Move license template parsing into submodule

This commit is contained in:
David Lukes 2018-02-26 12:49:12 +01:00
parent e48d7f3ebb
commit 310c1146f2
3 changed files with 177 additions and 175 deletions

View file

@ -0,0 +1,174 @@
use regex;
/// Convert the license template into a string which can be turned into a regex.
///
/// The license template could use regex syntax directly, but that would require a lot of manual
/// escaping, which is inconvenient. It is therefore literal by default, with optional regex
/// subparts delimited by `{` and `}`. Additionally:
///
/// - to insert literal `{`, `}` or `\`, escape it with `\`
/// - an empty regex placeholder (`{}`) is shorthand for `{.*?}`
///
/// This function parses this input format and builds a properly escaped *string* representation of
/// the equivalent regular expression. It **does not** however guarantee that the returned string is
/// a syntactically valid regular expression.
///
/// # Examples
///
/// ```
/// # use rustfmt_config::license;
/// assert_eq!(
/// license::parse_template(
/// r"
/// // Copyright {\d+} The \} Rust \\ Project \{ Developers. See the {([A-Z]+)}
/// // file at the top-level directory of this distribution and at
/// // {}.
/// //
/// // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
/// // http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
/// // <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
/// // option. This file may not be copied, modified, or distributed
/// // except according to those terms.
/// "
/// ).unwrap(),
/// r"^
/// // Copyright \d+ The \} Rust \\ Project \{ Developers\. See the ([A-Z]+)
/// // file at the top\-level directory of this distribution and at
/// // .*?\.
/// //
/// // Licensed under the Apache License, Version 2\.0 <LICENSE\-APACHE or
/// // http://www\.apache\.org/licenses/LICENSE\-2\.0> or the MIT license
/// // <LICENSE\-MIT or http://opensource\.org/licenses/MIT>, at your
/// // option\. This file may not be copied, modified, or distributed
/// // except according to those terms\.
/// "
/// );
/// ```
pub fn parse_template(template: &str) -> Result<String, String> {
// the template is parsed using a state machine
enum State {
Lit,
LitEsc,
// the u32 keeps track of brace nesting
Re(u32),
ReEsc(u32),
}
let mut parsed = String::from("^");
let mut buffer = String::new();
let mut state = State::Lit;
let mut linum = 1;
// keeps track of last line on which a regex placeholder was started
let mut open_brace_line = 0;
for chr in template.chars() {
if chr == '\n' {
linum += 1;
}
state = match state {
State::Lit => match chr {
'{' => {
parsed.push_str(&regex::escape(&buffer));
buffer.clear();
open_brace_line = linum;
State::Re(1)
}
'}' => return Err(format!("escape or balance closing brace on l. {}", linum)),
'\\' => State::LitEsc,
_ => {
buffer.push(chr);
State::Lit
}
},
State::LitEsc => {
buffer.push(chr);
State::Lit
}
State::Re(brace_nesting) => {
match chr {
'{' => {
buffer.push(chr);
State::Re(brace_nesting + 1)
}
'}' => {
match brace_nesting {
1 => {
// default regex for empty placeholder {}
if buffer.is_empty() {
buffer = ".*?".to_string();
}
parsed.push_str(&buffer);
buffer.clear();
State::Lit
}
_ => {
buffer.push(chr);
State::Re(brace_nesting - 1)
}
}
}
'\\' => {
buffer.push(chr);
State::ReEsc(brace_nesting)
}
_ => {
buffer.push(chr);
State::Re(brace_nesting)
}
}
}
State::ReEsc(brace_nesting) => {
buffer.push(chr);
State::Re(brace_nesting)
}
}
}
match state {
State::Re(_) | State::ReEsc(_) => {
return Err(format!(
"escape or balance opening brace on l. {}",
open_brace_line
));
}
State::LitEsc => return Err(format!("incomplete escape sequence on l. {}", linum)),
_ => (),
}
parsed.push_str(&regex::escape(&buffer));
Ok(parsed)
}
#[cfg(test)]
mod test {
use super::parse_template;
#[test]
fn test_parse_license_template() {
assert_eq!(
parse_template("literal (.*)").unwrap(),
r"^literal \(\.\*\)"
);
assert_eq!(parse_template(r"escaping \}").unwrap(), r"^escaping \}");
assert!(parse_template("unbalanced } without escape").is_err());
assert_eq!(
parse_template(r"{\d+} place{-?}holder{s?}").unwrap(),
r"^\d+ place-?holders?"
);
assert_eq!(parse_template("default {}").unwrap(), "^default .*?");
assert_eq!(
parse_template(r"unbalanced nested braces {\{{3}}").unwrap(),
r"^unbalanced nested braces \{{3}"
);
assert_eq!(
parse_template("parsing error }").unwrap_err(),
"escape or balance closing brace on l. 1"
);
assert_eq!(
parse_template("parsing error {\nsecond line").unwrap_err(),
"escape or balance opening brace on l. 1"
);
assert_eq!(
parse_template(r"parsing error \").unwrap_err(),
"incomplete escape sequence on l. 1"
);
}
}

View file

@ -408,7 +408,7 @@ macro_rules! create_config {
license_template_path, e);
return;
};
let license_template_parsed = match parse_license_template(&license_template_str) {
let license_template_parsed = match license::parse_template(&license_template_str) {
Ok(string) => string,
Err(e) => {
eprintln!("Warning: unable to parse license template file {:?}: {}",

View file

@ -25,6 +25,7 @@ mod options;
pub mod file_lines;
pub mod lists;
pub mod summary;
pub mod license;
use config::config_type::ConfigType;
use config::file_lines::FileLines;
@ -174,145 +175,9 @@ pub fn get_toml_path(dir: &Path) -> Result<Option<PathBuf>, Error> {
Ok(None)
}
/// Convert the license template into a string which can be turned into a regex.
///
/// The license template could use regex syntax directly, but that would require a lot of manual
/// escaping, which is inconvenient. It is therefore literal by default, with optional regex
/// subparts delimited by `{` and `}`. Additionally:
///
/// - to insert literal `{`, `}` or `\`, escape it with `\`
/// - an empty regex placeholder (`{}`) is shorthand for `{.*?}`
///
/// This function parses this input format and builds a properly escaped *string* representation of
/// the equivalent regular expression. It **does not** however guarantee that the returned string is
/// a syntactically valid regular expression.
///
/// # Examples
///
/// ```
/// assert_eq!(
/// rustfmt_config::parse_license_template(
/// r"
/// // Copyright {\d+} The \} Rust \\ Project \{ Developers. See the {([A-Z]+)}
/// // file at the top-level directory of this distribution and at
/// // {}.
/// //
/// // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
/// // http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
/// // <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
/// // option. This file may not be copied, modified, or distributed
/// // except according to those terms.
/// "
/// ).unwrap(),
/// r"^
/// // Copyright \d+ The \} Rust \\ Project \{ Developers\. See the ([A-Z]+)
/// // file at the top\-level directory of this distribution and at
/// // .*?\.
/// //
/// // Licensed under the Apache License, Version 2\.0 <LICENSE\-APACHE or
/// // http://www\.apache\.org/licenses/LICENSE\-2\.0> or the MIT license
/// // <LICENSE\-MIT or http://opensource\.org/licenses/MIT>, at your
/// // option\. This file may not be copied, modified, or distributed
/// // except according to those terms\.
/// "
/// );
/// ```
pub fn parse_license_template(template: &str) -> Result<String, String> {
// the template is parsed using a state machine
enum State {
Lit,
LitEsc,
// the u32 keeps track of brace nesting
Re(u32),
ReEsc(u32),
}
let mut parsed = String::from("^");
let mut buffer = String::new();
let mut state = State::Lit;
let mut linum = 1;
// keeps track of last line on which a regex placeholder was started
let mut open_brace_line = 0;
for chr in template.chars() {
if chr == '\n' {
linum += 1;
}
state = match state {
State::Lit => match chr {
'{' => {
parsed.push_str(&regex::escape(&buffer));
buffer.clear();
open_brace_line = linum;
State::Re(1)
}
'}' => return Err(format!("escape or balance closing brace on l. {}", linum)),
'\\' => State::LitEsc,
_ => {
buffer.push(chr);
State::Lit
}
},
State::LitEsc => {
buffer.push(chr);
State::Lit
}
State::Re(brace_nesting) => {
match chr {
'{' => {
buffer.push(chr);
State::Re(brace_nesting + 1)
}
'}' => {
match brace_nesting {
1 => {
// default regex for empty placeholder {}
if buffer.is_empty() {
buffer = ".*?".to_string();
}
parsed.push_str(&buffer);
buffer.clear();
State::Lit
}
_ => {
buffer.push(chr);
State::Re(brace_nesting - 1)
}
}
}
'\\' => {
buffer.push(chr);
State::ReEsc(brace_nesting)
}
_ => {
buffer.push(chr);
State::Re(brace_nesting)
}
}
}
State::ReEsc(brace_nesting) => {
buffer.push(chr);
State::Re(brace_nesting)
}
}
}
match state {
State::Re(_) | State::ReEsc(_) => {
return Err(format!(
"escape or balance opening brace on l. {}",
open_brace_line
));
}
State::LitEsc => return Err(format!("incomplete escape sequence on l. {}", linum)),
_ => (),
}
parsed.push_str(&regex::escape(&buffer));
Ok(parsed)
}
#[cfg(test)]
mod test {
use super::{parse_license_template, Config};
use super::Config;
#[test]
fn test_config_set() {
@ -349,43 +214,6 @@ mod test {
assert_eq!(config.was_set().verbose(), false);
}
#[test]
fn test_parse_license_template() {
assert_eq!(
parse_license_template("literal (.*)").unwrap(),
r"^literal \(\.\*\)"
);
assert_eq!(
parse_license_template(r"escaping \}").unwrap(),
r"^escaping \}"
);
assert!(parse_license_template("unbalanced } without escape").is_err());
assert_eq!(
parse_license_template(r"{\d+} place{-?}holder{s?}").unwrap(),
r"^\d+ place-?holders?"
);
assert_eq!(
parse_license_template("default {}").unwrap(),
"^default .*?"
);
assert_eq!(
parse_license_template(r"unbalanced nested braces {\{{3}}").unwrap(),
r"^unbalanced nested braces \{{3}"
);
assert_eq!(
parse_license_template("parsing error }").unwrap_err(),
"escape or balance closing brace on l. 1"
);
assert_eq!(
parse_license_template("parsing error {\nsecond line").unwrap_err(),
"escape or balance opening brace on l. 1"
);
assert_eq!(
parse_license_template(r"parsing error \").unwrap_err(),
"incomplete escape sequence on l. 1"
);
}
// FIXME(#2183) these tests cannot be run in parallel because they use env vars
// #[test]
// fn test_as_not_nightly_channel() {