// filepatterns.rs

//

// Copyright 2019 Raphaël Gomès <rgomes@octobus.net>

//

// This software may be used and distributed according to the terms of the

// GNU General Public License version 2 or any later version.

//! Handling of Mercurial-specific patterns.

use crate::{

    utils::{

        files::{canonical_path, get_bytes_from_path, get_path_from_bytes},

        hg_path::{path_to_hg_path_buf, HgPathBuf, HgPathError},

        SliceExt,

    },

    FastHashMap, PatternError,

};

use lazy_static::lazy_static;

use regex::bytes::{NoExpand, Regex};

use std::fs::File;

use std::io::Read;

use std::ops::Deref;

use std::path::{Path, PathBuf};

use std::vec::Vec;

lazy_static! {

    static ref RE_ESCAPE: Vec<Vec<u8>> = {

        let mut v: Vec<Vec<u8>> = (0..=255).map(|byte| vec![byte]).collect();

        let to_escape = b"()[]{}?*+-|^$\\.&~# \t\n\r\x0b\x0c";

        for byte in to_escape {

            v[*byte as usize].insert(0, b'\\');

        }

        v

    };

}

/// These are matched in order

const GLOB_REPLACEMENTS: &[(&[u8], &[u8])] =

    &[(b"*/", b"(?:.*/)?"), (b"*", b".*"), (b"", b"[^/]*")];

/// Appended to the regexp of globs

const GLOB_SUFFIX: &[u8; 7] = b"(?:/|$)";

#[derive(Debug, Copy, Clone, PartialEq, Eq)]

pub enum PatternSyntax {

    /// A regular expression

    Regexp,

    /// Glob that matches at the front of the path

    RootGlob,

    /// Glob that matches at any suffix of the path (still anchored at

    /// slashes)

    Glob,

    /// a path relative to repository root, which is matched recursively

    Path,

    /// A path relative to cwd

    RelPath,

    /// an unrooted glob (*.rs matches Rust files in all dirs)

    RelGlob,

    /// A regexp that needn't match the start of a name

    RelRegexp,

    /// A path relative to repository root, which is matched non-recursively

    /// (will not match subdirectories)

    RootFiles,

    /// A file of patterns to read and include

    Include,

    /// A file of patterns to match against files under the same directory

    SubInclude,

}

/// Transforms a glob pattern into a regex

fn glob_to_re(pat: &[u8]) -> Vec<u8> {

    let mut input = pat;

    let mut res: Vec<u8> = vec![];

    let mut group_depth = 0;

    while let Some((c, rest)) = input.split_first() {

        input = rest;

        match c {

            b'*' => {

                for (source, repl) in GLOB_REPLACEMENTS {

                    if let Some(rest) = input.drop_prefix(source) {

                        input = rest;

                        res.extend(*repl);

                        break;

                    }

                }

            }

            b'?' => res.extend(b"."),

            b'[' => {

                match input.iter().skip(1).position(|b| *b == b']') {

                    None => res.extend(b"\\["),

                    Some(end) => {

                        // Account for the one we skipped

                        let end = end + 1;

                        res.extend(b"[");

                        for (i, b) in input[..end].iter().enumerate() {

                            if *b == b'!' && i == 0 {

                                res.extend(b"^")

                            } else if *b == b'^' && i == 0 {

                                res.extend(b"\\^")

                            } else if *b == b'\\' {

                                res.extend(b"\\\\")

                            } else {

                                res.push(*b)

                            }

                        }

                        res.extend(b"]");

                        input = &input[end + 1..];

                    }

                }

            }

            b'{' => {

                group_depth += 1;

                res.extend(b"(?:")

            }

            b'}' if group_depth > 0 => {

                group_depth -= 1;

                res.extend(b")");

            }

            b',' if group_depth > 0 => res.extend(b"|"),

            b'\\' => {

                let c = {

                    if let Some((c, rest)) = input.split_first() {

                        input = rest;

                        c

                    } else {

                        c

                    }

                };

                res.extend(&RE_ESCAPE[*c as usize])

            }

            _ => res.extend(&RE_ESCAPE[*c as usize]),

        }

    }

    res

}

fn escape_pattern(pattern: &[u8]) -> Vec<u8> {

    pattern

        .iter()

        .flat_map(|c| RE_ESCAPE[*c as usize].clone())

        .collect()

}

pub fn parse_pattern_syntax(

    kind: &[u8],

) -> Result<PatternSyntax, PatternError> {

    match kind {

        b"re:" => Ok(PatternSyntax::Regexp),

        b"path:" => Ok(PatternSyntax::Path),

        b"relpath:" => Ok(PatternSyntax::RelPath),

        b"rootfilesin:" => Ok(PatternSyntax::RootFiles),

        b"relglob:" => Ok(PatternSyntax::RelGlob),

        b"relre:" => Ok(PatternSyntax::RelRegexp),

        b"glob:" => Ok(PatternSyntax::Glob),

        b"rootglob:" => Ok(PatternSyntax::RootGlob),

        b"include:" => Ok(PatternSyntax::Include),

        b"subinclude:" => Ok(PatternSyntax::SubInclude),

        _ => Err(PatternError::UnsupportedSyntax(

            String::from_utf8_lossy(kind).to_string(),

        )),

    }

}

/// Builds the regex that corresponds to the given pattern.

/// If within a `syntax: regexp` context, returns the pattern,

/// otherwise, returns the corresponding regex.

fn _build_single_regex(entry: &IgnorePattern) -> Vec<u8> {

    let IgnorePattern {

        syntax, pattern, ..

    } = entry;

    if pattern.is_empty() {

        return vec![];

    }

    match syntax {

        PatternSyntax::Regexp => pattern.to_owned(),

        PatternSyntax::RelRegexp => {

            // The `regex` crate accepts `**` while `re2` and Python's `re`

            // do not. Checking for `*` correctly triggers the same error all

            // engines.

            if pattern[0] == b'^'

                || pattern[0] == b'*'

                || pattern.starts_with(b".*")

            {

                return pattern.to_owned();

            }

            [&b".*"[..], pattern].concat()

        }

        PatternSyntax::Path | PatternSyntax::RelPath => {

            if pattern == b"." {

                return vec![];

            }

            [escape_pattern(pattern).as_slice(), b"(?:/|$)"].concat()

        }

        PatternSyntax::RootFiles => {

            let mut res = if pattern == b"." {

                vec![]

            } else {

                // Pattern is a directory name.

                [escape_pattern(pattern).as_slice(), b"/"].concat()

            };

            // Anything after the pattern must be a non-directory.

            res.extend(b"[^/]+$");

            res

        }

        PatternSyntax::RelGlob => {

            let glob_re = glob_to_re(pattern);

            if let Some(rest) = glob_re.drop_prefix(b"[^/]*") {

                [b".*", rest, GLOB_SUFFIX].concat()

            } else {

                [b"(?:.*/)?", glob_re.as_slice(), GLOB_SUFFIX].concat()

            }

        }

        PatternSyntax::Glob | PatternSyntax::RootGlob => {

            [glob_to_re(pattern).as_slice(), GLOB_SUFFIX].concat()

        }

        PatternSyntax::Include | PatternSyntax::SubInclude => unreachable!(),

    }

}

const GLOB_SPECIAL_CHARACTERS: [u8; 7] =

    [b'*', b'?', b'[', b']', b'{', b'}', b'\\'];

/// TODO support other platforms

#[cfg(unix)]

pub fn normalize_path_bytes(bytes: &[u8]) -> Vec<u8> {

    if bytes.is_empty() {

        return b".".to_vec();

    }

    let sep = b'/';

    let mut initial_slashes = bytes.iter().take_while(|b| **b == sep).count();

    if initial_slashes > 2 {

        // POSIX allows one or two initial slashes, but treats three or more

        // as single slash.

        initial_slashes = 1;

    }

    let components = bytes

        .split(|b| *b == sep)

        .filter(|c| !(c.is_empty() || c == b"."))

        .fold(vec![], |mut acc, component| {

            if component != b".."

                || (initial_slashes == 0 && acc.is_empty())

                || (!acc.is_empty() && acc[acc.len() - 1] == b"..")

            {

                acc.push(component)

            } else if !acc.is_empty() {

                acc.pop();

            }

            acc

        });

    let mut new_bytes = components.join(&sep);

    if initial_slashes > 0 {

        let mut buf: Vec<_> = (0..initial_slashes).map(|_| sep).collect();

        buf.extend(new_bytes);

        new_bytes = buf;

    }

    if new_bytes.is_empty() {

        b".".to_vec()

    } else {

        new_bytes

    }

}

/// Wrapper function to `_build_single_regex` that short-circuits 'exact' globs

/// that don't need to be transformed into a regex.

pub fn build_single_regex(

    entry: &IgnorePattern,

) -> Result<Option<Vec<u8>>, PatternError> {

    let IgnorePattern {

        pattern, syntax, ..

    } = entry;

    let pattern = match syntax {

        PatternSyntax::RootGlob

        | PatternSyntax::Path

        | PatternSyntax::RelGlob

        | PatternSyntax::RootFiles => normalize_path_bytes(&pattern),

        PatternSyntax::Include | PatternSyntax::SubInclude => {

            return Err(PatternError::NonRegexPattern(entry.clone()))

        }

        _ => pattern.to_owned(),

    };

    if *syntax == PatternSyntax::RootGlob

        && !pattern.iter().any(|b| GLOB_SPECIAL_CHARACTERS.contains(b))

    {

        Ok(None)

    } else {

        let mut entry = entry.clone();

        entry.pattern = pattern;

        Ok(Some(_build_single_regex(&entry)))

    }

}

lazy_static! {

    static ref SYNTAXES: FastHashMap<&'static [u8], &'static [u8]> = {

        let mut m = FastHashMap::default();

        m.insert(b"re".as_ref(), b"relre:".as_ref());

        m.insert(b"regexp".as_ref(), b"relre:".as_ref());

        m.insert(b"glob".as_ref(), b"relglob:".as_ref());

        m.insert(b"rootglob".as_ref(), b"rootglob:".as_ref());

        m.insert(b"include".as_ref(), b"include:".as_ref());

        m.insert(b"subinclude".as_ref(), b"subinclude:".as_ref());

        m

    };

}

#[derive(Debug)]

pub enum PatternFileWarning {

    /// (file path, syntax bytes)

    InvalidSyntax(PathBuf, Vec<u8>),

    /// File path

    NoSuchFile(PathBuf),

}

pub fn parse_pattern_file_contents<P: AsRef<Path>>(

    lines: &[u8],

    file_path: P,

    warn: bool,

) -> Result<(Vec<IgnorePattern>, Vec<PatternFileWarning>), PatternError> {

    let comment_regex = Regex::new(r"((?:^|[^\\])(?:\\\\)*)#.*").unwrap();

    let comment_escape_regex = Regex::new(r"\\#").unwrap();

    let mut inputs: Vec<IgnorePattern> = vec![];

    let mut warnings: Vec<PatternFileWarning> = vec![];

    let mut current_syntax = b"relre:".as_ref();

    for (line_number, mut line) in lines.split(|c| *c == b'\n').enumerate() {

        let line_number = line_number + 1;

        let line_buf;