diff options
Diffstat (limited to 'src/construct')
| -rw-r--r-- | src/construct/attention.rs | 9 | ||||
| -rw-r--r-- | src/construct/gfm_autolink_literal.rs | 10 | ||||
| -rw-r--r-- | src/construct/partial_mdx_jsx.rs | 79 | ||||
| -rw-r--r-- | src/construct/partial_space_or_tab_eol.rs | 36 |
4 files changed, 55 insertions, 79 deletions
diff --git a/src/construct/attention.rs b/src/construct/attention.rs index 947a79b..4a208df 100644 --- a/src/construct/attention.rs +++ b/src/construct/attention.rs @@ -80,8 +80,13 @@ use crate::event::{Event, Kind, Name, Point}; use crate::resolve::Name as ResolveName; use crate::state::{Name as StateName, State}; use crate::tokenizer::Tokenizer; -use crate::util::classify_character::{classify_opt, Kind as CharacterKind}; -use crate::util::slice::{char_after_index, char_before_index, Slice}; +use crate::util::{ + char::{ + after_index as char_after_index, before_index as char_before_index, classify_opt, + Kind as CharacterKind, + }, + slice::Slice, +}; use alloc::{vec, vec::Vec}; /// Attentention sequence that we can take markers from. diff --git a/src/construct/gfm_autolink_literal.rs b/src/construct/gfm_autolink_literal.rs index ae483a7..c25f04c 100644 --- a/src/construct/gfm_autolink_literal.rs +++ b/src/construct/gfm_autolink_literal.rs @@ -148,8 +148,8 @@ use crate::event::{Event, Kind, Name}; use crate::state::{Name as StateName, State}; use crate::tokenizer::Tokenizer; use crate::util::{ - classify_character::Kind as CharacterKind, - slice::{byte_to_kind, Position, Slice}, + char::{kind_after_index, Kind as CharacterKind}, + slice::{Position, Slice}, }; use alloc::vec::Vec; @@ -366,7 +366,7 @@ pub fn domain_inside(tokenizer: &mut Tokenizer) -> State { } _ => { // Source: <https://github.com/github/cmark-gfm/blob/ef1cfcb/extensions/autolink.c#L12>. - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Other { tokenizer.tokenize_state.seen = true; @@ -470,7 +470,7 @@ pub fn path_inside(tokenizer: &mut Tokenizer) -> State { } _ => { // Source: <https://github.com/github/cmark-gfm/blob/ef1cfcb/extensions/autolink.c#L12>. - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace { State::Retry(StateName::GfmAutolinkLiteralPathAfter) @@ -543,7 +543,7 @@ pub fn trail(tokenizer: &mut Tokenizer) -> State { } _ => { // Whitespace is the end of the URL, anything else is continuation. - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace { State::Ok diff --git a/src/construct/partial_mdx_jsx.rs b/src/construct/partial_mdx_jsx.rs index 1a51608..2daa448 100644 --- a/src/construct/partial_mdx_jsx.rs +++ b/src/construct/partial_mdx_jsx.rs @@ -164,14 +164,11 @@ use crate::event::Name; use crate::state::{Name as StateName, State}; use crate::tokenizer::Tokenizer; -use crate::util::{ - classify_character::Kind as CharacterKind, - slice::{byte_to_kind, char_after_index}, -}; -use alloc::{ - format, - string::{String, ToString}, +use crate::util::char::{ + after_index as char_after_index, format_byte, format_opt as format_char_opt, kind_after_index, + Kind as CharacterKind, }; +use alloc::format; use core::str; use unicode_id::UnicodeID; @@ -305,7 +302,8 @@ pub fn closing_tag_name_before(tokenizer: &mut Tokenizer) -> State { /// ``` pub fn primary_name(tokenizer: &mut Tokenizer) -> State { // End of name. - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) + == CharacterKind::Whitespace || matches!(tokenizer.current, Some(b'.' | b'/' | b':' | b'>' | b'{')) { tokenizer.exit(Name::MdxJsxTagNamePrimary); @@ -418,7 +416,8 @@ pub fn member_name_before(tokenizer: &mut Tokenizer) -> State { pub fn member_name(tokenizer: &mut Tokenizer) -> State { // End of name. // Note: no `:` allowed here. - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) + == CharacterKind::Whitespace || matches!(tokenizer.current, Some(b'.' | b'/' | b'>' | b'{')) { tokenizer.exit(Name::MdxJsxTagNameMember); @@ -529,7 +528,8 @@ pub fn local_name_before(tokenizer: &mut Tokenizer) -> State { /// ``` pub fn local_name(tokenizer: &mut Tokenizer) -> State { // End of local name (note that we don’t expect another colon, or a member). - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) + == CharacterKind::Whitespace || matches!(tokenizer.current, Some(b'/' | b'>' | b'{')) { tokenizer.exit(Name::MdxJsxTagNameLocal); @@ -645,7 +645,8 @@ pub fn attribute_before(tokenizer: &mut Tokenizer) -> State { /// ``` pub fn attribute_primary_name(tokenizer: &mut Tokenizer) -> State { // End of attribute name or tag. - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) + == CharacterKind::Whitespace || matches!(tokenizer.current, Some(b'/' | b':' | b'=' | b'>' | b'{')) { tokenizer.exit(Name::MdxJsxTagAttributePrimaryName); @@ -711,7 +712,7 @@ pub fn attribute_primary_name_after(tokenizer: &mut Tokenizer) -> State { } _ => { // End of tag / new attribute. - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace || matches!(tokenizer.current, Some(b'/' | b'>' | b'{')) || id_start(char_after_index( @@ -768,7 +769,8 @@ pub fn attribute_local_name_before(tokenizer: &mut Tokenizer) -> State { /// ``` pub fn attribute_local_name(tokenizer: &mut Tokenizer) -> State { // End of local name (note that we don’t expect another colon). - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) + == CharacterKind::Whitespace || matches!(tokenizer.current, Some(b'/' | b'=' | b'>' | b'{')) { tokenizer.exit(Name::MdxJsxTagAttributeNameLocal); @@ -986,7 +988,7 @@ pub fn es_whitespace_start(tokenizer: &mut Tokenizer) -> State { match tokenizer.current { Some(b'\n') => State::Retry(StateName::MdxJsxEsWhitespaceEol), _ => { - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace { tokenizer.enter(Name::MdxJsxEsWhitespace); @@ -1016,7 +1018,7 @@ pub fn es_whitespace_inside(tokenizer: &mut Tokenizer) -> State { State::Next(StateName::MdxJsxEsWhitespaceInside) } _ => { - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace { tokenizer.consume(); @@ -1044,7 +1046,7 @@ pub fn es_whitespace_eol(tokenizer: &mut Tokenizer) -> State { pub fn es_whitespace_eol_after(tokenizer: &mut Tokenizer) -> State { if tokenizer.tokenize_state.token_1 == Name::MdxJsxFlowTag && tokenizer.lazy { crash_lazy(tokenizer) - } else if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) + } else if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace { tokenizer.enter(Name::MdxJsxEsWhitespace); @@ -1064,7 +1066,7 @@ pub fn es_whitespace_eol_after_inside(tokenizer: &mut Tokenizer) -> State { State::Next(StateName::MdxJsxEsWhitespaceEolAfterInside) } _ => { - if byte_to_kind(tokenizer.parse_state.bytes, tokenizer.point.index) + if kind_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) == CharacterKind::Whitespace { tokenizer.consume(); @@ -1107,45 +1109,12 @@ fn crash(tokenizer: &Tokenizer, at: &str, expect: &str) -> State { char_after_index(tokenizer.parse_state.bytes, tokenizer.point.index) }; - // To do: externalize this, and the print mechanism in the tokenizer, - // to one proper formatter. - let actual = match char { - None => "end of file".to_string(), - Some(char) => format!("character {}", format_char(char)), - }; - State::Error(format!( "{}:{}: Unexpected {} {}, expected {}", - tokenizer.point.line, tokenizer.point.column, actual, at, expect + tokenizer.point.line, + tokenizer.point.column, + format_char_opt(char), + at, + expect )) } - -fn format_char(char: char) -> String { - let unicode = format!("U+{:>04X}", char as u32); - let printable = match char { - '`' => Some("`` ` ``".to_string()), - ' '..='~' => Some(format!("`{}`", char)), - _ => None, - }; - - if let Some(char) = printable { - format!("{} ({})", char, unicode) - } else { - unicode - } -} - -fn format_byte(byte: u8) -> String { - let unicode = format!("U+{:>04X}", byte); - let printable = match byte { - b'`' => Some("`` ` ``".to_string()), - b' '..=b'~' => Some(format!("`{}`", str::from_utf8(&[byte]).unwrap())), - _ => None, - }; - - if let Some(char) = printable { - format!("{} ({})", char, unicode) - } else { - unicode - } -} diff --git a/src/construct/partial_space_or_tab_eol.rs b/src/construct/partial_space_or_tab_eol.rs index 01f440e..1247639 100644 --- a/src/construct/partial_space_or_tab_eol.rs +++ b/src/construct/partial_space_or_tab_eol.rs @@ -64,24 +64,26 @@ pub fn space_or_tab_eol_with_options(tokenizer: &mut Tokenizer, options: Options /// | ␠␠b /// ``` pub fn start(tokenizer: &mut Tokenizer) -> State { - if matches!(tokenizer.current, Some(b'\t' | b'\n' | b' ')) { - tokenizer.attempt( - State::Next(StateName::SpaceOrTabEolAfterFirst), - State::Next(StateName::SpaceOrTabEolAtEol), - ); + match tokenizer.current { + Some(b'\t' | b' ') => { + tokenizer.attempt( + State::Next(StateName::SpaceOrTabEolAfterFirst), + State::Next(StateName::SpaceOrTabEolAtEol), + ); - State::Retry(space_or_tab_with_options( - tokenizer, - SpaceOrTabOptions { - kind: Name::SpaceOrTab, - min: 1, - max: usize::MAX, - content: tokenizer.tokenize_state.space_or_tab_eol_content.clone(), - connect: tokenizer.tokenize_state.space_or_tab_eol_connect, - }, - )) - } else { - State::Nok + State::Retry(space_or_tab_with_options( + tokenizer, + SpaceOrTabOptions { + kind: Name::SpaceOrTab, + min: 1, + max: usize::MAX, + content: tokenizer.tokenize_state.space_or_tab_eol_content.clone(), + connect: tokenizer.tokenize_state.space_or_tab_eol_connect, + }, + )) + } + Some(b'\n') => State::Retry(StateName::SpaceOrTabEolAtEol), + _ => State::Nok, } } |