This implied a lot of other core changes, which also pushed me into
improving some of the edit actions and PDB calls to be multi-layer aware
in the same time.
Note that it is still work-in-progress, but I just had to commit
something in an acceptable intermediate state otherwise I was just going
crazy.
In particular now the various transform tools are multi-layer aware and
work simultaneously on all selected layers (and the linked layers if any
of the selected layers is linked too). Both preview and final transform
processing works.
In the limitations, preview doesn't work well (only one layer in the
preview) when there is a selection (though the actual transform works).
Also I am left to wonder how we should process this case of canvas
selection+transform on multi-layers. Indeed currently I am just creating
a floating selection (like we used to for the selection+transform case)
containing a transform result of the composited version of all selected
layers. This is a possible expected result, but another could be to get
several transformed layers (without composition). But then should the
"Floating Selection" concept allow for multiple Floating Selections?
Sooo many questions left to answer.
In GimpDrawTool, update the tool widget on GIMP_TOOL_ACTION_RESUME,
so that it can respond to changes in the display-shell scale/
offset. We'd previously done that for individual tools/widgets,
but let's just do it in one place.
In the transform tools, when transforming the entire image in show-
all mode, don't crop the preview and the transform bounds to the
size of the canvas.
... which temporarily changes the transform-type of the tool,
restoring the original type once it's halted. To be used when
activating a transform tool through an action for a specific
transform type.
Add a new "Image" transform type to the transform tools, in
addition to the existing "Layer", "Selection", and "Path" transform
types. The "Image" mode transforms the entire image, rather than a
single item. In tools with a preview, the preview shows the
transformed image projection. The clipping mode controls how the
canvas is clipped during the transform; in particular, the ADJUST
clipping mode resizes the canvas to fit its transformed size.
Remove the special clipping-mode handling for channels throughout
the transform (and drawable-filter) code, and rather use
gimp_item_get_clip(), added in the previous commit, instead. As
mentioned in the previous commit, we only modify the clipping mode
in top-level code, while having lower-level code use the clipping
mode as-is. This not only hides the actual clipping-mode logic
from the transform code, but, in particular, allows code performing
transformation internally to use arbitrary clipping modes.
Also, this commit fixes a bunch of PDB bugs all over the place :)
In GimpTransformGridTool, allow performing simultaneous forward
(normal) and backward (corrective) transforms, by having each
transform direction operate on an independent set of parameters.
In other words, whereas the transform-grid tools previously had a
single transform, which could be applied either normally or
correctively using the "direction" tool-option, they now have two
independent transforms, one applied normally and the other
applied correctively, which are toggled using the "direction"
option. The overall transform is the combination of the backward
transform, followed by the forward transform.
Another way to think about it, is that the tool transforms a source
shape into a destination shape. The source shape is defined by the
backward transform, and the destination shape is defined by the
forward transform. Wherewas previously only one of these shapes
could be controlled (the other shape always being the item bounds),
it's now possible to control both shapes in a single transform.
The next commit will allow modifying both shapes simultaneously,
making this even more useful.
Note that since both transforms start off as the identity, using
only one of the transform directions has the same behavior as
before.
Add an undo_desc field to GimpTransformToolClass, which subclasses
should set to the tool's default undo description. Provide a
default implementation for the get_undo_desc() vfunc, which returns
(a copy of) undo_desc. This simplifies transform tools that have a
static undo descrption, as well as provides a fallback when a
detailed undo description can't be generated (not currently
relevant, but will be used in the next commit).
In all tools, when the current item can't be edited due to its lock
mask, use gimp_tools_blink_lock_box(), added in the previous
commit,to blink the lock box of the corresponding dockable, in
addition to showing an error message in the status bar, to hint at
the source of the error.
In the transform tools, when there is no item of the selected type
to transform, blink the move-type box widget in the tool options,
in addition to showing an error message in the status bar, to hint
at the source of the error.
Replace GimpTransformTool's 'drawable' field with an 'item' field,
and have GimpTransformGridTool set it to the active item, to which
the transformation is applied, during its initialization. In
gimp_transform_tool_get_active_item(), return the value of the
transform tool's 'item' field, if not NULL, instead of the image's
active item. This makes sure we apply that transform-grid tools
apply the transformation for the item for which they were
activated, even if the image's active item has changed.
In GimpTransformTools, precalculate the resulting size of the
transformed item(s), and request confirmation if the size grows to
over 10 times the size of the image (in either dimension). This
protects against transformations that can result in suprprisingly
large items, such as inverted transformations, and, specifically,
perspective-correction transformations performed using the measure
tool, which will be added in the following commits.
In gimp_transform_tool_transform(), use "active_item", instead of
"tool->drawable", when cutting/pasting the selected portion of a
layer for transformation. The latter is a remnant of the old
transform-tool code, and is not guaranteed to be correspond to the
correct drawable, or even to a valid drawable (i.e., it can
potentially produce wrong results, or segfault.)
- remove gimp_widget_flush_expose()
- remove the "now" argument to gimp_display_shell_flush() and make it
only update widget states
- rename gimp_display_flush_whenever() to gimp_display_flush_update_region()
and call gimp_display_shell_flush() separately in the only case we
passed FALSE to flush_whenever()
- remove th flush_now interval logic from GimpDisplay, as soon as we
have exposed the canvas, we are in the loop for the next frame clock
tick anyway, so delaying a useless and removed process_updates serves
no purpose
- in gimptool-progress.c, create the invisible grab widget also for
non-cencelable cases, so we can always safely run the main loop
manually to make the progress updates visible
- in gimp-gegl-apply-operation.c, always run the main loop manually
to make the progress updates visible
- in gimpstatusbar.c, leave some FIXME comments as reminder that
we might need the same logic as in gimptool-progress.c
Last commit reintroduced this bug.
Allow transforming invisible layers using transform-grid tools, by
adding a 'drawable' member to GimpTransformTool, and setting/
clearing it when initializing/halting a GimpTransformTool. In
gimp_transform_tool_check_active_item(), skip the visibility check
if the active item equals the GimpTransformTool's 'drawable'
member.
Split gimp_transform_tool_get_active_item() into two functions:
gimp_transform_tool_get_active_item(), which returns the item
without checking for errors, and
gimp_transform_tool_check_active_item(), which returns the active
item while checking for errors. Adapt the rest of the code to the
change.
Remove the invisible_layer_ok parameter of
gimp_transform_tool_check_active_item(), and always return an error
when the active layer is invisible. This causes the flip and
measure tools to correctly reject invisible layers. Un-hide the
active item in GimpTransformGridTool before transforming, to avoid
rejecting layers that were hidden by the tool.
While most of our transform tools use an interactive transform
grid, and have similar behavior, the flip tool is an odd one out.
The new "auto straighten" function of the measure tool introduces
another tool that performs transformations, while not behaving like
the rest of the transform tools.
Factor out the parts of GimpTransformTool that handle user
interaction into GimpTransformGridTool (with corresponding
GimpTransformGridOptions, and GimpTransformGridToolUndo), and only
leave the basic transform functionality and options in
GimpTransformTool (and GimpTransformOptions).
Derive all the transform tools (and transform-tool base classes)
that previously derived from GimpTransformTool, from
GimpTransformGridTool. The one exception is GimpFlipTool, which
still derives from GimpTransformTool directly. The next commit
will derive GimpMeasureTool from GimpTransformTool as well.
... or during rotation.
If checked before rotation, it works as expected, i.e. one sees only the
original or the rotated image.
While rotation is in progress: if unchecked, one sees neither the
original nor the image preview; if checked, one sees both original and
rotated preview.
Let's make the behavior consistent and only show exactly one version at
all time.
A tool commit can be triggered in various cases, and the tool manager
relies on gimp_tool_has_display() to decide whether to run a tool
action. This function does much more than just checking GimpTool's
display. It also checks status_displays, and for a transform tool in
particular, it checks GimpDrawTool's display. This may be right for
other tools (I have no idea), so I can't just change this function.
Anyway we have to assume it is not a programming error if a transform
tool gets a COMMIT action while display is NULL (i.e. tool is halted).
When this happens, let's simply ignore.
This fixes the edge case raised by Ell, in comment 2 of bug 793150: when
an image has no layer, transform tools can't work and display is NULL.
But it still outputs status messages and therefore status_displays is
not empty. So the tool manager will still run a COMMIT action, which is
not an error. We only have to discard such COMMIT silently.
... which specifies whether the transform matrix is valid.
Subclasses can then set this member to indicate that the current
transformation is invalid (which can happen in the unified,
perspective, and handle transform tools). When the transform is
invalid, GimpTransformTool doesn't apply it upon committing, and
returns an error instead.
This commit doesn't set the transform_valid member in any of the
subclasses, so there's no effective change. The next commit adds a
GimpGenericTransformTool subclass, that will use the new member.
In GimpCanvasTransformPreview, use the transform matrix to
determine if we're doing a perspective transform, rather than
relying on a separate property, so that we don't use the slow
perspective path unnecessarily.
Consequently, remove the does_perspective member of
GimpTransformTool, since it's no longer used.
Change gimp_tool_set_active_modifier_state() to honor the new
GimpToolControlSetting. Explicitly set the mode to SEPARATE in
all tools that require modifier keys during a stroke.
And here comes the actual fix: change GimpTransformTool and
GimpToolTransformGrid to use SAME mode, and remove their
active_modifer_key() and hover_modifier() impls, so it makes no
difference whether a modifier is pressed before of after mouse button
press/release.
... the matrix is not actually used
Don't abort transform-tool commits when the transformation matrix
is the identity, for transform tools that don't calculate a
transformation matrix to begin with (i.e., the flip tool),
otherwise they do nothing.
and update the grid as soon as a constraint is changed, not only on
the next motion. Change GimpTransformTool to forward the events to the
widget if it exists, but still handle them if it doesn't (yes this
code duplication is ugly, but the widget can hardly handle events if
it doesn't exist...).
More than 2000 lines of code less in app/, instead of
if (instance->member)
{
g_object_unref/g_free/g_whatever (instance->member);
instance->member = NULL;
}
we now simply use
g_clear_object/pointer (&instance->member);
Call HALT generically in gimp_tool_control() after calling COMMIT, and
remove all hacks in tools that call both COMMIT and HALT or call
halt() from commit().
Some tools interact with their subclasses (e.g. filter tool and
operation tool), and it's essential that COMMIT runs through the
entire class hierarchy before HALT.
Probably breaks something, please test.
to ::can_undo() and ::can_redo(). They still return description
strings and the new naming is slightly off :) but get_undo_desc() will
be needed for something else soon, and half of the time the functions
are indeed used to check whether there are undo/redo staps at all.
We were leaking all tool widgets set with gimp_draw_tool_set_widget(),
and those having signal connections to e.g. the display shell were
doing things when they were supposed to be gone. Fixes make check.
and a default key_press() handler that emits CONFIRM, CANCEL and RESET
responses. Remove code with the same purpose from all subclasses.
Change tools feed key_press() to the widget and connect to its
"response" instead of implementing key_press() themselves. This will
only be better and less code after the tool side of this is done
generically.
Jehan