Minifi README

[xonotic/gmqcc.git] / lexer.c

diff --git a/lexer.c b/lexer.c
index 068b4dc824bd5563d1c96722860c4e054b765d79..e5684c0ba2466aadce52bc3c46c4ba05f8e5fd11 100644 (file)
--- a/lexer.c
+++ b/lexer.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2012, 2013
+ * Copyright (C) 2012, 2013, 2014, 2015
  *     Wolfgang Bumiller
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy of
@@ -99,7 +99,7 @@ static int lex_getch(lex_file *lex);
 lex_file* lex_open(const char *file)
 {
     lex_file  *lex;
-    fs_file_t *in = fs_file_open(file, "rb");
+    FILE *in = fopen(file, "rb");
     uint32_t   read;
 
     if (!in) {
@@ -109,7 +109,7 @@ lex_file* lex_open(const char *file)
 
     lex = (lex_file*)mem_a(sizeof(*lex));
     if (!lex) {
-        fs_file_close(in);
+        fclose(in);
         lexerror(NULL, "out of memory\n");
         return NULL;
     }
@@ -187,7 +187,7 @@ void lex_close(lex_file *lex)
         vec_free(lex->modelname);
 
     if (lex->file)
-        fs_file_close(lex->file);
+        fclose(lex->file);
 
     vec_free(lex->tok.value);
 
@@ -201,15 +201,15 @@ static int lex_fgetc(lex_file *lex)
 {
     if (lex->file) {
         lex->column++;
-        return fs_file_getc(lex->file);
+        return fgetc(lex->file);
     }
     if (lex->open_string) {
         if (lex->open_string_pos >= lex->open_string_length)
-            return FS_FILE_EOF;
+            return EOF;
         lex->column++;
         return lex->open_string[lex->open_string_pos++];
     }
-    return FS_FILE_EOF;
+    return EOF;
 }
 
 /* Get or put-back data
@@ -424,7 +424,7 @@ static bool lex_try_pragma(lex_file *lex)
         goto unroll;
 
     lex->line = line;
-    while (ch != '\n' && ch != FS_FILE_EOF)
+    while (ch != '\n' && ch != EOF)
         ch = lex_getch(lex);
     vec_free(command);
     vec_free(param);
@@ -504,7 +504,7 @@ static int lex_skipwhite(lex_file *lex, bool hadwhite)
     do
     {
         ch = lex_getch(lex);
-        while (ch != FS_FILE_EOF && util_isspace(ch)) {
+        while (ch != EOF && util_isspace(ch)) {
             if (ch == '\n') {
                 if (lex_try_pragma(lex))
                     continue;
@@ -540,7 +540,7 @@ static int lex_skipwhite(lex_file *lex, bool hadwhite)
                     lex_tokench(lex, ' ');
                 }
 
-                while (ch != FS_FILE_EOF && ch != '\n') {
+                while (ch != EOF && ch != '\n') {
                     if (lex->flags.preprocessing)
                         lex_tokench(lex, ' '); /* ch); */
                     ch = lex_getch(lex);
@@ -561,7 +561,7 @@ static int lex_skipwhite(lex_file *lex, bool hadwhite)
                     lex_tokench(lex, ' ');
                 }
 
-                while (ch != FS_FILE_EOF)
+                while (ch != EOF)
                 {
                     ch = lex_getch(lex);
                     if (ch == '*') {
@@ -579,7 +579,7 @@ static int lex_skipwhite(lex_file *lex, bool hadwhite)
                         if (ch == '\n')
                             lex_tokench(lex, '\n');
                         else
-                            lex_tokench(lex, ' '); /* ch); */
+                            lex_tokench(lex, ' ');
                     }
                 }
                 ch = ' '; /* cause TRUE in the isspace check */
@@ -590,7 +590,7 @@ static int lex_skipwhite(lex_file *lex, bool hadwhite)
             ch = '/';
             break;
         }
-    } while (ch != FS_FILE_EOF && util_isspace(ch));
+    } while (ch != EOF && util_isspace(ch));
 
     if (haswhite) {
         lex_endtoken(lex);
@@ -606,7 +606,7 @@ static bool GMQCC_WARN lex_finish_ident(lex_file *lex)
     int ch;
 
     ch = lex_getch(lex);
-    while (ch != FS_FILE_EOF && isident(ch))
+    while (ch != EOF && isident(ch))
     {
         lex_tokench(lex, ch);
         ch = lex_getch(lex);
@@ -626,7 +626,7 @@ static int lex_parse_frame(lex_file *lex)
     lex_token_new(lex);
 
     ch = lex_getch(lex);
-    while (ch != FS_FILE_EOF && ch != '\n' && util_isspace(ch))
+    while (ch != EOF && ch != '\n' && util_isspace(ch))
         ch = lex_getch(lex);
 
     if (ch == '\n')
@@ -681,14 +681,14 @@ static bool lex_finish_frames(lex_file *lex)
 static int GMQCC_WARN lex_finish_string(lex_file *lex, int quote)
 {
     utf8ch_t chr = 0;
-    int ch = 0;
+    int ch = 0, texttype = 0;
     int nextch;
     bool hex;
     bool oct;
     char u8buf[8]; /* way more than enough */
     int  u8len, uc;
 
-    while (ch != FS_FILE_EOF)
+    while (ch != EOF)
     {
         ch = lex_getch(lex);
         if (ch == quote)
@@ -697,18 +697,18 @@ static int GMQCC_WARN lex_finish_string(lex_file *lex, int quote)
         if (lex->flags.preprocessing && ch == '\\') {
             lex_tokench(lex, ch);
             ch = lex_getch(lex);
-            if (ch == FS_FILE_EOF) {
+            if (ch == EOF) {
                 lexerror(lex, "unexpected end of file");
-                lex_ungetch(lex, FS_FILE_EOF); /* next token to be TOKEN_EOF */
+                lex_ungetch(lex, EOF); /* next token to be TOKEN_EOF */
                 return (lex->tok.ttype = TOKEN_ERROR);
             }
             lex_tokench(lex, ch);
         }
         else if (ch == '\\') {
             ch = lex_getch(lex);
-            if (ch == FS_FILE_EOF) {
+            if (ch == EOF) {
                 lexerror(lex, "unexpected end of file");
-                lex_ungetch(lex, FS_FILE_EOF); /* next token to be TOKEN_EOF */
+                lex_ungetch(lex, EOF); /* next token to be TOKEN_EOF */
                 return (lex->tok.ttype = TOKEN_ERROR);
             }
 
@@ -716,13 +716,12 @@ static int GMQCC_WARN lex_finish_string(lex_file *lex, int quote)
             case '\\': break;
             case '\'': break;
             case '"':  break;
-            case 'a':  ch = '\a'; break;
-            case 'b':  ch = '\b'; break;
-            case 'r':  ch = '\r'; break;
-            case 'n':  ch = '\n'; break;
-            case 't':  ch = '\t'; break;
-            case 'f':  ch = '\f'; break;
-            case 'v':  ch = '\v'; break;
+            case 'a': ch = '\a'; break;
+            case 'r': ch = '\r'; break;
+            case 'n': ch = '\n'; break;
+            case 't': ch = '\t'; break;
+            case 'f': ch = '\f'; break;
+            case 'v': ch = '\v'; break;
             case 'x':
             case 'X':
                 /* same procedure as in fteqcc */
@@ -825,7 +824,15 @@ static int GMQCC_WARN lex_finish_string(lex_file *lex, int quote)
                 else
                     ch = chr;
                 break;
-            case '\n':  ch = '\n'; break;
+
+            /* high bit text */
+            case 'b': case 's':
+                texttype ^= 128;
+                continue;
+
+            case '\n':
+                ch = '\n';
+                break;
 
             default:
                 lexwarn(lex, WARN_UNKNOWN_CONTROL_SEQUENCE, "unrecognized control sequence: \\%c", ch);
@@ -833,13 +840,13 @@ static int GMQCC_WARN lex_finish_string(lex_file *lex, int quote)
                 lex_tokench(lex, '\\');
             }
             /* add the character finally */
-            lex_tokench(lex, ch);
+            lex_tokench(lex, ch | texttype);
         }
         else
             lex_tokench(lex, ch);
     }
     lexerror(lex, "unexpected end of file within string constant");
-    lex_ungetch(lex, FS_FILE_EOF); /* next token to be TOKEN_EOF */
+    lex_ungetch(lex, EOF); /* next token to be TOKEN_EOF */
     return (lex->tok.ttype = TOKEN_ERROR);
 }
 
@@ -956,7 +963,7 @@ int lex_do(lex_file *lex)
     if (lex->eof)
         return (lex->tok.ttype = TOKEN_FATAL);
 
-    if (ch == FS_FILE_EOF) {
+    if (ch == EOF) {
         lex->eof = true;
         return (lex->tok.ttype = TOKEN_EOF);
     }
@@ -993,7 +1000,7 @@ int lex_do(lex_file *lex)
         if (!strcmp(v, "framevalue"))
         {
             ch = lex_getch(lex);
-            while (ch != FS_FILE_EOF && util_isspace(ch) && ch != '\n')
+            while (ch != EOF && util_isspace(ch) && ch != '\n')
                 ch = lex_getch(lex);
 
             if (!util_isdigit(ch)) {
@@ -1073,7 +1080,7 @@ int lex_do(lex_file *lex)
             vec_free(lex->frames);
             /* skip line (fteqcc does it too) */
             ch = lex_getch(lex);
-            while (ch != FS_FILE_EOF && ch != '\n')
+            while (ch != EOF && ch != '\n')
                 ch = lex_getch(lex);
             return lex_do(lex);
         }
@@ -1087,7 +1094,7 @@ int lex_do(lex_file *lex)
         {
             /* skip line */
             ch = lex_getch(lex);
-            while (ch != FS_FILE_EOF && ch != '\n')
+            while (ch != EOF && ch != '\n')
                 ch = lex_getch(lex);
             return lex_do(lex);
         }
@@ -1233,12 +1240,16 @@ int lex_do(lex_file *lex)
         ch == '~' || ch == '^'    /* ~=, ~, ^                        */
     )  {
         lex_tokench(lex, ch);
-
         nextch = lex_getch(lex);
-        if ((nextch == '=' && ch != '<') ||
-            (nextch == ch  && ch != '!') ||
-            (nextch == '<' && ch == '>')) {
+
+        if ((nextch == '=' && ch != '<') || (nextch == '<' && ch == '>'))
+            lex_tokench(lex, nextch);
+        else if (nextch == ch && ch != '!') {
             lex_tokench(lex, nextch);
+            if ((thirdch = lex_getch(lex)) == '=')
+                lex_tokench(lex, thirdch);
+            else
+                lex_ungetch(lex, thirdch);
         } else if (ch == '<' && nextch == '=') {
             lex_tokench(lex, nextch);
             if ((thirdch = lex_getch(lex)) == '>')
@@ -1328,6 +1339,8 @@ int lex_do(lex_file *lex)
         } else if (!strcmp(v, "vector")) {
             lex->tok.ttype = TOKEN_TYPENAME;
             lex->tok.constval.t = TYPE_VECTOR;
+        } else if (!strcmp(v, "_length")) {
+            lex->tok.ttype = TOKEN_OPERATOR;
         } else {
             size_t kw;
             for (kw = 0; kw < GMQCC_ARRAY_COUNT(keywords_qc); ++kw) {