userdiff.c

   1 #include "git-compat-util.h"
   2 #include "alloc.h"
   3 #include "config.h"
   4 #include "userdiff.h"
   5 #include "attr.h"
   6 #include "strbuf.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          /* Keywords */
  94          "!^[ \t]*(do|while|for|if|else|instanceof|new|return|switch|case|throw|catch|using)\n"
  95          /* Methods and constructors */
  96          "^[ \t]*(((static|public|internal|private|protected|new|virtual|sealed|override|unsafe|async)[ \t]+)*[][<>@.~_[:alnum:]]+[ \t]+[<>@._[:alnum:]]+[ \t]*\\(.*\\))[ \t]*$\n"
  97          /* Properties */
  98          "^[ \t]*(((static|public|internal|private|protected|new|virtual|sealed|override|unsafe)[ \t]+)*[][<>@.~_[:alnum:]]+[ \t]+[@._[:alnum:]]+)[ \t]*$\n"
  99          /* Type definitions */
 100          "^[ \t]*(((static|public|internal|private|protected|new|unsafe|sealed|abstract|partial)[ \t]+)*(class|enum|interface|struct|record)[ \t]+.*)$\n"
 101          /* Namespace */
 102          "^[ \t]*(namespace[ \t]+.*)$",
 103          /* -- */
 104          "[a-zA-Z_][a-zA-Z0-9_]*"
 105          "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lL]?"
 106          "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->"),
 107 IPATTERN("css",
 108          "![:;][[:space:]]*$\n"
 109          "^[:[@.#]?[_a-z0-9].*$",
 110          /* -- */
 111          /*
 112           * This regex comes from W3C CSS specs. Should theoretically also
 113           * allow ISO 10646 characters U+00A0 and higher,
 114           * but they are not handled in this regex.
 115           */
 116          "-?[_a-zA-Z][-_a-zA-Z0-9]*" /* identifiers */
 117          "|-?[0-9]+|\\#[0-9a-fA-F]+" /* numbers */
 118 ),
 119 PATTERNS("dts",
 120          "!;\n"
 121          "!=\n"
 122          /* lines beginning with a word optionally preceded by '&' or the root */
 123          "^[ \t]*((/[ \t]*\\{|&?[a-zA-Z_]).*)",
 124          /* -- */
 125          /* Property names and math operators */
 126          "[a-zA-Z0-9,._+?#-]+"
 127          "|[-+*/%&^|!~]|>>|<<|&&|\\|\\|"),
 128 PATTERNS("elixir",
 129          "^[ \t]*((def(macro|module|impl|protocol|p)?|test)[ \t].*)$",
 130          /* -- */
 131          /* Atoms, names, and module attributes */
 132          "[@:]?[a-zA-Z0-9@_?!]+"
 133          /* Numbers with specific base */
 134          "|[-+]?0[xob][0-9a-fA-F]+"
 135          /* Numbers */
 136          "|[-+]?[0-9][0-9_.]*([eE][-+]?[0-9_]+)?"
 137          /* Operators and atoms that represent them */
 138          "|:?(\\+\\+|--|\\.\\.|~~~|<>|\\^\\^\\^|<?\\|>|<<<?|>?>>|<<?~|~>?>|<~>|<=|>=|===?|!==?|=~|&&&?|\\|\\|\\|?|=>|<-|\\\\\\\\|->)"
 139          /* Not real operators, but should be grouped */
 140          "|:?%[A-Za-z0-9_.]\\{\\}?"),
 141 IPATTERN("fortran",
 142          /* Don't match comment lines */
 143          "!^([C*]|[ \t]*!)\n"
 144          /* Don't match 'module procedure' lines */
 145          "!^[ \t]*MODULE[ \t]+PROCEDURE[ \t]\n"
 146          /* Program, module, block data */
 147          "^[ \t]*((END[ \t]+)?(PROGRAM|MODULE|BLOCK[ \t]+DATA"
 148                 /* Subroutines and functions */
 149                 "|([^!'\" \t]+[ \t]+)*(SUBROUTINE|FUNCTION))[ \t]+[A-Z].*)$",
 150          /* -- */
 151          "[a-zA-Z][a-zA-Z0-9_]*"
 152          "|\\.([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])\\."
 153          /* numbers and format statements like 2E14.4, or ES12.6, 9X.
 154           * Don't worry about format statements without leading digits since
 155           * they would have been matched above as a variable anyway. */
 156          "|[-+]?[0-9.]+([AaIiDdEeFfLlTtXx][Ss]?[-+]?[0-9.]*)?(_[a-zA-Z0-9][a-zA-Z0-9_]*)?"
 157          "|//|\\*\\*|::|[/<>=]="),
 158 IPATTERN("fountain",
 159          "^((\\.[^.]|(int|ext|est|int\\.?/ext|i/e)[. ]).*)$",
 160          /* -- */
 161          "[^ \t-]+"),
 162 PATTERNS("golang",
 163          /* Functions */
 164          "^[ \t]*(func[ \t]*.*(\\{[ \t]*)?)\n"
 165          /* Structs and interfaces */
 166          "^[ \t]*(type[ \t].*(struct|interface)[ \t]*(\\{[ \t]*)?)",
 167          /* -- */
 168          "[a-zA-Z_][a-zA-Z0-9_]*"
 169          "|[-+0-9.eE]+i?|0[xX]?[0-9a-fA-F]+i?"
 170          "|[-+*/<>%&^|=!:]=|--|\\+\\+|<<=?|>>=?|&\\^=?|&&|\\|\\||<-|\\.{3}"),
 171 PATTERNS("html",
 172          "^[ \t]*(<[Hh][1-6]([ \t].*)?>.*)$",
 173          /* -- */
 174          "[^<>= \t]+"),
 175 PATTERNS("java",
 176          "!^[ \t]*(catch|do|for|if|instanceof|new|return|switch|throw|while)\n"
 177          /* Class, enum, interface, and record declarations */
 178          "^[ \t]*(([a-z-]+[ \t]+)*(class|enum|interface|record)[ \t]+.*)$\n"
 179          /* Method definitions; note that constructor signatures are not */
 180          /* matched because they are indistinguishable from method calls. */
 181          "^[ \t]*(([A-Za-z_<>&][][?&<>.,A-Za-z_0-9]*[ \t]+)+[A-Za-z_][A-Za-z_0-9]*[ \t]*\\([^;]*)$",
 182          /* -- */
 183          "[a-zA-Z_][a-zA-Z0-9_]*"
 184          "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lL]?"
 185          "|[-+*/<>%&^|=!]="
 186          "|--|\\+\\+|<<=?|>>>?=?|&&|\\|\\|"),
 187 PATTERNS("kotlin",
 188          "^[ \t]*(([a-z]+[ \t]+)*(fun|class|interface)[ \t]+.*)$",
 189          /* -- */
 190          "[a-zA-Z_][a-zA-Z0-9_]*"
 191          /* hexadecimal and binary numbers */
 192          "|0[xXbB][0-9a-fA-F_]+[lLuU]*"
 193          /* integers and floats */
 194          "|[0-9][0-9_]*([.][0-9_]*)?([Ee][-+]?[0-9]+)?[fFlLuU]*"
 195          /* floating point numbers beginning with decimal point */
 196          "|[.][0-9][0-9_]*([Ee][-+]?[0-9]+)?[fFlLuU]?"
 197          /* unary and binary operators */
 198          "|[-+*/<>%&^|=!]==?|--|\\+\\+|<<=|>>=|&&|\\|\\||->|\\.\\*|!!|[?:.][.:]"),
 199 PATTERNS("markdown",
 200          "^ {0,3}#{1,6}[ \t].*",
 201          /* -- */
 202          "[^<>= \t]+"),
 203 PATTERNS("matlab",
 204          /*
 205           * Octave pattern is mostly the same as matlab, except that '%%%' and
 206           * '##' can also be used to begin code sections, in addition to '%%'
 207           * that is understood by both.
 208           */
 209          "^[[:space:]]*((classdef|function)[[:space:]].*)$|^(%%%?|##)[[:space:]].*$",
 210          /* -- */
 211          "[a-zA-Z_][a-zA-Z0-9_]*|[-+0-9.e]+|[=~<>]=|\\.[*/\\^']|\\|\\||&&"),
 212 PATTERNS("objc",
 213          /* Negate C statements that can look like functions */
 214          "!^[ \t]*(do|for|if|else|return|switch|while)\n"
 215          /* Objective-C methods */
 216          "^[ \t]*([-+][ \t]*\\([ \t]*[A-Za-z_][A-Za-z_0-9* \t]*\\)[ \t]*[A-Za-z_].*)$\n"
 217          /* C functions */
 218          "^[ \t]*(([A-Za-z_][A-Za-z_0-9]*[ \t]+)+[A-Za-z_][A-Za-z_0-9]*[ \t]*\\([^;]*)$\n"
 219          /* Objective-C class/protocol definitions */
 220          "^(@(implementation|interface|protocol)[ \t].*)$",
 221          /* -- */
 222          "[a-zA-Z_][a-zA-Z0-9_]*"
 223          "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lL]?"
 224          "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->"),
 225 PATTERNS("pascal",
 226          "^(((class[ \t]+)?(procedure|function)|constructor|destructor|interface"
 227          "|implementation|initialization|finalization)[ \t]*.*)$\n"
 228          "^(.*=[ \t]*(class|record).*)$",
 229          /* -- */
 230          "[a-zA-Z_][a-zA-Z0-9_]*"
 231          "|[-+0-9.e]+|0[xXbB]?[0-9a-fA-F]+"
 232          "|<>|<=|>=|:=|\\.\\."),
 233 PATTERNS("perl",
 234          "^package .*\n"
 235          "^sub [[:alnum:]_':]+[ \t]*"
 236                 "(\\([^)]*\\)[ \t]*)?" /* prototype */
 237                 /*
 238                  * Attributes.  A regex can't count nested parentheses,
 239                  * so just slurp up whatever we see, taking care not
 240                  * to accept lines like "sub foo; # defined elsewhere".
 241                  *
 242                  * An attribute could contain a semicolon, but at that
 243                  * point it seems reasonable enough to give up.
 244                  */
 245                 "(:[^;#]*)?"
 246                 "(\\{[ \t]*)?" /* brace can come here or on the next line */
 247                 "(#.*)?$\n" /* comment */
 248          "^(BEGIN|END|INIT|CHECK|UNITCHECK|AUTOLOAD|DESTROY)[ \t]*"
 249                 "(\\{[ \t]*)?" /* brace can come here or on the next line */
 250                 "(#.*)?$\n"
 251          "^=head[0-9] .*",      /* POD */
 252          /* -- */
 253          "[[:alpha:]_'][[:alnum:]_']*"
 254          "|0[xb]?[0-9a-fA-F_]*"
 255          /* taking care not to interpret 3..5 as (3.)(.5) */
 256          "|[0-9a-fA-F_]+(\\.[0-9a-fA-F_]+)?([eE][-+]?[0-9_]+)?"
 257          "|=>|-[rwxoRWXOezsfdlpSugkbctTBMAC>]|~~|::"
 258          "|&&=|\\|\\|=|//=|\\*\\*="
 259          "|&&|\\|\\||//|\\+\\+|--|\\*\\*|\\.\\.\\.?"
 260          "|[-+*/%.^&<>=!|]="
 261          "|=~|!~"
 262          "|<<|<>|<=>|>>"),
 263 PATTERNS("php",
 264          "^[\t ]*(((public|protected|private|static|abstract|final)[\t ]+)*function.*)$\n"
 265          "^[\t ]*((((final|abstract)[\t ]+)?class|enum|interface|trait).*)$",
 266          /* -- */
 267          "[a-zA-Z_][a-zA-Z0-9_]*"
 268          "|[-+0-9.e]+|0[xXbB]?[0-9a-fA-F]+"
 269          "|[-+*/<>%&^|=!.]=|--|\\+\\+|<<=?|>>=?|===|&&|\\|\\||::|->"),
 270 PATTERNS("python",
 271          "^[ \t]*((class|(async[ \t]+)?def)[ \t].*)$",
 272          /* -- */
 273          "[a-zA-Z_][a-zA-Z0-9_]*"
 274          "|[-+0-9.e]+[jJlL]?|0[xX]?[0-9a-fA-F]+[lL]?"
 275          "|[-+*/<>%&^|=!]=|//=?|<<=?|>>=?|\\*\\*=?"),
 276          /* -- */
 277 PATTERNS("ruby",
 278          "^[ \t]*((class|module|def)[ \t].*)$",
 279          /* -- */
 280          "(@|@@|\\$)?[a-zA-Z_][a-zA-Z0-9_]*"
 281          "|[-+0-9.e]+|0[xXbB]?[0-9a-fA-F]+|\\?(\\\\C-)?(\\\\M-)?."
 282          "|//=?|[-+*/<>%&^|=!]=|<<=?|>>=?|===|\\.{1,3}|::|[!=]~"),
 283 PATTERNS("rust",
 284          "^[\t ]*((pub(\\([^\\)]+\\))?[\t ]+)?((async|const|unsafe|extern([\t ]+\"[^\"]+\"))[\t ]+)?(struct|enum|union|mod|trait|fn|impl|macro_rules!)[< \t]+[^;]*)$",
 285          /* -- */
 286          "[a-zA-Z_][a-zA-Z0-9_]*"
 287          "|[0-9][0-9_a-fA-Fiosuxz]*(\\.([0-9]*[eE][+-]?)?[0-9_fF]*)?"
 288          "|[-+*\\/<>%&^|=!:]=|<<=?|>>=?|&&|\\|\\||->|=>|\\.{2}=|\\.{3}|::"),
 289 PATTERNS("scheme",
 290          "^[\t ]*(\\(((define|def(struct|syntax|class|method|rules|record|proto|alias)?)[-*/ \t]|(library|module|struct|class)[*+ \t]).*)$",
 291          /*
 292           * R7RS valid identifiers include any sequence enclosed
 293           * within vertical lines having no backslashes
 294           */
 295          "\\|([^\\\\]*)\\|"
 296          /* All other words should be delimited by spaces or parentheses */
 297          "|([^][)(}{[ \t])+"),
 298 PATTERNS("tex", "^(\\\\((sub)*section|chapter|part)\\*{0,1}\\{.*)$",
 299          "\\\\[a-zA-Z@]+|\\\\.|([a-zA-Z0-9]|[^\x01-\x7f])+"),
 300 { "default", NULL, NULL, -1, { NULL, 0 } },
 301 };
 302 #undef PATTERNS
 303 #undef IPATTERN
 304
 305 static struct userdiff_driver driver_true = {
 306         .name = "diff=true",
 307         .binary = 0,
 308 };
 309
 310 static struct userdiff_driver driver_false = {
 311         .name = "!diff",
 312         .binary = 1,
 313 };
 314
 315 struct find_by_namelen_data {
 316         const char *name;
 317         size_t len;
 318         struct userdiff_driver *driver;
 319 };
 320
 321 static int userdiff_find_by_namelen_cb(struct userdiff_driver *driver,
 322                                        enum userdiff_driver_type type UNUSED,
 323                                        void *priv)
 324 {
 325         struct find_by_namelen_data *cb_data = priv;
 326
 327         if (!strncmp(driver->name, cb_data->name, cb_data->len) &&
 328             !driver->name[cb_data->len]) {
 329                 cb_data->driver = driver;
 330                 return 1; /* tell the caller to stop iterating */
 331         }
 332         return 0;
 333 }
 334
 335 static int regexec_supports_multi_byte_chars(void)
 336 {
 337         static const char not_space[] = "[^[:space:]]";
 338         static const char utf8_multi_byte_char[] = "\xc2\xa3";
 339         regex_t re;
 340         regmatch_t match;
 341         static int result = -1;
 342
 343         if (result != -1)
 344                 return result;
 345         if (regcomp(&re, not_space, REG_EXTENDED))
 346                 BUG("invalid regular expression: %s", not_space);
 347         result = !regexec(&re, utf8_multi_byte_char, 1, &match, 0) &&
 348                 match.rm_so == 0 &&
 349                 match.rm_eo == strlen(utf8_multi_byte_char);
 350         regfree(&re);
 351         return result;
 352 }
 353
 354 static struct userdiff_driver *userdiff_find_by_namelen(const char *name, size_t len)
 355 {
 356         struct find_by_namelen_data udcbdata = {
 357                 .name = name,
 358                 .len = len,
 359         };
 360         for_each_userdiff_driver(userdiff_find_by_namelen_cb, &udcbdata);
 361         return udcbdata.driver;
 362 }
 363
 364 static int parse_funcname(struct userdiff_funcname *f, const char *k,
 365                 const char *v, int cflags)
 366 {
 367         if (git_config_string(&f->pattern, k, v) < 0)
 368                 return -1;
 369         f->cflags = cflags;
 370         return 0;
 371 }
 372
 373 static int parse_tristate(int *b, const char *k, const char *v)
 374 {
 375         if (v && !strcasecmp(v, "auto"))
 376                 *b = -1;
 377         else
 378                 *b = git_config_bool(k, v);
 379         return 0;
 380 }
 381
 382 static int parse_bool(int *b, const char *k, const char *v)
 383 {
 384         *b = git_config_bool(k, v);
 385         return 0;
 386 }
 387
 388 int userdiff_config(const char *k, const char *v)
 389 {
 390         struct userdiff_driver *drv;
 391         const char *name, *type;
 392         size_t namelen;
 393
 394         if (parse_config_key(k, "diff", &name, &namelen, &type) || !name)
 395                 return 0;
 396
 397         drv = userdiff_find_by_namelen(name, namelen);
 398         if (!drv) {
 399                 ALLOC_GROW(drivers, ndrivers+1, drivers_alloc);
 400                 drv = &drivers[ndrivers++];
 401                 memset(drv, 0, sizeof(*drv));
 402                 drv->name = xmemdupz(name, namelen);
 403                 drv->binary = -1;
 404         }
 405
 406         if (!strcmp(type, "funcname"))
 407                 return parse_funcname(&drv->funcname, k, v, 0);
 408         if (!strcmp(type, "xfuncname"))
 409                 return parse_funcname(&drv->funcname, k, v, REG_EXTENDED);
 410         if (!strcmp(type, "binary"))
 411                 return parse_tristate(&drv->binary, k, v);
 412         if (!strcmp(type, "command"))
 413                 return git_config_string(&drv->external, k, v);
 414         if (!strcmp(type, "textconv"))
 415                 return git_config_string(&drv->textconv, k, v);
 416         if (!strcmp(type, "cachetextconv"))
 417                 return parse_bool(&drv->textconv_want_cache, k, v);
 418         if (!strcmp(type, "wordregex"))
 419                 return git_config_string(&drv->word_regex, k, v);
 420         if (!strcmp(type, "algorithm"))
 421                 return git_config_string(&drv->algorithm, k, v);
 422
 423         return 0;
 424 }
 425
 426 struct userdiff_driver *userdiff_find_by_name(const char *name)
 427 {
 428         int len = strlen(name);
 429         struct userdiff_driver *driver = userdiff_find_by_namelen(name, len);
 430         if (driver && driver->word_regex_multi_byte) {
 431                 if (regexec_supports_multi_byte_chars())
 432                         driver->word_regex = driver->word_regex_multi_byte;
 433                 driver->word_regex_multi_byte = NULL;
 434         }
 435         return driver;
 436 }
 437
 438 struct userdiff_driver *userdiff_find_by_path(struct index_state *istate,
 439                                               const char *path)
 440 {
 441         static struct attr_check *check;
 442
 443         if (!check)
 444                 check = attr_check_initl("diff", NULL);
 445         if (!path)
 446                 return NULL;
 447         git_check_attr(istate, path, check);
 448
 449         if (ATTR_TRUE(check->items[0].value))
 450                 return &driver_true;
 451         if (ATTR_FALSE(check->items[0].value))
 452                 return &driver_false;
 453         if (ATTR_UNSET(check->items[0].value))
 454                 return NULL;
 455         return userdiff_find_by_name(check->items[0].value);
 456 }
 457
 458 struct userdiff_driver *userdiff_get_textconv(struct repository *r,
 459                                               struct userdiff_driver *driver)
 460 {
 461         if (!driver->textconv)
 462                 return NULL;
 463
 464         if (driver->textconv_want_cache && !driver->textconv_cache) {
 465                 struct notes_cache *c = xmalloc(sizeof(*c));
 466                 struct strbuf name = STRBUF_INIT;
 467
 468                 strbuf_addf(&name, "textconv/%s", driver->name);
 469                 notes_cache_init(r, c, name.buf, driver->textconv);
 470                 driver->textconv_cache = c;
 471                 strbuf_release(&name);
 472         }
 473
 474         return driver;
 475 }
 476
 477 static int for_each_userdiff_driver_list(each_userdiff_driver_fn fn,
 478                                          enum userdiff_driver_type type, void *cb_data,
 479                                          struct userdiff_driver *drv,
 480                                          int drv_size)
 481 {
 482         int i;
 483         int ret;
 484         for (i = 0; i < drv_size; i++) {
 485                 struct userdiff_driver *item = drv + i;
 486                 if ((ret = fn(item, type, cb_data)))
 487                         return ret;
 488         }
 489         return 0;
 490 }
 491
 492 int for_each_userdiff_driver(each_userdiff_driver_fn fn, void *cb_data)
 493 {
 494         int ret;
 495
 496         ret = for_each_userdiff_driver_list(fn, USERDIFF_DRIVER_TYPE_CUSTOM,
 497                                             cb_data, drivers, ndrivers);
 498         if (ret)
 499                 return ret;
 500
 501         ret = for_each_userdiff_driver_list(fn, USERDIFF_DRIVER_TYPE_BUILTIN,
 502                                             cb_data, builtin_drivers,
 503                                             ARRAY_SIZE(builtin_drivers));
 504         if (ret)
 505                 return ret;
 506
 507         return 0;
 508 }