1use std::{
2 fs,
3 ops::Range,
4 path::{Path, PathBuf},
5};
6
7use clap::Args;
8use color_eyre::eyre::{self, Result};
9use fluent_templates::Loader;
10use hashbrown::HashSet;
11use novel_api::Timing;
12use pulldown_cmark::{Event, HeadingLevel, Options, Parser, Tag, TagEnd, TextMergeWithOffset};
13
14use crate::{
15 utils::{self, CurrentDir, Lang},
16 LANG_ID, LOCALES,
17};
18
19#[must_use]
20#[derive(Args)]
21#[command(arg_required_else_help = true,
22 about = LOCALES.lookup(&LANG_ID, "check_command"))]
23pub struct Check {
24 #[arg(help = LOCALES.lookup(&LANG_ID, "file_path"))]
25 pub file_path: PathBuf,
26
27 #[arg(long, default_value_t = false,
28 help = LOCALES.lookup(&LANG_ID, "basic_check"))]
29 pub basic_check: bool,
30
31 #[arg(long, default_value_t = false,
32 help = LOCALES.lookup(&LANG_ID, "word_count"))]
33 pub word_count: bool,
34}
35
36pub fn execute(config: Check) -> Result<()> {
37 let mut timing = Timing::new();
38
39 let input_file_path;
40 let input_file_parent_path;
41
42 if utils::is_markdown_or_txt_file(&config.file_path)? {
43 input_file_path = dunce::canonicalize(&config.file_path)?;
44 input_file_parent_path = input_file_path.parent().unwrap().to_path_buf();
45 } else if let Ok(Some(path)) =
46 utils::try_get_markdown_or_txt_file_name_in_dir(&config.file_path)
47 {
48 input_file_path = path;
49 input_file_parent_path = dunce::canonicalize(&config.file_path)?;
50 } else {
51 eyre::bail!("Invalid input path: `{}`", config.file_path.display());
52 }
53 tracing::info!("Input file path: `{}`", input_file_path.display());
54
55 let current_dir = CurrentDir::new(input_file_parent_path)?;
56
57 let bytes = fs::read(&input_file_path)?;
58 let markdown = simdutf8::basic::from_utf8(&bytes)?;
59 let mut parser = TextMergeWithOffset::new(
60 Parser::new_ext(markdown, Options::ENABLE_YAML_STYLE_METADATA_BLOCKS).into_offset_iter(),
61 );
62
63 let lang = check_metadata(&mut parser)?;
64
65 let max_width = (utils::terminal_size().0 / 2) as usize;
66 let mut char_set = HashSet::new();
67 let mut in_paragraph = false;
68 let mut word_count = 0;
69 parser.for_each(|(event, range)| match event {
70 Event::Start(tag) => match tag {
71 Tag::Heading { level, .. } => {
72 let title = markdown[range].trim_start_matches('#').trim();
73
74 if level == HeadingLevel::H1 {
75 if !check_volume_title(title, lang) {
76 println_msg(format!("Irregular volume title format: `{title}`"));
77 }
78 } else if level == HeadingLevel::H2 {
79 if !check_chapter_title(title, lang) {
80 println_msg(format!("Irregular chapter title format: `{title}`"));
81 }
82 } else {
83 println_msg(format!(
84 "Irregular heading level: `{level:?}`, content: `{title}`"
85 ));
86 }
87 }
88 Tag::Image { dest_url, .. } => {
89 let image_path = Path::new(dest_url.as_ref());
90
91 if !image_path.is_file() {
92 println_msg(format!("Image `{}` does not exist", image_path.display()));
93 }
94 }
95 Tag::Paragraph => {
96 in_paragraph = true;
97 }
98 Tag::BlockQuote(_)
99 | Tag::CodeBlock(_)
100 | Tag::List(_)
101 | Tag::Item
102 | Tag::FootnoteDefinition(_)
103 | Tag::Table(_)
104 | Tag::TableHead
105 | Tag::TableRow
106 | Tag::TableCell
107 | Tag::Emphasis
108 | Tag::Strong
109 | Tag::Strikethrough
110 | Tag::Link { .. }
111 | Tag::HtmlBlock
112 | Tag::MetadataBlock(_)
113 | Tag::DefinitionList
114 | Tag::DefinitionListTitle
115 | Tag::DefinitionListDefinition => {
116 if !config.basic_check {
117 let content = console::truncate_str(markdown[range].trim(), max_width, "...");
118
119 println_msg(format!(
120 "Markdown tag that should not appear: `{tag:?}`, content: `{content}`"
121 ));
122 }
123 }
124 },
125 Event::Text(text) => {
126 if !config.basic_check {
127 for c in text.chars() {
128 if !utils::is_cjk(c)
129 && !utils::is_punctuation(c)
130 && !c.is_ascii_alphanumeric()
131 && c != ' '
132 {
133 if char_set.contains(&c) {
134 continue;
135 } else {
136 char_set.insert(c);
137
138 println_msg(format!(
139 "Irregular char: `{}`, at `{}`",
140 c,
141 console::truncate_str(
142 markdown[range.clone()].trim(),
143 max_width,
144 "..."
145 )
146 ));
147 }
148 }
149 }
150 }
151
152 if config.word_count {
153 for c in text.chars() {
154 if utils::is_cjk(c) {
155 word_count += 1;
156 }
157 }
158 }
159 }
160 Event::End(tag) => {
161 if let TagEnd::Paragraph = tag {
162 in_paragraph = false;
163 }
164 }
165 Event::HardBreak
166 | Event::Code(_)
167 | Event::Html(_)
168 | Event::FootnoteReference(_)
169 | Event::SoftBreak
170 | Event::Rule
171 | Event::TaskListMarker(_)
172 | Event::InlineHtml(_)
173 | Event::InlineMath(_)
174 | Event::DisplayMath(_) => {
175 if !config.basic_check {
176 let content = console::truncate_str(markdown[range].trim(), max_width, "...");
177
178 println_msg(format!(
179 "Markdown event that should not appear: `{event:?}`, content: `{content}`"
180 ));
181 }
182 }
183 });
184
185 if config.word_count {
186 println!("Total number of words: {word_count}");
187 }
188
189 current_dir.restore()?;
190
191 tracing::debug!("Time spent on `check`: {}", timing.elapsed()?);
192
193 Ok(())
194}
195
196fn check_metadata<'a, T>(parser: &mut TextMergeWithOffset<'a, T>) -> Result<Lang>
197where
198 T: Iterator<Item = (Event<'a>, Range<usize>)>,
199{
200 let metadata = utils::get_metadata(parser)?;
201
202 eyre::ensure!(
203 metadata.cover_image_is_ok(),
204 "Cover image does not exist: `{}`",
205 metadata.cover_image.unwrap().display()
206 );
207
208 Ok(metadata.lang)
209}
210
211fn println_msg(msg: String) {
212 println!("{} {}", utils::emoji("⚠️"), msg);
213}
214
215macro_rules! regex {
216 ($re:literal $(,)?) => {{
217 static RE: std::sync::OnceLock<regex::Regex> = std::sync::OnceLock::new();
218 RE.get_or_init(|| regex::Regex::new($re).unwrap())
219 }};
220}
221
222#[must_use]
223fn check_chapter_title<T>(title: T, lang: Lang) -> bool
224where
225 T: AsRef<str>,
226{
227 let title = title.as_ref();
228
229 match lang {
230 Lang::ZhHant => {
231 let regex = regex!(r"第([零一二三四五六七八九十百千]|[0-9]){1,7}[章話] .+");
232 regex.is_match(title.as_ref())
233 }
234 Lang::ZhHans => {
235 let regex = regex!(r"第([零一二三四五六七八九十百千]|[0-9]){1,7}[章话] .+");
236 regex.is_match(title.as_ref())
237 }
238 }
239}
240
241#[must_use]
242fn check_volume_title<T>(title: T, lang: Lang) -> bool
243where
244 T: AsRef<str>,
245{
246 let title = title.as_ref();
247
248 match lang {
249 Lang::ZhHant => {
250 let regex = regex!(r"第([一二三四五六七八九十]|[0-9]){1,3}卷 .+");
251 regex.is_match(title) || title == "簡介"
252 }
253 Lang::ZhHans => {
254 let regex = regex!(r"第([一二三四五六七八九十]|[0-9]){1,3}卷 .+");
255 regex.is_match(title) || title == "简介"
256 }
257 }
258}
259
260#[cfg(test)]
261mod tests {
262 use super::*;
263
264 #[test]
265 fn check_chapter_title_test() {
266 assert!(check_chapter_title("第一章 被俘虏的开始", Lang::ZhHans));
267 assert!(check_chapter_title(
268 "第一百三十二章 标标标标标标标标标",
269 Lang::ZhHans
270 ));
271 assert!(check_chapter_title("第123章 标题标标标标", Lang::ZhHans));
272 assert!(!check_chapter_title("第一章 ", Lang::ZhHans));
273 assert!(!check_chapter_title("第1二3话", Lang::ZhHans));
274 assert!(!check_chapter_title("第123话标题", Lang::ZhHans));
275 assert!(!check_chapter_title("123话 标题", Lang::ZhHans));
276 }
277
278 #[test]
279 fn check_volume_title_test() {
280 assert!(check_volume_title(
281 "第三十二卷 标标标标标标标标标",
282 Lang::ZhHans
283 ));
284 assert!(!check_volume_title("第123话 标题标标标标", Lang::ZhHans));
285 assert!(!check_volume_title("第1卷 ", Lang::ZhHans));
286 }
287}