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