://problem/11338654> Fixing a bug where having a summary for a bitfield without a format specified would in certain cases crash LLDB - This has also led to refactoring the by-type accessors for the data formatter subsystem. These now belong in our internal layer, and are just invoked by the public API stratum llvm-svn: 156429
2012-05-09 05:49:57 +08:00
FormatManager::GetSyntheticForType (lldb::TypeNameSpecifierImplSP type_sp)
{
if (!type_sp)
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
return lldb::ScriptedSyntheticChildrenSP();
lldb::ScriptedSyntheticChildrenSP synth_chosen_sp;
<rdar://problem/11338654> Fixing a bug where having a summary for a bitfield without a format specified would in certain cases crash LLDB - This has also led to refactoring the by-type accessors for the data formatter subsystem. These now belong in our internal layer, and are just invoked by the public API stratum llvm-svn: 156429
2012-05-09 05:49:57 +08:00
uint32_t num_categories = m_categories_map.GetCount();
lldb::TypeCategoryImplSP category_sp;
uint32_t prio_category = UINT32_MAX;
for (uint32_t category_id = 0;
category_id < num_categories;
category_id++)
{
category_sp = GetCategoryAtIndex(category_id);
if (category_sp->IsEnabled() == false)
continue;
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
lldb::ScriptedSyntheticChildrenSP synth_current_sp((ScriptedSyntheticChildren*)category_sp->GetSyntheticForType(type_sp).get());
<rdar://problem/11338654> Fixing a bug where having a summary for a bitfield without a format specified would in certain cases crash LLDB - This has also led to refactoring the by-type accessors for the data formatter subsystem. These now belong in our internal layer, and are just invoked by the public API stratum llvm-svn: 156429
2012-05-09 05:49:57 +08:00
if (synth_current_sp && (synth_chosen_sp.get() == NULL || (prio_category > category_sp->GetEnabledPosition())))
{
prio_category = category_sp->GetEnabledPosition();
synth_chosen_sp = synth_current_sp;
}
}
return synth_chosen_sp;
}
Add LLDB_DISABLE_PYTHON around newly added methods in DataVisualization.h / DataVisualization.cpp / ValueObject.cpp and FormatManager.h / FormatManager.cpp llvm-svn: 156886
2012-05-16 08:38:08 +08:00
#endif
<rdar://problem/11338654> Fixing a bug where having a summary for a bitfield without a format specified would in certain cases crash LLDB - This has also led to refactoring the by-type accessors for the data formatter subsystem. These now belong in our internal layer, and are just invoked by the public API stratum llvm-svn: 156429
2012-05-09 05:49:57 +08:00
Add LLDB_DISABLE_PYTHON around newly added methods in DataVisualization.h / DataVisualization.cpp / ValueObject.cpp and FormatManager.h / FormatManager.cpp llvm-svn: 156886
2012-05-16 08:38:08 +08:00
#ifndef LLDB_DISABLE_PYTHON
<rdar://problem/11338654> Fixing a bug where having a summary for a bitfield without a format specified would in certain cases crash LLDB - This has also led to refactoring the by-type accessors for the data formatter subsystem. These now belong in our internal layer, and are just invoked by the public API stratum llvm-svn: 156429
2012-05-09 05:49:57 +08:00
lldb::SyntheticChildrenSP
FormatManager::GetSyntheticChildrenForType (lldb::TypeNameSpecifierImplSP type_sp)
{
if (!type_sp)
return lldb::SyntheticChildrenSP();
lldb::TypeFilterImplSP filter_sp = GetFilterForType(type_sp);
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
lldb::ScriptedSyntheticChildrenSP synth_sp = GetSyntheticForType(type_sp);
<rdar://problem/11338654> Fixing a bug where having a summary for a bitfield without a format specified would in certain cases crash LLDB - This has also led to refactoring the by-type accessors for the data formatter subsystem. These now belong in our internal layer, and are just invoked by the public API stratum llvm-svn: 156429
2012-05-09 05:49:57 +08:00
if (filter_sp->GetRevision() > synth_sp->GetRevision())
return lldb::SyntheticChildrenSP(filter_sp.get());
else
return lldb::SyntheticChildrenSP(synth_sp.get());
}
Add LLDB_DISABLE_PYTHON around newly added methods in DataVisualization.h / DataVisualization.cpp / ValueObject.cpp and FormatManager.h / FormatManager.cpp llvm-svn: 156886
2012-05-16 08:38:08 +08:00
#endif
<rdar://problem/11338654> Fixing a bug where having a summary for a bitfield without a format specified would in certain cases crash LLDB - This has also led to refactoring the by-type accessors for the data formatter subsystem. These now belong in our internal layer, and are just invoked by the public API stratum llvm-svn: 156429
2012-05-09 05:49:57 +08:00
<rdar://problem/10062621> New public API for handling formatters: creating, deleting, modifying categories, and formatters, and managing type/formatter association. This provides SB classes for each of the main object types involved in providing formatter support: SBTypeCategory SBTypeFilter SBTypeFormat SBTypeSummary SBTypeSynthetic plus, an SBTypeNameSpecifier class that is used on the public API layer to abstract the notion that formatters can be applied to plain type-names as well as to regular expressions For naming consistency, this patch also renames a lot of formatters-related classes. Plus, the changes in how flags are handled that started with summaries is now extended to other classes as well. A new enum (lldb::eTypeOption) is meant to support this on the public side. The patch also adds several new calls to the formatter infrastructure that are used to implement by-index accessing and several other design changes required to accommodate the new API layer. An architectural change is introduced in that backing objects for formatters now become writable. On the public API layer, CoW is implemented to prevent unwanted propagation of changes. Lastly, there are some modifications in how the "default" category is constructed and managed in relation to other categories. llvm-svn: 150558
2012-02-15 10:34:21 +08:00
lldb::TypeCategoryImplSP
Redesign of the interaction between Python and frozen objects: - introduced two new classes ValueObjectConstResultChild and ValueObjectConstResultImpl: the first one is a ValueObjectChild obtained from a ValueObjectConstResult, the second is a common implementation backend for VOCR and VOCRCh of method calls meant to read through pointers stored in frozen objects ; now such reads transparently move from host to target as required - as a consequence of the above, removed code that made target-memory copies of expression results in several places throughout LLDB, and also removed code that enabled to recognize an expression result VO as such - introduced a new GetPointeeData() method in ValueObject that lets you read a given amount of objects of type T from a VO representing a T* or T[], and doing dereferences transparently in private layer it returns a DataExtractor ; in public layer it returns an instance of a newly created lldb::SBData - as GetPointeeData() does the right thing for both frozen and non-frozen ValueObject's, reimplemented ReadPointedString() to use it en lieu of doing the raw read itself - introduced a new GetData() method in ValueObject that lets you get a copy of the data that backs the ValueObject (for pointers, this returns the address without any previous dereferencing steps ; for arrays it actually reads the whole chunk of memory) in public layer this returns an SBData, just like GetPointeeData() - introduced a new CreateValueFromData() method in SBValue that lets you create a new SBValue from a chunk of data wrapped in an SBData the limitation to remember for this kind of SBValue is that they have no address: extracting the address-of for these objects (with any of GetAddress(), GetLoadAddress() and AddressOf()) will return invalid values - added several tests to check that "p"-ing objects (STL classes, char* and char[]) will do the right thing Solved a bug where global pointers to global variables were not dereferenced correctly for display New target setting "max-string-summary-length" gives the maximum number of characters to show in a string when summarizing it, instead of the hardcoded 128 Solved a bug where the summary for char[] and char* would not be shown if the ValueObject's were dumped via the "p" command Removed m_pointers_point_to_load_addrs from ValueObject. Introduced a new m_address_type_of_children, which each ValueObject can set to tell the address type of any pointers and/or references it creates. In the current codebase, this is load address most of the time (the only notable exception being file addresses that generate file address children UNLESS we have a live process) Updated help text for summary-string Fixed an issue in STL formatters where std::stlcontainer::iterator would match the container's synthetic children providers Edited the syntax and help for some commands to have proper argument types llvm-svn: 139160
2011-09-07 03:20:51 +08:00
FormatManager::GetCategory (const ConstString& category_name,
Additional code cleanups ; Short option name for --python-script in type summary add moved from -s to -o (this is a preliminary step in moving the short option for --summary-string from -f to -s) ; Accordingly updated the test suite llvm-svn: 138315
2011-08-23 08:32:52 +08:00
bool can_create)
{
if (!category_name)
Redesign of the interaction between Python and frozen objects: - introduced two new classes ValueObjectConstResultChild and ValueObjectConstResultImpl: the first one is a ValueObjectChild obtained from a ValueObjectConstResult, the second is a common implementation backend for VOCR and VOCRCh of method calls meant to read through pointers stored in frozen objects ; now such reads transparently move from host to target as required - as a consequence of the above, removed code that made target-memory copies of expression results in several places throughout LLDB, and also removed code that enabled to recognize an expression result VO as such - introduced a new GetPointeeData() method in ValueObject that lets you read a given amount of objects of type T from a VO representing a T* or T[], and doing dereferences transparently in private layer it returns a DataExtractor ; in public layer it returns an instance of a newly created lldb::SBData - as GetPointeeData() does the right thing for both frozen and non-frozen ValueObject's, reimplemented ReadPointedString() to use it en lieu of doing the raw read itself - introduced a new GetData() method in ValueObject that lets you get a copy of the data that backs the ValueObject (for pointers, this returns the address without any previous dereferencing steps ; for arrays it actually reads the whole chunk of memory) in public layer this returns an SBData, just like GetPointeeData() - introduced a new CreateValueFromData() method in SBValue that lets you create a new SBValue from a chunk of data wrapped in an SBData the limitation to remember for this kind of SBValue is that they have no address: extracting the address-of for these objects (with any of GetAddress(), GetLoadAddress() and AddressOf()) will return invalid values - added several tests to check that "p"-ing objects (STL classes, char* and char[]) will do the right thing Solved a bug where global pointers to global variables were not dereferenced correctly for display New target setting "max-string-summary-length" gives the maximum number of characters to show in a string when summarizing it, instead of the hardcoded 128 Solved a bug where the summary for char[] and char* would not be shown if the ValueObject's were dumped via the "p" command Removed m_pointers_point_to_load_addrs from ValueObject. Introduced a new m_address_type_of_children, which each ValueObject can set to tell the address type of any pointers and/or references it creates. In the current codebase, this is load address most of the time (the only notable exception being file addresses that generate file address children UNLESS we have a live process) Updated help text for summary-string Fixed an issue in STL formatters where std::stlcontainer::iterator would match the container's synthetic children providers Edited the syntax and help for some commands to have proper argument types llvm-svn: 139160
2011-09-07 03:20:51 +08:00
return GetCategory(m_default_category_name);
<rdar://problem/10062621> New public API for handling formatters: creating, deleting, modifying categories, and formatters, and managing type/formatter association. This provides SB classes for each of the main object types involved in providing formatter support: SBTypeCategory SBTypeFilter SBTypeFormat SBTypeSummary SBTypeSynthetic plus, an SBTypeNameSpecifier class that is used on the public API layer to abstract the notion that formatters can be applied to plain type-names as well as to regular expressions For naming consistency, this patch also renames a lot of formatters-related classes. Plus, the changes in how flags are handled that started with summaries is now extended to other classes as well. A new enum (lldb::eTypeOption) is meant to support this on the public side. The patch also adds several new calls to the formatter infrastructure that are used to implement by-index accessing and several other design changes required to accommodate the new API layer. An architectural change is introduced in that backing objects for formatters now become writable. On the public API layer, CoW is implemented to prevent unwanted propagation of changes. Lastly, there are some modifications in how the "default" category is constructed and managed in relation to other categories. llvm-svn: 150558
2012-02-15 10:34:21 +08:00
lldb::TypeCategoryImplSP category;
Additional code cleanups ; Short option name for --python-script in type summary add moved from -s to -o (this is a preliminary step in moving the short option for --summary-string from -f to -s) ; Accordingly updated the test suite llvm-svn: 138315
2011-08-23 08:32:52 +08:00
if (m_categories_map.Get(category_name, category))
return category;
if (!can_create)
<rdar://problem/10062621> New public API for handling formatters: creating, deleting, modifying categories, and formatters, and managing type/formatter association. This provides SB classes for each of the main object types involved in providing formatter support: SBTypeCategory SBTypeFilter SBTypeFormat SBTypeSummary SBTypeSynthetic plus, an SBTypeNameSpecifier class that is used on the public API layer to abstract the notion that formatters can be applied to plain type-names as well as to regular expressions For naming consistency, this patch also renames a lot of formatters-related classes. Plus, the changes in how flags are handled that started with summaries is now extended to other classes as well. A new enum (lldb::eTypeOption) is meant to support this on the public side. The patch also adds several new calls to the formatter infrastructure that are used to implement by-index accessing and several other design changes required to accommodate the new API layer. An architectural change is introduced in that backing objects for formatters now become writable. On the public API layer, CoW is implemented to prevent unwanted propagation of changes. Lastly, there are some modifications in how the "default" category is constructed and managed in relation to other categories. llvm-svn: 150558
2012-02-15 10:34:21 +08:00
return lldb::TypeCategoryImplSP();
Additional code cleanups ; Short option name for --python-script in type summary add moved from -s to -o (this is a preliminary step in moving the short option for --summary-string from -f to -s) ; Accordingly updated the test suite llvm-svn: 138315
2011-08-23 08:32:52 +08:00
<rdar://problem/10062621> New public API for handling formatters: creating, deleting, modifying categories, and formatters, and managing type/formatter association. This provides SB classes for each of the main object types involved in providing formatter support: SBTypeCategory SBTypeFilter SBTypeFormat SBTypeSummary SBTypeSynthetic plus, an SBTypeNameSpecifier class that is used on the public API layer to abstract the notion that formatters can be applied to plain type-names as well as to regular expressions For naming consistency, this patch also renames a lot of formatters-related classes. Plus, the changes in how flags are handled that started with summaries is now extended to other classes as well. A new enum (lldb::eTypeOption) is meant to support this on the public side. The patch also adds several new calls to the formatter infrastructure that are used to implement by-index accessing and several other design changes required to accommodate the new API layer. An architectural change is introduced in that backing objects for formatters now become writable. On the public API layer, CoW is implemented to prevent unwanted propagation of changes. Lastly, there are some modifications in how the "default" category is constructed and managed in relation to other categories. llvm-svn: 150558
2012-02-15 10:34:21 +08:00
m_categories_map.Add(category_name,lldb::TypeCategoryImplSP(new TypeCategoryImpl(this, category_name)));
Redesign of the interaction between Python and frozen objects: - introduced two new classes ValueObjectConstResultChild and ValueObjectConstResultImpl: the first one is a ValueObjectChild obtained from a ValueObjectConstResult, the second is a common implementation backend for VOCR and VOCRCh of method calls meant to read through pointers stored in frozen objects ; now such reads transparently move from host to target as required - as a consequence of the above, removed code that made target-memory copies of expression results in several places throughout LLDB, and also removed code that enabled to recognize an expression result VO as such - introduced a new GetPointeeData() method in ValueObject that lets you read a given amount of objects of type T from a VO representing a T* or T[], and doing dereferences transparently in private layer it returns a DataExtractor ; in public layer it returns an instance of a newly created lldb::SBData - as GetPointeeData() does the right thing for both frozen and non-frozen ValueObject's, reimplemented ReadPointedString() to use it en lieu of doing the raw read itself - introduced a new GetData() method in ValueObject that lets you get a copy of the data that backs the ValueObject (for pointers, this returns the address without any previous dereferencing steps ; for arrays it actually reads the whole chunk of memory) in public layer this returns an SBData, just like GetPointeeData() - introduced a new CreateValueFromData() method in SBValue that lets you create a new SBValue from a chunk of data wrapped in an SBData the limitation to remember for this kind of SBValue is that they have no address: extracting the address-of for these objects (with any of GetAddress(), GetLoadAddress() and AddressOf()) will return invalid values - added several tests to check that "p"-ing objects (STL classes, char* and char[]) will do the right thing Solved a bug where global pointers to global variables were not dereferenced correctly for display New target setting "max-string-summary-length" gives the maximum number of characters to show in a string when summarizing it, instead of the hardcoded 128 Solved a bug where the summary for char[] and char* would not be shown if the ValueObject's were dumped via the "p" command Removed m_pointers_point_to_load_addrs from ValueObject. Introduced a new m_address_type_of_children, which each ValueObject can set to tell the address type of any pointers and/or references it creates. In the current codebase, this is load address most of the time (the only notable exception being file addresses that generate file address children UNLESS we have a live process) Updated help text for summary-string Fixed an issue in STL formatters where std::stlcontainer::iterator would match the container's synthetic children providers Edited the syntax and help for some commands to have proper argument types llvm-svn: 139160
2011-09-07 03:20:51 +08:00
return GetCategory(category_name);
Additional code cleanups ; Short option name for --python-script in type summary add moved from -s to -o (this is a preliminary step in moving the short option for --summary-string from -f to -s) ; Accordingly updated the test suite llvm-svn: 138315
2011-08-23 08:32:52 +08:00
}
smarter summary strings: - formats %s %char[] %c and %a now work to print 0-terminated c-strings if they are applied to a char* or char[] even without the [] operator (e.g. ${var%s}) - array formats (char[], intN[], ..) now work when applied to an array of a scalar type even without the [] operator (e.g. ${var%int32_t[]}) LLDB will not crash because of endless loop when trying to obtain a summary for an object that has no value and references itself in its summary string In many cases, a wrong summary string will now display an "<error>" message instead of giving out an empty string llvm-svn: 135007
2011-07-13 06:56:10 +08:00
lldb::Format
FormatManager::GetSingleItemFormat(lldb::Format vector_format)
{
switch(vector_format)
{
case eFormatVectorOfChar:
return eFormatCharArray;
case eFormatVectorOfSInt8:
case eFormatVectorOfSInt16:
case eFormatVectorOfSInt32:
case eFormatVectorOfSInt64:
return eFormatDecimal;
case eFormatVectorOfUInt8:
case eFormatVectorOfUInt16:
case eFormatVectorOfUInt32:
case eFormatVectorOfUInt64:
case eFormatVectorOfUInt128:
return eFormatHex;
case eFormatVectorOfFloat32:
case eFormatVectorOfFloat64:
return eFormatFloat;
default:
return lldb::eFormatInvalid;
}
While tracking down memory consumption issue a few things were needed: the ability to dump more information about modules in "target modules list". We can now dump the shared pointer reference count for modules, the pointer to the module itself (in case performance tools can help track down who has references to said pointer), and the modification time. Added "target delete [target-idx ...]" to be able to delete targets when they are no longer needed. This will help track down memory usage issues and help to resolve when module ref counts keep getting incremented. If the command gets no arguments, the currently selected target will be deleted. If any arguments are given, they must all be valid target indexes (use the "target list" command to get the current target indexes). Took care of a bunch of "no newline at end of file" warnings. TimeValue objects can now dump their time to a lldb_private::Stream object. Modified the "target modules list --global" command to not error out if there are no targets since it doesn't require a target. Fixed an issue in the MacOSX DYLD dynamic loader plug-in where if a shared library was updated on disk, we would keep using the older one, even if it was updated. Don't allow the ModuleList::GetSharedModule(...) to return an empty module. Previously we could specify a valid path on disc to a module, and specify an architecture that wasn't contained in that module and get a shared pointer to a module that wouldn't be able to return an object file or a symbol file. We now make sure an object file can be extracted prior to adding the shared pointer to the module to get added to the shared list. llvm-svn: 137196
2011-08-10 10:10:13 +08:00
}
First round of code cleanups: - all instances of "vobj" have been renamed to "valobj" - class Debugger::Formatting has been renamed to DataVisualization (defined in FormatManager.h/cpp) The interface to this class has not changed - FormatCategory now uses ConstString's as keys to the navigators instead of repeatedly casting from ConstString to const char* and back all the time Next step is making the same happen for categories themselves - category gnu-libstdc++ is defined in the constructor for a FormatManager The source code for it is defined in gnu_libstdcpp.py, drawn from examples/synthetic at compile time All references to previous 'osxcpp' name have been removed from both code and file names Functional changes: - the name of the option to use a summary string for 'type summary add' has changed from the previous --format-string to the new --summary-string. It is expected that the short option will change from -f to -s, and -s for --python-script will become -o llvm-svn: 137886
2011-08-18 06:13:59 +08:00
<rdar://problem/12042982> This radar extends the notion of one-liner summaries to automagically apply in a few interesting cases More specifically, this checkin changes the printout of ValueObjects to print on one-line (as if type summary add -c had been applied) iff: this ValueObject does not have a summary its children have no synthetic children its children are not a non-empty base class without a summary its children do not have a summary that asks for children to show up the aggregate length of all the names of all the children is <= 50 characters you did not ask to see the types during a printout your pointer depth is 0 This is meant to simplify the way LLDB shows data on screen for small structs and similarly compact data types (e.g. std::pair<int,int> anyone?) Feedback is especially welcome on how the feature feels and corner cases where we should apply this printout and don't (or viceversa, we are applying it when we shouldn't be) llvm-svn: 191996
2013-10-05 07:14:13 +08:00
bool
FormatManager::ShouldPrintAsOneLiner (ValueObject& valobj)
{
<rdar://problem/15319880> Introduce a new boolean setting enable-auto-oneliner This setting if set to false will force LLDB to not use the new compact one-line display By default, one-line mode stays on, at least until we can be confident it works. But now if it seriously impedes your workflow while it evolves/it works wonders but you still hate it, there's a way to turn it off llvm-svn: 193450
2013-10-26 07:09:40 +08:00
// if settings say no oneline whatsoever
Renaming the setting to enable/disable automatic one-lining of summaries as auto-one-line-summaries llvm-svn: 193801
2013-11-01 05:01:07 +08:00
if (valobj.GetTargetSP().get() && valobj.GetTargetSP()->GetDebugger().GetAutoOneLineSummaries() == false)
<rdar://problem/15319880> Introduce a new boolean setting enable-auto-oneliner This setting if set to false will force LLDB to not use the new compact one-line display By default, one-line mode stays on, at least until we can be confident it works. But now if it seriously impedes your workflow while it evolves/it works wonders but you still hate it, there's a way to turn it off llvm-svn: 193450
2013-10-26 07:09:40 +08:00
return false; // then don't oneline
<rdar://problem/12042982> This radar extends the notion of one-liner summaries to automagically apply in a few interesting cases More specifically, this checkin changes the printout of ValueObjects to print on one-line (as if type summary add -c had been applied) iff: this ValueObject does not have a summary its children have no synthetic children its children are not a non-empty base class without a summary its children do not have a summary that asks for children to show up the aggregate length of all the names of all the children is <= 50 characters you did not ask to see the types during a printout your pointer depth is 0 This is meant to simplify the way LLDB shows data on screen for small structs and similarly compact data types (e.g. std::pair<int,int> anyone?) Feedback is especially welcome on how the feature feels and corner cases where we should apply this printout and don't (or viceversa, we are applying it when we shouldn't be) llvm-svn: 191996
2013-10-05 07:14:13 +08:00
// if this object has a summary, don't try to do anything special to it
// if the user wants one-liner, they can ask for it in summary :)
if (valobj.GetSummaryFormat().get() != nullptr)
return false;
// no children, no party
if (valobj.GetNumChildren() == 0)
return false;
size_t total_children_name_len = 0;
for (size_t idx = 0;
idx < valobj.GetNumChildren();
idx++)
{
ValueObjectSP child_sp(valobj.GetChildAtIndex(idx, true));
// something is wrong here - bail out
if (!child_sp)
return false;
// if we decided to define synthetic children for a type, we probably care enough
// to show them, but avoid nesting children in children
if (child_sp->GetSyntheticChildren().get() != nullptr)
return false;
total_children_name_len += child_sp->GetName().GetLength();
// 50 itself is a "randomly" chosen number - the idea is that
// overly long structs should not get this treatment
// FIXME: maybe make this a user-tweakable setting?
if (total_children_name_len > 50)
return false;
// if a summary is there..
if (child_sp->GetSummaryFormat())
{
// and it wants children, then bail out
Allow summary formatters to take ValueObjects into account when deciding whether values/children should be printed and if child names should be shown This decision has always been statically-bound to the individual formatter. With this patch, the idea is that this decision could potentially be dynamic depending on the ValueObject itself llvm-svn: 207046
2014-04-24 07:16:25 +08:00
if (child_sp->GetSummaryFormat()->DoesPrintChildren(child_sp.get()))
<rdar://problem/12042982> This radar extends the notion of one-liner summaries to automagically apply in a few interesting cases More specifically, this checkin changes the printout of ValueObjects to print on one-line (as if type summary add -c had been applied) iff: this ValueObject does not have a summary its children have no synthetic children its children are not a non-empty base class without a summary its children do not have a summary that asks for children to show up the aggregate length of all the names of all the children is <= 50 characters you did not ask to see the types during a printout your pointer depth is 0 This is meant to simplify the way LLDB shows data on screen for small structs and similarly compact data types (e.g. std::pair<int,int> anyone?) Feedback is especially welcome on how the feature feels and corner cases where we should apply this printout and don't (or viceversa, we are applying it when we shouldn't be) llvm-svn: 191996
2013-10-05 07:14:13 +08:00
return false;
}
One should actually not do one-line printing of nested aggregates even if they are not the base class This check was overly strict. Relax it. While one could conceivably want nested one-lining: (Foo) aFoo = (x = 1, y = (t = 3, q = “Hello), z = 3.14) the spirit of this feature is mostly to make *SMALL LINEAR* structs come out more compact. Aggregates with children and no summary for now just disable the one-lining. Define a one-liner summary to override :) llvm-svn: 193218
2013-10-23 09:34:31 +08:00
// if this child has children..
if (child_sp->GetNumChildren())
<rdar://problem/12042982> This radar extends the notion of one-liner summaries to automagically apply in a few interesting cases More specifically, this checkin changes the printout of ValueObjects to print on one-line (as if type summary add -c had been applied) iff: this ValueObject does not have a summary its children have no synthetic children its children are not a non-empty base class without a summary its children do not have a summary that asks for children to show up the aggregate length of all the names of all the children is <= 50 characters you did not ask to see the types during a printout your pointer depth is 0 This is meant to simplify the way LLDB shows data on screen for small structs and similarly compact data types (e.g. std::pair<int,int> anyone?) Feedback is especially welcome on how the feature feels and corner cases where we should apply this printout and don't (or viceversa, we are applying it when we shouldn't be) llvm-svn: 191996
2013-10-05 07:14:13 +08:00
{
One should actually not do one-line printing of nested aggregates even if they are not the base class This check was overly strict. Relax it. While one could conceivably want nested one-lining: (Foo) aFoo = (x = 1, y = (t = 3, q = “Hello), z = 3.14) the spirit of this feature is mostly to make *SMALL LINEAR* structs come out more compact. Aggregates with children and no summary for now just disable the one-lining. Define a one-liner summary to override :) llvm-svn: 193218
2013-10-23 09:34:31 +08:00
// ...and no summary...
// (if it had a summary and the summary wanted children, we would have bailed out anyway
// so this only makes us bail out if this has no summary and we would then print children)
if (!child_sp->GetSummaryFormat())
return false; // then bail out
<rdar://problem/12042982> This radar extends the notion of one-liner summaries to automagically apply in a few interesting cases More specifically, this checkin changes the printout of ValueObjects to print on one-line (as if type summary add -c had been applied) iff: this ValueObject does not have a summary its children have no synthetic children its children are not a non-empty base class without a summary its children do not have a summary that asks for children to show up the aggregate length of all the names of all the children is <= 50 characters you did not ask to see the types during a printout your pointer depth is 0 This is meant to simplify the way LLDB shows data on screen for small structs and similarly compact data types (e.g. std::pair<int,int> anyone?) Feedback is especially welcome on how the feature feels and corner cases where we should apply this printout and don't (or viceversa, we are applying it when we shouldn't be) llvm-svn: 191996
2013-10-05 07:14:13 +08:00
}
}
return true;
}
Third round of code cleanups: - reorganizing the PTS (Partial Template Specializations) in FormatManager.h - applied a patch by Filipe Cabecinhas to make LLDB compile with GCC Functional changes: - fixed an issue where command type summary add for type "struct Foo" would not match any types. currently, "struct" will be stripped off and type "Foo" will be matched. similar behavior occurs for class, enum and union specifiers. llvm-svn: 138020
2011-08-19 09:14:49 +08:00
ConstString
FormatManager::GetValidTypeName (const ConstString& type)
{
return ::GetValidTypeName_Impl(type);
}
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
ConstString
GetTypeForCache (ValueObject& valobj,
lldb::DynamicValueType use_dynamic)
{
if (use_dynamic == lldb::eNoDynamicValues)
{
if (valobj.IsDynamic())
{
if (valobj.GetStaticValue())
return valobj.GetStaticValue()->GetQualifiedTypeName();
else
return ConstString();
}
else
return valobj.GetQualifiedTypeName();
}
if (valobj.IsDynamic())
return valobj.GetQualifiedTypeName();
if (valobj.GetDynamicValue(use_dynamic))
return valobj.GetDynamicValue(use_dynamic)->GetQualifiedTypeName();
return ConstString();
}
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
lldb::TypeFormatImplSP
FormatManager::GetHardcodedFormat (ValueObject& valobj,
lldb::DynamicValueType use_dynamic)
This checkin introduces the notion of hardcoded formatters, which LLDB can bind to a ValueObject internally depending on any criteria User-vended by-type formatters still would prevail on these hardcoded ones For the time being, while the infrastructure is there, no such formatters exist This can be useful for cases such as expanding vtables for C++ class pointers, when there is no clear cut notion of a typename matching, and the feature is low-level enough that it makes sense for the debugger core to be vending it llvm-svn: 193724
2013-10-31 07:46:27 +08:00
{
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
for (const auto& candidate: m_hardcoded_formats)
{
auto result = candidate(valobj,use_dynamic,*this);
if (result)
return result;
}
return nullptr;
This checkin introduces the notion of hardcoded formatters, which LLDB can bind to a ValueObject internally depending on any criteria User-vended by-type formatters still would prevail on these hardcoded ones For the time being, while the infrastructure is there, no such formatters exist This can be useful for cases such as expanding vtables for C++ class pointers, when there is no clear cut notion of a typename matching, and the feature is low-level enough that it makes sense for the debugger core to be vending it llvm-svn: 193724
2013-10-31 07:46:27 +08:00
}
<rdar://problem/11778815> Formats (as in "type format") are now included in categories The only bit missing is caching formats along with synthetic children and summaries, which might be now desirable llvm-svn: 192217
2013-10-09 03:03:22 +08:00
lldb::TypeFormatImplSP
FormatManager::GetFormat (ValueObject& valobj,
lldb::DynamicValueType use_dynamic)
{
TypeFormatImplSP retval;
This is the last piece of work for "formats in categories": we now cache formats as well as summaries and synthetics llvm-svn: 192928
2013-10-18 06:27:19 +08:00
Log *log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_TYPES));
ConstString valobj_type(GetTypeForCache(valobj, use_dynamic));
if (valobj_type)
{
if (log)
log->Printf("\n\n[FormatManager::GetFormat] Looking into cache for type %s", valobj_type.AsCString("<invalid>"));
if (m_format_cache.GetFormat(valobj_type,retval))
{
if (log)
{
log->Printf("[FormatManager::GetFormat] Cache search success. Returning.");
if (log->GetDebug())
log->Printf("[FormatManager::GetFormat] Cache hits: %" PRIu64 " - Cache Misses: %" PRIu64, m_format_cache.GetCacheHits(), m_format_cache.GetCacheMisses());
}
return retval;
}
if (log)
log->Printf("[FormatManager::GetFormat] Cache search failed. Going normal route");
}
<rdar://problem/11778815> Formats (as in "type format") are now included in categories The only bit missing is caching formats along with synthetic children and summaries, which might be now desirable llvm-svn: 192217
2013-10-09 03:03:22 +08:00
retval = m_categories_map.GetFormat(valobj, use_dynamic);
This checkin introduces the notion of hardcoded formatters, which LLDB can bind to a ValueObject internally depending on any criteria User-vended by-type formatters still would prevail on these hardcoded ones For the time being, while the infrastructure is there, no such formatters exist This can be useful for cases such as expanding vtables for C++ class pointers, when there is no clear cut notion of a typename matching, and the feature is low-level enough that it makes sense for the debugger core to be vending it llvm-svn: 193724
2013-10-31 07:46:27 +08:00
if (!retval)
{
if (log)
log->Printf("[FormatManager::GetFormat] Search failed. Giving hardcoded a chance.");
retval = GetHardcodedFormat(valobj, use_dynamic);
}
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
else if (valobj_type)
This is the last piece of work for "formats in categories": we now cache formats as well as summaries and synthetics llvm-svn: 192928
2013-10-18 06:27:19 +08:00
{
if (log)
sweep up -Wformat warnings from gcc This is a purely mechanical change explicitly casting any parameters for printf style conversion. This cleans up the warnings emitted by gcc 4.8 on Linux. llvm-svn: 205607
2014-04-04 12:06:10 +08:00
log->Printf("[FormatManager::GetFormat] Caching %p for type %s",
static_cast<void*>(retval.get()),
valobj_type.AsCString("<invalid>"));
This is the last piece of work for "formats in categories": we now cache formats as well as summaries and synthetics llvm-svn: 192928
2013-10-18 06:27:19 +08:00
m_format_cache.SetFormat(valobj_type,retval);
}
if (log && log->GetDebug())
log->Printf("[FormatManager::GetFormat] Cache hits: %" PRIu64 " - Cache Misses: %" PRIu64, m_format_cache.GetCacheHits(), m_format_cache.GetCacheMisses());
<rdar://problem/11778815> Formats (as in "type format") are now included in categories The only bit missing is caching formats along with synthetic children and summaries, which might be now desirable llvm-svn: 192217
2013-10-09 03:03:22 +08:00
return retval;
}
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
lldb::TypeSummaryImplSP
FormatManager::GetHardcodedSummaryFormat (ValueObject& valobj,
lldb::DynamicValueType use_dynamic)
This checkin introduces the notion of hardcoded formatters, which LLDB can bind to a ValueObject internally depending on any criteria User-vended by-type formatters still would prevail on these hardcoded ones For the time being, while the infrastructure is there, no such formatters exist This can be useful for cases such as expanding vtables for C++ class pointers, when there is no clear cut notion of a typename matching, and the feature is low-level enough that it makes sense for the debugger core to be vending it llvm-svn: 193724
2013-10-31 07:46:27 +08:00
{
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
for (const auto& candidate: m_hardcoded_summaries)
{
auto result = candidate(valobj,use_dynamic,*this);
if (result)
return result;
}
return nullptr;
This checkin introduces the notion of hardcoded formatters, which LLDB can bind to a ValueObject internally depending on any criteria User-vended by-type formatters still would prevail on these hardcoded ones For the time being, while the infrastructure is there, no such formatters exist This can be useful for cases such as expanding vtables for C++ class pointers, when there is no clear cut notion of a typename matching, and the feature is low-level enough that it makes sense for the debugger core to be vending it llvm-svn: 193724
2013-10-31 07:46:27 +08:00
}
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
lldb::TypeSummaryImplSP
FormatManager::GetSummaryFormat (ValueObject& valobj,
lldb::DynamicValueType use_dynamic)
{
TypeSummaryImplSP retval;
<rdar://problem/13521159> LLDB is crashing when logging is enabled from lldb-perf-clang. This has to do with the global destructor chain as the process and its threads are being torn down. All logging channels now make one and only one instance that is kept in a global pointer which is never freed. This guarantees that logging can correctly continue as the process tears itself down. llvm-svn: 178191
2013-03-28 07:08:40 +08:00
Log *log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_TYPES));
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
ConstString valobj_type(GetTypeForCache(valobj, use_dynamic));
if (valobj_type)
{
if (log)
<rdar://problem/12717717> Modifying our data formatters matching algorithm to ensure that "const X*" is treated as equivalent to "X*" Also, a couple improvements to the "lldb types" logging llvm-svn: 184215
2013-06-19 02:23:07 +08:00
log->Printf("\n\n[FormatManager::GetSummaryFormat] Looking into cache for type %s", valobj_type.AsCString("<invalid>"));
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
if (m_format_cache.GetSummary(valobj_type,retval))
<rdar://problem/12717717> Modifying our data formatters matching algorithm to ensure that "const X*" is treated as equivalent to "X*" Also, a couple improvements to the "lldb types" logging llvm-svn: 184215
2013-06-19 02:23:07 +08:00
{
if (log)
{
log->Printf("[FormatManager::GetSummaryFormat] Cache search success. Returning.");
Remove the #define USE_CACHE since the formatters cache has been operational for a while now and has not caused issues that warrant disabling it Also, print the cache hits statistics if the log is in debugging mode vs. LLDB being a debug build - this should make it easier to gather useful metrics on cache success rate for real users llvm-svn: 184900
2013-06-26 09:03:38 +08:00
if (log->GetDebug())
clean up about 22 warnings messages llvm-svn: 187900
2013-08-08 03:05:15 +08:00
log->Printf("[FormatManager::GetSummaryFormat] Cache hits: %" PRIu64 " - Cache Misses: %" PRIu64, m_format_cache.GetCacheHits(), m_format_cache.GetCacheMisses());
<rdar://problem/12717717> Modifying our data formatters matching algorithm to ensure that "const X*" is treated as equivalent to "X*" Also, a couple improvements to the "lldb types" logging llvm-svn: 184215
2013-06-19 02:23:07 +08:00
}
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
return retval;
<rdar://problem/12717717> Modifying our data formatters matching algorithm to ensure that "const X*" is treated as equivalent to "X*" Also, a couple improvements to the "lldb types" logging llvm-svn: 184215
2013-06-19 02:23:07 +08:00
}
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
if (log)
log->Printf("[FormatManager::GetSummaryFormat] Cache search failed. Going normal route");
}
retval = m_categories_map.GetSummaryFormat(valobj, use_dynamic);
This checkin introduces the notion of hardcoded formatters, which LLDB can bind to a ValueObject internally depending on any criteria User-vended by-type formatters still would prevail on these hardcoded ones For the time being, while the infrastructure is there, no such formatters exist This can be useful for cases such as expanding vtables for C++ class pointers, when there is no clear cut notion of a typename matching, and the feature is low-level enough that it makes sense for the debugger core to be vending it llvm-svn: 193724
2013-10-31 07:46:27 +08:00
if (!retval)
{
if (log)
log->Printf("[FormatManager::GetSummaryFormat] Search failed. Giving hardcoded a chance.");
retval = GetHardcodedSummaryFormat(valobj, use_dynamic);
}
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
else if (valobj_type)
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
{
if (log)
sweep up -Wformat warnings from gcc This is a purely mechanical change explicitly casting any parameters for printf style conversion. This cleans up the warnings emitted by gcc 4.8 on Linux. llvm-svn: 205607
2014-04-04 12:06:10 +08:00
log->Printf("[FormatManager::GetSummaryFormat] Caching %p for type %s",
static_cast<void*>(retval.get()),
valobj_type.AsCString("<invalid>"));
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
m_format_cache.SetSummary(valobj_type,retval);
}
Remove the #define USE_CACHE since the formatters cache has been operational for a while now and has not caused issues that warrant disabling it Also, print the cache hits statistics if the log is in debugging mode vs. LLDB being a debug build - this should make it easier to gather useful metrics on cache success rate for real users llvm-svn: 184900
2013-06-26 09:03:38 +08:00
if (log && log->GetDebug())
clean up about 22 warnings messages llvm-svn: 187900
2013-08-08 03:05:15 +08:00
log->Printf("[FormatManager::GetSummaryFormat] Cache hits: %" PRIu64 " - Cache Misses: %" PRIu64, m_format_cache.GetCacheHits(), m_format_cache.GetCacheMisses());
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
return retval;
}
#ifndef LLDB_DISABLE_PYTHON
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
lldb::SyntheticChildrenSP
FormatManager::GetHardcodedSyntheticChildren (ValueObject& valobj,
lldb::DynamicValueType use_dynamic)
This checkin introduces the notion of hardcoded formatters, which LLDB can bind to a ValueObject internally depending on any criteria User-vended by-type formatters still would prevail on these hardcoded ones For the time being, while the infrastructure is there, no such formatters exist This can be useful for cases such as expanding vtables for C++ class pointers, when there is no clear cut notion of a typename matching, and the feature is low-level enough that it makes sense for the debugger core to be vending it llvm-svn: 193724
2013-10-31 07:46:27 +08:00
{
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
for (const auto& candidate: m_hardcoded_synthetics)
{
auto result = candidate(valobj,use_dynamic,*this);
if (result)
return result;
}
return nullptr;
This checkin introduces the notion of hardcoded formatters, which LLDB can bind to a ValueObject internally depending on any criteria User-vended by-type formatters still would prevail on these hardcoded ones For the time being, while the infrastructure is there, no such formatters exist This can be useful for cases such as expanding vtables for C++ class pointers, when there is no clear cut notion of a typename matching, and the feature is low-level enough that it makes sense for the debugger core to be vending it llvm-svn: 193724
2013-10-31 07:46:27 +08:00
}
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
lldb::SyntheticChildrenSP
FormatManager::GetSyntheticChildren (ValueObject& valobj,
Remove the #define USE_CACHE since the formatters cache has been operational for a while now and has not caused issues that warrant disabling it Also, print the cache hits statistics if the log is in debugging mode vs. LLDB being a debug build - this should make it easier to gather useful metrics on cache success rate for real users llvm-svn: 184900
2013-06-26 09:03:38 +08:00
lldb::DynamicValueType use_dynamic)
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
{
SyntheticChildrenSP retval;
<rdar://problem/13521159> LLDB is crashing when logging is enabled from lldb-perf-clang. This has to do with the global destructor chain as the process and its threads are being torn down. All logging channels now make one and only one instance that is kept in a global pointer which is never freed. This guarantees that logging can correctly continue as the process tears itself down. llvm-svn: 178191
2013-03-28 07:08:40 +08:00
Log *log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_TYPES));
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
ConstString valobj_type(GetTypeForCache(valobj, use_dynamic));
if (valobj_type)
{
if (log)
<rdar://problem/12717717> Modifying our data formatters matching algorithm to ensure that "const X*" is treated as equivalent to "X*" Also, a couple improvements to the "lldb types" logging llvm-svn: 184215
2013-06-19 02:23:07 +08:00
log->Printf("\n\n[FormatManager::GetSyntheticChildren] Looking into cache for type %s", valobj_type.AsCString("<invalid>"));
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
if (m_format_cache.GetSynthetic(valobj_type,retval))
<rdar://problem/12717717> Modifying our data formatters matching algorithm to ensure that "const X*" is treated as equivalent to "X*" Also, a couple improvements to the "lldb types" logging llvm-svn: 184215
2013-06-19 02:23:07 +08:00
{
if (log)
{
log->Printf("[FormatManager::GetSyntheticChildren] Cache search success. Returning.");
Remove the #define USE_CACHE since the formatters cache has been operational for a while now and has not caused issues that warrant disabling it Also, print the cache hits statistics if the log is in debugging mode vs. LLDB being a debug build - this should make it easier to gather useful metrics on cache success rate for real users llvm-svn: 184900
2013-06-26 09:03:38 +08:00
if (log->GetDebug())
clean up about 22 warnings messages llvm-svn: 187900
2013-08-08 03:05:15 +08:00
log->Printf("[FormatManager::GetSyntheticChildren] Cache hits: %" PRIu64 " - Cache Misses: %" PRIu64, m_format_cache.GetCacheHits(), m_format_cache.GetCacheMisses());
<rdar://problem/12717717> Modifying our data formatters matching algorithm to ensure that "const X*" is treated as equivalent to "X*" Also, a couple improvements to the "lldb types" logging llvm-svn: 184215
2013-06-19 02:23:07 +08:00
}
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
return retval;
<rdar://problem/12717717> Modifying our data formatters matching algorithm to ensure that "const X*" is treated as equivalent to "X*" Also, a couple improvements to the "lldb types" logging llvm-svn: 184215
2013-06-19 02:23:07 +08:00
}
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
if (log)
<rdar://problem/12717717> Modifying our data formatters matching algorithm to ensure that "const X*" is treated as equivalent to "X*" Also, a couple improvements to the "lldb types" logging llvm-svn: 184215
2013-06-19 02:23:07 +08:00
log->Printf("[FormatManager::GetSyntheticChildren] Cache search failed. Going normal route");
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
}
retval = m_categories_map.GetSyntheticChildren(valobj, use_dynamic);
This checkin introduces the notion of hardcoded formatters, which LLDB can bind to a ValueObject internally depending on any criteria User-vended by-type formatters still would prevail on these hardcoded ones For the time being, while the infrastructure is there, no such formatters exist This can be useful for cases such as expanding vtables for C++ class pointers, when there is no clear cut notion of a typename matching, and the feature is low-level enough that it makes sense for the debugger core to be vending it llvm-svn: 193724
2013-10-31 07:46:27 +08:00
if (!retval)
{
if (log)
log->Printf("[FormatManager::GetSyntheticChildren] Search failed. Giving hardcoded a chance.");
retval = GetHardcodedSyntheticChildren(valobj, use_dynamic);
}
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
else if (valobj_type)
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
{
if (log)
sweep up -Wformat warnings from gcc This is a purely mechanical change explicitly casting any parameters for printf style conversion. This cleans up the warnings emitted by gcc 4.8 on Linux. llvm-svn: 205607
2014-04-04 12:06:10 +08:00
log->Printf("[FormatManager::GetSyntheticChildren] Caching %p for type %s",
static_cast<void*>(retval.get()),
valobj_type.AsCString("<invalid>"));
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
m_format_cache.SetSynthetic(valobj_type,retval);
}
Remove the #define USE_CACHE since the formatters cache has been operational for a while now and has not caused issues that warrant disabling it Also, print the cache hits statistics if the log is in debugging mode vs. LLDB being a debug build - this should make it easier to gather useful metrics on cache success rate for real users llvm-svn: 184900
2013-06-26 09:03:38 +08:00
if (log && log->GetDebug())
clean up about 22 warnings messages llvm-svn: 187900
2013-08-08 03:05:15 +08:00
log->Printf("[FormatManager::GetSyntheticChildren] Cache hits: %" PRIu64 " - Cache Misses: %" PRIu64, m_format_cache.GetCacheHits(), m_format_cache.GetCacheMisses());
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
return retval;
}
#endif
FormatManager::FormatManager() :
m_format_cache(),
First round of code cleanups: - all instances of "vobj" have been renamed to "valobj" - class Debugger::Formatting has been renamed to DataVisualization (defined in FormatManager.h/cpp) The interface to this class has not changed - FormatCategory now uses ConstString's as keys to the navigators instead of repeatedly casting from ConstString to const char* and back all the time Next step is making the same happen for categories themselves - category gnu-libstdc++ is defined in the constructor for a FormatManager The source code for it is defined in gnu_libstdcpp.py, drawn from examples/synthetic at compile time All references to previous 'osxcpp' name have been removed from both code and file names Functional changes: - the name of the option to use a summary string for 'type summary add' has changed from the previous --format-string to the new --summary-string. It is expected that the short option will change from -f to -s, and -s for --python-script will become -o llvm-svn: 137886
2011-08-18 06:13:59 +08:00
m_named_summaries_map(this),
m_last_revision(0),
m_categories_map(this),
Code cleanup and refactoring (round 4): - FormatCategories now are directly mapped by ConstString objects instead of going through const char* -> ConstString -> const char* - FormatCategory callback does not pass category name anymore. This is not necessary because FormatCategory objects themselves hold their name as a member variable llvm-svn: 138254
2011-08-23 02:36:52 +08:00
m_default_category_name(ConstString("default")),
m_system_category_name(ConstString("system")),
This commit provides a new default summary for Objective-C boolean variables, which shows YES or NO instead of the character value. A new category named objc is added to contain this summary provider. Any future Objective-C related formatters would probably fit here llvm-svn: 149388
2012-02-01 01:01:51 +08:00
m_gnu_cpp_category_name(ConstString("gnu-libstdc++")),
Added formatters for libc++ (http://libcxx.llvm.org): std::string has a summary provider std::vector std::list and std::map have both a summary and a synthetic children provider Given the usage of a custom namespace (std::__1::classname) for the implementation of libc++, we keep both libstdcpp and libc++ formatters enabled at the same time since that raises no conflicts and enabled for seamless transition between the two The formatters for libc++ reside in a libcxx category, and are loaded from libcxx.py (to be found in examples/synthetic) The formatters-stl test cases have been divided to be separate for libcxx and libstdcpp. This separation is necessary because (a) we need different compiler flags for libc++ than for libstdcpp (b) libc++ inlines a lot more than libstdcpp and some code changes were required to accommodate this difference llvm-svn: 152570
2012-03-13 03:47:17 +08:00
m_libcxx_category_name(ConstString("libcxx")),
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
m_objc_category_name(ConstString("objc")),
m_corefoundation_category_name(ConstString("CoreFoundation")),
m_coregraphics_category_name(ConstString("CoreGraphics")),
m_coreservices_category_name(ConstString("CoreServices")),
This patch provides a set of formatters for most of the commonly used Cocoa classes. The formatter for NSString is an improved version of the one previously shipped as an example, the others are new in design and implementation. A more robust and OO-compliant Objective-C runtime wrapper is provided for runtime versions 1 and 2 on 32 and 64 bit. The formatters are contained in a category named "AppKit", which is not enabled at startup. llvm-svn: 151299
2012-02-24 07:10:03 +08:00
m_vectortypes_category_name(ConstString("VectorTypes")),
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
m_appkit_category_name(ConstString("AppKit")),
m_hardcoded_formats(),
m_hardcoded_summaries(),
m_hardcoded_synthetics()
First round of code cleanups: - all instances of "vobj" have been renamed to "valobj" - class Debugger::Formatting has been renamed to DataVisualization (defined in FormatManager.h/cpp) The interface to this class has not changed - FormatCategory now uses ConstString's as keys to the navigators instead of repeatedly casting from ConstString to const char* and back all the time Next step is making the same happen for categories themselves - category gnu-libstdc++ is defined in the constructor for a FormatManager The source code for it is defined in gnu_libstdcpp.py, drawn from examples/synthetic at compile time All references to previous 'osxcpp' name have been removed from both code and file names Functional changes: - the name of the option to use a summary string for 'type summary add' has changed from the previous --format-string to the new --summary-string. It is expected that the short option will change from -f to -s, and -s for --python-script will become -o llvm-svn: 137886
2011-08-18 06:13:59 +08:00
{
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
LoadSystemFormatters();
<rdar://problem/12239827> Making a summary for std::wstring as provided by libstdc++ along with a relevant test case llvm-svn: 172286
2013-01-12 09:22:57 +08:00
LoadLibStdcppFormatters();
Added formatters for libc++ (http://libcxx.llvm.org): std::string has a summary provider std::vector std::list and std::map have both a summary and a synthetic children provider Given the usage of a custom namespace (std::__1::classname) for the implementation of libc++, we keep both libstdcpp and libc++ formatters enabled at the same time since that raises no conflicts and enabled for seamless transition between the two The formatters for libc++ reside in a libcxx category, and are loaded from libcxx.py (to be found in examples/synthetic) The formatters-stl test cases have been divided to be separate for libcxx and libstdcpp. This separation is necessary because (a) we need different compiler flags for libc++ than for libstdcpp (b) libc++ inlines a lot more than libstdcpp and some code changes were required to accommodate this difference llvm-svn: 152570
2012-03-13 03:47:17 +08:00
LoadLibcxxFormatters();
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
LoadObjCFormatters();
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
LoadHardcodedFormatters();
First round of code cleanups: - all instances of "vobj" have been renamed to "valobj" - class Debugger::Formatting has been renamed to DataVisualization (defined in FormatManager.h/cpp) The interface to this class has not changed - FormatCategory now uses ConstString's as keys to the navigators instead of repeatedly casting from ConstString to const char* and back all the time Next step is making the same happen for categories themselves - category gnu-libstdc++ is defined in the constructor for a FormatManager The source code for it is defined in gnu_libstdcpp.py, drawn from examples/synthetic at compile time All references to previous 'osxcpp' name have been removed from both code and file names Functional changes: - the name of the option to use a summary string for 'type summary add' has changed from the previous --format-string to the new --summary-string. It is expected that the short option will change from -f to -s, and -s for --python-script will become -o llvm-svn: 137886
2011-08-18 06:13:59 +08:00
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
EnableCategory(m_objc_category_name,TypeCategoryMap::Last);
EnableCategory(m_corefoundation_category_name,TypeCategoryMap::Last);
EnableCategory(m_appkit_category_name,TypeCategoryMap::Last);
EnableCategory(m_coreservices_category_name,TypeCategoryMap::Last);
EnableCategory(m_coregraphics_category_name,TypeCategoryMap::Last);
EnableCategory(m_gnu_cpp_category_name,TypeCategoryMap::Last);
EnableCategory(m_libcxx_category_name,TypeCategoryMap::Last);
EnableCategory(m_vectortypes_category_name,TypeCategoryMap::Last);
EnableCategory(m_system_category_name,TypeCategoryMap::Last);
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
}
<rdar://problem/12632394> Add a format for FourCharCode This is now safe to do thanks to the "formats in categories" feature llvm-svn: 192233
2013-10-09 04:59:02 +08:00
static void
AddFormat (TypeCategoryImpl::SharedPointer category_sp,
lldb::Format format,
ConstString type_name,
TypeFormatImpl::Flags flags,
bool regex = false)
{
Add a new way to bind a format to a type: by enum type The "type format add" command gets a new flag --type (-t). If you pass -t <sometype>, upon fetching the value for an object of your type, LLDB will display it as-if it was of enumeration type <sometype> This is useful in cases of non-contiguous enums where there are empty gaps of unspecified values, and as such one cannot type their variables as the enum type, but users would still like to see them as-if they were of the enum type (e.g. DWARF field types with their user-reserved ranges) The SB API has also been improved to handle both types of formats, and a test case is added llvm-svn: 198105
2013-12-28 16:44:02 +08:00
lldb::TypeFormatImplSP format_sp(new TypeFormatImpl_Format(format, flags));
<rdar://problem/12632394> Add a format for FourCharCode This is now safe to do thanks to the "formats in categories" feature llvm-svn: 192233
2013-10-09 04:59:02 +08:00
if (regex)
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
category_sp->GetRegexTypeFormatsContainer()->Add(RegularExpressionSP(new RegularExpression(type_name.AsCString())),format_sp);
<rdar://problem/12632394> Add a format for FourCharCode This is now safe to do thanks to the "formats in categories" feature llvm-svn: 192233
2013-10-09 04:59:02 +08:00
else
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
category_sp->GetTypeFormatsContainer()->Add(type_name, format_sp);
<rdar://problem/12632394> Add a format for FourCharCode This is now safe to do thanks to the "formats in categories" feature llvm-svn: 192233
2013-10-09 04:59:02 +08:00
}
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
static void
AddStringSummary(TypeCategoryImpl::SharedPointer category_sp,
const char* string,
ConstString type_name,
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
TypeSummaryImpl::Flags flags,
bool regex = false)
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
{
lldb::TypeSummaryImplSP summary_sp(new StringSummaryFormat(flags,
string));
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
if (regex)
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
category_sp->GetRegexTypeSummariesContainer()->Add(RegularExpressionSP(new RegularExpression(type_name.AsCString())),summary_sp);
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
else
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
category_sp->GetTypeSummariesContainer()->Add(type_name, summary_sp);
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
}
#ifndef LLDB_DISABLE_PYTHON
static void
AddCXXSummary (TypeCategoryImpl::SharedPointer category_sp,
CXXFunctionSummaryFormat::Callback funct,
const char* description,
ConstString type_name,
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
TypeSummaryImpl::Flags flags,
bool regex = false)
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
{
lldb::TypeSummaryImplSP summary_sp(new CXXFunctionSummaryFormat(flags,funct,description));
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
if (regex)
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
category_sp->GetRegexTypeSummariesContainer()->Add(RegularExpressionSP(new RegularExpression(type_name.AsCString())),summary_sp);
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
else
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
category_sp->GetTypeSummariesContainer()->Add(type_name, summary_sp);
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
}
#endif
#ifndef LLDB_DISABLE_PYTHON
static void AddCXXSynthetic (TypeCategoryImpl::SharedPointer category_sp,
CXXSyntheticChildren::CreateFrontEndCallback generator,
const char* description,
ConstString type_name,
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
ScriptedSyntheticChildren::Flags flags,
bool regex = false)
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
{
lldb::SyntheticChildrenSP synth_sp(new CXXSyntheticChildren(flags,description,generator));
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
if (regex)
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
category_sp->GetRegexTypeSyntheticsContainer()->Add(RegularExpressionSP(new RegularExpression(type_name.AsCString())), synth_sp);
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
else
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
category_sp->GetTypeSyntheticsContainer()->Add(type_name,synth_sp);
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
}
#endif
rdar://15648942 Provide a filter for libc++ std::atomic<T> This just hides some implementation clutter and promotes the actual content to only child status llvm-svn: 200984
2014-02-08 03:21:09 +08:00
#ifndef LLDB_DISABLE_PYTHON
static void AddFilter (TypeCategoryImpl::SharedPointer category_sp,
std::vector<std::string> children,
const char* description,
ConstString type_name,
ScriptedSyntheticChildren::Flags flags,
bool regex = false)
{
TypeFilterImplSP filter_sp(new TypeFilterImpl(flags));
for (auto child : children)
filter_sp->AddExpressionPath(child);
if (regex)
category_sp->GetRegexTypeFiltersContainer()->Add(RegularExpressionSP(new RegularExpression(type_name.AsCString())), filter_sp);
else
category_sp->GetTypeFiltersContainer()->Add(type_name,filter_sp);
}
#endif
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
void
<rdar://problem/12239827> Making a summary for std::wstring as provided by libstdc++ along with a relevant test case llvm-svn: 172286
2013-01-12 09:22:57 +08:00
FormatManager::LoadLibStdcppFormatters()
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
{
(a) adding formatters for: NSTimeZone and CFTimeZonRef SEL and related types CFGregorianDate llvm-svn: 151866
2012-03-02 08:55:53 +08:00
TypeSummaryImpl::Flags stl_summary_flags;
stl_summary_flags.SetCascades(true)
.SetSkipPointers(false)
.SetSkipReferences(false)
.SetDontShowChildren(true)
.SetDontShowValue(true)
.SetShowMembersOneLiner(false)
.SetHideItemNames(false);
lldb::TypeSummaryImplSP std_string_summary_sp(new StringSummaryFormat(stl_summary_flags,
<rdar://problem/10062621> New public API for handling formatters: creating, deleting, modifying categories, and formatters, and managing type/formatter association. This provides SB classes for each of the main object types involved in providing formatter support: SBTypeCategory SBTypeFilter SBTypeFormat SBTypeSummary SBTypeSynthetic plus, an SBTypeNameSpecifier class that is used on the public API layer to abstract the notion that formatters can be applied to plain type-names as well as to regular expressions For naming consistency, this patch also renames a lot of formatters-related classes. Plus, the changes in how flags are handled that started with summaries is now extended to other classes as well. A new enum (lldb::eTypeOption) is meant to support this on the public side. The patch also adds several new calls to the formatter infrastructure that are used to implement by-index accessing and several other design changes required to accommodate the new API layer. An architectural change is introduced in that backing objects for formatters now become writable. On the public API layer, CoW is implemented to prevent unwanted propagation of changes. Lastly, there are some modifications in how the "default" category is constructed and managed in relation to other categories. llvm-svn: 150558
2012-02-15 10:34:21 +08:00
"${var._M_dataplus._M_p}"));
Redesign of the interaction between Python and frozen objects: - introduced two new classes ValueObjectConstResultChild and ValueObjectConstResultImpl: the first one is a ValueObjectChild obtained from a ValueObjectConstResult, the second is a common implementation backend for VOCR and VOCRCh of method calls meant to read through pointers stored in frozen objects ; now such reads transparently move from host to target as required - as a consequence of the above, removed code that made target-memory copies of expression results in several places throughout LLDB, and also removed code that enabled to recognize an expression result VO as such - introduced a new GetPointeeData() method in ValueObject that lets you read a given amount of objects of type T from a VO representing a T* or T[], and doing dereferences transparently in private layer it returns a DataExtractor ; in public layer it returns an instance of a newly created lldb::SBData - as GetPointeeData() does the right thing for both frozen and non-frozen ValueObject's, reimplemented ReadPointedString() to use it en lieu of doing the raw read itself - introduced a new GetData() method in ValueObject that lets you get a copy of the data that backs the ValueObject (for pointers, this returns the address without any previous dereferencing steps ; for arrays it actually reads the whole chunk of memory) in public layer this returns an SBData, just like GetPointeeData() - introduced a new CreateValueFromData() method in SBValue that lets you create a new SBValue from a chunk of data wrapped in an SBData the limitation to remember for this kind of SBValue is that they have no address: extracting the address-of for these objects (with any of GetAddress(), GetLoadAddress() and AddressOf()) will return invalid values - added several tests to check that "p"-ing objects (STL classes, char* and char[]) will do the right thing Solved a bug where global pointers to global variables were not dereferenced correctly for display New target setting "max-string-summary-length" gives the maximum number of characters to show in a string when summarizing it, instead of the hardcoded 128 Solved a bug where the summary for char[] and char* would not be shown if the ValueObject's were dumped via the "p" command Removed m_pointers_point_to_load_addrs from ValueObject. Introduced a new m_address_type_of_children, which each ValueObject can set to tell the address type of any pointers and/or references it creates. In the current codebase, this is load address most of the time (the only notable exception being file addresses that generate file address children UNLESS we have a live process) Updated help text for summary-string Fixed an issue in STL formatters where std::stlcontainer::iterator would match the container's synthetic children providers Edited the syntax and help for some commands to have proper argument types llvm-svn: 139160
2011-09-07 03:20:51 +08:00
<rdar://problem/10062621> New public API for handling formatters: creating, deleting, modifying categories, and formatters, and managing type/formatter association. This provides SB classes for each of the main object types involved in providing formatter support: SBTypeCategory SBTypeFilter SBTypeFormat SBTypeSummary SBTypeSynthetic plus, an SBTypeNameSpecifier class that is used on the public API layer to abstract the notion that formatters can be applied to plain type-names as well as to regular expressions For naming consistency, this patch also renames a lot of formatters-related classes. Plus, the changes in how flags are handled that started with summaries is now extended to other classes as well. A new enum (lldb::eTypeOption) is meant to support this on the public side. The patch also adds several new calls to the formatter infrastructure that are used to implement by-index accessing and several other design changes required to accommodate the new API layer. An architectural change is introduced in that backing objects for formatters now become writable. On the public API layer, CoW is implemented to prevent unwanted propagation of changes. Lastly, there are some modifications in how the "default" category is constructed and managed in relation to other categories. llvm-svn: 150558
2012-02-15 10:34:21 +08:00
TypeCategoryImpl::SharedPointer gnu_category_sp = GetCategory(m_gnu_cpp_category_name);
Redesign of the interaction between Python and frozen objects: - introduced two new classes ValueObjectConstResultChild and ValueObjectConstResultImpl: the first one is a ValueObjectChild obtained from a ValueObjectConstResult, the second is a common implementation backend for VOCR and VOCRCh of method calls meant to read through pointers stored in frozen objects ; now such reads transparently move from host to target as required - as a consequence of the above, removed code that made target-memory copies of expression results in several places throughout LLDB, and also removed code that enabled to recognize an expression result VO as such - introduced a new GetPointeeData() method in ValueObject that lets you read a given amount of objects of type T from a VO representing a T* or T[], and doing dereferences transparently in private layer it returns a DataExtractor ; in public layer it returns an instance of a newly created lldb::SBData - as GetPointeeData() does the right thing for both frozen and non-frozen ValueObject's, reimplemented ReadPointedString() to use it en lieu of doing the raw read itself - introduced a new GetData() method in ValueObject that lets you get a copy of the data that backs the ValueObject (for pointers, this returns the address without any previous dereferencing steps ; for arrays it actually reads the whole chunk of memory) in public layer this returns an SBData, just like GetPointeeData() - introduced a new CreateValueFromData() method in SBValue that lets you create a new SBValue from a chunk of data wrapped in an SBData the limitation to remember for this kind of SBValue is that they have no address: extracting the address-of for these objects (with any of GetAddress(), GetLoadAddress() and AddressOf()) will return invalid values - added several tests to check that "p"-ing objects (STL classes, char* and char[]) will do the right thing Solved a bug where global pointers to global variables were not dereferenced correctly for display New target setting "max-string-summary-length" gives the maximum number of characters to show in a string when summarizing it, instead of the hardcoded 128 Solved a bug where the summary for char[] and char* would not be shown if the ValueObject's were dumped via the "p" command Removed m_pointers_point_to_load_addrs from ValueObject. Introduced a new m_address_type_of_children, which each ValueObject can set to tell the address type of any pointers and/or references it creates. In the current codebase, this is load address most of the time (the only notable exception being file addresses that generate file address children UNLESS we have a live process) Updated help text for summary-string Fixed an issue in STL formatters where std::stlcontainer::iterator would match the container's synthetic children providers Edited the syntax and help for some commands to have proper argument types llvm-svn: 139160
2011-09-07 03:20:51 +08:00
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::string"),
Redesign of the interaction between Python and frozen objects: - introduced two new classes ValueObjectConstResultChild and ValueObjectConstResultImpl: the first one is a ValueObjectChild obtained from a ValueObjectConstResult, the second is a common implementation backend for VOCR and VOCRCh of method calls meant to read through pointers stored in frozen objects ; now such reads transparently move from host to target as required - as a consequence of the above, removed code that made target-memory copies of expression results in several places throughout LLDB, and also removed code that enabled to recognize an expression result VO as such - introduced a new GetPointeeData() method in ValueObject that lets you read a given amount of objects of type T from a VO representing a T* or T[], and doing dereferences transparently in private layer it returns a DataExtractor ; in public layer it returns an instance of a newly created lldb::SBData - as GetPointeeData() does the right thing for both frozen and non-frozen ValueObject's, reimplemented ReadPointedString() to use it en lieu of doing the raw read itself - introduced a new GetData() method in ValueObject that lets you get a copy of the data that backs the ValueObject (for pointers, this returns the address without any previous dereferencing steps ; for arrays it actually reads the whole chunk of memory) in public layer this returns an SBData, just like GetPointeeData() - introduced a new CreateValueFromData() method in SBValue that lets you create a new SBValue from a chunk of data wrapped in an SBData the limitation to remember for this kind of SBValue is that they have no address: extracting the address-of for these objects (with any of GetAddress(), GetLoadAddress() and AddressOf()) will return invalid values - added several tests to check that "p"-ing objects (STL classes, char* and char[]) will do the right thing Solved a bug where global pointers to global variables were not dereferenced correctly for display New target setting "max-string-summary-length" gives the maximum number of characters to show in a string when summarizing it, instead of the hardcoded 128 Solved a bug where the summary for char[] and char* would not be shown if the ValueObject's were dumped via the "p" command Removed m_pointers_point_to_load_addrs from ValueObject. Introduced a new m_address_type_of_children, which each ValueObject can set to tell the address type of any pointers and/or references it creates. In the current codebase, this is load address most of the time (the only notable exception being file addresses that generate file address children UNLESS we have a live process) Updated help text for summary-string Fixed an issue in STL formatters where std::stlcontainer::iterator would match the container's synthetic children providers Edited the syntax and help for some commands to have proper argument types llvm-svn: 139160
2011-09-07 03:20:51 +08:00
std_string_summary_sp);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::basic_string<char>"),
Redesign of the interaction between Python and frozen objects: - introduced two new classes ValueObjectConstResultChild and ValueObjectConstResultImpl: the first one is a ValueObjectChild obtained from a ValueObjectConstResult, the second is a common implementation backend for VOCR and VOCRCh of method calls meant to read through pointers stored in frozen objects ; now such reads transparently move from host to target as required - as a consequence of the above, removed code that made target-memory copies of expression results in several places throughout LLDB, and also removed code that enabled to recognize an expression result VO as such - introduced a new GetPointeeData() method in ValueObject that lets you read a given amount of objects of type T from a VO representing a T* or T[], and doing dereferences transparently in private layer it returns a DataExtractor ; in public layer it returns an instance of a newly created lldb::SBData - as GetPointeeData() does the right thing for both frozen and non-frozen ValueObject's, reimplemented ReadPointedString() to use it en lieu of doing the raw read itself - introduced a new GetData() method in ValueObject that lets you get a copy of the data that backs the ValueObject (for pointers, this returns the address without any previous dereferencing steps ; for arrays it actually reads the whole chunk of memory) in public layer this returns an SBData, just like GetPointeeData() - introduced a new CreateValueFromData() method in SBValue that lets you create a new SBValue from a chunk of data wrapped in an SBData the limitation to remember for this kind of SBValue is that they have no address: extracting the address-of for these objects (with any of GetAddress(), GetLoadAddress() and AddressOf()) will return invalid values - added several tests to check that "p"-ing objects (STL classes, char* and char[]) will do the right thing Solved a bug where global pointers to global variables were not dereferenced correctly for display New target setting "max-string-summary-length" gives the maximum number of characters to show in a string when summarizing it, instead of the hardcoded 128 Solved a bug where the summary for char[] and char* would not be shown if the ValueObject's were dumped via the "p" command Removed m_pointers_point_to_load_addrs from ValueObject. Introduced a new m_address_type_of_children, which each ValueObject can set to tell the address type of any pointers and/or references it creates. In the current codebase, this is load address most of the time (the only notable exception being file addresses that generate file address children UNLESS we have a live process) Updated help text for summary-string Fixed an issue in STL formatters where std::stlcontainer::iterator would match the container's synthetic children providers Edited the syntax and help for some commands to have proper argument types llvm-svn: 139160
2011-09-07 03:20:51 +08:00
std_string_summary_sp);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::basic_string<char,std::char_traits<char>,std::allocator<char> >"),
Redesign of the interaction between Python and frozen objects: - introduced two new classes ValueObjectConstResultChild and ValueObjectConstResultImpl: the first one is a ValueObjectChild obtained from a ValueObjectConstResult, the second is a common implementation backend for VOCR and VOCRCh of method calls meant to read through pointers stored in frozen objects ; now such reads transparently move from host to target as required - as a consequence of the above, removed code that made target-memory copies of expression results in several places throughout LLDB, and also removed code that enabled to recognize an expression result VO as such - introduced a new GetPointeeData() method in ValueObject that lets you read a given amount of objects of type T from a VO representing a T* or T[], and doing dereferences transparently in private layer it returns a DataExtractor ; in public layer it returns an instance of a newly created lldb::SBData - as GetPointeeData() does the right thing for both frozen and non-frozen ValueObject's, reimplemented ReadPointedString() to use it en lieu of doing the raw read itself - introduced a new GetData() method in ValueObject that lets you get a copy of the data that backs the ValueObject (for pointers, this returns the address without any previous dereferencing steps ; for arrays it actually reads the whole chunk of memory) in public layer this returns an SBData, just like GetPointeeData() - introduced a new CreateValueFromData() method in SBValue that lets you create a new SBValue from a chunk of data wrapped in an SBData the limitation to remember for this kind of SBValue is that they have no address: extracting the address-of for these objects (with any of GetAddress(), GetLoadAddress() and AddressOf()) will return invalid values - added several tests to check that "p"-ing objects (STL classes, char* and char[]) will do the right thing Solved a bug where global pointers to global variables were not dereferenced correctly for display New target setting "max-string-summary-length" gives the maximum number of characters to show in a string when summarizing it, instead of the hardcoded 128 Solved a bug where the summary for char[] and char* would not be shown if the ValueObject's were dumped via the "p" command Removed m_pointers_point_to_load_addrs from ValueObject. Introduced a new m_address_type_of_children, which each ValueObject can set to tell the address type of any pointers and/or references it creates. In the current codebase, this is load address most of the time (the only notable exception being file addresses that generate file address children UNLESS we have a live process) Updated help text for summary-string Fixed an issue in STL formatters where std::stlcontainer::iterator would match the container's synthetic children providers Edited the syntax and help for some commands to have proper argument types llvm-svn: 139160
2011-09-07 03:20:51 +08:00
std_string_summary_sp);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::basic_string<char, std::char_traits<char>, std::allocator<char> >"),
Added a new --omit-names (-O, uppercase letter o) option to "type summary add". When used in conjunction with --inline-children, this option will cause the names of the values to be omitted from the output. This can be beneficial in cases such as vFloat, where it will compact the representation from ([0]=1,[1]=2,[2]=3,[3]=4) to (1, 2, 3, 4). Added a test case to check that the new option works correctly. Also took some time to revisit SummaryFormat and related classes and tweak them for added readability and maintainability. Finally, added a new class name to which the std::string summary should be applied. llvm-svn: 149644
2012-02-03 07:34:52 +08:00
std_string_summary_sp);
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
<rdar://problem/12239827> Making a summary for std::wstring as provided by libstdc++ along with a relevant test case llvm-svn: 172286
2013-01-12 09:22:57 +08:00
// making sure we force-pick the summary for printing wstring (_M_p is a wchar_t*)
lldb::TypeSummaryImplSP std_wstring_summary_sp(new StringSummaryFormat(stl_summary_flags,
"${var._M_dataplus._M_p%S}"));
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::wstring"),
<rdar://problem/12239827> Making a summary for std::wstring as provided by libstdc++ along with a relevant test case llvm-svn: 172286
2013-01-12 09:22:57 +08:00
std_wstring_summary_sp);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::basic_string<wchar_t>"),
<rdar://problem/12239827> Making a summary for std::wstring as provided by libstdc++ along with a relevant test case llvm-svn: 172286
2013-01-12 09:22:57 +08:00
std_wstring_summary_sp);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::basic_string<wchar_t,std::char_traits<wchar_t>,std::allocator<wchar_t> >"),
<rdar://problem/12239827> Making a summary for std::wstring as provided by libstdc++ along with a relevant test case llvm-svn: 172286
2013-01-12 09:22:57 +08:00
std_wstring_summary_sp);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::basic_string<wchar_t, std::char_traits<wchar_t>, std::allocator<wchar_t> >"),
<rdar://problem/12239827> Making a summary for std::wstring as provided by libstdc++ along with a relevant test case llvm-svn: 172286
2013-01-12 09:22:57 +08:00
std_wstring_summary_sp);
Redesign of the interaction between Python and frozen objects: - introduced two new classes ValueObjectConstResultChild and ValueObjectConstResultImpl: the first one is a ValueObjectChild obtained from a ValueObjectConstResult, the second is a common implementation backend for VOCR and VOCRCh of method calls meant to read through pointers stored in frozen objects ; now such reads transparently move from host to target as required - as a consequence of the above, removed code that made target-memory copies of expression results in several places throughout LLDB, and also removed code that enabled to recognize an expression result VO as such - introduced a new GetPointeeData() method in ValueObject that lets you read a given amount of objects of type T from a VO representing a T* or T[], and doing dereferences transparently in private layer it returns a DataExtractor ; in public layer it returns an instance of a newly created lldb::SBData - as GetPointeeData() does the right thing for both frozen and non-frozen ValueObject's, reimplemented ReadPointedString() to use it en lieu of doing the raw read itself - introduced a new GetData() method in ValueObject that lets you get a copy of the data that backs the ValueObject (for pointers, this returns the address without any previous dereferencing steps ; for arrays it actually reads the whole chunk of memory) in public layer this returns an SBData, just like GetPointeeData() - introduced a new CreateValueFromData() method in SBValue that lets you create a new SBValue from a chunk of data wrapped in an SBData the limitation to remember for this kind of SBValue is that they have no address: extracting the address-of for these objects (with any of GetAddress(), GetLoadAddress() and AddressOf()) will return invalid values - added several tests to check that "p"-ing objects (STL classes, char* and char[]) will do the right thing Solved a bug where global pointers to global variables were not dereferenced correctly for display New target setting "max-string-summary-length" gives the maximum number of characters to show in a string when summarizing it, instead of the hardcoded 128 Solved a bug where the summary for char[] and char* would not be shown if the ValueObject's were dumped via the "p" command Removed m_pointers_point_to_load_addrs from ValueObject. Introduced a new m_address_type_of_children, which each ValueObject can set to tell the address type of any pointers and/or references it creates. In the current codebase, this is load address most of the time (the only notable exception being file addresses that generate file address children UNLESS we have a live process) Updated help text for summary-string Fixed an issue in STL formatters where std::stlcontainer::iterator would match the container's synthetic children providers Edited the syntax and help for some commands to have proper argument types llvm-svn: 139160
2011-09-07 03:20:51 +08:00
Fixed the Xcode project building of LLVM to be a bit more user friendly: - If you download and build the sources in the Xcode project, x86_64 builds by default using the "llvm.zip" checkpointed LLVM. - If you delete the "lldb/llvm.zip" and the "lldb/llvm" folder, and build the Xcode project will download the right LLVM sources and build them from scratch - If you have a "lldb/llvm" folder already that contains a "lldb/llvm/lib" directory, we will use the sources you have placed in the LLDB directory. Python can now be disabled for platforms that don't support it. Changed the way the libllvmclang.a files get used. They now all get built into arch specific directories and never get merged into universal binaries as this was causing issues where you would have to go and delete the file if you wanted to build an extra architecture slice. llvm-svn: 143678
2011-11-04 11:34:56 +08:00
#ifndef LLDB_DISABLE_PYTHON
<rdar://problem/10062621> New public API for handling formatters: creating, deleting, modifying categories, and formatters, and managing type/formatter association. This provides SB classes for each of the main object types involved in providing formatter support: SBTypeCategory SBTypeFilter SBTypeFormat SBTypeSummary SBTypeSynthetic plus, an SBTypeNameSpecifier class that is used on the public API layer to abstract the notion that formatters can be applied to plain type-names as well as to regular expressions For naming consistency, this patch also renames a lot of formatters-related classes. Plus, the changes in how flags are handled that started with summaries is now extended to other classes as well. A new enum (lldb::eTypeOption) is meant to support this on the public side. The patch also adds several new calls to the formatter infrastructure that are used to implement by-index accessing and several other design changes required to accommodate the new API layer. An architectural change is introduced in that backing objects for formatters now become writable. On the public API layer, CoW is implemented to prevent unwanted propagation of changes. Lastly, there are some modifications in how the "default" category is constructed and managed in relation to other categories. llvm-svn: 150558
2012-02-15 10:34:21 +08:00
SyntheticChildren::Flags stl_synth_flags;
stl_synth_flags.SetCascades(true).SetSkipPointers(false).SetSkipReferences(false);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetRegexTypeSyntheticsContainer()->Add(RegularExpressionSP(new RegularExpression("^std::vector<.+>(( )?&)?$")),
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
SyntheticChildrenSP(new ScriptedSyntheticChildren(stl_synth_flags,
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
"lldb.formatters.cpp.gnu_libstdcpp.StdVectorSynthProvider")));
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetRegexTypeSyntheticsContainer()->Add(RegularExpressionSP(new RegularExpression("^std::map<.+> >(( )?&)?$")),
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
SyntheticChildrenSP(new ScriptedSyntheticChildren(stl_synth_flags,
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
"lldb.formatters.cpp.gnu_libstdcpp.StdMapSynthProvider")));
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetRegexTypeSyntheticsContainer()->Add(RegularExpressionSP(new RegularExpression("^std::list<.+>(( )?&)?$")),
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
SyntheticChildrenSP(new ScriptedSyntheticChildren(stl_synth_flags,
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
"lldb.formatters.cpp.gnu_libstdcpp.StdListSynthProvider")));
(a) adding formatters for: NSTimeZone and CFTimeZonRef SEL and related types CFGregorianDate llvm-svn: 151866
2012-03-02 08:55:53 +08:00
See: http://stackoverflow.com/questions/16166897/c-how-to-obtain-address-of-stdvector-object-in-lldb-xcode The user was trying to obtain the address-of an std::vector and the experience was more painful than necessary because data formatters were kicking in for vector* objects We got this right for libc++ - we should get it right for libstdc++ too llvm-svn: 180219
2013-04-25 05:52:24 +08:00
stl_summary_flags.SetDontShowChildren(false);stl_summary_flags.SetSkipPointers(true);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetRegexTypeSummariesContainer()->Add(RegularExpressionSP(new RegularExpression("^std::vector<.+>(( )?&)?$")),
Synthetic values are now automatically enabled and active by default. SBValue is set up to always wrap a synthetic value when one is available. A new setting enable-synthetic-value is provided on the target to disable this behavior. There also is a new GetNonSyntheticValue() API call on SBValue to go back from synthetic to non-synthetic. There is no call to go from non-synthetic to synthetic. The test suite has been changed accordingly. Fallout from changes to type searching: an hack has to be played to make it possible to use maps that contain std::string due to the special name replacement operated by clang Fixing a test case that was using libstdcpp instead of libc++ - caught as a consequence of said changes to type searching llvm-svn: 153495
2012-03-27 10:35:13 +08:00
TypeSummaryImplSP(new StringSummaryFormat(stl_summary_flags,
"size=${svar%#}")));
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetRegexTypeSummariesContainer()->Add(RegularExpressionSP(new RegularExpression("^std::map<.+> >(( )?&)?$")),
Synthetic values are now automatically enabled and active by default. SBValue is set up to always wrap a synthetic value when one is available. A new setting enable-synthetic-value is provided on the target to disable this behavior. There also is a new GetNonSyntheticValue() API call on SBValue to go back from synthetic to non-synthetic. There is no call to go from non-synthetic to synthetic. The test suite has been changed accordingly. Fallout from changes to type searching: an hack has to be played to make it possible to use maps that contain std::string due to the special name replacement operated by clang Fixing a test case that was using libstdcpp instead of libc++ - caught as a consequence of said changes to type searching llvm-svn: 153495
2012-03-27 10:35:13 +08:00
TypeSummaryImplSP(new StringSummaryFormat(stl_summary_flags,
"size=${svar%#}")));
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetRegexTypeSummariesContainer()->Add(RegularExpressionSP(new RegularExpression("^std::list<.+>(( )?&)?$")),
(a) adding formatters for: NSTimeZone and CFTimeZonRef SEL and related types CFGregorianDate llvm-svn: 151866
2012-03-02 08:55:53 +08:00
TypeSummaryImplSP(new StringSummaryFormat(stl_summary_flags,
"size=${svar%#}")));
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
AddCXXSynthetic(gnu_category_sp, lldb_private::formatters::LibStdcppVectorIteratorSyntheticFrontEndCreator, "std::vector iterator synthetic children", ConstString("^__gnu_cxx::__normal_iterator<.+>$"), stl_synth_flags, true);
AddCXXSynthetic(gnu_category_sp, lldb_private::formatters::LibstdcppMapIteratorSyntheticFrontEndCreator, "std::map iterator synthetic children", ConstString("^std::_Rb_tree_iterator<.+>$"), stl_synth_flags, true);
<rdar://problem/12953018> Synthetic children and summary for std::vector<bool> (for both libcxx and libstdcpp). std::vector<bool> is a special case and is custom-implemented to be a vector of bits, which means we failed to handle it with the standard std::vector<T> formatter. This checkin provides custom formatters that work correctly llvm-svn: 174333
2013-02-05 06:54:42 +08:00
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::vector<std::allocator<bool> >"),
<rdar://problem/12953018> Synthetic children and summary for std::vector<bool> (for both libcxx and libstdcpp). std::vector<bool> is a special case and is custom-implemented to be a vector of bits, which means we failed to handle it with the standard std::vector<T> formatter. This checkin provides custom formatters that work correctly llvm-svn: 174333
2013-02-05 06:54:42 +08:00
TypeSummaryImplSP(new StringSummaryFormat(stl_summary_flags, "size=${svar%#}")));
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
gnu_category_sp->GetTypeSyntheticsContainer()->Add(ConstString("std::vector<std::allocator<bool> >"),
<rdar://problem/12953018> Synthetic children and summary for std::vector<bool> (for both libcxx and libstdcpp). std::vector<bool> is a special case and is custom-implemented to be a vector of bits, which means we failed to handle it with the standard std::vector<T> formatter. This checkin provides custom formatters that work correctly llvm-svn: 174333
2013-02-05 06:54:42 +08:00
SyntheticChildrenSP(new CXXSyntheticChildren(stl_synth_flags,"libc++ std::vector<bool> synthetic children",lldb_private::formatters::LibstdcppVectorBoolSyntheticFrontEndCreator)));
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
#endif
}
Added formatters for libc++ (http://libcxx.llvm.org): std::string has a summary provider std::vector std::list and std::map have both a summary and a synthetic children provider Given the usage of a custom namespace (std::__1::classname) for the implementation of libc++, we keep both libstdcpp and libc++ formatters enabled at the same time since that raises no conflicts and enabled for seamless transition between the two The formatters for libc++ reside in a libcxx category, and are loaded from libcxx.py (to be found in examples/synthetic) The formatters-stl test cases have been divided to be separate for libcxx and libstdcpp. This separation is necessary because (a) we need different compiler flags for libc++ than for libstdcpp (b) libc++ inlines a lot more than libstdcpp and some code changes were required to accommodate this difference llvm-svn: 152570
2012-03-13 03:47:17 +08:00
void
FormatManager::LoadLibcxxFormatters()
{
TypeSummaryImpl::Flags stl_summary_flags;
stl_summary_flags.SetCascades(true)
.SetSkipPointers(false)
.SetSkipReferences(false)
.SetDontShowChildren(true)
.SetDontShowValue(true)
.SetShowMembersOneLiner(false)
.SetHideItemNames(false);
Mark newly added synthesizer calls as being #ifndef LLDB_DISABLE_PYTHON llvm-svn: 152599
2012-03-13 09:19:42 +08:00
#ifndef LLDB_DISABLE_PYTHON
<rdar://problem/12239827> Providing a data formatter for libc++ std::wstring In the process, refactoring the std::string data formatter to be written in C++ so that commonalities between the two can be exploited Also, providing a new API on the ValueObject to navigate a hierarchy by index-path Lastly, an appropriate test case is included llvm-svn: 172282
2013-01-12 09:00:22 +08:00
//std::string code(" lldb.formatters.cpp.libcxx.stdstring_SummaryProvider(valobj,internal_dict)");
//lldb::TypeSummaryImplSP std_string_summary_sp(new ScriptSummaryFormat(stl_summary_flags, "lldb.formatters.cpp.libcxx.stdstring_SummaryProvider",code.c_str()));
Added formatters for libc++ (http://libcxx.llvm.org): std::string has a summary provider std::vector std::list and std::map have both a summary and a synthetic children provider Given the usage of a custom namespace (std::__1::classname) for the implementation of libc++, we keep both libstdcpp and libc++ formatters enabled at the same time since that raises no conflicts and enabled for seamless transition between the two The formatters for libc++ reside in a libcxx category, and are loaded from libcxx.py (to be found in examples/synthetic) The formatters-stl test cases have been divided to be separate for libcxx and libstdcpp. This separation is necessary because (a) we need different compiler flags for libc++ than for libstdcpp (b) libc++ inlines a lot more than libstdcpp and some code changes were required to accommodate this difference llvm-svn: 152570
2012-03-13 03:47:17 +08:00
<rdar://problem/12239827> Providing a data formatter for libc++ std::wstring In the process, refactoring the std::string data formatter to be written in C++ so that commonalities between the two can be exploited Also, providing a new API on the ValueObject to navigate a hierarchy by index-path Lastly, an appropriate test case is included llvm-svn: 172282
2013-01-12 09:00:22 +08:00
lldb::TypeSummaryImplSP std_string_summary_sp(new CXXFunctionSummaryFormat(stl_summary_flags, lldb_private::formatters::LibcxxStringSummaryProvider, "std::string summary provider"));
lldb::TypeSummaryImplSP std_wstring_summary_sp(new CXXFunctionSummaryFormat(stl_summary_flags, lldb_private::formatters::LibcxxWStringSummaryProvider, "std::wstring summary provider"));
Added formatters for libc++ (http://libcxx.llvm.org): std::string has a summary provider std::vector std::list and std::map have both a summary and a synthetic children provider Given the usage of a custom namespace (std::__1::classname) for the implementation of libc++, we keep both libstdcpp and libc++ formatters enabled at the same time since that raises no conflicts and enabled for seamless transition between the two The formatters for libc++ reside in a libcxx category, and are loaded from libcxx.py (to be found in examples/synthetic) The formatters-stl test cases have been divided to be separate for libcxx and libstdcpp. This separation is necessary because (a) we need different compiler flags for libc++ than for libstdcpp (b) libc++ inlines a lot more than libstdcpp and some code changes were required to accommodate this difference llvm-svn: 152570
2012-03-13 03:47:17 +08:00
TypeCategoryImpl::SharedPointer libcxx_category_sp = GetCategory(m_libcxx_category_name);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
libcxx_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::__1::string"),
Added formatters for libc++ (http://libcxx.llvm.org): std::string has a summary provider std::vector std::list and std::map have both a summary and a synthetic children provider Given the usage of a custom namespace (std::__1::classname) for the implementation of libc++, we keep both libstdcpp and libc++ formatters enabled at the same time since that raises no conflicts and enabled for seamless transition between the two The formatters for libc++ reside in a libcxx category, and are loaded from libcxx.py (to be found in examples/synthetic) The formatters-stl test cases have been divided to be separate for libcxx and libstdcpp. This separation is necessary because (a) we need different compiler flags for libc++ than for libstdcpp (b) libc++ inlines a lot more than libstdcpp and some code changes were required to accommodate this difference llvm-svn: 152570
2012-03-13 03:47:17 +08:00
std_string_summary_sp);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
libcxx_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >"),
Added formatters for libc++ (http://libcxx.llvm.org): std::string has a summary provider std::vector std::list and std::map have both a summary and a synthetic children provider Given the usage of a custom namespace (std::__1::classname) for the implementation of libc++, we keep both libstdcpp and libc++ formatters enabled at the same time since that raises no conflicts and enabled for seamless transition between the two The formatters for libc++ reside in a libcxx category, and are loaded from libcxx.py (to be found in examples/synthetic) The formatters-stl test cases have been divided to be separate for libcxx and libstdcpp. This separation is necessary because (a) we need different compiler flags for libc++ than for libstdcpp (b) libc++ inlines a lot more than libstdcpp and some code changes were required to accommodate this difference llvm-svn: 152570
2012-03-13 03:47:17 +08:00
std_string_summary_sp);
Mark newly added synthesizer calls as being #ifndef LLDB_DISABLE_PYTHON llvm-svn: 152599
2012-03-13 09:19:42 +08:00
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
libcxx_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::__1::wstring"),
<rdar://problem/12239827> Providing a data formatter for libc++ std::wstring In the process, refactoring the std::string data formatter to be written in C++ so that commonalities between the two can be exploited Also, providing a new API on the ValueObject to navigate a hierarchy by index-path Lastly, an appropriate test case is included llvm-svn: 172282
2013-01-12 09:00:22 +08:00
std_wstring_summary_sp);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
libcxx_category_sp->GetTypeSummariesContainer()->Add(ConstString("std::__1::basic_string<wchar_t, std::__1::char_traits<wchar_t>, std::__1::allocator<wchar_t> >"),
<rdar://problem/12239827> Providing a data formatter for libc++ std::wstring In the process, refactoring the std::string data formatter to be written in C++ so that commonalities between the two can be exploited Also, providing a new API on the ValueObject to navigate a hierarchy by index-path Lastly, an appropriate test case is included llvm-svn: 172282
2013-01-12 09:00:22 +08:00
std_wstring_summary_sp);
Added formatters for libc++ (http://libcxx.llvm.org): std::string has a summary provider std::vector std::list and std::map have both a summary and a synthetic children provider Given the usage of a custom namespace (std::__1::classname) for the implementation of libc++, we keep both libstdcpp and libc++ formatters enabled at the same time since that raises no conflicts and enabled for seamless transition between the two The formatters for libc++ reside in a libcxx category, and are loaded from libcxx.py (to be found in examples/synthetic) The formatters-stl test cases have been divided to be separate for libcxx and libstdcpp. This separation is necessary because (a) we need different compiler flags for libc++ than for libstdcpp (b) libc++ inlines a lot more than libstdcpp and some code changes were required to accommodate this difference llvm-svn: 152570
2012-03-13 03:47:17 +08:00
SyntheticChildren::Flags stl_synth_flags;
stl_synth_flags.SetCascades(true).SetSkipPointers(false).SetSkipReferences(false);
The formatters for std::shared_ptr, std::weak_ptr, std::list, std::vector and std::map as provided by libc++ are now written in C++ instead of Python std::deque is still in Python but is much less commonly used llvm-svn: 177454
2013-03-20 06:58:48 +08:00
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxStdVectorSyntheticFrontEndCreator, "libc++ std::vector synthetic children", ConstString("^std::__1::vector<.+>(( )?&)?$"), stl_synth_flags, true);
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxStdListSyntheticFrontEndCreator, "libc++ std::list synthetic children", ConstString("^std::__1::list<.+>(( )?&)?$"), stl_synth_flags, true);
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxStdMapSyntheticFrontEndCreator, "libc++ std::map synthetic children", ConstString("^std::__1::map<.+> >(( )?&)?$"), stl_synth_flags, true);
<rdar://problem/11669154> Make a summary format for libc++ STL containers that shows the number of items as before, but also shows the pointer value for pointer-to-container llvm-svn: 181236
2013-05-07 02:55:52 +08:00
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxVectorBoolSyntheticFrontEndCreator, "libc++ std::vector<bool> synthetic children", ConstString("std::__1::vector<std::__1::allocator<bool> >"), stl_synth_flags);
Enable the data formatter for std::vector<bool> on libc++ again. In recent clang builds, we are vended a different typename, which the formatter needs to match against. llvm-svn: 215801
2014-08-16 09:02:36 +08:00
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxVectorBoolSyntheticFrontEndCreator, "libc++ std::vector<bool> synthetic children", ConstString("std::__1::vector<bool, std::__1::allocator<bool> >"), stl_synth_flags);
<rdar://problem/13125225> Adding data formatters for std::set, std::multiset and std::multimap for libc++ The underlying data structure is the same as std::map, so this change is very minimal and mostly consists of test cases llvm-svn: 183323
2013-06-06 01:47:23 +08:00
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxStdMapSyntheticFrontEndCreator, "libc++ std::set synthetic children", ConstString("^std::__1::set<.+> >(( )?&)?$"), stl_synth_flags, true);
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxStdMapSyntheticFrontEndCreator, "libc++ std::multiset synthetic children", ConstString("^std::__1::multiset<.+> >(( )?&)?$"), stl_synth_flags, true);
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxStdMapSyntheticFrontEndCreator, "libc++ std::multimap synthetic children", ConstString("^std::__1::multimap<.+> >(( )?&)?$"), stl_synth_flags, true);
<rdar://problem/14071463> SVN r189964 provided a sample Python script to inspect unordered(multi){set|map} with synthetic children, contribued by Jared Grubb This checkin converts that sample script to a C++ provider built into LLDB A test case is also provided llvm-svn: 190564
2013-09-12 08:48:47 +08:00
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxStdUnorderedMapSyntheticFrontEndCreator, "libc++ std::unordered containers synthetic children", ConstString("^(std::__1::)unordered_(multi)?(map|set)<.+> >$"), stl_synth_flags, true);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
libcxx_category_sp->GetRegexTypeSyntheticsContainer()->Add(RegularExpressionSP(new RegularExpression("^(std::__1::)deque<.+>(( )?&)?$")),
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
SyntheticChildrenSP(new ScriptedSyntheticChildren(stl_synth_flags,
Data formatters for libc++ deque and shared/weak ptrs - a contribution by Jared Grubb llvm-svn: 162680
2012-08-28 01:42:50 +08:00
"lldb.formatters.cpp.libcxx.stddeque_SynthProvider")));
The formatters for std::shared_ptr, std::weak_ptr, std::list, std::vector and std::map as provided by libc++ are now written in C++ instead of Python std::deque is still in Python but is much less commonly used llvm-svn: 177454
2013-03-20 06:58:48 +08:00
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxSharedPtrSyntheticFrontEndCreator, "shared_ptr synthetic children", ConstString("^(std::__1::)shared_ptr<.+>(( )?&)?$"), stl_synth_flags, true);
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxSharedPtrSyntheticFrontEndCreator, "weak_ptr synthetic children", ConstString("^(std::__1::)weak_ptr<.+>(( )?&)?$"), stl_synth_flags, true);
Data formatters for libc++ deque and shared/weak ptrs - a contribution by Jared Grubb llvm-svn: 162680
2012-08-28 01:42:50 +08:00
<rdar://problem/11669154> Make a summary format for libc++ STL containers that shows the number of items as before, but also shows the pointer value for pointer-to-container llvm-svn: 181236
2013-05-07 02:55:52 +08:00
stl_summary_flags.SetDontShowChildren(false);stl_summary_flags.SetSkipPointers(false);
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibcxxVectorBoolSyntheticFrontEndCreator, "libc++ std::vector<bool> synthetic children", ConstString("std::__1::vector<bool, std::__1::allocator<bool> >"), stl_synth_flags);
<rdar://problem/11669154> Make a summary format for libc++ STL containers that shows the number of items as before, but also shows the pointer value for pointer-to-container llvm-svn: 181236
2013-05-07 02:55:52 +08:00
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::vector summary provider", ConstString("^std::__1::vector<.+>(( )?&)?$"), stl_summary_flags, true);
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::list summary provider", ConstString("^std::__1::list<.+>(( )?&)?$"), stl_summary_flags, true);
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::map summary provider", ConstString("^std::__1::map<.+>(( )?&)?$"), stl_summary_flags, true);
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::deque summary provider", ConstString("^std::__1::deque<.+>(( )?&)?$"), stl_summary_flags, true);
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::vector<bool> summary provider", ConstString("std::__1::vector<std::__1::allocator<bool> >"), stl_summary_flags);
Enable the data formatter for std::vector<bool> on libc++ again. In recent clang builds, we are vended a different typename, which the formatter needs to match against. llvm-svn: 215801
2014-08-16 09:02:36 +08:00
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::vector<bool> summary provider", ConstString("std::__1::vector<bool, std::__1::allocator<bool> >"), stl_summary_flags);
<rdar://problem/13125225> Adding data formatters for std::set, std::multiset and std::multimap for libc++ The underlying data structure is the same as std::map, so this change is very minimal and mostly consists of test cases llvm-svn: 183323
2013-06-06 01:47:23 +08:00
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::set summary provider", ConstString("^std::__1::set<.+>(( )?&)?$"), stl_summary_flags, true);
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::multiset summary provider", ConstString("^std::__1::multiset<.+>(( )?&)?$"), stl_summary_flags, true);
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::multimap summary provider", ConstString("^std::__1::multimap<.+>(( )?&)?$"), stl_summary_flags, true);
<rdar://problem/14071463> SVN r189964 provided a sample Python script to inspect unordered(multi){set|map} with synthetic children, contribued by Jared Grubb This checkin converts that sample script to a C++ provider built into LLDB A test case is also provided llvm-svn: 190564
2013-09-12 08:48:47 +08:00
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::unordered containers summary provider", ConstString("^(std::__1::)unordered_(multi)?(map|set)<.+> >$"), stl_summary_flags, true);
<rdar://problem/11669154> Make a summary format for libc++ STL containers that shows the number of items as before, but also shows the pointer value for pointer-to-container llvm-svn: 181236
2013-05-07 02:55:52 +08:00
stl_summary_flags.SetSkipPointers(true);
New and improved data formatter for std::shared_ptr<> and std::weak_ptr<> llvm-svn: 198724
2014-01-08 09:36:59 +08:00
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxSmartPointerSummaryProvider, "libc++ std::shared_ptr summary provider", ConstString("^std::__1::shared_ptr<.+>(( )?&)?$"), stl_summary_flags, true);
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxSmartPointerSummaryProvider, "libc++ std::weak_ptr summary provider", ConstString("^std::__1::weak_ptr<.+>(( )?&)?$"), stl_summary_flags, true);
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibCxxVectorIteratorSyntheticFrontEndCreator, "std::vector iterator synthetic children", ConstString("^std::__1::__wrap_iter<.+>$"), stl_synth_flags, true);
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
AddCXXSummary(libcxx_category_sp, lldb_private::formatters::LibcxxContainerSummaryProvider, "libc++ std::vector<bool> summary provider", ConstString("std::__1::vector<bool, std::__1::allocator<bool> >"), stl_summary_flags);
<rdar://problem/4529976> Adding data formatters for iterators for std::map and std::vector (both libc++ and libstdcpp) This does not include reverse iterators since they are both trickier (due to requirements the standard imposes on them) and much less useful llvm-svn: 175787
2013-02-22 03:57:10 +08:00
AddCXXSynthetic(libcxx_category_sp, lldb_private::formatters::LibCxxMapIteratorSyntheticFrontEndCreator, "std::map iterator synthetic children", ConstString("^std::__1::__map_iterator<.+>$"), stl_synth_flags, true);
Adding a custom summary for libc++ std::vector<bool> vector<bool> is specialized and the existing general summary for vectors would lie to the user. Tackling libstdc++ and synthetic children is the following, less critical, part of this task llvm-svn: 172671
2013-01-17 07:17:30 +08:00
rdar://15648942 Provide a filter for libc++ std::atomic<T> This just hides some implementation clutter and promotes the actual content to only child status llvm-svn: 200984
2014-02-08 03:21:09 +08:00
AddFilter(libcxx_category_sp, {"__a_"}, "libc++ std::atomic filter", ConstString("^std::__1::atomic<.*>$"), stl_synth_flags, true);
Added formatters for libc++ (http://libcxx.llvm.org): std::string has a summary provider std::vector std::list and std::map have both a summary and a synthetic children provider Given the usage of a custom namespace (std::__1::classname) for the implementation of libc++, we keep both libstdcpp and libc++ formatters enabled at the same time since that raises no conflicts and enabled for seamless transition between the two The formatters for libc++ reside in a libcxx category, and are loaded from libcxx.py (to be found in examples/synthetic) The formatters-stl test cases have been divided to be separate for libcxx and libstdcpp. This separation is necessary because (a) we need different compiler flags for libc++ than for libstdcpp (b) libc++ inlines a lot more than libstdcpp and some code changes were required to accommodate this difference llvm-svn: 152570
2012-03-13 03:47:17 +08:00
#endif
}
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
void
FormatManager::LoadSystemFormatters()
{
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
TypeSummaryImpl::Flags string_flags;
The summary for const char* was not cascading. This was preventing us from providing a summary for the result of std::string.c_str() with libc++ llvm-svn: 175841
2013-02-22 08:37:31 +08:00
string_flags.SetCascades(true)
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
.SetSkipPointers(true)
.SetSkipReferences(false)
.SetDontShowChildren(true)
.SetDontShowValue(false)
.SetShowMembersOneLiner(false)
.SetHideItemNames(false);
lldb::TypeSummaryImplSP string_format(new StringSummaryFormat(string_flags, "${var%s}"));
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
lldb::TypeSummaryImplSP string_array_format(new StringSummaryFormat(TypeSummaryImpl::Flags().SetCascades(false)
.SetSkipPointers(true)
.SetSkipReferences(false)
<rdar://problem/12410225> By default, omit the children for a char[] and just show the string contents Can be overridden by appropriate command-line flags llvm-svn: 178292
2013-03-29 05:36:58 +08:00
.SetDontShowChildren(true)
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
.SetDontShowValue(true)
.SetShowMembersOneLiner(false)
.SetHideItemNames(false),
"${var%s}"));
lldb::RegularExpressionSP any_size_char_arr(new RegularExpression("char \\[[0-9]+\\]"));
TypeCategoryImpl::SharedPointer sys_category_sp = GetCategory(m_system_category_name);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
sys_category_sp->GetTypeSummariesContainer()->Add(ConstString("char *"), string_format);
sys_category_sp->GetTypeSummariesContainer()->Add(ConstString("unsigned char *"), string_format);
sys_category_sp->GetRegexTypeSummariesContainer()->Add(any_size_char_arr, string_array_format);
Get rid of the FooStructSynth, it was a testing thing I put in and forgot to remove Hopefully nobody had a struct Foo in their app:-) llvm-svn: 193092
2013-10-22 01:29:51 +08:00
<rdar://problem/11628688> Making sure to use the ostype format for OSType values llvm-svn: 160180
2012-07-14 02:54:40 +08:00
lldb::TypeSummaryImplSP ostype_summary(new StringSummaryFormat(TypeSummaryImpl::Flags().SetCascades(false)
.SetSkipPointers(true)
.SetSkipReferences(true)
.SetDontShowChildren(true)
.SetDontShowValue(false)
.SetShowMembersOneLiner(false)
.SetHideItemNames(false),
"${var%O}"));
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
sys_category_sp->GetTypeSummariesContainer()->Add(ConstString("OSType"), ostype_summary);
This patch provides a set of formatters for most of the commonly used Cocoa classes. The formatter for NSString is an improved version of the one previously shipped as an example, the others are new in design and implementation. A more robust and OO-compliant Objective-C runtime wrapper is provided for runtime versions 1 and 2 on 32 and 64 bit. The formatters are contained in a category named "AppKit", which is not enabled at startup. llvm-svn: 151299
2012-02-24 07:10:03 +08:00
Add ifdef LLDB_DISABLE_PYTHON around newly added use of AddCXXSummary in FormatManager::LoadSystemFormatters(); that function pulls in code that assumes python; can't be used without the ifdef. llvm-svn: 172300
2013-01-12 12:24:50 +08:00
#ifndef LLDB_DISABLE_PYTHON
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
// FIXME because of a bug in the FormattersContainer we need to add a summary for both X* and const X* (<rdar://problem/12717717>)
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
AddCXXSummary(sys_category_sp, lldb_private::formatters::Char16StringSummaryProvider, "char16_t * summary provider", ConstString("char16_t *"), string_flags);
This patch provides a set of formatters for most of the commonly used Cocoa classes. The formatter for NSString is an improved version of the one previously shipped as an example, the others are new in design and implementation. A more robust and OO-compliant Objective-C runtime wrapper is provided for runtime versions 1 and 2 on 32 and 64 bit. The formatters are contained in a category named "AppKit", which is not enabled at startup. llvm-svn: 151299
2012-02-24 07:10:03 +08:00
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
AddCXXSummary(sys_category_sp, lldb_private::formatters::Char32StringSummaryProvider, "char32_t * summary provider", ConstString("char32_t *"), string_flags);
<rdar://problem/11383764> Making a data formatter for wchar_t * llvm-svn: 172165
2013-01-11 10:44:00 +08:00
AddCXXSummary(sys_category_sp, lldb_private::formatters::WCharStringSummaryProvider, "wchar_t * summary provider", ConstString("wchar_t *"), string_flags);
<rdar://problem/12790664> Single-character Unicode data formatters llvm-svn: 172492
2013-01-15 07:53:26 +08:00
AddCXXSummary(sys_category_sp, lldb_private::formatters::Char16StringSummaryProvider, "unichar * summary provider", ConstString("unichar *"), string_flags);
TypeSummaryImpl::Flags widechar_flags;
widechar_flags.SetDontShowValue(true)
.SetSkipPointers(true)
.SetSkipReferences(false)
.SetCascades(true)
.SetDontShowChildren(true)
.SetHideItemNames(true)
.SetShowMembersOneLiner(false);
AddCXXSummary(sys_category_sp, lldb_private::formatters::Char16SummaryProvider, "char16_t summary provider", ConstString("char16_t"), widechar_flags);
AddCXXSummary(sys_category_sp, lldb_private::formatters::Char32SummaryProvider, "char32_t summary provider", ConstString("char32_t"), widechar_flags);
AddCXXSummary(sys_category_sp, lldb_private::formatters::WCharSummaryProvider, "wchar_t summary provider", ConstString("wchar_t"), widechar_flags);
AddCXXSummary(sys_category_sp, lldb_private::formatters::Char16SummaryProvider, "unichar summary provider", ConstString("unichar"), widechar_flags);
<rdar://problem/12632394> Add a format for FourCharCode This is now safe to do thanks to the "formats in categories" feature llvm-svn: 192233
2013-10-09 04:59:02 +08:00
TypeFormatImpl::Flags fourchar_flags;
fourchar_flags.SetCascades(true).SetSkipPointers(true).SetSkipReferences(true);
AddFormat(sys_category_sp, lldb::eFormatOSType, ConstString("FourCharCode"), fourchar_flags);
Add ifdef LLDB_DISABLE_PYTHON around newly added use of AddCXXSummary in FormatManager::LoadSystemFormatters(); that function pulls in code that assumes python; can't be used without the ifdef. llvm-svn: 172300
2013-01-12 12:24:50 +08:00
#endif
This patch provides a set of formatters for most of the commonly used Cocoa classes. The formatter for NSString is an improved version of the one previously shipped as an example, the others are new in design and implementation. A more robust and OO-compliant Objective-C runtime wrapper is provided for runtime versions 1 and 2 on 32 and 64 bit. The formatters are contained in a category named "AppKit", which is not enabled at startup. llvm-svn: 151299
2012-02-24 07:10:03 +08:00
}
<rdar://problem/11086338> Implementing support for synthetic children generated by running C++ code instead of Python scripts ; Adding a bunch of value-generating APIs to our private code layer ; Providing synthetic children for NSArray llvm-svn: 163818
2012-09-14 02:27:09 +08:00
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
void
FormatManager::LoadObjCFormatters()
{
TypeSummaryImpl::Flags objc_flags;
objc_flags.SetCascades(false)
<rdar://problem/11773899> Formatters for BOOL* and BOOL& llvm-svn: 160181
2012-07-14 02:55:41 +08:00
.SetSkipPointers(true)
.SetSkipReferences(true)
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
.SetDontShowChildren(true)
.SetDontShowValue(true)
.SetShowMembersOneLiner(false)
.SetHideItemNames(false);
Making the right thing with regards to disabling summaries on LLDB_DISABLE_PYTHON builds llvm-svn: 163170
2012-09-05 06:04:50 +08:00
TypeCategoryImpl::SharedPointer objc_category_sp = GetCategory(m_objc_category_name);
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
<rdar://problem/11449953> Change Debugger::SetOutputFileHandle() so that it does not automatically initialize the script interpreter in order to transfer its output file handle to it This should delay initialization of Python until strictly necessary and speed-up debugger startup Also, convert formatters for SEL and BOOL ObjC data-types from Python to C++, in order to reap more performance benefits from the above changes llvm-svn: 166967
2012-10-30 05:18:03 +08:00
lldb::TypeSummaryImplSP ObjC_BOOL_summary(new CXXFunctionSummaryFormat(objc_flags, lldb_private::formatters::ObjCBOOLSummaryProvider,""));
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
objc_category_sp->GetTypeSummariesContainer()->Add(ConstString("BOOL"),
This commit provides a new default summary for Objective-C boolean variables, which shows YES or NO instead of the character value. A new category named objc is added to contain this summary provider. Any future Objective-C related formatters would probably fit here llvm-svn: 149388
2012-02-01 01:01:51 +08:00
ObjC_BOOL_summary);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
objc_category_sp->GetTypeSummariesContainer()->Add(ConstString("BOOL &"),
<rdar://problem/11449953> Change Debugger::SetOutputFileHandle() so that it does not automatically initialize the script interpreter in order to transfer its output file handle to it This should delay initialization of Python until strictly necessary and speed-up debugger startup Also, convert formatters for SEL and BOOL ObjC data-types from Python to C++, in order to reap more performance benefits from the above changes llvm-svn: 166967
2012-10-30 05:18:03 +08:00
ObjC_BOOL_summary);
FormatNavigator has long stopped navigating anything - the generation of possible formatters matches is now done elsewhere So, rename the class for what it truly is: a FormattersContainer Also do a bunch of related text substitutions in the interest of overall naming clarity llvm-svn: 197795
2013-12-20 17:38:13 +08:00
objc_category_sp->GetTypeSummariesContainer()->Add(ConstString("BOOL *"),
<rdar://problem/11449953> Change Debugger::SetOutputFileHandle() so that it does not automatically initialize the script interpreter in order to transfer its output file handle to it This should delay initialization of Python until strictly necessary and speed-up debugger startup Also, convert formatters for SEL and BOOL ObjC data-types from Python to C++, in order to reap more performance benefits from the above changes llvm-svn: 166967
2012-10-30 05:18:03 +08:00
ObjC_BOOL_summary);
Change DataExtractor::Dump() to use a series of if..else if statements instead of a switch for the size of the floating point types; some architectures sizeof double and sizeof long double are the same and that's invalid in a switch. Fix the LLDB_DISABLE_PYTHON ifdef block in FormatManager::LoadObjCFormatters so it builds on arm again. llvm-svn: 167263
2012-11-02 07:35:19 +08:00
#ifndef LLDB_DISABLE_PYTHON
(a) adding formatters for: NSTimeZone and CFTimeZonRef SEL and related types CFGregorianDate llvm-svn: 151866
2012-03-02 08:55:53 +08:00
// we need to skip pointers here since we are special casing a SEL* when retrieving its value
objc_flags.SetSkipPointers(true);
<rdar://problem/12725746> Providing data formatters for char16_t* and char32_t* C++11-style Unicode strings Using this chance to refactor the UTF data reader used for data formatters for added generality Added a relevant test case llvm-svn: 172119
2013-01-11 06:08:35 +08:00
AddCXXSummary(objc_category_sp, lldb_private::formatters::ObjCSELSummaryProvider<false>, "SEL summary provider", ConstString("SEL"), objc_flags);
AddCXXSummary(objc_category_sp, lldb_private::formatters::ObjCSELSummaryProvider<false>, "SEL summary provider", ConstString("struct objc_selector"), objc_flags);
AddCXXSummary(objc_category_sp, lldb_private::formatters::ObjCSELSummaryProvider<false>, "SEL summary provider", ConstString("objc_selector"), objc_flags);
AddCXXSummary(objc_category_sp, lldb_private::formatters::ObjCSELSummaryProvider<true>, "SEL summary provider", ConstString("objc_selector *"), objc_flags);
<rdar://problem/13204647> The SEL data formatter was working hard to ensure that pointers-to-selectors could be formatted by the same block of code. In that effort, we were taking the address-of a SEL. This operation fails when the SEL lives in a register, and was causing problems. The formatter has been fixed to work correctly without assuming &selector will be a valid object. llvm-svn: 175227
2013-02-15 08:06:04 +08:00
AddCXXSummary(objc_category_sp, lldb_private::formatters::ObjCSELSummaryProvider<true>, "SEL summary provider", ConstString("SEL *"), objc_flags);
<rdar://problem/11449953> Change Debugger::SetOutputFileHandle() so that it does not automatically initialize the script interpreter in order to transfer its output file handle to it This should delay initialization of Python until strictly necessary and speed-up debugger startup Also, convert formatters for SEL and BOOL ObjC data-types from Python to C++, in order to reap more performance benefits from the above changes llvm-svn: 166967
2012-10-30 05:18:03 +08:00
Porting the Objective-C Class data type’s summary from Python to C++ llvm-svn: 177172
2013-03-16 02:55:30 +08:00
AddCXXSummary(objc_category_sp, lldb_private::formatters::ObjCClassSummaryProvider, "Class summary provider", ConstString("Class"), objc_flags);
Make a synthetic children provider for the ObjC Class type llvm-svn: 180588
2013-04-26 08:59:02 +08:00
SyntheticChildren::Flags class_synth_flags;
class_synth_flags.SetCascades(true).SetSkipPointers(false).SetSkipReferences(false);
AddCXXSynthetic(objc_category_sp, lldb_private::formatters::ObjCClassSyntheticFrontEndCreator, "Class synthetic children", ConstString("Class"), class_synth_flags);
Making the right thing with regards to disabling summaries on LLDB_DISABLE_PYTHON builds llvm-svn: 163170
2012-09-05 06:04:50 +08:00
#endif // LLDB_DISABLE_PYTHON
(a) adding formatters for: NSTimeZone and CFTimeZonRef SEL and related types CFGregorianDate llvm-svn: 151866
2012-03-02 08:55:53 +08:00
<rdar://problem/11449953> Change Debugger::SetOutputFileHandle() so that it does not automatically initialize the script interpreter in order to transfer its output file handle to it This should delay initialization of Python until strictly necessary and speed-up debugger startup Also, convert formatters for SEL and BOOL ObjC data-types from Python to C++, in order to reap more performance benefits from the above changes llvm-svn: 166967
2012-10-30 05:18:03 +08:00
objc_flags.SetSkipPointers(false);
Adding a summary for ObjC blocks llvm-svn: 186023
2013-07-11 02:25:45 +08:00
objc_flags.SetCascades(true);
objc_flags.SetSkipReferences(false);
AddStringSummary (objc_category_sp,
"${var.__FuncPtr%A}",
ConstString("__block_literal_generic"),
objc_flags);
<rdar://problem/11449953> Change Debugger::SetOutputFileHandle() so that it does not automatically initialize the script interpreter in order to transfer its output file handle to it This should delay initialization of Python until strictly necessary and speed-up debugger startup Also, convert formatters for SEL and BOOL ObjC data-types from Python to C++, in order to reap more performance benefits from the above changes llvm-svn: 166967
2012-10-30 05:18:03 +08:00
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
TypeCategoryImpl::SharedPointer corefoundation_category_sp = GetCategory(m_corefoundation_category_name);
<rdar://problem/12709976> Adding a summary for NSError llvm-svn: 169792
2012-12-11 07:30:25 +08:00
AddStringSummary(corefoundation_category_sp,
"${var.years} years, ${var.months} months, ${var.days} days, ${var.hours} hours, ${var.minutes} minutes ${var.seconds} seconds",
ConstString("CFGregorianUnits"),
objc_flags);
AddStringSummary(corefoundation_category_sp,
"location=${var.location} length=${var.length}",
ConstString("CFRange"),
objc_flags);
AddStringSummary(corefoundation_category_sp,
"(x=${var.x}, y=${var.y})",
ConstString("NSPoint"),
objc_flags);
AddStringSummary(corefoundation_category_sp,
"location=${var.location}, length=${var.length}",
ConstString("NSRange"),
objc_flags);
AddStringSummary(corefoundation_category_sp,
"${var.origin}, ${var.size}",
ConstString("NSRect"),
objc_flags);
AddStringSummary(corefoundation_category_sp,
"(${var.origin}, ${var.size}), ...",
ConstString("NSRectArray"),
objc_flags);
AddStringSummary(objc_category_sp,
"(width=${var.width}, height=${var.height})",
ConstString("NSSize"),
objc_flags);
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
TypeCategoryImpl::SharedPointer coregraphics_category_sp = GetCategory(m_coregraphics_category_name);
<rdar://problem/12709976> Adding a summary for NSError llvm-svn: 169792
2012-12-11 07:30:25 +08:00
AddStringSummary(coregraphics_category_sp,
"(width=${var.width}, height=${var.height})",
ConstString("CGSize"),
objc_flags);
AddStringSummary(coregraphics_category_sp,
"(x=${var.x}, y=${var.y})",
ConstString("CGPoint"),
objc_flags);
AddStringSummary(coregraphics_category_sp,
"origin=${var.origin} size=${var.size}",
ConstString("CGRect"),
objc_flags);
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
TypeCategoryImpl::SharedPointer coreservices_category_sp = GetCategory(m_coreservices_category_name);
<rdar://problem/12709976> Adding a summary for NSError llvm-svn: 169792
2012-12-11 07:30:25 +08:00
AddStringSummary(coreservices_category_sp,
"red=${var.red} green=${var.green} blue=${var.blue}",
ConstString("RGBColor"),
objc_flags);
AddStringSummary(coreservices_category_sp,
"(t=${var.top}, l=${var.left}, b=${var.bottom}, r=${var.right})",
ConstString("Rect"),
objc_flags);
AddStringSummary(coreservices_category_sp,
"(v=${var.v}, h=${var.h})",
ConstString("Point"),
objc_flags);
AddStringSummary(coreservices_category_sp,
"${var.month}/${var.day}/${var.year} ${var.hour} :${var.minute} :${var.second} dayOfWeek:${var.dayOfWeek}",
ConstString("DateTimeRect *"),
objc_flags);
AddStringSummary(coreservices_category_sp,
"${var.ld.month}/${var.ld.day}/${var.ld.year} ${var.ld.hour} :${var.ld.minute} :${var.ld.second} dayOfWeek:${var.ld.dayOfWeek}",
ConstString("LongDateRect"),
objc_flags);
AddStringSummary(coreservices_category_sp,
"(x=${var.x}, y=${var.y})",
ConstString("HIPoint"),
objc_flags);
AddStringSummary(coreservices_category_sp,
"origin=${var.origin} size=${var.size}",
ConstString("HIRect"),
objc_flags);
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
This patch provides a set of formatters for most of the commonly used Cocoa classes. The formatter for NSString is an improved version of the one previously shipped as an example, the others are new in design and implementation. A more robust and OO-compliant Objective-C runtime wrapper is provided for runtime versions 1 and 2 on 32 and 64 bit. The formatters are contained in a category named "AppKit", which is not enabled at startup. llvm-svn: 151299
2012-02-24 07:10:03 +08:00
TypeCategoryImpl::SharedPointer appkit_category_sp = GetCategory(m_appkit_category_name);
TypeSummaryImpl::Flags appkit_flags;
appkit_flags.SetCascades(true)
.SetSkipPointers(false)
.SetSkipReferences(false)
.SetDontShowChildren(true)
.SetDontShowValue(false)
.SetShowMembersOneLiner(false)
.SetHideItemNames(false);
<rdar://problem/11086338> Implementing support for synthetic children generated by running C++ code instead of Python scripts ; Adding a bunch of value-generating APIs to our private code layer ; Providing synthetic children for NSArray llvm-svn: 163818
2012-09-14 02:27:09 +08:00
appkit_flags.SetDontShowChildren(false);
Update LLDB_DISABLE_PYTHON #ifdefs in FormatManager.cpp to get it building on no-Python systems again. llvm-svn: 163961
2012-09-15 09:59:02 +08:00
Update the LLDB_DISABLE_PYTHON ifdef in FormatManager::LoadObjCFormatters to get FormatManager.cpp to build on no-python platforms again. llvm-svn: 164284
2012-09-20 14:06:59 +08:00
#ifndef LLDB_DISABLE_PYTHON
<rdar://problem/11086338> Implementing support for synthetic children generated by running C++ code instead of Python scripts ; Adding a bunch of value-generating APIs to our private code layer ; Providing synthetic children for NSArray llvm-svn: 163818
2012-09-14 02:27:09 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSArraySummaryProvider, "NSArray summary provider", ConstString("NSArray"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSArraySummaryProvider, "NSArray summary provider", ConstString("NSMutableArray"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSArraySummaryProvider, "NSArray summary provider", ConstString("__NSArrayI"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSArraySummaryProvider, "NSArray summary provider", ConstString("__NSArrayM"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSArraySummaryProvider, "NSArray summary provider", ConstString("__NSCFArray"), appkit_flags);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSArraySummaryProvider, "NSArray summary provider", ConstString("CFArrayRef"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSArraySummaryProvider, "NSArray summary provider", ConstString("CFMutableArrayRef"), appkit_flags);
<rdar://problem/11086338> Implementing support for synthetic children generated by running C++ code instead of Python scripts ; Adding a bunch of value-generating APIs to our private code layer ; Providing synthetic children for NSArray llvm-svn: 163818
2012-09-14 02:27:09 +08:00
<rdar://problem/11988289> Making C++ synthetic children provider for NSDictionary and related classes llvm-svn: 164144
2012-09-19 01:43:16 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDictionarySummaryProvider<false>, "NSDictionary summary provider", ConstString("NSDictionary"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDictionarySummaryProvider<false>, "NSDictionary summary provider", ConstString("NSMutableDictionary"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDictionarySummaryProvider<false>, "NSDictionary summary provider", ConstString("__NSCFDictionary"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDictionarySummaryProvider<false>, "NSDictionary summary provider", ConstString("__NSDictionaryI"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDictionarySummaryProvider<false>, "NSDictionary summary provider", ConstString("__NSDictionaryM"), appkit_flags);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSDictionarySummaryProvider<true>, "NSDictionary summary provider", ConstString("CFDictionaryRef"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSDictionarySummaryProvider<true>, "NSDictionary summary provider", ConstString("CFMutableDictionaryRef"), appkit_flags);
<rdar://problem/11988289> Making C++ synthetic children provider for NSDictionary and related classes llvm-svn: 164144
2012-09-19 01:43:16 +08:00
<rdar://problem/12529957> Synthetic children provider for NSSet llvm-svn: 175468
2013-02-19 07:16:23 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "NSSet summary", ConstString("NSSet"), appkit_flags);
<rdar://problem/12529989> Synthetic children provider for NSOrderedSet llvm-svn: 180655
2013-04-27 08:27:20 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "NSMutableSet summary", ConstString("NSMutableSet"), appkit_flags);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSSetSummaryProvider<true>, "CFSetRef summary", ConstString("CFSetRef"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSSetSummaryProvider<true>, "CFMutableSetRef summary", ConstString("CFMutableSetRef"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "__NSCFSet summary", ConstString("__NSCFSet"), appkit_flags);
<rdar://problem/12529957> Synthetic children provider for NSSet llvm-svn: 175468
2013-02-19 07:16:23 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "__NSSetI summary", ConstString("__NSSetI"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "__NSSetM summary", ConstString("__NSSetM"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "NSCountedSet summary", ConstString("NSCountedSet"), appkit_flags);
<rdar://problem/12529989> Synthetic children provider for NSOrderedSet llvm-svn: 180655
2013-04-27 08:27:20 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "NSMutableSet summary", ConstString("NSMutableSet"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "NSOrderedSet summary", ConstString("NSOrderedSet"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "__NSOrderedSetI summary", ConstString("__NSOrderedSetI"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSSetSummaryProvider<false>, "__NSOrderedSetM summary", ConstString("__NSOrderedSetM"), appkit_flags);
<rdar://problem/11988289> Making C++ synthetic children provider for NSDictionary and related classes llvm-svn: 164144
2012-09-19 01:43:16 +08:00
// AddSummary(appkit_category_sp, "${var.key%@} -> ${var.value%@}", ConstString("$_lldb_typegen_nspair"), appkit_flags);
<rdar://problem/11086338> Implementing support for synthetic children generated by running C++ code instead of Python scripts ; Adding a bunch of value-generating APIs to our private code layer ; Providing synthetic children for NSArray llvm-svn: 163818
2012-09-14 02:27:09 +08:00
appkit_flags.SetDontShowChildren(true);
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSArraySyntheticFrontEndCreator, "NSArray synthetic children", ConstString("__NSArrayM"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSArraySyntheticFrontEndCreator, "NSArray synthetic children", ConstString("__NSArrayI"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSArraySyntheticFrontEndCreator, "NSArray synthetic children", ConstString("NSArray"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSArraySyntheticFrontEndCreator, "NSArray synthetic children", ConstString("NSMutableArray"), ScriptedSyntheticChildren::Flags());
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSynthetic(corefoundation_category_sp, lldb_private::formatters::NSArraySyntheticFrontEndCreator, "NSArray synthetic children", ConstString("__NSCFArray"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(corefoundation_category_sp, lldb_private::formatters::NSArraySyntheticFrontEndCreator, "NSArray synthetic children", ConstString("CFMutableArrayRef"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(corefoundation_category_sp, lldb_private::formatters::NSArraySyntheticFrontEndCreator, "NSArray synthetic children", ConstString("CFArrayRef"), ScriptedSyntheticChildren::Flags());
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSDictionarySyntheticFrontEndCreator, "NSDictionary synthetic children", ConstString("__NSDictionaryM"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSDictionarySyntheticFrontEndCreator, "NSDictionary synthetic children", ConstString("__NSDictionaryI"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSDictionarySyntheticFrontEndCreator, "NSDictionary synthetic children", ConstString("NSDictionary"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSDictionarySyntheticFrontEndCreator, "NSDictionary synthetic children", ConstString("NSMutableDictionary"), ScriptedSyntheticChildren::Flags());
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSynthetic(corefoundation_category_sp, lldb_private::formatters::NSDictionarySyntheticFrontEndCreator, "NSDictionary synthetic children", ConstString("CFDictionaryRef"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(corefoundation_category_sp, lldb_private::formatters::NSDictionarySyntheticFrontEndCreator, "NSDictionary synthetic children", ConstString("CFMutableDictionaryRef"), ScriptedSyntheticChildren::Flags());
<rdar://problem/11988289> Making C++ synthetic children provider for NSDictionary and related classes llvm-svn: 164144
2012-09-19 01:43:16 +08:00
<rdar://problem/12529957> Synthetic children provider for NSSet llvm-svn: 175468
2013-02-19 07:16:23 +08:00
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSSetSyntheticFrontEndCreator, "NSSet synthetic children", ConstString("NSSet"), ScriptedSyntheticChildren::Flags());
<rdar://problem/12529989> Synthetic children provider for NSOrderedSet llvm-svn: 180655
2013-04-27 08:27:20 +08:00
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSSetSyntheticFrontEndCreator, "__NSSetI synthetic children", ConstString("__NSSetI"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSSetSyntheticFrontEndCreator, "__NSSetM synthetic children", ConstString("__NSSetM"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSSetSyntheticFrontEndCreator, "NSMutableSet synthetic children", ConstString("NSMutableSet"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSSetSyntheticFrontEndCreator, "NSOrderedSet synthetic children", ConstString("NSOrderedSet"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSSetSyntheticFrontEndCreator, "__NSOrderedSetI synthetic children", ConstString("__NSOrderedSetI"), ScriptedSyntheticChildren::Flags());
AddCXXSynthetic(appkit_category_sp, lldb_private::formatters::NSSetSyntheticFrontEndCreator, "__NSOrderedSetM synthetic children", ConstString("__NSOrderedSetM"), ScriptedSyntheticChildren::Flags());
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp,lldb_private::formatters::CFBagSummaryProvider, "CFBag summary provider", ConstString("CFBagRef"), appkit_flags);
AddCXXSummary(corefoundation_category_sp,lldb_private::formatters::CFBagSummaryProvider, "CFBag summary provider", ConstString("__CFBag"), appkit_flags);
AddCXXSummary(corefoundation_category_sp,lldb_private::formatters::CFBagSummaryProvider, "CFBag summary provider", ConstString("const struct __CFBag"), appkit_flags);
AddCXXSummary(corefoundation_category_sp,lldb_private::formatters::CFBagSummaryProvider, "CFBag summary provider", ConstString("CFMutableBagRef"), appkit_flags);
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp,lldb_private::formatters::CFBinaryHeapSummaryProvider, "CFBinaryHeap summary provider", ConstString("CFBinaryHeapRef"), appkit_flags);
AddCXXSummary(corefoundation_category_sp,lldb_private::formatters::CFBinaryHeapSummaryProvider, "CFBinaryHeap summary provider", ConstString("__CFBinaryHeap"), appkit_flags);
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
<rdar://problem/11086338> Implementing support for synthetic children generated by running C++ code instead of Python scripts ; Adding a bunch of value-generating APIs to our private code layer ; Providing synthetic children for NSArray llvm-svn: 163818
2012-09-14 02:27:09 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("NSString"), appkit_flags);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("CFStringRef"), appkit_flags);
__CFString should also format as an NSString llvm-svn: 198727
2014-01-08 10:34:42 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("__CFString"), appkit_flags);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("CFMutableStringRef"), appkit_flags);
<rdar://problem/12898191> Added a summary for NSMutableAttributedString In the process, converted formatters for other NSString-based classes over to C++ code llvm-svn: 174693
2013-02-08 09:55:46 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("NSMutableString"), appkit_flags);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("__NSCFConstantString"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("__NSCFString"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("NSCFConstantString"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("NSCFString"), appkit_flags);
<rdar://problem/11086338> Implementing support for synthetic children generated by running C++ code instead of Python scripts ; Adding a bunch of value-generating APIs to our private code layer ; Providing synthetic children for NSArray llvm-svn: 163818
2012-09-14 02:27:09 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSStringSummaryProvider, "NSString summary provider", ConstString("NSPathStore2"), appkit_flags);
<rdar://problem/11485744> Implement important data formatters in C++. Have the Objective-C language runtime plugin expose class descriptors objects akin to the objc_runtime.py Pythonic implementation. Rewrite the data formatters for some core Cocoa classes in C++ instead of Python. llvm-svn: 163155
2012-09-05 02:47:54 +08:00
<rdar://problem/12898191> Added a summary for NSMutableAttributedString In the process, converted formatters for other NSString-based classes over to C++ code llvm-svn: 174693
2013-02-08 09:55:46 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSAttributedStringSummaryProvider, "NSAttributedString summary provider", ConstString("NSAttributedString"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSMutableAttributedStringSummaryProvider, "NSMutableAttributedString summary provider", ConstString("NSMutableAttributedString"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSMutableAttributedStringSummaryProvider, "NSMutableAttributedString summary provider", ConstString("NSConcreteMutableAttributedString"), appkit_flags);
Moving CFBag and NSBundle summaries from Python to C++ Temporarily disabled non-code-running summaries for CF*Dictionary and NSCountedSet llvm-svn: 177171
2013-03-16 02:44:08 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSBundleSummaryProvider, "NSBundle summary provider", ConstString("NSBundle"), appkit_flags);
Making the right thing with regards to disabling summaries on LLDB_DISABLE_PYTHON builds llvm-svn: 163170
2012-09-05 06:04:50 +08:00
<rdar://problem/11086338> Implementing support for synthetic children generated by running C++ code instead of Python scripts ; Adding a bunch of value-generating APIs to our private code layer ; Providing synthetic children for NSArray llvm-svn: 163818
2012-09-14 02:27:09 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDataSummaryProvider<false>, "NSData summary provider", ConstString("NSData"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDataSummaryProvider<false>, "NSData summary provider", ConstString("NSConcreteData"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDataSummaryProvider<false>, "NSData summary provider", ConstString("NSConcreteMutableData"), appkit_flags);
rdar://16361422 Add NSMutableData to the list of types that the NSData formatter knows to represent llvm-svn: 204289
2014-03-20 09:15:44 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDataSummaryProvider<false>, "NSData summary provider", ConstString("NSMutableData"), appkit_flags);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSDataSummaryProvider<false>, "NSData summary provider", ConstString("__NSCFData"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSDataSummaryProvider<true>, "NSData summary provider", ConstString("CFDataRef"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSDataSummaryProvider<true>, "NSData summary provider", ConstString("CFMutableDataRef"), appkit_flags);
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
Converting more data formatters to C++ - NSBundle, CFBinaryHeap, NSMachPort and NSNotification llvm-svn: 177213
2013-03-16 08:50:25 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSMachPortSummaryProvider, "NSMachPort summary provider", ConstString("NSMachPort"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSNotificationSummaryProvider, "NSNotification summary provider", ConstString("NSNotification"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSNotificationSummaryProvider, "NSNotification summary provider", ConstString("NSConcreteNotification"), appkit_flags);
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
<rdar://problem/12709976> Adding a summary for NSError llvm-svn: 169792
2012-12-11 07:30:25 +08:00
AddStringSummary(appkit_category_sp, "domain: ${var._domain} - code: ${var._code}", ConstString("NSError"), appkit_flags);
<rdar://problem/12978143> Data formatters now cache themselves. This commit provides a new formatter cache mechanism. Upon resolving a formatter (summary or synthetic), LLDB remembers the resolution for later faster retrieval. Also moved the data formatters subsystem from the core to its own group and folder for easier management, and done some code reorganization. The ObjC runtime v1 now returns a class name if asked for the dynamic type of an object. This is required for formatters caching to work with the v1 runtime. Lastly, this commit disposes of the old hack where ValueObjects had to remember whether they were queried for formatters with their static or dynamic type. Now the ValueObjectDynamicValue class works well enough that we can use its dynamic value setting for the same purpose. llvm-svn: 173728
2013-01-29 07:47:25 +08:00
AddStringSummary(appkit_category_sp,"name:${var.name%S} reason:${var.reason%S}",ConstString("NSException"),appkit_flags);
<rdar://problem/11086338> Implementing support for synthetic children generated by running C++ code instead of Python scripts ; Adding a bunch of value-generating APIs to our private code layer ; Providing synthetic children for NSArray llvm-svn: 163818
2012-09-14 02:27:09 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSNumberSummaryProvider, "NSNumber summary provider", ConstString("NSNumber"), appkit_flags);
<rdar://problem/15154623> Move a couple formatters from category AppKit to CoreFoundation llvm-svn: 200713
2014-02-04 03:46:31 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSNumberSummaryProvider, "CFNumberRef summary provider", ConstString("CFNumberRef"), appkit_flags);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSNumberSummaryProvider, "NSNumber summary provider", ConstString("__NSCFBoolean"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSNumberSummaryProvider, "NSNumber summary provider", ConstString("__NSCFNumber"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSNumberSummaryProvider, "NSNumber summary provider", ConstString("NSCFBoolean"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSNumberSummaryProvider, "NSNumber summary provider", ConstString("NSCFNumber"), appkit_flags);
<rdar://problem/12099592> Adding back a bunch of code-running summaries llvm-svn: 165186
2012-10-04 07:53:45 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::RuntimeSpecificDescriptionSummaryProvider, "NSDecimalNumber summary provider", ConstString("NSDecimalNumber"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::RuntimeSpecificDescriptionSummaryProvider, "NSHost summary provider", ConstString("NSHost"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::RuntimeSpecificDescriptionSummaryProvider, "NSTask summary provider", ConstString("NSTask"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::RuntimeSpecificDescriptionSummaryProvider, "NSValue summary provider", ConstString("NSValue"), appkit_flags);
NSSet formatter is now C++ code Split some NS* formatters in their own source files Refactored a utility function for the C++ formatters to use Fixed the skip-summary test case to be explicit about requiring libstdc++ for operation llvm-svn: 175323
2013-02-16 07:38:37 +08:00
This checkin implements the data formatter for NSURL in C++ code llvm-svn: 174735
2013-02-09 03:28:04 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSURLSummaryProvider, "NSURL summary provider", ConstString("NSURL"), appkit_flags);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSURLSummaryProvider, "NSURL summary provider", ConstString("CFURLRef"), appkit_flags);
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
This checkin removes the last Cocoa formatters that were implemented in Python and reimplements them in C++. The Python Cocoa formatters are not shipped as part of LLDB anymore, but still exist in the source repository for user reference. Python formatters still exist for STL classes and users can still define their own Python formatters llvm-svn: 177366
2013-03-19 08:27:22 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDateSummaryProvider, "NSDate summary provider", ConstString("NSDate"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDateSummaryProvider, "NSDate summary provider", ConstString("__NSDate"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDateSummaryProvider, "NSDate summary provider", ConstString("__NSTaggedDate"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSDateSummaryProvider, "NSDate summary provider", ConstString("NSCalendarDate"), appkit_flags);
Returning data formatters to their previous working condition - Plus fixing an issue that was preventing Python oneliners from executing llvm-svn: 155563
2012-04-26 01:53:41 +08:00
C++ formatters for NSTimeZone and for CFBitVector llvm-svn: 177219
2013-03-16 09:50:07 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSTimeZoneSummaryProvider, "NSTimeZone summary provider", ConstString("NSTimeZone"), appkit_flags);
<rdar://problem/15154623> Move a couple formatters from category AppKit to CoreFoundation llvm-svn: 200713
2014-02-04 03:46:31 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::NSTimeZoneSummaryProvider, "NSTimeZone summary provider", ConstString("CFTimeZoneRef"), appkit_flags);
C++ formatters for NSTimeZone and for CFBitVector llvm-svn: 177219
2013-03-16 09:50:07 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSTimeZoneSummaryProvider, "NSTimeZone summary provider", ConstString("__NSTimeZone"), appkit_flags);
(a) adding formatters for: NSTimeZone and CFTimeZonRef SEL and related types CFGregorianDate llvm-svn: 151866
2012-03-02 08:55:53 +08:00
(a) adding an introspection formatter for NS(Mutable)IndexSet (b) fixes and improvements to the formatters for NSDate and NSString (c) adding an introspection formatter for NSCountedSet (d) making the Objective-C formatters test cases pass on both 64 and 32 bit one of the test cases is marked as expected failure on i386 - support needs to be added to the LLDB core for it to pass llvm-svn: 151826
2012-03-02 03:32:33 +08:00
// CFAbsoluteTime is actually a double rather than a pointer to an object
// we do not care about the numeric value, since it is probably meaningless to users
appkit_flags.SetDontShowValue(true);
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::CFAbsoluteTimeSummaryProvider, "CFAbsoluteTime summary provider", ConstString("CFAbsoluteTime"), appkit_flags);
(a) adding an introspection formatter for NS(Mutable)IndexSet (b) fixes and improvements to the formatters for NSDate and NSString (c) adding an introspection formatter for NSCountedSet (d) making the Objective-C formatters test cases pass on both 64 and 32 bit one of the test cases is marked as expected failure on i386 - support needs to be added to the LLDB core for it to pass llvm-svn: 151826
2012-03-02 03:32:33 +08:00
appkit_flags.SetDontShowValue(false);
NS(Mutable)IndexSet formatter moves from Python to C++ llvm-svn: 177217
2013-03-16 09:18:00 +08:00
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSIndexSetSummaryProvider, "NSIndexSet summary provider", ConstString("NSIndexSet"), appkit_flags);
AddCXXSummary(appkit_category_sp, lldb_private::formatters::NSIndexSetSummaryProvider, "NSIndexSet summary provider", ConstString("NSMutableIndexSet"), appkit_flags);
(a) adding an introspection formatter for NS(Mutable)IndexSet (b) fixes and improvements to the formatters for NSDate and NSString (c) adding an introspection formatter for NSCountedSet (d) making the Objective-C formatters test cases pass on both 64 and 32 bit one of the test cases is marked as expected failure on i386 - support needs to be added to the LLDB core for it to pass llvm-svn: 151826
2012-03-02 03:32:33 +08:00
<rdar://problem/15154623> Move a couple formatters from category AppKit to CoreFoundation llvm-svn: 200713
2014-02-04 03:46:31 +08:00
AddStringSummary(corefoundation_category_sp,
<rdar://problem/12709976> Adding a summary for NSError llvm-svn: 169792
2012-12-11 07:30:25 +08:00
"@\"${var.month%d}/${var.day%d}/${var.year%d} ${var.hour%d}:${var.minute%d}:${var.second}\"",
ConstString("CFGregorianDate"),
appkit_flags);
(a) adding formatters for: NSTimeZone and CFTimeZonRef SEL and related types CFGregorianDate llvm-svn: 151866
2012-03-02 08:55:53 +08:00
<rdar://problem/15154623> Several CF* data formatters were in the AppKit category This puts them back where they belong, i.e. in category CoreFoundation llvm-svn: 192008
2013-10-05 08:03:07 +08:00
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::CFBitVectorSummaryProvider, "CFBitVector summary provider", ConstString("CFBitVectorRef"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::CFBitVectorSummaryProvider, "CFBitVector summary provider", ConstString("CFMutableBitVectorRef"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::CFBitVectorSummaryProvider, "CFBitVector summary provider", ConstString("__CFBitVector"), appkit_flags);
AddCXXSummary(corefoundation_category_sp, lldb_private::formatters::CFBitVectorSummaryProvider, "CFBitVector summary provider", ConstString("__CFMutableBitVector"), appkit_flags);
Making the right thing with regards to disabling summaries on LLDB_DISABLE_PYTHON builds llvm-svn: 163170
2012-09-05 06:04:50 +08:00
#endif // LLDB_DISABLE_PYTHON
Adding formatters for several useful Objective-C/Cocoa data types. The new categories are not enabled at startup, but can be manually activated if desired. Adding new API calls to SBValue to be able to retrieve the associated formatters Some refactoring to FormatNavigator::Get() in order to shrink its size down to more manageable terms (a future, massive, refactoring effort will still be needed) Test cases added for the above llvm-svn: 150784
2012-02-17 11:18:30 +08:00
TypeCategoryImpl::SharedPointer vectors_category_sp = GetCategory(m_vectortypes_category_name);
TypeSummaryImpl::Flags vector_flags;
vector_flags.SetCascades(true)
.SetSkipPointers(true)
.SetSkipReferences(false)
.SetDontShowChildren(true)
.SetDontShowValue(false)
.SetShowMembersOneLiner(true)
.SetHideItemNames(true);
<rdar://problem/12709976> Adding a summary for NSError llvm-svn: 169792
2012-12-11 07:30:25 +08:00
AddStringSummary(vectors_category_sp,
"${var.uint128}",
ConstString("builtin_type_vec128"),
objc_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("float [4]"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("int32_t [4]"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("int16_t [8]"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vDouble"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vFloat"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vSInt8"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vSInt16"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vSInt32"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vUInt16"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vUInt8"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vUInt16"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vUInt32"),
vector_flags);
AddStringSummary(vectors_category_sp,
"",
ConstString("vBool32"),
vector_flags);
Fix preprocessor warnings for no newline at the end of the source files. llvm-svn: 141755
2011-10-12 08:53:29 +08:00
}
Refactor the hardcoded formatters facility to use sequences of lambdas - still no feature change as none are present now, but this feels cleaner. Also, hardcoded formatters do not need to be per-type, so disable caching thereof llvm-svn: 216004
2014-08-20 02:47:58 +08:00
void
FormatManager::LoadHardcodedFormatters()
{
{
// insert code to load formats here
}
{
// insert code to load summaries here
}
{
// insert code to load synthetics here
}
}