From ff5f81498ba1807ab06ffb5dadb1c99c102e0284 Mon Sep 17 00:00:00 2001 From: Titus Wormer Date: Tue, 13 Dec 2022 12:57:14 +0400 Subject: Replace build script with private crate Closes GH-34. Closes GH-35. --- build.rs | 155 --------------------------------------------------------------- 1 file changed, 155 deletions(-) delete mode 100644 build.rs (limited to 'build.rs') diff --git a/build.rs b/build.rs deleted file mode 100644 index 658cb0a..0000000 --- a/build.rs +++ /dev/null @@ -1,155 +0,0 @@ -use regex::Regex; -use std::fs; - -#[tokio::main] -async fn main() { - commonmark().await; - punctuation().await; -} - -async fn commonmark() { - let url = "https://raw.githubusercontent.com/commonmark/commonmark-spec/0.30/spec.txt"; - let data_url = "commonmark-data.txt"; - let code_url = "tests/commonmark.rs"; - - let value = if let Ok(value) = fs::read_to_string(data_url) { - value - } else { - let value = reqwest::get(url).await.unwrap().text().await.unwrap(); - - fs::write(data_url, value.clone()).unwrap(); - - value - }; - - let re = Regex::new(r"(?m)(?:^`{32} example\n[\s\S]*?\n`{32}$|^#{1,6} *(.*)$)").unwrap(); - let re_heading_prefix = Regex::new(r"#{1,6} ").unwrap(); - let re_in_out = Regex::new(r"\n\.(?:\n|$)").unwrap(); - let mut current_heading = None; - let mut number = 1; - - let value = Regex::new(r"[\s\S]*") - .unwrap() - .replace(&value, ""); - let value = Regex::new(r"โ†’").unwrap().replace_all(&value, "\t"); - let mut cases = vec![]; - - for mat in re.find_iter(&value) { - let mut lines = mat.as_str().lines().collect::>(); - - if lines.len() == 1 { - current_heading = Some(re_heading_prefix.replace(lines[0], "").to_string()); - } else { - lines.remove(0); - lines.pop(); - let section = current_heading.as_ref().unwrap(); - let case = lines.join("\n"); - let parts = re_in_out.split(&case).collect::>(); - let input = format!("{}\n", parts[0]); - let output = if parts[1].is_empty() { - "".into() - } else { - format!("{}\n", parts[1]) - }; - - let test = format!(" assert_eq!(\n to_html_with_options(\n r###\"{}\"###,\n &danger\n )?,\n r###\"{}\"###,\n r###\"{} ({})\"###\n);", input, output, section, number); - - cases.push(test); - - number += 1; - } - } - - let doc = format!( - "//! `CommonMark` test suite. - -// > ๐Ÿ‘‰ **Important**: this module is generated by `build.rs`. -// > It is generate from the latest CommonMark website. - -use markdown::{{to_html_with_options, CompileOptions, Options}}; -use pretty_assertions::assert_eq; - -#[rustfmt::skip] -#[test] -fn commonmark() -> Result<(), String> {{ - let danger = Options {{ - compile: CompileOptions {{ - allow_dangerous_html: true, - allow_dangerous_protocol: true, - ..CompileOptions::default() - }}, - ..Options::default() - }}; - -{} - - Ok(()) -}} -", - cases.join("\n\n") - ); - - fs::write(code_url, doc).unwrap(); -} - -async fn punctuation() { - let url = "https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt"; - let data_url = "unicode-data.txt"; - let code_url = "src/util/unicode.rs"; - - let value = if let Ok(value) = fs::read_to_string(data_url) { - value - } else { - let value = reqwest::get(url).await.unwrap().text().await.unwrap(); - - fs::write(data_url, value.clone()).unwrap(); - - value - }; - - let search = [ - "Pc", // Punctuation, Connector - "Pd", // Punctuation, Dash - "Pe", // Punctuation, Close - "Pf", // Punctuation, FinalQuote - "Pi", // Punctuation, InitialQuote - "Po", // Punctuation, Other - "Ps", // Punctuation, Open - ]; - - let found = value - .lines() - .map(|line| line.split(';').collect::>()) - .map(|cells| (cells[0], cells[2])) - .filter(|c| search.contains(&c.1)) - .map(|c| c.0) - .collect::>(); - - let doc = format!( - "//! Info on Unicode. - -/// List of characters that are considered punctuation. -/// -/// > ๐Ÿ‘‰ **Important**: this module is generated by `build.rs`. -/// > It is generate from the latest Unicode data. -/// -/// Rust does not contain an `is_punctuation` method on `char`, while it does -/// support [`is_ascii_alphanumeric`](char::is_ascii_alphanumeric). -/// -/// `CommonMark` handles attention (emphasis, strong) markers based on what -/// comes before or after them. -/// One such difference is if those characters are Unicode punctuation. -/// -/// ## References -/// -/// * [*ยง 2.1 Characters and lines* in `CommonMark`](https://spec.commonmark.org/0.30/#unicode-punctuation-character) -pub const PUNCTUATION: [char; {}] = [ -{} -]; -", - found.len(), - found.iter().map(|d| format!(" '\\u{{{}}}',", d)).collect::>().join("\n") - ); - - fs::write(code_url, doc).unwrap(); -} -- cgit