| Age | Commit message (Collapse) | Author |
|---|
|
Not quite fully functional yet. The window positioning still needs to be written to prefs.cfg.
|
|
# Conflicts:
# res/lang/eo.bin
# res/lang/ie.bin
|
|
Borrowed Emoji glyphs are laid out double-wide in monospace runs, because they most likely are wider than the monospace advance.
|
|
The fontpack documentation says that when a glyph isn't found, all available fonts are checked in priority order. However, the implementation didn't actually do this.
Now there is a separate priority mapping for loaded fonts so they can be quickly scanned for any missing glyphs in the right order.
"iosevka-body" was prioritized higher because it provides a number of UI icons.
|
|
# Conflicts:
# CMakeLists.txt
# res/fonts/SmolEmoji-Regular.ttf
# res/lang/gl.bin
# src/app.c
|
|
|
|
Brighten dark colors on a (nearly) black background.
On light backgrounds, reduce the darkening as it was a bit too strong; more of the color is now shown.
|
|
Added the missing BG escapes 100-107, and fixed how the FG is adjusted to keep text legible. Previously it was not considering the actual BG color being applied to a text run.
|
|
|
|
IssueID #435
|
|
When drawing glyphs, each one of them may be individually cached. Allocate less heap memory during these calls.
|
|
Improve mixed-language word wrapping, and position U+3001 and U+3002 near the baseline.
IssueID #380
|
|
IssueID #345
|
|
Tell HarfBuzz when Devanagari has been detected.
TODO: Check for other languages supported by HarfBuzz (there's a long list).
IssueID #345
|
|
SGR codes 2 and 10.
|
|
|
|
|
|
Other formatting besides line breaks is ignored for now, although attributed strings could be used here.
|
|
|
|
|
|
Skip the font download suggestion, and fixed an uninitialized value.
|
|
Fixed issue with very short text not being truncated at all. A short truncated label will now just show the icon, if one is set.
The tab close buttons don't appear if the buttons are too small.
|
|
The background color of the current text run can be changed via ANSI SGR control sequences.
Improved the ANSI escape sequence regex to detect more than just SGR sequences, so they can be properly filtered out.
|
|
# Conflicts:
# CMakeLists.txt
# Depends.cmake
# src/lang.c
|
|
If some fonts have the same human-readable name, the unique font ID will be added to show which font it actually is.
|
|
# Conflicts:
# src/lang.c
|
|
IssueID #372
|
|
Resources are now stored in a ZIP archive, and it comes with an explicit version number. All the old embed/bincat stuff was removed as unnecessary.
resources.lgr is also a valid fontpack, for loading the built-in fonts.
IssueID #363
|
|
|
|
|
|
Warn the user about missing glyphs and potentially unsupported ANSI escapes.
TODO: Site-specific setting for dismissed warning; fonts preference about missing glyph warnings.
|
|
|
|
FG color and font style escapes can be enabled separately. FG color remains enabled by default like before.
|
|
When GlyphBuffers were freed, the array holding them wasn't.
|
|
When the last safe break position was not in the current attributed run, the calculated wrap advance came out incorrect. This was possible when the first glyph in an attributed run didn't fit.
|
|
It can be argued that using ANSI escapes to style text goes against the intended use of Gemtext. Therefore, all ANSI escapes are now disabled by default in Gemtext.
|
|
These changes concern the situation when the attributes of text (i.e., font, color) are changed via escape sequences.
The concept of "base attributes" was added so that the low-level text renderer knows which font/color to set when a "reset" escape sequence is encountered. This depends on what kind of text is being renderer, e.g., preformatted or regular paragraphs.
The base attributes were added as variables in Text because it was getting unwieldy to pass all the information via the draw/measure/WrapText functions.
GmDocument now has a GmTheme struct that collects the font and color information into a single place.
|
|
|
|
Convert Markdown to Gemtext and use ANSI escape sequences to switch fonts (bold, italic, monospace). The conversion is still a bit buggy...
|
|
Now all the available fonts are selectable in the dialog. Added preformatted and UI font settings and shuffled the order a bit to make more sense.
|
|
Added a second version of Iosevka with more line spacing, to be used as the default monospace document font.
|
|
The idea is that the amount of content should be the same per line regardless of the font.
|
|
Check auxiliary fonts first, and then all primary fonts that haven't been checked yet.
|
|
Scaling of monospace and symbols; cache reset.
|
|
The built-in fonts are loaded via FontPack, and the font table is now constructed dynamically based on available fonts.
A full set of variants (style, size) are prepared for each font, but some of the data gets allocated lazily when needed.
GmRun needed a larger allocation for fonts, so now all the fields are combined into a single bit field.
TODO: Glyph scaling, vertical offsets, and symbol lookup are still not fully working.
|
|
|
|
|
|
The 80% size is used in the toolbar.
|
|
|
|
|
