Annotation of Amaya/amaya/AHTURLTools.c, revision 1.223
1.7 cvs 1: /*
2: *
1.223 ! vatton 3: * (c) COPYRIGHT INRIA and W3C, 1996-2009
1.7 cvs 4: * Please first read the full copyright statement in file COPYRIGHT.
5: *
6: */
1.9 cvs 7:
1.10 cvs 8: /*
9: * AHTURLTools.c: contains all the functions for testing, manipulating,
1.25 cvs 10: * and normalizing URLs. It also contains a local copy of the libWWW
11: * URL parsing functions.
1.10 cvs 12: *
13: * Authors: J. Kahan, I. Vatton
1.106 cvs 14: * R. Guetari: Windows.
1.10 cvs 15: *
16: */
1.15 cvs 17: #define THOT_EXPORT extern
1.3 cvs 18: #include "amaya.h"
19:
1.8 cvs 20: #include "init_f.h"
21: #include "AHTURLTools_f.h"
22:
1.24 cvs 23: #define MAX_PRINT_URL_LENGTH 50
1.106 cvs 24: typedef struct _HTURI
25: {
1.195 vatton 26: char *access; /* Now known as "scheme" */
27: char *host;
28: char *absolute;
29: char *relative;
30: char *fragment;
1.29 cvs 31: } HTURI;
1.24 cvs 32:
1.175 cvs 33: #ifdef _WINDOWS
1.164 kahan 34: #ifndef PATH_MAX
1.163 cvs 35: #define PATH_MAX MAX_PATH
1.164 kahan 36: #endif
1.155 cvs 37: #define stat _stat
38: #define uint64_t unsigned __int64
39: #define getpid _getpid
1.175 cvs 40: #else /* _WINDOWS */
1.161 kahan 41: #if HAVE_STDINT_H
42: #include <stdint.h>
43: #endif /* HAVE_STDINT_H */
1.175 cvs 44: #endif /* _WINDOWS */
1.28 cvs 45:
46: /*----------------------------------------------------------------------
47: ConvertToLowerCase
48: Converts a string to lowercase.
49: ----------------------------------------------------------------------*/
1.124 vatton 50: void ConvertToLowerCase (char *string)
1.28 cvs 51: {
1.195 vatton 52: int i;
1.93 cvs 53:
1.195 vatton 54: if (!string)
55: return;
1.28 cvs 56:
1.195 vatton 57: for (i = 0; string[i] != EOS; i++)
58: string[i] = tolower (string[i]);
1.28 cvs 59: }
1.22 cvs 60:
1.8 cvs 61: /*----------------------------------------------------------------------
1.75 cvs 62: EscapeChar
63: writes the equivalent escape code of a char in a string
64: ----------------------------------------------------------------------*/
1.109 cvs 65: void EscapeChar (char *string, char c)
1.75 cvs 66: {
1.109 cvs 67: unsigned int i;
68:
1.195 vatton 69: i = (unsigned char) c & 0xFF;
70: sprintf (string, "%02x", i);
1.75 cvs 71: }
72:
73: /*----------------------------------------------------------------------
1.96 cvs 74: UnEscapeChar
75: writes the equivalent hex code to a %xx coded char
76: ----------------------------------------------------------------------*/
1.109 cvs 77: static char UnEscapeChar (char c)
1.96 cvs 78: {
1.195 vatton 79: return c >= '0' && c <= '9' ? c - '0'
80: : c >= 'A' && c <= 'F' ? c - 'A' + 10
81: : c - 'a' + 10; /* accept small letters just in case */
1.96 cvs 82: }
83:
84: /*----------------------------------------------------------------------
1.75 cvs 85: EscapeURL
86: Takes a URL and escapes all protected chars into
87: %xx sequences. Also, removes any leading white spaces
88: Returns either NULL or a new buffer, which must be freed by the caller
89: ----------------------------------------------------------------------*/
1.106 cvs 90: char *EscapeURL (const char *url)
91: {
92: char *buffer;
1.208 vatton 93: char *ptr, *server, *param;
1.215 vatton 94: int buffer_len, par_len = 0;
95: int new_chars, len = 0, url_len;
1.106 cvs 96: int buffer_free_mem;
1.75 cvs 97: void *status;
98:
99: if (url && *url)
100: {
1.208 vatton 101: url_len = strlen (url);
102: buffer_free_mem = url_len + 20;
103: // a patch for sweetwiki
1.215 vatton 104: server = TtaGetEnvString ("WIKI_SERVER");
105: param = TtaGetEnvString ("WIKI_GET_PARAMS");
106: if (server)
107: len = strlen(server);
108: if (len == 0 || param == NULL ||
109: strncmp (url, server, len) ||
1.208 vatton 110: // or already included
1.215 vatton 111: url_len < 4 || strncmp (&url[url_len-4], ".jsp", 4))
1.208 vatton 112: // it's not necessary to add these parameters
113: par_len = 0;
1.215 vatton 114: else
115: par_len = strlen (param);
1.208 vatton 116:
117: buffer_free_mem += par_len;
1.216 vatton 118: buffer = (char *)TtaGetMemory (buffer_free_mem + 2);
1.107 kahan 119: ptr = (char *) url;
1.75 cvs 120: buffer_len = 0;
121: while (*ptr)
122: {
123: switch (*ptr)
124: {
125: /* put here below all the chars that need to
126: be escaped into %xx */
1.81 cvs 127: case 0x20: /* space */
1.195 vatton 128: /*case 0x26:*/ /* & */
1.140 kahan 129: case 0x27: /* antislash */
1.75 cvs 130: new_chars = 3;
131: break;
132:
133: default:
1.195 vatton 134: if ((unsigned char )*ptr > 127)
135: new_chars = 3;
136: else
137: new_chars = 1;
1.75 cvs 138: break;
139: }
140:
141: /* see if we need extra room in the buffer */
142: if (new_chars > buffer_free_mem)
143: {
1.76 cvs 144: buffer_free_mem = 20;
1.106 cvs 145: status = TtaRealloc (buffer, sizeof (char)
1.216 vatton 146: * (buffer_len + buffer_free_mem + 2));
1.75 cvs 147: if (status)
1.114 cvs 148: buffer = (char *) status;
1.106 cvs 149: else
1.195 vatton 150: {
151: /* @@ maybe we should do some other behavior here, like
152: freeing the buffer and return a void thing */
153: buffer[buffer_len] = EOS;
154: break;
155: }
1.75 cvs 156: }
1.195 vatton 157: /* escape the char */
158: if (new_chars == 3)
159: {
1.106 cvs 160: buffer[buffer_len] = '%';
1.75 cvs 161: EscapeChar (&buffer[buffer_len+1], *ptr);
162: }
1.140 kahan 163: else
164: buffer[buffer_len] = *ptr;
165:
166: /* update the status */
167: buffer_len += new_chars;
168: buffer_free_mem -= new_chars;
169: /* examine the next char */
170: ptr++;
1.208 vatton 171:
172: if (*ptr == EOS && par_len)
173: {
174: // add parameters
1.216 vatton 175: if (param[0] != '?' && strstr (url, "?") == NULL)
176: {
177: // add the missing character
178: buffer[buffer_len++] = '?';
179: buffer_free_mem--;
180: }
1.208 vatton 181: ptr = param;
182: par_len = 0;
183: }
1.140 kahan 184: }
185: buffer[buffer_len] = EOS;
186: }
187: else
188: buffer = NULL;
189:
190: return (buffer);
191: }
192:
193: /*----------------------------------------------------------------------
194: EscapeXML
195: Takes a string and escapes all protected chars into entity
196: sequences.
197: Returns either NULL or a new buffer, which must be freed by the caller
198: ----------------------------------------------------------------------*/
199: char *EscapeXML (const char *string)
200: {
201: char *buffer;
202: int buffer_len;
203: int buffer_free_mem;
204: char *ptr;
1.203 vatton 205: const char *entity = NULL;
1.140 kahan 206: int new_chars;
207: void *status;
208:
209: if (string && *string)
210: {
211: buffer_free_mem = strlen (string) + 20;
1.171 gully 212: buffer = (char *)TtaGetMemory (buffer_free_mem + 1);
1.140 kahan 213: ptr = (char *) string;
214: buffer_len = 0;
215:
216: while (*ptr)
217: {
218: switch (*ptr)
219: {
1.195 vatton 220: case 0x26: /* & */
221: entity = "&";
222: new_chars = sizeof (entity) - 1;
223: break;
1.140 kahan 224:
1.195 vatton 225: case '<': /* < */
226: entity = "<";
227: new_chars = sizeof (entity) - 1;
228: break;
229:
230: case '>': /* > */
231: entity = ">";
232: new_chars = sizeof (entity) - 1;
233: break;
234:
235: case '"': /* "e; */
236: entity = ""e;";
237: new_chars = sizeof (entity) - 1;
238: break;
1.140 kahan 239:
240: default:
1.195 vatton 241: new_chars = 1;
1.140 kahan 242: break;
243: }
244:
245: /* see if we need extra room in the buffer */
246: if (new_chars > buffer_free_mem)
247: {
248: buffer_free_mem = 20;
249: status = TtaRealloc (buffer, sizeof (char)
1.195 vatton 250: * (buffer_len + buffer_free_mem + 1));
1.140 kahan 251: if (status)
252: buffer = (char *) status;
253: else
1.195 vatton 254: {
255: /* @@ maybe we should do some other behavior here, like
256: freeing the buffer and return a void thing */
257: buffer[buffer_len] = EOS;
258: break;
259: }
260: }
261: /* escape the char */
262: if (entity)
263: {
264: sprintf (&buffer[buffer_len], "%s", entity);
265: entity = NULL;
1.140 kahan 266: }
1.75 cvs 267: else
268: buffer[buffer_len] = *ptr;
269:
270: /* update the status */
271: buffer_len += new_chars;
272: buffer_free_mem -= new_chars;
273: /* examine the next char */
274: ptr++;
275: }
1.106 cvs 276: buffer[buffer_len] = EOS;
1.75 cvs 277: }
1.76 cvs 278: else
279: buffer = NULL;
280:
1.75 cvs 281: return (buffer);
1.122 kahan 282: }
283:
1.75 cvs 284:
285: /*----------------------------------------------------------------------
1.11 cvs 286: ExplodeURL
1.8 cvs 287: ----------------------------------------------------------------------*/
1.106 cvs 288: void ExplodeURL (char *url, char **proto, char **host, char **dir,
1.195 vatton 289: char **file)
1.8 cvs 290: {
1.195 vatton 291: char *curr, *temp;
292: char used_sep;
1.32 cvs 293:
1.195 vatton 294: if (url && strchr (url, URL_SEP))
295: used_sep = URL_SEP;
296: else
297: used_sep = DIR_SEP;
298:
299: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
300: (dir == NULL) || (file == NULL))
301: return;
1.8 cvs 302:
1.195 vatton 303: /* initialize every pointer */
304: *proto = *host = *dir = *file = NULL;
305:
306: /* skip any leading space */
307: while ((*url == SPACE) || (*url == TAB))
308: url++;
309: curr = url;
310: if (*curr == 0)
311: goto finished;
312:
313: /* go to the end of the URL */
314: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
315: (*curr != __CR__) && (*curr != EOL))
316: curr++;
317:
318: /* mark the end of the chain */
319: *curr = EOS;
320: curr--;
321: if (curr <= url)
322: goto finished;
323:
324: /* search the next DIR_SEP indicating the beginning of the file name */
325: do
326: curr--;
327: while ((curr >= url) && (*curr != used_sep));
328:
329: if (curr < url)
330: goto finished;
331: *file = curr + 1;
332:
333: /* mark the end of the dir */
334: *curr = EOS;
335: curr--;
336: if (curr < url)
337: goto finished;
338:
339: /* search for the DIR_STR indicating the host name start */
340: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
341: curr--;
342:
343: /* if we found it, separate the host name from the directory */
344: if ((*curr == used_sep) && (*(curr + 1) == used_sep))
345: {
346: *host = temp = curr + 2;
347: while ((*temp != 0) && (*temp != used_sep))
348: temp++;
349: if (*temp == used_sep)
350: {
351: *temp = EOS;
352: *dir = temp + 1;
353: }
354: }
355: else
356: *dir = curr;
1.8 cvs 357:
1.195 vatton 358: if (curr <= url)
359: goto finished;
1.8 cvs 360:
1.195 vatton 361: /* mark the end of the proto */
362: *curr = EOS;
363: curr--;
364: if (curr < url)
365: goto finished;
1.8 cvs 366:
1.195 vatton 367: if (*curr == ':')
368: {
369: *curr = EOS;
1.9 cvs 370: curr--;
1.195 vatton 371: }
372: else
373: goto finished;
1.8 cvs 374:
1.195 vatton 375: if (curr < url)
376: goto finished;
377: while ((curr > url) && (isalpha (*curr)))
378: curr--;
379: *proto = curr;
1.8 cvs 380:
381: finished:;
382:
383: #ifdef AMAYA_DEBUG
1.195 vatton 384: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
385: if (*proto)
386: fprintf (stderr, "proto : %s, ", *proto);
387: if (*host)
388: fprintf (stderr, "host : %s, ", *host);
389: if (*dir)
390: fprintf (stderr, "dir : %s, ", *dir);
391: if (*file)
392: fprintf (stderr, "file : %s ", *file);
393: fprintf (stderr, "\n");
1.8 cvs 394: #endif
395:
396: }
1.3 cvs 397:
1.116 kahan 398: /*----------------------------------------------------------------------
1.195 vatton 399: PicTypeToMime
400: Converts a Thot PicType into the equivalent MIME type. If no convertion
401: is possible, it returns NULL.
1.116 kahan 402: ----------------------------------------------------------------------*/
1.203 vatton 403: const char *PicTypeToMIME (PicType contentType)
1.116 kahan 404: {
1.203 vatton 405: const char *mime_type;
1.116 kahan 406:
407: switch (contentType)
408: {
1.179 vatton 409: case xbm_type:
410: mime_type ="image/x-xbitmap";
411: break;
1.116 kahan 412: case eps_type:
413: mime_type ="application/postscript";
414: break;
1.195 vatton 415: case xpm_type:
1.116 kahan 416: mime_type ="image/x-xpicmap";
1.195 vatton 417: break;
1.116 kahan 418: case gif_type:
419: mime_type ="image/gif";
420: break;
421: case jpeg_type:
422: mime_type ="image/jpeg";
423: break;
424: case png_type:
425: mime_type ="image/png";
426: break;
427: case svg_type:
1.165 cvs 428: mime_type = AM_SVG_MIME_TYPE;
429: break;
430: case html_type:
431: mime_type = AM_XHTML_MIME_TYPE;
432: break;
433: case mathml_type:
434: mime_type = AM_MATHML_MIME_TYPE;
1.116 kahan 435: break;
1.195 vatton 436: case unknown_type:
437: default:
438: mime_type = NULL;
439: }
1.116 kahan 440:
441: return mime_type;
442: }
1.61 cvs 443:
444: /*----------------------------------------------------------------------
1.195 vatton 445: ImageElement
446: Returns the element (image parameter) and URL (url parameter) of an
447: image in a docImage document. The user must free the memory associated
448: with the url parameter if the function is succesful.
449: If the url parameter is NULL, we won't initialize it.
450: Returns TRUE if succesful, FALSE otherwise.
1.117 kahan 451: ----------------------------------------------------------------------*/
452: ThotBool ImageElement (Document doc, char **url, Element *image)
453: {
454: Element el, imgEl;
455: Attribute attr, srcAttr;
456: AttributeType attrType;
457: int length;
1.176 vatton 458: char *utf8value;
1.117 kahan 459:
460: if (DocumentTypes[doc] != docImage)
461: return FALSE;
462:
463: /* find the value of the src attribute */
464: attrType.AttrSSchema = TtaGetSSchema ("HTML", doc);
465: attrType.AttrTypeNum = HTML_ATTR_SRC;
466: el = TtaGetRootElement (doc);
467: TtaSearchAttribute (attrType, SearchInTree, el, &imgEl, &srcAttr);
468:
469: if (!imgEl)
470: return FALSE;
471: *image = imgEl;
472:
1.120 kahan 473: if (url)
474: {
475: attr = TtaGetAttribute (imgEl, attrType);
476: length = TtaGetTextAttributeLength (srcAttr) + 1;
1.176 vatton 477: utf8value = (char *)TtaGetMemory (length);
478: TtaGiveTextAttributeValue (srcAttr, utf8value, &length);
479: *url = (char *)TtaConvertMbsToByte ((unsigned char *)utf8value,
1.195 vatton 480: TtaGetDefaultCharset ());
1.176 vatton 481: TtaFreeMemory (utf8value);
1.120 kahan 482: }
1.117 kahan 483: return TRUE;
484: }
485:
486: /*----------------------------------------------------------------------
1.195 vatton 487: DocImageMimeType
488: Returns the MIME type of a docImage document.
1.117 kahan 489: ----------------------------------------------------------------------*/
1.203 vatton 490: const char *DocImageMimeType (Document doc)
1.117 kahan 491: {
1.203 vatton 492: const char *mime_type;
1.117 kahan 493: LoadedImageDesc *pImage;
494: PicType type;
495: Element image;
496:
497: if (DocumentTypes[doc] != docImage)
498: return NULL;
499:
500: mime_type = NULL;
501: if (!IsHTTPPath (DocumentURLs[doc]))
502: {
503: /* it is a local image */
1.120 kahan 504: if (ImageElement (doc, NULL, &image))
1.195 vatton 505: {
506: type = TtaGetPictureType (image);
507: mime_type = PicTypeToMIME (type);
508: }
1.117 kahan 509: }
510: else
511: {
512: /* find the value of the src attribute */
513: pImage = ImageURLs;
514: while (pImage != NULL)
1.195 vatton 515: {
516: if (pImage->document == doc)
517: {
518: if (pImage->content_type)
519: mime_type = pImage->content_type;
520: else if (pImage->elImage && pImage->elImage->currentElement)
521: {
522: type = TtaGetPictureType (pImage->elImage->currentElement);
523: mime_type = PicTypeToMIME (type);
524: }
525: break;
526: }
527: }
1.117 kahan 528: }
529: return (mime_type);
530: }
531:
1.4 cvs 532: /*----------------------------------------------------------------------
1.9 cvs 533: IsHTMLName
534: returns TRUE if path points to an HTML resource.
1.4 cvs 535: ----------------------------------------------------------------------*/
1.109 cvs 536: ThotBool IsHTMLName (const char *path)
1.106 cvs 537: {
1.183 quint 538: char *temppath;
539: char *suffix;
1.136 cvs 540: char nsuffix[MAX_LENGTH];
541: int i;
1.5 cvs 542:
1.101 cvs 543: if (!path)
544: return (FALSE);
1.5 cvs 545:
1.183 quint 546: temppath = TtaStrdup ((char *)path);
547: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 548: TtaExtractSuffix (temppath, suffix);
1.101 cvs 549: i = 0;
1.106 cvs 550: while (suffix[i] != EOS)
1.101 cvs 551: {
552: /* Normalize the suffix */
553: i = 0;
1.106 cvs 554: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.195 vatton 555: {
556: nsuffix[i] = tolower (suffix[i]);
557: i++;
558: }
1.106 cvs 559: nsuffix[i] = EOS;
560: if (!strcmp (nsuffix, "html") ||
1.195 vatton 561: !strcmp (nsuffix, "htm") ||
562: !strcmp (nsuffix, "shtml") ||
1.197 vatton 563: !strcmp (nsuffix, "asp") ||
1.195 vatton 564: !strcmp (nsuffix, "jsp") ||
565: !strcmp (nsuffix, "tpl") ||
566: !strcmp (nsuffix, "xht") ||
567: !strcmp (nsuffix, "xhtm") ||
568: !strcmp (nsuffix, "lhtml") ||
569: !strcmp (nsuffix, "xhtml"))
570: {
571: TtaFreeMemory (temppath);
572: TtaFreeMemory (suffix);
573: return (TRUE);
574: }
1.106 cvs 575: else if (!strcmp (nsuffix, "gz"))
1.195 vatton 576: {
577: /* take into account compressed files */
578: TtaExtractSuffix (temppath, suffix);
579: /* Normalize the suffix */
580: i = 0;
581: while (suffix[i] != EOS && i < MAX_LENGTH -1)
582: {
583: nsuffix[i] = tolower (suffix[i]);
584: i++;
585: }
586: nsuffix[i] = EOS;
587: TtaFreeMemory (temppath);
588: TtaFreeMemory (suffix);
589: if (!strcmp (nsuffix, "html") ||
590: !strcmp (nsuffix, "htm") ||
591: !strcmp (nsuffix, "shtml") ||
1.197 vatton 592: !strcmp (nsuffix, "asp") ||
1.195 vatton 593: !strcmp (nsuffix, "jsp") ||
594: !strcmp (nsuffix, "tpl") ||
595: !strcmp (nsuffix, "xht") ||
596: !strcmp (nsuffix, "xhtm") ||
597: !strcmp (nsuffix, "lhtml") ||
598: !strcmp (nsuffix, "xhtml"))
599: return (TRUE);
600: else
601: return (FALSE);
602: }
1.101 cvs 603: else
1.195 vatton 604: /* check if there is another suffix */
605: TtaExtractSuffix (temppath, suffix);
1.101 cvs 606: }
1.183 quint 607: TtaFreeMemory (temppath);
608: TtaFreeMemory (suffix);
609: return (FALSE);
1.3 cvs 610: }
611:
1.4 cvs 612: /*----------------------------------------------------------------------
1.136 cvs 613: IsMathMLName
614: returns TRUE if path points to an MathML resource.
1.56 cvs 615: ----------------------------------------------------------------------*/
1.136 cvs 616: ThotBool IsMathMLName (const char *path)
1.56 cvs 617: {
1.195 vatton 618: char *temppath;
619: char *suffix;
620: ThotBool ret;
621:
622: if (!path)
623: return (FALSE);
624:
625: temppath = TtaStrdup ((char *)path);
626: suffix = (char *)TtaGetMemory (strlen (path) + 1);
627: TtaExtractSuffix (temppath, suffix);
628:
629: if (!strcasecmp (suffix, "mml"))
630: ret = TRUE;
631: else if (!strcmp (suffix, "gz"))
632: {
633: /* take into account compressed files */
634: TtaExtractSuffix (temppath, suffix);
635: if (!strcasecmp (suffix, "mml"))
636: ret = TRUE;
637: else
638: ret = FALSE;
639: }
640: else
641: ret = FALSE;
1.183 quint 642:
643: TtaFreeMemory (temppath);
644: TtaFreeMemory (suffix);
645: return (ret);
1.60 cvs 646: }
647:
648: /*----------------------------------------------------------------------
1.136 cvs 649: IsSVGName
650: returns TRUE if path points to an SVG resource.
1.133 vatton 651: ----------------------------------------------------------------------*/
1.136 cvs 652: ThotBool IsSVGName (const char *path)
1.133 vatton 653: {
1.195 vatton 654: char *temppath;
655: char *suffix;
656: ThotBool ret;
657:
658: if (!path)
659: return (FALSE);
660:
661: temppath = TtaStrdup ((char *)path);
662: suffix = (char *)TtaGetMemory (strlen (path) + 1);
663: TtaExtractSuffix (temppath, suffix);
664:
665: if (!strcasecmp (suffix, "svg") || !strcasecmp (suffix, "svgz"))
666: ret = TRUE;
667: else if (!strcmp (suffix, "gz"))
668: {
669: /* take into account compressed files */
670: TtaExtractSuffix (temppath, suffix);
671: if (!strcasecmp (suffix, "svg"))
672: ret = TRUE;
673: else
674: ret = FALSE;
675: }
676: else
677: ret = FALSE;
1.183 quint 678:
679: TtaFreeMemory (temppath);
680: TtaFreeMemory (suffix);
681: return (ret);
1.133 vatton 682: }
683:
684: /*----------------------------------------------------------------------
1.136 cvs 685: IsXMLName
686: returns TRUE if path points to an XML resource.
1.103 cvs 687: ----------------------------------------------------------------------*/
1.136 cvs 688: ThotBool IsXMLName (const char *path)
1.103 cvs 689: {
1.195 vatton 690: char *temppath;
691: char *suffix;
692: ThotBool ret;
693:
694: if (!path)
695: return (FALSE);
696:
697: temppath = TtaStrdup ((char *)path);
698: suffix = (char *)TtaGetMemory (strlen (path) + 1);
699: TtaExtractSuffix (temppath, suffix);
700:
701: if (!strcasecmp (suffix, "xml") ||
702: !strcasecmp (suffix, "xht") ||
1.197 vatton 703: !strcasecmp (suffix, "xtd") ||
1.200 kia 704: !strcasecmp (suffix, "xtl") ||
1.195 vatton 705: !strcmp (suffix, "xhtm") ||
706: !strcmp (suffix, "xhtml") ||
707: !strcmp (suffix, "smi") ||
1.201 vatton 708: !strcmp (suffix, "xsd") |
709: !strcmp (suffix, "xslt") |
710: !strcmp (suffix, "xsl") |
1.195 vatton 711: !strcmp (suffix, "zsl"))
712: ret = TRUE;
713: else if (!strcmp (suffix, "gz"))
714: {
715: /* take into account compressed files */
716: TtaExtractSuffix (temppath, suffix);
717: if (!strcasecmp (suffix, "xml") ||
718: !strcasecmp (suffix, "xht") ||
1.197 vatton 719: !strcasecmp (suffix, "xtd") ||
1.200 kia 720: !strcasecmp (suffix, "xtl") ||
1.195 vatton 721: !strcmp (suffix, "xhtm") ||
722: !strcmp (suffix, "xhtml") ||
1.201 vatton 723: !strcmp (suffix, "xsd") |
724: !strcmp (suffix, "xslt") |
725: !strcmp (suffix, "xsl") |
726: !strcmp (suffix, "smi"))
727: ret = TRUE;
728: else
729: ret = FALSE;
730: }
731: else
732: ret = FALSE;
733:
734: TtaFreeMemory (temppath);
735: TtaFreeMemory (suffix);
736: return (ret);
737: }
738:
739: /*----------------------------------------------------------------------
740: IsXMLStruct
741: returns TRUE if path points to an XML transformation or schema.
742: ----------------------------------------------------------------------*/
743: ThotBool IsXMLStruct (const char *path)
744: {
745: char *temppath;
746: char *suffix;
747: ThotBool ret;
748:
749: if (!path)
750: return (FALSE);
751:
752: temppath = TtaStrdup ((char *)path);
753: suffix = (char *)TtaGetMemory (strlen (path) + 1);
754: TtaExtractSuffix (temppath, suffix);
755:
756: if (!strcmp (suffix, "xsd") |
757: !strcmp (suffix, "xslt") |
758: !strcmp (suffix, "xsl"))
759: ret = TRUE;
760: else if (!strcmp (suffix, "gz"))
761: {
762: /* take into account compressed files */
763: TtaExtractSuffix (temppath, suffix);
764: if (!strcmp (suffix, "xsd") |
765: !strcmp (suffix, "xslt") |
1.195 vatton 766: !strcmp (suffix, "xsl"))
767: ret = TRUE;
768: else
769: ret = FALSE;
770: }
771: else
772: ret = FALSE;
1.183 quint 773:
774: TtaFreeMemory (temppath);
775: TtaFreeMemory (suffix);
776: return (ret);
1.103 cvs 777: }
778:
779: /*----------------------------------------------------------------------
1.207 kia 780: IsXTigerLibrary
781: returns TRUE if path points to an XTiger resource.
782: ----------------------------------------------------------------------*/
783: ThotBool IsXTigerLibrary (const char *path)
784: {
785: char *temppath;
786: char *suffix;
787: ThotBool ret;
788:
789: if (!path)
790: return (FALSE);
791:
792: temppath = TtaStrdup ((char *)path);
793: suffix = (char *)TtaGetMemory (strlen (path) + 1);
794: TtaExtractSuffix (temppath, suffix);
795:
796: if (!strcasecmp (suffix, "xtl"))
797: ret = TRUE;
798: else if (!strcmp (suffix, "gz"))
799: {
800: /* take into account compressed files */
801: TtaExtractSuffix (temppath, suffix);
802: if (!strcasecmp (suffix, "xtl"))
803: ret = TRUE;
804: else
805: ret = FALSE;
806: }
807: else
808: ret = FALSE;
809:
810: TtaFreeMemory (temppath);
811: TtaFreeMemory (suffix);
812: return (ret);
813: }
814:
815: /*----------------------------------------------------------------------
816: IsXTiger
1.199 vatton 817: returns TRUE if path points to an XTiger resource.
818: ----------------------------------------------------------------------*/
819: ThotBool IsXTiger (const char *path)
820: {
821: char *temppath;
822: char *suffix;
823: ThotBool ret;
824:
825: if (!path)
826: return (FALSE);
827:
828: temppath = TtaStrdup ((char *)path);
829: suffix = (char *)TtaGetMemory (strlen (path) + 1);
830: TtaExtractSuffix (temppath, suffix);
831:
1.207 kia 832: if (!strcasecmp (suffix, "xtd"))
1.199 vatton 833: ret = TRUE;
834: else if (!strcmp (suffix, "gz"))
835: {
836: /* take into account compressed files */
837: TtaExtractSuffix (temppath, suffix);
1.207 kia 838: if (!strcasecmp (suffix, "xtd"))
1.199 vatton 839: ret = TRUE;
840: else
841: ret = FALSE;
842: }
843: else
844: ret = FALSE;
845:
846: TtaFreeMemory (temppath);
847: TtaFreeMemory (suffix);
848: return (ret);
849: }
850:
851: /*----------------------------------------------------------------------
1.223 ! vatton 852: IsResourceName
! 853: returns TRUE if path points to an undisplayed resource.
! 854: ----------------------------------------------------------------------*/
! 855: ThotBool IsResourceName (const char *path)
! 856: {
! 857: char *temppath, *suffix = NULL, *ptr = NULL;
! 858: ThotBool ret = FALSE;
! 859:
! 860: temppath = TtaStrdup ((char *)path);
! 861: if (temppath == NULL)
! 862: return FALSE;
! 863:
! 864: if (!strncmp (temppath, "http://", 7))
! 865: ptr = &temppath[7];
! 866: else if (!strncmp (temppath, "http://", 8))
! 867: ptr = &temppath[8];
! 868: else
! 869: ptr = temppath;
! 870: if (ptr != temppath)
! 871: {
! 872: // skip the host name
! 873: while (*ptr != EOS && *ptr != '/')
! 874: ptr++;
! 875: }
! 876: if (*ptr == EOS)
! 877: // no name
! 878: return FALSE;
! 879:
! 880: suffix = (char *)TtaGetMemory (strlen (ptr) + 1);
! 881: TtaExtractSuffix (ptr, suffix);
! 882: if (*suffix == EOS || *suffix == '/' ||
! 883: !strncasecmp (suffix, "htm", 3) ||
! 884: !strncasecmp (suffix,"xhtm", 4) ||
! 885: !strncasecmp (suffix, "php", 3) ||
! 886: !strncasecmp (suffix, "mml", 3) ||
! 887: !strncasecmp (suffix, "svg", 3) ||
! 888: !strncasecmp (suffix, "xml", 3))
! 889: ret = FALSE;
! 890: else
! 891: ret = TRUE;
! 892:
! 893: TtaFreeMemory (temppath);
! 894: TtaFreeMemory (suffix);
! 895: return ret;
! 896: }
! 897:
! 898: /*----------------------------------------------------------------------
! 899: IsUndisplayedName
1.136 cvs 900: returns TRUE if path points to an undisplayed resource.
1.103 cvs 901: ----------------------------------------------------------------------*/
1.136 cvs 902: ThotBool IsUndisplayedName (const char *path)
1.103 cvs 903: {
1.195 vatton 904: char *temppath;
905: char *suffix;
906: ThotBool ret;
907:
908: if (!path)
909: return (FALSE);
910:
911: temppath = TtaStrdup ((char *)path);
912: suffix = (char *)TtaGetMemory (strlen (path) + 1);
913: TtaExtractSuffix (temppath, suffix);
914:
915: if (!strcasecmp (suffix, "exe") ||
916: !strcasecmp (suffix, "zip") ||
917: !strcasecmp (suffix, "ppt") ||
918: !strcasecmp (suffix, "pdf") ||
919: !strcasecmp (suffix, "ps") ||
920: !strcasecmp (suffix, "eps") ||
921: !strcasecmp (suffix, "tar") ||
922: !strcasecmp (suffix, "tgz") ||
923: !strcasecmp (suffix, "ddl") ||
1.196 vatton 924: !strcasecmp (suffix, "deb") ||
925: !strcasecmp (suffix, "dmg") ||
1.195 vatton 926: !strcasecmp (suffix, "mpg") ||
927: !strcasecmp (suffix, "mpeg") ||
1.196 vatton 928: !strcasecmp (suffix, "rpm") ||
1.195 vatton 929: !strcasecmp (suffix, "wmv") ||
930: !strcasecmp (suffix, "wma") ||
1.223 ! vatton 931: !strcasecmp (suffix, "doc") ||
! 932: !strcasecmp (suffix, "odt") ||
1.195 vatton 933: !strcasecmp (suffix, "o"))
934: ret = TRUE;
935: else if (!strcmp (suffix, "gz"))
936: {
937: /* take into account compressed files */
938: TtaExtractSuffix (temppath, suffix);
939: if (!strcasecmp (suffix, "exe") ||
940: !strcasecmp (suffix, "zip") ||
941: !strcasecmp (suffix, "ppt") ||
942: !strcasecmp (suffix, "pdf") ||
943: !strcasecmp (suffix, "ps") ||
944: !strcasecmp (suffix, "eps") ||
945: !strcasecmp (suffix, "tar") ||
946: !strcasecmp (suffix, "ddl") ||
1.196 vatton 947: !strcasecmp (suffix, "dmg") ||
1.195 vatton 948: !strcasecmp (suffix, "mpg") ||
949: !strcasecmp (suffix, "mpeg") ||
950: !strcasecmp (suffix, "wmv") ||
951: !strcasecmp (suffix, "wma") ||
1.223 ! vatton 952: !strcasecmp (suffix, "doc") ||
! 953: !strcasecmp (suffix, "odt") ||
1.195 vatton 954: !strcasecmp (suffix, "o"))
955: ret = TRUE;
956: else
957: ret = FALSE;
958: }
959: else
960: ret = FALSE;
961:
962: TtaFreeMemory (temppath);
963: TtaFreeMemory (suffix);
964: return (ret);
1.103 cvs 965: }
966:
967: /*----------------------------------------------------------------------
1.60 cvs 968: IsCSSName
969: returns TRUE if path points to an XML resource.
970: ----------------------------------------------------------------------*/
1.111 cvs 971: ThotBool IsCSSName (const char *path)
1.60 cvs 972: {
1.195 vatton 973: char *temppath;
974: char *suffix;
975: ThotBool ret;
976:
977: if (!path)
978: return (FALSE);
979:
980: temppath = TtaStrdup ((char *)path);
981: suffix = (char *)TtaGetMemory (strlen (path) + 1);
982: TtaExtractSuffix (temppath, suffix);
983:
984: if (!strcasecmp (suffix, "css"))
985: ret = TRUE;
986: else if (!strcmp (suffix, "gz"))
987: {
988: /* take into account compressed files */
989: TtaExtractSuffix (temppath, suffix);
990: if (!strcasecmp (suffix, "css"))
991: ret = TRUE;
992: else
993: ret = FALSE;
994: }
995: else
996: ret = FALSE;
997:
998: TtaFreeMemory (temppath);
999: TtaFreeMemory (suffix);
1000: return (ret);
1.56 cvs 1001: }
1002:
1003: /*----------------------------------------------------------------------
1.177 kahan 1004: MultipleBookmarks
1005: returns the value of the multiple bookmarks environment variable.
1006: (TRUE if enabled).
1007: ----------------------------------------------------------------------*/
1008: ThotBool MultipleBookmarks (void)
1009: {
1010: ThotBool multiple_bm;
1011:
1.180 kahan 1012: TtaGetEnvBoolean ("DISABLE_MULTIPLE_BM", &multiple_bm);
1.177 kahan 1013:
1.180 kahan 1014: return (multiple_bm == FALSE);
1.177 kahan 1015: }
1016:
1017: /*----------------------------------------------------------------------
1.172 kahan 1018: IsRDFName
1019: returns TRUE if path points to an RDF resource.
1020: ----------------------------------------------------------------------*/
1021: ThotBool IsRDFName (const char *path)
1022: {
1.195 vatton 1023: char *temppath;
1024: char *suffix;
1025: ThotBool ret;
1026:
1027: /* temporarily disabling this function */
1028: if (!MultipleBookmarks ())
1029: return (FALSE);
1030:
1031: if (!path)
1032: return (FALSE);
1033:
1034: temppath = TtaStrdup ((char *)path);
1035: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1036: TtaExtractSuffix (temppath, suffix);
1037:
1038: if (!strcasecmp (suffix, "rdf"))
1039: ret = TRUE;
1040: else if (!strcmp (suffix, "gz"))
1041: {
1042: /* take into account compressed files */
1043: TtaExtractSuffix (temppath, suffix);
1044: if (!strcasecmp (suffix, "rdf"))
1045: ret = TRUE;
1046: else
1047: ret = FALSE;
1048: }
1049: else
1050: ret = FALSE;
1.183 quint 1051:
1.195 vatton 1052: TtaFreeMemory (temppath);
1053: TtaFreeMemory (suffix);
1054: return (ret);
1.172 kahan 1055: }
1056:
1057: /*----------------------------------------------------------------------
1.9 cvs 1058: IsImageName
1059: returns TRUE if path points to an image resource.
1.4 cvs 1060: ----------------------------------------------------------------------*/
1.111 cvs 1061: ThotBool IsImageName (const char *path)
1.106 cvs 1062: {
1.195 vatton 1063: char *temppath;
1064: char *suffix;
1065: char nsuffix[MAX_LENGTH];
1066: int i;
1067: ThotBool ret;
1068:
1069: if (!path)
1070: return (FALSE);
1071:
1072: temppath = TtaStrdup ((char *)path);
1073: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1074: TtaExtractSuffix (temppath, suffix);
1075:
1076: /* Normalize the suffix */
1077: i = 0;
1078: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1079: {
1080: nsuffix[i] = tolower (suffix[i]);
1081: i++;
1082: }
1083: nsuffix[i] = EOS;
1084: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
1085: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
1086: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1087: ret = TRUE;
1088: else
1089: ret = FALSE;
1090:
1091: TtaFreeMemory (temppath);
1092: TtaFreeMemory (suffix);
1093: return (ret);
1.3 cvs 1094: }
1095:
1.4 cvs 1096: /*----------------------------------------------------------------------
1.58 cvs 1097: IsImageType
1098: returns TRUE if type points to an image resource.
1099: ----------------------------------------------------------------------*/
1.111 cvs 1100: ThotBool IsImageType (const char *type)
1.58 cvs 1101: {
1.195 vatton 1102: char *temptype;
1103: int i;
1104: ThotBool ret;
1105:
1106: if (!type)
1107: return (FALSE);
1108:
1109: temptype = TtaStrdup ((char *)type);
1110: /* Normalize the type */
1111: i = 0;
1112: while (temptype[i] != EOS)
1113: {
1114: temptype[i] = tolower (temptype[i]);
1115: i++;
1116: }
1.166 vatton 1117: if (!strncmp (temptype, "image/", sizeof ("image/") - 1))
1.195 vatton 1118: i = sizeof ("image/") - 1;
1119: else
1120: i = 0;
1121: if (!strcmp (&temptype[i], "gif") ||
1122: !strcmp (&temptype[i], "x-xbitmap") ||
1123: !strcmp (&temptype[i], "x-xpixmap") ||
1124: !strcmp (&temptype[i], "jpeg") ||
1125: !strcmp (&temptype[i], "png"))
1126: ret = TRUE;
1127: else
1128: ret = FALSE;
1129: TtaFreeMemory (temptype);
1130: return (ret);
1.58 cvs 1131: }
1132:
1133: /*----------------------------------------------------------------------
1.9 cvs 1134: IsTextName
1.4 cvs 1135: ----------------------------------------------------------------------*/
1.111 cvs 1136: ThotBool IsTextName (const char *path)
1.106 cvs 1137: {
1.195 vatton 1138: char *temppath;
1139: char *suffix;
1140: char nsuffix[MAX_LENGTH];
1141: int i;
1142: ThotBool ret;
1143:
1144: if (!path)
1145: return (FALSE);
1146:
1147: temppath = TtaStrdup ((char *)path);
1148: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1149: TtaExtractSuffix (temppath, suffix);
1150:
1151: /* Normalize the suffix */
1152: i = 0;
1153: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1154: {
1155: nsuffix[i] = tolower (suffix[i]);
1156: i++;
1157: }
1158: nsuffix[i] = EOS;
1159:
1160: if (!strcmp (nsuffix, "txt") ||
1161: !strcmp (nsuffix, "dtd"))
1162: ret = TRUE;
1163: else if (!strcmp (nsuffix, "gz"))
1164: {
1165: /* take into account compressed files */
1166: TtaExtractSuffix (temppath, suffix);
1167: /* Normalize the suffix */
1168: i = 0;
1169: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1170: {
1171: nsuffix[i] = tolower (suffix[i]);
1172: i++;
1173: }
1174: nsuffix[i] = EOS;
1175: if (!strcmp (nsuffix, "txt") ||
1176: !strcmp (nsuffix, "dtd"))
1177: ret = TRUE;
1178: else
1179: ret = FALSE;
1180: }
1181: else
1182: ret = FALSE;
1183:
1184: TtaFreeMemory (temppath);
1185: TtaFreeMemory (suffix);
1186: return (ret);
1.3 cvs 1187: }
1188:
1.4 cvs 1189: /*----------------------------------------------------------------------
1.9 cvs 1190: IsHTTPPath
1191: returns TRUE if path is in fact an http URL.
1.4 cvs 1192: ----------------------------------------------------------------------*/
1.112 cvs 1193: ThotBool IsHTTPPath (const char *path)
1.3 cvs 1194: {
1.195 vatton 1195: if (!path)
1196: return FALSE;
1.3 cvs 1197:
1.195 vatton 1198: if ((!strncmp (path, "http:", 5) != 0)
1.204 vatton 1199: || !strncmp (path, "ftp:", 4)
1.195 vatton 1200: || !strncmp (path, "internal:", 9))
1201: return TRUE;
1202: return FALSE;
1.3 cvs 1203: }
1204:
1.4 cvs 1205: /*----------------------------------------------------------------------
1.9 cvs 1206: IsWithParameters
1207: returns TRUE if url has a concatenated query string.
1.4 cvs 1208: ----------------------------------------------------------------------*/
1.133 vatton 1209: ThotBool IsWithParameters (const char *url)
1.3 cvs 1210: {
1.195 vatton 1211: int i;
1212:
1213: if ((!url) || (url[0] == EOS))
1214: return FALSE;
1.3 cvs 1215:
1.195 vatton 1216: i = strlen (url) - 1;
1217: while (i > 0 && url[i--] != '?')
1218: if (i < 0)
1.5 cvs 1219: return FALSE;
1.3 cvs 1220:
1.195 vatton 1221: /* There is a parameter */
1222: return TRUE;
1.3 cvs 1223: }
1224:
1.4 cvs 1225: /*----------------------------------------------------------------------
1.9 cvs 1226: IsW3Path
1227: returns TRUE if path is in fact a URL.
1.4 cvs 1228: ----------------------------------------------------------------------*/
1.133 vatton 1229: ThotBool IsW3Path (const char *path)
1.106 cvs 1230: {
1.170 quint 1231: if (path == NULL)
1232: return FALSE;
1.106 cvs 1233: if (strncmp (path, "http:", 5) &&
1234: strncmp (path, "ftp:", 4) &&
1235: strncmp (path, "telnet:", 7) &&
1236: strncmp (path, "wais:", 5) &&
1237: strncmp (path, "news:", 5) &&
1238: strncmp (path, "gopher:", 7) &&
1239: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 1240: strncmp (path, "archie:", 7) &&
1241: strncmp (path, "https:", 6))
1.72 cvs 1242: return FALSE;
1243: return TRUE;
1.3 cvs 1244: }
1245:
1.4 cvs 1246: /*----------------------------------------------------------------------
1.90 cvs 1247: IsFilePath
1248: returns TRUE if path is in fact a URL.
1249: ----------------------------------------------------------------------*/
1.133 vatton 1250: ThotBool IsFilePath (const char *path)
1.90 cvs 1251: {
1.106 cvs 1252: if (strncmp (path, "file:", 5))
1.90 cvs 1253: return FALSE;
1254: return TRUE;
1255: }
1256:
1257: /*----------------------------------------------------------------------
1.9 cvs 1258: IsValidProtocol
1259: returns true if the url protocol is supported by Amaya.
1.4 cvs 1260: ----------------------------------------------------------------------*/
1.133 vatton 1261: ThotBool IsValidProtocol (const char *url)
1.106 cvs 1262: {
1.195 vatton 1263: if (!strncmp (url, "http:", 5)
1.106 cvs 1264: || !strncmp (url, "internal:", 9)
1.204 vatton 1265: || !strncmp (url, "ftp:", 4))
1.195 vatton 1266: /* experimental */
1267: /*** || !strncmp (path, "news:", 5)***/
1268: return (TRUE);
1269: else
1270: return (FALSE);
1.3 cvs 1271: }
1272:
1.31 cvs 1273:
1274: /*----------------------------------------------------------------------
1.195 vatton 1275: GetBaseURL
1276: normalizes orgName according to a base associated with doc, and
1277: following the standard URL format rules.
1278: The function returns the base used to solve relative URL and SRC:
1279: - the base of the document,
1280: - or the document path (without document name).
1.31 cvs 1281: ----------------------------------------------------------------------*/
1.106 cvs 1282: char *GetBaseURL (Document doc)
1.31 cvs 1283: {
1284: Element el;
1285: ElementType elType;
1286: AttributeType attrType;
1287: Attribute attr;
1.176 vatton 1288: char *ptr, *basename, *utf8path;
1.217 vatton 1289: char documentname[MAX_LENGTH];
1.31 cvs 1290: int length;
1.151 kahan 1291: ThotBool hasDocBase;
1.31 cvs 1292:
1.113 cvs 1293: if (doc == 0 || !DocumentURLs[doc])
1.195 vatton 1294: return NULL;
1.148 kahan 1295: /* the other functions expect basename to have no more than MAX_LENGTH chars */
1.171 gully 1296: basename = (char *)TtaGetMemory (MAX_LENGTH);
1.148 kahan 1297: basename[0] = EOS;
1.31 cvs 1298: length = MAX_LENGTH -1;
1.151 kahan 1299: hasDocBase = FALSE;
1300:
1301: /* If the document has a base URL, it has a priority over the headers. */
1302: /* @@ We need to do this too when we support XML:base */
1303:
1304: /* is it a HTML document ? */
1305: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1306: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
1307: /* it's a HTML document */
1308: {
1309: /* get the document element */
1310: el = TtaGetMainRoot (doc);
1311: /* search the BASE element */
1312: elType.ElTypeNum = HTML_EL_HEAD;
1313: el = TtaSearchTypedElement (elType, SearchForward, el);
1314: if (el)
1.195 vatton 1315: /* there is a HEAD element */
1316: {
1317: /* look for a BASE element within the HEAD */
1318: elType.ElTypeNum = HTML_EL_BASE;
1319: el = TtaSearchTypedElement (elType, SearchInTree, el);
1320: }
1.151 kahan 1321: if (el)
1.195 vatton 1322: {
1323: /* The document has a BASE element. Get the HREF attribute of the
1324: BASE element */
1325: hasDocBase = TRUE;
1326: attrType.AttrSSchema = elType.ElSSchema;
1327: attrType.AttrTypeNum = HTML_ATTR_HREF_;
1328: attr = TtaGetAttribute (el, attrType);
1329: if (attr)
1330: {
1331: /* Use the base path of the document */
1332: TtaGiveTextAttributeValue (attr, basename, &length);
1333: utf8path = (char *)TtaConvertMbsToByte ((unsigned char *)basename,
1334: TtaGetDefaultCharset ());
1.217 vatton 1335: if (IsW3Path (DocumentURLs[doc]) && utf8path[0] == '/')
1336: NormalizeURL (utf8path, 0, basename, documentname, DocumentURLs[doc]);
1337: else
1338: strncpy (basename, utf8path, MAX_LENGTH-1);
1.195 vatton 1339: TtaFreeMemory (utf8path);
1340: }
1341: }
1.151 kahan 1342: }
1343:
1344: /* there was no BASE. Do we have a location header? */
1345: if (!hasDocBase && DocumentMeta[doc] && DocumentMeta[doc]->full_content_location
1.148 kahan 1346: && DocumentMeta[doc]->full_content_location[0] != EOS)
1.65 cvs 1347: {
1.148 kahan 1348: strncpy (basename, DocumentMeta[doc]->full_content_location, MAX_LENGTH-1);
1349: basename[MAX_LENGTH-1] = EOS;
1350: length = strlen (basename);
1351: }
1352:
1353: if (basename[0] != EOS)
1354: {
1355: /* base and orgName have to be separated by a DIR_SEP */
1356: length--;
1357: if (basename[0] != EOS && basename[length] != URL_SEP &&
1.195 vatton 1358: basename[length] != DIR_SEP)
1359: /* verify if the base has the form "protocol://server:port" */
1360: {
1361: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1362: AMAYA_PARSE_HOST |
1363: AMAYA_PARSE_PUNCTUATION);
1364: if (ptr && !strcmp (ptr, basename))
1365: {
1366: /* it has this form, complete it by adding a URL_STR */
1367: if (strchr (basename, DIR_SEP))
1368: strcat (basename, DIR_STR);
1369: else
1370: strcat (basename, URL_STR);
1371: length++;
1372: }
1373: else if (!ptr || ptr[0] == EOS)
1374: {
1375: /* no host was detected, we may have a relative URL. We test
1376: if it begins with a URL_SEP, DIR_SEP or period. If yes, it's
1377: relative. */
1378: if (! (basename[0] == '.' || basename[0] == URL_SEP
1379: || basename[0] == DIR_SEP))
1380: basename[0] = EOS;
1381: }
1382: if (ptr)
1383: TtaFreeMemory (ptr);
1384: }
1.113 cvs 1385: }
1386:
1.148 kahan 1387: /* there was no base element and no location header, we use the DocumentURL */
1388: if (basename[0] == EOS)
1389: {
1390: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1391: basename[MAX_LENGTH-1] = EOS;
1392: }
1393:
1.31 cvs 1394: /* Remove anything after the last DIR_SEP char. If no such char is found,
1395: * then search for the first ":" char, hoping that what's before that is a
1396: * protocol. If found, end the string there. If neither char is found,
1397: * then discard the whole base element.
1398: */
1.106 cvs 1399: length = strlen (basename) - 1;
1.31 cvs 1400: /* search for the last DIR_SEP char */
1.220 vatton 1401: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 1402: length--;
1.221 vatton 1403: if (length > 1 && basename[length-1] == URL_SEP && basename[length-2] == ':')
1.219 vatton 1404: return (basename);
1.31 cvs 1405: if (length >= 0)
1406: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1407: basename[length + 1] = EOS;
1.31 cvs 1408: else
1409: /* search for the first PATH_STR char */
1410: {
1.106 cvs 1411: for (length = 0; basename[length] != ':' &&
1.195 vatton 1412: basename[length] != EOS; length ++);
1.106 cvs 1413: if (basename[length] == ':')
1.195 vatton 1414: /* found, so end the string there */
1415: basename[length + 1] = EOS;
1.31 cvs 1416: else
1.195 vatton 1417: /* not found, discard the base */
1418: basename[0] = EOS;
1.31 cvs 1419: }
1420: return (basename);
1421: }
1422:
1423:
1.4 cvs 1424: /*----------------------------------------------------------------------
1.195 vatton 1425: GetLocalPath
1426: Allocate and return the local document path associated to the url
1.40 cvs 1427: ----------------------------------------------------------------------*/
1.150 vatton 1428: char *GetLocalPath (Document doc, char *url)
1.106 cvs 1429: {
1430: char *ptr;
1431: char *n;
1432: char *documentname;
1433: char url_sep;
1.83 cvs 1434: int len;
1.67 cvs 1435: ThotBool noFile;
1.40 cvs 1436:
1.153 vatton 1437: if (url)
1.40 cvs 1438: {
1439: /* check whether the file name exists */
1.106 cvs 1440: len = strlen (url) - 1;
1.71 cvs 1441: if (IsW3Path (url))
1.195 vatton 1442: url_sep = URL_SEP;
1.41 cvs 1443: else
1.195 vatton 1444: url_sep = DIR_SEP;
1.41 cvs 1445: noFile = (url[len] == url_sep);
1.40 cvs 1446: if (noFile)
1.195 vatton 1447: url[len] = EOS;
1.171 gully 1448: ptr = (char *)TtaGetMemory (MAX_LENGTH);
1449: documentname = (char *)TtaGetMemory (MAX_LENGTH);
1.78 cvs 1450: TtaExtractName (url, ptr, documentname);
1.106 cvs 1451: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1452: if (!TtaCheckDirectory (ptr))
1.195 vatton 1453: /* directory did not exist */
1454: TtaMakeDirectory (ptr);
1.47 cvs 1455:
1.153 vatton 1456: if (doc == 0)
1.195 vatton 1457: {
1458: n = strrchr (documentname, '.');
1459: if (n)
1460: *n = EOS;
1461: if (documentname[0] == EOS)
1462: strcpy (documentname, "noname");
1463: n = GetTempName (ptr, documentname);
1464: TtaFreeMemory (ptr);
1465: ptr = n;
1466: }
1.69 cvs 1467: else
1.195 vatton 1468: {
1469: /* don't include the query string within document name */
1470: n = strrchr (documentname, '?');
1471: if (n)
1472: *n = EOS;
1473: /* don't include ':' within document name */
1474: n = strchr (documentname, ':');
1475: if (n)
1476: *n = EOS;
1477: /* if after all this operations document name
1478: is empty, let's use noname.html instead */
1479: if (documentname[0] == EOS)
1480: strcat (ptr, "noname.html");
1481: else
1482: strcat (ptr, documentname);
1483: }
1.40 cvs 1484: TtaFreeMemory (documentname);
1.157 kahan 1485: /* substitute invalid chars in file names by a _ */
1.188 gully 1486: n = strrchr(ptr,DIR_SEP);
1.157 kahan 1487: while (*n)
1.195 vatton 1488: {
1489: if (*n == '*' || *n == ',')
1490: *n = '_';
1491: if ((unsigned char)*n >= 0x80) /* avoid non-ASCII */
1492: *n = 'A' + ((unsigned char)*n % 26);
1493: n++;
1494: }
1.40 cvs 1495: /* restore the url */
1496: if (noFile)
1.195 vatton 1497: url[len] = url_sep;
1.40 cvs 1498: return (ptr);
1499: }
1500: else
1501: return (NULL);
1502: }
1503:
1.73 cvs 1504: /*----------------------------------------------------------------------
1.195 vatton 1505: ExtractTarget extract the target name from document nane.
1.79 cvs 1506: ----------------------------------------------------------------------*/
1.150 vatton 1507: void ExtractTarget (char *aName, char *target)
1.79 cvs 1508: {
1.195 vatton 1509: long int lg, i;
1510: char *ptr;
1511: char *oldptr;
1512:
1513: if (!target || !aName)
1514: /* bad target */
1515: return;
1516:
1517: target[0] = EOS;
1518: lg = strlen (aName);
1519: if (lg)
1520: {
1521: /* the name is not empty */
1522: oldptr = ptr = &aName[0];
1523: do
1524: {
1525: ptr = strrchr (oldptr, '#');
1526: if (ptr)
1527: oldptr = &ptr[1];
1528: }
1529: while (ptr);
1530:
1531: i = (long int) (oldptr) - (long int) (aName); /* name length */
1532: if (i > 1)
1533: {
1534: aName[i - 1] = EOS;
1535: if (i != lg)
1536: strcpy (target, oldptr);
1537: }
1538: }
1.79 cvs 1539: }
1540:
1541: /*----------------------------------------------------------------------
1.195 vatton 1542: RemoveNewLines (text)
1543: Removes any '\n' chars that are found in text.
1544: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1545: ----------------------------------------------------------------------*/
1.106 cvs 1546: ThotBool RemoveNewLines (char *text)
1547: {
1548: ThotBool change = FALSE;
1549: char *src;
1550: char *dest;
1.90 cvs 1551:
1552: src = text;
1553: dest = text;
1.115 kahan 1554:
1555: /* remove any preceding whitespace */
1556: while (*src && *src == ' ')
1557: {
1558: src++;
1559: change = 1;
1560: }
1561:
1.90 cvs 1562: while (*src)
1563: {
1564: switch (*src)
1.195 vatton 1565: {
1566: case '\n':
1567: /* don't copy the newline */
1568: change = 1;
1569: break;
1570: default:
1571: *dest = *src;
1572: dest++;
1573: break;
1574: }
1.90 cvs 1575: src++;
1576: }
1577: /* copy the last EOS char */
1578: *dest = *src;
1579:
1580: return (change);
1581: }
1582:
1583: /*----------------------------------------------------------------------
1.195 vatton 1584: CleanCopyFileURL
1585: Copies a file url from a src string to destination string.
1586: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1587: we want to do).
1.90 cvs 1588: ----------------------------------------------------------------------*/
1.106 cvs 1589: static void CleanCopyFileURL (char *dest, char *src,
1.195 vatton 1590: ConvertionType convertion)
1.90 cvs 1591: {
1592: while (*src)
1.89 cvs 1593: {
1.90 cvs 1594: switch (*src)
1.195 vatton 1595: {
1.184 gully 1596: #ifdef _WINDOWS
1.195 vatton 1597: case URL_SEP:
1598: /* make DIR_SEP transformation */
1599: if (convertion & AM_CONV_URL_SEP)
1600: *dest = DIR_SEP;
1601: else
1602: *dest = *src;
1603: dest++;
1604: src++;
1605: break;
1.184 gully 1606: #endif /* _WINDOWS */
1.96 cvs 1607:
1.195 vatton 1608: case '%':
1609: if (convertion & AM_CONV_PERCENT)
1610: {
1611: /* (code adapted from libwww's HTUnEscape function */
1612: src++;
1613: if (*src != EOS)
1614: {
1615: *dest = UnEscapeChar (*src) * 16;
1616: src++;
1617: }
1618: if (*src != EOS)
1619: {
1620: *dest = *dest + UnEscapeChar (*src);
1621: src++;
1622: }
1623: dest++;
1624: }
1625: else
1626: {
1627: *dest = *src;
1628: dest++;
1629: src++;
1630: }
1631: break;
1632:
1633: default:
1634: *dest = *src;
1635: dest++;
1636: src++;
1637: break;
1638: }
1.89 cvs 1639: }
1.90 cvs 1640: /* copy the EOS char */
1641: *dest = *src;
1.73 cvs 1642: }
1.40 cvs 1643:
1644: /*----------------------------------------------------------------------
1.195 vatton 1645: NormalizeURL
1646: normalizes orgName according to a base associated with doc, and
1647: following the standard URL format rules.
1648: if doc is < 0, use as a base the URL of the document that contains
1649: (or contained) the elements that are now in the copy/cut buffer.
1650: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1651: other path.
1652: The function returns the new complete and normalized URL
1653: or file name path (newName) and the name of the document (docName).
1654: N.B. If the function can't find out what's the docName, it assigns
1655: the name "noname.html".
1.4 cvs 1656: ----------------------------------------------------------------------*/
1.106 cvs 1657: void NormalizeURL (char *orgName, Document doc, char *newName,
1.206 kia 1658: char *docName, const char *otherPath)
1.106 cvs 1659: {
1.195 vatton 1660: char *basename;
1661: char tempOrgName[MAX_LENGTH];
1662: char *ptr;
1663: char used_sep;
1664: int length;
1665: ThotBool check;
1.5 cvs 1666:
1.184 gully 1667: #ifdef _WINDOWS
1.195 vatton 1668: int ndx;
1.184 gully 1669: #endif /* _WINDOWS */
1.44 cvs 1670:
1.195 vatton 1671: if (!newName || !docName)
1672: return;
1673:
1674: if (doc < 0)
1675: basename = TtaStrdup (SavedDocumentURL);
1676: else if (doc > 0)
1677: basename = GetBaseURL (doc);
1.217 vatton 1678: else if (otherPath)
1.195 vatton 1679: basename = TtaStrdup (otherPath);
1680: else
1681: basename = NULL;
1682:
1683: /*
1684: * Clean orgName
1685: * Make sure we have a complete orgName, without any leading or trailing
1686: * white spaces, or trailinbg new lines
1687: */
1688: ptr = orgName;
1689: /* skip leading white space and new line characters */
1690: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1691: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1692: tempOrgName[MAX_LENGTH -1] = EOS;
1693: /*
1694: * Make orgName a complete URL
1695: * If the URL does not include a protocol, then try to calculate
1696: * one using the doc's base element (if it exists),
1697: */
1698: if (tempOrgName[0] == EOS)
1699: {
1700: newName[0] = EOS;
1701: docName[0] = EOS;
1702: TtaFreeMemory (basename);
1.5 cvs 1703: return;
1.195 vatton 1704: }
1705:
1706: /* clean trailing white space */
1707: length = strlen (tempOrgName) - 1;
1708: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1709: {
1710: tempOrgName[length] = EOS;
1711: length--;
1712: }
1.18 cvs 1713:
1.195 vatton 1714: /* remove extra dot (which dot???) */
1715: /* ugly, but faster than a strcmp */
1716: if (tempOrgName[length] == '.'
1717: && (length == 0 || tempOrgName[length-1] != '.'))
1718: tempOrgName[length] = EOS;
1719:
1720: if (IsW3Path (tempOrgName))
1721: {
1722: /* the name is complete, go to the Sixth Step */
1723: strcpy (newName, tempOrgName);
1724: SimplifyUrl (&newName);
1725: /* verify if the URL has the form "protocol://server:port" */
1726: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1727: AMAYA_PARSE_HOST |
1728: AMAYA_PARSE_PUNCTUATION);
1729: if (ptr && !strcmp (ptr, newName))
1730: /* it has this form, we complete it by adding a DIR_STR */
1731: strcat (newName, URL_STR);
1732:
1733: if (ptr)
1734: TtaFreeMemory (ptr);
1735: }
1736: else if (basename == NULL)
1737: /* the name is complete, go to the Sixth Step */
1738: strcpy (newName, tempOrgName);
1739: else
1740: {
1741: /* Calculate the absolute URL, using the base or document URL */
1.184 gully 1742: #ifdef _WINDOWS
1.195 vatton 1743: if (!IsW3Path (basename))
1744: {
1745: length = strlen (tempOrgName);
1746: for (ndx = 0; ndx < length; ndx++)
1.211 vatton 1747: if (tempOrgName[ndx] == '/')
1748: tempOrgName[ndx] = '\\';
1.195 vatton 1749: }
1.184 gully 1750: #endif /* _WINDOWS */
1.195 vatton 1751: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1752: if (ptr)
1753: {
1754: SimplifyUrl (&ptr);
1755: strcpy (newName, ptr);
1756: TtaFreeMemory (ptr);
1757: }
1758: else
1759: newName[0] = EOS;
1760: }
1761:
1762: TtaFreeMemory (basename);
1763: /*
1764: * Prepare the docname that will refer to this ressource in the
1765: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1766: * noname.html as a default ressource name
1.18 cvs 1767: */
1.195 vatton 1768: if (newName[0] != EOS)
1769: {
1770: length = strlen (newName) - 1;
1771: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1772: {
1773: used_sep = newName[length];
1774: check = TRUE;
1775: while (check)
1776: {
1777: length--;
1778: while (length >= 0 && newName[length] != used_sep)
1779: length--;
1780: if (!strncmp (&newName[length+1], "..", 2))
1781: {
1782: newName[length+1] = EOS;
1783: /* remove also previous directory */
1784: length--;
1785: while (length >= 0 && newName[length] != used_sep)
1786: length--;
1787: if (strncmp (&newName[length+1], "//", 2))
1788: /* don't remove server name */
1789: newName[length+1] = EOS;
1790: }
1791: else if (!strncmp (&newName[length+1], ".", 1))
1792: newName[length+1] = EOS;
1793: else
1794: check = FALSE;
1795: }
1796: /* docname was not comprised inside the URL, so let's */
1797: /* assign the default ressource name */
1798: strcpy (docName, "noname.html");
1799: }
1800: else
1801: { /* docname is comprised inside the URL */
1802: while (length >= 0 && newName[length] != URL_SEP &&
1803: newName[length] != DIR_SEP)
1804: length--;
1805: if (length < 0)
1806: strcpy (docName, newName);
1807: else
1808: strcpy (docName, &newName[length+1]);
1809: }
1810: }
1811: else
1812: docName[0] = EOS;
1.18 cvs 1813: }
1.3 cvs 1814:
1.4 cvs 1815: /*----------------------------------------------------------------------
1.9 cvs 1816: IsSameHost
1.4 cvs 1817: ----------------------------------------------------------------------*/
1.106 cvs 1818: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1819: {
1.106 cvs 1820: char *basename_ptr1, *basename_ptr2;
1821: ThotBool result;
1.3 cvs 1822:
1.106 cvs 1823: basename_ptr1 = AmayaParseUrl (url1, "",
1.195 vatton 1824: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.106 cvs 1825: basename_ptr2 = AmayaParseUrl (url2, "",
1.195 vatton 1826: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1827:
1.106 cvs 1828: if (strcmp (basename_ptr1, basename_ptr2))
1829: result = FALSE;
1830: else
1831: result = TRUE;
1832: TtaFreeMemory (basename_ptr1);
1833: TtaFreeMemory (basename_ptr2);
1834: return (result);
1.3 cvs 1835: }
1836:
1837:
1.4 cvs 1838: /*----------------------------------------------------------------------
1.22 cvs 1839: HasKnownFileSuffix
1840: returns TRUE if path points to a file ending with a suffix.
1841: ----------------------------------------------------------------------*/
1.153 vatton 1842: ThotBool HasKnownFileSuffix (const char *path)
1.106 cvs 1843: {
1.195 vatton 1844: char *root;
1845: char temppath[MAX_LENGTH];
1846: char suffix[MAX_LENGTH];
1847:
1848: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1849: return (FALSE);
1850:
1851: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1852:
1853: if (root)
1854: {
1855: strcpy (temppath, root);
1856: TtaFreeMemory (root);
1857: /* Get the suffix */
1858: TtaExtractSuffix (temppath, suffix);
1859:
1860: if( suffix[0] == EOS)
1861: /* no suffix */
1862: return (FALSE);
1863:
1864: /* Normalize the suffix */
1865: ConvertToLowerCase (suffix);
1866:
1867: if (!strcmp (suffix, "gz"))
1868: /* skip the compressed suffix */
1869: {
1870: TtaExtractSuffix (temppath, suffix);
1871: if(suffix[0] == EOS)
1872: /* no suffix */
1873: return (FALSE);
1874: /* Normalize the suffix */
1875: ConvertToLowerCase (suffix);
1876: }
1877:
1878: if (strcmp (suffix, "gif") &&
1879: strcmp (suffix, "xbm") &&
1880: strcmp (suffix, "xpm") &&
1881: strcmp (suffix, "jpg") &&
1882: strcmp (suffix, "pdf") &&
1883: strcmp (suffix, "png") &&
1884: strcmp (suffix, "tgz") &&
1885: strcmp (suffix, "xpg") &&
1886: strcmp (suffix, "xpd") &&
1887: strcmp (suffix, "ps") &&
1888: strcmp (suffix, "au") &&
1889: strcmp (suffix, "html") &&
1890: strcmp (suffix, "htm") &&
1891: strcmp (suffix, "shtml") &&
1892: strcmp (suffix, "xht") &&
1893: strcmp (suffix, "xhtm") &&
1894: strcmp (suffix, "xhtml") &&
1895: strcmp (suffix, "txt") &&
1896: strcmp (suffix, "css") &&
1897: strcmp (suffix, "eps"))
1898: return (FALSE);
1899: else
1900: return (TRUE);
1901: }
1902: else
1903: return (FALSE);
1.22 cvs 1904: }
1905:
1906:
1907: /*----------------------------------------------------------------------
1.24 cvs 1908: ChopURL
1909: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1910: If inputURL is bigger than that size, outputURL receives
1911: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1912: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1913: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1914: copied into outputURL.
1915: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1916: chars.
1917: ----------------------------------------------------------------------*/
1.106 cvs 1918: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1919: {
1920: int len;
1.9 cvs 1921:
1.106 cvs 1922: len = strlen (inputURL);
1.24 cvs 1923: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1924: strcpy (outputURL, inputURL);
1.24 cvs 1925: else
1926: /* make a truncated urlName on the status window */
1927: {
1.106 cvs 1928: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1929: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1930: strcat (outputURL, "...");
1931: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1932: }
1.25 cvs 1933: }
1934:
1935:
1936: /*----------------------------------------------------------------------
1.195 vatton 1937: scan
1938: Scan a filename for its constituents
1939: -----------------------------------
1940:
1941: On entry,
1942: name points to a document name which may be incomplete.
1943: On exit,
1944: absolute or relative may be nonzero (but not both).
1945: host, fragment and access may be nonzero if they were specified.
1946: Any which are nonzero point to zero terminated strings.
1.25 cvs 1947: ----------------------------------------------------------------------*/
1.106 cvs 1948: static void scan (char *name, HTURI *parts)
1.25 cvs 1949: {
1.106 cvs 1950: char * p;
1951: char * after_access = name;
1.32 cvs 1952:
1.211 vatton 1953: memset (parts, 0, sizeof (HTURI));
1.28 cvs 1954: /* Look for fragment identifier */
1.106 cvs 1955: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1956: {
1.211 vatton 1957: *p++ = EOS;
1.28 cvs 1958: parts->fragment = p;
1.25 cvs 1959: }
1960:
1.28 cvs 1961: for (p=name; *p; p++)
1962: {
1.106 cvs 1963: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.195 vatton 1964: break;
1.106 cvs 1965: if (*p == ':')
1.195 vatton 1966: {
1967: *p = 0;
1968: parts->access = after_access; /* Scheme has been specified */
1969:
1970: /* The combination of gcc, the "-O" flag and the HP platform is
1971: unhealthy. The following three lines is a quick & dirty fix, but is
1972: not recommended. Rather, turn off "-O". */
1973:
1974: /* after_access = p;*/
1975: /* while (*after_access == 0)*/
1976: /* after_access++;*/
1977: after_access = p+1;
1978: if (!strcasecmp("URL", parts->access))
1979: /* Ignore IETF's URL: pre-prefix */
1980: parts->access = NULL;
1981: else
1982: break;
1983: }
1.25 cvs 1984: }
1985:
1.195 vatton 1986: p = after_access;
1987: if (*p == URL_SEP || *p == DIR_SEP)
1988: {
1989: if (p[1] == URL_SEP)
1990: {
1991: parts->host = p+2; /* host has been specified */
1992: *p = 0; /* Terminate access */
1993: /* look for end of host name if any */
1994: p = strchr (parts->host, URL_SEP);
1995: if (p)
1996: {
1997: *p = EOS; /* Terminate host */
1998: parts->absolute = p+1; /* Root has been found */
1999: }
2000: }
2001: else
2002: /* Root found but no host */
2003: parts->absolute = p+1;
2004: }
2005: else
2006: {
2007: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
2008: }
1.25 cvs 2009: }
2010:
2011:
2012: /*----------------------------------------------------------------------
1.28 cvs 2013: AmayaParseUrl: parse a Name relative to another name
2014:
2015: This returns those parts of a name which are given (and requested)
2016: substituting bits from the related name where necessary.
1.25 cvs 2017:
1.28 cvs 2018: On entry,
1.195 vatton 2019: aName A filename given
2020: relatedName A name relative to which aName is to be parsed. Give
2021: it an empty string if aName is absolute.
2022: wanted A mask for the bits which are wanted.
1.25 cvs 2023:
1.28 cvs 2024: On exit,
1.195 vatton 2025: returns A pointer to a malloc'd string which MUST BE FREED
1.25 cvs 2026: ----------------------------------------------------------------------*/
1.206 kia 2027: char *AmayaParseUrl (const char *aName, const char *relatedName, int wanted)
1.106 cvs 2028: {
2029: char *return_value;
2030: char result[MAX_LENGTH];
2031: char name[MAX_LENGTH];
2032: char rel[MAX_LENGTH];
2033: char *p, *access;
1.29 cvs 2034: HTURI given, related;
1.205 vatton 2035: int len, l;
1.106 cvs 2036: char used_sep;
1.206 kia 2037: const char*used_str;
1.32 cvs 2038:
1.106 cvs 2039: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 2040: {
1.106 cvs 2041: used_str = DIR_STR;
2042: used_sep = DIR_SEP;
1.33 cvs 2043: }
1.32 cvs 2044: else
1.33 cvs 2045: {
1.106 cvs 2046: used_str = URL_STR;
2047: used_sep = URL_SEP;
1.33 cvs 2048: }
1.32 cvs 2049:
1.29 cvs 2050: /* Make working copies of input strings to cut up: */
2051: return_value = NULL;
2052: result[0] = 0; /* Clear string */
1.169 quint 2053: rel[0] = EOS;
2054: strncpy (name, aName, MAX_LENGTH - 1);
2055: name[MAX_LENGTH - 1] = EOS;
2056: if (relatedName != NULL)
2057: {
2058: strncpy (rel, relatedName, MAX_LENGTH - 1);
2059: rel[MAX_LENGTH - 1] = EOS;
2060: }
1.29 cvs 2061:
2062: scan (name, &given);
2063: scan (rel, &related);
2064: access = given.access ? given.access : related.access;
2065: if (wanted & AMAYA_PARSE_ACCESS)
2066: if (access)
2067: {
1.195 vatton 2068: strcat (result, access);
1.211 vatton 2069: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.195 vatton 2070: strcat (result, ":");
1.29 cvs 2071: }
2072:
2073: if (given.access && related.access)
2074: /* If different, inherit nothing. */
1.106 cvs 2075: if (strcmp (given.access, related.access) != 0)
1.29 cvs 2076: {
1.195 vatton 2077: related.host = 0;
2078: related.absolute = 0;
2079: related.relative = 0;
2080: related.fragment = 0;
1.29 cvs 2081: }
2082:
2083: if (wanted & AMAYA_PARSE_HOST)
1.211 vatton 2084: if (given.host || related.host)
1.29 cvs 2085: {
1.211 vatton 2086: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.195 vatton 2087: strcat (result, "//");
2088: strcat (result, given.host ? given.host : related.host);
1.29 cvs 2089: }
2090:
2091: if (given.host && related.host)
2092: /* If different hosts, inherit no path. */
1.106 cvs 2093: if (strcmp (given.host, related.host) != 0)
1.29 cvs 2094: {
1.195 vatton 2095: related.absolute = 0;
2096: related.relative = 0;
2097: related.fragment = 0;
1.29 cvs 2098: }
2099:
1.205 vatton 2100: len = MAX_LENGTH - 1 - strlen (result);
1.29 cvs 2101: if (wanted & AMAYA_PARSE_PATH)
2102: {
2103: if (given.absolute)
1.195 vatton 2104: {
2105: /* All is given */
2106: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.205 vatton 2107: {
2108: strcat (result, used_str);
2109: len--;
2110: }
2111: l = strlen (given.absolute);
2112: if (l <= len)
2113: {
2114: strcat (result, given.absolute);
2115: len -= l;
2116: }
1.195 vatton 2117: }
1.29 cvs 2118: else if (related.absolute)
1.195 vatton 2119: {
2120: /* Adopt path not name */
2121: strcat (result, used_str);
1.205 vatton 2122: len--;
2123: l = strlen (related.absolute);
2124: if (l <= len)
2125: {
2126: strcat (result, related.absolute);
2127: len -= l;
2128: }
1.195 vatton 2129: if (given.relative)
2130: {
2131: /* Search part? */
2132: p = strchr (result, '?');
2133: if (!p)
1.205 vatton 2134: p = result + strlen(result) - 1;
2135: for (; *p != used_sep; p--); /* last / */
1.195 vatton 2136: /* Remove filename */
1.205 vatton 2137: p[1] = EOS;
1.195 vatton 2138: /* Add given one */
1.205 vatton 2139: l = strlen (given.relative);
2140: if (l <= len)
2141: {
2142: strcat (result, given.relative);
2143: len -= l;
2144: }
2145: }
2146: }
2147: else if (given.relative)
2148: {
2149: /* what we've got */
2150: l = strlen (given.relative);
2151: if (l <= len)
2152: {
1.195 vatton 2153: strcat (result, given.relative);
1.205 vatton 2154: len -= l;
1.195 vatton 2155: }
2156: }
1.29 cvs 2157: else if (related.relative)
1.205 vatton 2158: {
2159: l = strlen (related.relative);
2160: if (l <= len)
2161: {
2162: strcat (result, related.relative);
2163: len -= l;
2164: }
2165: }
1.29 cvs 2166: else
1.205 vatton 2167: {
2168: /* No inheritance */
2169: strcat (result, used_str);
2170: len--;
2171: }
1.25 cvs 2172: }
1.29 cvs 2173:
2174: if (wanted & AMAYA_PARSE_ANCHOR)
1.205 vatton 2175: if (len && (given.fragment || related.fragment))
1.29 cvs 2176: {
1.195 vatton 2177: if (given.absolute && given.fragment)
2178: {
2179: /*Fixes for relURLs...*/
2180: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.205 vatton 2181: {
2182: strcat (result, "#");
2183: len --;
2184: }
2185: l = strlen (given.fragment);
2186: if (l <= len)
2187: strcat (result, given.fragment);
1.195 vatton 2188: }
1.205 vatton 2189: else if (given.absolute || given.fragment)
1.195 vatton 2190: {
2191: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.205 vatton 2192: {
2193: strcat (result, "#");
2194: len--;
2195: }
2196: if (given.fragment)
2197: {
2198: l = strlen (given.fragment);
2199: if (l <= len)
2200: strcat (result, given.fragment);
2201: }
2202: else
2203: {
2204: l = strlen (given.fragment);
2205: if (l <= len)
2206: strcat (result, related.fragment);
2207: }
1.195 vatton 2208: }
1.29 cvs 2209: }
1.205 vatton 2210: return_value = TtaStrdup (result);
1.29 cvs 2211: return (return_value); /* exactly the right length */
1.25 cvs 2212: }
2213:
2214: /*----------------------------------------------------------------------
1.195 vatton 2215: HTCanon
2216: Canonicalizes the URL in the following manner starting from the host
2217: pointer:
1.25 cvs 2218:
1.195 vatton 2219: 1) The host name is converted to lowercase
2220: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1.25 cvs 2221:
1.195 vatton 2222: Return: OK The position of the current path part of the URL
2223: which might be the old one or a new one.
1.25 cvs 2224:
2225: ----------------------------------------------------------------------*/
1.223 ! vatton 2226: static char *HTCanon (char **filename, char *host)
1.106 cvs 2227: {
1.195 vatton 2228: char *newname = NULL;
2229: char used_sep;
2230: char *path;
2231: char *strptr;
2232: char *port;
2233: char *access = host-3;
2234:
2235: if (*filename && strchr (*filename, URL_SEP))
2236: used_sep = URL_SEP;
2237: else
2238: used_sep = DIR_SEP;
2239:
2240: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
2241: access--;
2242: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
2243: path = host + strlen (host);
2244: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
2245: host = strptr;
2246: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
2247: port = NULL;
1.25 cvs 2248:
1.195 vatton 2249: strptr = host; /* Convert to lower-case */
2250: while (strptr < path)
2251: {
2252: *strptr = tolower (*strptr);
2253: strptr++;
2254: }
2255:
2256: /* Does the URL contain a full domain name? This also works for a
2257: numerical host name. The domain name is already made lower-case
2258: and without a trailing dot. */
2259: {
2260: char *dot = port ? port : path;
2261: if (dot > *filename && *--dot == '.')
1.33 cvs 2262: {
1.195 vatton 2263: char *orig = dot;
2264: char *dest = dot + 1;
2265: while ((*orig++ = *dest++));
2266: if (port) port--;
2267: path--;
2268: }
2269: }
2270: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
2271: if (port)
2272: {
2273: if (!*(port+1) || *(port+1) == used_sep)
2274: {
2275: if (!newname)
2276: {
2277: char *orig = port;
2278: char *dest = port + 1;
2279: while ((*orig++ = *dest++));
2280: }
2281: }
2282: else if ((!strncmp (access, "http", 4) &&
2283: (*(port + 1) == '8' &&
2284: *(port+2) == '0' &&
2285: (*(port+3) == used_sep || !*(port + 3)))) ||
2286: (!strncmp (access, "gopher", 6) &&
2287: (*(port+1) == '7' &&
2288: *(port+2) == '0' &&
2289: (*(port+3) == used_sep || !*(port+3)))) ||
2290: (!strncmp (access, "ftp", 3) &&
2291: (*(port+1) == '2' &&
2292: *(port + 2) == '1' &&
2293: (*(port+3) == used_sep || !*(port+3))))) {
2294: if (!newname)
2295: {
2296: char *orig = port;
2297: char *dest = port + 3;
2298: while((*orig++ = *dest++));
2299: /* Update path position, Henry Minsky */
2300: path -= 3;
2301: }
1.33 cvs 2302: }
1.195 vatton 2303: else if (newname)
2304: strncat (newname, port, (int) (path - port));
1.25 cvs 2305: }
2306:
1.195 vatton 2307: if (newname)
2308: {
2309: char *newpath = newname + strlen (newname);
2310: strcat (newname, path);
2311: path = newpath;
2312: /* Free old copy */
2313: TtaFreeMemory(*filename);
2314: *filename = newname;
2315: }
2316: return path;
1.25 cvs 2317: }
2318:
2319:
2320: /*----------------------------------------------------------------------
1.29 cvs 2321: SimplifyUrl: simplify a URI
1.32 cvs 2322: A URI is allowed to contain the sequence xxx/../ which may be
2323: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 2324: Simplification helps us recognize duplicate URIs.
1.25 cvs 2325:
1.28 cvs 2326: Thus, /etc/junk/../fred becomes /etc/fred
1.195 vatton 2327: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 2328:
1.28 cvs 2329: but we should NOT change
1.195 vatton 2330: http://fred.xxx.edu/../..
1.25 cvs 2331:
1.195 vatton 2332: or ../../albert.html
1.25 cvs 2333:
1.28 cvs 2334: In order to avoid empty URLs the following URLs become:
1.25 cvs 2335:
1.195 vatton 2336: /fred/.. becomes /fred/..
2337: /fred/././.. becomes /fred/..
2338: /fred/.././junk/.././ becomes /fred/..
1.25 cvs 2339:
1.28 cvs 2340: If more than one set of `://' is found (several proxies in cascade) then
2341: only the part after the last `://' is simplified.
1.25 cvs 2342:
1.28 cvs 2343: Returns: A string which might be the old one or a new one.
1.25 cvs 2344: ----------------------------------------------------------------------*/
1.223 ! vatton 2345: void SimplifyUrl (char **url)
1.106 cvs 2346: {
2347: char *path;
2348: char *access;
2349: char *newptr;
2350: char *p;
2351: char *orig, *dest, *end;
1.28 cvs 2352:
1.106 cvs 2353: char used_sep;
1.77 cvs 2354: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1.195 vatton 2355: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 2356:
1.28 cvs 2357: if (!url || !*url)
2358: return;
2359:
1.106 cvs 2360: if (strchr (*url, URL_SEP))
1.195 vatton 2361: used_sep = URL_SEP;
1.32 cvs 2362: else
1.195 vatton 2363: used_sep = DIR_SEP;
1.32 cvs 2364:
1.77 cvs 2365: /* should we simplify double dot? */
2366: path = *url;
1.106 cvs 2367: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 2368: ddot_simplify = FALSE;
2369: else
2370: ddot_simplify = TRUE;
2371:
1.28 cvs 2372: /* Find any scheme name */
1.106 cvs 2373: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 2374: {
2375: /* Find host name */
1.28 cvs 2376: access = *url;
1.123 vatton 2377: while (access < path && (*access = tolower (*access)))
1.195 vatton 2378: access++;
1.28 cvs 2379: path += 3;
1.106 cvs 2380: while ((newptr = strstr (path, "://")) != NULL)
1.195 vatton 2381: /* For proxies */
2382: path = newptr + 3;
2383: /* We have a host name */
1.84 cvs 2384: path = HTCanon (url, path);
1.25 cvs 2385: }
1.106 cvs 2386: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 2387: path += 2;
2388: else
2389: path = *url;
1.222 vatton 2390: #ifndef _WINDOWS
1.84 cvs 2391: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 2392: /* Some URLs start //<foo> */
2393: path += 1;
1.222 vatton 2394: else
2395: #endif /* _WINDOWS */
2396: if (IsFilePath (path))
1.94 cvs 2397: {
2398: /* doesn't need to do anything more */
2399: return;
2400: }
1.106 cvs 2401: else if (!strncmp (path, "news:", 5))
1.28 cvs 2402: {
1.106 cvs 2403: newptr = strchr (path+5, '@');
1.28 cvs 2404: if (!newptr)
1.195 vatton 2405: newptr = path + 5;
1.28 cvs 2406: while (*newptr)
1.195 vatton 2407: {
2408: /* Make group or host lower case */
2409: *newptr = tolower (*newptr);
2410: newptr++;
2411: }
1.28 cvs 2412: /* Doesn't need to do any more */
2413: return;
1.25 cvs 2414: }
1.126 cheyroul 2415:
1.189 vatton 2416: /* remove initial multiple /'s */
2417: p = path - 1;
2418: if (path != *url && *p == used_sep && *path == used_sep)
2419: {
2420: while (*path == used_sep)
1.195 vatton 2421: {
2422: orig = path;
2423: dest = path + 1;
2424: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2425: end = orig-1;
2426: }
1.189 vatton 2427: }
2428:
2429: if (path)
1.28 cvs 2430: {
1.106 cvs 2431: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1.195 vatton 2432: (end = strchr (path, '#'))))
2433: end = path + strlen (path);
1.28 cvs 2434:
2435: /* Parse string second time to simplify */
2436: p = path;
2437: while (p < end)
1.195 vatton 2438: {
2439: /* if we're pointing to a char, it's safe to reactivate the
2440: ../ convertion */
2441: if (!ddot_simplify && *p != '.' && *p != used_sep)
2442: ddot_simplify = TRUE;
2443:
1.222 vatton 2444: if (*p == used_sep)
1.195 vatton 2445: {
2446: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
2447: {
2448: orig = p + 1;
2449: dest = (*(p+2) != used_sep) ? p+2 : p+3;
2450: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
2451: end = orig - 1;
2452: }
2453: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
2454: && (*(p+3) == used_sep || !*(p+3)))
2455: {
2456: newptr = p;
2457: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
2458: if (*newptr == used_sep)
2459: orig = newptr + 1;
2460: else
2461: orig = newptr;
2462:
2463: dest = (*(p+3) != used_sep) ? p+3 : p+4;
2464: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
2465: end = orig-1;
2466: /* Start again with prev slash */
2467: p = newptr;
2468: }
1.222 vatton 2469: else if (p != path && *(p+1) == used_sep)
1.195 vatton 2470: {
2471: while (*(p+1) == used_sep)
2472: {
2473: orig = p;
2474: dest = p + 1;
2475: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2476: end = orig-1;
2477: }
2478: }
2479: else
2480: p++;
2481: }
2482: else
2483: p++;
2484: }
2485: }
2486: /*
2487: ** Check for host/../.. kind of things
2488: */
2489: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
2490: && (!*(path+3) || *(path+3) == used_sep))
2491: *(path+1) = EOS;
1.28 cvs 2492: return;
2493: }
2494:
2495:
2496: /*----------------------------------------------------------------------
1.202 vatton 2497: NormalizeFile normalizes local names.
2498: convertion is AM_CONV_NONE or AM_CONV_ALL
1.195 vatton 2499: Return TRUE if target and src differ.
1.28 cvs 2500: ----------------------------------------------------------------------*/
1.106 cvs 2501: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2502: {
1.214 vatton 2503: #ifndef _WINDOWS
1.195 vatton 2504: char *s;
2505: int i;
1.214 vatton 2506: #endif /* _WINDOWS */
1.198 cvs 2507: int start_index; /* the first char that we'll copy */
1.195 vatton 2508: ThotBool change;
1.28 cvs 2509:
1.195 vatton 2510: change = FALSE;
2511: start_index = 0;
1.90 cvs 2512:
1.195 vatton 2513: if (!src || src[0] == EOS)
2514: {
2515: target[0] = EOS;
2516: return FALSE;
2517: }
2518:
2519: /* @@ do I need file: or file:/ here? */
2520: if (strncmp (src, "file:", 5) == 0)
2521: {
2522: /* remove the prefix file: */
2523: start_index += 5;
1.90 cvs 2524:
1.195 vatton 2525: /* remove the localhost prefix */
2526: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
2527: start_index += 11;
1.94 cvs 2528:
1.195 vatton 2529: /* remove the first two slashes in / / /path */
1.222 vatton 2530: if (src[start_index] == '/' && src[start_index + 1] == '/')
2531: start_index +=2;
2532: #ifdef _WINDOWS
1.195 vatton 2533: /* remove any extra slash before the drive name */
1.222 vatton 2534: if (src[start_index] == '/' && src[start_index+2] == ':')
1.195 vatton 2535: start_index++;
1.184 gully 2536: #endif /* _WINDOWS */
1.90 cvs 2537:
1.195 vatton 2538: if (src[start_index] == EOS)
2539: /* if there's nothing afterwards, add a DIR_STR */
2540: strcpy (target, DIR_STR);
2541: else
2542: /* as we're inside a file: URL, we'll apply all the convertions
2543: we know */
2544: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
2545:
2546: change = TRUE;
2547: }
2548: else if (convertion != AM_CONV_NONE)
2549: {
2550: /* we are following a "local" relative link, we do all the
2551: convertions except for the HOME_DIR ~ one */
2552: CleanCopyFileURL (target, src, convertion);
2553: }
1.198 cvs 2554: #ifdef _WINDOWS
1.211 vatton 2555: #ifdef IV
1.198 cvs 2556: else if (src[0] == DIR_SEP && src[1] == DIR_SEP)
2557: {
2558: s = getenv ("HOMEDRIVE");
2559: strcpy (target, s);
2560: i = strlen (target);
2561: strcpy (&target[i], &src[1]);
2562: change = TRUE;
2563: }
1.211 vatton 2564: #endif /* IV */
1.198 cvs 2565: #else /* _WINDOWS */
1.195 vatton 2566: else if (src[0] == '~')
2567: {
2568: /* it must be a URL typed in a text input field */
2569: /* do the HOME_DIR ~ substitution */
2570: s = TtaGetEnvString ("HOME");
2571: strcpy (target, s);
2572: i = strlen (target);
2573: strcpy (&target[i], &src[1]);
2574: change = TRUE;
2575: }
1.184 gully 2576: #endif /* _WINDOWS */
1.195 vatton 2577: else
2578: /* leave it as it is */
2579: strcpy (target, src);
1.96 cvs 2580:
1.195 vatton 2581: /* remove /../ and /./ */
2582: SimplifyUrl (&target);
2583: if (!change)
2584: change = (strcmp (src, target) != 0);
2585: return (change);
1.25 cvs 2586: }
2587:
1.28 cvs 2588:
1.25 cvs 2589: /*----------------------------------------------------------------------
1.31 cvs 2590: MakeRelativeURL: make relative name
1.25 cvs 2591:
1.28 cvs 2592: This function creates and returns a string which gives an expression of
2593: one address as related to another. Where there is no relation, an absolute
2594: address is retured.
1.25 cvs 2595:
1.28 cvs 2596: On entry,
1.195 vatton 2597: Both names must be absolute, fully qualified names of nodes
2598: (no fragment bits)
1.25 cvs 2599:
1.28 cvs 2600: On exit,
1.195 vatton 2601: The return result points to a newly allocated name which, if
2602: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2603: The caller is responsible for freeing the resulting name later.
1.25 cvs 2604: ----------------------------------------------------------------------*/
1.206 kia 2605: char *MakeRelativeURL (const char *aName, const char *relatedName)
1.106 cvs 2606: {
1.206 kia 2607: char *return_value;
2608: char result[MAX_LENGTH];
2609: const char *p;
2610: const char *q;
2611: const char *after_access;
2612: const char *last_slash = NULL;
2613: int slashes, levels, len;
1.184 gully 2614: #ifdef _WINDOWS
1.44 cvs 2615: int ndx;
1.184 gully 2616: #endif /* _WINDOWS */
1.44 cvs 2617:
1.29 cvs 2618: if (aName == NULL || relatedName == NULL)
2619: return (NULL);
2620:
2621: slashes = 0;
2622: after_access = NULL;
2623: p = aName;
2624: q = relatedName;
1.147 vatton 2625: len = 0;
2626: for (; *p && !strncasecmp (p, q, 1); p++, q++, len++)
1.27 cvs 2627: {
2628: /* Find extent of match */
1.106 cvs 2629: if (*p == ':')
1.195 vatton 2630: {
2631: after_access = p + 1;
1.184 gully 2632: #ifdef _WINDOWS
1.195 vatton 2633: if (len == 1)
2634: {
2635: /* it's a local Windows path like c:... */
2636: slashes+=2;
2637: }
1.184 gully 2638: #endif /* _WINDOWS */
1.195 vatton 2639: }
1.168 cvs 2640: if (*p == DIR_SEP)
1.195 vatton 2641: {
2642: /* memorize the last slash position and count them */
2643: last_slash = p;
2644: slashes++;
2645: }
1.25 cvs 2646: }
2647:
1.31 cvs 2648: /* q, p point to the first non-matching character or zero */
1.106 cvs 2649: if (*q == EOS)
1.31 cvs 2650: {
2651: /* New name is a subset of the related name */
2652: /* exactly the right length */
1.106 cvs 2653: len = strlen (p);
1.171 gully 2654: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.195 vatton 2655: strcpy (return_value, p);
1.31 cvs 2656: }
2657: else if ((slashes < 2 && after_access == NULL)
1.195 vatton 2658: || (slashes < 3 && after_access != NULL))
2659: {
1.31 cvs 2660: /* Two names whitout common path */
2661: /* exactly the right length */
1.106 cvs 2662: len = strlen (aName);
1.171 gully 2663: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.195 vatton 2664: strcpy (return_value, aName);
1.31 cvs 2665: }
2666: else
2667: {
2668: /* Some path in common */
1.218 vatton 2669: #ifdef IV
1.106 cvs 2670: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.195 vatton 2671: /* just the same server */
2672: strcpy (result, last_slash);
1.31 cvs 2673: else
1.218 vatton 2674: #endif
1.195 vatton 2675: {
2676: levels= 0;
2677: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
2678: if (*q == DIR_SEP)
2679: levels++;
1.31 cvs 2680:
1.195 vatton 2681: result[0] = EOS;
2682: for (;levels; levels--)
2683: strcat (result, "../");
2684: strcat (result, last_slash+1);
2685: }
1.52 cvs 2686:
2687: if (!*result)
1.195 vatton 2688: strcat (result, "./");
1.52 cvs 2689:
1.31 cvs 2690: /* exactly the right length */
1.106 cvs 2691: len = strlen (result);
1.171 gully 2692: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.195 vatton 2693: strcpy (return_value, result);
1.52 cvs 2694:
1.25 cvs 2695: }
1.184 gully 2696: #ifdef _WINDOWS
1.106 cvs 2697: len = strlen (return_value);
1.195 vatton 2698: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2699: if (return_value[ndx] == '\\')
1.195 vatton 2700: return_value[ndx] = '/' ;
1.184 gully 2701: #endif /* _WINDOWS */
1.29 cvs 2702: return (return_value);
1.24 cvs 2703: }
1.35 cvs 2704:
1.104 kahan 2705: /*----------------------------------------------------------------------
2706: AM_GetFileSize
2707: Returns TRUE and the filesize in the 2nd parameter.
2708: Otherwise, in case of a system error, returns FALSE, with a
2709: filesize of 0L.
2710: ---------------------------------------------------------------------*/
1.206 kia 2711: ThotBool AM_GetFileSize (const char *filename, unsigned long *file_size)
1.104 kahan 2712: {
2713: if (!TtaFileExist (filename))
2714: return FALSE;
1.185 vatton 2715: *file_size = TtaGetFileSize (filename);
2716: return TRUE;
1.104 kahan 2717: }
1.139 kahan 2718:
2719: /*----------------------------------------------------------------------
2720: AM_UseXHTMLMimeType
2721: Returns TRUE if the user has configured Amaya to use this MIME type,
2722: FALSE otherwise.
2723: ---------------------------------------------------------------------*/
2724: ThotBool AM_UseXHTMLMimeType (void)
2725: {
2726: ThotBool xhtml_mimetype;
2727:
2728: /* does the user wants to use the new MIME type? */
2729: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2730:
2731: return (xhtml_mimetype);
1.152 kahan 2732: }
2733:
1.154 kahan 2734:
2735: /********************************************
2736: The following routines were adapted from the GNU libc functions
2737: for generating a tmpnam.
2738: *********************************************/
2739:
2740: /* These are the characters used in temporary filenames. */
2741: static const char letters[] =
2742: "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
2743:
2744: /* Generate a temporary file name based on TMPL. TMPL must match the
2745: rules for mk[s]temp (i.e. end in "XXXXXX"). The name constructed
2746: does not exist at the time of the call to __gen_tempname. TMPL is
2747: overwritten with the result.
2748:
2749: We use a clever algorithm to get hard-to-predict names. */
2750: void
2751: AM_gen_tempname (char *tmpl)
2752: {
2753: int len;
2754: char *XXXXXX;
2755: static uint64_t value;
2756: uint64_t random_time_bits;
2757: unsigned int count;
2758: int save_errno = errno;
2759: struct stat st;
2760:
2761: /* A lower bound on the number of temporary files to attempt to
2762: generate. The maximum total number of temporary file names that
2763: can exist for a given template is 62**6. It should never be
2764: necessary to try all these combinations. Instead if a reasonable
2765: number of names is tried (we define reasonable as 62**3) fail to
2766: give the system administrator the chance to remove the problems. */
2767: unsigned int attempts_min = 62 * 62 * 62;
2768:
2769: /* The number of times to attempt to generate a temporary file. To
2770: conform to POSIX, this must be no smaller than TMP_MAX. */
2771: unsigned int attempts = attempts_min < TMP_MAX ? TMP_MAX : attempts_min;
2772:
2773: len = strlen (tmpl);
2774: if (len < 6 || strcmp (&tmpl[len - 6], "XXXXXX"))
2775: {
2776: /* @@ JK ? */
2777: errno = EINVAL;
2778: return;
2779: }
2780:
2781: /* This is where the Xs start. */
2782: XXXXXX = &tmpl[len - 6];
2783:
2784: /* Get some more or less random data. */
2785: #ifdef RANDOM_BITS
2786: RANDOM_BITS (random_time_bits);
2787: #else
2788: # if HAVE_GETTIMEOFDAY || _LIBC
2789: {
2790: struct timeval tv;
2791: gettimeofday (&tv, NULL);
2792: random_time_bits = ((uint64_t) tv.tv_usec << 16) ^ tv.tv_sec;
2793: }
2794: # else
2795: random_time_bits = time (NULL);
2796: # endif
2797: #endif
2798: value += random_time_bits ^ getpid ();
2799:
2800: for (count = 0; count < attempts; value += 7777, ++count)
2801: {
2802: uint64_t v = value;
2803:
2804: /* Fill in the random bits. */
2805: XXXXXX[0] = letters[v % 62];
2806: v /= 62;
2807: XXXXXX[1] = letters[v % 62];
2808: v /= 62;
2809: XXXXXX[2] = letters[v % 62];
2810: v /= 62;
2811: XXXXXX[3] = letters[v % 62];
2812: v /= 62;
2813: XXXXXX[4] = letters[v % 62];
2814: v /= 62;
2815: XXXXXX[5] = letters[v % 62];
2816:
2817: /* This case is backward from the other three. AM_gen_tempname
1.195 vatton 2818: succeeds if __xstat fails because the name does not exist.
2819: Note the continue to bypass the common logic at the bottom
2820: of the loop. */
1.154 kahan 2821: if (stat (tmpl, &st) < 0)
1.195 vatton 2822: break;
1.154 kahan 2823:
2824: continue;
2825: }
2826:
2827: if (count == attempts || errno != ENOENT)
2828: tmpl[0] = EOS;
2829: else
2830: errno = save_errno;
2831:
2832: return;
2833: }
2834:
2835: #define JOSE 1
2836:
1.152 kahan 2837: /*-----------------------------------------------------------------------
2838: GetTempName
2839: Front end to the Unix tempnam function, which is independent of the
2840: value of the TMPDIR env value
2841: Returns a dynamically allocated string with a tempname. The user
2842: must free this memory.
2843: -----------------------------------------------------------------------*/
2844: char *GetTempName (const char *dir, const char *prefix)
2845: {
1.154 kahan 2846: #ifdef JOSE
2847:
1.162 kahan 2848: static char tmpbufmem[PATH_MAX + 1];
1.154 kahan 2849: int len;
2850: int i;
2851:
1.155 cvs 2852: if (!dir || *dir == EOS || !TtaDirExists (dir))
1.154 kahan 2853: return NULL;
2854:
1.162 kahan 2855: /* make sure that the name is no bigger than PATH_MAX + the 6 tempname chars we
1.195 vatton 2856: will add */
1.154 kahan 2857:
1.156 cvs 2858: len = strlen (dir);
1.162 kahan 2859: if (len + 6 > PATH_MAX)
1.154 kahan 2860: return NULL;
2861:
2862: /* copy the dir name, and add a DIR_SEP if it's missing */
2863: if (dir[strlen (dir) - 1] == DIR_SEP)
2864: strcpy (tmpbufmem, dir);
2865: else
1.195 vatton 2866: {
2867: sprintf (tmpbufmem, "%s%c", dir, DIR_SEP);
2868: len++;
2869: }
1.154 kahan 2870:
1.161 kahan 2871: /* copy the prefix (no more than L_tmpnam chars, to respect POSIX). Save
1.156 cvs 2872: space for the 6 X and EOS chars that will become the random bits */
2873: if (prefix)
1.195 vatton 2874: {
1.156 cvs 2875: i = 0;
1.195 vatton 2876: while (prefix[i] != EOS && i < L_tmpnam - 8)
2877: tmpbufmem[len++] = prefix[i++];
2878: tmpbufmem[len] = EOS;
2879: }
1.156 cvs 2880:
2881: /* Add the 6 X chars */
2882: len = strlen (tmpbufmem);
2883: i = 0;
2884: while (i < 6)
1.195 vatton 2885: {
2886: tmpbufmem[len++] = 'X';
2887: i++;
2888: }
1.156 cvs 2889: tmpbufmem[len] = EOS;
1.154 kahan 2890:
2891: AM_gen_tempname (tmpbufmem);
2892:
2893: if (tmpbufmem[0] == EOS)
2894: return NULL;
2895: else
2896: return (TtaStrdup (tmpbufmem));
2897:
2898: #else
1.152 kahan 2899: char *tmpdir;
1.181 vatton 2900: char *tmp = NULL;
1.152 kahan 2901: char *name = NULL;
2902:
2903: /* save the value of TMPDIR */
1.181 vatton 2904: tmpdir = getenv ("TMPDIR");
1.152 kahan 2905: if (tmpdir)
2906: {
1.181 vatton 2907: /* remove TMPDIR from the environment */
2908: tmp = TtaGetMemory (strlen (tmpdir) + 20);
2909: sprintf (tmp, "TMPDIR=");
1.184 gully 2910: #ifdef _WINDOWS
1.152 kahan 2911: _putenv (tmp);
2912: #else
2913: putenv (tmp);
1.184 gully 2914: #endif /* _WINDOWS */
1.181 vatton 2915: /* prepare the string to restore the value of TMPDIR */
2916: strrcat (tmp, tmpdir);
1.152 kahan 2917: }
2918:
2919: /* create the tempname */
1.184 gully 2920: #ifdef _WINDOWS
1.152 kahan 2921: /* Under Windows, _tempnam returns the same name until the file is created */
2922: {
2923: char *altprefix;
2924: name = tmpnam (NULL); /* get a possibly unique string */
1.181 vatton 2925: altprefix = TtaGetMemory (strlen (prefix) + strlen(name) + 1);
1.152 kahan 2926: sprintf (altprefix, "%s%s", prefix, name + strlen(_P_tmpdir));
2927: name = _tempnam (dir, altprefix); /* get a name that isn't yet in use */
2928: TtaFreeMemory (altprefix);
2929: }
2930: #else
2931: name = tempnam (dir, prefix);
1.184 gully 2932: #endif /* _WINDOWS */
1.152 kahan 2933:
2934: if (tmpdir)
2935: {
1.181 vatton 2936: /* restore the value of TMPDIR */
1.184 gully 2937: #ifdef _WINDOWS
1.152 kahan 2938: _putenv (tmpdir);
2939: #else
2940: putenv (tmpdir);
1.184 gully 2941: #endif /* _WINDOWS */
1.181 vatton 2942: TtaFreeMemory (tmpdir);
1.152 kahan 2943: }
2944: return (name);
1.154 kahan 2945: #endif
1.139 kahan 2946: }
Webmaster
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to AHTURLTools.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Public] / Amaya / amaya