projects / cparser / blobdiff
? search:
summary | shortlog | log | commit | commitdiff | tree
raw | inline | side by side
condeval is called jump threading now
[cparser] / lexer.c
diff --git a/lexer.c b/lexer.c
index b90edcd..a4bace0 100644 (file)
--- a/lexer.c
+++ b/lexer.c
@@ -1,6 +1,6 @@
 /*
  * This file is part of cparser.
- * Copyright (C) 2007-2008 Matthias Braun <matze@braunis.de>
+ * Copyright (C) 2007-2009 Matthias Braun <matze@braunis.de>
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License
@@ -37,10 +37,13 @@
 #include <assert.h>
 #include <errno.h>
 #include <string.h>
-#include <strings.h>
 #include <stdbool.h>
 #include <ctype.h>
 
+#ifndef _WIN32
+#include <strings.h>
+#endif
+
 //#define DEBUG_CHARS
 #define MAX_PUTBACK 3
 #define BUF_SIZE    1024
@@ -253,6 +256,54 @@ realign:
        } while (bufpos == bufend);
 }
 
+static void decode_windows_1252(void)
+{
+       unsigned char read_buf[BUF_SIZE];
+       size_t const s = read_block(read_buf, sizeof(read_buf));
+       if (s == 0)
+               return;
+
+       unsigned char const *src = read_buf;
+       unsigned char const *end = read_buf + s;
+       utf32               *dst = buf + MAX_PUTBACK;
+       while (src != end) {
+               utf32 tc = *src++;
+               switch (tc) {
+                       case 0x80: tc = 0x20AC; break; // €
+                       case 0x82: tc = 0x201A; break; // ‚
+                       case 0x83: tc = 0x0192; break; // ƒ
+                       case 0x84: tc = 0x201E; break; // „
+                       case 0x85: tc = 0x2026; break; // …
+                       case 0x86: tc = 0x2020; break; // †
+                       case 0x87: tc = 0x2021; break; // ‡
+                       case 0x88: tc = 0x02C6; break; // ˆ
+                       case 0x89: tc = 0x2030; break; // ‰
+                       case 0x8A: tc = 0x0160; break; // Š
+                       case 0x8B: tc = 0x2039; break; // ‹
+                       case 0x8C: tc = 0x0152; break; // Œ
+                       case 0x8E: tc = 0x017D; break; // Ž
+                       case 0x91: tc = 0x2018; break; // ‘
+                       case 0x92: tc = 0x2019; break; // ’
+                       case 0x93: tc = 0x201C; break; // “
+                       case 0x94: tc = 0x201D; break; // ”
+                       case 0x95: tc = 0x2022; break; // •
+                       case 0x96: tc = 0x2013; break; // –
+                       case 0x97: tc = 0x2014; break; // —
+                       case 0x98: tc = 0x02DC; break; // ˜
+                       case 0x99: tc = 0x2122; break; // ™
+                       case 0x9A: tc = 0x0161; break; // š
+                       case 0x9B: tc = 0x203A; break; // ›
+                       case 0x9C: tc = 0x0153; break; // œ
+                       case 0x9E: tc = 0x017E; break; // ž
+                       case 0x9F: tc = 0x0178; break; // Ÿ
+               }
+               *dst++ = tc;
+       }
+
+       bufpos = buf + MAX_PUTBACK;
+       bufend = dst;
+}
+
 typedef void (*decoder_t)(void);
 
 static decoder_t decoder = decode_utf8;
@@ -263,23 +314,25 @@ typedef struct named_decoder_t {
 } named_decoder_t;
 
 static named_decoder_t const decoders[] = {
-       { "CP819",           decode_iso_8859_1  }, // offical alias
-       { "IBM819",          decode_iso_8859_1  }, // offical alias
-       { "ISO-8859-1",      decode_iso_8859_1  }, // offical alias
-       { "ISO-8859-15",     decode_iso_8859_15 }, // offical name
-       { "ISO8859-1",       decode_iso_8859_1  },
-       { "ISO8859-15",      decode_iso_8859_15 },
-       { "ISO_8859-1",      decode_iso_8859_1  }, // offical alias
-       { "ISO_8859-15",     decode_iso_8859_15 }, // offical alias
-       { "ISO_8859-1:1987", decode_iso_8859_1  }, // offical name
-       { "Latin-9",         decode_iso_8859_15 }, // offical alias
-       { "UTF-8",           decode_utf8        }, // offical name
-       { "csISOLatin1",     decode_iso_8859_1  }, // offical alias
-       { "iso-ir-100",      decode_iso_8859_1  }, // offical alias
-       { "l1",              decode_iso_8859_1  }, // offical alias
-       { "latin1",          decode_iso_8859_1  }, // offical alias
-
-       { NULL,              NULL               }
+       { "CP819",           decode_iso_8859_1   }, // offical alias
+       { "IBM819",          decode_iso_8859_1   }, // offical alias
+       { "ISO-8859-1",      decode_iso_8859_1   }, // offical alias
+       { "ISO-8859-15",     decode_iso_8859_15  }, // offical name
+       { "ISO8859-1",       decode_iso_8859_1   },
+       { "ISO8859-15",      decode_iso_8859_15  },
+       { "ISO_8859-1",      decode_iso_8859_1   }, // offical alias
+       { "ISO_8859-15",     decode_iso_8859_15  }, // offical alias
+       { "ISO_8859-1:1987", decode_iso_8859_1   }, // offical name
+       { "Latin-9",         decode_iso_8859_15  }, // offical alias
+       { "UTF-8",           decode_utf8         }, // offical name
+       { "csISOLatin1",     decode_iso_8859_1   }, // offical alias
+       { "cp1252",          decode_windows_1252 },
+       { "iso-ir-100",      decode_iso_8859_1   }, // offical alias
+       { "l1",              decode_iso_8859_1   }, // offical alias
+       { "latin1",          decode_iso_8859_1   }, // offical alias
+       { "windows-1252",    decode_windows_1252 }, // official name
+
+       { NULL,              NULL                }
 };
 
 void select_input_encoding(char const* const encoding)
@@ -1693,7 +1746,7 @@ static void parse_preprocessor_directive(void)
                parse_line_directive();
                break;
        case '\n':
-               /* NULL directive, see § 6.10.7 */
+               /* NULL directive, see §6.10.7 */
                break;
        default:
                parse_error("invalid preprocessor directive");
@@ -1713,6 +1766,16 @@ static void parse_preprocessor_directive(void)
                                        lexer_token.type = set_type;           \
                                        return;
 
+/* must use this as last thing */
+#define MAYBE_MODE(ch, set_type, mode)                     \
+                               case ch:                                   \
+                                       if (c_mode & mode) {                   \
+                                               next_char();                       \
+                                               lexer_token.type = set_type;       \
+                                               return;                            \
+                                       }                                      \
+                                       /* fallthrough */
+
 #define ELSE_CODE(code)                                    \
                                default:                                   \
                                        code                                   \
@@ -1868,6 +1931,7 @@ void lexer_next_preprocessing_token(void)
                case ':':
                        MAYBE_PROLOG
                        MAYBE('>', ']')
+                       MAYBE_MODE(':', T_COLONCOLON, _CXX)
                        ELSE(':')
                case '=':
                        MAYBE_PROLOG
cparser