Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat: allow literal to be more flexible about escaped characters #38

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

feat: allow literal to be more flexible about escaped characters #38

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
90 changes: 55 additions & 35 deletions src/token/parse.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,8 @@
#[cfg(feature = "miette")]
use miette::{self, Diagnostic, LabeledSpan, SourceCode};
use nom::error::{VerboseError as NomError, VerboseErrorKind as NomErrorKind};
use nom::{InputTakeAtPosition, FindToken};
use nom::character::complete::none_of;
use nom::error::{VerboseError as NomError, VerboseErrorKind as NomErrorKind, ErrorKind};
use pori::{Located, Location, Stateful};
use std::borrow::Cow;
use std::fmt::{self, Display, Formatter};
Expand Down Expand Up @@ -235,37 +237,54 @@ pub fn parse(expression: &str) -> Result<Tokenized, ParseError> {
flags(move |_| move |input: Input<'i>| Ok((input, ())))(input)
}

fn literal(input: Input) -> ParseResult<TokenKind<Annotation>> {
combinator::map(
combinator::verify(
bytes::escaped_transform(
bytes::is_not("/?*$:<>()[]{},\\"),
'\\',
branch::alt((
combinator::value("?", bytes::tag("?")),
combinator::value("*", bytes::tag("*")),
combinator::value("$", bytes::tag("$")),
combinator::value(":", bytes::tag(":")),
combinator::value("<", bytes::tag("<")),
combinator::value(">", bytes::tag(">")),
combinator::value("(", bytes::tag("(")),
combinator::value(")", bytes::tag(")")),
combinator::value("[", bytes::tag("[")),
combinator::value("]", bytes::tag("]")),
combinator::value("{", bytes::tag("{")),
combinator::value("}", bytes::tag("}")),
combinator::value(",", bytes::tag(",")),
)),

fn literal<'i>(
cant_contain: &'i str,
) -> impl FnMut(Input<'i>) -> ParseResult<'i, TokenKind<'i, Annotation>>
{
fn is_not_both<'i>(
arr1: &'i str,
arr2: &'i str,
) -> impl Fn(Input<'i>) -> ParseResult<'i,Input<'i>>
{
move |i: Input| {
let e: ErrorKind = ErrorKind::IsNot;
i.split_at_position1_complete(|c| arr1.find_token(c) || arr2.find_token(c), e)
}
}

move |input: Input| {
combinator::map(
combinator::verify(
bytes::escaped_transform(
is_not_both("/?*$<>()[]{}\\", cant_contain),
'\\',
branch::alt((
combinator::value("?", bytes::tag("?")),
combinator::value("*", bytes::tag("*")),
combinator::value("$", bytes::tag("$")),
combinator::value("<", bytes::tag("<")),
combinator::value(">", bytes::tag(">")),
combinator::value("(", bytes::tag("(")),
combinator::value(")", bytes::tag(")")),
combinator::value("[", bytes::tag("[")),
combinator::value("]", bytes::tag("]")),
combinator::value("{", bytes::tag("{")),
combinator::value("}", bytes::tag("}")),
combinator::value(",", bytes::tag(",")),
combinator::value(":", bytes::tag(":")),
)),
),
|text: &str| !text.is_empty(),
),
|text: &str| !text.is_empty(),
),
move |text| {
TokenKind::Literal(Literal {
text: text.into(),
is_case_insensitive: input.state.flags.is_case_insensitive,
})
},
)(input)
move |text| {
TokenKind::Literal(Literal {
text: text.into(),
is_case_insensitive: input.state.flags.is_case_insensitive,
})
},
)(input)
}
}

fn separator(input: Input) -> ParseResult<TokenKind<Annotation>> {
Expand Down Expand Up @@ -404,7 +423,7 @@ pub fn parse(expression: &str) -> Result<Tokenized, ParseError> {
"sub-glob",
glob(move |input| {
combinator::peek(branch::alt((bytes::tag(":"), bytes::tag(">"))))(input)
}),
}, ":"),
),
error::context("bounds", bounds),
)),
Expand Down Expand Up @@ -469,7 +488,7 @@ pub fn parse(expression: &str) -> Result<Tokenized, ParseError> {
"sub-glob",
glob(move |input| {
combinator::peek(branch::alt((bytes::tag(","), bytes::tag("}"))))(input)
}),
}, ","),
),
),
|alternatives: Vec<Vec<_>>| Alternative::from(alternatives).into(),
Expand All @@ -480,6 +499,7 @@ pub fn parse(expression: &str) -> Result<Tokenized, ParseError> {

fn glob<'i>(
terminator: impl 'i + Clone + Parser<Input<'i>, Input<'i>, ErrorStack<'i>>,
literal_cant_contain: &'i str,
) -> impl Parser<Input<'i>, Vec<Token<'i, Annotation>>, ErrorStack<'i>> {
fn annotate<'i, F>(
parser: F,
Expand All @@ -496,7 +516,7 @@ pub fn parse(expression: &str) -> Result<Tokenized, ParseError> {
multi::many1(branch::alt((
annotate(error::context(
"literal",
sequence::preceded(flags_with_state, literal),
sequence::preceded(flags_with_state, literal(literal_cant_contain)),
)),
annotate(error::context(
"repetition",
Expand Down Expand Up @@ -533,7 +553,7 @@ pub fn parse(expression: &str) -> Result<Tokenized, ParseError> {
}
else {
let input = Input::new(Expression::from(expression), ParserState::default());
let tokens = combinator::all_consuming(glob(combinator::eof))(input)
let tokens = combinator::all_consuming(glob(combinator::eof, ""))(input)
.map(|(_, tokens)| tokens)
.map_err(|error| ParseError::new(expression, error))?;
Ok(Tokenized {
Expand Down
124 changes: 124 additions & 0 deletions tests/parse.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,124 @@
use std::path::PathBuf;

use build_fs_tree::{FileSystemTree, dir};

use wax::{Glob};

use build_fs_tree::{file, Build};
use tempfile::{self, TempDir};


#[test]
fn test_parse_litteral(){
let (_root, path) = temptree();

test("README.md", &["README.md"], &path);
test("tests/walk.rs", &["walk.rs"], &path);

test("extra:dots.txt", &["extra:dots.txt"], &path);
test("extra,comma.txt", &["extra,comma.txt"], &path);
}

#[test]
fn test_parse_wildcards(){
let (_root, path) = temptree();

test("RE*.md", &["README.md"], &path);
test("READM?.md", &["README.md"], &path);
test("**/*.md", &["guide.md", "README.md"], &path);

test("*:dots.txt", &["extra:dots.txt"], &path);
test("*,comma.txt", &["extra,comma.txt"], &path);
}

#[test]
fn test_parse_char_classes(){
let (_root, path) = temptree();

test("READM[AEIOU].md", &["README.md"], &path);
test("READM[A-Z].md", &["README.md"], &path);
test("READM[a-zA-Z].md", &["README.md"], &path);

test("READM[!AIOU].md", &["README.md"], &path);

test("extr[a-z]:dots.txt", &["extra:dots.txt"], &path);
test("extr[a-z],comma.txt", &["extra,comma.txt"], &path);
}

#[test]
fn test_parse_alternatives(){
let (_root, path) = temptree();

test("READM{A,E}.md", &["README.md"], &path);
test("READ{M?,Z?}.md", &["README.md"], &path);
test("README.{txt,md}", &["README.md"], &path);

test("extra{:dots,:nots}.txt", &["extra:dots.txt"], &path);
test("extra{\\,comma,\\,nots}.txt", &["extra,comma.txt"], &path);
}

#[test]
fn test_parse_repetitions(){
let (_root, path) = temptree();

test("<t*/>walk.rs", &["walk.rs"], &path);
test("<t*/:1>walk.rs", &["walk.rs"], &path);
test("<t*/:0,1>walk.rs", &["walk.rs"], &path);
test("fl<?:2>t.txt", &["fleet.txt"], &path);
test("fl<e:2>t.txt", &["fleet.txt"], &path);

test("extra:dots.txt", &["extra:dots.txt"], &path);
test("extra,comma.txt", &["extra,comma.txt"], &path);
}

#[test]
fn test_parse_combined(){
let (_root, path) = temptree();

test("extra{[:,\\-]dots,:nots}.txt", &["extra:dots.txt"], &path);
}




fn test(expression: &str, expected: &[&str], path: &PathBuf){
println!("\n\n\nTest {expression} {:?}", expected);
let glob = Glob::new(expression).unwrap();
println!("Captures: {:?}", glob.captures().collect::<Vec<_>>());
println!("Glob: {:#?}", glob);

let names: Vec<String> = glob.walk(path).filter_map(|e| {
Some(e.unwrap().path().file_name().unwrap().to_string_lossy().to_string())
}).collect();

assert_eq!(expected, names);
}

// TODO: Rust's testing framework does not provide a mechanism for maintaining
// shared state. This means that tests that write to the file system must
// do so individually rather than writing before and after all tests have
// run. This should probably be avoided.

/// Writes a testing directory tree to a temporary location on the file system.
fn temptree() -> (TempDir, PathBuf) {
let root = tempfile::tempdir().unwrap();
let tree: FileSystemTree<&str, &str> = dir! {
"doc" => dir! {
"guide.md" => file!(""),
},
"src" => dir! {
"glob.rs" => file!(""),
"lib.rs" => file!(""),
},
"tests" => dir! {
"walk.rs" => file!(""),
},
"README.md" => file!(""),
"fleet.txt" => file!(""),
"extra:dots.txt" => file!(""),
"extra,comma.txt" => file!(""),
};
let path = root.path().join("project");
tree.build(&path).unwrap();
(root, path)
}