]>
Commit | Line | Data |
---|---|---|
1 | #include "git-compat-util.h" | |
2 | #include "config.h" | |
3 | #include "userdiff.h" | |
4 | #include "attr.h" | |
5 | #include "strbuf.h" | |
6 | #include "environment.h" | |
7 | ||
8 | static struct userdiff_driver *drivers; | |
9 | static int ndrivers; | |
10 | static int drivers_alloc; | |
11 | ||
12 | #define PATTERNS(lang, rx, wrx) { \ | |
13 | .name = lang, \ | |
14 | .binary = -1, \ | |
15 | .funcname = { \ | |
16 | .pattern = rx, \ | |
17 | .cflags = REG_EXTENDED, \ | |
18 | }, \ | |
19 | .word_regex = wrx "|[^[:space:]]|[\xc0-\xff][\x80-\xbf]+", \ | |
20 | .word_regex_multi_byte = wrx "|[^[:space:]]", \ | |
21 | } | |
22 | #define IPATTERN(lang, rx, wrx) { \ | |
23 | .name = lang, \ | |
24 | .binary = -1, \ | |
25 | .funcname = { \ | |
26 | .pattern = rx, \ | |
27 | .cflags = REG_EXTENDED | REG_ICASE, \ | |
28 | }, \ | |
29 | .word_regex = wrx "|[^[:space:]]|[\xc0-\xff][\x80-\xbf]+", \ | |
30 | .word_regex_multi_byte = wrx "|[^[:space:]]", \ | |
31 | } | |
32 | ||
33 | /* | |
34 | * Built-in drivers for various languages, sorted by their names | |
35 | * (except that the "default" is left at the end). | |
36 | * | |
37 | * When writing or updating patterns, assume that the contents these | |
38 | * patterns are applied to are syntactically correct. The patterns | |
39 | * can be simple without implementing all syntactical corner cases, as | |
40 | * long as they are sufficiently permissive. | |
41 | */ | |
42 | static struct userdiff_driver builtin_drivers[] = { | |
43 | IPATTERN("ada", | |
44 | "!^(.*[ \t])?(is[ \t]+new|renames|is[ \t]+separate)([ \t].*)?$\n" | |
45 | "!^[ \t]*with[ \t].*$\n" | |
46 | "^[ \t]*((procedure|function)[ \t]+.*)$\n" | |
47 | "^[ \t]*((package|protected|task)[ \t]+.*)$", | |
48 | /* -- */ | |
49 | "[a-zA-Z][a-zA-Z0-9_]*" | |
50 | "|[-+]?[0-9][0-9#_.aAbBcCdDeEfF]*([eE][+-]?[0-9_]+)?" | |
51 | "|=>|\\.\\.|\\*\\*|:=|/=|>=|<=|<<|>>|<>"), | |
52 | PATTERNS("bash", | |
53 | /* Optional leading indentation */ | |
54 | "^[ \t]*" | |
55 | /* Start of captured text */ | |
56 | "(" | |
57 | "(" | |
58 | /* POSIX identifier with mandatory parentheses */ | |
59 | "[a-zA-Z_][a-zA-Z0-9_]*[ \t]*\\([ \t]*\\))" | |
60 | "|" | |
61 | /* Bashism identifier with optional parentheses */ | |
62 | "(function[ \t]+[a-zA-Z_][a-zA-Z0-9_]*(([ \t]*\\([ \t]*\\))|([ \t]+))" | |
63 | ")" | |
64 | /* Optional whitespace */ | |
65 | "[ \t]*" | |
66 | /* Compound command starting with `{`, `(`, `((` or `[[` */ | |
67 | "(\\{|\\(\\(?|\\[\\[)" | |
68 | /* End of captured text */ | |
69 | ")", | |
70 | /* -- */ | |
71 | /* Characters not in the default $IFS value */ | |
72 | "[^ \t]+"), | |
73 | PATTERNS("bibtex", | |
74 | "(@[a-zA-Z]{1,}[ \t]*\\{{0,1}[ \t]*[^ \t\"@',\\#}{~%]*).*$", | |
75 | /* -- */ | |
76 | "[={}\"]|[^={}\" \t]+"), | |
77 | PATTERNS("cpp", | |
78 | /* Jump targets or access declarations */ | |
79 | "!^[ \t]*[A-Za-z_][A-Za-z_0-9]*:[[:space:]]*($|/[/*])\n" | |
80 | /* functions/methods, variables, and compounds at top level */ | |
81 | "^((::[[:space:]]*)?[A-Za-z_].*)$", | |
82 | /* -- */ | |
83 | /* identifiers and keywords */ | |
84 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
85 | /* decimal and octal integers as well as floatingpoint numbers */ | |
86 | "|[0-9][0-9.]*([Ee][-+]?[0-9]+)?[fFlLuU]*" | |
87 | /* hexadecimal and binary integers */ | |
88 | "|0[xXbB][0-9a-fA-F]+[lLuU]*" | |
89 | /* floatingpoint numbers that begin with a decimal point */ | |
90 | "|\\.[0-9][0-9]*([Ee][-+]?[0-9]+)?[fFlL]?" | |
91 | "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*|<=>"), | |
92 | PATTERNS("csharp", | |
93 | /* | |
94 | * Jump over reserved keywords which are illegal method names, but which | |
95 | * can be followed by parentheses without special characters in between, | |
96 | * making them look like methods. | |
97 | */ | |
98 | "!(^|[ \t]+)" /* Start of line or whitespace. */ | |
99 | "(do|while|for|foreach|if|else|new|default|return|switch|case|throw" | |
100 | "|catch|using|lock|fixed)" | |
101 | "([ \t(]+|$)\n" /* Whitespace, "(", or end of line. */ | |
102 | /* | |
103 | * Methods/constructors: | |
104 | * The strategy is to identify a minimum of two groups (any combination | |
105 | * of keywords/type/name) before the opening parenthesis, and without | |
106 | * final unexpected characters, normally only used in ordinary statements. | |
107 | */ | |
108 | "^[ \t]*" /* Remove leading whitespace. */ | |
109 | "(" /* Start chunk header capture. */ | |
110 | "(" /* First group. */ | |
111 | "[][[:alnum:]@_.]" /* Name. */ | |
112 | "(<[][[:alnum:]@_, \t<>]+>)?" /* Optional generic parameters. */ | |
113 | ")+" | |
114 | "([ \t]+" /* Subsequent groups, prepended with space. */ | |
115 | "([][[:alnum:]@_.](<[][[:alnum:]@_, \t<>]+>)?)+" | |
116 | ")+" | |
117 | "[ \t]*" /* Optional space before parameters start. */ | |
118 | "\\(" /* Start of method parameters. */ | |
119 | "[^;]*" /* Allow complex parameters, but exclude statements (;). */ | |
120 | ")$\n" /* Close chunk header capture. */ | |
121 | /* | |
122 | * Properties: | |
123 | * As with methods, expect a minimum of two groups. But, more trivial than | |
124 | * methods, the vast majority of properties long enough to be worth | |
125 | * showing a chunk header for don't include "=:;,()" on the line they are | |
126 | * defined, since they don't have a parameter list. | |
127 | */ | |
128 | "^[ \t]*(" | |
129 | "([][[:alnum:]@_.](<[][[:alnum:]@_, \t<>]+>)?)+" | |
130 | "([ \t]+" | |
131 | "([][[:alnum:]@_.](<[][[:alnum:]@_, \t<>]+>)?)+" | |
132 | ")+" /* Up to here, same as methods regex. */ | |
133 | "[^;=:,()]*" /* Compared to methods, no parameter list allowed. */ | |
134 | ")$\n" | |
135 | /* Type definitions */ | |
136 | "^[ \t]*(((static|public|internal|private|protected|new|unsafe|sealed|abstract|partial)[ \t]+)*(class|enum|interface|struct|record)[ \t]+.*)$\n" | |
137 | /* Namespace */ | |
138 | "^[ \t]*(namespace[ \t]+.*)$", | |
139 | /* -- */ | |
140 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
141 | "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lL]?" | |
142 | "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->"), | |
143 | IPATTERN("css", | |
144 | "![:;][[:space:]]*$\n" | |
145 | "^[:[@.#]?[_a-z0-9].*$", | |
146 | /* -- */ | |
147 | /* | |
148 | * This regex comes from W3C CSS specs. Should theoretically also | |
149 | * allow ISO 10646 characters U+00A0 and higher, | |
150 | * but they are not handled in this regex. | |
151 | */ | |
152 | "-?[_a-zA-Z][-_a-zA-Z0-9]*" /* identifiers */ | |
153 | "|-?[0-9]+|\\#[0-9a-fA-F]+" /* numbers */ | |
154 | ), | |
155 | PATTERNS("dts", | |
156 | "!;\n" | |
157 | "!=\n" | |
158 | /* lines beginning with a word optionally preceded by '&' or the root */ | |
159 | "^[ \t]*((/[ \t]*\\{|&?[a-zA-Z_]).*)", | |
160 | /* -- */ | |
161 | /* Property names and math operators */ | |
162 | "[a-zA-Z0-9,._+?#-]+" | |
163 | "|[-+*/%&^|!~]|>>|<<|&&|\\|\\|"), | |
164 | PATTERNS("elixir", | |
165 | "^[ \t]*((def(macro|module|impl|protocol|p)?|test)[ \t].*)$", | |
166 | /* -- */ | |
167 | /* Atoms, names, and module attributes */ | |
168 | "[@:]?[a-zA-Z0-9@_?!]+" | |
169 | /* Numbers with specific base */ | |
170 | "|[-+]?0[xob][0-9a-fA-F]+" | |
171 | /* Numbers */ | |
172 | "|[-+]?[0-9][0-9_.]*([eE][-+]?[0-9_]+)?" | |
173 | /* Operators and atoms that represent them */ | |
174 | "|:?(\\+\\+|--|\\.\\.|~~~|<>|\\^\\^\\^|<?\\|>|<<<?|>?>>|<<?~|~>?>|<~>|<=|>=|===?|!==?|=~|&&&?|\\|\\|\\|?|=>|<-|\\\\\\\\|->)" | |
175 | /* Not real operators, but should be grouped */ | |
176 | "|:?%[A-Za-z0-9_.]\\{\\}?"), | |
177 | IPATTERN("fortran", | |
178 | /* Don't match comment lines */ | |
179 | "!^([C*]|[ \t]*!)\n" | |
180 | /* Don't match 'module procedure' lines */ | |
181 | "!^[ \t]*MODULE[ \t]+PROCEDURE[ \t]\n" | |
182 | /* Program, module, block data */ | |
183 | "^[ \t]*((END[ \t]+)?(PROGRAM|MODULE|BLOCK[ \t]+DATA" | |
184 | /* Subroutines and functions */ | |
185 | "|([^!'\" \t]+[ \t]+)*(SUBROUTINE|FUNCTION))[ \t]+[A-Z].*)$", | |
186 | /* -- */ | |
187 | "[a-zA-Z][a-zA-Z0-9_]*" | |
188 | "|\\.([Ee][Qq]|[Nn][Ee]|[Gg][TtEe]|[Ll][TtEe]|[Tt][Rr][Uu][Ee]|[Ff][Aa][Ll][Ss][Ee]|[Aa][Nn][Dd]|[Oo][Rr]|[Nn]?[Ee][Qq][Vv]|[Nn][Oo][Tt])\\." | |
189 | /* numbers and format statements like 2E14.4, or ES12.6, 9X. | |
190 | * Don't worry about format statements without leading digits since | |
191 | * they would have been matched above as a variable anyway. */ | |
192 | "|[-+]?[0-9.]+([AaIiDdEeFfLlTtXx][Ss]?[-+]?[0-9.]*)?(_[a-zA-Z0-9][a-zA-Z0-9_]*)?" | |
193 | "|//|\\*\\*|::|[/<>=]="), | |
194 | IPATTERN("fountain", | |
195 | "^((\\.[^.]|(int|ext|est|int\\.?/ext|i/e)[. ]).*)$", | |
196 | /* -- */ | |
197 | "[^ \t-]+"), | |
198 | PATTERNS("golang", | |
199 | /* Functions */ | |
200 | "^[ \t]*(func[ \t]*.*(\\{[ \t]*)?)\n" | |
201 | /* Structs and interfaces */ | |
202 | "^[ \t]*(type[ \t].*(struct|interface)[ \t]*(\\{[ \t]*)?)", | |
203 | /* -- */ | |
204 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
205 | "|[-+0-9.eE]+i?|0[xX]?[0-9a-fA-F]+i?" | |
206 | "|[-+*/<>%&^|=!:]=|--|\\+\\+|<<=?|>>=?|&\\^=?|&&|\\|\\||<-|\\.{3}"), | |
207 | PATTERNS("html", | |
208 | "^[ \t]*(<[Hh][1-6]([ \t].*)?>.*)$", | |
209 | /* -- */ | |
210 | "[^<>= \t]+"), | |
211 | PATTERNS("java", | |
212 | "!^[ \t]*(catch|do|for|if|instanceof|new|return|switch|throw|while)\n" | |
213 | /* Class, enum, interface, and record declarations */ | |
214 | "^[ \t]*(([a-z-]+[ \t]+)*(class|enum|interface|record)[ \t]+.*)$\n" | |
215 | /* Method definitions; note that constructor signatures are not */ | |
216 | /* matched because they are indistinguishable from method calls. */ | |
217 | "^[ \t]*(([A-Za-z_<>&][][?&<>.,A-Za-z_0-9]*[ \t]+)+[A-Za-z_][A-Za-z_0-9]*[ \t]*\\([^;]*)$", | |
218 | /* -- */ | |
219 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
220 | "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lL]?" | |
221 | "|[-+*/<>%&^|=!]=" | |
222 | "|--|\\+\\+|<<=?|>>>?=?|&&|\\|\\|"), | |
223 | PATTERNS("kotlin", | |
224 | "^[ \t]*(([a-z]+[ \t]+)*(fun|class|interface)[ \t]+.*)$", | |
225 | /* -- */ | |
226 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
227 | /* hexadecimal and binary numbers */ | |
228 | "|0[xXbB][0-9a-fA-F_]+[lLuU]*" | |
229 | /* integers and floats */ | |
230 | "|[0-9][0-9_]*([.][0-9_]*)?([Ee][-+]?[0-9]+)?[fFlLuU]*" | |
231 | /* floating point numbers beginning with decimal point */ | |
232 | "|[.][0-9][0-9_]*([Ee][-+]?[0-9]+)?[fFlLuU]?" | |
233 | /* unary and binary operators */ | |
234 | "|[-+*/<>%&^|=!]==?|--|\\+\\+|<<=|>>=|&&|\\|\\||->|\\.\\*|!!|[?:.][.:]"), | |
235 | PATTERNS("markdown", | |
236 | "^ {0,3}#{1,6}[ \t].*", | |
237 | /* -- */ | |
238 | "[^<>= \t]+"), | |
239 | PATTERNS("matlab", | |
240 | /* | |
241 | * Octave pattern is mostly the same as matlab, except that '%%%' and | |
242 | * '##' can also be used to begin code sections, in addition to '%%' | |
243 | * that is understood by both. | |
244 | */ | |
245 | "^[[:space:]]*((classdef|function)[[:space:]].*)$|^(%%%?|##)[[:space:]].*$", | |
246 | /* -- */ | |
247 | "[a-zA-Z_][a-zA-Z0-9_]*|[-+0-9.e]+|[=~<>]=|\\.[*/\\^']|\\|\\||&&"), | |
248 | PATTERNS("objc", | |
249 | /* Negate C statements that can look like functions */ | |
250 | "!^[ \t]*(do|for|if|else|return|switch|while)\n" | |
251 | /* Objective-C methods */ | |
252 | "^[ \t]*([-+][ \t]*\\([ \t]*[A-Za-z_][A-Za-z_0-9* \t]*\\)[ \t]*[A-Za-z_].*)$\n" | |
253 | /* C functions */ | |
254 | "^[ \t]*(([A-Za-z_][A-Za-z_0-9]*[ \t]+)+[A-Za-z_][A-Za-z_0-9]*[ \t]*\\([^;]*)$\n" | |
255 | /* Objective-C class/protocol definitions */ | |
256 | "^(@(implementation|interface|protocol)[ \t].*)$", | |
257 | /* -- */ | |
258 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
259 | "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lL]?" | |
260 | "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->"), | |
261 | PATTERNS("pascal", | |
262 | "^(((class[ \t]+)?(procedure|function)|constructor|destructor|interface" | |
263 | "|implementation|initialization|finalization)[ \t]*.*)$\n" | |
264 | "^(.*=[ \t]*(class|record).*)$", | |
265 | /* -- */ | |
266 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
267 | "|[-+0-9.e]+|0[xXbB]?[0-9a-fA-F]+" | |
268 | "|<>|<=|>=|:=|\\.\\."), | |
269 | PATTERNS("perl", | |
270 | "^package .*\n" | |
271 | "^sub [[:alnum:]_':]+[ \t]*" | |
272 | "(\\([^)]*\\)[ \t]*)?" /* prototype */ | |
273 | /* | |
274 | * Attributes. A regex can't count nested parentheses, | |
275 | * so just slurp up whatever we see, taking care not | |
276 | * to accept lines like "sub foo; # defined elsewhere". | |
277 | * | |
278 | * An attribute could contain a semicolon, but at that | |
279 | * point it seems reasonable enough to give up. | |
280 | */ | |
281 | "(:[^;#]*)?" | |
282 | "(\\{[ \t]*)?" /* brace can come here or on the next line */ | |
283 | "(#.*)?$\n" /* comment */ | |
284 | "^(BEGIN|END|INIT|CHECK|UNITCHECK|AUTOLOAD|DESTROY)[ \t]*" | |
285 | "(\\{[ \t]*)?" /* brace can come here or on the next line */ | |
286 | "(#.*)?$\n" | |
287 | "^=head[0-9] .*", /* POD */ | |
288 | /* -- */ | |
289 | "[[:alpha:]_'][[:alnum:]_']*" | |
290 | "|0[xb]?[0-9a-fA-F_]*" | |
291 | /* taking care not to interpret 3..5 as (3.)(.5) */ | |
292 | "|[0-9a-fA-F_]+(\\.[0-9a-fA-F_]+)?([eE][-+]?[0-9_]+)?" | |
293 | "|=>|-[rwxoRWXOezsfdlpSugkbctTBMAC>]|~~|::" | |
294 | "|&&=|\\|\\|=|//=|\\*\\*=" | |
295 | "|&&|\\|\\||//|\\+\\+|--|\\*\\*|\\.\\.\\.?" | |
296 | "|[-+*/%.^&<>=!|]=" | |
297 | "|=~|!~" | |
298 | "|<<|<>|<=>|>>"), | |
299 | PATTERNS("php", | |
300 | "^[\t ]*(((public|protected|private|static|abstract|final)[\t ]+)*function.*)$\n" | |
301 | "^[\t ]*((((final|abstract)[\t ]+)?class|enum|interface|trait).*)$", | |
302 | /* -- */ | |
303 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
304 | "|[-+0-9.e]+|0[xXbB]?[0-9a-fA-F]+" | |
305 | "|[-+*/<>%&^|=!.]=|--|\\+\\+|<<=?|>>=?|===|&&|\\|\\||::|->"), | |
306 | PATTERNS("python", | |
307 | "^[ \t]*((class|(async[ \t]+)?def)[ \t].*)$", | |
308 | /* -- */ | |
309 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
310 | "|[-+0-9.e]+[jJlL]?|0[xX]?[0-9a-fA-F]+[lL]?" | |
311 | "|[-+*/<>%&^|=!]=|//=?|<<=?|>>=?|\\*\\*=?"), | |
312 | /* -- */ | |
313 | PATTERNS("ruby", | |
314 | "^[ \t]*((class|module|def)[ \t].*)$", | |
315 | /* -- */ | |
316 | "(@|@@|\\$)?[a-zA-Z_][a-zA-Z0-9_]*" | |
317 | "|[-+0-9.e]+|0[xXbB]?[0-9a-fA-F]+|\\?(\\\\C-)?(\\\\M-)?." | |
318 | "|//=?|[-+*/<>%&^|=!]=|<<=?|>>=?|===|\\.{1,3}|::|[!=]~"), | |
319 | PATTERNS("rust", | |
320 | "^[\t ]*((pub(\\([^\\)]+\\))?[\t ]+)?((async|const|unsafe|extern([\t ]+\"[^\"]+\"))[\t ]+)?(struct|enum|union|mod|trait|fn|impl|macro_rules!)[< \t]+[^;]*)$", | |
321 | /* -- */ | |
322 | "[a-zA-Z_][a-zA-Z0-9_]*" | |
323 | "|[0-9][0-9_a-fA-Fiosuxz]*(\\.([0-9]*[eE][+-]?)?[0-9_fF]*)?" | |
324 | "|[-+*\\/<>%&^|=!:]=|<<=?|>>=?|&&|\\|\\||->|=>|\\.{2}=|\\.{3}|::"), | |
325 | PATTERNS("scheme", | |
326 | "^[\t ]*(\\(((define|def(struct|syntax|class|method|rules|record|proto|alias)?)[-*/ \t]|(library|module|struct|class)[*+ \t]).*)$", | |
327 | /* | |
328 | * R7RS valid identifiers include any sequence enclosed | |
329 | * within vertical lines having no backslashes | |
330 | */ | |
331 | "\\|([^\\\\]*)\\|" | |
332 | /* All other words should be delimited by spaces or parentheses */ | |
333 | "|([^][)(}{[ \t])+"), | |
334 | PATTERNS("tex", "^(\\\\((sub)*section|chapter|part)\\*{0,1}\\{.*)$", | |
335 | "\\\\[a-zA-Z@]+|\\\\.|([a-zA-Z0-9]|[^\x01-\x7f])+"), | |
336 | { "default", NULL, NULL, -1, { NULL, 0 } }, | |
337 | }; | |
338 | #undef PATTERNS | |
339 | #undef IPATTERN | |
340 | ||
341 | static struct userdiff_driver driver_true = { | |
342 | .name = "diff=true", | |
343 | .binary = 0, | |
344 | }; | |
345 | ||
346 | static struct userdiff_driver driver_false = { | |
347 | .name = "!diff", | |
348 | .binary = 1, | |
349 | }; | |
350 | ||
351 | struct find_by_namelen_data { | |
352 | const char *name; | |
353 | size_t len; | |
354 | struct userdiff_driver *driver; | |
355 | }; | |
356 | ||
357 | static int userdiff_find_by_namelen_cb(struct userdiff_driver *driver, | |
358 | enum userdiff_driver_type type UNUSED, | |
359 | void *priv) | |
360 | { | |
361 | struct find_by_namelen_data *cb_data = priv; | |
362 | ||
363 | if (!xstrncmpz(driver->name, cb_data->name, cb_data->len)) { | |
364 | cb_data->driver = driver; | |
365 | return 1; /* tell the caller to stop iterating */ | |
366 | } | |
367 | return 0; | |
368 | } | |
369 | ||
370 | static int regexec_supports_multi_byte_chars(void) | |
371 | { | |
372 | static const char not_space[] = "[^[:space:]]"; | |
373 | static const char utf8_multi_byte_char[] = "\xc2\xa3"; | |
374 | regex_t re; | |
375 | regmatch_t match; | |
376 | static int result = -1; | |
377 | ||
378 | if (result != -1) | |
379 | return result; | |
380 | if (regcomp(&re, not_space, REG_EXTENDED)) | |
381 | BUG("invalid regular expression: %s", not_space); | |
382 | result = !regexec(&re, utf8_multi_byte_char, 1, &match, 0) && | |
383 | match.rm_so == 0 && | |
384 | match.rm_eo == strlen(utf8_multi_byte_char); | |
385 | regfree(&re); | |
386 | return result; | |
387 | } | |
388 | ||
389 | static struct userdiff_driver *userdiff_find_by_namelen(const char *name, size_t len) | |
390 | { | |
391 | struct find_by_namelen_data udcbdata = { | |
392 | .name = name, | |
393 | .len = len, | |
394 | }; | |
395 | for_each_userdiff_driver(userdiff_find_by_namelen_cb, &udcbdata); | |
396 | return udcbdata.driver; | |
397 | } | |
398 | ||
399 | static int parse_funcname(struct userdiff_funcname *f, const char *k, | |
400 | const char *v, int cflags) | |
401 | { | |
402 | if (git_config_string(&f->pattern, k, v) < 0) | |
403 | return -1; | |
404 | f->cflags = cflags; | |
405 | return 0; | |
406 | } | |
407 | ||
408 | static int parse_tristate(int *b, const char *k, const char *v) | |
409 | { | |
410 | if (v && !strcasecmp(v, "auto")) | |
411 | *b = -1; | |
412 | else | |
413 | *b = git_config_bool(k, v); | |
414 | return 0; | |
415 | } | |
416 | ||
417 | static int parse_bool(int *b, const char *k, const char *v) | |
418 | { | |
419 | *b = git_config_bool(k, v); | |
420 | return 0; | |
421 | } | |
422 | ||
423 | int userdiff_config(const char *k, const char *v) | |
424 | { | |
425 | struct userdiff_driver *drv; | |
426 | const char *name, *type; | |
427 | size_t namelen; | |
428 | ||
429 | if (parse_config_key(k, "diff", &name, &namelen, &type) || !name) | |
430 | return 0; | |
431 | ||
432 | drv = userdiff_find_by_namelen(name, namelen); | |
433 | if (!drv) { | |
434 | ALLOC_GROW(drivers, ndrivers+1, drivers_alloc); | |
435 | drv = &drivers[ndrivers++]; | |
436 | memset(drv, 0, sizeof(*drv)); | |
437 | drv->name = xmemdupz(name, namelen); | |
438 | drv->binary = -1; | |
439 | } | |
440 | ||
441 | if (!strcmp(type, "funcname")) | |
442 | return parse_funcname(&drv->funcname, k, v, 0); | |
443 | if (!strcmp(type, "xfuncname")) | |
444 | return parse_funcname(&drv->funcname, k, v, REG_EXTENDED); | |
445 | if (!strcmp(type, "binary")) | |
446 | return parse_tristate(&drv->binary, k, v); | |
447 | if (!strcmp(type, "command")) | |
448 | return git_config_string(&drv->external, k, v); | |
449 | if (!strcmp(type, "textconv")) | |
450 | return git_config_string(&drv->textconv, k, v); | |
451 | if (!strcmp(type, "cachetextconv")) | |
452 | return parse_bool(&drv->textconv_want_cache, k, v); | |
453 | if (!strcmp(type, "wordregex")) | |
454 | return git_config_string(&drv->word_regex, k, v); | |
455 | if (!strcmp(type, "algorithm")) | |
456 | return git_config_string(&drv->algorithm, k, v); | |
457 | ||
458 | return 0; | |
459 | } | |
460 | ||
461 | struct userdiff_driver *userdiff_find_by_name(const char *name) | |
462 | { | |
463 | int len = strlen(name); | |
464 | struct userdiff_driver *driver = userdiff_find_by_namelen(name, len); | |
465 | if (driver && driver->word_regex_multi_byte) { | |
466 | if (regexec_supports_multi_byte_chars()) | |
467 | driver->word_regex = driver->word_regex_multi_byte; | |
468 | driver->word_regex_multi_byte = NULL; | |
469 | } | |
470 | return driver; | |
471 | } | |
472 | ||
473 | struct userdiff_driver *userdiff_find_by_path(struct index_state *istate, | |
474 | const char *path) | |
475 | { | |
476 | static struct attr_check *check; | |
477 | ||
478 | if (!check) | |
479 | check = attr_check_initl("diff", NULL); | |
480 | if (!path) | |
481 | return NULL; | |
482 | git_check_attr(istate, path, check); | |
483 | ||
484 | if (ATTR_TRUE(check->items[0].value)) | |
485 | return &driver_true; | |
486 | if (ATTR_FALSE(check->items[0].value)) | |
487 | return &driver_false; | |
488 | if (ATTR_UNSET(check->items[0].value)) | |
489 | return NULL; | |
490 | return userdiff_find_by_name(check->items[0].value); | |
491 | } | |
492 | ||
493 | struct userdiff_driver *userdiff_get_textconv(struct repository *r, | |
494 | struct userdiff_driver *driver) | |
495 | { | |
496 | if (!driver->textconv) | |
497 | return NULL; | |
498 | ||
499 | if (driver->textconv_want_cache && !driver->textconv_cache && | |
500 | have_git_dir()) { | |
501 | struct notes_cache *c = xmalloc(sizeof(*c)); | |
502 | struct strbuf name = STRBUF_INIT; | |
503 | ||
504 | strbuf_addf(&name, "textconv/%s", driver->name); | |
505 | notes_cache_init(r, c, name.buf, driver->textconv); | |
506 | driver->textconv_cache = c; | |
507 | strbuf_release(&name); | |
508 | } | |
509 | ||
510 | return driver; | |
511 | } | |
512 | ||
513 | static int for_each_userdiff_driver_list(each_userdiff_driver_fn fn, | |
514 | enum userdiff_driver_type type, void *cb_data, | |
515 | struct userdiff_driver *drv, | |
516 | int drv_size) | |
517 | { | |
518 | int i; | |
519 | int ret; | |
520 | for (i = 0; i < drv_size; i++) { | |
521 | struct userdiff_driver *item = drv + i; | |
522 | if ((ret = fn(item, type, cb_data))) | |
523 | return ret; | |
524 | } | |
525 | return 0; | |
526 | } | |
527 | ||
528 | int for_each_userdiff_driver(each_userdiff_driver_fn fn, void *cb_data) | |
529 | { | |
530 | int ret; | |
531 | ||
532 | ret = for_each_userdiff_driver_list(fn, USERDIFF_DRIVER_TYPE_CUSTOM, | |
533 | cb_data, drivers, ndrivers); | |
534 | if (ret) | |
535 | return ret; | |
536 | ||
537 | ret = for_each_userdiff_driver_list(fn, USERDIFF_DRIVER_TYPE_BUILTIN, | |
538 | cb_data, builtin_drivers, | |
539 | ARRAY_SIZE(builtin_drivers)); | |
540 | if (ret) | |
541 | return ret; | |
542 | ||
543 | return 0; | |
544 | } |