src/libgit2/userdiff.h

   1 /*
   2  * Copyright (C) the libgit2 contributors. All rights reserved.
   3  *
   4  * This file is part of libgit2, distributed under the GNU GPL v2 with
   5  * a Linking Exception. For full terms see the included COPYING file.
   6  */
   7 #ifndef INCLUDE_userdiff_h__
   8 #define INCLUDE_userdiff_h__
   9
  10 #include "regexp.h"
  11
  12 /*
  13  * This file isolates the built in diff driver function name patterns.
  14  * Most of these patterns are taken from Git (with permission from the
  15  * original authors for relicensing to libgit2).
  16  */
  17
  18 typedef struct {
  19         const char *name;
  20         const char *fns;
  21         const char *words;
  22         int flags;
  23 } git_diff_driver_definition;
  24
  25 #define WORD_DEFAULT "|[^[:space:]]|[\xc0-\xff][\x80-\xbf]+"
  26
  27 /*
  28  * These builtin driver definition macros have same signature as in core
  29  * git userdiff.c so that the data can be extracted verbatim
  30  */
  31 #define PATTERNS(NAME, FN_PATS, WORD_PAT) \
  32         { NAME, FN_PATS, WORD_PAT WORD_DEFAULT, 0 }
  33 #define IPATTERN(NAME, FN_PATS, WORD_PAT) \
  34         { NAME, FN_PATS, WORD_PAT WORD_DEFAULT, GIT_REGEXP_ICASE }
  35
  36 /*
  37  * The table of diff driver patterns
  38  *
  39  * Function name patterns are a list of newline separated patterns that
  40  * match a function declaration (i.e. the line you want in the hunk header),
  41  * or a negative pattern prefixed with a '!' to reject a pattern (such as
  42  * rejecting goto labels in C code).
  43  *
  44  * Word boundary patterns are just a simple pattern that will be OR'ed with
  45  * the default value above (i.e. whitespace or non-ASCII characters).
  46  */
  47 static git_diff_driver_definition builtin_defs[] = {
  48
  49 IPATTERN("ada",
  50          "!^(.*[ \t])?(is[ \t]+new|renames|is[ \t]+separate)([ \t].*)?$\n"
  51          "!^[ \t]*with[ \t].*$\n"
  52          "^[ \t]*((procedure|function)[ \t]+.*)$\n"
  53          "^[ \t]*((package|protected|task)[ \t]+.*)$",
  54          /* -- */
  55          "[a-zA-Z][a-zA-Z0-9_]*"
  56          "|[-+]?[0-9][0-9#_.aAbBcCdDeEfF]*([eE][+-]?[0-9_]+)?"
  57          "|=>|\\.\\.|\\*\\*|:=|/=|>=|<=|<<|>>|<>"),
  58
  59 IPATTERN("fortran",
  60          "!^([C*]|[ \t]*!)\n"
  61          "!^[ \t]*MODULE[ \t]+PROCEDURE[ \t]\n"
  62          "^[ \t]*((END[ \t]+)?(PROGRAM|MODULE|BLOCK[ \t]+DATA"
  63                 "|([^'\" \t]+[ \t]+)*(SUBROUTINE|FUNCTION))[ \t]+[A-Z].*)$",
  64          /* -- */
  65          "[a-zA-Z][a-zA-Z0-9_]*"
  66          "|\\.([Ee][Qq]|[Nn][Ee]|[Gg][TtEe]|[Ll][TtEe]|[Tt][Rr][Uu][Ee]|[Ff][Aa][Ll][Ss][Ee]|[Aa][Nn][Dd]|[Oo][Rr]|[Nn]?[Ee][Qq][Vv]|[Nn][Oo][Tt])\\."
  67          /* numbers and format statements like 2E14.4, or ES12.6, 9X.
  68           * Don't worry about format statements without leading digits since
  69           * they would have been matched above as a variable anyway. */
  70          "|[-+]?[0-9.]+([AaIiDdEeFfLlTtXx][Ss]?[-+]?[0-9.]*)?(_[a-zA-Z0-9][a-zA-Z0-9_]*)?"
  71          "|//|\\*\\*|::|[/<>=]="),
  72
  73 PATTERNS("html", "^[ \t]*(<[Hh][1-6][ \t].*>.*)$",
  74          "[^<>= \t]+"),
  75
  76 PATTERNS("java",
  77          "!^[ \t]*(catch|do|for|if|instanceof|new|return|switch|throw|while)\n"
  78          "^[ \t]*(([A-Za-z_][A-Za-z_0-9]*[ \t]+)+[A-Za-z_][A-Za-z_0-9]*[ \t]*\\([^;]*)$",
  79          /* -- */
  80          "[a-zA-Z_][a-zA-Z0-9_]*"
  81          "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lL]?"
  82          "|[-+*/<>%&^|=!]="
  83          "|--|\\+\\+|<<=?|>>>?=?|&&|\\|\\|"),
  84
  85 PATTERNS("matlab",
  86          "^[[:space:]]*((classdef|function)[[:space:]].*)$|^%%[[:space:]].*$",
  87          "[a-zA-Z_][a-zA-Z0-9_]*|[-+0-9.e]+|[=~<>]=|\\.[*/\\^']|\\|\\||&&"),
  88
  89 PATTERNS("objc",
  90          /* Negate C statements that can look like functions */
  91          "!^[ \t]*(do|for|if|else|return|switch|while)\n"
  92          /* Objective-C methods */
  93          "^[ \t]*([-+][ \t]*\\([ \t]*[A-Za-z_][A-Za-z_0-9* \t]*\\)[ \t]*[A-Za-z_].*)$\n"
  94          /* C functions */
  95          "^[ \t]*(([A-Za-z_][A-Za-z_0-9]*[ \t]+)+[A-Za-z_][A-Za-z_0-9]*[ \t]*\\([^;]*)$\n"
  96          /* Objective-C class/protocol definitions */
  97          "^(@(implementation|interface|protocol)[ \t].*)$",
  98          /* -- */
  99          "[a-zA-Z_][a-zA-Z0-9_]*"
 100          "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lL]?"
 101          "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->"),
 102
 103 PATTERNS("pascal",
 104          "^(((class[ \t]+)?(procedure|function)|constructor|destructor|interface|"
 105                 "implementation|initialization|finalization)[ \t]*.*)$"
 106          "\n"
 107          "^(.*=[ \t]*(class|record).*)$",
 108          /* -- */
 109          "[a-zA-Z_][a-zA-Z0-9_]*"
 110          "|[-+0-9.e]+|0[xXbB]?[0-9a-fA-F]+"
 111          "|<>|<=|>=|:=|\\.\\."),
 112
 113 PATTERNS("perl",
 114          "^package .*\n"
 115          "^sub [[:alnum:]_':]+[ \t]*"
 116                 "(\\([^)]*\\)[ \t]*)?" /* prototype */
 117                 /*
 118                  * Attributes.  A regex can't count nested parentheses,
 119                  * so just slurp up whatever we see, taking care not
 120                  * to accept lines like "sub foo; # defined elsewhere".
 121                  *
 122                  * An attribute could contain a semicolon, but at that
 123                  * point it seems reasonable enough to give up.
 124                  */
 125                 "(:[^;#]*)?"
 126                 "(\\{[ \t]*)?" /* brace can come here or on the next line */
 127                 "(#.*)?$\n" /* comment */
 128          "^(BEGIN|END|INIT|CHECK|UNITCHECK|AUTOLOAD|DESTROY)[ \t]*"
 129                 "(\\{[ \t]*)?" /* brace can come here or on the next line */
 130                 "(#.*)?$\n"
 131          "^=head[0-9] .*",      /* POD */
 132          /* -- */
 133          "[[:alpha:]_'][[:alnum:]_']*"
 134          "|0[xb]?[0-9a-fA-F_]*"
 135          /* taking care not to interpret 3..5 as (3.)(.5) */
 136          "|[0-9a-fA-F_]+(\\.[0-9a-fA-F_]+)?([eE][-+]?[0-9_]+)?"
 137          "|=>|-[rwxoRWXOezsfdlpSugkbctTBMAC>]|~~|::"
 138          "|&&=|\\|\\|=|//=|\\*\\*="
 139          "|&&|\\|\\||//|\\+\\+|--|\\*\\*|\\.\\.\\.?"
 140          "|[-+*/%.^&<>=!|]="
 141          "|=~|!~"
 142          "|<<|<>|<=>|>>"),
 143
 144 PATTERNS("python", "^[ \t]*((class|def)[ \t].*)$",
 145          /* -- */
 146          "[a-zA-Z_][a-zA-Z0-9_]*"
 147          "|[-+0-9.e]+[jJlL]?|0[xX]?[0-9a-fA-F]+[lL]?"
 148          "|[-+*/<>%&^|=!]=|//=?|<<=?|>>=?|\\*\\*=?"),
 149
 150 PATTERNS("ruby", "^[ \t]*((class|module|def)[ \t].*)$",
 151          /* -- */
 152          "(@|@@|\\$)?[a-zA-Z_][a-zA-Z0-9_]*"
 153          "|[-+0-9.e]+|0[xXbB]?[0-9a-fA-F]+|\\?(\\\\C-)?(\\\\M-)?."
 154          "|//=?|[-+*/<>%&^|=!]=|<<=?|>>=?|===|\\.{1,3}|::|[!=]~"),
 155
 156 PATTERNS("bibtex", "(@[a-zA-Z]{1,}[ \t]*\\{{0,1}[ \t]*[^ \t\"@',\\#}{~%]*).*$",
 157          "[={}\"]|[^={}\" \t]+"),
 158
 159 PATTERNS("tex", "^(\\\\((sub)*section|chapter|part)\\*{0,1}\\{.*)$",
 160          "\\\\[a-zA-Z@]+|\\\\.|[a-zA-Z0-9\x80-\xff]+"),
 161
 162 PATTERNS("cpp",
 163          /* Jump targets or access declarations */
 164          "!^[ \t]*[A-Za-z_][A-Za-z_0-9]*:[[:space:]]*($|/[/*])\n"
 165          /* functions/methods, variables, and compounds at top level */
 166          "^((::[[:space:]]*)?[A-Za-z_].*)$",
 167          /* -- */
 168          "[a-zA-Z_][a-zA-Z0-9_]*"
 169          "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lLuU]*"
 170          "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
 171
 172 PATTERNS("csharp",
 173          /* Keywords */
 174          "!^[ \t]*(do|while|for|if|else|instanceof|new|return|switch|case|throw|catch|using)\n"
 175          /* Methods and constructors */
 176          "^[ \t]*(((static|public|internal|private|protected|new|virtual|sealed|override|unsafe)[ \t]+)*[][<>@.~_[:alnum:]]+[ \t]+[<>@._[:alnum:]]+[ \t]*\\(.*\\))[ \t]*$\n"
 177          /* Properties */
 178          "^[ \t]*(((static|public|internal|private|protected|new|virtual|sealed|override|unsafe)[ \t]+)*[][<>@.~_[:alnum:]]+[ \t]+[@._[:alnum:]]+)[ \t]*$\n"
 179          /* Type definitions */
 180          "^[ \t]*(((static|public|internal|private|protected|new|unsafe|sealed|abstract|partial)[ \t]+)*(class|enum|interface|struct)[ \t]+.*)$\n"
 181          /* Namespace */
 182          "^[ \t]*(namespace[ \t]+.*)$",
 183          /* -- */
 184          "[a-zA-Z_][a-zA-Z0-9_]*"
 185          "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lL]?"
 186          "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->"),
 187
 188 PATTERNS("php",
 189          "^[ \t]*(((public|private|protected|static|final)[ \t]+)*((class|function)[ \t].*))$",
 190          /* -- */
 191          "[a-zA-Z_][a-zA-Z0-9_]*"
 192          "|[-+0-9.e]+[fFlL]?|0[xX]?[0-9a-fA-F]+[lL]?"
 193          "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->"),
 194
 195 PATTERNS("javascript",
 196          "([a-zA-Z_$][a-zA-Z0-9_$]*(\\.[a-zA-Z0-9_$]+)*[ \t]*=[ \t]*function([ \t][a-zA-Z_$][a-zA-Z0-9_$]*)?[^\\{]*)\n"
 197          "([a-zA-Z_$][a-zA-Z0-9_$]*[ \t]*:[ \t]*function([ \t][a-zA-Z_$][a-zA-Z0-9_$]*)?[^\\{]*)\n"
 198          "[^a-zA-Z0-9_\\$](function([ \t][a-zA-Z_$][a-zA-Z0-9_$]*)?[^\\{]*)",
 199          /* -- */
 200          "[a-zA-Z_][a-zA-Z0-9_]*"
 201          "|[-+0-9.e]+[fFlL]?|0[xX]?[0-9a-fA-F]+[lL]?"
 202          "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->"),
 203 };
 204
 205 #undef IPATTERN
 206 #undef PATTERNS
 207 #undef WORD_DEFAULT
 208
 209 #endif
 210