maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/lib/AST/CommentSema.cpp

823 lines
29 KiB
C++
Raw Normal View History

Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
//===--- CommentSema.cpp - Doxygen comment semantic analysis --------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "clang/AST/CommentSema.h"
Pull the Attr iteration parts out of Attr.h, so including DeclBase.h doesn't pull in all the generated Attr code. Required to pull some functions out of line, but this shouldn't have a perf impact. No functionality change. llvm-svn: 169092
2012-12-01 23:09:41 +08:00
#include "clang/AST/Attr.h"
Comment parsing: extract TableGen'able pieces into new CommandTraits class. llvm-svn: 161548
2012-08-09 08:03:17 +08:00
#include "clang/AST/CommentCommandTraits.h"
Pull the Attr iteration parts out of Attr.h, so including DeclBase.h doesn't pull in all the generated Attr code. Required to pull some functions out of line, but this shouldn't have a perf impact. No functionality change. llvm-svn: 169092
2012-12-01 23:09:41 +08:00
#include "clang/AST/CommentDiagnostic.h"
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
#include "clang/AST/Decl.h"
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
#include "clang/AST/DeclTemplate.h"
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
#include "clang/Basic/SourceManager.h"
Move the 'find macro by spelling' infrastructure to the Preprocessor class and use it to suggest appropriate macro for __attribute__((deprecated)) in -Wdocumentation-deprecated-sync. llvm-svn: 164892
2012-09-29 19:40:46 +08:00
#include "clang/Lex/Preprocessor.h"
#include "llvm/ADT/SmallString.h"
Sort all of Clang's files under 'lib', and fix up the broken headers uncovered. This required manually correcting all of the incorrect main-module headers I could find, and running the new llvm/utils/sort_includes.py script over the files. I also manually added quite a few missing headers that were uncovered by shuffling the order or moving headers up to be main-module-headers. llvm-svn: 169237
2012-12-04 17:13:33 +08:00
#include "llvm/ADT/StringSwitch.h"
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
namespace clang {
namespace comments {
Comment HTML tag name machers: move from StringSwitch to an efficient TableGen-generated string matcher. llvm-svn: 162969
2012-08-31 10:21:44 +08:00
namespace {
#include "clang/AST/CommentHTMLTagsProperties.inc"
} // unnamed namespace
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
Sema::Sema(llvm::BumpPtrAllocator &Allocator, const SourceManager &SourceMgr,
Move the 'find macro by spelling' infrastructure to the Preprocessor class and use it to suggest appropriate macro for __attribute__((deprecated)) in -Wdocumentation-deprecated-sync. llvm-svn: 164892
2012-09-29 19:40:46 +08:00
DiagnosticsEngine &Diags, CommandTraits &Traits,
const Preprocessor *PP) :
Comment parsing: extract TableGen'able pieces into new CommandTraits class. llvm-svn: 161548
2012-08-09 08:03:17 +08:00
Allocator(Allocator), SourceMgr(SourceMgr), Diags(Diags), Traits(Traits),
Move the 'find macro by spelling' infrastructure to the Preprocessor class and use it to suggest appropriate macro for __attribute__((deprecated)) in -Wdocumentation-deprecated-sync. llvm-svn: 164892
2012-09-29 19:40:46 +08:00
PP(PP), ThisDeclInfo(NULL), BriefCommand(NULL), ReturnsCommand(NULL) {
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
}
void Sema::setDecl(const Decl *D) {
Comment AST: add DeclInfo to store information about the declaration. Sema was already extracting most of this, but discarding at the end of semantic analysis. llvm-svn: 161140
2012-08-02 07:08:09 +08:00
if (!D)
return;
ThisDeclInfo = new (Allocator) DeclInfo;
[Doc parsing] This patch searches overridden objc/c++ methods looking for documentation on a particular base class inherited by any method that overrides the base class. In case of redeclaration, as when objc method is defined in the implementation, it also looks up for documentation in class/class extension being redeclared. llvm-svn: 165643
2012-10-11 02:34:52 +08:00
ThisDeclInfo->CommentDecl = D;
Initialize flag in DeclInfo. llvm-svn: 161141
2012-08-02 07:21:57 +08:00
ThisDeclInfo->IsFilled = false;
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
ParagraphComment *Sema::actOnParagraphComment(
ArrayRef<InlineContentComment *> Content) {
return new (Allocator) ParagraphComment(Content);
}
BlockCommandComment *Sema::actOnBlockCommandStart(SourceLocation LocBegin,
SourceLocation LocEnd,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
unsigned CommandID) {
return new (Allocator) BlockCommandComment(LocBegin, LocEnd, CommandID);
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
void Sema::actOnBlockCommandArgs(BlockCommandComment *Command,
ArrayRef<BlockCommandComment::Argument> Args) {
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
Command->setArgs(Args);
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
void Sema::actOnBlockCommandFinish(BlockCommandComment *Command,
ParagraphComment *Paragraph) {
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
Command->setParagraph(Paragraph);
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
checkBlockCommandEmptyParagraph(Command);
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
checkBlockCommandDuplicate(Command);
Comment diagnostics: warn if \returns is used in a non-function comment or if the function returns void. llvm-svn: 161261
2012-08-04 05:15:32 +08:00
checkReturnsCommand(Command);
Comment sema: warn when comment has \deprecated but declaration does not have a deprecation attribute ('deprecated', 'availability' or 'unavailable'). This warning is under a separate flag, -Wdocumentation-deprecated-sync, so it can be turned off easily while leaving other -Wdocumentation warnings on. llvm-svn: 164467
2012-09-23 05:47:50 +08:00
checkDeprecatedCommand(Command);
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
ParamCommandComment *Sema::actOnParamCommandStart(SourceLocation LocBegin,
SourceLocation LocEnd,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
unsigned CommandID) {
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
ParamCommandComment *Command =
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
new (Allocator) ParamCommandComment(LocBegin, LocEnd, CommandID);
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
Comment Sema: refactor handling of 'ParmVarDecl's and save them in Sema members. llvm-svn: 160634
2012-07-24 01:40:30 +08:00
if (!isFunctionDecl())
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
Diag(Command->getLocation(),
diag::warn_doc_param_not_attached_to_a_function_decl)
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
<< Command->getCommandNameRange(Traits);
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
return Command;
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
void Sema::actOnParamCommandDirectionArg(ParamCommandComment *Command,
SourceLocation ArgLocBegin,
SourceLocation ArgLocEnd,
StringRef Arg) {
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
ParamCommandComment::PassDirection Direction;
std::string ArgLower = Arg.lower();
// TODO: optimize: lower Name first (need an API in SmallString for that),
// after that StringSwitch.
if (ArgLower == "[in]")
Direction = ParamCommandComment::In;
else if (ArgLower == "[out]")
Direction = ParamCommandComment::Out;
else if (ArgLower == "[in,out]" || ArgLower == "[out,in]")
Direction = ParamCommandComment::InOut;
else {
// Remove spaces.
std::string::iterator O = ArgLower.begin();
for (std::string::iterator I = ArgLower.begin(), E = ArgLower.end();
I != E; ++I) {
const char C = *I;
if (C != ' ' && C != '\n' && C != '\r' &&
C != '\t' && C != '\v' && C != '\f')
*O++ = C;
}
ArgLower.resize(O - ArgLower.begin());
bool RemovingWhitespaceHelped = false;
if (ArgLower == "[in]") {
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
Direction = ParamCommandComment::In;
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
RemovingWhitespaceHelped = true;
} else if (ArgLower == "[out]") {
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
Direction = ParamCommandComment::Out;
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
RemovingWhitespaceHelped = true;
} else if (ArgLower == "[in,out]" || ArgLower == "[out,in]") {
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
Direction = ParamCommandComment::InOut;
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
RemovingWhitespaceHelped = true;
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
} else {
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
Direction = ParamCommandComment::In;
RemovingWhitespaceHelped = false;
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
SourceRange ArgRange(ArgLocBegin, ArgLocEnd);
if (RemovingWhitespaceHelped)
Diag(ArgLocBegin, diag::warn_doc_param_spaces_in_direction)
<< ArgRange
<< FixItHint::CreateReplacement(
ArgRange,
ParamCommandComment::getDirectionAsString(Direction));
else
Diag(ArgLocBegin, diag::warn_doc_param_invalid_direction)
<< ArgRange;
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
Command->setDirection(Direction, /* Explicit = */ true);
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
void Sema::actOnParamCommandParamNameArg(ParamCommandComment *Command,
SourceLocation ArgLocBegin,
SourceLocation ArgLocEnd,
StringRef Arg) {
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
// Parser will not feed us more arguments than needed.
Comment AST nodes: rename getXXXCount() methods to getNumXXXs() to be in line with Statement AST nodes. llvm-svn: 160182
2012-07-14 03:02:42 +08:00
assert(Command->getNumArgs() == 0);
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
if (!Command->isDirectionExplicit()) {
// User didn't provide a direction argument.
Command->setDirection(ParamCommandComment::In, /* Explicit = */ false);
}
typedef BlockCommandComment::Argument Argument;
Argument *A = new (Allocator) Argument(SourceRange(ArgLocBegin,
ArgLocEnd),
Arg);
Command->setArgs(llvm::makeArrayRef(A, 1));
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
void Sema::actOnParamCommandFinish(ParamCommandComment *Command,
ParagraphComment *Paragraph) {
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
Command->setParagraph(Paragraph);
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
checkBlockCommandEmptyParagraph(Command);
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
TParamCommandComment *Sema::actOnTParamCommandStart(SourceLocation LocBegin,
SourceLocation LocEnd,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
unsigned CommandID) {
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
TParamCommandComment *Command =
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
new (Allocator) TParamCommandComment(LocBegin, LocEnd, CommandID);
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
Comment AST: DeclInfo: collapse a bunch of boolean flags into an enum. llvm-svn: 161352
2012-08-07 05:31:15 +08:00
if (!isTemplateOrSpecialization())
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
Diag(Command->getLocation(),
diag::warn_doc_tparam_not_attached_to_a_template_decl)
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
<< Command->getCommandNameRange(Traits);
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
return Command;
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
void Sema::actOnTParamCommandParamNameArg(TParamCommandComment *Command,
SourceLocation ArgLocBegin,
SourceLocation ArgLocEnd,
StringRef Arg) {
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
// Parser will not feed us more arguments than needed.
assert(Command->getNumArgs() == 0);
typedef BlockCommandComment::Argument Argument;
Argument *A = new (Allocator) Argument(SourceRange(ArgLocBegin,
ArgLocEnd),
Arg);
Command->setArgs(llvm::makeArrayRef(A, 1));
Comment AST: DeclInfo: collapse a bunch of boolean flags into an enum. llvm-svn: 161352
2012-08-07 05:31:15 +08:00
if (!isTemplateOrSpecialization()) {
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
// We already warned that this \\tparam is not attached to a template decl.
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
return;
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
}
Comment AST: add DeclInfo to store information about the declaration. Sema was already extracting most of this, but discarding at the end of semantic analysis. llvm-svn: 161140
2012-08-02 07:08:09 +08:00
const TemplateParameterList *TemplateParameters =
ThisDeclInfo->TemplateParameters;
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
SmallVector<unsigned, 2> Position;
if (resolveTParamReference(Arg, TemplateParameters, &Position)) {
Command->setPosition(copyArray(llvm::makeArrayRef(Position)));
llvm::StringMap<TParamCommandComment *>::iterator PrevCommandIt =
TemplateParameterDocs.find(Arg);
if (PrevCommandIt != TemplateParameterDocs.end()) {
SourceRange ArgRange(ArgLocBegin, ArgLocEnd);
Diag(ArgLocBegin, diag::warn_doc_tparam_duplicate)
<< Arg << ArgRange;
TParamCommandComment *PrevCommand = PrevCommandIt->second;
Diag(PrevCommand->getLocation(), diag::note_doc_tparam_previous)
<< PrevCommand->getParamNameRange();
}
TemplateParameterDocs[Arg] = Command;
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
return;
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
}
SourceRange ArgRange(ArgLocBegin, ArgLocEnd);
Diag(ArgLocBegin, diag::warn_doc_tparam_not_found)
<< Arg << ArgRange;
if (!TemplateParameters || TemplateParameters->size() == 0)
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
return;
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
StringRef CorrectedName;
if (TemplateParameters->size() == 1) {
const NamedDecl *Param = TemplateParameters->getParam(0);
const IdentifierInfo *II = Param->getIdentifier();
if (II)
CorrectedName = II->getName();
} else {
CorrectedName = correctTypoInTParamReference(Arg, TemplateParameters);
}
if (!CorrectedName.empty()) {
Diag(ArgLocBegin, diag::note_doc_tparam_name_suggestion)
<< CorrectedName
<< FixItHint::CreateReplacement(ArgRange, CorrectedName);
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
return;
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
void Sema::actOnTParamCommandFinish(TParamCommandComment *Command,
ParagraphComment *Paragraph) {
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
Command->setParagraph(Paragraph);
checkBlockCommandEmptyParagraph(Command);
}
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
InlineCommandComment *Sema::actOnInlineCommand(SourceLocation CommandLocBegin,
SourceLocation CommandLocEnd,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
unsigned CommandID) {
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
ArrayRef<InlineCommandComment::Argument> Args;
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
StringRef CommandName = Traits.getCommandInfo(CommandID)->Name;
Comment AST: add InlineContentComment::RenderKind to specify a default rendering mode for clients that don't want to interpret Doxygen commands. Also add a libclang API to query this information. llvm-svn: 160633
2012-07-24 00:43:01 +08:00
return new (Allocator) InlineCommandComment(
CommandLocBegin,
CommandLocEnd,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
CommandID,
Comment AST: add InlineContentComment::RenderKind to specify a default rendering mode for clients that don't want to interpret Doxygen commands. Also add a libclang API to query this information. llvm-svn: 160633
2012-07-24 00:43:01 +08:00
getInlineCommandRenderKind(CommandName),
Args);
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
InlineCommandComment *Sema::actOnInlineCommand(SourceLocation CommandLocBegin,
SourceLocation CommandLocEnd,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
unsigned CommandID,
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
SourceLocation ArgLocBegin,
SourceLocation ArgLocEnd,
StringRef Arg) {
typedef InlineCommandComment::Argument Argument;
Argument *A = new (Allocator) Argument(SourceRange(ArgLocBegin,
ArgLocEnd),
Arg);
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
StringRef CommandName = Traits.getCommandInfo(CommandID)->Name;
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
Comment AST: add InlineContentComment::RenderKind to specify a default rendering mode for clients that don't want to interpret Doxygen commands. Also add a libclang API to query this information. llvm-svn: 160633
2012-07-24 00:43:01 +08:00
return new (Allocator) InlineCommandComment(
CommandLocBegin,
CommandLocEnd,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
CommandID,
Comment AST: add InlineContentComment::RenderKind to specify a default rendering mode for clients that don't want to interpret Doxygen commands. Also add a libclang API to query this information. llvm-svn: 160633
2012-07-24 00:43:01 +08:00
getInlineCommandRenderKind(CommandName),
llvm::makeArrayRef(A, 1));
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
InlineContentComment *Sema::actOnUnknownCommand(SourceLocation LocBegin,
SourceLocation LocEnd,
Comment parsing: handle non-builtin commands correctly. After semantic analysis registers a command, it becomes a "known" command for the lexer, since it has an ID. Having this freedom of choice to register a command is a good thing since BriefParser does not need this. But the parser should still invoke the correct semantic analysis method (actOnUnknownCommand) in this case. llvm-svn: 163646
2012-09-12 03:22:03 +08:00
StringRef CommandName) {
unsigned CommandID = Traits.registerUnknownCommand(CommandName)->getID();
return actOnUnknownCommand(LocBegin, LocEnd, CommandID);
}
InlineContentComment *Sema::actOnUnknownCommand(SourceLocation LocBegin,
SourceLocation LocEnd,
unsigned CommandID) {
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
ArrayRef<InlineCommandComment::Argument> Args;
Comment AST: add InlineContentComment::RenderKind to specify a default rendering mode for clients that don't want to interpret Doxygen commands. Also add a libclang API to query this information. llvm-svn: 160633
2012-07-24 00:43:01 +08:00
return new (Allocator) InlineCommandComment(
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
LocBegin, LocEnd, CommandID,
Comment AST: add InlineContentComment::RenderKind to specify a default rendering mode for clients that don't want to interpret Doxygen commands. Also add a libclang API to query this information. llvm-svn: 160633
2012-07-24 00:43:01 +08:00
InlineCommandComment::RenderNormal,
Args);
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
TextComment *Sema::actOnText(SourceLocation LocBegin,
SourceLocation LocEnd,
StringRef Text) {
return new (Allocator) TextComment(LocBegin, LocEnd, Text);
}
VerbatimBlockComment *Sema::actOnVerbatimBlockStart(SourceLocation Loc,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
unsigned CommandID) {
StringRef CommandName = Traits.getCommandInfo(CommandID)->Name;
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
return new (Allocator) VerbatimBlockComment(
Loc,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
Loc.getLocWithOffset(1 + CommandName.size()),
CommandID);
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
VerbatimBlockLineComment *Sema::actOnVerbatimBlockLine(SourceLocation Loc,
StringRef Text) {
return new (Allocator) VerbatimBlockLineComment(Loc, Text);
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
void Sema::actOnVerbatimBlockFinish(
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
VerbatimBlockComment *Block,
SourceLocation CloseNameLocBegin,
StringRef CloseName,
ArrayRef<VerbatimBlockLineComment *> Lines) {
Block->setCloseName(CloseName, CloseNameLocBegin);
Block->setLines(Lines);
}
VerbatimLineComment *Sema::actOnVerbatimLine(SourceLocation LocBegin,
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
unsigned CommandID,
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
SourceLocation TextBegin,
StringRef Text) {
return new (Allocator) VerbatimLineComment(
LocBegin,
TextBegin.getLocWithOffset(Text.size()),
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
CommandID,
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
TextBegin,
Text);
}
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
HTMLStartTagComment *Sema::actOnHTMLStartTagStart(SourceLocation LocBegin,
StringRef TagName) {
return new (Allocator) HTMLStartTagComment(LocBegin, TagName);
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
Comment parser and sema: remove useless return values llvm-svn: 161331
2012-08-07 03:03:12 +08:00
void Sema::actOnHTMLStartTagFinish(
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
HTMLStartTagComment *Tag,
ArrayRef<HTMLStartTagComment::Attribute> Attrs,
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
SourceLocation GreaterLoc,
bool IsSelfClosing) {
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
Tag->setAttrs(Attrs);
Tag->setGreaterLoc(GreaterLoc);
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
if (IsSelfClosing)
Tag->setSelfClosing();
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
else if (!isHTMLEndTagForbidden(Tag->getTagName()))
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
HTMLOpenTags.push_back(Tag);
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
HTMLEndTagComment *Sema::actOnHTMLEndTag(SourceLocation LocBegin,
SourceLocation LocEnd,
StringRef TagName) {
HTMLEndTagComment *HET =
new (Allocator) HTMLEndTagComment(LocBegin, LocEnd, TagName);
if (isHTMLEndTagForbidden(TagName)) {
Diag(HET->getLocation(), diag::warn_doc_html_end_forbidden)
<< TagName << HET->getSourceRange();
return HET;
Make concepts of optional and forbidden end tags separate. Thanks Jordan for pointing this! llvm-svn: 160149
2012-07-13 07:37:09 +08:00
}
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
bool FoundOpen = false;
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
for (SmallVectorImpl<HTMLStartTagComment *>::const_reverse_iterator
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
I = HTMLOpenTags.rbegin(), E = HTMLOpenTags.rend();
I != E; ++I) {
if ((*I)->getTagName() == TagName) {
FoundOpen = true;
break;
}
}
if (!FoundOpen) {
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
Diag(HET->getLocation(), diag::warn_doc_html_end_unbalanced)
<< HET->getSourceRange();
return HET;
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
}
while (!HTMLOpenTags.empty()) {
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
const HTMLStartTagComment *HST = HTMLOpenTags.back();
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
HTMLOpenTags.pop_back();
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
StringRef LastNotClosedTagName = HST->getTagName();
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
if (LastNotClosedTagName == TagName)
break;
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
if (isHTMLEndTagOptional(LastNotClosedTagName))
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
continue;
bool OpenLineInvalid;
const unsigned OpenLine = SourceMgr.getPresumedLineNumber(
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
HST->getLocation(),
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
&OpenLineInvalid);
bool CloseLineInvalid;
const unsigned CloseLine = SourceMgr.getPresumedLineNumber(
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
HET->getLocation(),
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
&CloseLineInvalid);
if (OpenLineInvalid || CloseLineInvalid || OpenLine == CloseLine)
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
Diag(HST->getLocation(), diag::warn_doc_html_start_end_mismatch)
<< HST->getTagName() << HET->getTagName()
<< HST->getSourceRange() << HET->getSourceRange();
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
else {
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
Diag(HST->getLocation(), diag::warn_doc_html_start_end_mismatch)
<< HST->getTagName() << HET->getTagName()
<< HST->getSourceRange();
Diag(HET->getLocation(), diag::note_doc_html_end_tag)
<< HET->getSourceRange();
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
}
}
Comment parsing: repaint the bikesched: rename 'HTML open tags' to 'HTML start tags' and 'HTML close tags' to 'HTML end tags' according to HTML spec. llvm-svn: 160153
2012-07-13 08:44:24 +08:00
return HET;
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
FullComment *Sema::actOnFullComment(
ArrayRef<BlockContentComment *> Blocks) {
search for overridden methods with comment when overriding method has none of its own. Factor in Doug's comments. // rdar://12378793 llvm-svn: 165771
2012-10-12 07:52:50 +08:00
FullComment *FC = new (Allocator) FullComment(Blocks, ThisDeclInfo);
Comment diagnostics: for unresolved parameters, do not suggest parameter fixit with parameter that is documented. Fixes PR13670, <rdar://problem/12155840>. llvm-svn: 162570
2012-08-25 01:45:39 +08:00
resolveParamCommandIndexes(FC);
return FC;
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
}
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
void Sema::checkBlockCommandEmptyParagraph(BlockCommandComment *Command) {
Comment parsing: handle \deprecated command. It is a block command, but it should be fine to use it without further explanations in the attached paragraph, so the warning about empty paragraph was turned off for it. llvm-svn: 163836
2012-09-14 04:36:01 +08:00
if (Traits.getCommandInfo(Command->getCommandID())->IsEmptyParagraphAllowed)
return;
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
ParagraphComment *Paragraph = Command->getParagraph();
if (Paragraph->isWhitespace()) {
SourceLocation DiagLoc;
Comment AST nodes: rename getXXXCount() methods to getNumXXXs() to be in line with Statement AST nodes. llvm-svn: 160182
2012-07-14 03:02:42 +08:00
if (Command->getNumArgs() > 0)
DiagLoc = Command->getArgRange(Command->getNumArgs() - 1).getEnd();
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
if (!DiagLoc.isValid())
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
DiagLoc = Command->getCommandNameRange(Traits).getEnd();
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
Diag(DiagLoc, diag::warn_doc_block_command_empty_paragraph)
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
<< Command->getCommandName(Traits)
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
<< Command->getSourceRange();
}
}
Comment diagnostics: warn if \returns is used in a non-function comment or if the function returns void. llvm-svn: 161261
2012-08-04 05:15:32 +08:00
void Sema::checkReturnsCommand(const BlockCommandComment *Command) {
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
if (!Traits.getCommandInfo(Command->getCommandID())->IsReturnsCommand)
Comment diagnostics: warn if \returns is used in a non-function comment or if the function returns void. llvm-svn: 161261
2012-08-04 05:15:32 +08:00
return;
if (isFunctionDecl()) {
if (ThisDeclInfo->ResultType->isVoidType()) {
unsigned DiagKind;
[Doc parsing] This patch searches overridden objc/c++ methods looking for documentation on a particular base class inherited by any method that overrides the base class. In case of redeclaration, as when objc method is defined in the implementation, it also looks up for documentation in class/class extension being redeclared. llvm-svn: 165643
2012-10-11 02:34:52 +08:00
switch (ThisDeclInfo->CommentDecl->getKind()) {
Comment diagnostics: warn if \returns is used in a non-function comment or if the function returns void. llvm-svn: 161261
2012-08-04 05:15:32 +08:00
default:
Comment diagnostics: \return in void function: specialize diagnostic text for ObjC methods. llvm-svn: 161324
2012-08-07 00:29:26 +08:00
if (ThisDeclInfo->IsObjCMethod)
DiagKind = 3;
else
DiagKind = 0;
Comment diagnostics: warn if \returns is used in a non-function comment or if the function returns void. llvm-svn: 161261
2012-08-04 05:15:32 +08:00
break;
case Decl::CXXConstructor:
DiagKind = 1;
break;
case Decl::CXXDestructor:
DiagKind = 2;
break;
}
Diag(Command->getLocation(),
diag::warn_doc_returns_attached_to_a_void_function)
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
<< Command->getCommandName(Traits)
Comment diagnostics: warn if \returns is used in a non-function comment or if the function returns void. llvm-svn: 161261
2012-08-04 05:15:32 +08:00
<< DiagKind
<< Command->getSourceRange();
}
return;
}
Diag(Command->getLocation(),
diag::warn_doc_returns_not_attached_to_a_function_decl)
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
<< Command->getCommandName(Traits)
Comment diagnostics: warn if \returns is used in a non-function comment or if the function returns void. llvm-svn: 161261
2012-08-04 05:15:32 +08:00
<< Command->getSourceRange();
}
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
void Sema::checkBlockCommandDuplicate(const BlockCommandComment *Command) {
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
const CommandInfo *Info = Traits.getCommandInfo(Command->getCommandID());
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
const BlockCommandComment *PrevCommand = NULL;
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
if (Info->IsBriefCommand) {
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
if (!BriefCommand) {
BriefCommand = Command;
return;
}
PrevCommand = BriefCommand;
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
} else if (Info->IsReturnsCommand) {
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
if (!ReturnsCommand) {
ReturnsCommand = Command;
return;
}
PrevCommand = ReturnsCommand;
} else {
// We don't want to check this command for duplicates.
return;
}
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
StringRef CommandName = Command->getCommandName(Traits);
StringRef PrevCommandName = PrevCommand->getCommandName(Traits);
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
Diag(Command->getLocation(), diag::warn_doc_block_command_duplicate)
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
<< CommandName
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
<< Command->getSourceRange();
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
if (CommandName == PrevCommandName)
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
Diag(PrevCommand->getLocation(), diag::note_doc_block_command_previous)
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
<< PrevCommandName
<< PrevCommand->getSourceRange();
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
else
Diag(PrevCommand->getLocation(),
diag::note_doc_block_command_previous_alias)
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
<< PrevCommandName
<< CommandName;
Comment diagnostics: warn on duplicate \brief and \return commands. Doxygen manual claims that multiple \brief or \returns commands will be merged together, but actual behavior is different (second \brief command becomes a part of a discussion, second \returns becomes a "Returns: blah" paragraph on its own). Anyway, it seems to be a bad idea to use multiple \brief or \returns commands in a single command. llvm-svn: 161325
2012-08-07 01:08:27 +08:00
}
Comment sema: warn when comment has \deprecated but declaration does not have a deprecation attribute ('deprecated', 'availability' or 'unavailable'). This warning is under a separate flag, -Wdocumentation-deprecated-sync, so it can be turned off easily while leaving other -Wdocumentation warnings on. llvm-svn: 164467
2012-09-23 05:47:50 +08:00
void Sema::checkDeprecatedCommand(const BlockCommandComment *Command) {
if (!Traits.getCommandInfo(Command->getCommandID())->IsDeprecatedCommand)
return;
[Doc parsing] This patch searches overridden objc/c++ methods looking for documentation on a particular base class inherited by any method that overrides the base class. In case of redeclaration, as when objc method is defined in the implementation, it also looks up for documentation in class/class extension being redeclared. llvm-svn: 165643
2012-10-11 02:34:52 +08:00
const Decl *D = ThisDeclInfo->CommentDecl;
Comment sema: warn when comment has \deprecated but declaration does not have a deprecation attribute ('deprecated', 'availability' or 'unavailable'). This warning is under a separate flag, -Wdocumentation-deprecated-sync, so it can be turned off easily while leaving other -Wdocumentation warnings on. llvm-svn: 164467
2012-09-23 05:47:50 +08:00
if (!D)
return;
if (D->hasAttr<DeprecatedAttr>() ||
D->hasAttr<AvailabilityAttr>() ||
D->hasAttr<UnavailableAttr>())
return;
Diag(Command->getLocation(),
diag::warn_doc_deprecated_not_sync)
<< Command->getSourceRange();
// Try to emit a fixit with a deprecation attribute.
if (const FunctionDecl *FD = dyn_cast<FunctionDecl>(D)) {
// Don't emit a Fix-It for non-member function definitions. GCC does not
// accept attributes on them.
const DeclContext *Ctx = FD->getDeclContext();
if ((!Ctx || !Ctx->isRecord()) &&
FD->doesThisDeclarationHaveABody())
return;
Move the 'find macro by spelling' infrastructure to the Preprocessor class and use it to suggest appropriate macro for __attribute__((deprecated)) in -Wdocumentation-deprecated-sync. llvm-svn: 164892
2012-09-29 19:40:46 +08:00
StringRef AttributeSpelling = "__attribute__((deprecated))";
if (PP) {
TokenValue Tokens[] = {
tok::kw___attribute, tok::l_paren, tok::l_paren,
PP->getIdentifierInfo("deprecated"),
tok::r_paren, tok::r_paren
};
StringRef MacroName = PP->getLastMacroWithSpelling(FD->getLocation(),
Tokens);
if (!MacroName.empty())
AttributeSpelling = MacroName;
}
SmallString<64> TextToInsert(" ");
TextToInsert += AttributeSpelling;
Comment sema: warn when comment has \deprecated but declaration does not have a deprecation attribute ('deprecated', 'availability' or 'unavailable'). This warning is under a separate flag, -Wdocumentation-deprecated-sync, so it can be turned off easily while leaving other -Wdocumentation warnings on. llvm-svn: 164467
2012-09-23 05:47:50 +08:00
Diag(FD->getLocEnd(),
diag::note_add_deprecation_attr)
<< FixItHint::CreateInsertion(FD->getLocEnd().getLocWithOffset(1),
Move the 'find macro by spelling' infrastructure to the Preprocessor class and use it to suggest appropriate macro for __attribute__((deprecated)) in -Wdocumentation-deprecated-sync. llvm-svn: 164892
2012-09-29 19:40:46 +08:00
TextToInsert);
Comment sema: warn when comment has \deprecated but declaration does not have a deprecation attribute ('deprecated', 'availability' or 'unavailable'). This warning is under a separate flag, -Wdocumentation-deprecated-sync, so it can be turned off easily while leaving other -Wdocumentation warnings on. llvm-svn: 164467
2012-09-23 05:47:50 +08:00
}
}
Comment diagnostics: for unresolved parameters, do not suggest parameter fixit with parameter that is documented. Fixes PR13670, <rdar://problem/12155840>. llvm-svn: 162570
2012-08-25 01:45:39 +08:00
void Sema::resolveParamCommandIndexes(const FullComment *FC) {
if (!isFunctionDecl()) {
// We already warned that \\param commands are not attached to a function
// decl.
return;
}
Remove useless 'llvm::' qualifier from names like StringRef and others that are brought into 'clang' namespace by clang/Basic/LLVM.h llvm-svn: 172323
2013-01-13 03:30:44 +08:00
SmallVector<ParamCommandComment *, 8> UnresolvedParamCommands;
Comment diagnostics: for unresolved parameters, do not suggest parameter fixit with parameter that is documented. Fixes PR13670, <rdar://problem/12155840>. llvm-svn: 162570
2012-08-25 01:45:39 +08:00
// Comment AST nodes that correspond to \c ParamVars for which we have
// found a \\param command or NULL if no documentation was found so far.
Remove useless 'llvm::' qualifier from names like StringRef and others that are brought into 'clang' namespace by clang/Basic/LLVM.h llvm-svn: 172323
2013-01-13 03:30:44 +08:00
SmallVector<ParamCommandComment *, 8> ParamVarDocs;
Comment diagnostics: for unresolved parameters, do not suggest parameter fixit with parameter that is documented. Fixes PR13670, <rdar://problem/12155840>. llvm-svn: 162570
2012-08-25 01:45:39 +08:00
ArrayRef<const ParmVarDecl *> ParamVars = getParamVars();
ParamVarDocs.resize(ParamVars.size(), NULL);
// First pass over all \\param commands: resolve all parameter names.
for (Comment::child_iterator I = FC->child_begin(), E = FC->child_end();
I != E; ++I) {
ParamCommandComment *PCC = dyn_cast<ParamCommandComment>(*I);
if (!PCC || !PCC->hasParamName())
continue;
[doc parsing] use getParamName to access parameter for current(rewritten) comment and getParamNameAsWritten to access param name coming with \param marker. llvm-svn: 166231
2012-10-19 05:42:42 +08:00
StringRef ParamName = PCC->getParamNameAsWritten();
Comment diagnostics: for unresolved parameters, do not suggest parameter fixit with parameter that is documented. Fixes PR13670, <rdar://problem/12155840>. llvm-svn: 162570
2012-08-25 01:45:39 +08:00
// Check that referenced parameter name is in the function decl.
const unsigned ResolvedParamIndex = resolveParmVarReference(ParamName,
ParamVars);
if (ResolvedParamIndex == ParamCommandComment::InvalidParamIndex) {
UnresolvedParamCommands.push_back(PCC);
continue;
}
PCC->setParamIndex(ResolvedParamIndex);
if (ParamVarDocs[ResolvedParamIndex]) {
SourceRange ArgRange = PCC->getParamNameRange();
Diag(ArgRange.getBegin(), diag::warn_doc_param_duplicate)
<< ParamName << ArgRange;
ParamCommandComment *PrevCommand = ParamVarDocs[ResolvedParamIndex];
Diag(PrevCommand->getLocation(), diag::note_doc_param_previous)
<< PrevCommand->getParamNameRange();
}
ParamVarDocs[ResolvedParamIndex] = PCC;
}
// Find parameter declarations that have no corresponding \\param.
Remove useless 'llvm::' qualifier from names like StringRef and others that are brought into 'clang' namespace by clang/Basic/LLVM.h llvm-svn: 172323
2013-01-13 03:30:44 +08:00
SmallVector<const ParmVarDecl *, 8> OrphanedParamDecls;
Comment diagnostics: for unresolved parameters, do not suggest parameter fixit with parameter that is documented. Fixes PR13670, <rdar://problem/12155840>. llvm-svn: 162570
2012-08-25 01:45:39 +08:00
for (unsigned i = 0, e = ParamVarDocs.size(); i != e; ++i) {
if (!ParamVarDocs[i])
OrphanedParamDecls.push_back(ParamVars[i]);
}
// Second pass over unresolved \\param commands: do typo correction.
// Suggest corrections from a set of parameter declarations that have no
// corresponding \\param.
for (unsigned i = 0, e = UnresolvedParamCommands.size(); i != e; ++i) {
const ParamCommandComment *PCC = UnresolvedParamCommands[i];
SourceRange ArgRange = PCC->getParamNameRange();
[doc parsing] use getParamName to access parameter for current(rewritten) comment and getParamNameAsWritten to access param name coming with \param marker. llvm-svn: 166231
2012-10-19 05:42:42 +08:00
StringRef ParamName = PCC->getParamNameAsWritten();
Comment diagnostics: for unresolved parameters, do not suggest parameter fixit with parameter that is documented. Fixes PR13670, <rdar://problem/12155840>. llvm-svn: 162570
2012-08-25 01:45:39 +08:00
Diag(ArgRange.getBegin(), diag::warn_doc_param_not_found)
<< ParamName << ArgRange;
// All parameters documented -- can't suggest a correction.
if (OrphanedParamDecls.size() == 0)
continue;
unsigned CorrectedParamIndex = ParamCommandComment::InvalidParamIndex;
if (OrphanedParamDecls.size() == 1) {
// If one parameter is not documented then that parameter is the only
// possible suggestion.
CorrectedParamIndex = 0;
} else {
// Do typo correction.
CorrectedParamIndex = correctTypoInParmVarReference(ParamName,
OrphanedParamDecls);
}
if (CorrectedParamIndex != ParamCommandComment::InvalidParamIndex) {
const ParmVarDecl *CorrectedPVD = OrphanedParamDecls[CorrectedParamIndex];
if (const IdentifierInfo *CorrectedII = CorrectedPVD->getIdentifier())
Diag(ArgRange.getBegin(), diag::note_doc_param_name_suggestion)
<< CorrectedII->getName()
<< FixItHint::CreateReplacement(ArgRange, CorrectedII->getName());
}
}
}
Comment Sema: refactor handling of 'ParmVarDecl's and save them in Sema members. llvm-svn: 160634
2012-07-24 01:40:30 +08:00
bool Sema::isFunctionDecl() {
Comment AST: add DeclInfo to store information about the declaration. Sema was already extracting most of this, but discarding at the end of semantic analysis. llvm-svn: 161140
2012-08-02 07:08:09 +08:00
if (!ThisDeclInfo)
return false;
if (!ThisDeclInfo->IsFilled)
CommentSema: simplify functions, per Jordan's comment. llvm-svn: 160689
2012-07-25 04:58:46 +08:00
inspectThisDecl();
Comments AST: refactor DeclInfo to use an enum for decl kind instead of separate flags. llvm-svn: 161217
2012-08-03 05:45:39 +08:00
return ThisDeclInfo->getKind() == DeclInfo::FunctionKind;
Comment Sema: refactor handling of 'ParmVarDecl's and save them in Sema members. llvm-svn: 160634
2012-07-24 01:40:30 +08:00
}
Comment AST: DeclInfo: collapse a bunch of boolean flags into an enum. llvm-svn: 161352
2012-08-07 05:31:15 +08:00
bool Sema::isTemplateOrSpecialization() {
Comment AST: add DeclInfo to store information about the declaration. Sema was already extracting most of this, but discarding at the end of semantic analysis. llvm-svn: 161140
2012-08-02 07:08:09 +08:00
if (!ThisDeclInfo)
return false;
if (!ThisDeclInfo->IsFilled)
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
inspectThisDecl();
Comment AST: DeclInfo: collapse a bunch of boolean flags into an enum. llvm-svn: 161352
2012-08-07 05:31:15 +08:00
return ThisDeclInfo->getTemplateKind() != DeclInfo::NotTemplate;
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
}
Comment Sema: refactor handling of 'ParmVarDecl's and save them in Sema members. llvm-svn: 160634
2012-07-24 01:40:30 +08:00
ArrayRef<const ParmVarDecl *> Sema::getParamVars() {
Comment AST: add DeclInfo to store information about the declaration. Sema was already extracting most of this, but discarding at the end of semantic analysis. llvm-svn: 161140
2012-08-02 07:08:09 +08:00
if (!ThisDeclInfo->IsFilled)
CommentSema: simplify functions, per Jordan's comment. llvm-svn: 160689
2012-07-25 04:58:46 +08:00
inspectThisDecl();
Comment AST: add DeclInfo to store information about the declaration. Sema was already extracting most of this, but discarding at the end of semantic analysis. llvm-svn: 161140
2012-08-02 07:08:09 +08:00
return ThisDeclInfo->ParamVars;
Comment Sema: refactor handling of 'ParmVarDecl's and save them in Sema members. llvm-svn: 160634
2012-07-24 01:40:30 +08:00
}
void Sema::inspectThisDecl() {
Comment AST: add DeclInfo to store information about the declaration. Sema was already extracting most of this, but discarding at the end of semantic analysis. llvm-svn: 161140
2012-08-02 07:08:09 +08:00
ThisDeclInfo->fill();
Comment Sema: refactor handling of 'ParmVarDecl's and save them in Sema members. llvm-svn: 160634
2012-07-24 01:40:30 +08:00
}
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
unsigned Sema::resolveParmVarReference(StringRef Name,
Comment Sema: refactor handling of 'ParmVarDecl's and save them in Sema members. llvm-svn: 160634
2012-07-24 01:40:30 +08:00
ArrayRef<const ParmVarDecl *> ParamVars) {
for (unsigned i = 0, e = ParamVars.size(); i != e; ++i) {
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
const IdentifierInfo *II = ParamVars[i]->getIdentifier();
if (II && II->getName() == Name)
return i;
}
return ParamCommandComment::InvalidParamIndex;
}
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
namespace {
class SimpleTypoCorrector {
StringRef Typo;
const unsigned MaxEditDistance;
const NamedDecl *BestDecl;
unsigned BestEditDistance;
unsigned BestIndex;
unsigned NextIndex;
public:
SimpleTypoCorrector(StringRef Typo) :
Typo(Typo), MaxEditDistance((Typo.size() + 2) / 3),
BestDecl(NULL), BestEditDistance(MaxEditDistance + 1),
BestIndex(0), NextIndex(0)
{ }
void addDecl(const NamedDecl *ND);
const NamedDecl *getBestDecl() const {
if (BestEditDistance > MaxEditDistance)
return NULL;
return BestDecl;
}
unsigned getBestDeclIndex() const {
assert(getBestDecl());
return BestIndex;
}
};
void SimpleTypoCorrector::addDecl(const NamedDecl *ND) {
unsigned CurrIndex = NextIndex++;
const IdentifierInfo *II = ND->getIdentifier();
if (!II)
return;
StringRef Name = II->getName();
unsigned MinPossibleEditDistance = abs((int)Name.size() - (int)Typo.size());
if (MinPossibleEditDistance > 0 &&
Typo.size() / MinPossibleEditDistance < 3)
return;
unsigned EditDistance = Typo.edit_distance(Name, true, MaxEditDistance);
if (EditDistance < BestEditDistance) {
BestEditDistance = EditDistance;
BestDecl = ND;
BestIndex = CurrIndex;
}
}
} // unnamed namespace
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
unsigned Sema::correctTypoInParmVarReference(
StringRef Typo,
Comment Sema: refactor handling of 'ParmVarDecl's and save them in Sema members. llvm-svn: 160634
2012-07-24 01:40:30 +08:00
ArrayRef<const ParmVarDecl *> ParamVars) {
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
SimpleTypoCorrector Corrector(Typo);
for (unsigned i = 0, e = ParamVars.size(); i != e; ++i)
Corrector.addDecl(ParamVars[i]);
if (Corrector.getBestDecl())
return Corrector.getBestDeclIndex();
else
Remove redundant semicolons which are null statements. llvm-svn: 163546
2012-09-11 05:20:09 +08:00
return ParamCommandComment::InvalidParamIndex;
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
}
namespace {
bool ResolveTParamReferenceHelper(
StringRef Name,
const TemplateParameterList *TemplateParameters,
SmallVectorImpl<unsigned> *Position) {
for (unsigned i = 0, e = TemplateParameters->size(); i != e; ++i) {
const NamedDecl *Param = TemplateParameters->getParam(i);
const IdentifierInfo *II = Param->getIdentifier();
if (II && II->getName() == Name) {
Position->push_back(i);
return true;
}
if (const TemplateTemplateParmDecl *TTP =
dyn_cast<TemplateTemplateParmDecl>(Param)) {
Position->push_back(i);
if (ResolveTParamReferenceHelper(Name, TTP->getTemplateParameters(),
Position))
return true;
Position->pop_back();
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
}
}
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
return false;
}
} // unnamed namespace
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
Comment parsing: add support for \tparam command on all levels. The only caveat is renumbering CXCommentKind enum for aesthetic reasons -- this breaks libclang binary compatibility, but should not be a problem since API is so new. This also fixes PR13372 as a side-effect. llvm-svn: 161087
2012-08-01 06:37:06 +08:00
bool Sema::resolveTParamReference(
StringRef Name,
const TemplateParameterList *TemplateParameters,
SmallVectorImpl<unsigned> *Position) {
Position->clear();
if (!TemplateParameters)
return false;
return ResolveTParamReferenceHelper(Name, TemplateParameters, Position);
}
namespace {
void CorrectTypoInTParamReferenceHelper(
const TemplateParameterList *TemplateParameters,
SimpleTypoCorrector &Corrector) {
for (unsigned i = 0, e = TemplateParameters->size(); i != e; ++i) {
const NamedDecl *Param = TemplateParameters->getParam(i);
Corrector.addDecl(Param);
if (const TemplateTemplateParmDecl *TTP =
dyn_cast<TemplateTemplateParmDecl>(Param))
CorrectTypoInTParamReferenceHelper(TTP->getTemplateParameters(),
Corrector);
}
}
} // unnamed namespace
StringRef Sema::correctTypoInTParamReference(
StringRef Typo,
const TemplateParameterList *TemplateParameters) {
SimpleTypoCorrector Corrector(Typo);
CorrectTypoInTParamReferenceHelper(TemplateParameters, Corrector);
if (const NamedDecl *ND = Corrector.getBestDecl()) {
const IdentifierInfo *II = ND->getIdentifier();
assert(II && "SimpleTypoCorrector should not return this decl");
return II->getName();
}
return StringRef();
Enable comment parsing and semantic analysis to emit diagnostics. A few diagnostics implemented -- see testcases. I created a new TableGen file for comment diagnostics, DiagnosticCommentKinds.td, because comment diagnostics don't logically fit into AST diagnostics file. But I don't feel strongly about it. This also implements support for self-closing HTML tags in comment lexer and parser (for example, <br />). In order to issue precise diagnostics CommentSema needs to know the declaration the comment is attached to. There is no easy way to find a decl by comment, so we match comments and decls in lockstep: after parsing one declgroup we check if we have any new, not yet attached comments. If we do -- then we do the usual comment-finding process. It is interesting that this automatically handles trailing comments. We pick up not only comments that precede the declaration, but also comments that *follow* the declaration -- thanks to the lookahead in the lexer: after parsing the declgroup we've consumed the semicolon and looked ahead through comments. Added -Wdocumentation-html flag for semantic HTML errors to allow the user to disable only HTML warnings (but not HTML parse errors, which we emit as warnings in -Wdocumentation). llvm-svn: 160078
2012-07-12 05:38:39 +08:00
}
Comment AST: add InlineContentComment::RenderKind to specify a default rendering mode for clients that don't want to interpret Doxygen commands. Also add a libclang API to query this information. llvm-svn: 160633
2012-07-24 00:43:01 +08:00
InlineCommandComment::RenderKind
Sema::getInlineCommandRenderKind(StringRef Name) const {
Comment AST: TableGen'ize all command lists in CommentCommandTraits.cpp. Now we have a list of all commands. This is a good thing in itself, but it also enables us to easily implement typo correction for command names. With this change we have objects that contain information about each command, so it makes sense to resolve command name just once during lexing (currently we store command names as strings and do a linear search every time some property value is needed). Thus comment token and AST nodes were changed to contain a command ID -- index into a tables of builtin and registered commands. Unknown commands are registered during parsing and thus are also uniformly assigned an ID. Using an ID instead of a StringRef is also a nice memory optimization since ID is a small integer that fits into a common bitfield in Comment class. This change implies that to get any information about a command (even a command name) we need a CommandTraits object to resolve the command ID to CommandInfo*. Currently a fresh temporary CommandTraits object is created whenever it is needed since it does not have any state. But with this change it has state -- new commands can be registered, so a CommandTraits object was added to ASTContext. Also, in libclang CXComment has to be expanded to include a CXTranslationUnit so that all functions working on comment AST nodes can get a CommandTraits object. This breaks binary compatibility of CXComment APIs. Now clang_FullComment_getAsXML(CXTranslationUnit TU, CXComment CXC) doesn't need TU parameter anymore, so it was removed. This is a source-incompatible change for this C API. llvm-svn: 163540
2012-09-11 04:32:42 +08:00
assert(Traits.getCommandInfo(Name)->IsInlineCommand);
Comment AST: add InlineContentComment::RenderKind to specify a default rendering mode for clients that don't want to interpret Doxygen commands. Also add a libclang API to query this information. llvm-svn: 160633
2012-07-24 00:43:01 +08:00
return llvm::StringSwitch<InlineCommandComment::RenderKind>(Name)
.Case("b", InlineCommandComment::RenderBold)
.Cases("c", "p", InlineCommandComment::RenderMonospaced)
.Cases("a", "e", "em", InlineCommandComment::RenderEmphasized)
.Default(InlineCommandComment::RenderNormal);
}
Implement AST classes for comments, a real parser for Doxygen comments and a very simple semantic analysis that just builds the AST; minor changes for lexer to pick up source locations I didn't think about before. Comments AST is modelled along the ideas of HTML AST: block and inline content. * Block content is a paragraph or a command that has a paragraph as an argument or verbatim command. * Inline content is placed within some block. Inline content includes plain text, inline commands and HTML as tag soup. llvm-svn: 159790
2012-07-06 08:28:32 +08:00
} // end namespace comments
} // end namespace clang