home *** CD-ROM | disk | FTP | other *** search
- /*
- tidy.c - HTML parser and pretty printer
-
- Copyright (c) 1998 World Wide Web Consortium (Massachusetts
- Institute of Technology, Institut National de Recherche en
- Informatique et en Automatique, Keio University). All Rights
- Reserved.
-
- Contributing Author(s):
-
- Dave Raggett <dsr@w3.org>
-
- The contributing author(s) would like to thank all those who
- helped with testing, bug fixes, and patience. This wouldn't
- have been possible without all of you.
-
- COPYRIGHT NOTICE:
-
- This software and documentation is provided "as is," and
- the copyright holders and contributing author(s) make no
- representations or warranties, express or implied, including
- but not limited to, warranties of merchantability or fitness
- for any particular purpose or that the use of the software or
- documentation will not infringe any third party patents,
- copyrights, trademarks or other rights.
-
- The copyright holders and contributing author(s) will not be
- liable for any direct, indirect, special or consequential damages
- arising out of any use of the software or documentation, even if
- advised of the possibility of such damage.
-
- Permission is hereby granted to use, copy, modify, and distribute
- this source code, or portions hereof, documentation and executables,
- for any purpose, without fee, subject to the following restrictions:
-
- 1. The origin of this source code must not be misrepresented.
- 2. Altered versions must be plainly marked as such and must
- not be misrepresented as being the original source.
- 3. This Copyright notice may not be removed or altered from any
- source or altered source distribution.
-
- The copyright holders and contributing author(s) specifically
- permit, without fee, and encourage the use of this source code
- as a component for supporting the Hypertext Markup Language in
- commercial products. If you use this source code in a product,
- acknowledgment is not required but would be appreciated.
- */
-
- #include "platform.h"
- #include "html.h"
-
- void InitTidy(void);
- void DeInitTidy(void);
-
- extern char *release_date;
-
- Bool debug_flag = no;
- Node *debug_element = null;
- Lexer *debug_lexer = null;
- uint totalerrors = 0;
- uint totalwarnings = 0;
-
- jmp_buf error_exit; /* Address for long jump to jump to */
- FILE *errout; /* set to stderr or stdout */
-
- void FatalError(char *msg)
- {
- fprintf(stderr, "Fatal error: %s\n", msg);
- DeInitTidy();
- longjmp(error_exit, -1);
- }
-
- void *MemAlloc(uint size)
- {
- void *p;
-
- p = malloc(size);
-
- if (!p)
- FatalError("Out of memory!");
-
- return p;
- }
-
- void *MemRealloc(void *mem, uint newsize)
- {
- void *p;
-
- if (mem == (void *)null)
- return MemAlloc(newsize);
-
- p = realloc(mem, newsize);
-
- if (!p)
- FatalError("Out of memory!");
-
- return p;
- }
-
- void MemFree(void *mem)
- {
- if (mem != (void *)null)
- free(mem);
- }
-
- void ClearMemory(void *mem, uint size)
- {
- memset(mem, 0, size);
- }
-
- StreamIn *OpenInput(FILE *fp)
- {
- StreamIn *in;
-
- in = (StreamIn *)MemAlloc(sizeof(StreamIn));
- in->file = fp;
- in->pushed = no;
- in->c = '\0';
- in->tabs = 0;
- in->curline = 1;
- in->curcol = 1;
- in->encoding = CharEncoding;
- in->state = FSM_ASCII;
-
- return in;
- }
-
- /* read char from stream */
- int ReadCharFromStream(StreamIn *in)
- {
- uint n, c, i, count;
-
- if (feof(in->file))
- return -1;
-
- c = getc(in->file);
-
- /*
- A document in ISO-2022 based encoding uses some ESC sequences
- called "designator" to switch character sets. The designators
- defined and used in ISO-2022-JP are:
-
- "ESC" + "(" + ? for ISO646 variants
-
- "ESC" + "$" + ? and
- "ESC" + "$" + "(" + ? for multibyte character sets
-
- Where ? stands for a single character used to indicate the
- character set for multibyte characters.
-
- Tidy handles this by preserving the escape sequence and
- setting the top bit of each byte for non-ascii chars. This
- bit is then cleared on output. The input stream keeps track
- of the state to determine when to set/clear the bit.
- */
-
- if (in->encoding == ISO2022)
- {
- if (c == 0x1b) /* ESC */
- {
- in->state = FSM_ESC;
- return c;
- }
-
- switch (in->state)
- {
- case FSM_ESC:
- if (c == '$')
- in->state = FSM_ESCD;
- else if (c == '(')
- in->state = FSM_ESCP;
- else
- in->state = FSM_ASCII;
- break;
-
- case FSM_ESCD:
- if (c == '(')
- in->state = FSM_ESCDP;
- else
- in->state = FSM_NONASCII;
- break;
-
- case FSM_ESCDP:
- in->state = FSM_NONASCII;
- break;
-
- case FSM_ESCP:
- in->state = FSM_ASCII;
- break;
-
- case FSM_NONASCII:
- c |= 0x80;
- break;
- }
-
- return c;
- }
-
- if (in->encoding != UTF8)
- return c;
-
- /* deal with UTF-8 encoded char */
-
- if ((c & 0xE0) == 0xC0) /* 110X XXXX two bytes */
- {
- n = c & 31;
- count = 1;
- }
- else if ((c & 0xF0) == 0xE0) /* 1110 XXXX three bytes */
- {
- n = c & 15;
- count = 2;
- }
- else if ((c & 0xF8) == 0xF0) /* 1111 0XXX four bytes */
- {
- n = c & 7;
- count = 3;
- }
- else if ((c & 0xFC) == 0xF8) /* 1111 10XX five bytes */
- {
- n = c & 3;
- count = 4;
- }
- else if ((c & 0xFE) == 0xFC) /* 1111 110X six bytes */
- {
- n = c & 1;
- count = 5;
- }
- else /* 0XXX XXXX one byte */
- return c;
-
- /* successor bytes should have the form 10XX XXXX */
- for (i = 1; i <= count; ++i)
- {
- if (feof(in->file))
- return -1;
-
- c = getc(in->file);
-
- n = (n << 6) | (c & 0x3F);
- }
-
- return n;
- }
-
- int ReadChar(StreamIn *in)
- {
- int c;
-
- if (in->pushed)
- {
- in->pushed = no;
- c = in->c;
-
- if (c == '\n')
- {
- in->curcol = 1;
- in->curline++;
- return c;
- }
-
- in->curcol++;
- return c;
- }
-
- in->lastcol = in->curcol;
-
- if (in->tabs > 0)
- {
- in->curcol++;
- in->tabs--;
- return ' ';
- }
-
- for (;;)
- {
- c = ReadCharFromStream(in);
-
- if (c < 0)
- return EndOfStream;
-
- if (c == '\n')
- {
- in->curcol = 1;
- in->curline++;
- break;
- }
-
- if (c == '\t')
- {
- in->tabs = tabsize - ((in->curcol - 1) % tabsize) - 1;
- in->curcol++;
- c = ' ';
- break;
- }
-
- /* strip control characters, except for Esc */
-
- if (c == '\033')
- break;
-
- if (0 < c && c < 32)
- continue;
-
- /* watch out for IS02022 */
-
- if (in->encoding == RAW || in->encoding == ISO2022)
- {
- in->curcol++;
- break;
- }
-
- /* produced e.g. as a side-effect of smart quotes in Word */
-
- if (127 < c && c < 160)
- {
- ReportEncodingError(in->lexer, WINDOWS_CHARS, c);
-
- if (c == 150)
- c = 8211; /* en dash */
- else if (c == 151)
- c = 8212; /* em dash */
- else if (c == 138)
- c = 352; /* latin capital letter S with caron */
- else if (c == 154)
- c = 353; /* latin small letter s with caron */
- else if (c == 159)
- c = 376; /* latin capital letter Y with diaeresis */
- else if (c == 140)
- c = 338; /* latin capital ligature OE */
- else if (c == 156)
- c = 339; /* latin capital ligature OE */
- else if (c == 153)
- c = 8482; /* TM */
- else if (c == 134)
- c = 8224; /* dagger */
- else if (c == 135)
- c = 8225; /* double dagger */
- else if (c == 137)
- c = 8240; /* per mille sign */
- else if (c == 130)
- c = 8218; /* single low quotation mark */
- else if (c == 132)
- c = 8222; /* double low quotation mark */
- else if (c == 145)
- c = 8216; /* single left quotation mark */
- else if (c == 146)
- c = 8217; /* single right quotation mark */
- else if (c == 147)
- c = 8220; /* double left quotation mark */
- else if (c == 148)
- c = 8221; /* double right quotation mark */
- else if (c == 139)
- c = 8249; /* single left-pointing angle quotation mark */
- else if (c == 155)
- c = 8250; /* single right-pointing angle quotation mark */
- else
- continue;
- }
-
- in->curcol++;
- break;
- }
-
- return c;
- }
-
- void UngetChar(int c, StreamIn *in)
- {
- in->pushed = yes;
- in->c = c;
-
- if (c == '\n')
- --(in->curline);
-
- in->curcol = in->lastcol;
- }
-
- /* like strdup but using MemAlloc */
- char *wstrdup(char *str)
- {
- char *s, *p;
- int len;
-
- if (str == null)
- return null;
-
- for (len = 0; str[len] != '\0'; ++len);
-
- s = (char *)MemAlloc(sizeof(char)*(1+len));
- for (p = s; *p++ = *str++;);
- return s;
- }
-
- /* like strndup but using MemAlloc */
- char *wstrndup(char *str, int len)
- {
- char *s, *p;
-
- if (str == null || len < 0)
- return null;
-
- s = (char *)MemAlloc(sizeof(char)*(1+len));
-
- p = s;
-
- while (len-- > 0 && (*p++ = *str++));
-
- *p = '\0';
- return s;
- }
-
- /* exactly same as strncpy */
- void wstrncpy(char *s1, char *s2, int size)
- {
- if (s1 != null && s2 != null)
- {
- if (size >= 0)
- {
- while (size--)
- *s1++ = *s2++;
- }
- else
- while (*s1++ = *s2++);
- }
- }
-
- void wstrcpy(char *s1, char *s2)
- {
- while (*s1++ = *s2++);
- }
-
- /* exactly same as strcmp */
- int wstrcmp(char *s1, char *s2)
- {
- int c;
-
- while ((c = *s1) == *s2)
- {
- if (c == '\0')
- return 0;
-
- ++s1;
- ++s2;
- }
-
- return (*s1 > *s2 ? 1 : -1);
- }
-
- /* returns byte count, not char count */
- int wstrlen(char *str)
- {
- int len = 0;
-
- while(*str++)
- ++len;
-
- return len;
- }
-
- /*
- MS C 4.2 doesn't include strcasecmp.
- Note that tolower and toupper won't
- work on chars > 127
- */
- int wstrcasecmp(char *s1, char *s2)
- {
- uint c;
-
- while (c = (uint)(*s1), ToLower(c) == ToLower((uint)(*s2)))
- {
- if (c == '\0')
- return 0;
-
- ++s1;
- ++s2;
- }
-
- return (*s1 > *s2 ? 1 : -1);
- }
-
- int wstrncmp(char *s1, char *s2, int n)
- {
- int c;
-
- while ((c = *s1) == *s2)
- {
- if (c == '\0')
- return 0;
-
- if (n == 0)
- return 0;
-
- ++s1;
- ++s2;
- --n;
- }
-
- if (n == 0)
- return 0;
-
- return (*s1 > *s2 ? 1 : -1);
- }
-
- int wstrncasecmp(char *s1, char *s2, int n)
- {
- int c;
-
- while (c = *s1, tolower(c) == tolower(*s2))
- {
- if (c == '\0')
- return 0;
-
- if (n == 0)
- return 0;
-
- ++s1;
- ++s2;
- --n;
- }
-
- if (n == 0)
- return 0;
-
- return (*s1 > *s2 ? 1 : -1);
- }
-
- Bool wsubstr(char *s1, char *s2)
- {
- int i, len1 = wstrlen(s1), len2 = wstrlen(s2);
-
- for (i = 0; i <= len1 - len2; ++i)
- {
- if (wstrncasecmp(s1+i, s2, len2) == 0)
- return yes;
- }
-
- return no;
- }
-
-
- void outc(uint c, Out *out)
- {
- uint ch;
-
- if (out->encoding == UTF8)
- {
- if (c < 128)
- putc(c, out->fp);
- else if (c <= 0x7FF)
- {
- ch = (0xC0 | (c >> 6)); putc(ch, out->fp);
- ch = (0x80 | (c & 0x3F)); putc(ch, out->fp);
- }
- else if (c <= 0xFFFF)
- {
- ch = (0xE0 | (c >> 12)); putc(ch, out->fp);
- ch = (0x80 | ((c >> 6) & 0x3F)); putc(ch, out->fp);
- ch = (0x80 | (c & 0x3F)); putc(ch, out->fp);
- }
- else if (c <= 0x1FFFFF)
- {
- ch = (0xF0 | (c >> 18)); putc(ch, out->fp);
- ch = (0x80 | ((c >> 12) & 0x3F)); putc(ch, out->fp);
- ch = (0x80 | ((c >> 6) & 0x3F)); putc(ch, out->fp);
- ch = (0x80 | (c & 0x3F)); putc(ch, out->fp);
- }
- else
- {
- ch = (0xF8 | (c >> 24)); putc(ch, out->fp);
- ch = (0x80 | ((c >> 18) & 0x3F)); putc(ch, out->fp);
- ch = (0x80 | ((c >> 12) & 0x3F)); putc(ch, out->fp);
- ch = (0x80 | ((c >> 6) & 0x3F)); putc(ch, out->fp);
- ch = (0x80 | (c & 0x3F)); putc(ch, out->fp);
- }
- }
- else if (out->encoding == ISO2022)
- {
- if (c == 0x1b) /* ESC */
- out->state = FSM_ESC;
- else
- {
- switch (out->state)
- {
- case FSM_ESC:
- if (c == '$')
- out->state = FSM_ESCD;
- else if (c == '(')
- out->state = FSM_ESCP;
- else
- out->state = FSM_ASCII;
- break;
-
- case FSM_ESCD:
- if (c == '(')
- out->state = FSM_ESCDP;
- else
- out->state = FSM_NONASCII;
- break;
-
- case FSM_ESCDP:
- out->state = FSM_NONASCII;
- break;
-
- case FSM_ESCP:
- out->state = FSM_ASCII;
- break;
-
- case FSM_NONASCII:
- c &= 0x7F;
- break;
- }
- }
-
- putc(c, out->fp);
- }
- else
- putc(c, out->fp);
- }
-
- /*
- first time initialization which should
- precede reading the command line
- */
- void InitTidy(void)
- {
- InitMap();
- InitAttrs();
- InitTags();
- InitEntities();
- InitConfig();
-
- totalerrors = totalwarnings = 0;
- XmlTags = XmlOut = HideEndTags = UpperCaseTags =
- MakeClean = writeback = OnlyErrors = no;
-
- errfile = null;
- errout = stderr;
-
- #ifdef CONFIG_FILE
- ParseConfigFile(CONFIG_FILE);
- #endif
- }
-
- /*
- call this when you have finished with tidy
- to free the hash tables and other resources
- */
- void DeInitTidy(void)
- {
- FreeTags();
- FreeAttrTable();
- FreeEntities();
- FreeConfig();
- FreePrintBuf();
- }
-
- int main(int argc, char **argv)
- {
- char *file, *prog;
- FILE *fp = null;
- Node *node;
- Lexer *lexer;
- char *s, c, *arg, *current_errorfile = "stderr";
- int jmpret;
- Out out; /* normal output stream */
-
- /*
- set up for long jump back to here on severe errors
- */
- jmpret = setjmp(error_exit);
-
- /*
- return on a severe error after long jump
- */
-
- if (jmpret != 0)
- {
- /* ensure input is closed */
- if (fp && fp != stdin)
- fclose(fp);
-
- /* 2 signifies a serious error */
- return 2;
- }
-
- InitTidy();
-
- /* look for env var "HTML_TIDY" */
-
- if ((file = getenv("HTML_TIDY")))
- ParseConfigFile(file);
-
- /* read command line */
-
- prog = argv[0];
-
- while (argc > 0)
- {
- if (argc > 1 && argv[1][0] == '-')
- {
- /* support -foo and --foo */
- arg = argv[1] + 1;
-
- if (arg[0] == '-')
- ++arg;
-
- if (strcmp(arg, "indent") == 0)
- IndentContent = yes;
- else if (strcmp(arg, "xml") == 0)
- XmlTags = yes;
- else if (strcmp(arg, "asxml") == 0)
- xHTML = yes;
- else if (strcmp(arg, "indent") == 0)
- IndentContent = yes;
- else if (strcmp(arg, "omit") == 0)
- HideEndTags = yes;
- else if (strcmp(arg, "upper") == 0)
- UpperCaseTags = yes;
- else if (strcmp(arg, "clean") == 0)
- MakeClean = yes;
- else if (strcmp(arg, "raw") == 0)
- CharEncoding = RAW;
- else if (strcmp(arg, "ascii") == 0)
- CharEncoding = ASCII;
- else if (strcmp(arg, "latin1") == 0)
- CharEncoding = LATIN1;
- else if (strcmp(arg, "utf8") == 0)
- CharEncoding = UTF8;
- else if (strcmp(arg, "iso2022") == 0)
- CharEncoding = ISO2022;
- else if (strcmp(arg, "numeric") == 0)
- NumEntities = yes;
- else if (strcmp(arg, "modify") == 0)
- writeback = yes;
- else if (strcmp(arg, "change") == 0) /* obsolete */
- writeback = yes;
- else if (strcmp(arg, "update") == 0) /* obsolete */
- writeback = yes;
- else if (strcmp(arg, "errors") == 0)
- OnlyErrors = yes;
- else if (strcmp(arg, "slides") == 0)
- BurstSlides = yes;
- else if (strcmp(arg, "help") == 0 ||
- argv[1][1] == '?'|| argv[1][1] == 'h')
- {
- HelpText(stdout, prog);
- return 1;
- }
- else if (strcmp(arg, "config") == 0)
- {
- if (argc >= 3)
- {
- ParseConfigFile(argv[2]);
- --argc;
- ++argv;
- }
- }
- else if (strcmp(argv[1], "-file") == 0 ||
- strcmp(argv[1], "--file") == 0 ||
- strcmp(argv[1], "-f") == 0)
- {
- if (argc >= 3)
- {
- /* create copy that can be freed by FreeConfig() */
- errfile = wstrdup(argv[2]);
- --argc;
- ++argv;
- }
- }
- else if (strcmp(argv[1], "-wrap") == 0 ||
- strcmp(argv[1], "--wrap") == 0 ||
- strcmp(argv[1], "-w") == 0)
- {
- if (argc >= 3)
- {
- sscanf(argv[2], "%d", &wraplen);
- --argc;
- ++argv;
- }
- }
- else
- {
- s = argv[1];
-
- while ((c = *++s))
- {
- if (c == 'i')
- IndentContent = yes;
- else if (c == 'o')
- HideEndTags = yes;
- else if (c == 'u')
- UpperCaseTags = yes;
- else if (c == 'c')
- MakeClean = yes;
- else if (c == 'n')
- NumEntities = yes;
- else if (c == 'm')
- writeback = yes;
- else if (c == 'e')
- OnlyErrors = yes;
- else
- UnknownOption(stderr, c);
- }
- }
-
- --argc;
- ++argv;
- continue;
- }
-
- /* ensure config is self-consistent */
- AdjustConfig();
-
- /* user specified error file */
- if (errfile)
- {
- /* is it same as the currently opened file? */
- if (wstrcmp(errfile, current_errorfile) != 0)
- {
- /* no so close previous error file */
-
- if (errout != stderr)
- fclose(errout);
-
- /* and try to open the new error file */
- fp = fopen(errfile, "w");
-
- if (fp != null)
- {
- errout = fp;
- current_errorfile = errfile;
- }
- else /* can't be opened so fall back to stderr */
- {
- errout = stderr;
- current_errorfile = "stderr";
- }
- }
- }
-
- if (argc > 1)
- {
- file = argv[1];
- fp = fopen(file, "r");
- }
- else
- {
- fp = stdin;
- file = "stdin";
- }
-
- if (fp != null)
- {
- lexer = NewLexer(OpenInput(fp));
- lexer->errout = errout;
-
- /*
- store pointer to lexer in input stream
- to allow character encoding errors to be
- reported
- */
- lexer->in->lexer = lexer;
-
- /* Tidy doesn't alter the doctype for generic XML docs */
- if (XmlTags)
- node = ParseXMLDocument(lexer);
- else
- {
- lexer->warnings = 0;
- HelloMessage(errout, release_date, file);
- node = ParseDocument(lexer);
-
- /* replaces i by em and b by strong */
- if (LogicalEmphasis)
- EmFromI(node);
-
- /* replaces presentational markup by style rules */
- if (MakeClean)
- CleanTree(lexer, node);
-
- if (node->content)
- {
- if (xHTML)
- SetXHTMLDocType(lexer, node);
- else
- FixDocType(lexer, node);
- }
-
- /* ensure presence of initial <?XML version="1.0"?> */
- if (XmlOut && XmlPi)
- FixXMLPI(lexer, node);
-
- totalwarnings += lexer->warnings;
- totalerrors += lexer->errors;
-
- if(node->content)
- {
- ReportVersion(errout, file, HTMLVersionName(lexer));
- ReportNumWarnings(errout, lexer);
- }
- }
-
- if (fp != stdin)
- fclose(fp);
-
- MemFree(lexer->in);
-
- if (lexer->errors > 0)
- NeedsAuthorIntervention(errout);
-
- out.state = FSM_ASCII;
- out.encoding = CharEncoding;
-
- if (!OnlyErrors && lexer->errors == 0)
- {
- if (BurstSlides)
- {
- Node *body;
-
- /*
- remove doctype to avoid potential clash with
- markup introduced when bursting into slides
- */
- DiscardDocType(node);
-
- /* slides use transitional features */
- lexer->versions |= VERS_LOOSE;
-
- /* and patch up doctype to match */
- if (xHTML)
- SetXHTMLDocType(lexer, node);
- else
- FixDocType(lexer, node);
-
-
- /* find the body element which may be implicit */
- body = FindBody(node);
-
- if (body)
- {
- ReportNumberOfSlides(errout, CountSlides(body));
- CreateSlides(lexer, node);
- }
- else
- MissingBody(errout);
- }
- else if (writeback && (fp = fopen(file, "w")))
- {
- out.fp = fp;
-
- if (XmlTags)
- PPrintXMLTree(&out, null, 0, lexer, node);
- else
- PPrintTree(&out, null, 0, lexer, node);
-
- PFlushLine(&out, 0);
- fclose(fp);
- }
- else
- {
- out.fp = stdout;
-
- if (XmlTags)
- PPrintXMLTree(&out, null, 0, lexer, node);
- else
- PPrintTree(&out, null, 0, lexer, node);
-
- PFlushLine(&out, 0);
- }
-
- }
-
- ErrorSummary(lexer);
- FreeNode(node);
- FreeLexer(lexer);
- }
- else
- UnknownFile(errout, prog, file);
-
- --argc;
- ++argv;
-
- if (argc <= 1)
- break;
- }
-
- if (totalerrors + totalwarnings > 0)
- GeneralInfo(errout);
-
- if (errout != stderr)
- fclose(errout);
-
- /* called to free hash tables etc. */
- DeInitTidy();
-
- /* return status can be used by scripts */
-
- if (totalerrors > 0)
- return 2;
-
- if (totalwarnings > 0)
- return 1;
-
- /* 0 signifies all is ok */
- return 0;
- }
-
-