Users can now synchronize the content of the pattern editor with a file
on disk when exporting a pattern. Any modifications made in the pattern
editor are automatically saved to the file and when the pattern editor
gains focus it will automatically load any changes to the file on disk.
This feature is brand new and can have unforeseen problems so test it
fully before risking losing work.
Added major changes to how the text editor handles utf8 characters with
the added ability to obtain them using the [] operator and column
coordinates as indices. Sanitize coordinates was renamed to Set
Coordinates and can be used instead of the constructor to guarantee no
out of range asserts and/or exceptions. Improved move left and right
readability and cursor accuracy when clicking. Fixed single character
selection by double-clicking it. Deleted old and unused code and other
changes that I can't remember about atm.
These changes are part of an effort aimed at removing tabs from ImHex
that started some time ago. Here text preprocessing is removed from all
the places were it was done before and moved to the places where files
are read that go in the pattern editor with two notable exceptions.
1) Pattern import reads patterns in order to present a list that can be
filtered.That can safely ignore preprocessing since only needs to get
information needed to filter.
2) The pattern editor can incorporate text from the clipboard so that
needs to be preprocessed as well.
find/replace is unable to add tabs or carriage returns so this should
cover all angles.
Added clion-like bracket matching feature with shortcut to go to the
other one. Also improved some cumbersome repeated function call.
Added support for using negative indices in coordinates so -1 is the last
line or column, -2 the previous, etc..
Pattern library has fixes for column errors being incorrectly set for lines
containing tabs that are not replaced by 4 spaces.
allows the ability to assign colors to global placed and non-placed
variables, pattern, local and calculated pointer variables, template
arguments, function variables and arguments, etc etc etc. It
accomplishes this using the parser and the token sequence generated by
the lexer. It still uses the original colorizing code but the underlying
data holding the pattern has been updated to be easier to use and to
debug. The changes are too numerous to cite here.It is a big but
necessary step to bring the pattern editor to a somewhat useful state.
There may be one commit in the pattern language repo needed to be able
to run this code
The recent update that made importing patterns undoable had the side
effect of undoing tab changes as well. When working on a fix for that it
became clear that the undo/redo stacks were being shared by all
providers so that you could undo changes done in a separate file. The
problem was fixed by making the text editors (and the console editors as
well) to be per provider which gives better control on things like
breakpoints and selections and a more robust focusing system that
preserves the cursor positions and removes the need to click on a
pattern to start editing. There are a lot of changes, but they are
mostly all the uses of text editors being now coded differently to
manage the providers. File imports are still undoable, but switching
providers is not.
Further changes suggested by reviewer were implemented namely a mull provider was being used and there was a get function missing which prevented the use of the preferred syntax.
The recent changes to the text editor to fix the longest line length
problems broke the console horizontal scrollbar. The code that displays
the console editor was more complicated that it needed be, and it had
the bad side effect of resetting the cursor which prevented horizontal
scrolling. Adding a function that appends lines to the text editor fixes
all problems and makes the code clearer. To accommodate for strings
containing zeros, the code that inserts text was changed to print a '.'
when zeros are encountered thus keeping the line length the same.
Proof of concept for implementing subpixel processing in ImGui. This is
work in progress, and it is bound to have problems.
What it does:
1) Uses freetype own subpixel processing implementation to build a
32-bit color atlas for the default font only (no icons, no unifont) . 2)
Avoids pixel perfect font conversion when possible. 3) Self contained,
no ImGui source code changes.
4) Results in much improved legibility of fonts rendered on low dpi LCD
screens that use horizontal RGB pixel layouts (no BRG or OLED or CRT if
they even exist anymore)
What it doesn't:
1) Fancy class based interface. The code is barely the minimum needed to
show it can work. 2) Dual source color blending. That needs to be
implemented in shader code, so it needs to change ImGui source code
although minimally. This will result in some characters appearing dimmer
than others. Easily fixed with small fragment and vertex shaders. 3)
subpixel positioning. If characters are very thin they will look
colored, or they can be moved to improve legibility. 4) deal with
detection of fringe cases including rare pixel layouts, non LCD screens,
Mac-OS not handling subpixel rendering and any other deviation from the
standard LCD. 5) tries to be efficient in speed or memory use. Font
Atlases will be 4 times the size they were before, but there are no
noticeable delays in font loading in the examples I have tried.
Any comments and code improvements are welcome.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
The horizontal scroll bar length is set using the maximum line length across the input file. The original setup had the lengths of imported and included files added so changes are made to insure that only changes from the input file are taken. This required changes to the Pattern Language library so the library is updated to the latest version as well.
The function SetSelection() in the text editor is used as its name implies to set the part of text that will be shown highlighted as a selection. It has two parameter for the selection start and end.
Strangely it also uses a third argument to choose one of 3 selection modes (line, word and normal). This is strange because it seems that if one wanted to select a line one would simply choose the line start and end as the selection values.
Furthermore using selection mode creates a bug when the word boundaries are advanced twice during a single word left or right selection and a second bug when SetSelection is used to set the cursor when the window acquires focus breaking selections in the console editor.
This PR simply eliminates the extraneous argument to SetSelection() ensuring that any problems related to it are fixed once and for all improving code readability by removing the use of inconsistent argument values to cover for the functionality duplication design flaw.
### Problem description
This PR addresses two small issues regarding the cursor in the pattern
editor (TextEditor.cpp):
1. It was not possible to move the cursor to the start of a line, if it
contained leading white space characters. With my fix, the editor will
behave more like other code editors. Pressing Home once will jump to the
first non-whitespace character (as it did before), but pressing it again
will jump to column 0. Subsequent presses will alternate between the two
positions.
2. When expanding a selection with Ctrl+Shift+{Left/Right}, the new
selection position would skip an additional word, resulting in the
cursor landing "inside" the selection. This PR fixes this bug.
### Implementation description
1. To fix the first issue, I simply added a condition in
`TextEditor::MoveHome` to check if the cursor is already on the first
meaningful character, or on one of the whitespaces preceding it, in
which case the jump offset is set to 0. If we're already on column 0,
then jump forwards to the first non-whitespace character.
2. This bug was happening because the word boundary jump calculations
were essentially happening twice. Once in
`TextEditor::MoveLeft`/`TextEditor::MoveRight`, and then a second time
in `TextEditor::SetSelection`, leading to the selection skipping an
additional word. I fixed this by replacing the ternary operator with
just `SelectionMode::Normal`.
---------
Co-authored-by: paxcut <53811119+paxcut@users.noreply.github.com>
fixes Issue #1621.
Using shift-tab on an empty line caused a crash. Additionally, changed
the hard coded value of 4 to the tab size variable it really needs to be.
Now that line numbers are not part of the line of code clicking them
makes the text editor lose focus. This PR changes that by allowing the
user to toggle breakpoints by clicking the field where the line number
is located.
Not only will the text editor retain focus when breakpoints are set, but
if other parts of ImHex had focus, then it will be transferred to the
text editor's current cursor position when the line number field is
clicked.
It is also possible to keep the focus where it was and only retain the
focus if the text editor was focused when the break point is set. The
change is very trivial so if that is preferred I can easily switch it.
A user complained that they imported a file by accident when they meant
to export it and as a result had trouble recovering the changes they
were trying to save. Auto-save saved the day but there is no reason for
not being able to undo changes after importing a pattern.
In fact, the previous implementation treated importing a pattern as a
reset on the editor instance which actually erased all previous undo
entries. Importing now is treated as a normal editing operation where
the entire file is replaced with the imported pattern.
Since all imports use AddText it was easy to add an undo entry to that
function while removing the part where the previous undo records were
being deleted.
Care is taken to add the preprocessed version of the imported file to
the undo buffer so that unwanted chars don't sneak in. A bug was found
in the handling of a tab char as well but hopefully it wont need to be
used anymore.
The pattern editor was resetting the flag that marked text changes on
every frame, but resolving a text change may take more than one frame,
so instead I created a function that allows the resetting of the text
changed boolean from view pattern editor.
Currently, when a pattern is evaluated the pattern editor will lose
focus regardless of how evaluation was triggered. This is specially
annoying when using f5 or auto-evaluate. Now whenever a pattern is
evaluated, focus is given to the pattern editor when evaluation ends.
JumptoLine didn't work for empty lines.
Co-authored-by: Nik <werwolv98@gmail.com>
