diff options
Diffstat (limited to 'src/xine-utils/xmllexer.c')
| -rw-r--r-- | src/xine-utils/xmllexer.c | 140 |
1 files changed, 99 insertions, 41 deletions
diff --git a/src/xine-utils/xmllexer.c b/src/xine-utils/xmllexer.c index 047dbb1a6..bb03e5a79 100644 --- a/src/xine-utils/xmllexer.c +++ b/src/xine-utils/xmllexer.c @@ -43,14 +43,11 @@ #include "bswap.h" /* private constants*/ -#define NORMAL 0 /* normal lex mode */ -#define DATA 1 /* data lex mode */ /* private global variables */ static const char * lexbuf; static int lexbuf_size = 0; static int lexbuf_pos = 0; -static int lex_mode = NORMAL; static int in_comment = 0; static char *lex_malloc = NULL; @@ -92,6 +89,12 @@ static void lex_convert (const char * buf, int size, enum utf utf) lexbuf = lex_malloc = realloc (utf8, lexbuf_size + 1); } +static enum { + NORMAL, + DATA, + CDATA, +} lex_mode = NORMAL; + void lexer_init(const char * buf, int size) { static const char boms[] = { 0xFF, 0xFE, 0, 0, 0xFE, 0xFF }, bom_utf8[] = { 0xEF, 0xBB, 0xBF }; @@ -123,83 +126,102 @@ void lexer_init(const char * buf, int size) { lprintf("buffer length %d\n", size); } +typedef enum { + STATE_UNKNOWN = -1, + STATE_IDLE, + STATE_EOL, + STATE_SEPAR, + STATE_T_M_START, + STATE_T_M_STOP_1, + STATE_T_M_STOP_2, + STATE_T_EQUAL, + STATE_T_STRING_SINGLE, + STATE_T_STRING_DOUBLE, + STATE_T_COMMENT, + STATE_T_TI_STOP, + STATE_T_DASHDASH, + STATE_T_C_STOP, + STATE_IDENT /* must be last */ +} lexer_state_t; + int lexer_get_token(char * tok, int tok_size) { int tok_pos = 0; - int state = 0; + lexer_state_t state = STATE_IDLE; char c; if (tok) { while ((tok_pos < tok_size) && (lexbuf_pos < lexbuf_size)) { c = lexbuf[lexbuf_pos]; - lprintf("c=%c, state=%d, in_comment=%d\n", c, state, in_comment); + lprintf("c=%c, state=%d, lex_mode=%d, in_comment=%d\n", c, state, lex_mode, in_comment); - if (lex_mode == NORMAL) { - /* normal mode */ + switch (lex_mode) { + case NORMAL: switch (state) { /* init state */ - case 0: + case STATE_IDLE: switch (c) { case '\n': case '\r': - state = 1; + state = STATE_EOL; tok[tok_pos] = c; tok_pos++; break; case ' ': case '\t': - state = 2; + state = STATE_SEPAR; tok[tok_pos] = c; tok_pos++; break; case '<': - state = 3; + state = STATE_T_M_START; tok[tok_pos] = c; tok_pos++; break; case '>': - state = 4; + state = STATE_T_M_STOP_1; tok[tok_pos] = c; tok_pos++; break; case '/': if (!in_comment) - state = 5; + state = STATE_T_M_STOP_2; tok[tok_pos] = c; tok_pos++; break; case '=': - state = 6; + state = STATE_T_EQUAL; tok[tok_pos] = c; tok_pos++; break; case '\"': /* " */ - state = 7; + state = STATE_T_STRING_DOUBLE; break; case '\'': /* " */ - state = 12; + state = STATE_T_STRING_SINGLE; break; case '-': - state = 10; + state = STATE_T_DASHDASH; tok[tok_pos] = c; tok_pos++; break; case '?': - state = 9; + if (!in_comment) + state = STATE_T_TI_STOP; tok[tok_pos] = c; tok_pos++; break; default: - state = 100; + state = STATE_IDENT; tok[tok_pos] = c; tok_pos++; break; @@ -208,7 +230,7 @@ int lexer_get_token(char * tok, int tok_size) { break; /* end of line */ - case 1: + case STATE_EOL: if (c == '\n' || (c == '\r')) { tok[tok_pos] = c; lexbuf_pos++; @@ -220,7 +242,7 @@ int lexer_get_token(char * tok, int tok_size) { break; /* T_SEPAR */ - case 2: + case STATE_SEPAR: if (c == ' ' || (c == '\t')) { tok[tok_pos] = c; lexbuf_pos++; @@ -232,7 +254,7 @@ int lexer_get_token(char * tok, int tok_size) { break; /* T_M_START < or </ or <! or <? */ - case 3: + case STATE_T_M_START: switch (c) { case '/': tok[tok_pos] = c; @@ -245,7 +267,7 @@ int lexer_get_token(char * tok, int tok_size) { tok[tok_pos] = c; lexbuf_pos++; tok_pos++; - state = 8; + state = STATE_T_COMMENT; break; case '?': tok[tok_pos] = c; @@ -261,7 +283,7 @@ int lexer_get_token(char * tok, int tok_size) { break; /* T_M_STOP_1 */ - case 4: + case STATE_T_M_STOP_1: tok[tok_pos] = '\0'; if (!in_comment) lex_mode = DATA; @@ -269,7 +291,7 @@ int lexer_get_token(char * tok, int tok_size) { break; /* T_M_STOP_2 */ - case 5: + case STATE_T_M_STOP_2: if (c == '>') { tok[tok_pos] = c; lexbuf_pos++; @@ -285,13 +307,13 @@ int lexer_get_token(char * tok, int tok_size) { break; /* T_EQUAL */ - case 6: + case STATE_T_EQUAL: tok[tok_pos] = '\0'; return T_EQUAL; break; /* T_STRING */ - case 7: + case STATE_T_STRING_DOUBLE: tok[tok_pos] = c; lexbuf_pos++; if (c == '\"') { /* " */ @@ -301,8 +323,8 @@ int lexer_get_token(char * tok, int tok_size) { tok_pos++; break; - /* T_C_START or T_DOCTYPE_START */ - case 8: + /* T_C_START or T_DOCTYPE_START or T_CDATA_START */ + case STATE_T_COMMENT: switch (c) { case '-': lexbuf_pos++; @@ -326,6 +348,17 @@ int lexer_get_token(char * tok, int tok_size) { return T_ERROR; } break; + case '[': + lexbuf_pos++; + if (strncmp(lexbuf + lexbuf_pos, "CDATA[", 6) == 0) { + strncpy (tok + tok_pos, "[CDATA[", 7); /* FIXME */ + lexbuf_pos += 6; + lex_mode = CDATA; + return T_CDATA_START; + } else{ + return T_ERROR; + } + break; default: /* error */ return T_ERROR; @@ -333,12 +366,14 @@ int lexer_get_token(char * tok, int tok_size) { break; /* T_TI_STOP */ - case 9: + case STATE_T_TI_STOP: if (c == '>') { tok[tok_pos] = c; lexbuf_pos++; tok_pos++; /* FIXME */ tok[tok_pos] = '\0'; + if (!in_comment) + lex_mode = DATA; return T_TI_STOP; } else { tok[tok_pos] = '\0'; @@ -347,24 +382,24 @@ int lexer_get_token(char * tok, int tok_size) { break; /* -- */ - case 10: + case STATE_T_DASHDASH: switch (c) { case '-': tok[tok_pos] = c; tok_pos++; lexbuf_pos++; - state = 11; + state = STATE_T_C_STOP; break; default: tok[tok_pos] = c; tok_pos++; lexbuf_pos++; - state = 100; + state = STATE_IDENT; } break; /* --> */ - case 11: + case STATE_T_C_STOP: switch (c) { case '>': tok[tok_pos] = c; @@ -384,12 +419,12 @@ int lexer_get_token(char * tok, int tok_size) { tok[tok_pos] = c; tok_pos++; lexbuf_pos++; - state = 100; + state = STATE_IDENT; } break; /* T_STRING (single quotes) */ - case 12: + case STATE_T_STRING_SINGLE: tok[tok_pos] = c; lexbuf_pos++; if (c == '\'') { /* " */ @@ -400,7 +435,7 @@ int lexer_get_token(char * tok, int tok_size) { break; /* IDENT */ - case 100: + case STATE_IDENT: switch (c) { case '<': case '>': @@ -417,13 +452,13 @@ int lexer_get_token(char * tok, int tok_size) { tok[tok_pos] = c; tok_pos++; lexbuf_pos++; - state = 9; + state = STATE_T_TI_STOP; break; case '-': tok[tok_pos] = c; tok_pos++; lexbuf_pos++; - state = 10; + state = STATE_T_DASHDASH; break; default: tok[tok_pos] = c; @@ -435,8 +470,9 @@ int lexer_get_token(char * tok, int tok_size) { lprintf("expected char \'%c\'\n", tok[tok_pos - 1]); /* FIX ME */ return T_ERROR; } - } else { - /* data mode, stop if char equal '<' */ + break; + + case DATA: /* data mode, stop if char equal '<' */ switch (c) { case '<': @@ -448,6 +484,28 @@ int lexer_get_token(char * tok, int tok_size) { tok_pos++; lexbuf_pos++; } + break; + + case CDATA: /* cdata mode, stop if next token is "]]>" */ + switch (c) + { + case ']': + if (strncmp(lexbuf + lexbuf_pos, "]]>", 3) == 0) { + lexbuf_pos += 3; + lex_mode = DATA; + return T_CDATA_STOP; + } else { + tok[tok_pos] = c; + tok_pos++; + lexbuf_pos++; + } + break; + default: + tok[tok_pos] = c; + tok_pos++; + lexbuf_pos++; + } + break; } } lprintf ("loop done tok_pos = %d, tok_size=%d, lexbuf_pos=%d, lexbuf_size=%d\n", |