Show More
@@ -1,874 +1,874 b'' | |||||
1 | // filepatterns.rs |
|
1 | // filepatterns.rs | |
2 | // |
|
2 | // | |
3 | // Copyright 2019 Raphaël Gomès <rgomes@octobus.net> |
|
3 | // Copyright 2019 Raphaël Gomès <rgomes@octobus.net> | |
4 | // |
|
4 | // | |
5 | // This software may be used and distributed according to the terms of the |
|
5 | // This software may be used and distributed according to the terms of the | |
6 | // GNU General Public License version 2 or any later version. |
|
6 | // GNU General Public License version 2 or any later version. | |
7 |
|
7 | |||
8 | //! Handling of Mercurial-specific patterns. |
|
8 | //! Handling of Mercurial-specific patterns. | |
9 |
|
9 | |||
10 | use crate::{ |
|
10 | use crate::{ | |
11 | utils::{ |
|
11 | utils::{ | |
12 | files::{canonical_path, get_bytes_from_path, get_path_from_bytes}, |
|
12 | files::{canonical_path, get_bytes_from_path, get_path_from_bytes}, | |
13 | hg_path::{path_to_hg_path_buf, HgPathBuf, HgPathError}, |
|
13 | hg_path::{path_to_hg_path_buf, HgPathBuf, HgPathError}, | |
14 | SliceExt, |
|
14 | SliceExt, | |
15 | }, |
|
15 | }, | |
16 | FastHashMap, PatternError, |
|
16 | FastHashMap, PatternError, | |
17 | }; |
|
17 | }; | |
18 | use lazy_static::lazy_static; |
|
18 | use lazy_static::lazy_static; | |
19 | use regex::bytes::{NoExpand, Regex}; |
|
19 | use regex::bytes::{NoExpand, Regex}; | |
20 | use std::ops::Deref; |
|
20 | use std::ops::Deref; | |
21 | use std::path::{Path, PathBuf}; |
|
21 | use std::path::{Path, PathBuf}; | |
22 | use std::vec::Vec; |
|
22 | use std::vec::Vec; | |
23 |
|
23 | |||
24 | lazy_static! { |
|
24 | lazy_static! { | |
25 | static ref RE_ESCAPE: Vec<Vec<u8>> = { |
|
25 | static ref RE_ESCAPE: Vec<Vec<u8>> = { | |
26 | let mut v: Vec<Vec<u8>> = (0..=255).map(|byte| vec![byte]).collect(); |
|
26 | let mut v: Vec<Vec<u8>> = (0..=255).map(|byte| vec![byte]).collect(); | |
27 | let to_escape = b"()[]{}?*+-|^$\\.&~#\t\n\r\x0b\x0c"; |
|
27 | let to_escape = b"()[]{}?*+-|^$\\.&~#\t\n\r\x0b\x0c"; | |
28 | for byte in to_escape { |
|
28 | for byte in to_escape { | |
29 | v[*byte as usize].insert(0, b'\\'); |
|
29 | v[*byte as usize].insert(0, b'\\'); | |
30 | } |
|
30 | } | |
31 | v |
|
31 | v | |
32 | }; |
|
32 | }; | |
33 | } |
|
33 | } | |
34 |
|
34 | |||
35 | /// These are matched in order |
|
35 | /// These are matched in order | |
36 | const GLOB_REPLACEMENTS: &[(&[u8], &[u8])] = |
|
36 | const GLOB_REPLACEMENTS: &[(&[u8], &[u8])] = | |
37 | &[(b"*/", b"(?:.*/)?"), (b"*", b".*"), (b"", b"[^/]*")]; |
|
37 | &[(b"*/", b"(?:.*/)?"), (b"*", b".*"), (b"", b"[^/]*")]; | |
38 |
|
38 | |||
39 | #[derive(Debug, Clone, PartialEq, Eq)] |
|
39 | #[derive(Debug, Clone, PartialEq, Eq)] | |
40 | pub enum PatternSyntax { |
|
40 | pub enum PatternSyntax { | |
41 | /// A regular expression |
|
41 | /// A regular expression | |
42 | Regexp, |
|
42 | Regexp, | |
43 | /// Glob that matches at the front of the path |
|
43 | /// Glob that matches at the front of the path | |
44 | RootGlob, |
|
44 | RootGlob, | |
45 | /// Glob that matches at any suffix of the path (still anchored at |
|
45 | /// Glob that matches at any suffix of the path (still anchored at | |
46 | /// slashes) |
|
46 | /// slashes) | |
47 | Glob, |
|
47 | Glob, | |
48 | /// a path relative to repository root, which is matched recursively |
|
48 | /// a path relative to repository root, which is matched recursively | |
49 | Path, |
|
49 | Path, | |
50 | /// a single exact path relative to repository root |
|
50 | /// a single exact path relative to repository root | |
51 | FilePath, |
|
51 | FilePath, | |
52 | /// A path relative to cwd |
|
52 | /// A path relative to cwd | |
53 | RelPath, |
|
53 | RelPath, | |
54 | /// an unrooted glob (*.rs matches Rust files in all dirs) |
|
54 | /// an unrooted glob (*.rs matches Rust files in all dirs) | |
55 | RelGlob, |
|
55 | RelGlob, | |
56 | /// A regexp that needn't match the start of a name |
|
56 | /// A regexp that needn't match the start of a name | |
57 | RelRegexp, |
|
57 | RelRegexp, | |
58 | /// A path relative to repository root, which is matched non-recursively |
|
58 | /// A path relative to repository root, which is matched non-recursively | |
59 | /// (will not match subdirectories) |
|
59 | /// (will not match subdirectories) | |
60 | RootFiles, |
|
60 | RootFiles, | |
61 | /// A file of patterns to read and include |
|
61 | /// A file of patterns to read and include | |
62 | Include, |
|
62 | Include, | |
63 | /// A file of patterns to match against files under the same directory |
|
63 | /// A file of patterns to match against files under the same directory | |
64 | SubInclude, |
|
64 | SubInclude, | |
65 | /// SubInclude with the result of parsing the included file |
|
65 | /// SubInclude with the result of parsing the included file | |
66 | /// |
|
66 | /// | |
67 | /// Note: there is no ExpandedInclude because that expansion can be done |
|
67 | /// Note: there is no ExpandedInclude because that expansion can be done | |
68 | /// in place by replacing the Include pattern by the included patterns. |
|
68 | /// in place by replacing the Include pattern by the included patterns. | |
69 | /// SubInclude requires more handling. |
|
69 | /// SubInclude requires more handling. | |
70 | /// |
|
70 | /// | |
71 | /// Note: `Box` is used to minimize size impact on other enum variants |
|
71 | /// Note: `Box` is used to minimize size impact on other enum variants | |
72 | ExpandedSubInclude(Box<SubInclude>), |
|
72 | ExpandedSubInclude(Box<SubInclude>), | |
73 | } |
|
73 | } | |
74 |
|
74 | |||
75 | /// Transforms a glob pattern into a regex |
|
75 | /// Transforms a glob pattern into a regex | |
76 | fn glob_to_re(pat: &[u8]) -> Vec<u8> { |
|
76 | pub fn glob_to_re(pat: &[u8]) -> Vec<u8> { | |
77 | let mut input = pat; |
|
77 | let mut input = pat; | |
78 | let mut res: Vec<u8> = vec![]; |
|
78 | let mut res: Vec<u8> = vec![]; | |
79 | let mut group_depth = 0; |
|
79 | let mut group_depth = 0; | |
80 |
|
80 | |||
81 | while let Some((c, rest)) = input.split_first() { |
|
81 | while let Some((c, rest)) = input.split_first() { | |
82 | input = rest; |
|
82 | input = rest; | |
83 |
|
83 | |||
84 | match c { |
|
84 | match c { | |
85 | b'*' => { |
|
85 | b'*' => { | |
86 | for (source, repl) in GLOB_REPLACEMENTS { |
|
86 | for (source, repl) in GLOB_REPLACEMENTS { | |
87 | if let Some(rest) = input.drop_prefix(source) { |
|
87 | if let Some(rest) = input.drop_prefix(source) { | |
88 | input = rest; |
|
88 | input = rest; | |
89 | res.extend(*repl); |
|
89 | res.extend(*repl); | |
90 | break; |
|
90 | break; | |
91 | } |
|
91 | } | |
92 | } |
|
92 | } | |
93 | } |
|
93 | } | |
94 | b'?' => res.extend(b"."), |
|
94 | b'?' => res.extend(b"."), | |
95 | b'[' => { |
|
95 | b'[' => { | |
96 | match input.iter().skip(1).position(|b| *b == b']') { |
|
96 | match input.iter().skip(1).position(|b| *b == b']') { | |
97 | None => res.extend(b"\\["), |
|
97 | None => res.extend(b"\\["), | |
98 | Some(end) => { |
|
98 | Some(end) => { | |
99 | // Account for the one we skipped |
|
99 | // Account for the one we skipped | |
100 | let end = end + 1; |
|
100 | let end = end + 1; | |
101 |
|
101 | |||
102 | res.extend(b"["); |
|
102 | res.extend(b"["); | |
103 |
|
103 | |||
104 | for (i, b) in input[..end].iter().enumerate() { |
|
104 | for (i, b) in input[..end].iter().enumerate() { | |
105 | if *b == b'!' && i == 0 { |
|
105 | if *b == b'!' && i == 0 { | |
106 | res.extend(b"^") |
|
106 | res.extend(b"^") | |
107 | } else if *b == b'^' && i == 0 { |
|
107 | } else if *b == b'^' && i == 0 { | |
108 | res.extend(b"\\^") |
|
108 | res.extend(b"\\^") | |
109 | } else if *b == b'\\' { |
|
109 | } else if *b == b'\\' { | |
110 | res.extend(b"\\\\") |
|
110 | res.extend(b"\\\\") | |
111 | } else { |
|
111 | } else { | |
112 | res.push(*b) |
|
112 | res.push(*b) | |
113 | } |
|
113 | } | |
114 | } |
|
114 | } | |
115 | res.extend(b"]"); |
|
115 | res.extend(b"]"); | |
116 | input = &input[end + 1..]; |
|
116 | input = &input[end + 1..]; | |
117 | } |
|
117 | } | |
118 | } |
|
118 | } | |
119 | } |
|
119 | } | |
120 | b'{' => { |
|
120 | b'{' => { | |
121 | group_depth += 1; |
|
121 | group_depth += 1; | |
122 | res.extend(b"(?:") |
|
122 | res.extend(b"(?:") | |
123 | } |
|
123 | } | |
124 | b'}' if group_depth > 0 => { |
|
124 | b'}' if group_depth > 0 => { | |
125 | group_depth -= 1; |
|
125 | group_depth -= 1; | |
126 | res.extend(b")"); |
|
126 | res.extend(b")"); | |
127 | } |
|
127 | } | |
128 | b',' if group_depth > 0 => res.extend(b"|"), |
|
128 | b',' if group_depth > 0 => res.extend(b"|"), | |
129 | b'\\' => { |
|
129 | b'\\' => { | |
130 | let c = { |
|
130 | let c = { | |
131 | if let Some((c, rest)) = input.split_first() { |
|
131 | if let Some((c, rest)) = input.split_first() { | |
132 | input = rest; |
|
132 | input = rest; | |
133 | c |
|
133 | c | |
134 | } else { |
|
134 | } else { | |
135 | c |
|
135 | c | |
136 | } |
|
136 | } | |
137 | }; |
|
137 | }; | |
138 | res.extend(&RE_ESCAPE[*c as usize]) |
|
138 | res.extend(&RE_ESCAPE[*c as usize]) | |
139 | } |
|
139 | } | |
140 | _ => res.extend(&RE_ESCAPE[*c as usize]), |
|
140 | _ => res.extend(&RE_ESCAPE[*c as usize]), | |
141 | } |
|
141 | } | |
142 | } |
|
142 | } | |
143 | res |
|
143 | res | |
144 | } |
|
144 | } | |
145 |
|
145 | |||
146 | fn escape_pattern(pattern: &[u8]) -> Vec<u8> { |
|
146 | fn escape_pattern(pattern: &[u8]) -> Vec<u8> { | |
147 | pattern |
|
147 | pattern | |
148 | .iter() |
|
148 | .iter() | |
149 | .flat_map(|c| RE_ESCAPE[*c as usize].clone()) |
|
149 | .flat_map(|c| RE_ESCAPE[*c as usize].clone()) | |
150 | .collect() |
|
150 | .collect() | |
151 | } |
|
151 | } | |
152 |
|
152 | |||
153 | pub fn parse_pattern_syntax( |
|
153 | pub fn parse_pattern_syntax( | |
154 | kind: &[u8], |
|
154 | kind: &[u8], | |
155 | ) -> Result<PatternSyntax, PatternError> { |
|
155 | ) -> Result<PatternSyntax, PatternError> { | |
156 | match kind { |
|
156 | match kind { | |
157 | b"re:" => Ok(PatternSyntax::Regexp), |
|
157 | b"re:" => Ok(PatternSyntax::Regexp), | |
158 | b"path:" => Ok(PatternSyntax::Path), |
|
158 | b"path:" => Ok(PatternSyntax::Path), | |
159 | b"filepath:" => Ok(PatternSyntax::FilePath), |
|
159 | b"filepath:" => Ok(PatternSyntax::FilePath), | |
160 | b"relpath:" => Ok(PatternSyntax::RelPath), |
|
160 | b"relpath:" => Ok(PatternSyntax::RelPath), | |
161 | b"rootfilesin:" => Ok(PatternSyntax::RootFiles), |
|
161 | b"rootfilesin:" => Ok(PatternSyntax::RootFiles), | |
162 | b"relglob:" => Ok(PatternSyntax::RelGlob), |
|
162 | b"relglob:" => Ok(PatternSyntax::RelGlob), | |
163 | b"relre:" => Ok(PatternSyntax::RelRegexp), |
|
163 | b"relre:" => Ok(PatternSyntax::RelRegexp), | |
164 | b"glob:" => Ok(PatternSyntax::Glob), |
|
164 | b"glob:" => Ok(PatternSyntax::Glob), | |
165 | b"rootglob:" => Ok(PatternSyntax::RootGlob), |
|
165 | b"rootglob:" => Ok(PatternSyntax::RootGlob), | |
166 | b"include:" => Ok(PatternSyntax::Include), |
|
166 | b"include:" => Ok(PatternSyntax::Include), | |
167 | b"subinclude:" => Ok(PatternSyntax::SubInclude), |
|
167 | b"subinclude:" => Ok(PatternSyntax::SubInclude), | |
168 | _ => Err(PatternError::UnsupportedSyntax( |
|
168 | _ => Err(PatternError::UnsupportedSyntax( | |
169 | String::from_utf8_lossy(kind).to_string(), |
|
169 | String::from_utf8_lossy(kind).to_string(), | |
170 | )), |
|
170 | )), | |
171 | } |
|
171 | } | |
172 | } |
|
172 | } | |
173 |
|
173 | |||
174 | lazy_static! { |
|
174 | lazy_static! { | |
175 | static ref FLAG_RE: Regex = Regex::new(r"^\(\?[aiLmsux]+\)").unwrap(); |
|
175 | static ref FLAG_RE: Regex = Regex::new(r"^\(\?[aiLmsux]+\)").unwrap(); | |
176 | } |
|
176 | } | |
177 |
|
177 | |||
178 | /// Builds the regex that corresponds to the given pattern. |
|
178 | /// Builds the regex that corresponds to the given pattern. | |
179 | /// If within a `syntax: regexp` context, returns the pattern, |
|
179 | /// If within a `syntax: regexp` context, returns the pattern, | |
180 | /// otherwise, returns the corresponding regex. |
|
180 | /// otherwise, returns the corresponding regex. | |
181 | fn _build_single_regex(entry: &IgnorePattern, glob_suffix: &[u8]) -> Vec<u8> { |
|
181 | fn _build_single_regex(entry: &IgnorePattern, glob_suffix: &[u8]) -> Vec<u8> { | |
182 | let IgnorePattern { |
|
182 | let IgnorePattern { | |
183 | syntax, pattern, .. |
|
183 | syntax, pattern, .. | |
184 | } = entry; |
|
184 | } = entry; | |
185 | if pattern.is_empty() { |
|
185 | if pattern.is_empty() { | |
186 | return vec![]; |
|
186 | return vec![]; | |
187 | } |
|
187 | } | |
188 | match syntax { |
|
188 | match syntax { | |
189 | PatternSyntax::Regexp => pattern.to_owned(), |
|
189 | PatternSyntax::Regexp => pattern.to_owned(), | |
190 | PatternSyntax::RelRegexp => { |
|
190 | PatternSyntax::RelRegexp => { | |
191 | // The `regex` crate accepts `**` while `re2` and Python's `re` |
|
191 | // The `regex` crate accepts `**` while `re2` and Python's `re` | |
192 | // do not. Checking for `*` correctly triggers the same error all |
|
192 | // do not. Checking for `*` correctly triggers the same error all | |
193 | // engines. |
|
193 | // engines. | |
194 | if pattern[0] == b'^' |
|
194 | if pattern[0] == b'^' | |
195 | || pattern[0] == b'*' |
|
195 | || pattern[0] == b'*' | |
196 | || pattern.starts_with(b".*") |
|
196 | || pattern.starts_with(b".*") | |
197 | { |
|
197 | { | |
198 | return pattern.to_owned(); |
|
198 | return pattern.to_owned(); | |
199 | } |
|
199 | } | |
200 | match FLAG_RE.find(pattern) { |
|
200 | match FLAG_RE.find(pattern) { | |
201 | Some(mat) => { |
|
201 | Some(mat) => { | |
202 | let s = mat.start(); |
|
202 | let s = mat.start(); | |
203 | let e = mat.end(); |
|
203 | let e = mat.end(); | |
204 | [ |
|
204 | [ | |
205 | &b"(?"[..], |
|
205 | &b"(?"[..], | |
206 | &pattern[s + 2..e - 1], |
|
206 | &pattern[s + 2..e - 1], | |
207 | &b":"[..], |
|
207 | &b":"[..], | |
208 | if pattern[e] == b'^' |
|
208 | if pattern[e] == b'^' | |
209 | || pattern[e] == b'*' |
|
209 | || pattern[e] == b'*' | |
210 | || pattern[e..].starts_with(b".*") |
|
210 | || pattern[e..].starts_with(b".*") | |
211 | { |
|
211 | { | |
212 | &b""[..] |
|
212 | &b""[..] | |
213 | } else { |
|
213 | } else { | |
214 | &b".*"[..] |
|
214 | &b".*"[..] | |
215 | }, |
|
215 | }, | |
216 | &pattern[e..], |
|
216 | &pattern[e..], | |
217 | &b")"[..], |
|
217 | &b")"[..], | |
218 | ] |
|
218 | ] | |
219 | .concat() |
|
219 | .concat() | |
220 | } |
|
220 | } | |
221 | None => [&b".*"[..], pattern].concat(), |
|
221 | None => [&b".*"[..], pattern].concat(), | |
222 | } |
|
222 | } | |
223 | } |
|
223 | } | |
224 | PatternSyntax::Path | PatternSyntax::RelPath => { |
|
224 | PatternSyntax::Path | PatternSyntax::RelPath => { | |
225 | if pattern == b"." { |
|
225 | if pattern == b"." { | |
226 | return vec![]; |
|
226 | return vec![]; | |
227 | } |
|
227 | } | |
228 | [escape_pattern(pattern).as_slice(), b"(?:/|$)"].concat() |
|
228 | [escape_pattern(pattern).as_slice(), b"(?:/|$)"].concat() | |
229 | } |
|
229 | } | |
230 | PatternSyntax::RootFiles => { |
|
230 | PatternSyntax::RootFiles => { | |
231 | let mut res = if pattern == b"." { |
|
231 | let mut res = if pattern == b"." { | |
232 | vec![] |
|
232 | vec![] | |
233 | } else { |
|
233 | } else { | |
234 | // Pattern is a directory name. |
|
234 | // Pattern is a directory name. | |
235 | [escape_pattern(pattern).as_slice(), b"/"].concat() |
|
235 | [escape_pattern(pattern).as_slice(), b"/"].concat() | |
236 | }; |
|
236 | }; | |
237 |
|
237 | |||
238 | // Anything after the pattern must be a non-directory. |
|
238 | // Anything after the pattern must be a non-directory. | |
239 | res.extend(b"[^/]+$"); |
|
239 | res.extend(b"[^/]+$"); | |
240 | res |
|
240 | res | |
241 | } |
|
241 | } | |
242 | PatternSyntax::RelGlob => { |
|
242 | PatternSyntax::RelGlob => { | |
243 | let glob_re = glob_to_re(pattern); |
|
243 | let glob_re = glob_to_re(pattern); | |
244 | if let Some(rest) = glob_re.drop_prefix(b"[^/]*") { |
|
244 | if let Some(rest) = glob_re.drop_prefix(b"[^/]*") { | |
245 | [b".*", rest, glob_suffix].concat() |
|
245 | [b".*", rest, glob_suffix].concat() | |
246 | } else { |
|
246 | } else { | |
247 | [b"(?:.*/)?", glob_re.as_slice(), glob_suffix].concat() |
|
247 | [b"(?:.*/)?", glob_re.as_slice(), glob_suffix].concat() | |
248 | } |
|
248 | } | |
249 | } |
|
249 | } | |
250 | PatternSyntax::Glob | PatternSyntax::RootGlob => { |
|
250 | PatternSyntax::Glob | PatternSyntax::RootGlob => { | |
251 | [glob_to_re(pattern).as_slice(), glob_suffix].concat() |
|
251 | [glob_to_re(pattern).as_slice(), glob_suffix].concat() | |
252 | } |
|
252 | } | |
253 | PatternSyntax::Include |
|
253 | PatternSyntax::Include | |
254 | | PatternSyntax::SubInclude |
|
254 | | PatternSyntax::SubInclude | |
255 | | PatternSyntax::ExpandedSubInclude(_) |
|
255 | | PatternSyntax::ExpandedSubInclude(_) | |
256 | | PatternSyntax::FilePath => unreachable!(), |
|
256 | | PatternSyntax::FilePath => unreachable!(), | |
257 | } |
|
257 | } | |
258 | } |
|
258 | } | |
259 |
|
259 | |||
260 | const GLOB_SPECIAL_CHARACTERS: [u8; 7] = |
|
260 | const GLOB_SPECIAL_CHARACTERS: [u8; 7] = | |
261 | [b'*', b'?', b'[', b']', b'{', b'}', b'\\']; |
|
261 | [b'*', b'?', b'[', b']', b'{', b'}', b'\\']; | |
262 |
|
262 | |||
263 | /// TODO support other platforms |
|
263 | /// TODO support other platforms | |
264 | #[cfg(unix)] |
|
264 | #[cfg(unix)] | |
265 | pub fn normalize_path_bytes(bytes: &[u8]) -> Vec<u8> { |
|
265 | pub fn normalize_path_bytes(bytes: &[u8]) -> Vec<u8> { | |
266 | if bytes.is_empty() { |
|
266 | if bytes.is_empty() { | |
267 | return b".".to_vec(); |
|
267 | return b".".to_vec(); | |
268 | } |
|
268 | } | |
269 | let sep = b'/'; |
|
269 | let sep = b'/'; | |
270 |
|
270 | |||
271 | let mut initial_slashes = bytes.iter().take_while(|b| **b == sep).count(); |
|
271 | let mut initial_slashes = bytes.iter().take_while(|b| **b == sep).count(); | |
272 | if initial_slashes > 2 { |
|
272 | if initial_slashes > 2 { | |
273 | // POSIX allows one or two initial slashes, but treats three or more |
|
273 | // POSIX allows one or two initial slashes, but treats three or more | |
274 | // as single slash. |
|
274 | // as single slash. | |
275 | initial_slashes = 1; |
|
275 | initial_slashes = 1; | |
276 | } |
|
276 | } | |
277 | let components = bytes |
|
277 | let components = bytes | |
278 | .split(|b| *b == sep) |
|
278 | .split(|b| *b == sep) | |
279 | .filter(|c| !(c.is_empty() || c == b".")) |
|
279 | .filter(|c| !(c.is_empty() || c == b".")) | |
280 | .fold(vec![], |mut acc, component| { |
|
280 | .fold(vec![], |mut acc, component| { | |
281 | if component != b".." |
|
281 | if component != b".." | |
282 | || (initial_slashes == 0 && acc.is_empty()) |
|
282 | || (initial_slashes == 0 && acc.is_empty()) | |
283 | || (!acc.is_empty() && acc[acc.len() - 1] == b"..") |
|
283 | || (!acc.is_empty() && acc[acc.len() - 1] == b"..") | |
284 | { |
|
284 | { | |
285 | acc.push(component) |
|
285 | acc.push(component) | |
286 | } else if !acc.is_empty() { |
|
286 | } else if !acc.is_empty() { | |
287 | acc.pop(); |
|
287 | acc.pop(); | |
288 | } |
|
288 | } | |
289 | acc |
|
289 | acc | |
290 | }); |
|
290 | }); | |
291 | let mut new_bytes = components.join(&sep); |
|
291 | let mut new_bytes = components.join(&sep); | |
292 |
|
292 | |||
293 | if initial_slashes > 0 { |
|
293 | if initial_slashes > 0 { | |
294 | let mut buf: Vec<_> = (0..initial_slashes).map(|_| sep).collect(); |
|
294 | let mut buf: Vec<_> = (0..initial_slashes).map(|_| sep).collect(); | |
295 | buf.extend(new_bytes); |
|
295 | buf.extend(new_bytes); | |
296 | new_bytes = buf; |
|
296 | new_bytes = buf; | |
297 | } |
|
297 | } | |
298 | if new_bytes.is_empty() { |
|
298 | if new_bytes.is_empty() { | |
299 | b".".to_vec() |
|
299 | b".".to_vec() | |
300 | } else { |
|
300 | } else { | |
301 | new_bytes |
|
301 | new_bytes | |
302 | } |
|
302 | } | |
303 | } |
|
303 | } | |
304 |
|
304 | |||
305 | /// Wrapper function to `_build_single_regex` that short-circuits 'exact' globs |
|
305 | /// Wrapper function to `_build_single_regex` that short-circuits 'exact' globs | |
306 | /// that don't need to be transformed into a regex. |
|
306 | /// that don't need to be transformed into a regex. | |
307 | pub fn build_single_regex( |
|
307 | pub fn build_single_regex( | |
308 | entry: &IgnorePattern, |
|
308 | entry: &IgnorePattern, | |
309 | glob_suffix: &[u8], |
|
309 | glob_suffix: &[u8], | |
310 | ) -> Result<Option<Vec<u8>>, PatternError> { |
|
310 | ) -> Result<Option<Vec<u8>>, PatternError> { | |
311 | let IgnorePattern { |
|
311 | let IgnorePattern { | |
312 | pattern, syntax, .. |
|
312 | pattern, syntax, .. | |
313 | } = entry; |
|
313 | } = entry; | |
314 | let pattern = match syntax { |
|
314 | let pattern = match syntax { | |
315 | PatternSyntax::RootGlob |
|
315 | PatternSyntax::RootGlob | |
316 | | PatternSyntax::Path |
|
316 | | PatternSyntax::Path | |
317 | | PatternSyntax::RelGlob |
|
317 | | PatternSyntax::RelGlob | |
318 | | PatternSyntax::RelPath |
|
318 | | PatternSyntax::RelPath | |
319 | | PatternSyntax::RootFiles => normalize_path_bytes(pattern), |
|
319 | | PatternSyntax::RootFiles => normalize_path_bytes(pattern), | |
320 | PatternSyntax::Include | PatternSyntax::SubInclude => { |
|
320 | PatternSyntax::Include | PatternSyntax::SubInclude => { | |
321 | return Err(PatternError::NonRegexPattern(entry.clone())) |
|
321 | return Err(PatternError::NonRegexPattern(entry.clone())) | |
322 | } |
|
322 | } | |
323 | _ => pattern.to_owned(), |
|
323 | _ => pattern.to_owned(), | |
324 | }; |
|
324 | }; | |
325 | let is_simple_rootglob = *syntax == PatternSyntax::RootGlob |
|
325 | let is_simple_rootglob = *syntax == PatternSyntax::RootGlob | |
326 | && !pattern.iter().any(|b| GLOB_SPECIAL_CHARACTERS.contains(b)); |
|
326 | && !pattern.iter().any(|b| GLOB_SPECIAL_CHARACTERS.contains(b)); | |
327 | if is_simple_rootglob || syntax == &PatternSyntax::FilePath { |
|
327 | if is_simple_rootglob || syntax == &PatternSyntax::FilePath { | |
328 | Ok(None) |
|
328 | Ok(None) | |
329 | } else { |
|
329 | } else { | |
330 | let mut entry = entry.clone(); |
|
330 | let mut entry = entry.clone(); | |
331 | entry.pattern = pattern; |
|
331 | entry.pattern = pattern; | |
332 | Ok(Some(_build_single_regex(&entry, glob_suffix))) |
|
332 | Ok(Some(_build_single_regex(&entry, glob_suffix))) | |
333 | } |
|
333 | } | |
334 | } |
|
334 | } | |
335 |
|
335 | |||
336 | lazy_static! { |
|
336 | lazy_static! { | |
337 | static ref SYNTAXES: FastHashMap<&'static [u8], PatternSyntax> = { |
|
337 | static ref SYNTAXES: FastHashMap<&'static [u8], PatternSyntax> = { | |
338 | let mut m = FastHashMap::default(); |
|
338 | let mut m = FastHashMap::default(); | |
339 |
|
339 | |||
340 | m.insert(b"re:".as_ref(), PatternSyntax::Regexp); |
|
340 | m.insert(b"re:".as_ref(), PatternSyntax::Regexp); | |
341 | m.insert(b"regexp:".as_ref(), PatternSyntax::Regexp); |
|
341 | m.insert(b"regexp:".as_ref(), PatternSyntax::Regexp); | |
342 | m.insert(b"path:".as_ref(), PatternSyntax::Path); |
|
342 | m.insert(b"path:".as_ref(), PatternSyntax::Path); | |
343 | m.insert(b"filepath:".as_ref(), PatternSyntax::FilePath); |
|
343 | m.insert(b"filepath:".as_ref(), PatternSyntax::FilePath); | |
344 | m.insert(b"relpath:".as_ref(), PatternSyntax::RelPath); |
|
344 | m.insert(b"relpath:".as_ref(), PatternSyntax::RelPath); | |
345 | m.insert(b"rootfilesin:".as_ref(), PatternSyntax::RootFiles); |
|
345 | m.insert(b"rootfilesin:".as_ref(), PatternSyntax::RootFiles); | |
346 | m.insert(b"relglob:".as_ref(), PatternSyntax::RelGlob); |
|
346 | m.insert(b"relglob:".as_ref(), PatternSyntax::RelGlob); | |
347 | m.insert(b"relre:".as_ref(), PatternSyntax::RelRegexp); |
|
347 | m.insert(b"relre:".as_ref(), PatternSyntax::RelRegexp); | |
348 | m.insert(b"glob:".as_ref(), PatternSyntax::Glob); |
|
348 | m.insert(b"glob:".as_ref(), PatternSyntax::Glob); | |
349 | m.insert(b"rootglob:".as_ref(), PatternSyntax::RootGlob); |
|
349 | m.insert(b"rootglob:".as_ref(), PatternSyntax::RootGlob); | |
350 | m.insert(b"include:".as_ref(), PatternSyntax::Include); |
|
350 | m.insert(b"include:".as_ref(), PatternSyntax::Include); | |
351 | m.insert(b"subinclude:".as_ref(), PatternSyntax::SubInclude); |
|
351 | m.insert(b"subinclude:".as_ref(), PatternSyntax::SubInclude); | |
352 |
|
352 | |||
353 | m |
|
353 | m | |
354 | }; |
|
354 | }; | |
355 | } |
|
355 | } | |
356 |
|
356 | |||
357 | #[derive(Debug)] |
|
357 | #[derive(Debug)] | |
358 | pub enum PatternFileWarning { |
|
358 | pub enum PatternFileWarning { | |
359 | /// (file path, syntax bytes) |
|
359 | /// (file path, syntax bytes) | |
360 | InvalidSyntax(PathBuf, Vec<u8>), |
|
360 | InvalidSyntax(PathBuf, Vec<u8>), | |
361 | /// File path |
|
361 | /// File path | |
362 | NoSuchFile(PathBuf), |
|
362 | NoSuchFile(PathBuf), | |
363 | } |
|
363 | } | |
364 |
|
364 | |||
365 | pub fn parse_one_pattern( |
|
365 | pub fn parse_one_pattern( | |
366 | pattern: &[u8], |
|
366 | pattern: &[u8], | |
367 | source: &Path, |
|
367 | source: &Path, | |
368 | default: PatternSyntax, |
|
368 | default: PatternSyntax, | |
369 | normalize: bool, |
|
369 | normalize: bool, | |
370 | ) -> IgnorePattern { |
|
370 | ) -> IgnorePattern { | |
371 | let mut pattern_bytes: &[u8] = pattern; |
|
371 | let mut pattern_bytes: &[u8] = pattern; | |
372 | let mut syntax = default; |
|
372 | let mut syntax = default; | |
373 |
|
373 | |||
374 | for (s, val) in SYNTAXES.iter() { |
|
374 | for (s, val) in SYNTAXES.iter() { | |
375 | if let Some(rest) = pattern_bytes.drop_prefix(s) { |
|
375 | if let Some(rest) = pattern_bytes.drop_prefix(s) { | |
376 | syntax = val.clone(); |
|
376 | syntax = val.clone(); | |
377 | pattern_bytes = rest; |
|
377 | pattern_bytes = rest; | |
378 | break; |
|
378 | break; | |
379 | } |
|
379 | } | |
380 | } |
|
380 | } | |
381 |
|
381 | |||
382 | let pattern = match syntax { |
|
382 | let pattern = match syntax { | |
383 | PatternSyntax::RootGlob |
|
383 | PatternSyntax::RootGlob | |
384 | | PatternSyntax::Path |
|
384 | | PatternSyntax::Path | |
385 | | PatternSyntax::Glob |
|
385 | | PatternSyntax::Glob | |
386 | | PatternSyntax::RelGlob |
|
386 | | PatternSyntax::RelGlob | |
387 | | PatternSyntax::RelPath |
|
387 | | PatternSyntax::RelPath | |
388 | | PatternSyntax::RootFiles |
|
388 | | PatternSyntax::RootFiles | |
389 | if normalize => |
|
389 | if normalize => | |
390 | { |
|
390 | { | |
391 | normalize_path_bytes(pattern_bytes) |
|
391 | normalize_path_bytes(pattern_bytes) | |
392 | } |
|
392 | } | |
393 | _ => pattern_bytes.to_vec(), |
|
393 | _ => pattern_bytes.to_vec(), | |
394 | }; |
|
394 | }; | |
395 |
|
395 | |||
396 | IgnorePattern { |
|
396 | IgnorePattern { | |
397 | syntax, |
|
397 | syntax, | |
398 | pattern, |
|
398 | pattern, | |
399 | source: source.to_owned(), |
|
399 | source: source.to_owned(), | |
400 | } |
|
400 | } | |
401 | } |
|
401 | } | |
402 |
|
402 | |||
403 | pub fn parse_pattern_file_contents( |
|
403 | pub fn parse_pattern_file_contents( | |
404 | lines: &[u8], |
|
404 | lines: &[u8], | |
405 | file_path: &Path, |
|
405 | file_path: &Path, | |
406 | default_syntax_override: Option<PatternSyntax>, |
|
406 | default_syntax_override: Option<PatternSyntax>, | |
407 | warn: bool, |
|
407 | warn: bool, | |
408 | relativize: bool, |
|
408 | relativize: bool, | |
409 | ) -> Result<(Vec<IgnorePattern>, Vec<PatternFileWarning>), PatternError> { |
|
409 | ) -> Result<(Vec<IgnorePattern>, Vec<PatternFileWarning>), PatternError> { | |
410 | let comment_regex = Regex::new(r"((?:^|[^\\])(?:\\\\)*)#.*").unwrap(); |
|
410 | let comment_regex = Regex::new(r"((?:^|[^\\])(?:\\\\)*)#.*").unwrap(); | |
411 |
|
411 | |||
412 | #[allow(clippy::trivial_regex)] |
|
412 | #[allow(clippy::trivial_regex)] | |
413 | let comment_escape_regex = Regex::new(r"\\#").unwrap(); |
|
413 | let comment_escape_regex = Regex::new(r"\\#").unwrap(); | |
414 | let mut inputs: Vec<IgnorePattern> = vec![]; |
|
414 | let mut inputs: Vec<IgnorePattern> = vec![]; | |
415 | let mut warnings: Vec<PatternFileWarning> = vec![]; |
|
415 | let mut warnings: Vec<PatternFileWarning> = vec![]; | |
416 |
|
416 | |||
417 | let mut current_syntax = |
|
417 | let mut current_syntax = | |
418 | default_syntax_override.unwrap_or(PatternSyntax::RelRegexp); |
|
418 | default_syntax_override.unwrap_or(PatternSyntax::RelRegexp); | |
419 |
|
419 | |||
420 | for mut line in lines.split(|c| *c == b'\n') { |
|
420 | for mut line in lines.split(|c| *c == b'\n') { | |
421 | let line_buf; |
|
421 | let line_buf; | |
422 | if line.contains(&b'#') { |
|
422 | if line.contains(&b'#') { | |
423 | if let Some(cap) = comment_regex.captures(line) { |
|
423 | if let Some(cap) = comment_regex.captures(line) { | |
424 | line = &line[..cap.get(1).unwrap().end()] |
|
424 | line = &line[..cap.get(1).unwrap().end()] | |
425 | } |
|
425 | } | |
426 | line_buf = comment_escape_regex.replace_all(line, NoExpand(b"#")); |
|
426 | line_buf = comment_escape_regex.replace_all(line, NoExpand(b"#")); | |
427 | line = &line_buf; |
|
427 | line = &line_buf; | |
428 | } |
|
428 | } | |
429 |
|
429 | |||
430 | let line = line.trim_end(); |
|
430 | let line = line.trim_end(); | |
431 |
|
431 | |||
432 | if line.is_empty() { |
|
432 | if line.is_empty() { | |
433 | continue; |
|
433 | continue; | |
434 | } |
|
434 | } | |
435 |
|
435 | |||
436 | if let Some(syntax) = line.drop_prefix(b"syntax:") { |
|
436 | if let Some(syntax) = line.drop_prefix(b"syntax:") { | |
437 | let syntax = syntax.trim(); |
|
437 | let syntax = syntax.trim(); | |
438 |
|
438 | |||
439 | if let Some(parsed) = |
|
439 | if let Some(parsed) = | |
440 | SYNTAXES.get([syntax, &b":"[..]].concat().as_slice()) |
|
440 | SYNTAXES.get([syntax, &b":"[..]].concat().as_slice()) | |
441 | { |
|
441 | { | |
442 | current_syntax = parsed.clone(); |
|
442 | current_syntax = parsed.clone(); | |
443 | } else if warn { |
|
443 | } else if warn { | |
444 | warnings.push(PatternFileWarning::InvalidSyntax( |
|
444 | warnings.push(PatternFileWarning::InvalidSyntax( | |
445 | file_path.to_owned(), |
|
445 | file_path.to_owned(), | |
446 | syntax.to_owned(), |
|
446 | syntax.to_owned(), | |
447 | )); |
|
447 | )); | |
448 | } |
|
448 | } | |
449 | } else { |
|
449 | } else { | |
450 | let pattern = parse_one_pattern( |
|
450 | let pattern = parse_one_pattern( | |
451 | line, |
|
451 | line, | |
452 | file_path, |
|
452 | file_path, | |
453 | current_syntax.clone(), |
|
453 | current_syntax.clone(), | |
454 | false, |
|
454 | false, | |
455 | ); |
|
455 | ); | |
456 | inputs.push(if relativize { |
|
456 | inputs.push(if relativize { | |
457 | pattern.to_relative() |
|
457 | pattern.to_relative() | |
458 | } else { |
|
458 | } else { | |
459 | pattern |
|
459 | pattern | |
460 | }) |
|
460 | }) | |
461 | } |
|
461 | } | |
462 | } |
|
462 | } | |
463 | Ok((inputs, warnings)) |
|
463 | Ok((inputs, warnings)) | |
464 | } |
|
464 | } | |
465 |
|
465 | |||
466 | pub fn parse_pattern_args( |
|
466 | pub fn parse_pattern_args( | |
467 | patterns: Vec<Vec<u8>>, |
|
467 | patterns: Vec<Vec<u8>>, | |
468 | cwd: &Path, |
|
468 | cwd: &Path, | |
469 | root: &Path, |
|
469 | root: &Path, | |
470 | ) -> Result<Vec<IgnorePattern>, HgPathError> { |
|
470 | ) -> Result<Vec<IgnorePattern>, HgPathError> { | |
471 | let mut ignore_patterns: Vec<IgnorePattern> = Vec::new(); |
|
471 | let mut ignore_patterns: Vec<IgnorePattern> = Vec::new(); | |
472 | for pattern in patterns { |
|
472 | for pattern in patterns { | |
473 | let pattern = parse_one_pattern( |
|
473 | let pattern = parse_one_pattern( | |
474 | &pattern, |
|
474 | &pattern, | |
475 | Path::new("<args>"), |
|
475 | Path::new("<args>"), | |
476 | PatternSyntax::RelPath, |
|
476 | PatternSyntax::RelPath, | |
477 | true, |
|
477 | true, | |
478 | ); |
|
478 | ); | |
479 | match pattern.syntax { |
|
479 | match pattern.syntax { | |
480 | PatternSyntax::RelGlob | PatternSyntax::RelPath => { |
|
480 | PatternSyntax::RelGlob | PatternSyntax::RelPath => { | |
481 | let name = get_path_from_bytes(&pattern.pattern); |
|
481 | let name = get_path_from_bytes(&pattern.pattern); | |
482 | let canon = canonical_path(root, cwd, name)?; |
|
482 | let canon = canonical_path(root, cwd, name)?; | |
483 | ignore_patterns.push(IgnorePattern { |
|
483 | ignore_patterns.push(IgnorePattern { | |
484 | syntax: pattern.syntax, |
|
484 | syntax: pattern.syntax, | |
485 | pattern: get_bytes_from_path(canon), |
|
485 | pattern: get_bytes_from_path(canon), | |
486 | source: pattern.source, |
|
486 | source: pattern.source, | |
487 | }) |
|
487 | }) | |
488 | } |
|
488 | } | |
489 | _ => ignore_patterns.push(pattern.to_owned()), |
|
489 | _ => ignore_patterns.push(pattern.to_owned()), | |
490 | }; |
|
490 | }; | |
491 | } |
|
491 | } | |
492 | Ok(ignore_patterns) |
|
492 | Ok(ignore_patterns) | |
493 | } |
|
493 | } | |
494 |
|
494 | |||
495 | pub fn read_pattern_file( |
|
495 | pub fn read_pattern_file( | |
496 | file_path: &Path, |
|
496 | file_path: &Path, | |
497 | warn: bool, |
|
497 | warn: bool, | |
498 | inspect_pattern_bytes: &mut impl FnMut(&Path, &[u8]), |
|
498 | inspect_pattern_bytes: &mut impl FnMut(&Path, &[u8]), | |
499 | ) -> Result<(Vec<IgnorePattern>, Vec<PatternFileWarning>), PatternError> { |
|
499 | ) -> Result<(Vec<IgnorePattern>, Vec<PatternFileWarning>), PatternError> { | |
500 | match std::fs::read(file_path) { |
|
500 | match std::fs::read(file_path) { | |
501 | Ok(contents) => { |
|
501 | Ok(contents) => { | |
502 | inspect_pattern_bytes(file_path, &contents); |
|
502 | inspect_pattern_bytes(file_path, &contents); | |
503 | parse_pattern_file_contents(&contents, file_path, None, warn, true) |
|
503 | parse_pattern_file_contents(&contents, file_path, None, warn, true) | |
504 | } |
|
504 | } | |
505 | Err(e) if e.kind() == std::io::ErrorKind::NotFound => Ok(( |
|
505 | Err(e) if e.kind() == std::io::ErrorKind::NotFound => Ok(( | |
506 | vec![], |
|
506 | vec![], | |
507 | vec![PatternFileWarning::NoSuchFile(file_path.to_owned())], |
|
507 | vec![PatternFileWarning::NoSuchFile(file_path.to_owned())], | |
508 | )), |
|
508 | )), | |
509 | Err(e) => Err(e.into()), |
|
509 | Err(e) => Err(e.into()), | |
510 | } |
|
510 | } | |
511 | } |
|
511 | } | |
512 |
|
512 | |||
513 | /// Represents an entry in an "ignore" file. |
|
513 | /// Represents an entry in an "ignore" file. | |
514 | #[derive(Debug, Eq, PartialEq, Clone)] |
|
514 | #[derive(Debug, Eq, PartialEq, Clone)] | |
515 | pub struct IgnorePattern { |
|
515 | pub struct IgnorePattern { | |
516 | pub syntax: PatternSyntax, |
|
516 | pub syntax: PatternSyntax, | |
517 | pub pattern: Vec<u8>, |
|
517 | pub pattern: Vec<u8>, | |
518 | pub source: PathBuf, |
|
518 | pub source: PathBuf, | |
519 | } |
|
519 | } | |
520 |
|
520 | |||
521 | impl IgnorePattern { |
|
521 | impl IgnorePattern { | |
522 | pub fn new(syntax: PatternSyntax, pattern: &[u8], source: &Path) -> Self { |
|
522 | pub fn new(syntax: PatternSyntax, pattern: &[u8], source: &Path) -> Self { | |
523 | Self { |
|
523 | Self { | |
524 | syntax, |
|
524 | syntax, | |
525 | pattern: pattern.to_owned(), |
|
525 | pattern: pattern.to_owned(), | |
526 | source: source.to_owned(), |
|
526 | source: source.to_owned(), | |
527 | } |
|
527 | } | |
528 | } |
|
528 | } | |
529 |
|
529 | |||
530 | pub fn to_relative(self) -> Self { |
|
530 | pub fn to_relative(self) -> Self { | |
531 | let Self { |
|
531 | let Self { | |
532 | syntax, |
|
532 | syntax, | |
533 | pattern, |
|
533 | pattern, | |
534 | source, |
|
534 | source, | |
535 | } = self; |
|
535 | } = self; | |
536 | Self { |
|
536 | Self { | |
537 | syntax: match syntax { |
|
537 | syntax: match syntax { | |
538 | PatternSyntax::Regexp => PatternSyntax::RelRegexp, |
|
538 | PatternSyntax::Regexp => PatternSyntax::RelRegexp, | |
539 | PatternSyntax::Glob => PatternSyntax::RelGlob, |
|
539 | PatternSyntax::Glob => PatternSyntax::RelGlob, | |
540 | x => x, |
|
540 | x => x, | |
541 | }, |
|
541 | }, | |
542 | pattern, |
|
542 | pattern, | |
543 | source, |
|
543 | source, | |
544 | } |
|
544 | } | |
545 | } |
|
545 | } | |
546 | } |
|
546 | } | |
547 |
|
547 | |||
548 | pub type PatternResult<T> = Result<T, PatternError>; |
|
548 | pub type PatternResult<T> = Result<T, PatternError>; | |
549 |
|
549 | |||
550 | /// Wrapper for `read_pattern_file` that also recursively expands `include:` |
|
550 | /// Wrapper for `read_pattern_file` that also recursively expands `include:` | |
551 | /// and `subinclude:` patterns. |
|
551 | /// and `subinclude:` patterns. | |
552 | /// |
|
552 | /// | |
553 | /// The former are expanded in place, while `PatternSyntax::ExpandedSubInclude` |
|
553 | /// The former are expanded in place, while `PatternSyntax::ExpandedSubInclude` | |
554 | /// is used for the latter to form a tree of patterns. |
|
554 | /// is used for the latter to form a tree of patterns. | |
555 | pub fn get_patterns_from_file( |
|
555 | pub fn get_patterns_from_file( | |
556 | pattern_file: &Path, |
|
556 | pattern_file: &Path, | |
557 | root_dir: &Path, |
|
557 | root_dir: &Path, | |
558 | inspect_pattern_bytes: &mut impl FnMut(&Path, &[u8]), |
|
558 | inspect_pattern_bytes: &mut impl FnMut(&Path, &[u8]), | |
559 | ) -> PatternResult<(Vec<IgnorePattern>, Vec<PatternFileWarning>)> { |
|
559 | ) -> PatternResult<(Vec<IgnorePattern>, Vec<PatternFileWarning>)> { | |
560 | let (patterns, mut warnings) = |
|
560 | let (patterns, mut warnings) = | |
561 | read_pattern_file(pattern_file, true, inspect_pattern_bytes)?; |
|
561 | read_pattern_file(pattern_file, true, inspect_pattern_bytes)?; | |
562 | let patterns = patterns |
|
562 | let patterns = patterns | |
563 | .into_iter() |
|
563 | .into_iter() | |
564 | .flat_map(|entry| -> PatternResult<_> { |
|
564 | .flat_map(|entry| -> PatternResult<_> { | |
565 | Ok(match &entry.syntax { |
|
565 | Ok(match &entry.syntax { | |
566 | PatternSyntax::Include => { |
|
566 | PatternSyntax::Include => { | |
567 | let inner_include = |
|
567 | let inner_include = | |
568 | root_dir.join(get_path_from_bytes(&entry.pattern)); |
|
568 | root_dir.join(get_path_from_bytes(&entry.pattern)); | |
569 | let (inner_pats, inner_warnings) = get_patterns_from_file( |
|
569 | let (inner_pats, inner_warnings) = get_patterns_from_file( | |
570 | &inner_include, |
|
570 | &inner_include, | |
571 | root_dir, |
|
571 | root_dir, | |
572 | inspect_pattern_bytes, |
|
572 | inspect_pattern_bytes, | |
573 | )?; |
|
573 | )?; | |
574 | warnings.extend(inner_warnings); |
|
574 | warnings.extend(inner_warnings); | |
575 | inner_pats |
|
575 | inner_pats | |
576 | } |
|
576 | } | |
577 | PatternSyntax::SubInclude => { |
|
577 | PatternSyntax::SubInclude => { | |
578 | let mut sub_include = SubInclude::new( |
|
578 | let mut sub_include = SubInclude::new( | |
579 | root_dir, |
|
579 | root_dir, | |
580 | &entry.pattern, |
|
580 | &entry.pattern, | |
581 | &entry.source, |
|
581 | &entry.source, | |
582 | )?; |
|
582 | )?; | |
583 | let (inner_patterns, inner_warnings) = |
|
583 | let (inner_patterns, inner_warnings) = | |
584 | get_patterns_from_file( |
|
584 | get_patterns_from_file( | |
585 | &sub_include.path, |
|
585 | &sub_include.path, | |
586 | &sub_include.root, |
|
586 | &sub_include.root, | |
587 | inspect_pattern_bytes, |
|
587 | inspect_pattern_bytes, | |
588 | )?; |
|
588 | )?; | |
589 | sub_include.included_patterns = inner_patterns; |
|
589 | sub_include.included_patterns = inner_patterns; | |
590 | warnings.extend(inner_warnings); |
|
590 | warnings.extend(inner_warnings); | |
591 | vec![IgnorePattern { |
|
591 | vec![IgnorePattern { | |
592 | syntax: PatternSyntax::ExpandedSubInclude(Box::new( |
|
592 | syntax: PatternSyntax::ExpandedSubInclude(Box::new( | |
593 | sub_include, |
|
593 | sub_include, | |
594 | )), |
|
594 | )), | |
595 | ..entry |
|
595 | ..entry | |
596 | }] |
|
596 | }] | |
597 | } |
|
597 | } | |
598 | _ => vec![entry], |
|
598 | _ => vec![entry], | |
599 | }) |
|
599 | }) | |
600 | }) |
|
600 | }) | |
601 | .flatten() |
|
601 | .flatten() | |
602 | .collect(); |
|
602 | .collect(); | |
603 |
|
603 | |||
604 | Ok((patterns, warnings)) |
|
604 | Ok((patterns, warnings)) | |
605 | } |
|
605 | } | |
606 |
|
606 | |||
607 | /// Holds all the information needed to handle a `subinclude:` pattern. |
|
607 | /// Holds all the information needed to handle a `subinclude:` pattern. | |
608 | #[derive(Debug, PartialEq, Eq, Clone)] |
|
608 | #[derive(Debug, PartialEq, Eq, Clone)] | |
609 | pub struct SubInclude { |
|
609 | pub struct SubInclude { | |
610 | /// Will be used for repository (hg) paths that start with this prefix. |
|
610 | /// Will be used for repository (hg) paths that start with this prefix. | |
611 | /// It is relative to the current working directory, so comparing against |
|
611 | /// It is relative to the current working directory, so comparing against | |
612 | /// repository paths is painless. |
|
612 | /// repository paths is painless. | |
613 | pub prefix: HgPathBuf, |
|
613 | pub prefix: HgPathBuf, | |
614 | /// The file itself, containing the patterns |
|
614 | /// The file itself, containing the patterns | |
615 | pub path: PathBuf, |
|
615 | pub path: PathBuf, | |
616 | /// Folder in the filesystem where this it applies |
|
616 | /// Folder in the filesystem where this it applies | |
617 | pub root: PathBuf, |
|
617 | pub root: PathBuf, | |
618 |
|
618 | |||
619 | pub included_patterns: Vec<IgnorePattern>, |
|
619 | pub included_patterns: Vec<IgnorePattern>, | |
620 | } |
|
620 | } | |
621 |
|
621 | |||
622 | impl SubInclude { |
|
622 | impl SubInclude { | |
623 | pub fn new( |
|
623 | pub fn new( | |
624 | root_dir: &Path, |
|
624 | root_dir: &Path, | |
625 | pattern: &[u8], |
|
625 | pattern: &[u8], | |
626 | source: &Path, |
|
626 | source: &Path, | |
627 | ) -> Result<SubInclude, HgPathError> { |
|
627 | ) -> Result<SubInclude, HgPathError> { | |
628 | let normalized_source = |
|
628 | let normalized_source = | |
629 | normalize_path_bytes(&get_bytes_from_path(source)); |
|
629 | normalize_path_bytes(&get_bytes_from_path(source)); | |
630 |
|
630 | |||
631 | let source_root = get_path_from_bytes(&normalized_source); |
|
631 | let source_root = get_path_from_bytes(&normalized_source); | |
632 | let source_root = source_root.parent().unwrap_or(source_root); |
|
632 | let source_root = source_root.parent().unwrap_or(source_root); | |
633 |
|
633 | |||
634 | let path = source_root.join(get_path_from_bytes(pattern)); |
|
634 | let path = source_root.join(get_path_from_bytes(pattern)); | |
635 | let new_root = path.parent().unwrap_or_else(|| path.deref()); |
|
635 | let new_root = path.parent().unwrap_or_else(|| path.deref()); | |
636 |
|
636 | |||
637 | let prefix = canonical_path(root_dir, root_dir, new_root)?; |
|
637 | let prefix = canonical_path(root_dir, root_dir, new_root)?; | |
638 |
|
638 | |||
639 | Ok(Self { |
|
639 | Ok(Self { | |
640 | prefix: path_to_hg_path_buf(prefix).map(|mut p| { |
|
640 | prefix: path_to_hg_path_buf(prefix).map(|mut p| { | |
641 | if !p.is_empty() { |
|
641 | if !p.is_empty() { | |
642 | p.push_byte(b'/'); |
|
642 | p.push_byte(b'/'); | |
643 | } |
|
643 | } | |
644 | p |
|
644 | p | |
645 | })?, |
|
645 | })?, | |
646 | path: path.to_owned(), |
|
646 | path: path.to_owned(), | |
647 | root: new_root.to_owned(), |
|
647 | root: new_root.to_owned(), | |
648 | included_patterns: Vec::new(), |
|
648 | included_patterns: Vec::new(), | |
649 | }) |
|
649 | }) | |
650 | } |
|
650 | } | |
651 | } |
|
651 | } | |
652 |
|
652 | |||
653 | /// Separate and pre-process subincludes from other patterns for the "ignore" |
|
653 | /// Separate and pre-process subincludes from other patterns for the "ignore" | |
654 | /// phase. |
|
654 | /// phase. | |
655 | pub fn filter_subincludes( |
|
655 | pub fn filter_subincludes( | |
656 | ignore_patterns: Vec<IgnorePattern>, |
|
656 | ignore_patterns: Vec<IgnorePattern>, | |
657 | ) -> Result<(Vec<SubInclude>, Vec<IgnorePattern>), HgPathError> { |
|
657 | ) -> Result<(Vec<SubInclude>, Vec<IgnorePattern>), HgPathError> { | |
658 | let mut subincludes = vec![]; |
|
658 | let mut subincludes = vec![]; | |
659 | let mut others = vec![]; |
|
659 | let mut others = vec![]; | |
660 |
|
660 | |||
661 | for pattern in ignore_patterns { |
|
661 | for pattern in ignore_patterns { | |
662 | if let PatternSyntax::ExpandedSubInclude(sub_include) = pattern.syntax |
|
662 | if let PatternSyntax::ExpandedSubInclude(sub_include) = pattern.syntax | |
663 | { |
|
663 | { | |
664 | subincludes.push(*sub_include); |
|
664 | subincludes.push(*sub_include); | |
665 | } else { |
|
665 | } else { | |
666 | others.push(pattern) |
|
666 | others.push(pattern) | |
667 | } |
|
667 | } | |
668 | } |
|
668 | } | |
669 | Ok((subincludes, others)) |
|
669 | Ok((subincludes, others)) | |
670 | } |
|
670 | } | |
671 |
|
671 | |||
672 | #[cfg(test)] |
|
672 | #[cfg(test)] | |
673 | mod tests { |
|
673 | mod tests { | |
674 | use super::*; |
|
674 | use super::*; | |
675 | use pretty_assertions::assert_eq; |
|
675 | use pretty_assertions::assert_eq; | |
676 |
|
676 | |||
677 | #[test] |
|
677 | #[test] | |
678 | fn escape_pattern_test() { |
|
678 | fn escape_pattern_test() { | |
679 | let untouched = |
|
679 | let untouched = | |
680 | br#"!"%',/0123456789:;<=>@ABCDEFGHIJKLMNOPQRSTUVWXYZ_`abcdefghijklmnopqrstuvwxyz"#; |
|
680 | br#"!"%',/0123456789:;<=>@ABCDEFGHIJKLMNOPQRSTUVWXYZ_`abcdefghijklmnopqrstuvwxyz"#; | |
681 | assert_eq!(escape_pattern(untouched), untouched.to_vec()); |
|
681 | assert_eq!(escape_pattern(untouched), untouched.to_vec()); | |
682 | // All escape codes |
|
682 | // All escape codes | |
683 | assert_eq!( |
|
683 | assert_eq!( | |
684 | escape_pattern(br"()[]{}?*+-|^$\\.&~#\t\n\r\v\f"), |
|
684 | escape_pattern(br"()[]{}?*+-|^$\\.&~#\t\n\r\v\f"), | |
685 | br"\(\)\[\]\{\}\?\*\+\-\|\^\$\\\\\.\&\~\#\\t\\n\\r\\v\\f".to_vec() |
|
685 | br"\(\)\[\]\{\}\?\*\+\-\|\^\$\\\\\.\&\~\#\\t\\n\\r\\v\\f".to_vec() | |
686 | ); |
|
686 | ); | |
687 | } |
|
687 | } | |
688 |
|
688 | |||
689 | #[test] |
|
689 | #[test] | |
690 | fn glob_test() { |
|
690 | fn glob_test() { | |
691 | assert_eq!(glob_to_re(br"?"), br"."); |
|
691 | assert_eq!(glob_to_re(br"?"), br"."); | |
692 | assert_eq!(glob_to_re(br"*"), br"[^/]*"); |
|
692 | assert_eq!(glob_to_re(br"*"), br"[^/]*"); | |
693 | assert_eq!(glob_to_re(br"**"), br".*"); |
|
693 | assert_eq!(glob_to_re(br"**"), br".*"); | |
694 | assert_eq!(glob_to_re(br"**/a"), br"(?:.*/)?a"); |
|
694 | assert_eq!(glob_to_re(br"**/a"), br"(?:.*/)?a"); | |
695 | assert_eq!(glob_to_re(br"a/**/b"), br"a/(?:.*/)?b"); |
|
695 | assert_eq!(glob_to_re(br"a/**/b"), br"a/(?:.*/)?b"); | |
696 | assert_eq!(glob_to_re(br"[a*?!^][^b][!c]"), br"[a*?!^][\^b][^c]"); |
|
696 | assert_eq!(glob_to_re(br"[a*?!^][^b][!c]"), br"[a*?!^][\^b][^c]"); | |
697 | assert_eq!(glob_to_re(br"{a,b}"), br"(?:a|b)"); |
|
697 | assert_eq!(glob_to_re(br"{a,b}"), br"(?:a|b)"); | |
698 | assert_eq!(glob_to_re(br".\*\?"), br"\.\*\?"); |
|
698 | assert_eq!(glob_to_re(br".\*\?"), br"\.\*\?"); | |
699 | } |
|
699 | } | |
700 |
|
700 | |||
701 | #[test] |
|
701 | #[test] | |
702 | fn test_parse_pattern_file_contents() { |
|
702 | fn test_parse_pattern_file_contents() { | |
703 | let lines = b"syntax: glob\n*.elc"; |
|
703 | let lines = b"syntax: glob\n*.elc"; | |
704 |
|
704 | |||
705 | assert_eq!( |
|
705 | assert_eq!( | |
706 | parse_pattern_file_contents( |
|
706 | parse_pattern_file_contents( | |
707 | lines, |
|
707 | lines, | |
708 | Path::new("file_path"), |
|
708 | Path::new("file_path"), | |
709 | None, |
|
709 | None, | |
710 | false, |
|
710 | false, | |
711 | true, |
|
711 | true, | |
712 | ) |
|
712 | ) | |
713 | .unwrap() |
|
713 | .unwrap() | |
714 | .0, |
|
714 | .0, | |
715 | vec![IgnorePattern::new( |
|
715 | vec![IgnorePattern::new( | |
716 | PatternSyntax::RelGlob, |
|
716 | PatternSyntax::RelGlob, | |
717 | b"*.elc", |
|
717 | b"*.elc", | |
718 | Path::new("file_path") |
|
718 | Path::new("file_path") | |
719 | )], |
|
719 | )], | |
720 | ); |
|
720 | ); | |
721 |
|
721 | |||
722 | let lines = b"syntax: include\nsyntax: glob"; |
|
722 | let lines = b"syntax: include\nsyntax: glob"; | |
723 |
|
723 | |||
724 | assert_eq!( |
|
724 | assert_eq!( | |
725 | parse_pattern_file_contents( |
|
725 | parse_pattern_file_contents( | |
726 | lines, |
|
726 | lines, | |
727 | Path::new("file_path"), |
|
727 | Path::new("file_path"), | |
728 | None, |
|
728 | None, | |
729 | false, |
|
729 | false, | |
730 | true, |
|
730 | true, | |
731 | ) |
|
731 | ) | |
732 | .unwrap() |
|
732 | .unwrap() | |
733 | .0, |
|
733 | .0, | |
734 | vec![] |
|
734 | vec![] | |
735 | ); |
|
735 | ); | |
736 | let lines = b"glob:**.o"; |
|
736 | let lines = b"glob:**.o"; | |
737 | assert_eq!( |
|
737 | assert_eq!( | |
738 | parse_pattern_file_contents( |
|
738 | parse_pattern_file_contents( | |
739 | lines, |
|
739 | lines, | |
740 | Path::new("file_path"), |
|
740 | Path::new("file_path"), | |
741 | None, |
|
741 | None, | |
742 | false, |
|
742 | false, | |
743 | true, |
|
743 | true, | |
744 | ) |
|
744 | ) | |
745 | .unwrap() |
|
745 | .unwrap() | |
746 | .0, |
|
746 | .0, | |
747 | vec![IgnorePattern::new( |
|
747 | vec![IgnorePattern::new( | |
748 | PatternSyntax::RelGlob, |
|
748 | PatternSyntax::RelGlob, | |
749 | b"**.o", |
|
749 | b"**.o", | |
750 | Path::new("file_path") |
|
750 | Path::new("file_path") | |
751 | )] |
|
751 | )] | |
752 | ); |
|
752 | ); | |
753 | } |
|
753 | } | |
754 |
|
754 | |||
755 | #[test] |
|
755 | #[test] | |
756 | fn test_build_single_regex() { |
|
756 | fn test_build_single_regex() { | |
757 | assert_eq!( |
|
757 | assert_eq!( | |
758 | build_single_regex( |
|
758 | build_single_regex( | |
759 | &IgnorePattern::new( |
|
759 | &IgnorePattern::new( | |
760 | PatternSyntax::RelGlob, |
|
760 | PatternSyntax::RelGlob, | |
761 | b"rust/target/", |
|
761 | b"rust/target/", | |
762 | Path::new("") |
|
762 | Path::new("") | |
763 | ), |
|
763 | ), | |
764 | b"(?:/|$)" |
|
764 | b"(?:/|$)" | |
765 | ) |
|
765 | ) | |
766 | .unwrap(), |
|
766 | .unwrap(), | |
767 | Some(br"(?:.*/)?rust/target(?:/|$)".to_vec()), |
|
767 | Some(br"(?:.*/)?rust/target(?:/|$)".to_vec()), | |
768 | ); |
|
768 | ); | |
769 | assert_eq!( |
|
769 | assert_eq!( | |
770 | build_single_regex( |
|
770 | build_single_regex( | |
771 | &IgnorePattern::new( |
|
771 | &IgnorePattern::new( | |
772 | PatternSyntax::Regexp, |
|
772 | PatternSyntax::Regexp, | |
773 | br"rust/target/\d+", |
|
773 | br"rust/target/\d+", | |
774 | Path::new("") |
|
774 | Path::new("") | |
775 | ), |
|
775 | ), | |
776 | b"(?:/|$)" |
|
776 | b"(?:/|$)" | |
777 | ) |
|
777 | ) | |
778 | .unwrap(), |
|
778 | .unwrap(), | |
779 | Some(br"rust/target/\d+".to_vec()), |
|
779 | Some(br"rust/target/\d+".to_vec()), | |
780 | ); |
|
780 | ); | |
781 | } |
|
781 | } | |
782 |
|
782 | |||
783 | #[test] |
|
783 | #[test] | |
784 | fn test_build_single_regex_shortcut() { |
|
784 | fn test_build_single_regex_shortcut() { | |
785 | assert_eq!( |
|
785 | assert_eq!( | |
786 | build_single_regex( |
|
786 | build_single_regex( | |
787 | &IgnorePattern::new( |
|
787 | &IgnorePattern::new( | |
788 | PatternSyntax::RootGlob, |
|
788 | PatternSyntax::RootGlob, | |
789 | b"", |
|
789 | b"", | |
790 | Path::new("") |
|
790 | Path::new("") | |
791 | ), |
|
791 | ), | |
792 | b"(?:/|$)" |
|
792 | b"(?:/|$)" | |
793 | ) |
|
793 | ) | |
794 | .unwrap(), |
|
794 | .unwrap(), | |
795 | None, |
|
795 | None, | |
796 | ); |
|
796 | ); | |
797 | assert_eq!( |
|
797 | assert_eq!( | |
798 | build_single_regex( |
|
798 | build_single_regex( | |
799 | &IgnorePattern::new( |
|
799 | &IgnorePattern::new( | |
800 | PatternSyntax::RootGlob, |
|
800 | PatternSyntax::RootGlob, | |
801 | b"whatever", |
|
801 | b"whatever", | |
802 | Path::new("") |
|
802 | Path::new("") | |
803 | ), |
|
803 | ), | |
804 | b"(?:/|$)" |
|
804 | b"(?:/|$)" | |
805 | ) |
|
805 | ) | |
806 | .unwrap(), |
|
806 | .unwrap(), | |
807 | None, |
|
807 | None, | |
808 | ); |
|
808 | ); | |
809 | assert_eq!( |
|
809 | assert_eq!( | |
810 | build_single_regex( |
|
810 | build_single_regex( | |
811 | &IgnorePattern::new( |
|
811 | &IgnorePattern::new( | |
812 | PatternSyntax::RootGlob, |
|
812 | PatternSyntax::RootGlob, | |
813 | b"*.o", |
|
813 | b"*.o", | |
814 | Path::new("") |
|
814 | Path::new("") | |
815 | ), |
|
815 | ), | |
816 | b"(?:/|$)" |
|
816 | b"(?:/|$)" | |
817 | ) |
|
817 | ) | |
818 | .unwrap(), |
|
818 | .unwrap(), | |
819 | Some(br"[^/]*\.o(?:/|$)".to_vec()), |
|
819 | Some(br"[^/]*\.o(?:/|$)".to_vec()), | |
820 | ); |
|
820 | ); | |
821 | } |
|
821 | } | |
822 |
|
822 | |||
823 | #[test] |
|
823 | #[test] | |
824 | fn test_build_single_relregex() { |
|
824 | fn test_build_single_relregex() { | |
825 | assert_eq!( |
|
825 | assert_eq!( | |
826 | build_single_regex( |
|
826 | build_single_regex( | |
827 | &IgnorePattern::new( |
|
827 | &IgnorePattern::new( | |
828 | PatternSyntax::RelRegexp, |
|
828 | PatternSyntax::RelRegexp, | |
829 | b"^ba{2}r", |
|
829 | b"^ba{2}r", | |
830 | Path::new("") |
|
830 | Path::new("") | |
831 | ), |
|
831 | ), | |
832 | b"(?:/|$)" |
|
832 | b"(?:/|$)" | |
833 | ) |
|
833 | ) | |
834 | .unwrap(), |
|
834 | .unwrap(), | |
835 | Some(b"^ba{2}r".to_vec()), |
|
835 | Some(b"^ba{2}r".to_vec()), | |
836 | ); |
|
836 | ); | |
837 | assert_eq!( |
|
837 | assert_eq!( | |
838 | build_single_regex( |
|
838 | build_single_regex( | |
839 | &IgnorePattern::new( |
|
839 | &IgnorePattern::new( | |
840 | PatternSyntax::RelRegexp, |
|
840 | PatternSyntax::RelRegexp, | |
841 | b"ba{2}r", |
|
841 | b"ba{2}r", | |
842 | Path::new("") |
|
842 | Path::new("") | |
843 | ), |
|
843 | ), | |
844 | b"(?:/|$)" |
|
844 | b"(?:/|$)" | |
845 | ) |
|
845 | ) | |
846 | .unwrap(), |
|
846 | .unwrap(), | |
847 | Some(b".*ba{2}r".to_vec()), |
|
847 | Some(b".*ba{2}r".to_vec()), | |
848 | ); |
|
848 | ); | |
849 | assert_eq!( |
|
849 | assert_eq!( | |
850 | build_single_regex( |
|
850 | build_single_regex( | |
851 | &IgnorePattern::new( |
|
851 | &IgnorePattern::new( | |
852 | PatternSyntax::RelRegexp, |
|
852 | PatternSyntax::RelRegexp, | |
853 | b"(?ia)ba{2}r", |
|
853 | b"(?ia)ba{2}r", | |
854 | Path::new("") |
|
854 | Path::new("") | |
855 | ), |
|
855 | ), | |
856 | b"(?:/|$)" |
|
856 | b"(?:/|$)" | |
857 | ) |
|
857 | ) | |
858 | .unwrap(), |
|
858 | .unwrap(), | |
859 | Some(b"(?ia:.*ba{2}r)".to_vec()), |
|
859 | Some(b"(?ia:.*ba{2}r)".to_vec()), | |
860 | ); |
|
860 | ); | |
861 | assert_eq!( |
|
861 | assert_eq!( | |
862 | build_single_regex( |
|
862 | build_single_regex( | |
863 | &IgnorePattern::new( |
|
863 | &IgnorePattern::new( | |
864 | PatternSyntax::RelRegexp, |
|
864 | PatternSyntax::RelRegexp, | |
865 | b"(?ia)^ba{2}r", |
|
865 | b"(?ia)^ba{2}r", | |
866 | Path::new("") |
|
866 | Path::new("") | |
867 | ), |
|
867 | ), | |
868 | b"(?:/|$)" |
|
868 | b"(?:/|$)" | |
869 | ) |
|
869 | ) | |
870 | .unwrap(), |
|
870 | .unwrap(), | |
871 | Some(b"(?ia:^ba{2}r)".to_vec()), |
|
871 | Some(b"(?ia:^ba{2}r)".to_vec()), | |
872 | ); |
|
872 | ); | |
873 | } |
|
873 | } | |
874 | } |
|
874 | } |
General Comments 0
You need to be logged in to leave comments.
Login now