Browse Source
Remove text_processing
Remove text_processing
Unwrap text_processing into melib In preparation for uploading meli as a separate crate on crates.io.async
27 changed files with 129 additions and 156 deletions
-
17Cargo.lock
-
5Cargo.toml
-
20melib/Cargo.toml
-
78melib/build.rs
-
2melib/src/email/compose/mime.rs
-
2melib/src/lib.rs
-
8melib/src/text_processing/grapheme_clusters.rs
-
16melib/src/text_processing/line_break.rs
-
2melib/src/text_processing/mod.rs
-
4melib/src/text_processing/tables.rs
-
0melib/src/text_processing/types.rs
-
0melib/src/text_processing/wcwidth.rs
-
2melib/src/thread.rs
-
2src/bin.rs
-
1src/components.rs
-
1src/components/contacts/contact_list.rs
-
2src/conf.rs
-
2src/conf/accounts.rs
-
2src/mailcap.rs
-
2src/terminal/cells.rs
-
2src/terminal/embed/grid.rs
-
2src/terminal/text_editing.rs
-
2src/workers.rs
-
3testing/Cargo.toml
-
4testing/src/email_parse.rs
-
10text_processing/Cargo.toml
-
94text_processing/build.rs
@ -1,5 +1,5 @@ |
|||
/*
|
|||
* meli - text_processing crate.
|
|||
* meli - text_processing mod.
|
|||
*
|
|||
* Copyright 2017-2020 Manos Pitsidianakis
|
|||
*
|
|||
@ -1,10 +0,0 @@ |
|||
[package] |
|||
name = "text_processing" |
|||
version = "0.4.1" |
|||
authors = ["Manos Pitsidianakis <el13635@mail.ntua.gr>"] |
|||
workspace = ".." |
|||
edition = "2018" |
|||
build = "build.rs" |
|||
|
|||
[dependencies] |
|||
unicode-segmentation = "1.2.1" |
|||
@ -1,94 +0,0 @@ |
|||
/*
|
|||
* meli - text_processing crate.
|
|||
*
|
|||
* Copyright 2017-2020 Manos Pitsidianakis
|
|||
*
|
|||
* This file is part of meli.
|
|||
*
|
|||
* meli is free software: you can redistribute it and/or modify
|
|||
* it under the terms of the GNU General Public License as published by
|
|||
* the Free Software Foundation, either version 3 of the License, or
|
|||
* (at your option) any later version.
|
|||
*
|
|||
* meli is distributed in the hope that it will be useful,
|
|||
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|||
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|||
* GNU General Public License for more details.
|
|||
*
|
|||
* You should have received a copy of the GNU General Public License
|
|||
* along with meli. If not, see <http://www.gnu.org/licenses/>.
|
|||
*/
|
|||
|
|||
const LINE_BREAK_TABLE_URL: &str = "http://www.unicode.org/Public/UCD/latest/ucd/LineBreak.txt";
|
|||
use std::fs::File;
|
|||
use std::io::prelude::*;
|
|||
use std::io::BufReader;
|
|||
use std::path::{Path, PathBuf};
|
|||
use std::process::Command;
|
|||
|
|||
include!("src/types.rs");
|
|||
|
|||
fn main() -> Result<(), std::io::Error> {
|
|||
let mod_path = Path::new("src/tables.rs");
|
|||
if mod_path.exists() {
|
|||
eprintln!(
|
|||
"{} already exists, delete it if you want to replace it.",
|
|||
mod_path.display()
|
|||
);
|
|||
std::process::exit(0);
|
|||
}
|
|||
let mut tmpdir_path = PathBuf::from(
|
|||
std::str::from_utf8(&Command::new("mktemp").arg("-d").output()?.stdout)
|
|||
.unwrap()
|
|||
.trim(),
|
|||
);
|
|||
tmpdir_path.push("LineBreak.txt");
|
|||
Command::new("curl")
|
|||
.args(&["-o", tmpdir_path.to_str().unwrap(), LINE_BREAK_TABLE_URL])
|
|||
.output()?;
|
|||
|
|||
let file = File::open(&tmpdir_path)?;
|
|||
let buf_reader = BufReader::new(file);
|
|||
|
|||
let mut line_break_table: Vec<(u32, u32, LineBreakClass)> = Vec::with_capacity(3800);
|
|||
for line in buf_reader.lines() {
|
|||
let line = line.unwrap();
|
|||
if line.starts_with('#') || line.starts_with(' ') || line.is_empty() {
|
|||
continue;
|
|||
}
|
|||
let tokens: &str = line.split_whitespace().next().unwrap();
|
|||
|
|||
let semicolon_idx: usize = tokens.chars().position(|c| c == ';').unwrap();
|
|||
/* LineBreak.txt list is ascii encoded so we can assume each char takes one byte: */
|
|||
let chars_str: &str = &tokens[..semicolon_idx];
|
|||
|
|||
let mut codepoint_iter = chars_str.split("..");
|
|||
|
|||
let first_codepoint: u32 =
|
|||
u32::from_str_radix(std::dbg!(codepoint_iter.next().unwrap()), 16).unwrap();
|
|||
|
|||
let sec_codepoint: u32 = codepoint_iter
|
|||
.next()
|
|||
.map(|v| u32::from_str_radix(std::dbg!(v), 16).unwrap())
|
|||
.unwrap_or(first_codepoint);
|
|||
let class = &tokens[semicolon_idx + 1..semicolon_idx + 1 + 2];
|
|||
line_break_table.push((first_codepoint, sec_codepoint, LineBreakClass::from(class)));
|
|||
}
|
|||
|
|||
let mut file = File::create(&mod_path)?;
|
|||
file.write_all(b"use crate::types::LineBreakClass::*;\n")
|
|||
.unwrap();
|
|||
file.write_all(b"use crate::types::LineBreakClass;\n\n")
|
|||
.unwrap();
|
|||
file.write_all(b"const line_break_rules: &'static [(u32, u32, LineBreakClass)] = &[\n")
|
|||
.unwrap();
|
|||
for l in &line_break_table {
|
|||
file.write_all(format!(" (0x{:X}, 0x{:X}, {:?}),\n", l.0, l.1, l.2).as_bytes())
|
|||
.unwrap();
|
|||
}
|
|||
file.write_all(b"];").unwrap();
|
|||
std::fs::remove_file(&tmpdir_path).unwrap();
|
|||
tmpdir_path.pop();
|
|||
std::fs::remove_dir(&tmpdir_path).unwrap();
|
|||
Ok(())
|
|||
}
|
|||
Write
Preview
Loading…
Cancel
Save
Reference in new issue