diff options
Diffstat (limited to 'Userland/Libraries/LibWeb')
| -rw-r--r-- | Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp | 49 | ||||
| -rw-r--r-- | Userland/Libraries/LibWeb/DOM/Document.cpp | 4 | ||||
| -rw-r--r-- | Userland/Libraries/LibWeb/HTML/Parser/HTMLTokenizer.cpp | 70 | ||||
| -rw-r--r-- | Userland/Libraries/LibWeb/Layout/TextNode.cpp | 14 |
4 files changed, 51 insertions, 86 deletions
diff --git a/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp b/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp index d15e1664ef..113b240cde 100644 --- a/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp +++ b/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp @@ -4,11 +4,11 @@ * SPDX-License-Identifier: BSD-2-Clause */ +#include <AK/CharacterTypes.h> #include <AK/SourceLocation.h> #include <AK/Vector.h> #include <LibTextCodec/Decoder.h> #include <LibWeb/CSS/Parser/Tokenizer.h> -#include <ctype.h> #define CSS_TOKENIZER_TRACE 0 @@ -20,11 +20,6 @@ static inline void log_parse_error(const SourceLocation& location = SourceLocati dbgln_if(CSS_TOKENIZER_TRACE, "Parse error (css tokenization) {} ", location); } -static inline bool is_surrogate(u32 code_point) -{ - return (code_point & 0xfffff800) == 0xd800; -} - static inline bool is_quotation_mark(u32 code_point) { return code_point == 0x22; @@ -35,24 +30,14 @@ static inline bool is_greater_than_maximum_allowed_code_point(u32 code_point) return code_point > 0x10FFFF; } -static inline bool is_hex_digit(u32 code_point) -{ - return isxdigit(code_point); -} - static inline bool is_low_line(u32 code_point) { return code_point == 0x5F; } -static inline bool is_non_ascii(u32 code_point) -{ - return code_point >= 0x80; -} - static inline bool is_name_start_code_point(u32 code_point) { - return isalpha(code_point) || is_non_ascii(code_point) || is_low_line(code_point); + return is_ascii_alpha(code_point) || !is_ascii(code_point) || is_low_line(code_point); } static inline bool is_hyphen_minus(u32 code_point) @@ -62,7 +47,7 @@ static inline bool is_hyphen_minus(u32 code_point) static inline bool is_name_code_point(u32 code_point) { - return is_name_start_code_point(code_point) || isdigit(code_point) || is_hyphen_minus(code_point); + return is_name_start_code_point(code_point) || is_ascii_digit(code_point) || is_hyphen_minus(code_point); } static inline bool is_non_printable(u32 code_point) @@ -303,12 +288,12 @@ u32 Tokenizer::consume_escaped_code_point() auto input = code_point.value(); - if (is_hex_digit(input)) { + if (is_ascii_hex_digit(input)) { StringBuilder builder; builder.append_code_point(input); size_t counter = 0; - while (is_hex_digit(peek_code_point().value()) && counter++ < 5) { + while (is_ascii_hex_digit(peek_code_point().value()) && counter++ < 5) { builder.append_code_point(next_code_point().value()); } @@ -317,7 +302,7 @@ u32 Tokenizer::consume_escaped_code_point() } auto unhexed = strtoul(builder.to_string().characters(), nullptr, 16); - if (unhexed == 0 || is_surrogate(unhexed) || is_greater_than_maximum_allowed_code_point(unhexed)) { + if (unhexed == 0 || is_unicode_surrogate(unhexed) || is_greater_than_maximum_allowed_code_point(unhexed)) { return REPLACEMENT_CHARACTER; } @@ -378,14 +363,14 @@ CSSNumber Tokenizer::consume_a_number() for (;;) { auto digits = peek_code_point().value(); - if (!isdigit(digits)) + if (!is_ascii_digit(digits)) break; repr.append_code_point(next_code_point().value()); } auto maybe_number = peek_twin().value(); - if (is_full_stop(maybe_number.first) && isdigit(maybe_number.second)) { + if (is_full_stop(maybe_number.first) && is_ascii_digit(maybe_number.second)) { repr.append_code_point(next_code_point().value()); repr.append_code_point(next_code_point().value()); @@ -393,7 +378,7 @@ CSSNumber Tokenizer::consume_a_number() for (;;) { auto digits = peek_code_point(); - if (digits.has_value() && !isdigit(digits.value())) + if (digits.has_value() && !is_ascii_digit(digits.value())) break; repr.append_code_point(next_code_point().value()); @@ -403,12 +388,12 @@ CSSNumber Tokenizer::consume_a_number() auto maybe_exp = peek_triplet().value(); if (is_E(maybe_exp.first) || is_e(maybe_exp.first)) { if (is_plus_sign(maybe_exp.second) || is_hyphen_minus(maybe_exp.second)) { - if (isdigit(maybe_exp.third)) { + if (is_ascii_digit(maybe_exp.third)) { repr.append_code_point(next_code_point().value()); repr.append_code_point(next_code_point().value()); repr.append_code_point(next_code_point().value()); } - } else if (isdigit(maybe_exp.second)) { + } else if (is_ascii_digit(maybe_exp.second)) { repr.append_code_point(next_code_point().value()); repr.append_code_point(next_code_point().value()); } @@ -417,7 +402,7 @@ CSSNumber Tokenizer::consume_a_number() for (;;) { auto digits = peek_code_point().value(); - if (!isdigit(digits)) + if (!is_ascii_digit(digits)) break; repr.append_code_point(next_code_point().value()); @@ -588,19 +573,19 @@ bool Tokenizer::starts_with_a_number() const bool Tokenizer::starts_with_a_number(U32Triplet values) { if (is_plus_sign(values.first) || is_hyphen_minus(values.first)) { - if (isdigit(values.second)) + if (is_ascii_digit(values.second)) return true; - if (is_full_stop(values.second) && isdigit(values.third)) + if (is_full_stop(values.second) && is_ascii_digit(values.third)) return true; return false; } if (is_full_stop(values.first)) - return isdigit(values.second); + return is_ascii_digit(values.second); - if (isdigit(values.first)) + if (is_ascii_digit(values.first)) return true; return false; @@ -902,7 +887,7 @@ Token Tokenizer::consume_a_token() return create_new_token(Token::TokenType::CloseCurly); } - if (isdigit(input)) { + if (is_ascii_digit(input)) { dbgln_if(CSS_TOKENIZER_TRACE, "is digit"); reconsume_current_input_code_point(); return consume_a_numeric_token(); diff --git a/Userland/Libraries/LibWeb/DOM/Document.cpp b/Userland/Libraries/LibWeb/DOM/Document.cpp index b9887a8553..27c8ba5bb0 100644 --- a/Userland/Libraries/LibWeb/DOM/Document.cpp +++ b/Userland/Libraries/LibWeb/DOM/Document.cpp @@ -6,6 +6,7 @@ * SPDX-License-Identifier: BSD-2-Clause */ +#include <AK/CharacterTypes.h> #include <AK/StringBuilder.h> #include <AK/Utf8View.h> #include <LibCore/Timer.h> @@ -53,7 +54,6 @@ #include <LibWeb/Page/BrowsingContext.h> #include <LibWeb/SVG/TagNames.h> #include <LibWeb/UIEvents/MouseEvent.h> -#include <ctype.h> namespace Web::DOM { @@ -253,7 +253,7 @@ String Document::title() const StringBuilder builder; bool last_was_space = false; for (auto code_point : Utf8View(raw_title)) { - if (isspace(code_point)) { + if (is_ascii_space(code_point)) { last_was_space = true; } else { if (last_was_space && !builder.is_empty()) diff --git a/Userland/Libraries/LibWeb/HTML/Parser/HTMLTokenizer.cpp b/Userland/Libraries/LibWeb/HTML/Parser/HTMLTokenizer.cpp index 7d716a8532..638ba37bea 100644 --- a/Userland/Libraries/LibWeb/HTML/Parser/HTMLTokenizer.cpp +++ b/Userland/Libraries/LibWeb/HTML/Parser/HTMLTokenizer.cpp @@ -4,13 +4,13 @@ * SPDX-License-Identifier: BSD-2-Clause */ +#include <AK/CharacterTypes.h> #include <AK/Debug.h> #include <AK/SourceLocation.h> #include <LibTextCodec/Decoder.h> #include <LibWeb/HTML/Parser/Entities.h> #include <LibWeb/HTML/Parser/HTMLToken.h> #include <LibWeb/HTML/Parser/HTMLTokenizer.h> -#include <ctype.h> #include <string.h> namespace Web::HTML { @@ -93,25 +93,25 @@ namespace Web::HTML { if (!current_input_character.has_value()) #define ON_ASCII_ALPHA \ - if (current_input_character.has_value() && isalpha(current_input_character.value())) + if (current_input_character.has_value() && is_ascii_alpha(current_input_character.value())) #define ON_ASCII_ALPHANUMERIC \ - if (current_input_character.has_value() && isalnum(current_input_character.value())) + if (current_input_character.has_value() && is_ascii_alphanumeric(current_input_character.value())) #define ON_ASCII_UPPER_ALPHA \ - if (current_input_character.has_value() && current_input_character.value() >= 'A' && current_input_character.value() <= 'Z') + if (current_input_character.has_value() && is_ascii_upper_alpha(current_input_character.value())) #define ON_ASCII_LOWER_ALPHA \ - if (current_input_character.has_value() && current_input_character.value() >= 'a' && current_input_character.value() <= 'z') + if (current_input_character.has_value() && is_ascii_lower_alpha(current_input_character.value())) #define ON_ASCII_DIGIT \ - if (current_input_character.has_value() && isdigit(current_input_character.value())) + if (current_input_character.has_value() && is_ascii_digit(current_input_character.value())) #define ON_ASCII_HEX_DIGIT \ - if (current_input_character.has_value() && isxdigit(current_input_character.value())) + if (current_input_character.has_value() && is_ascii_hex_digit(current_input_character.value())) #define ON_WHITESPACE \ - if (current_input_character.has_value() && strchr("\t\n\f ", current_input_character.value())) + if (current_input_character.has_value() && is_ascii(current_input_character.value()) && "\t\n\f "sv.contains(current_input_character.value())) #define ANYTHING_ELSE if (1) @@ -172,26 +172,6 @@ static inline void log_parse_error(const SourceLocation& location = SourceLocati dbgln_if(TOKENIZER_TRACE_DEBUG, "Parse error (tokenization) {}", location); } -static inline bool is_surrogate(u32 code_point) -{ - return (code_point & 0xfffff800) == 0xd800; -} - -static inline bool is_noncharacter(u32 code_point) -{ - return code_point >= 0xfdd0 && (code_point <= 0xfdef || (code_point & 0xfffe) == 0xfffe) && code_point <= 0x10ffff; -} - -static inline bool is_c0_control(u32 code_point) -{ - return code_point <= 0x1f; -} - -static inline bool is_control(u32 code_point) -{ - return is_c0_control(code_point) || (code_point >= 0x7f && code_point <= 0x9f); -} - Optional<u32> HTMLTokenizer::next_code_point() { if (m_utf8_iterator == m_utf8_view.end()) @@ -322,7 +302,7 @@ _StartOfFunction: } ON_ASCII_UPPER_ALPHA { - m_current_token.m_tag.tag_name.append(tolower(current_input_character.value())); + m_current_token.m_tag.tag_name.append(to_ascii_lowercase(current_input_character.value())); m_current_token.m_end_position = nth_last_position(0); continue; } @@ -458,7 +438,7 @@ _StartOfFunction: ON_ASCII_UPPER_ALPHA { create_new_token(HTMLToken::Type::DOCTYPE); - m_current_token.m_doctype.name.append(tolower(current_input_character.value())); + m_current_token.m_doctype.name.append(to_ascii_lowercase(current_input_character.value())); m_current_token.m_doctype.missing_name = false; SWITCH_TO(DOCTYPEName); } @@ -507,7 +487,7 @@ _StartOfFunction: } ON_ASCII_UPPER_ALPHA { - m_current_token.m_doctype.name.append(tolower(current_input_character.value())); + m_current_token.m_doctype.name.append(to_ascii_lowercase(current_input_character.value())); continue; } ON(0) @@ -550,10 +530,10 @@ _StartOfFunction: } ANYTHING_ELSE { - if (toupper(current_input_character.value()) == 'P' && consume_next_if_match("UBLIC", CaseSensitivity::CaseInsensitive)) { + if (to_ascii_uppercase(current_input_character.value()) == 'P' && consume_next_if_match("UBLIC", CaseSensitivity::CaseInsensitive)) { SWITCH_TO(AfterDOCTYPEPublicKeyword); } - if (toupper(current_input_character.value()) == 'S' && consume_next_if_match("YSTEM", CaseSensitivity::CaseInsensitive)) { + if (to_ascii_uppercase(current_input_character.value()) == 'S' && consume_next_if_match("YSTEM", CaseSensitivity::CaseInsensitive)) { SWITCH_TO(AfterDOCTYPESystemKeyword); } log_parse_error(); @@ -1068,7 +1048,7 @@ _StartOfFunction: } ON_ASCII_UPPER_ALPHA { - m_current_token.m_tag.attributes.last().local_name_builder.append_code_point(tolower(current_input_character.value())); + m_current_token.m_tag.attributes.last().local_name_builder.append_code_point(to_ascii_lowercase(current_input_character.value())); continue; } ON(0) @@ -1558,7 +1538,7 @@ _StartOfFunction: if (consumed_as_part_of_an_attribute() && !match.value().entity.ends_with(';')) { auto next_code_point = peek_code_point(0); - if (next_code_point.has_value() && (next_code_point.value() == '=' || isalnum(next_code_point.value()))) { + if (next_code_point.has_value() && (next_code_point.value() == '=' || is_ascii_alphanumeric(next_code_point.value()))) { FLUSH_CODEPOINTS_CONSUMED_AS_A_CHARACTER_REFERENCE; SWITCH_TO_RETURN_STATE; } @@ -1720,14 +1700,14 @@ _StartOfFunction: log_parse_error(); m_character_reference_code = 0xFFFD; } - if (is_surrogate(m_character_reference_code)) { + if (is_unicode_surrogate(m_character_reference_code)) { log_parse_error(); m_character_reference_code = 0xFFFD; } - if (is_noncharacter(m_character_reference_code)) { + if (is_unicode_noncharacter(m_character_reference_code)) { log_parse_error(); } - if (m_character_reference_code == 0xd || (is_control(m_character_reference_code) && !isspace(m_character_reference_code))) { + if (m_character_reference_code == 0xd || (is_unicode_control(m_character_reference_code) && !is_ascii_space(m_character_reference_code))) { log_parse_error(); constexpr struct { u32 number; @@ -1870,7 +1850,7 @@ _StartOfFunction: } ON_ASCII_UPPER_ALPHA { - m_current_token.m_tag.tag_name.append(tolower(current_input_character.value())); + m_current_token.m_tag.tag_name.append(to_ascii_lowercase(current_input_character.value())); m_temporary_buffer.append(current_input_character.value()); continue; } @@ -1980,7 +1960,7 @@ _StartOfFunction: } ON_ASCII_UPPER_ALPHA { - m_current_token.m_tag.tag_name.append(tolower(current_input_character.value())); + m_current_token.m_tag.tag_name.append(to_ascii_lowercase(current_input_character.value())); m_temporary_buffer.append(current_input_character.value()); continue; } @@ -2193,7 +2173,7 @@ _StartOfFunction: } ON_ASCII_UPPER_ALPHA { - m_current_token.m_tag.tag_name.append(tolower(current_input_character.value())); + m_current_token.m_tag.tag_name.append(to_ascii_lowercase(current_input_character.value())); m_temporary_buffer.append(current_input_character.value()); continue; } @@ -2247,7 +2227,7 @@ _StartOfFunction: } ON_ASCII_UPPER_ALPHA { - m_temporary_buffer.append(tolower(current_input_character.value())); + m_temporary_buffer.append(to_ascii_lowercase(current_input_character.value())); EMIT_CURRENT_CHARACTER; } ON_ASCII_LOWER_ALPHA @@ -2393,7 +2373,7 @@ _StartOfFunction: } ON_ASCII_UPPER_ALPHA { - m_temporary_buffer.append(tolower(current_input_character.value())); + m_temporary_buffer.append(to_ascii_lowercase(current_input_character.value())); EMIT_CURRENT_CHARACTER; } ON_ASCII_LOWER_ALPHA @@ -2512,7 +2492,7 @@ _StartOfFunction: } ON_ASCII_UPPER_ALPHA { - m_current_token.m_tag.tag_name.append(tolower(current_input_character.value())); + m_current_token.m_tag.tag_name.append(to_ascii_lowercase(current_input_character.value())); m_temporary_buffer.append(current_input_character.value()); continue; } @@ -2598,7 +2578,7 @@ bool HTMLTokenizer::consume_next_if_match(const StringView& string, CaseSensitiv // FIXME: This should be more Unicode-aware. if (case_sensitivity == CaseSensitivity::CaseInsensitive) { if (code_point.value() < 0x80) { - if (tolower(code_point.value()) != tolower(string[i])) + if (to_ascii_lowercase(code_point.value()) != to_ascii_lowercase(string[i])) return false; continue; } diff --git a/Userland/Libraries/LibWeb/Layout/TextNode.cpp b/Userland/Libraries/LibWeb/Layout/TextNode.cpp index c5aaeb47d8..6f7818c55f 100644 --- a/Userland/Libraries/LibWeb/Layout/TextNode.cpp +++ b/Userland/Libraries/LibWeb/Layout/TextNode.cpp @@ -4,6 +4,7 @@ * SPDX-License-Identifier: BSD-2-Clause */ +#include <AK/CharacterTypes.h> #include <AK/ScopeGuard.h> #include <AK/StringBuilder.h> #include <LibGfx/Painter.h> @@ -13,7 +14,6 @@ #include <LibWeb/Layout/Label.h> #include <LibWeb/Layout/TextNode.h> #include <LibWeb/Page/BrowsingContext.h> -#include <ctype.h> namespace Web::Layout { @@ -30,7 +30,7 @@ TextNode::~TextNode() static bool is_all_whitespace(const StringView& string) { for (size_t i = 0; i < string.length(); ++i) { - if (!isspace(string[i])) + if (!is_ascii_space(string[i])) return false; } return true; @@ -116,7 +116,7 @@ void TextNode::compute_text_for_rendering(bool collapse, bool previous_is_empty_ auto it = utf8_view.begin(); auto skip_over_whitespace = [&] { auto prev = it; - while (it != utf8_view.end() && isspace(*it)) { + while (it != utf8_view.end() && is_ascii_space(*it)) { prev = it; ++it; } @@ -125,7 +125,7 @@ void TextNode::compute_text_for_rendering(bool collapse, bool previous_is_empty_ if (previous_is_empty_or_ends_in_whitespace) skip_over_whitespace(); for (; it != utf8_view.end(); ++it) { - if (!isspace(*it)) { + if (!is_ascii_space(*it)) { builder.append(utf8_view.as_string().characters_without_null_termination() + utf8_view.byte_offset_of(it), it.code_point_length_in_bytes()); } else { builder.append(' '); @@ -160,7 +160,7 @@ void TextNode::split_into_lines_by_rules(InlineFormattingContext& context, Layou float chunk_width; if (do_wrap_lines) { - if (do_collapse && isspace(*chunk.view.begin()) && line_boxes.last().is_empty_or_ends_in_whitespace()) { + if (do_collapse && is_ascii_space(*chunk.view.begin()) && line_boxes.last().is_empty_or_ends_in_whitespace()) { // This is a non-empty chunk that starts with collapsible whitespace. // We are at either at the start of a new line, or after something that ended in whitespace, // so we don't need to contribute our own whitespace to the line. Skip over it instead! @@ -264,7 +264,7 @@ TextNode::ChunkIterator::ChunkIterator(StringView const& text, LayoutMode layout , m_start_of_chunk(m_utf8_view.begin()) , m_iterator(m_utf8_view.begin()) { - m_last_was_space = !text.is_empty() && isspace(*m_utf8_view.begin()); + m_last_was_space = !text.is_empty() && is_ascii_space(*m_utf8_view.begin()); } Optional<TextNode::Chunk> TextNode::ChunkIterator::next() @@ -286,7 +286,7 @@ Optional<TextNode::Chunk> TextNode::ChunkIterator::next() return result.release_value(); } if (m_wrap_lines) { - bool is_space = isspace(*m_iterator); + bool is_space = is_ascii_space(*m_iterator); if (is_space != m_last_was_space) { m_last_was_space = is_space; if (auto result = try_commit_chunk(m_iterator, false); result.has_value()) |