Magic-Mirror/gecko-dev
1
0
Fork 0
mirror of https://github.com/mozilla/gecko-dev.git synced 2024-11-23 12:51:06 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Bug 1911550 - Make text-transform:capitalize work even if no textrun is available, for <select> intrinsic sizing. r=layout-reviewers,emilio

Browse Source 
Differential Revision: https://phabricator.services.mozilla.com/D218533
This commit is contained in:
Jonathan Kew 2024-08-06 08:36:20 +00:00
parent ba0b2b57c8
commit c2cefa5aae
3 changed files with 127 additions and 98 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

104
dom/base/nsLineBreaker.cpp
View File

@ -78,63 +78,71 @@ nsLineBreaker::~nsLineBreaker() {
"Should have Reset() before destruction!");
}
/* static */
bool nsLineBreaker::ShouldCapitalize(uint32_t aChar, bool& aCapitalizeNext) {
using mozilla::intl::GeneralCategory;
auto category = UnicodeProperties::CharType(aChar);
switch (category) {
case GeneralCategory::Uppercase_Letter:
case GeneralCategory::Lowercase_Letter:
case GeneralCategory::Titlecase_Letter:
case GeneralCategory::Modifier_Letter:
case GeneralCategory::Other_Letter:
case GeneralCategory::Decimal_Number:
case GeneralCategory::Letter_Number:
case GeneralCategory::Other_Number:
if (aCapitalizeNext) {
aCapitalizeNext = false;
return true;
}
break;
case GeneralCategory::Space_Separator:
case GeneralCategory::Line_Separator:
case GeneralCategory::Paragraph_Separator:
case GeneralCategory::Dash_Punctuation:
case GeneralCategory::Initial_Punctuation:
/* These punctuation categories are excluded, for examples like
* "what colo[u]r" -> "What Colo[u]r?" (rather than "What Colo[U]R?")
* and
* "snake_case" -> "Snake_case" (to match word selection behavior)
case GeneralCategory::Open_Punctuation:
case GeneralCategory::Close_Punctuation:
case GeneralCategory::Connector_Punctuation:
*/
aCapitalizeNext = true;
break;
case GeneralCategory::Final_Punctuation:
/* Special-case: exclude Unicode single-close-quote/apostrophe,
for examples like "Lowe’s" etc. */
if (aChar != 0x2019) {
aCapitalizeNext = true;
}
break;
case GeneralCategory::Other_Punctuation:
/* Special-case: exclude ASCII apostrophe, for "Lowe's" etc.,
and MIDDLE DOT, for Catalan "l·l". */
if (aChar != '\'' && aChar != 0x00B7) {
aCapitalizeNext = true;
}
break;
default:
break;
}
return false;
}
static void SetupCapitalization(const char16_t* aWord, uint32_t aLength,
bool* aCapitalization) {
// Capitalize the first alphanumeric character after a space or punctuation.
using mozilla::intl::GeneralCategory;
bool capitalizeNextChar = true;
for (uint32_t i = 0; i < aLength; ++i) {
uint32_t ch = aWord[i];
if (i + 1 < aLength && NS_IS_SURROGATE_PAIR(ch, aWord[i + 1])) {
ch = SURROGATE_TO_UCS4(ch, aWord[i + 1]);
}
auto category = UnicodeProperties::CharType(ch);
switch (category) {
case GeneralCategory::Uppercase_Letter:
case GeneralCategory::Lowercase_Letter:
case GeneralCategory::Titlecase_Letter:
case GeneralCategory::Modifier_Letter:
case GeneralCategory::Other_Letter:
case GeneralCategory::Decimal_Number:
case GeneralCategory::Letter_Number:
case GeneralCategory::Other_Number:
if (capitalizeNextChar) {
aCapitalization[i] = true;
capitalizeNextChar = false;
}
break;
case GeneralCategory::Space_Separator:
case GeneralCategory::Line_Separator:
case GeneralCategory::Paragraph_Separator:
case GeneralCategory::Dash_Punctuation:
case GeneralCategory::Initial_Punctuation:
/* These punctuation categories are excluded, for examples like
* "what colo[u]r" -> "What Colo[u]r?" (rather than "What Colo[U]R?")
* and
* "snake_case" -> "Snake_case" (to match word selection behavior)
case GeneralCategory::Open_Punctuation:
case GeneralCategory::Close_Punctuation:
case GeneralCategory::Connector_Punctuation:
*/
capitalizeNextChar = true;
break;
case GeneralCategory::Final_Punctuation:
/* Special-case: exclude Unicode single-close-quote/apostrophe,
for examples like "Lowe’s" etc. */
if (ch != 0x2019) {
capitalizeNextChar = true;
}
break;
case GeneralCategory::Other_Punctuation:
/* Special-case: exclude ASCII apostrophe, for "Lowe's" etc.,
and MIDDLE DOT, for Catalan "l·l". */
if (ch != '\'' && ch != 0x00B7) {
capitalizeNextChar = true;
}
break;
default:
break;
}
aCapitalization[i] =
nsLineBreaker::ShouldCapitalize(ch, capitalizeNextChar);
if (!IS_IN_BMP(ch)) {
++i;
}

7
dom/base/nsLineBreaker.h
View File

@ -72,6 +72,13 @@ class nsLineBreaker {
return mozilla::intl::NS_IsSpace(u);
}
// Helper also used by nsCaseTransformTextRunFactory::TransformString.
// aChar is the current character to be examined;
// aCapitalizeNext is a state variable: initialize it to true at start-of-
// text, then pass it back to this function as each successive character is
// considered.
static bool ShouldCapitalize(uint32_t aChar, bool& aCapitalizeNext);
// Break opportunities exist at the end of each run of breakable whitespace
// (see IsSpace above). Break opportunities can also exist between pairs of
// non-whitespace characters, as determined by mozilla::intl::LineBreaker.

114
layout/generic/nsTextRunTransformations.cpp
View File

@ -18,6 +18,7 @@
#include "mozilla/TextEditor.h"
#include "mozilla/gfx/2D.h"
#include "nsGkAtoms.h"
#include "nsLineBreaker.h"
#include "nsSpecialCasingData.h"
#include "nsStyleConsts.h"
#include "nsTextFrameUtils.h"
@ -335,6 +336,10 @@ bool nsCaseTransformTextRunFactory::TransformString(
// tonos added (if it is disjunctive eta)
const char16_t kGreekUpperEta = 0x0397;
// If we're doing capitalization and don't have a textrun, this is the state
// to be passed to each call to nsLineBreaker::ShouldCapitalize.
bool capitalizeNext = true;
for (uint32_t i = 0; i < length; ++i, ++aOffsetInTextRun) {
uint32_t ch = str[i];
@ -365,7 +370,7 @@ bool nsCaseTransformTextRunFactory::TransformString(
bool maskPassword = (charStyle && charStyle->mMaskPassword) || aMaskChar;
int extraChars = 0;
const mozilla::unicode::MultiCharMapping* mcm;
const unicode::MultiCharMapping* mcm;
bool inhibitBreakBefore = false; // have we just deleted preceding hyphen?
if (i < length - 1 && NS_IS_SURROGATE_PAIR(ch, str[i + 1])) {
@ -448,7 +453,7 @@ bool nsCaseTransformTextRunFactory::TransformString(
}
}
cat = mozilla::unicode::GetGenCategory(ch);
cat = unicode::GetGenCategory(ch);
if (languageSpecificCasing == eLSCB_Irish &&
cat == nsUGenCategory::kLetter) {
@ -528,7 +533,7 @@ bool nsCaseTransformTextRunFactory::TransformString(
sigmaIndex = uint32_t(-1);
}
mcm = mozilla::unicode::SpecialLower(ch);
mcm = unicode::SpecialLower(ch);
if (mcm) {
int j = 0;
while (j < 2 && mcm->mMappedChars[j + 1]) {
@ -664,7 +669,7 @@ bool nsCaseTransformTextRunFactory::TransformString(
break;
}
mcm = mozilla::unicode::SpecialUpper(ch);
mcm = unicode::SpecialUpper(ch);
if (mcm) {
int j = 0;
while (j < 2 && mcm->mMappedChars[j + 1]) {
@ -686,61 +691,70 @@ bool nsCaseTransformTextRunFactory::TransformString(
}
break;
case StyleTextTransform::CAPITALIZE._0:
case StyleTextTransform::CAPITALIZE._0: {
if (capitalizeDutchIJ && ch == 'j') {
ch = 'J';
capitalizeDutchIJ = false;
break;
}
capitalizeDutchIJ = false;
// If we have a textrun, its mCapitalize array tells us which chars
// are to be capitalized. If not, we track the state locally, and
// assume there's no context to be considered.
bool doCapitalize = false;
if (aTextRun) {
if (capitalizeDutchIJ && ch == 'j') {
ch = 'J';
capitalizeDutchIJ = false;
if (aOffsetInTextRun < aTextRun->mCapitalize.Length()) {
doCapitalize = aTextRun->mCapitalize[aOffsetInTextRun];
}
} else {
doCapitalize = nsLineBreaker::ShouldCapitalize(ch, capitalizeNext);
}
if (doCapitalize) {
if (languageSpecificCasing == eLSCB_Turkish && ch == 'i') {
ch = LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE;
break;
}
capitalizeDutchIJ = false;
if (aOffsetInTextRun < aTextRun->mCapitalize.Length() &&
aTextRun->mCapitalize[aOffsetInTextRun]) {
if (languageSpecificCasing == eLSCB_Turkish && ch == 'i') {
ch = LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE;
if (languageSpecificCasing == eLSCB_Dutch && ch == 'i') {
ch = 'I';
capitalizeDutchIJ = true;
break;
}
if (languageSpecificCasing == eLSCB_Lithuanian) {
/*
* # Remove DOT ABOVE after "i" with upper or titlecase
*
* 0307; 0307; ; ; lt After_Soft_Dotted; # COMBINING DOT ABOVE
*/
if (ch == 'i' || ch == 'j' || ch == 0x012F) {
seenSoftDotted = true;
ch = ToTitleCase(ch);
break;
}
if (languageSpecificCasing == eLSCB_Dutch && ch == 'i') {
ch = 'I';
capitalizeDutchIJ = true;
break;
}
if (languageSpecificCasing == eLSCB_Lithuanian) {
/*
* # Remove DOT ABOVE after "i" with upper or titlecase
*
* 0307; 0307; ; ; lt After_Soft_Dotted; # COMBINING DOT ABOVE
*/
if (ch == 'i' || ch == 'j' || ch == 0x012F) {
seenSoftDotted = true;
ch = ToTitleCase(ch);
if (seenSoftDotted) {
seenSoftDotted = false;
if (ch == 0x0307) {
ch = uint32_t(-1);
break;
}
if (seenSoftDotted) {
seenSoftDotted = false;
if (ch == 0x0307) {
ch = uint32_t(-1);
break;
}
}
}
mcm = mozilla::unicode::SpecialTitle(ch);
if (mcm) {
int j = 0;
while (j < 2 && mcm->mMappedChars[j + 1]) {
aConvertedString.Append(mcm->mMappedChars[j]);
++extraChars;
++j;
}
ch = mcm->mMappedChars[j];
break;
}
ch = ToTitleCase(ch);
}
mcm = unicode::SpecialTitle(ch);
if (mcm) {
int j = 0;
while (j < 2 && mcm->mMappedChars[j + 1]) {
aConvertedString.Append(mcm->mMappedChars[j]);
++extraChars;
++j;
}
ch = mcm->mMappedChars[j];
break;
}
ch = ToTitleCase(ch);
}
break;
}
case StyleTextTransform::MATH_AUTO._0:
// text-transform: math-auto is used for automatic italicization of
@ -773,7 +787,7 @@ bool nsCaseTransformTextRunFactory::TransformString(
if (!aCaseTransformsOnly) {
if (!forceNonFullWidth && (style & StyleTextTransform::FULL_WIDTH)) {
ch = mozilla::unicode::GetFullWidth(ch);
ch = unicode::GetFullWidth(ch);
}
if (style & StyleTextTransform::FULL_SIZE_KANA) {
@ -825,7 +839,7 @@ bool nsCaseTransformTextRunFactory::TransformString(
}
if (forceNonFullWidth) {
ch = mozilla::unicode::GetFullWidthInverse(ch);
ch = unicode::GetFullWidthInverse(ch);
}
}