-
Notifications
You must be signed in to change notification settings - Fork 35
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
This commit analyzes script runs based on Unicode data. It also starts using a "LayoutSession" type, which will eventually support queries of substrings. Access to the layout is done through iterators, which will improve flexibility. These iterators don't allocate. In addition, the iterator now breaks out a run of glyphs of the same font, which will very likely improve performance downstream. It's still work in progress, but I wanted to checkpoint. Progress towards #4
- Loading branch information
Showing
5 changed files
with
298 additions
and
11 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,163 @@ | ||
| //! Retained layout that supports substring queries. | ||
|
|
||
| use harfbuzz::sys::{hb_script_t, HB_SCRIPT_COMMON, HB_SCRIPT_INHERITED, HB_SCRIPT_UNKNOWN}; | ||
|
|
||
| use euclid::Vector2D; | ||
|
|
||
| use crate::hb_layout::{layout_fragment, HbFace}; | ||
| use crate::unicode_funcs::lookup_script; | ||
| use crate::{FontCollection, FontRef, Glyph, TextStyle}; | ||
|
|
||
| pub struct LayoutSession<'a> { | ||
| text: &'a str, | ||
| fragments: Vec<LayoutFragment>, | ||
| } | ||
|
|
||
| pub(crate) struct LayoutFragment { | ||
| // Length of substring covered by this fragment. | ||
| pub(crate) substr_len: usize, | ||
| pub(crate) script: hb_script_t, | ||
| pub(crate) advance: Vector2D<f32>, | ||
| pub(crate) glyphs: Vec<Glyph>, | ||
| pub(crate) hb_face: HbFace, | ||
| pub(crate) font: FontRef, | ||
| } | ||
|
|
||
| pub struct LayoutRangeIter<'a> { | ||
| // This probably wants to be a mut ref so we can stash resources in the session. | ||
| session: &'a LayoutSession<'a>, | ||
| offset: Vector2D<f32>, | ||
| fragment_ix: usize, | ||
| } | ||
|
|
||
| pub struct LayoutRun<'a> { | ||
| // This should potentially be in fragment (would make it easier to binary search) | ||
| offset: Vector2D<f32>, | ||
| fragment: &'a LayoutFragment, | ||
| } | ||
|
|
||
| pub struct RunIter<'a> { | ||
| offset: Vector2D<f32>, | ||
| fragment: &'a LayoutFragment, | ||
| glyph_ix: usize, | ||
| } | ||
|
|
||
| pub struct GlyphInfo { | ||
| pub glyph_id: u32, | ||
| pub offset: Vector2D<f32>, | ||
| } | ||
|
|
||
| impl<'a> LayoutSession<'a> { | ||
| pub fn create( | ||
| text: &'a str, | ||
| style: &TextStyle, | ||
| collection: &FontCollection, | ||
| ) -> LayoutSession<'a> { | ||
| let mut i = 0; | ||
| let mut fragments = Vec::new(); | ||
| while i < text.len() { | ||
| let (script, script_len) = get_script_run(&text[i..]); | ||
| let script_substr = &text[i..i + script_len]; | ||
| for (range, font) in collection.itemize(script_substr) { | ||
| let fragment = layout_fragment(style, font, script, &script_substr[range]); | ||
| fragments.push(fragment); | ||
| } | ||
| i += script_len; | ||
| } | ||
| LayoutSession { text, fragments } | ||
| } | ||
|
|
||
| pub fn iter_all(&self) -> LayoutRangeIter { | ||
| LayoutRangeIter { | ||
| offset: Vector2D::zero(), | ||
| session: &self, | ||
| fragment_ix: 0, | ||
| } | ||
| } | ||
|
|
||
| // TODO: similar function as iter_all but takes a range (maybe subsumes iter_all, as | ||
| // it has the same behavior with [0..text.len()]). | ||
| } | ||
|
|
||
| impl<'a> Iterator for LayoutRangeIter<'a> { | ||
| type Item = LayoutRun<'a>; | ||
|
|
||
| fn next(&mut self) -> Option<LayoutRun<'a>> { | ||
| if self.fragment_ix == self.session.fragments.len() { | ||
| None | ||
| } else { | ||
| let fragment = &self.session.fragments[self.fragment_ix]; | ||
| self.fragment_ix += 1; | ||
| let offset = self.offset; | ||
| self.offset += fragment.advance; | ||
| Some(LayoutRun { offset, fragment }) | ||
| } | ||
| } | ||
| } | ||
|
|
||
| impl<'a> LayoutRun<'a> { | ||
| pub fn font(&self) -> &FontRef { | ||
| &self.fragment.font | ||
| } | ||
|
|
||
| pub fn glyphs(&self) -> RunIter<'a> { | ||
| RunIter { | ||
| offset: self.offset, | ||
| fragment: self.fragment, | ||
| glyph_ix: 0, | ||
| } | ||
| } | ||
| } | ||
|
|
||
| impl<'a> Iterator for RunIter<'a> { | ||
| type Item = GlyphInfo; | ||
|
|
||
| fn next(&mut self) -> Option<GlyphInfo> { | ||
| if self.glyph_ix == self.fragment.glyphs.len() { | ||
| None | ||
| } else { | ||
| let glyph = &self.fragment.glyphs[self.glyph_ix]; | ||
| self.glyph_ix += 1; | ||
| Some(GlyphInfo { | ||
| glyph_id: glyph.glyph_id, | ||
| offset: self.offset + glyph.offset, | ||
| }) | ||
| } | ||
| } | ||
| } | ||
|
|
||
| /// Figure out the script for the initial part of the buffer, and also | ||
| /// return the length of the run where that script is valid. | ||
| pub(crate) fn get_script_run(text: &str) -> (hb_script_t, usize) { | ||
| let mut char_iter = text.chars(); | ||
| if let Some(cp) = char_iter.next() { | ||
| let mut current_script = lookup_script(cp.into()); | ||
| let mut len = cp.len_utf8(); | ||
| while let Some(cp) = char_iter.next() { | ||
| let script = lookup_script(cp.into()); | ||
| if script != current_script { | ||
| if current_script == HB_SCRIPT_INHERITED || current_script == HB_SCRIPT_COMMON { | ||
| current_script = script; | ||
| } else if script != HB_SCRIPT_INHERITED && script != HB_SCRIPT_COMMON { | ||
| break; | ||
| } | ||
| } | ||
| len += cp.len_utf8(); | ||
| } | ||
| if current_script == HB_SCRIPT_INHERITED { | ||
| current_script = HB_SCRIPT_COMMON; | ||
| } | ||
| (current_script, len) | ||
| } else { | ||
| (HB_SCRIPT_UNKNOWN, 0) | ||
| } | ||
| } | ||
|
|
||
| fn debug_script_runs(text: &str) { | ||
| let mut text_substr = text; | ||
| while !text_substr.is_empty() { | ||
| let (script, len) = get_script_run(text_substr); | ||
| println!("text {:?} script {:x}", &text_substr[..len], script); | ||
| text_substr = &text_substr[len..]; | ||
| } | ||
| } |