Skip to content

Commit

Permalink
Titlecasing renames (#3843)
Browse files Browse the repository at this point in the history
  • Loading branch information
@Manishearth
Manishearth authored Aug 11, 2023
1 parent 0c1b106 commit 57b9ea4
Show file tree
Hide file tree
Showing 42 changed files with 479 additions and 549 deletions.
2 changes: 1 addition & 1 deletion components/casemap/benches/casemap.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,7 @@ fn overview_bench(c: &mut Criterion) {
c.bench_function("icu_casemap/titlecase_segment", |b| {
b.iter(|| {
for s in TEST_STRING_EN.split(' ') {
black_box(casemapper.titlecase_segment_legacy_to_string(
black_box(casemapper.titlecase_segment_with_only_case_data_to_string(
black_box(s),
&root,
Default::default(),
Expand Down
111 changes: 62 additions & 49 deletions components/casemap/src/casemapper.rs
View file

Large diffs are not rendered by default.

10 changes: 5 additions & 5 deletions components/casemap/src/internals.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,7 @@ use crate::provider::data::{DotType, MappingKind};
use crate::provider::exception_helpers::ExceptionSlot;
use crate::provider::{CaseMapUnfoldV1, CaseMapV1};
use crate::set::ClosureSink;
use crate::titlecase::TailCasing;
use crate::titlecase::TrailingCase;
use core::fmt;
use icu_locid::LanguageIdentifier;
use writeable::Writeable;
Expand Down Expand Up @@ -54,7 +54,7 @@ pub(crate) struct FullCaseWriteable<'a, const IS_TITLE_CONTEXT: bool> {
src: &'a str,
locale: CaseMapLocale,
mapping: MappingKind,
titlecase_tail_casing: TailCasing,
titlecase_tail_casing: TrailingCase,
}

impl<'a, const IS_TITLE_CONTEXT: bool> Writeable for FullCaseWriteable<'a, IS_TITLE_CONTEXT> {
Expand All @@ -68,15 +68,15 @@ impl<'a, const IS_TITLE_CONTEXT: bool> Writeable for FullCaseWriteable<'a, IS_TI
self.data
.full_helper::<IS_TITLE_CONTEXT, W>(c, context, self.locale, mapping, sink)?;
if IS_TITLE_CONTEXT {
if self.titlecase_tail_casing == TailCasing::Lowercase {
if self.titlecase_tail_casing == TrailingCase::Lower {
mapping = MappingKind::Lower;
} else {
break;
}
}
}
// Write the rest of the string
if IS_TITLE_CONTEXT && self.titlecase_tail_casing == TailCasing::PreserveCase {
if IS_TITLE_CONTEXT && self.titlecase_tail_casing == TrailingCase::Unchanged {
sink.write_str(iter.as_str())?;
}
Ok(())
Expand Down Expand Up @@ -516,7 +516,7 @@ impl<'data> CaseMapV1<'data> {
src: &'a str,
locale: CaseMapLocale,
mapping: MappingKind,
titlecase_tail_casing: TailCasing,
titlecase_tail_casing: TrailingCase,
) -> FullCaseWriteable<'a, IS_TITLE_CONTEXT> {
// Ensure that they are either both true or both false, i.e. an XNOR operation
debug_assert!(!(IS_TITLE_CONTEXT ^ (mapping == MappingKind::Title)));
Expand Down
233 changes: 110 additions & 123 deletions components/casemap/src/titlecase.rs
View file

Large diffs are not rendered by default.

62 changes: 31 additions & 31 deletions components/casemap/tests/conversions.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -209,27 +209,27 @@ fn test_armenian() {
let ew = "և";
let yerevan = "Երևանի";
assert_eq!(
cm.titlecase_segment_legacy_to_string(ew, &root, default_options),
cm.titlecase_segment_with_only_case_data_to_string(ew, &root, default_options),
"Եւ"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string(yerevan, &root, default_options),
cm.titlecase_segment_with_only_case_data_to_string(yerevan, &root, default_options),
"Երևանի"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string(ew, &east, default_options),
cm.titlecase_segment_with_only_case_data_to_string(ew, &east, default_options),
"Եվ"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string(yerevan, &east, default_options),
cm.titlecase_segment_with_only_case_data_to_string(yerevan, &east, default_options),
"Երևանի"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string(ew, &west, default_options),
cm.titlecase_segment_with_only_case_data_to_string(ew, &west, default_options),
"Եւ"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string(yerevan, &west, default_options),
cm.titlecase_segment_with_only_case_data_to_string(yerevan, &west, default_options),
"Երևանի"
);
}
Expand All @@ -241,105 +241,105 @@ fn test_dutch() {
let default_options = Default::default();

assert_eq!(
cm.titlecase_segment_legacy_to_string("ijssel", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("ijssel", &nl, default_options),
"IJssel"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("igloo", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("igloo", &nl, default_options),
"Igloo"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("IJMUIDEN", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("IJMUIDEN", &nl, default_options),
"IJmuiden"
);

assert_eq!(
cm.titlecase_segment_legacy_to_string("ij", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("ij", &nl, default_options),
"IJ"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("IJ", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("IJ", &nl, default_options),
"IJ"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íj́", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íj́", &nl, default_options),
"ÍJ́"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("ÍJ́", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("ÍJ́", &nl, default_options),
"ÍJ́"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íJ́", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íJ́", &nl, default_options),
"ÍJ́"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("Ij́", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("Ij́", &nl, default_options),
"Ij́"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("ij́", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("ij́", &nl, default_options),
"Ij́"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("ïj́", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("ïj́", &nl, default_options),
"Ïj́"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íj\u{0308}", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íj\u{0308}", &nl, default_options),
"Íj\u{0308}"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íj́\u{1D16E}", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íj́\u{1D16E}", &nl, default_options),
"Íj́\u{1D16E}"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íj\u{1ABE}", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íj\u{1ABE}", &nl, default_options),
"Íj\u{1ABE}"
);

assert_eq!(
cm.titlecase_segment_legacy_to_string("ijabc", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("ijabc", &nl, default_options),
"IJabc"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("IJabc", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("IJabc", &nl, default_options),
"IJabc"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íj́abc", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íj́abc", &nl, default_options),
"ÍJ́abc"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("ÍJ́abc", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("ÍJ́abc", &nl, default_options),
"ÍJ́abc"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íJ́abc", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íJ́abc", &nl, default_options),
"ÍJ́abc"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("Ij́abc", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("Ij́abc", &nl, default_options),
"Ij́abc"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("ij́abc", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("ij́abc", &nl, default_options),
"Ij́abc"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("ïj́abc", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("ïj́abc", &nl, default_options),
"Ïj́abc"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íjabc\u{0308}", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íjabc\u{0308}", &nl, default_options),
"Íjabc\u{0308}"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íj́abc\u{1D16E}", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íj́abc\u{1D16E}", &nl, default_options),
"ÍJ́abc\u{1D16E}"
);
assert_eq!(
cm.titlecase_segment_legacy_to_string("íjabc\u{1ABE}", &nl, default_options),
cm.titlecase_segment_with_only_case_data_to_string("íjabc\u{1ABE}", &nl, default_options),
"Íjabc\u{1ABE}"
);
}
Expand Down
2 changes: 1 addition & 1 deletion ffi/diplomat/c/include/ICU4XCaseMapper.h
View file

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

15 changes: 8 additions & 7 deletions ffi/diplomat/c/include/ICU4XHeadAdjustment.h → ...plomat/c/include/ICU4XLeadingAdjustment.h
View file

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

2 changes: 0 additions & 2 deletions ffi/diplomat/c/include/ICU4XTitlecaseMapper.h
View file

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

12 changes: 6 additions & 6 deletions ffi/diplomat/c/include/ICU4XTitlecaseOptionsV1.h
View file

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

14 changes: 7 additions & 7 deletions ffi/diplomat/c/include/ICU4XTailCasing.h → ffi/diplomat/c/include/ICU4XTrailingCase.h
View file

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

Loading

0 comments on commit 57b9ea4

Please sign in to comment.