1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
// Copyright 2016 The rust-url developers.
//
// 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.

use parser::{self, SchemeType, to_u32};
use std::str;
use Url;

/// Exposes methods to manipulate the path of an URL that is not cannot-be-base.
///
/// The path always starts with a `/` slash, and is made of slash-separated segments.
/// There is always at least one segment (which may be the empty string).
///
/// Examples:
///
/// ```rust
/// # use url::Url;
/// let mut url = Url::parse("mailto:me@example.com").unwrap();
/// assert!(url.path_segments_mut().is_err());
///
/// let mut url = Url::parse("http://example.net/foo/index.html").unwrap();
/// url.path_segments_mut().unwrap().pop().push("img").push("2/100%.png");
/// assert_eq!(url.as_str(), "http://example.net/foo/img/2%2F100%25.png");
/// ```
pub struct PathSegmentsMut<'a> {
    url: &'a mut Url,
    after_first_slash: usize,
    after_path: String,
    old_after_path_position: u32,
}

// Not re-exported outside the crate
pub fn new(url: &mut Url) -> PathSegmentsMut {
    let after_path = url.take_after_path();
    let old_after_path_position = to_u32(url.serialization.len()).unwrap();
    debug_assert!(url.byte_at(url.path_start) == b'/');
    PathSegmentsMut {
        after_first_slash: url.path_start as usize + "/".len(),
        url: url,
        old_after_path_position: old_after_path_position,
        after_path: after_path,
    }
}

impl<'a> Drop for PathSegmentsMut<'a> {
    fn drop(&mut self) {
        self.url.restore_after_path(self.old_after_path_position, &self.after_path)
    }
}

impl<'a> PathSegmentsMut<'a> {
    /// Remove all segments in the path, leaving the minimal `url.path() == "/"`.
    ///
    /// Returns `&mut Self` so that method calls can be chained.
    ///
    /// Example:
    ///
    /// ```rust
    /// # use url::Url;
    /// let mut url = Url::parse("https://github.com/servo/rust-url/").unwrap();
    /// url.path_segments_mut().unwrap().clear().push("logout");
    /// assert_eq!(url.as_str(), "https://github.com/logout");
    /// ```
    pub fn clear(&mut self) -> &mut Self {
        self.url.serialization.truncate(self.after_first_slash);
        self
    }

    /// Remove the last segment of this URL’s path if it is empty,
    /// except if these was only one segment to begin with.
    ///
    /// In other words, remove one path trailing slash, if any,
    /// unless it is also the initial slash (so this does nothing if `url.path() == "/")`.
    ///
    /// Returns `&mut Self` so that method calls can be chained.
    ///
    /// Example:
    ///
    /// ```rust
    /// # use url::Url;
    /// let mut url = Url::parse("https://github.com/servo/rust-url/").unwrap();
    /// url.path_segments_mut().unwrap().push("pulls");
    /// assert_eq!(url.as_str(), "https://github.com/servo/rust-url//pulls");
    ///
    /// let mut url = Url::parse("https://github.com/servo/rust-url/").unwrap();
    /// url.path_segments_mut().unwrap().pop_if_empty().push("pulls");
    /// assert_eq!(url.as_str(), "https://github.com/servo/rust-url/pulls");
    /// ```
    pub fn pop_if_empty(&mut self) -> &mut Self {
        if self.url.serialization[self.after_first_slash..].ends_with('/') {
            self.url.serialization.pop();
        }
        self
    }

    /// Remove the last segment of this URL’s path.
    ///
    /// If the path only has one segment, make it empty such that `url.path() == "/"`.
    ///
    /// Returns `&mut Self` so that method calls can be chained.
    pub fn pop(&mut self) -> &mut Self {
        let last_slash = self.url.serialization[self.after_first_slash..].rfind('/').unwrap_or(0);
        self.url.serialization.truncate(self.after_first_slash + last_slash);
        self
    }

    /// Append the given segment at the end of this URL’s path.
    ///
    /// See the documentation for `.extend()`.
    ///
    /// Returns `&mut Self` so that method calls can be chained.
    pub fn push(&mut self, segment: &str) -> &mut Self {
        self.extend(Some(segment))
    }

    /// Append each segment from the given iterator at the end of this URL’s path.
    ///
    /// Each segment is percent-encoded like in `Url::parse` or `Url::join`,
    /// except that `%` and `/` characters are also encoded (to `%25` and `%2F`).
    /// This is unlike `Url::parse` where `%` is left as-is in case some of the input
    /// is already percent-encoded, and `/` denotes a path segment separator.)
    ///
    /// Note that, in addition to slashes between new segments,
    /// this always adds a slash between the existing path and the new segments
    /// *except* if the existing path is `"/"`.
    /// If the previous last segment was empty (if the path had a trailing slash)
    /// the path after `.extend()` will contain two consecutive slashes.
    /// If that is undesired, call `.pop_if_empty()` first.
    ///
    /// To obtain a behavior similar to `Url::join`, call `.pop()` unconditionally first.
    ///
    /// Returns `&mut Self` so that method calls can be chained.
    ///
    /// Example:
    ///
    /// ```rust
    /// # use url::Url;
    /// let mut url = Url::parse("https://github.com/").unwrap();
    /// let org = "servo";
    /// let repo = "rust-url";
    /// let issue_number = "188";
    /// url.path_segments_mut().unwrap().extend(&[org, repo, "issues", issue_number]);
    /// assert_eq!(url.as_str(), "https://github.com/servo/rust-url/issues/188");
    /// ```
    ///
    /// In order to make sure that parsing the serialization of an URL gives the same URL,
    /// a segment is ignored if it is `"."` or `".."`:
    ///
    /// ```rust
    /// # use url::Url;
    /// let mut url = Url::parse("https://github.com/servo").unwrap();
    /// url.path_segments_mut().unwrap().extend(&["..", "rust-url", ".", "pulls"]);
    /// assert_eq!(url.as_str(), "https://github.com/servo/rust-url/pulls");
    /// ```
    pub fn extend<I>(&mut self, segments: I) -> &mut Self
    where I: IntoIterator, I::Item: AsRef<str> {
        let scheme_type = SchemeType::from(self.url.scheme());
        let path_start = self.url.path_start as usize;
        self.url.mutate(|parser| {
            parser.context = parser::Context::PathSegmentSetter;
            for segment in segments {
                let segment = segment.as_ref();
                if matches!(segment, "." | "..") {
                    continue
                }
                if parser.serialization.len() > path_start + 1 {
                    parser.serialization.push('/');
                }
                let mut has_host = true;  // FIXME account for this?
                parser.parse_path(scheme_type, &mut has_host, path_start,
                                  parser::Input::new(segment));
            }
        });
        self
    }

    /// For internal testing, not part of the public API.
    #[doc(hidden)]
    pub fn assert_url_invariants(&mut self) -> &mut Self {
        self.url.assert_invariants();
        self
    }
}