konevsa/typst
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Rename is_cjk_script() to is_cj_script() (#2829)

Browse Source
This commit is contained in:
Leedehai 2023-12-05 13:04:38 -05:00 committed by GitHub
parent 8a69d3bfb6
commit 058f24f95e
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 22 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
crates/typst/src/layout/inline/mod.rs
View File

@ -7,7 +7,7 @@ use unicode_script::{Script, UnicodeScript};
use self::linebreak::{breakpoints, Breakpoint};
use self::shaping::{
is_gb_style, is_of_cjk_script, shape, ShapedGlyph, ShapedText, BEGIN_PUNCT_PAT,
is_gb_style, is_of_cj_script, shape, ShapedGlyph, ShapedText, BEGIN_PUNCT_PAT,
END_PUNCT_PAT,
};
use crate::diag::{bail, SourceResult};
@ -639,16 +639,16 @@ fn add_cjk_latin_spacing(items: &mut [Item]) {
.and_then(|shaped| shaped.glyphs.first())
});
// Case 1: CJK followed by a Latin character
if glyph.is_cjk_script() && next.map_or(false, |g| g.is_letter_or_number()) {
// Case 1: CJ followed by a Latin character
if glyph.is_cj_script() && next.map_or(false, |g| g.is_letter_or_number()) {
// The spacing is default to 1/4 em, and can be shrunk to 1/8 em.
glyph.x_advance += Em::new(0.25);
glyph.adjustability.shrinkability.1 += Em::new(0.125);
text.width += Em::new(0.25).at(text.size);
}
// Case 2: Latin followed by a CJK character
if glyph.is_cjk_script() && prev.map_or(false, |g| g.is_letter_or_number()) {
// Case 2: Latin followed by a CJ character
if glyph.is_cj_script() && prev.map_or(false, |g| g.is_letter_or_number()) {
glyph.x_advance += Em::new(0.25);
glyph.x_offset += Em::new(0.25);
glyph.adjustability.shrinkability.0 += Em::new(0.125);
@ -1028,7 +1028,7 @@ fn line<'a>(
// Deal with CJK punctuation at line ends.
let gb_style = is_gb_style(shaped.lang, shaped.region);
let maybe_adjust_last_glyph = trimmed.ends_with(END_PUNCT_PAT)
|| (p.cjk_latin_spacing && trimmed.ends_with(is_of_cjk_script));
|| (p.cjk_latin_spacing && trimmed.ends_with(is_of_cj_script));
// Usually, we don't want to shape an empty string because:
// - We don't want the height of trimmed whitespace in a different
@ -1056,7 +1056,7 @@ fn line<'a>(
punct.shrink_right(shrink_amount);
reshaped.width -= shrink_amount.at(reshaped.size);
} else if p.cjk_latin_spacing
&& last_glyph.is_cjk_script()
&& last_glyph.is_cj_script()
&& (last_glyph.x_advance - last_glyph.x_offset) > Em::one()
{
// If the last glyph is a CJK character adjusted by [`add_cjk_latin_spacing`],
@ -1078,10 +1078,10 @@ fn line<'a>(
}
}
// Deal with CJK characters at line starts.
// Deal with CJ characters at line starts.
let text = &p.bidi.text[range.start..end];
let maybe_adjust_first_glyph = text.starts_with(BEGIN_PUNCT_PAT)
|| (p.cjk_latin_spacing && text.starts_with(is_of_cjk_script));
|| (p.cjk_latin_spacing && text.starts_with(is_of_cj_script));
// Reshape the start item if it's split in half.
let mut first = None;
@ -1116,7 +1116,7 @@ fn line<'a>(
reshaped.width -= amount_abs;
width -= amount_abs;
} else if p.cjk_latin_spacing
&& first_glyph.is_cjk_script()
&& first_glyph.is_cj_script()
&& first_glyph.x_offset > Em::zero()
{
// If the first glyph is a CJK character adjusted by [`add_cjk_latin_spacing`],

22
crates/typst/src/layout/inline/shaping.rs
View File

@ -107,9 +107,9 @@ impl ShapedGlyph {
self.is_justifiable
}
/// Whether the glyph is part of a CJK script.
pub fn is_cjk_script(&self) -> bool {
is_cjk_script(self.c, self.script)
/// Whether the glyph is part of Chinese or Japanese script (i.e. CJ, not CJK).
pub fn is_cj_script(&self) -> bool {
is_cj_script(self.c, self.script)
}
pub fn is_cjk_punctuation(&self) -> bool {
@ -360,7 +360,7 @@ impl<'a> ShapedText<'a> {
pub fn cjk_justifiable_at_last(&self) -> bool {
self.glyphs
.last()
.map(|g| g.is_cjk_script() || g.is_cjk_punctuation())
.map(|g| g.is_cj_script() || g.is_cjk_punctuation())
.unwrap_or(false)
}
@ -934,15 +934,17 @@ fn is_space(c: char) -> bool {
matches!(c, ' ' | '\u{00A0}' | '　')
}
/// Whether the glyph is part of a CJK script.
/// Whether the glyph is part of Chinese or Japanese script (i.e. CJ, not CJK).
#[inline]
pub(super) fn is_of_cjk_script(c: char) -> bool {
is_cjk_script(c, c.script())
pub(super) fn is_of_cj_script(c: char) -> bool {
is_cj_script(c, c.script())
}
/// Whether the glyph is part of a CJK script.
/// Whether the glyph is part of Chinese or Japanese script (i.e. CJ, not CJK).
/// The function is dedicated to typesetting Chinese or Japanese, which do not
/// have spaces between words, so K is not checked here.
#[inline]
fn is_cjk_script(c: char, script: Script) -> bool {
fn is_cj_script(c: char, script: Script) -> bool {
use Script::*;
// U+30FC: Katakana-Hiragana Prolonged Sound Mark
matches!(script, Hiragana | Katakana | Han) || c == '\u{30FC}'
@ -1016,7 +1018,7 @@ fn is_justifiable(
) -> bool {
// GB style is not relevant here.
is_space(c)
|| is_cjk_script(c, script)
|| is_cj_script(c, script)
|| is_cjk_left_aligned_punctuation(c, x_advance, stretchability, true)
|| is_cjk_right_aligned_punctuation(c, x_advance, stretchability)
|| is_cjk_center_aligned_punctuation(c, true)