// Copyright 2018 Google LLC
//
// 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.

//! Link label parsing and matching.

use unicase::UniCase;

use crate::scanners::{is_ascii_whitespace, scan_eol, is_ascii_punctuation};
use crate::strings::CowStr;

#[derive(Debug)]
pub(crate) enum ReferenceLabel<'a> {
    Link(CowStr<'a>),
    Footnote(CowStr<'a>),
}

pub(crate) type LinkLabel<'a> = UniCase<CowStr<'a>>;

pub(crate) type FootnoteLabel<'a> = UniCase<CowStr<'a>>;

/// Assumes the opening bracket has already been scanned.
/// The line break handler determines what happens when a linebreak
/// is found. It is passed the bytes following the line break and
/// either returns `Some(k)`, where `k` is the number of bytes to skip,
/// or `None` to abort parsing the label.
/// Returns the number of bytes read (including closing bracket) and label on success.
pub(crate) fn scan_link_label_rest<'t>(
    text: &'t str,
    linebreak_handler: &dyn Fn(&[u8]) -> Option<usize>,
    is_in_table: bool,
) -> Option<(usize, CowStr<'t>)> {
    let bytes = text.as_bytes();
    let mut ix = 0;
    let mut only_white_space = true;
    let mut codepoints = 0;
    // no worries, doesn't allocate until we push things onto it
    let mut label = String::new();
    let mut mark = 0;

    loop {
        if codepoints >= 1000 {
            return None;
        }
        match *bytes.get(ix)? {
            b'[' => return None,
            b']' => break,
            // Backslash escapes in link references are normally untouched, but
            // tables are an exception, because they're parsed as-if the tables
            // were parsed in a discrete pass, changing `\|` to `|`, and then
            // passing the changed string to the inline parser.
            b'|' if is_in_table && ix != 0 && bytes.get(ix - 1) == Some(&b'\\') => {
                // only way to reach this spot is to have `\\|` (even number of `\` before `|`)
                label.push_str(&text[mark..ix - 1]);
                label.push('|');
                ix += 1;
                only_white_space = false;
                mark = ix;
            }
            b'\\' if is_in_table && bytes.get(ix + 1) == Some(&b'|') => {
                // only way to reach this spot is to have `\|` (odd number of `\` before `|`)
                label.push_str(&text[mark..ix]);
                label.push('|');
                ix += 2;
                codepoints += 1;
                only_white_space = false;
                mark = ix;
            }
            b'\\' if is_ascii_punctuation(*bytes.get(ix + 1)?) => {
                ix += 2;
                codepoints += 2;
                only_white_space = false;
            }
            b if is_ascii_whitespace(b) => {
                // normalize labels by collapsing whitespaces, including linebreaks
                let mut whitespaces = 0;
                let mut linebreaks = 0;
                let whitespace_start = ix;

                while ix < bytes.len() && is_ascii_whitespace(bytes[ix]) {
                    if let Some(eol_bytes) = scan_eol(&bytes[ix..]) {
                        linebreaks += 1;
                        if linebreaks > 1 {
                            return None;
                        }
                        ix += eol_bytes;
                        ix += linebreak_handler(&bytes[ix..])?;
                        whitespaces += 2; // indicate that we need to replace
                    } else {
                        whitespaces += if bytes[ix] == b' ' { 1 } else { 2 };
                        ix += 1;
                    }
                }
                if whitespaces > 1 {
                    label.push_str(&text[mark..whitespace_start]);
                    label.push(' ');
                    mark = ix;
                    codepoints += ix - whitespace_start;
                } else {
                    codepoints += 1;
                }
            }
            b => {
                only_white_space = false;
                ix += 1;
                if b & 0b1000_0000 != 0 {
                    codepoints += 1;
                }
            }
        }
    }

    if only_white_space {
        None
    } else {
        let cow = if mark == 0 {
            let asciiws = &[' ', '\r', '\n', '\t'][..];
            text[..ix].trim_matches(asciiws).into()
        } else {
            label.push_str(&text[mark..ix]);
            while matches!(label.as_bytes().last(), Some(&b' ' | &b'\r' | &b'\n' | &b'\t')) {
                label.pop();
            }
            while matches!(label.as_bytes().first(), Some(&b' ' | &b'\r' | &b'\n' | &b'\t')) {
                label.remove(0);
            }
            label.into()
        };
        Some((ix + 1, cow))
    }
}

#[cfg(test)]
mod test {
    use super::scan_link_label_rest;

    #[test]
    fn whitespace_normalization() {
        let input = "«\t\tBlurry Eyes\t\t»][blurry_eyes]";
        let expected_output = "« Blurry Eyes »"; // regular spaces!

        let (_bytes, normalized_label) = scan_link_label_rest(input, &|_| None, false).unwrap();
        assert_eq!(expected_output, normalized_label.as_ref());
    }

    #[test]
    fn return_carriage_linefeed_ok() {
        let input = "hello\r\nworld\r\n]";
        assert!(scan_link_label_rest(input, &|_| Some(0), false).is_some());
    }
}