escape8259-0.5.3/.cargo_vcs_info.json0000644000000001360000000000100126310ustar { "git": { "sha1": "da55df5b65157ac2251b3fab6e4232f4d47bfa70" }, "path_in_vcs": "" }escape8259-0.5.3/.gitignore000064400000000000000000000000231046102023000134040ustar 00000000000000/target Cargo.lock escape8259-0.5.3/Cargo.toml0000644000000017120000000000100106300ustar # THIS FILE IS AUTOMATICALLY GENERATED BY CARGO # # When uploading crates to the registry Cargo will automatically # "normalize" Cargo.toml files for maximal compatibility # with all versions of Cargo and also rewrite `path` dependencies # to registry (e.g., crates.io) dependencies. # # If you are reading this file be aware that the original Cargo.toml # will likely look very different (and much more reasonable). # See Cargo.toml.orig for the original contents. [package] edition = "2018" name = "escape8259" version = "0.5.3" authors = ["Eric Seppanen "] exclude = ["Cranky.toml"] description = "RFC8259-compliant string escaping and un-escaping" readme = "README.md" keywords = [ "json", "rfc8259", "cddl", ] categories = ["encoding"] license = "MIT" repository = "https://github.com/ericseppanen/escape8259" [package.metadata.release] pre-release-commit-message = "release {{version}}" [dev-dependencies.rustversion] version = "1.0" escape8259-0.5.3/Cargo.toml.orig000064400000000000000000000007401046102023000143110ustar 00000000000000[package] name = "escape8259" description = "RFC8259-compliant string escaping and un-escaping" keywords = ["json", "rfc8259", "cddl"] categories = ["encoding"] version = "0.5.3" repository = "https://github.com/ericseppanen/escape8259" license = "MIT" authors = ["Eric Seppanen "] readme = "README.md" edition = "2018" exclude = ["Cranky.toml"] [dev-dependencies] rustversion = "1.0" [package.metadata.release] pre-release-commit-message = "release {{version}}" escape8259-0.5.3/LICENSE000064400000000000000000000020561046102023000124310ustar 00000000000000MIT License Copyright (c) 2019 Eric Seppanen Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. escape8259-0.5.3/README.md000064400000000000000000000015471046102023000127070ustar 00000000000000 `escape8259` performs RFC8259-compliant string escaping and un-escaping. [RFC8259] is a JSON encoding standard. Many JSON encoders exist, but other RFCs use the same string escaping mechanism, so it's useful to be able to access the string escaping functions by themselves. # Examples ```rust use escape8259::{escape, unescape}; let s = "A null (\0) and a double-quote (\")"; assert_eq!(escape(s), r#"A null (\u0000) and a double-quote (\")"#); let crab = r#"This is a crab: \ud83e\udd80"#; assert_eq!(unescape(crab).unwrap(), "This is a crab: 🦀"); // We accept encodings that weren't really necessary. assert_eq!(unescape(r#"\u0041\n"#).unwrap(), "A\n"); let multiline = r#"hello world"#; assert_eq!(escape(multiline), r#"hello\n world"#); ``` [RFC8259]: https://tools.ietf.org/html/rfc8259 escape8259-0.5.3/src/lib.rs000064400000000000000000000236751046102023000133410ustar 00000000000000//! `escape8259` performs RFC8259-compliant string escaping and un-escaping. //! //! [RFC8259] is a JSON encoding standard. Many JSON encoders exist, but other //! RFCs use the same string escaping mechanism, so it's useful to be able to //! access the string escaping functions by themselves. //! //! # Examples //! //! ``` //! use escape8259::{escape, unescape}; //! //! let s = "A null (\0) and a double-quote (\")"; //! assert_eq!(escape(s), r#"A null (\u0000) and a double-quote (\")"#); //! //! let crab = r#"This is a crab: \ud83e\udd80"#; //! assert_eq!(unescape(crab).unwrap(), "This is a crab: 🦀"); //! //! // We accept encodings that weren't really necessary. //! assert_eq!(unescape(r#"\u0041\n"#).unwrap(), "A\n"); //! //! let multiline = r#"hello //! world"#; //! assert_eq!(escape(multiline), r#"hello\n world"#); //! ``` //! //! [RFC8259]: https://tools.ietf.org/html/rfc8259 #![warn(missing_docs)] #![forbid(unsafe_code)] #![warn(clippy::cast_possible_truncation)] use std::char::decode_utf16; use std::fmt::{Display, Write}; /// An error occurred while unescaping. #[allow(clippy::empty_structs_with_brackets)] // FIXME: correct this if releasing a new major version. #[derive(Debug, Clone, Eq, PartialEq)] pub struct UnescapeError {} impl Display for UnescapeError { fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { f.write_str("failed rfc8259 unescape") } } impl std::error::Error for UnescapeError {} type UnescapeResult = Result; // Used to collect output characters and queue u16 values for translation. struct UnescapeState { // The accumulated characters out: String, // Store a fragment of a large character for later decoding stash: u16, } impl UnescapeState { fn new() -> UnescapeState { UnescapeState { out: String::new(), stash: 0, } } // Collect a new character fn push_char(&mut self, c: char) -> UnescapeResult<()> { if self.stash != 0 { return Err(UnescapeError {}); } self.out.push(c); Ok(()) } // Collect a new UTF16 word. This can either be one whole character, // or part of a larger character. fn push_u16(&mut self, x: u16) -> UnescapeResult<()> { let surrogate = (0xD800..=0xDFFF).contains(&x); match (self.stash, surrogate) { (0, false) => { // The std library only provides utf16 decode of an iterator, // so to decode a single character we wrap it in an array. // Hopefully the compiler will elide most of this extra work. let words = [x]; match decode_utf16(words.iter().copied()).next() { Some(Ok(c)) => { self.out.push(c); } _ => return Err(UnescapeError {}), } } (0, true) => self.stash = x, (_, false) => { return Err(UnescapeError {}); } (w, true) => { let words = [w, x]; match decode_utf16(words.iter().copied()).next() { Some(Ok(c)) => { self.out.push(c); self.stash = 0; } _ => return Err(UnescapeError {}), } } } Ok(()) } // If we queued up part of a UTF-16 encoded word but didn't // finish it, return an error. Otherwise, consume self and // return the accumulated String. fn finalize(self) -> UnescapeResult { if self.stash != 0 { return Err(UnescapeError {}); } Ok(self.out) } } fn parse_u16(s: &mut S) -> UnescapeResult where S: Iterator, { // Placeholder character in case the input doesn't have the 4 chars we want. let placeholders = std::iter::repeat('\0'); let hexnum: String = s.chain(placeholders).take(4).collect(); u16::from_str_radix(&hexnum, 16).map_err(|_| UnescapeError {}) } // RFC8259 says non-escaped characters must be in one of the following ranges: // %x20-21 / %x23-5B / %x5D-10FFFF fn is_safe_char(c: char) -> bool { let safe_ranges = [(0x20..=0x21), (0x23..=0x5B), (0x5D..=0x10FFFF)]; let cv = c as u32; safe_ranges.iter().any(|range| range.contains(&cv)) } /// Un-escape a string, following RFC8259 rules. /// /// The only allowed single-character escapes are: /// `\" \\ \/ /b /f /n /r /t` /// /// Any other character may be escaped in UTF-16 form: /// `\uXXXX` or `\uXXXX\uXXXX` /// /// Characters in the ranges `0x20-21`, `0x23-5B`, `0x5D-10FFFF` /// may appear un-escaped in the input. #[inline] pub fn unescape(s: S) -> UnescapeResult where S: AsRef, { unescape_inner(s.as_ref()) } fn unescape_inner(s: &str) -> UnescapeResult { let mut state = UnescapeState::new(); let mut ins = s.chars(); while let Some(c) = ins.next() { if c == '\\' { match ins.next() { None => { return Err(UnescapeError {}); } Some(d) => { match d { '"' | '\\' | '/' => state.push_char(d)?, 'b' => state.push_char('\x08')?, // backspace 'f' => state.push_char('\x0C')?, // formfeed 'n' => state.push_char('\n')?, // linefeed 'r' => state.push_char('\r')?, // carriage return 't' => state.push_char('\t')?, // tab 'u' => { let val = parse_u16(&mut ins)?; state.push_u16(val)?; } _ => { return Err(UnescapeError {}); } } } } } else if is_safe_char(c) { state.push_char(c)?; } else { return Err(UnescapeError {}); } } state.finalize() } // %x22 / ; " quotation mark U+0022 // %x5C / ; \ reverse solidus U+005C // %x2F / ; / solidus U+002F // %x62 / ; b backspace U+0008 // %x66 / ; f form feed U+000C // %x6E / ; n line feed U+000A // %x72 / ; r carriage return U+000D // %x74 / ; t tab U+0009 // %x75 4HEXDIG ) ; uXXXX U+XXXX fn force_escape(c: char, out: &mut String) { let c = c as u32; match c { 0x08 => out.push_str("\\b"), 0x09 => out.push_str("\\t"), 0x0A => out.push_str("\\n"), 0x0C => out.push_str("\\f"), 0x0D => out.push_str("\\r"), 0x22 => out.push_str("\\\""), 0x5C => out.push_str("\\\\"), _ => { // RFC8259 allows unicode characters natively, so there is no need // to convert everything into \uXXXX form. The only thing that's // required to use that form are the ASCII control characters, // which will never require more than one \uXXXX value. if c >= 0x20 { panic!("force_escape unnecessary encoding requested"); } write!(out, "\\u{:04x}", c).unwrap(); } } } /// Escape a string, following RFC8259 rules. /// /// Only characters that require escaping will be escaped: /// quotation mark `?`, /// reverse solidus `\` (backslash), /// and the control characters (`0x00-1F`). #[inline] pub fn escape(s: S) -> String where S: AsRef, { escape_inner(s.as_ref()) } fn escape_inner(s: &str) -> String { let mut out = String::new(); for c in s.chars() { if is_safe_char(c) { out.push(c); } else { force_escape(c, &mut out); } } out } #[cfg(test)] mod tests { use super::*; #[rustversion::attr(since(1.46), track_caller)] fn assert_round_trip(s: &str) { assert_eq!(s, unescape(&escape(s)).unwrap()); } #[test] fn test_round_trip() { assert_round_trip("abc"); assert_round_trip("\n\r\t\x08\x0C\x00"); assert_round_trip(r#"\"#); assert_round_trip(r#"""#); assert_round_trip("Σ𝄞"); assert_round_trip(r#"\𝄞"#); assert_round_trip(r#"(╯°□°）╯︵ ┻━┻"#); } #[test] fn test_escape() { assert_eq!(escape("\0"), r#"\u0000"#); assert_eq!(escape("\n"), r#"\n"#); assert_eq!(escape(r#"\"#), r#"\\"#); assert_eq!(escape(r#"""#), r#"\""#); assert_eq!(escape("Σ"), "Σ"); // U+03A3 assert_eq!(escape("𝄞"), "𝄞"); // U+1D11E } #[test] fn test_unescape() { assert_eq!(unescape(&r#"abc"#), Ok("abc".into())); assert_eq!(unescape(&r#"ab\nc"#), Ok("ab\nc".into())); assert_eq!(unescape(r#"ab\zc"#), Err(UnescapeError {})); assert_eq!(unescape(r#" \"abc\" "#), Ok(" \"abc\" ".into())); assert_eq!(unescape(r#"𝄞"#), Ok("𝄞".into())); assert_eq!(unescape(r#"\𝄞"#), Err(UnescapeError {})); assert_eq!(unescape(r#"\uD834\uDD1E"#), Ok("𝄞".into())); assert_eq!(unescape(r#"\uD834"#), Err(UnescapeError {})); assert_eq!(unescape(r#"\uDD1E"#), Err(UnescapeError {})); assert_eq!(unescape("\t"), Err(UnescapeError {})); } #[test] fn test_generic_asref() { assert_eq!(escape("\n"), r#"\n"#); assert_eq!(escape(String::from("\n")), r#"\n"#); assert_eq!(escape(&String::from("\n")), r#"\n"#); assert_eq!(unescape("abc"), Ok("abc".into())); assert_eq!(unescape(String::from("abc")), Ok("abc".into())); assert_eq!(unescape(&String::from("abc")), Ok("abc".into())); } #[test] fn test_error_impl() { // This won't compile if UnescapeError doesn't impl Display + Error. let e = UnescapeError {}; let _x: Box = e.into(); } }