summaryrefslogtreecommitdiffstats
path: root/globset/src/pathutil.rs
blob: 6c2fb1e99b20addf1f0e9c437dc5b7d44597d46a (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
use std::borrow::Cow;

use bstr::BStr;

/// The final component of the path, if it is a normal file.
///
/// If the path terminates in ., .., or consists solely of a root of prefix,
/// file_name will return None.
pub fn file_name<'a>(path: &Cow<'a, BStr>) -> Option<Cow<'a, BStr>> {
    if path.is_empty() {
        return None;
    } else if path.last() == Some(b'.') {
        return None;
    }
    let last_slash = path.rfind_byte(b'/').map(|i| i + 1).unwrap_or(0);
    Some(match *path {
        Cow::Borrowed(path) => Cow::Borrowed(&path[last_slash..]),
        Cow::Owned(ref path) => {
            let mut path = path.clone();
            path.drain_bytes(..last_slash);
            Cow::Owned(path)
        }
    })
}

/// Return a file extension given a path's file name.
///
/// Note that this does NOT match the semantics of std::path::Path::extension.
/// Namely, the extension includes the `.` and matching is otherwise more
/// liberal. Specifically, the extenion is:
///
/// * None, if the file name given is empty;
/// * None, if there is no embedded `.`;
/// * Otherwise, the portion of the file name starting with the final `.`.
///
/// e.g., A file name of `.rs` has an extension `.rs`.
///
/// N.B. This is done to make certain glob match optimizations easier. Namely,
/// a pattern like `*.rs` is obviously trying to match files with a `rs`
/// extension, but it also matches files like `.rs`, which doesn't have an
/// extension according to std::path::Path::extension.
pub fn file_name_ext<'a>(name: &Cow<'a, BStr>) -> Option<Cow<'a, BStr>> {
    if name.is_empty() {
        return None;
    }
    let last_dot_at = match name.rfind_byte(b'.') {
        None => return None,
        Some(i) => i,
    };
    Some(match *name {
        Cow::Borrowed(name) => Cow::Borrowed(&name[last_dot_at..]),
        Cow::Owned(ref name) => {
            let mut name = name.clone();
            name.drain_bytes(..last_dot_at);
            Cow::Owned(name)
        }
    })
}

/// Normalizes a path to use `/` as a separator everywhere, even on platforms
/// that recognize other characters as separators.
#[cfg(unix)]
pub fn normalize_path(path: Cow<BStr>) -> Cow<BStr> {
    // UNIX only uses /, so we're good.
    path
}

/// Normalizes a path to use `/` as a separator everywhere, even on platforms
/// that recognize other characters as separators.
#[cfg(not(unix))]
pub fn normalize_path(mut path: Cow<BStr>) -> Cow<BStr> {
    use std::path::is_separator;

    for i in 0..path.len() {
        if path[i] == b'/' || !is_separator(path[i] as char) {
            continue;
        }
        path.to_mut()[i] = b'/';
    }
    path
}

#[cfg(test)]
mod tests {
    use std::borrow::Cow;

    use bstr::{B, BString};

    use super::{file_name_ext, normalize_path};

    macro_rules! ext {
        ($name:ident, $file_name:expr, $ext:expr) => {
            #[test]
            fn $name() {
                let bs = BString::from($file_name);
                let got = file_name_ext(&Cow::Owned(bs));
                assert_eq!($ext.map(|s| Cow::Borrowed(B(s))), got);
            }
        };
    }

    ext!(ext1, "foo.rs", Some(".rs"));
    ext!(ext2, ".rs", Some(".rs"));
    ext!(ext3, "..rs", Some(".rs"));
    ext!(ext4, "", None::<&str>);
    ext!(ext5, "foo", None::<&str>);

    macro_rules! normalize {
        ($name:ident, $path:expr, $expected:expr) => {
            #[test]
            fn $name() {
                let bs = BString::from_slice($path);
                let got = normalize_path(Cow::Owned(bs));
                assert_eq!($expected.to_vec(), got.into_owned());
            }
        };
    }

    normalize!(normal1, b"foo", b"foo");
    normalize!(normal2, b"foo/bar", b"foo/bar");
    #[cfg(unix)]
    normalize!(normal3, b"foo\\bar", b"foo\\bar");
    #[cfg(not(unix))]
    normalize!(normal3, b"foo\\bar", b"foo/bar");
    #[cfg(unix)]
    normalize!(normal4, b"foo\\bar/baz", b"foo\\bar/baz");
    #[cfg(not(unix))]
    normalize!(normal4, b"foo\\bar/baz", b"foo/bar/baz");
}