summaryrefslogtreecommitdiffstats
path: root/third_party/rust/url/UPGRADING.md
blob: f156130f687b9a63f89a204d98928e8a17baf1e9 (plain)
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
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
# Guide to upgrading from url 0.x to 1.x

* The fields of `Url` are now private because the `Url` constructor, parser,
  and setters maintain invariants that could be violated if you were to set the fields directly.
  Instead of accessing, for example, `url.scheme`, use the getter method, such as `url.scheme()`.
  Instead of assigning directly to a field, for example `url.scheme = "https".to_string()`,
  use the setter method, such as `url.set_scheme("https").unwrap()`.
  (Some setters validate the new value and return a `Result` that must be used).

* The methods of `Url` now return `&str` instead of `String`,
  thus reducing allocations and making serialization cheap.

* The `path()` method on `url::Url` instances used to return `Option<&[String]>`;
  now it returns `&str`.
  If you would like functionality more similar to the old behavior of `path()`,
  use `path_segments()` that returns `Option<str::Split<char>>`.

    Before upgrading:

    ```rust
    let issue_list_url = Url::parse(
         "https://github.com/rust-lang/rust/issues?labels=E-easy&state=open"
    ).unwrap();
    assert_eq!(issue_list_url.path(), Some(&["rust-lang".to_string(),
                                             "rust".to_string(),
                                             "issues".to_string()][..]));
    ```

    After upgrading:

    ```rust
    let issue_list_url = Url::parse(
         "https://github.com/rust-lang/rust/issues?labels=E-easy&state=open"
    ).unwrap();
    assert_eq!(issue_list_url.path(), "/rust-lang/rust/issues");
    assert_eq!(issue_list_url.path_segments().map(|c| c.collect::<Vec<_>>()),
               Some(vec!["rust-lang", "rust", "issues"]));
    ```

* The `path_mut()` method on `url::Url` instances that allowed modification of a URL's path
  has been replaced by `path_segments_mut()`.

    Before upgrading:

    ```rust
    let mut url = Url::parse("https://github.com/rust-lang/rust").unwrap();
    url.path_mut().unwrap().push("issues");
    ```

    After upgrading:

    ```rust
    let mut url = Url::parse("https://github.com/rust-lang/rust").unwrap();
    url.path_segments_mut().unwrap().push("issues");
    ```

* The `domain_mut()` method on `url::Url` instances that allowed modification of a URL's domain
  has been replaced by `set_host()` and `set_ip_host()`.

* The `host()` method on `url::Url` instances used to return `Option<&Host>`;
  now it returns `Option<Host<&str>>`.
  The `serialize_host()` method that returned `Option<String>`
  has been replaced by the `host_str()` method that returns `Option<&str>`.

* The `serialize()` method on `url::Url` instances that returned `String`
  has been replaced by an `as_str()` method that returns `&str`.

    Before upgrading:

    ```rust
    let this_document = Url::parse("http://servo.github.io/rust-url/url/index.html").unwrap();
    assert_eq!(this_document.serialize(), "http://servo.github.io/rust-url/url/index.html".to_string());
    ```

    After upgrading:

    ```rust
    let this_document = Url::parse("http://servo.github.io/rust-url/url/index.html").unwrap();
    assert_eq!(this_document.as_str(), "http://servo.github.io/rust-url/url/index.html");
    ```

* `url::UrlParser` has been replaced by `url::Url::parse()` and `url::Url::join()`.

    Before upgrading:

    ```rust
    let this_document = Url::parse("http://servo.github.io/rust-url/url/index.html").unwrap();
    let css_url = UrlParser::new().base_url(&this_document).parse("../main.css").unwrap();
    assert_eq!(css_url.serialize(), "http://servo.github.io/rust-url/main.css".to_string());
    ```

    After upgrading:

    ```rust
    let this_document = Url::parse("http://servo.github.io/rust-url/url/index.html").unwrap();
    let css_url = this_document.join("../main.css").unwrap();
    assert_eq!(css_url.as_str(), "http://servo.github.io/rust-url/main.css");
    ```

* `url::parse_path()` and `url::UrlParser::parse_path()` have been removed without replacement.
  As a workaround, you can give a base URL that you then ignore too `url::Url::parse()`.

    Before upgrading:

    ```rust
    let (path, query, fragment) = url::parse_path("/foo/bar/../baz?q=42").unwrap();
    assert_eq!(path, vec!["foo".to_string(), "baz".to_string()]);
    assert_eq!(query, Some("q=42".to_string()));
    assert_eq!(fragment, None);
    ```

    After upgrading:

    ```rust
    let base = Url::parse("http://example.com").unwrap();
    let with_path = base.join("/foo/bar/../baz?q=42").unwrap();
    assert_eq!(with_path.path(), "/foo/baz");
    assert_eq!(with_path.query(), Some("q=42"));
    assert_eq!(with_path.fragment(), None);
    ```

* The `url::form_urlencoded::serialize()` method
  has been replaced with the `url::form_urlencoded::Serializer` struct.
  Instead of calling `serialize()` with key/value pairs,
  create a new `Serializer` with a new string,
  call the `extend_pairs()` method on the `Serializer` instance with the key/value pairs as the argument,
  then call `finish()`.

    Before upgrading:

    ```rust
    let form = url::form_urlencoded::serialize(form.iter().map(|(k, v)| {
        (&k[..], &v[..])
    }));
    ```

    After upgrading:

    ```rust
    let form = url::form_urlencoded::Serializer::new(String::new()).extend_pairs(
        form.iter().map(|(k, v)| { (&k[..], &v[..]) })
    ).finish();
    ```

* The `set_query_from_pairs()` method on `url::Url` instances that took key/value pairs
  has been replaced with `query_pairs_mut()`, which allows you to modify the `url::Url`'s query pairs.

    Before upgrading:

    ```rust
    let mut url = Url::parse("https://duckduckgo.com/").unwrap();
    let pairs = vec![
        ("q", "test"),
        ("ia", "images"),
    ];
    url.set_query_from_pairs(pairs.iter().map(|&(k, v)| {
        (&k[..], &v[..])
    }));
    ```

    After upgrading:

    ```rust
    let mut url = Url::parse("https://duckduckgo.com/").unwrap();
    let pairs = vec![
        ("q", "test"),
        ("ia", "images"),
    ];
    url.query_pairs_mut().clear().extend_pairs(
      pairs.iter().map(|&(k, v)| { (&k[..], &v[..]) })
    );
    ```

* `url::SchemeData`, its variants `Relative` and `NonRelative`,
  and the struct `url::RelativeSchemeData` have been removed.
  Instead of matching on these variants
  to determine if you have a URL in a relative scheme such as HTTP
  versus a URL in a non-relative scheme as data,
  use the `cannot_be_a_base()` method to determine which kind you have.

    Before upgrading:

    ```rust
    match url.scheme_data {
        url::SchemeData::Relative(..) => {}
        url::SchemeData::NonRelative(..) => {
            return Err(human(format!("`{}` must have relative scheme \
                                      data: {}", field, url)))
        }
    }
    ```

    After upgrading:

    ```rust
    if url.cannot_be_a_base() {
        return Err(human(format!("`{}` must have relative scheme \
                                  data: {}", field, url)))
    }
    ```

* The functions `url::whatwg_scheme_type_mapper()`, the `SchemeType` enum,
  and the `scheme_type_mapper()` method on `url::UrlParser` instances have been removed.
  `SchemeType` had a method for getting the `default_port()`;
  to replicate this functionality, use the method `port_or_known_default()` on `url::Url` instances.
  The `port_or_default()` method on `url::Url` instances has been removed;
  use `port_or_known_default()` instead.

    Before upgrading:

    ```rust
    let port = match whatwg_scheme_type_mapper(&url.scheme) {
        SchemeType::Relative(port) => port,
        _ => return Err(format!("Invalid special scheme: `{}`",
                                raw_url.scheme)),
    };
    ```

    After upgrading:

    ```rust
    let port = match url.port_or_known_default() {
        Some(port) => port,
        _ => return Err(format!("Invalid special scheme: `{}`",
                                url.scheme())),
    };
    ```

* The following formatting utilities have been removed without replacement;
  look at their linked previous implementations
  if you would like to replicate the functionality in your code:
  * [`url::format::PathFormatter`](https://github.com/servo/rust-url/pull/176/commits/9e759f18726c8e1343162922b87163d4dd08fe3c#diff-0bb16ac13b75e9b568fa4aff61b0e71dL24)
  * [`url::format::UserInfoFormatter`](https://github.com/servo/rust-url/pull/176/commits/9e759f18726c8e1343162922b87163d4dd08fe3c#diff-0bb16ac13b75e9b568fa4aff61b0e71dL50)
  * [`url::format::UrlNoFragmentFormatter`](https://github.com/servo/rust-url/pull/176/commits/9e759f18726c8e1343162922b87163d4dd08fe3c#diff-0bb16ac13b75e9b568fa4aff61b0e71dL70)

* `url::percent_encoding::percent_decode()` used to have a return type of `Vec<u8>`;
  now it returns an iterator of decoded `u8` bytes that also implements `Into<Cow<u8>>`.
  Use `.into().to_owned()` to obtain a `Vec<u8>`.
  (`.collect()` also works but might not be as efficient.)

* The `url::percent_encoding::EncodeSet` struct and constant instances
  used with `url::percent_encoding::percent_encode()`
  have been changed to structs that implement the trait `url::percent_encoding::EncodeSet`.
  * `SIMPLE_ENCODE_SET`, `QUERY_ENCODE_SET`, `DEFAULT_ENCODE_SET`,
    and `USERINFO_ENCODE_SET` have the same behavior.
  * `USERNAME_ENCODE_SET` and `PASSWORD_ENCODE_SET` have been removed;
    use `USERINFO_ENCODE_SET` instead.
  * `HTTP_VALUE_ENCODE_SET` has been removed;
    an implementation of it in the new types can be found [in hyper's source](
    https://github.com/hyperium/hyper/blob/67436c5bf615cf5a55a71e32b788afef5985570e/src/header/parsing.rs#L131-L138)
    if you need to replicate this functionality in your code.
  * `FORM_URLENCODED_ENCODE_SET` has been removed;
    instead, use the functionality in `url::form_urlencoded`.
  * `PATH_SEGMENT_ENCODE_SET` has been added for use on '/'-separated path segments.

* `url::percent_encoding::percent_decode_to()` has been removed.
  Use `url::percent_encoding::percent_decode()` which returns an iterator.
  You can then use the iterator’s `collect()` method
  or give it to some data structure’s `extend()` method.
* A number of `ParseError` variants have changed.
  [See the documentation for the current set](http://servo.github.io/rust-url/url/enum.ParseError.html).
* `url::OpaqueOrigin::new()` and `url::Origin::UID(OpaqueOrigin)`
  have been replaced by `url::Origin::new_opaque()` and `url::Origin::Opaque(OpaqueOrigin)`, respectively.