// the code here is taken (and slightly modified) from https://github.com/XadillaX/crlf2lf // // convert CRLF (0D 0A) to LF ( 0A ) // use std::fs::File; use std::io::prelude::*; use std::io::{self, Read}; use std::path::Path; struct Bytes { bytes: Vec, len: usize, index: usize, } impl Bytes { fn new(bytes: Vec) -> Bytes { let l = bytes.len(); Bytes { bytes, len: l, index: 0, } } } #[derive(Debug)] enum Item { Pair(u8, u8), Single(u8), } impl Iterator for Bytes { type Item = Item; fn next(&mut self) -> Option { let cur_index = self.index; self.index += 1; if cur_index == self.len { return None; } if cur_index + 1 == self.len { Some(Item::Single(self.bytes[cur_index])) } else { // if self.index < self.len { Some(Item::Pair(self.bytes[cur_index], self.bytes[cur_index + 1])) } } } // The caller has to make sure that crlf2lf is only // used for text files pub fn wrong_line_endings2crlf(fname: &str) -> Result<(), io::Error> { let path = Path::new(fname); let mut hdl_in; match File::open(path) { Ok(f) => { hdl_in = f; } Err(e) => return Err(e), }; let mut buffer: Vec = Vec::new(); // read the whole file if let Err(e) = hdl_in.read_to_end(&mut buffer) { return Err(e); }; let another_vec: Vec = fix_line_endings2crlf(&buffer); let mut hdl_out; match File::create(path) { Ok(f) => { hdl_out = f; } Err(e) => return Err(e), }; // write back match hdl_out.write(&another_vec) { Ok(_) => Ok(()), Err(e) => Err(e), } } // The caller has to make sure that crlf2lf is only // used for text files pub fn wrong_line_endings2lf(fname: &str) -> Result<(), io::Error> { let path = Path::new(fname); let mut hdl_in; match File::open(path) { Ok(f) => { hdl_in = f; } Err(e) => return Err(e), }; let mut buffer: Vec = Vec::new(); // read the whole file if let Err(e) = hdl_in.read_to_end(&mut buffer) { return Err(e); }; let another_vec: Vec = fix_line_endings2lf(&buffer); let mut hdl_out; match File::create(path) { Ok(f) => { hdl_out = f; } Err(e) => return Err(e), }; // write back match hdl_out.write(&another_vec) { Ok(_) => Ok(()), Err(e) => Err(e), } } pub fn fix_line_endings2lf(buffer: &[u8]) -> Vec { let mut another_vec: Vec = Vec::new(); const CR: u8 = 0x0d; // 13 const LF: u8 = 0x0a; // 10 let b = Bytes::new(buffer.to_vec()); let mut skip = false; for v in b { if skip { skip = false; continue; } match v { Item::Pair(CR, LF) | Item::Pair(LF, CR) => { another_vec.push(LF); skip = true; } Item::Pair(CR, _) | Item::Pair(LF, _) | Item::Single(CR) | Item::Single(LF) => { another_vec.push(LF); } Item::Pair(c, _) | Item::Single(c) => { another_vec.push(c); } } } // if there is no line ending at the end of a fill add it if another_vec[another_vec.len() - 1] != LF { another_vec.push(LF); } another_vec } pub fn fix_line_endings2crlf(buffer: &[u8]) -> Vec { let mut another_vec: Vec = Vec::new(); const CR: u8 = 0x0d; // 13 \r const LF: u8 = 0x0a; // 10 \n let b = Bytes::new(buffer.to_vec()); let mut skip = false; for v in b { if skip { skip = false; continue; } match v { Item::Pair(CR, LF) | Item::Pair(LF, CR) => { another_vec.push(CR); another_vec.push(LF); skip = true; } Item::Pair(CR, _) | Item::Pair(LF, _) | Item::Single(CR) | Item::Single(LF) => { another_vec.push(CR); another_vec.push(LF); } Item::Pair(c, _) | Item::Single(c) => { another_vec.push(c); } } } let avlen = another_vec.len(); if avlen == 1 || (another_vec[avlen - 1] != LF && another_vec[avlen - 2] != CR) { another_vec.push(CR); another_vec.push(LF); } another_vec } #[test] fn test_fix_line_endings2crlf() { let testfiles = vec![ ("a", "a\r\n"), ("no line ending at all", "no line ending at all\r\n"), ("\rbla", "\r\nbla\r\n"), ("\r\r", "\r\n\r\n"), ("\r\nbla", "\r\nbla\r\n"), ("\r\n\r\nbla", "\r\n\r\nbla\r\n"), ("bla\r\nbla", "bla\r\nbla\r\n"), ("bla\r\n\r\nbla", "bla\r\n\r\nbla\r\n"), ("bla\r\nbla\r\nbla", "bla\r\nbla\r\nbla\r\n"), ("\n\r", "\r\n"), ("\n\r\n\r", "\r\n\r\n"), ("\n\rbla\n\r", "\r\nbla\r\n"), ]; for (s, rs) in testfiles { let bs = s.as_bytes().to_vec(); let brs = rs.as_bytes().to_vec(); assert_eq!(fix_line_endings2crlf(&bs), brs); } } #[test] fn test_fix_line_endings() { let testfiles = vec![ ("no line ending at all", "no line ending at all\n"), ("\r", "\n"), ("\n\r", "\n"), ("\r\n", "\n"), ("\r\n\n", "\n\n"), ("CR at end of file\r", "CR at end of file\n"), ("LF at end of file\n", "LF at end of file\n"), ("CRLF at end of file\r\n", "CRLF at end of file\n"), ("Two CR at end of file\r\r", "Two CR at end of file\n\n"), ("Two LF at end of file\n\n", "Two LF at end of file\n\n"), ( "Two CRLF at end of file\r\n\r\n", "Two CRLF at end of file\n\n", ), ( "Two CR in the middle\r\r of file", "Two CR in the middle\n\n of file\n", ), ( "Two LF in the middle\n\n of file", "Two LF in the middle\n\n of file\n", ), ( "Two CRLF in the middle\r\n\r\n of file", "Two CRLF in the middle\n\n of file\n", ), ( "Mixed CR, CRLF\rline2\r\nline3\r\n", "Mixed CR, CRLF\nline2\nline3\n", ), ( "Mixed CR, CRLF\r\r\nline2\r\nline3\r\n", "Mixed CR, CRLF\n\nline2\nline3\n", ), // ("Mixed CR, CRLF\n\r\r\nline2\r\nline3\r\n", "Mixed CR, CRLF\n\n\nline2\nline3\n"), ( "Mixed CR, LFCR\rline2\n\rline3\r\n", "Mixed CR, LFCR\nline2\nline3\n", ), //("line1\nebla", "line1\rebla"), //("mor\r\nebla", "mor\rebla"), ]; for (s, rs) in testfiles { let bs = s.as_bytes().to_vec(); let brs = rs.as_bytes().to_vec(); assert_eq!(fix_line_endings2lf(&bs), brs); } }