Annotation of Amaya/amaya/AHTURLTools.c, revision 1.226
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 == '/' ||
1.225 vatton 883: strstr (suffix, "#") || strstr (suffix, "?") ||
1.223 vatton 884: !strncasecmp (suffix, "htm", 3) ||
1.225 vatton 885: !strncasecmp (suffix, "shtm", 4) ||
1.223 vatton 886: !strncasecmp (suffix,"xhtm", 4) ||
887: !strncasecmp (suffix, "php", 3) ||
888: !strncasecmp (suffix, "mml", 3) ||
889: !strncasecmp (suffix, "svg", 3) ||
890: !strncasecmp (suffix, "xml", 3))
891: ret = FALSE;
892: else
893: ret = TRUE;
894:
895: TtaFreeMemory (temppath);
896: TtaFreeMemory (suffix);
897: return ret;
898: }
899:
900: /*----------------------------------------------------------------------
901: IsUndisplayedName
1.136 cvs 902: returns TRUE if path points to an undisplayed resource.
1.103 cvs 903: ----------------------------------------------------------------------*/
1.136 cvs 904: ThotBool IsUndisplayedName (const char *path)
1.103 cvs 905: {
1.195 vatton 906: char *temppath;
907: char *suffix;
908: ThotBool ret;
909:
910: if (!path)
911: return (FALSE);
912:
913: temppath = TtaStrdup ((char *)path);
914: suffix = (char *)TtaGetMemory (strlen (path) + 1);
915: TtaExtractSuffix (temppath, suffix);
916:
917: if (!strcasecmp (suffix, "exe") ||
918: !strcasecmp (suffix, "zip") ||
919: !strcasecmp (suffix, "ppt") ||
920: !strcasecmp (suffix, "pdf") ||
921: !strcasecmp (suffix, "ps") ||
922: !strcasecmp (suffix, "eps") ||
923: !strcasecmp (suffix, "tar") ||
924: !strcasecmp (suffix, "tgz") ||
925: !strcasecmp (suffix, "ddl") ||
1.196 vatton 926: !strcasecmp (suffix, "deb") ||
927: !strcasecmp (suffix, "dmg") ||
1.195 vatton 928: !strcasecmp (suffix, "mpg") ||
929: !strcasecmp (suffix, "mpeg") ||
1.196 vatton 930: !strcasecmp (suffix, "rpm") ||
1.195 vatton 931: !strcasecmp (suffix, "wmv") ||
932: !strcasecmp (suffix, "wma") ||
1.223 vatton 933: !strcasecmp (suffix, "doc") ||
934: !strcasecmp (suffix, "odt") ||
1.195 vatton 935: !strcasecmp (suffix, "o"))
936: ret = TRUE;
937: else if (!strcmp (suffix, "gz"))
938: {
939: /* take into account compressed files */
940: TtaExtractSuffix (temppath, suffix);
941: if (!strcasecmp (suffix, "exe") ||
942: !strcasecmp (suffix, "zip") ||
943: !strcasecmp (suffix, "ppt") ||
944: !strcasecmp (suffix, "pdf") ||
945: !strcasecmp (suffix, "ps") ||
946: !strcasecmp (suffix, "eps") ||
947: !strcasecmp (suffix, "tar") ||
948: !strcasecmp (suffix, "ddl") ||
1.196 vatton 949: !strcasecmp (suffix, "dmg") ||
1.195 vatton 950: !strcasecmp (suffix, "mpg") ||
951: !strcasecmp (suffix, "mpeg") ||
952: !strcasecmp (suffix, "wmv") ||
953: !strcasecmp (suffix, "wma") ||
1.223 vatton 954: !strcasecmp (suffix, "doc") ||
955: !strcasecmp (suffix, "odt") ||
1.195 vatton 956: !strcasecmp (suffix, "o"))
957: ret = TRUE;
958: else
959: ret = FALSE;
960: }
961: else
962: ret = FALSE;
963:
964: TtaFreeMemory (temppath);
965: TtaFreeMemory (suffix);
966: return (ret);
1.103 cvs 967: }
968:
969: /*----------------------------------------------------------------------
1.60 cvs 970: IsCSSName
971: returns TRUE if path points to an XML resource.
972: ----------------------------------------------------------------------*/
1.111 cvs 973: ThotBool IsCSSName (const char *path)
1.60 cvs 974: {
1.195 vatton 975: char *temppath;
976: char *suffix;
977: ThotBool ret;
978:
979: if (!path)
980: return (FALSE);
981:
982: temppath = TtaStrdup ((char *)path);
983: suffix = (char *)TtaGetMemory (strlen (path) + 1);
984: TtaExtractSuffix (temppath, suffix);
985:
986: if (!strcasecmp (suffix, "css"))
987: ret = TRUE;
988: else if (!strcmp (suffix, "gz"))
989: {
990: /* take into account compressed files */
991: TtaExtractSuffix (temppath, suffix);
992: if (!strcasecmp (suffix, "css"))
993: ret = TRUE;
994: else
995: ret = FALSE;
996: }
997: else
998: ret = FALSE;
999:
1000: TtaFreeMemory (temppath);
1001: TtaFreeMemory (suffix);
1002: return (ret);
1.56 cvs 1003: }
1004:
1005: /*----------------------------------------------------------------------
1.177 kahan 1006: MultipleBookmarks
1007: returns the value of the multiple bookmarks environment variable.
1008: (TRUE if enabled).
1009: ----------------------------------------------------------------------*/
1010: ThotBool MultipleBookmarks (void)
1011: {
1012: ThotBool multiple_bm;
1013:
1.180 kahan 1014: TtaGetEnvBoolean ("DISABLE_MULTIPLE_BM", &multiple_bm);
1.177 kahan 1015:
1.180 kahan 1016: return (multiple_bm == FALSE);
1.177 kahan 1017: }
1018:
1019: /*----------------------------------------------------------------------
1.172 kahan 1020: IsRDFName
1021: returns TRUE if path points to an RDF resource.
1022: ----------------------------------------------------------------------*/
1023: ThotBool IsRDFName (const char *path)
1024: {
1.195 vatton 1025: char *temppath;
1026: char *suffix;
1027: ThotBool ret;
1028:
1029: /* temporarily disabling this function */
1030: if (!MultipleBookmarks ())
1031: return (FALSE);
1032:
1033: if (!path)
1034: return (FALSE);
1035:
1036: temppath = TtaStrdup ((char *)path);
1037: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1038: TtaExtractSuffix (temppath, suffix);
1039:
1040: if (!strcasecmp (suffix, "rdf"))
1041: ret = TRUE;
1042: else if (!strcmp (suffix, "gz"))
1043: {
1044: /* take into account compressed files */
1045: TtaExtractSuffix (temppath, suffix);
1046: if (!strcasecmp (suffix, "rdf"))
1047: ret = TRUE;
1048: else
1049: ret = FALSE;
1050: }
1051: else
1052: ret = FALSE;
1.183 quint 1053:
1.195 vatton 1054: TtaFreeMemory (temppath);
1055: TtaFreeMemory (suffix);
1056: return (ret);
1.172 kahan 1057: }
1058:
1059: /*----------------------------------------------------------------------
1.9 cvs 1060: IsImageName
1061: returns TRUE if path points to an image resource.
1.4 cvs 1062: ----------------------------------------------------------------------*/
1.111 cvs 1063: ThotBool IsImageName (const char *path)
1.106 cvs 1064: {
1.195 vatton 1065: char *temppath;
1066: char *suffix;
1067: char nsuffix[MAX_LENGTH];
1068: int i;
1069: ThotBool ret;
1070:
1071: if (!path)
1072: return (FALSE);
1073:
1074: temppath = TtaStrdup ((char *)path);
1075: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1076: TtaExtractSuffix (temppath, suffix);
1077:
1078: /* Normalize the suffix */
1079: i = 0;
1080: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1081: {
1082: nsuffix[i] = tolower (suffix[i]);
1083: i++;
1084: }
1085: nsuffix[i] = EOS;
1086: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
1087: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
1088: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1089: ret = TRUE;
1090: else
1091: ret = FALSE;
1092:
1093: TtaFreeMemory (temppath);
1094: TtaFreeMemory (suffix);
1095: return (ret);
1.3 cvs 1096: }
1097:
1.4 cvs 1098: /*----------------------------------------------------------------------
1.58 cvs 1099: IsImageType
1100: returns TRUE if type points to an image resource.
1101: ----------------------------------------------------------------------*/
1.111 cvs 1102: ThotBool IsImageType (const char *type)
1.58 cvs 1103: {
1.195 vatton 1104: char *temptype;
1105: int i;
1106: ThotBool ret;
1107:
1108: if (!type)
1109: return (FALSE);
1110:
1111: temptype = TtaStrdup ((char *)type);
1112: /* Normalize the type */
1113: i = 0;
1114: while (temptype[i] != EOS)
1115: {
1116: temptype[i] = tolower (temptype[i]);
1117: i++;
1118: }
1.166 vatton 1119: if (!strncmp (temptype, "image/", sizeof ("image/") - 1))
1.195 vatton 1120: i = sizeof ("image/") - 1;
1121: else
1122: i = 0;
1123: if (!strcmp (&temptype[i], "gif") ||
1124: !strcmp (&temptype[i], "x-xbitmap") ||
1125: !strcmp (&temptype[i], "x-xpixmap") ||
1126: !strcmp (&temptype[i], "jpeg") ||
1127: !strcmp (&temptype[i], "png"))
1128: ret = TRUE;
1129: else
1130: ret = FALSE;
1131: TtaFreeMemory (temptype);
1132: return (ret);
1.58 cvs 1133: }
1134:
1135: /*----------------------------------------------------------------------
1.9 cvs 1136: IsTextName
1.4 cvs 1137: ----------------------------------------------------------------------*/
1.111 cvs 1138: ThotBool IsTextName (const char *path)
1.106 cvs 1139: {
1.195 vatton 1140: char *temppath;
1141: char *suffix;
1142: char nsuffix[MAX_LENGTH];
1143: int i;
1144: ThotBool ret;
1145:
1146: if (!path)
1147: return (FALSE);
1148:
1149: temppath = TtaStrdup ((char *)path);
1150: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1151: TtaExtractSuffix (temppath, suffix);
1152:
1153: /* Normalize the suffix */
1154: i = 0;
1155: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1156: {
1157: nsuffix[i] = tolower (suffix[i]);
1158: i++;
1159: }
1160: nsuffix[i] = EOS;
1161:
1162: if (!strcmp (nsuffix, "txt") ||
1163: !strcmp (nsuffix, "dtd"))
1164: ret = TRUE;
1165: else if (!strcmp (nsuffix, "gz"))
1166: {
1167: /* take into account compressed files */
1168: TtaExtractSuffix (temppath, suffix);
1169: /* Normalize the suffix */
1170: i = 0;
1171: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1172: {
1173: nsuffix[i] = tolower (suffix[i]);
1174: i++;
1175: }
1176: nsuffix[i] = EOS;
1177: if (!strcmp (nsuffix, "txt") ||
1178: !strcmp (nsuffix, "dtd"))
1179: ret = TRUE;
1180: else
1181: ret = FALSE;
1182: }
1183: else
1184: ret = FALSE;
1185:
1186: TtaFreeMemory (temppath);
1187: TtaFreeMemory (suffix);
1188: return (ret);
1.3 cvs 1189: }
1190:
1.4 cvs 1191: /*----------------------------------------------------------------------
1.9 cvs 1192: IsHTTPPath
1193: returns TRUE if path is in fact an http URL.
1.4 cvs 1194: ----------------------------------------------------------------------*/
1.112 cvs 1195: ThotBool IsHTTPPath (const char *path)
1.3 cvs 1196: {
1.195 vatton 1197: if (!path)
1198: return FALSE;
1.3 cvs 1199:
1.195 vatton 1200: if ((!strncmp (path, "http:", 5) != 0)
1.226 ! carcone 1201: #ifdef SSL
! 1202: || !strncmp (path, "https:", 6)
! 1203: #endif /* SSL */
1.204 vatton 1204: || !strncmp (path, "ftp:", 4)
1.195 vatton 1205: || !strncmp (path, "internal:", 9))
1206: return TRUE;
1207: return FALSE;
1.3 cvs 1208: }
1209:
1.4 cvs 1210: /*----------------------------------------------------------------------
1.9 cvs 1211: IsWithParameters
1212: returns TRUE if url has a concatenated query string.
1.4 cvs 1213: ----------------------------------------------------------------------*/
1.133 vatton 1214: ThotBool IsWithParameters (const char *url)
1.3 cvs 1215: {
1.195 vatton 1216: int i;
1217:
1218: if ((!url) || (url[0] == EOS))
1219: return FALSE;
1.3 cvs 1220:
1.195 vatton 1221: i = strlen (url) - 1;
1222: while (i > 0 && url[i--] != '?')
1223: if (i < 0)
1.5 cvs 1224: return FALSE;
1.3 cvs 1225:
1.195 vatton 1226: /* There is a parameter */
1227: return TRUE;
1.3 cvs 1228: }
1229:
1.4 cvs 1230: /*----------------------------------------------------------------------
1.9 cvs 1231: IsW3Path
1232: returns TRUE if path is in fact a URL.
1.4 cvs 1233: ----------------------------------------------------------------------*/
1.133 vatton 1234: ThotBool IsW3Path (const char *path)
1.106 cvs 1235: {
1.170 quint 1236: if (path == NULL)
1237: return FALSE;
1.106 cvs 1238: if (strncmp (path, "http:", 5) &&
1.226 ! carcone 1239: strncmp (path, "https:", 6) &&
1.106 cvs 1240: strncmp (path, "ftp:", 4) &&
1241: strncmp (path, "telnet:", 7) &&
1242: strncmp (path, "wais:", 5) &&
1243: strncmp (path, "news:", 5) &&
1244: strncmp (path, "gopher:", 7) &&
1245: strncmp (path, "mailto:", 7) &&
1.226 ! carcone 1246: strncmp (path, "archie:", 7))
1.72 cvs 1247: return FALSE;
1248: return TRUE;
1.3 cvs 1249: }
1250:
1.4 cvs 1251: /*----------------------------------------------------------------------
1.90 cvs 1252: IsFilePath
1253: returns TRUE if path is in fact a URL.
1254: ----------------------------------------------------------------------*/
1.133 vatton 1255: ThotBool IsFilePath (const char *path)
1.90 cvs 1256: {
1.106 cvs 1257: if (strncmp (path, "file:", 5))
1.90 cvs 1258: return FALSE;
1259: return TRUE;
1260: }
1261:
1262: /*----------------------------------------------------------------------
1.9 cvs 1263: IsValidProtocol
1264: returns true if the url protocol is supported by Amaya.
1.4 cvs 1265: ----------------------------------------------------------------------*/
1.133 vatton 1266: ThotBool IsValidProtocol (const char *url)
1.106 cvs 1267: {
1.195 vatton 1268: if (!strncmp (url, "http:", 5)
1.226 ! carcone 1269: #ifdef SSL
! 1270: || !strncmp (url, "https:", 6)
! 1271: #endif /* SSL */
1.106 cvs 1272: || !strncmp (url, "internal:", 9)
1.204 vatton 1273: || !strncmp (url, "ftp:", 4))
1.195 vatton 1274: /* experimental */
1275: /*** || !strncmp (path, "news:", 5)***/
1276: return (TRUE);
1277: else
1278: return (FALSE);
1.3 cvs 1279: }
1280:
1.31 cvs 1281:
1282: /*----------------------------------------------------------------------
1.195 vatton 1283: GetBaseURL
1284: normalizes orgName according to a base associated with doc, and
1285: following the standard URL format rules.
1286: The function returns the base used to solve relative URL and SRC:
1287: - the base of the document,
1288: - or the document path (without document name).
1.31 cvs 1289: ----------------------------------------------------------------------*/
1.106 cvs 1290: char *GetBaseURL (Document doc)
1.31 cvs 1291: {
1292: Element el;
1293: ElementType elType;
1294: AttributeType attrType;
1295: Attribute attr;
1.176 vatton 1296: char *ptr, *basename, *utf8path;
1.217 vatton 1297: char documentname[MAX_LENGTH];
1.31 cvs 1298: int length;
1.151 kahan 1299: ThotBool hasDocBase;
1.31 cvs 1300:
1.113 cvs 1301: if (doc == 0 || !DocumentURLs[doc])
1.195 vatton 1302: return NULL;
1.148 kahan 1303: /* the other functions expect basename to have no more than MAX_LENGTH chars */
1.171 gully 1304: basename = (char *)TtaGetMemory (MAX_LENGTH);
1.148 kahan 1305: basename[0] = EOS;
1.31 cvs 1306: length = MAX_LENGTH -1;
1.151 kahan 1307: hasDocBase = FALSE;
1308:
1309: /* If the document has a base URL, it has a priority over the headers. */
1310: /* @@ We need to do this too when we support XML:base */
1311:
1312: /* is it a HTML document ? */
1313: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1314: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
1315: /* it's a HTML document */
1316: {
1317: /* get the document element */
1318: el = TtaGetMainRoot (doc);
1319: /* search the BASE element */
1320: elType.ElTypeNum = HTML_EL_HEAD;
1321: el = TtaSearchTypedElement (elType, SearchForward, el);
1322: if (el)
1.195 vatton 1323: /* there is a HEAD element */
1324: {
1325: /* look for a BASE element within the HEAD */
1326: elType.ElTypeNum = HTML_EL_BASE;
1327: el = TtaSearchTypedElement (elType, SearchInTree, el);
1328: }
1.151 kahan 1329: if (el)
1.195 vatton 1330: {
1331: /* The document has a BASE element. Get the HREF attribute of the
1332: BASE element */
1333: hasDocBase = TRUE;
1334: attrType.AttrSSchema = elType.ElSSchema;
1335: attrType.AttrTypeNum = HTML_ATTR_HREF_;
1336: attr = TtaGetAttribute (el, attrType);
1337: if (attr)
1338: {
1339: /* Use the base path of the document */
1340: TtaGiveTextAttributeValue (attr, basename, &length);
1341: utf8path = (char *)TtaConvertMbsToByte ((unsigned char *)basename,
1342: TtaGetDefaultCharset ());
1.217 vatton 1343: if (IsW3Path (DocumentURLs[doc]) && utf8path[0] == '/')
1344: NormalizeURL (utf8path, 0, basename, documentname, DocumentURLs[doc]);
1345: else
1346: strncpy (basename, utf8path, MAX_LENGTH-1);
1.195 vatton 1347: TtaFreeMemory (utf8path);
1348: }
1349: }
1.151 kahan 1350: }
1351:
1352: /* there was no BASE. Do we have a location header? */
1353: if (!hasDocBase && DocumentMeta[doc] && DocumentMeta[doc]->full_content_location
1.148 kahan 1354: && DocumentMeta[doc]->full_content_location[0] != EOS)
1.65 cvs 1355: {
1.148 kahan 1356: strncpy (basename, DocumentMeta[doc]->full_content_location, MAX_LENGTH-1);
1357: basename[MAX_LENGTH-1] = EOS;
1358: length = strlen (basename);
1359: }
1360:
1361: if (basename[0] != EOS)
1362: {
1363: /* base and orgName have to be separated by a DIR_SEP */
1364: length--;
1365: if (basename[0] != EOS && basename[length] != URL_SEP &&
1.195 vatton 1366: basename[length] != DIR_SEP)
1367: /* verify if the base has the form "protocol://server:port" */
1368: {
1369: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1370: AMAYA_PARSE_HOST |
1371: AMAYA_PARSE_PUNCTUATION);
1372: if (ptr && !strcmp (ptr, basename))
1373: {
1374: /* it has this form, complete it by adding a URL_STR */
1375: if (strchr (basename, DIR_SEP))
1376: strcat (basename, DIR_STR);
1377: else
1378: strcat (basename, URL_STR);
1379: length++;
1380: }
1381: else if (!ptr || ptr[0] == EOS)
1382: {
1383: /* no host was detected, we may have a relative URL. We test
1384: if it begins with a URL_SEP, DIR_SEP or period. If yes, it's
1385: relative. */
1386: if (! (basename[0] == '.' || basename[0] == URL_SEP
1387: || basename[0] == DIR_SEP))
1388: basename[0] = EOS;
1389: }
1390: if (ptr)
1391: TtaFreeMemory (ptr);
1392: }
1.113 cvs 1393: }
1394:
1.148 kahan 1395: /* there was no base element and no location header, we use the DocumentURL */
1396: if (basename[0] == EOS)
1397: {
1398: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1399: basename[MAX_LENGTH-1] = EOS;
1400: }
1401:
1.31 cvs 1402: /* Remove anything after the last DIR_SEP char. If no such char is found,
1403: * then search for the first ":" char, hoping that what's before that is a
1404: * protocol. If found, end the string there. If neither char is found,
1405: * then discard the whole base element.
1406: */
1.106 cvs 1407: length = strlen (basename) - 1;
1.31 cvs 1408: /* search for the last DIR_SEP char */
1.220 vatton 1409: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 1410: length--;
1.221 vatton 1411: if (length > 1 && basename[length-1] == URL_SEP && basename[length-2] == ':')
1.219 vatton 1412: return (basename);
1.31 cvs 1413: if (length >= 0)
1414: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1415: basename[length + 1] = EOS;
1.31 cvs 1416: else
1417: /* search for the first PATH_STR char */
1418: {
1.106 cvs 1419: for (length = 0; basename[length] != ':' &&
1.195 vatton 1420: basename[length] != EOS; length ++);
1.106 cvs 1421: if (basename[length] == ':')
1.195 vatton 1422: /* found, so end the string there */
1423: basename[length + 1] = EOS;
1.31 cvs 1424: else
1.195 vatton 1425: /* not found, discard the base */
1426: basename[0] = EOS;
1.31 cvs 1427: }
1428: return (basename);
1429: }
1430:
1431:
1.4 cvs 1432: /*----------------------------------------------------------------------
1.195 vatton 1433: GetLocalPath
1434: Allocate and return the local document path associated to the url
1.40 cvs 1435: ----------------------------------------------------------------------*/
1.150 vatton 1436: char *GetLocalPath (Document doc, char *url)
1.106 cvs 1437: {
1438: char *ptr;
1439: char *n;
1440: char *documentname;
1441: char url_sep;
1.83 cvs 1442: int len;
1.67 cvs 1443: ThotBool noFile;
1.40 cvs 1444:
1.153 vatton 1445: if (url)
1.40 cvs 1446: {
1447: /* check whether the file name exists */
1.106 cvs 1448: len = strlen (url) - 1;
1.71 cvs 1449: if (IsW3Path (url))
1.195 vatton 1450: url_sep = URL_SEP;
1.41 cvs 1451: else
1.195 vatton 1452: url_sep = DIR_SEP;
1.41 cvs 1453: noFile = (url[len] == url_sep);
1.40 cvs 1454: if (noFile)
1.195 vatton 1455: url[len] = EOS;
1.171 gully 1456: ptr = (char *)TtaGetMemory (MAX_LENGTH);
1457: documentname = (char *)TtaGetMemory (MAX_LENGTH);
1.78 cvs 1458: TtaExtractName (url, ptr, documentname);
1.106 cvs 1459: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1460: if (!TtaCheckDirectory (ptr))
1.195 vatton 1461: /* directory did not exist */
1462: TtaMakeDirectory (ptr);
1.47 cvs 1463:
1.153 vatton 1464: if (doc == 0)
1.195 vatton 1465: {
1466: n = strrchr (documentname, '.');
1467: if (n)
1468: *n = EOS;
1469: if (documentname[0] == EOS)
1470: strcpy (documentname, "noname");
1471: n = GetTempName (ptr, documentname);
1472: TtaFreeMemory (ptr);
1473: ptr = n;
1474: }
1.69 cvs 1475: else
1.195 vatton 1476: {
1477: /* don't include the query string within document name */
1478: n = strrchr (documentname, '?');
1479: if (n)
1480: *n = EOS;
1481: /* don't include ':' within document name */
1482: n = strchr (documentname, ':');
1483: if (n)
1484: *n = EOS;
1485: /* if after all this operations document name
1486: is empty, let's use noname.html instead */
1487: if (documentname[0] == EOS)
1488: strcat (ptr, "noname.html");
1489: else
1490: strcat (ptr, documentname);
1491: }
1.40 cvs 1492: TtaFreeMemory (documentname);
1.157 kahan 1493: /* substitute invalid chars in file names by a _ */
1.188 gully 1494: n = strrchr(ptr,DIR_SEP);
1.157 kahan 1495: while (*n)
1.195 vatton 1496: {
1497: if (*n == '*' || *n == ',')
1498: *n = '_';
1499: if ((unsigned char)*n >= 0x80) /* avoid non-ASCII */
1500: *n = 'A' + ((unsigned char)*n % 26);
1501: n++;
1502: }
1.40 cvs 1503: /* restore the url */
1504: if (noFile)
1.195 vatton 1505: url[len] = url_sep;
1.40 cvs 1506: return (ptr);
1507: }
1508: else
1509: return (NULL);
1510: }
1511:
1.73 cvs 1512: /*----------------------------------------------------------------------
1.195 vatton 1513: ExtractTarget extract the target name from document nane.
1.79 cvs 1514: ----------------------------------------------------------------------*/
1.150 vatton 1515: void ExtractTarget (char *aName, char *target)
1.79 cvs 1516: {
1.195 vatton 1517: long int lg, i;
1518: char *ptr;
1519: char *oldptr;
1520:
1521: if (!target || !aName)
1522: /* bad target */
1523: return;
1524:
1525: target[0] = EOS;
1526: lg = strlen (aName);
1527: if (lg)
1528: {
1529: /* the name is not empty */
1530: oldptr = ptr = &aName[0];
1531: do
1532: {
1533: ptr = strrchr (oldptr, '#');
1534: if (ptr)
1535: oldptr = &ptr[1];
1536: }
1537: while (ptr);
1538:
1539: i = (long int) (oldptr) - (long int) (aName); /* name length */
1540: if (i > 1)
1541: {
1542: aName[i - 1] = EOS;
1543: if (i != lg)
1544: strcpy (target, oldptr);
1545: }
1546: }
1.79 cvs 1547: }
1548:
1549: /*----------------------------------------------------------------------
1.195 vatton 1550: RemoveNewLines (text)
1551: Removes any '\n' chars that are found in text.
1552: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1553: ----------------------------------------------------------------------*/
1.106 cvs 1554: ThotBool RemoveNewLines (char *text)
1555: {
1556: ThotBool change = FALSE;
1557: char *src;
1558: char *dest;
1.90 cvs 1559:
1560: src = text;
1561: dest = text;
1.115 kahan 1562:
1563: /* remove any preceding whitespace */
1564: while (*src && *src == ' ')
1565: {
1566: src++;
1567: change = 1;
1568: }
1569:
1.90 cvs 1570: while (*src)
1571: {
1572: switch (*src)
1.195 vatton 1573: {
1574: case '\n':
1575: /* don't copy the newline */
1576: change = 1;
1577: break;
1578: default:
1579: *dest = *src;
1580: dest++;
1581: break;
1582: }
1.90 cvs 1583: src++;
1584: }
1585: /* copy the last EOS char */
1586: *dest = *src;
1587:
1588: return (change);
1589: }
1590:
1591: /*----------------------------------------------------------------------
1.195 vatton 1592: CleanCopyFileURL
1593: Copies a file url from a src string to destination string.
1594: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1595: we want to do).
1.90 cvs 1596: ----------------------------------------------------------------------*/
1.106 cvs 1597: static void CleanCopyFileURL (char *dest, char *src,
1.195 vatton 1598: ConvertionType convertion)
1.90 cvs 1599: {
1600: while (*src)
1.89 cvs 1601: {
1.90 cvs 1602: switch (*src)
1.195 vatton 1603: {
1.184 gully 1604: #ifdef _WINDOWS
1.195 vatton 1605: case URL_SEP:
1606: /* make DIR_SEP transformation */
1607: if (convertion & AM_CONV_URL_SEP)
1608: *dest = DIR_SEP;
1609: else
1610: *dest = *src;
1611: dest++;
1612: src++;
1613: break;
1.184 gully 1614: #endif /* _WINDOWS */
1.96 cvs 1615:
1.195 vatton 1616: case '%':
1617: if (convertion & AM_CONV_PERCENT)
1618: {
1619: /* (code adapted from libwww's HTUnEscape function */
1620: src++;
1621: if (*src != EOS)
1622: {
1623: *dest = UnEscapeChar (*src) * 16;
1624: src++;
1625: }
1626: if (*src != EOS)
1627: {
1628: *dest = *dest + UnEscapeChar (*src);
1629: src++;
1630: }
1631: dest++;
1632: }
1633: else
1634: {
1635: *dest = *src;
1636: dest++;
1637: src++;
1638: }
1639: break;
1640:
1641: default:
1642: *dest = *src;
1643: dest++;
1644: src++;
1645: break;
1646: }
1.89 cvs 1647: }
1.90 cvs 1648: /* copy the EOS char */
1649: *dest = *src;
1.73 cvs 1650: }
1.40 cvs 1651:
1652: /*----------------------------------------------------------------------
1.195 vatton 1653: NormalizeURL
1654: normalizes orgName according to a base associated with doc, and
1655: following the standard URL format rules.
1656: if doc is < 0, use as a base the URL of the document that contains
1657: (or contained) the elements that are now in the copy/cut buffer.
1658: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1659: other path.
1660: The function returns the new complete and normalized URL
1661: or file name path (newName) and the name of the document (docName).
1662: N.B. If the function can't find out what's the docName, it assigns
1663: the name "noname.html".
1.4 cvs 1664: ----------------------------------------------------------------------*/
1.106 cvs 1665: void NormalizeURL (char *orgName, Document doc, char *newName,
1.206 kia 1666: char *docName, const char *otherPath)
1.106 cvs 1667: {
1.195 vatton 1668: char *basename;
1669: char tempOrgName[MAX_LENGTH];
1670: char *ptr;
1671: char used_sep;
1672: int length;
1673: ThotBool check;
1.5 cvs 1674:
1.184 gully 1675: #ifdef _WINDOWS
1.195 vatton 1676: int ndx;
1.184 gully 1677: #endif /* _WINDOWS */
1.44 cvs 1678:
1.195 vatton 1679: if (!newName || !docName)
1680: return;
1681:
1682: if (doc < 0)
1683: basename = TtaStrdup (SavedDocumentURL);
1684: else if (doc > 0)
1685: basename = GetBaseURL (doc);
1.217 vatton 1686: else if (otherPath)
1.195 vatton 1687: basename = TtaStrdup (otherPath);
1688: else
1689: basename = NULL;
1690:
1691: /*
1692: * Clean orgName
1693: * Make sure we have a complete orgName, without any leading or trailing
1694: * white spaces, or trailinbg new lines
1695: */
1696: ptr = orgName;
1697: /* skip leading white space and new line characters */
1698: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1699: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1700: tempOrgName[MAX_LENGTH -1] = EOS;
1701: /*
1702: * Make orgName a complete URL
1703: * If the URL does not include a protocol, then try to calculate
1704: * one using the doc's base element (if it exists),
1705: */
1706: if (tempOrgName[0] == EOS)
1707: {
1708: newName[0] = EOS;
1709: docName[0] = EOS;
1710: TtaFreeMemory (basename);
1.5 cvs 1711: return;
1.195 vatton 1712: }
1713:
1714: /* clean trailing white space */
1715: length = strlen (tempOrgName) - 1;
1716: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1717: {
1718: tempOrgName[length] = EOS;
1719: length--;
1720: }
1.18 cvs 1721:
1.195 vatton 1722: /* remove extra dot (which dot???) */
1723: /* ugly, but faster than a strcmp */
1724: if (tempOrgName[length] == '.'
1725: && (length == 0 || tempOrgName[length-1] != '.'))
1726: tempOrgName[length] = EOS;
1727:
1728: if (IsW3Path (tempOrgName))
1729: {
1730: /* the name is complete, go to the Sixth Step */
1731: strcpy (newName, tempOrgName);
1732: SimplifyUrl (&newName);
1733: /* verify if the URL has the form "protocol://server:port" */
1734: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1735: AMAYA_PARSE_HOST |
1736: AMAYA_PARSE_PUNCTUATION);
1737: if (ptr && !strcmp (ptr, newName))
1738: /* it has this form, we complete it by adding a DIR_STR */
1739: strcat (newName, URL_STR);
1740:
1741: if (ptr)
1742: TtaFreeMemory (ptr);
1743: }
1744: else if (basename == NULL)
1745: /* the name is complete, go to the Sixth Step */
1746: strcpy (newName, tempOrgName);
1747: else
1748: {
1749: /* Calculate the absolute URL, using the base or document URL */
1.184 gully 1750: #ifdef _WINDOWS
1.195 vatton 1751: if (!IsW3Path (basename))
1752: {
1753: length = strlen (tempOrgName);
1754: for (ndx = 0; ndx < length; ndx++)
1.211 vatton 1755: if (tempOrgName[ndx] == '/')
1756: tempOrgName[ndx] = '\\';
1.195 vatton 1757: }
1.184 gully 1758: #endif /* _WINDOWS */
1.195 vatton 1759: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1760: if (ptr)
1761: {
1762: SimplifyUrl (&ptr);
1763: strcpy (newName, ptr);
1764: TtaFreeMemory (ptr);
1765: }
1766: else
1767: newName[0] = EOS;
1768: }
1769:
1770: TtaFreeMemory (basename);
1771: /*
1772: * Prepare the docname that will refer to this ressource in the
1773: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1774: * noname.html as a default ressource name
1.18 cvs 1775: */
1.195 vatton 1776: if (newName[0] != EOS)
1777: {
1778: length = strlen (newName) - 1;
1779: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1780: {
1781: used_sep = newName[length];
1782: check = TRUE;
1783: while (check)
1784: {
1785: length--;
1786: while (length >= 0 && newName[length] != used_sep)
1787: length--;
1788: if (!strncmp (&newName[length+1], "..", 2))
1789: {
1790: newName[length+1] = EOS;
1791: /* remove also previous directory */
1792: length--;
1793: while (length >= 0 && newName[length] != used_sep)
1794: length--;
1795: if (strncmp (&newName[length+1], "//", 2))
1796: /* don't remove server name */
1797: newName[length+1] = EOS;
1798: }
1799: else if (!strncmp (&newName[length+1], ".", 1))
1800: newName[length+1] = EOS;
1801: else
1802: check = FALSE;
1803: }
1804: /* docname was not comprised inside the URL, so let's */
1805: /* assign the default ressource name */
1806: strcpy (docName, "noname.html");
1807: }
1808: else
1809: { /* docname is comprised inside the URL */
1810: while (length >= 0 && newName[length] != URL_SEP &&
1811: newName[length] != DIR_SEP)
1812: length--;
1813: if (length < 0)
1814: strcpy (docName, newName);
1815: else
1816: strcpy (docName, &newName[length+1]);
1817: }
1818: }
1819: else
1820: docName[0] = EOS;
1.18 cvs 1821: }
1.3 cvs 1822:
1.4 cvs 1823: /*----------------------------------------------------------------------
1.9 cvs 1824: IsSameHost
1.4 cvs 1825: ----------------------------------------------------------------------*/
1.106 cvs 1826: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1827: {
1.106 cvs 1828: char *basename_ptr1, *basename_ptr2;
1829: ThotBool result;
1.3 cvs 1830:
1.106 cvs 1831: basename_ptr1 = AmayaParseUrl (url1, "",
1.195 vatton 1832: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.106 cvs 1833: basename_ptr2 = AmayaParseUrl (url2, "",
1.195 vatton 1834: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1835:
1.106 cvs 1836: if (strcmp (basename_ptr1, basename_ptr2))
1837: result = FALSE;
1838: else
1839: result = TRUE;
1840: TtaFreeMemory (basename_ptr1);
1841: TtaFreeMemory (basename_ptr2);
1842: return (result);
1.3 cvs 1843: }
1844:
1845:
1.4 cvs 1846: /*----------------------------------------------------------------------
1.22 cvs 1847: HasKnownFileSuffix
1848: returns TRUE if path points to a file ending with a suffix.
1849: ----------------------------------------------------------------------*/
1.153 vatton 1850: ThotBool HasKnownFileSuffix (const char *path)
1.106 cvs 1851: {
1.195 vatton 1852: char *root;
1853: char temppath[MAX_LENGTH];
1854: char suffix[MAX_LENGTH];
1855:
1856: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1857: return (FALSE);
1858:
1859: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1860:
1861: if (root)
1862: {
1863: strcpy (temppath, root);
1864: TtaFreeMemory (root);
1865: /* Get the suffix */
1866: TtaExtractSuffix (temppath, suffix);
1867:
1868: if( suffix[0] == EOS)
1869: /* no suffix */
1870: return (FALSE);
1871:
1872: /* Normalize the suffix */
1873: ConvertToLowerCase (suffix);
1874:
1875: if (!strcmp (suffix, "gz"))
1876: /* skip the compressed suffix */
1877: {
1878: TtaExtractSuffix (temppath, suffix);
1879: if(suffix[0] == EOS)
1880: /* no suffix */
1881: return (FALSE);
1882: /* Normalize the suffix */
1883: ConvertToLowerCase (suffix);
1884: }
1885:
1886: if (strcmp (suffix, "gif") &&
1887: strcmp (suffix, "xbm") &&
1888: strcmp (suffix, "xpm") &&
1889: strcmp (suffix, "jpg") &&
1890: strcmp (suffix, "pdf") &&
1891: strcmp (suffix, "png") &&
1892: strcmp (suffix, "tgz") &&
1893: strcmp (suffix, "xpg") &&
1894: strcmp (suffix, "xpd") &&
1895: strcmp (suffix, "ps") &&
1896: strcmp (suffix, "au") &&
1897: strcmp (suffix, "html") &&
1898: strcmp (suffix, "htm") &&
1899: strcmp (suffix, "shtml") &&
1900: strcmp (suffix, "xht") &&
1901: strcmp (suffix, "xhtm") &&
1902: strcmp (suffix, "xhtml") &&
1903: strcmp (suffix, "txt") &&
1904: strcmp (suffix, "css") &&
1905: strcmp (suffix, "eps"))
1906: return (FALSE);
1907: else
1908: return (TRUE);
1909: }
1910: else
1911: return (FALSE);
1.22 cvs 1912: }
1913:
1914:
1915: /*----------------------------------------------------------------------
1.24 cvs 1916: ChopURL
1917: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1918: If inputURL is bigger than that size, outputURL receives
1919: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1920: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1921: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1922: copied into outputURL.
1923: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1924: chars.
1925: ----------------------------------------------------------------------*/
1.106 cvs 1926: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1927: {
1928: int len;
1.9 cvs 1929:
1.106 cvs 1930: len = strlen (inputURL);
1.24 cvs 1931: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1932: strcpy (outputURL, inputURL);
1.24 cvs 1933: else
1934: /* make a truncated urlName on the status window */
1935: {
1.106 cvs 1936: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1937: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1938: strcat (outputURL, "...");
1939: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1940: }
1.25 cvs 1941: }
1942:
1943:
1944: /*----------------------------------------------------------------------
1.195 vatton 1945: scan
1946: Scan a filename for its constituents
1947: -----------------------------------
1948:
1949: On entry,
1950: name points to a document name which may be incomplete.
1951: On exit,
1952: absolute or relative may be nonzero (but not both).
1953: host, fragment and access may be nonzero if they were specified.
1954: Any which are nonzero point to zero terminated strings.
1.25 cvs 1955: ----------------------------------------------------------------------*/
1.106 cvs 1956: static void scan (char *name, HTURI *parts)
1.25 cvs 1957: {
1.106 cvs 1958: char * p;
1959: char * after_access = name;
1.32 cvs 1960:
1.211 vatton 1961: memset (parts, 0, sizeof (HTURI));
1.28 cvs 1962: /* Look for fragment identifier */
1.106 cvs 1963: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1964: {
1.211 vatton 1965: *p++ = EOS;
1.28 cvs 1966: parts->fragment = p;
1.25 cvs 1967: }
1968:
1.28 cvs 1969: for (p=name; *p; p++)
1970: {
1.106 cvs 1971: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.195 vatton 1972: break;
1.106 cvs 1973: if (*p == ':')
1.195 vatton 1974: {
1975: *p = 0;
1976: parts->access = after_access; /* Scheme has been specified */
1977:
1978: /* The combination of gcc, the "-O" flag and the HP platform is
1979: unhealthy. The following three lines is a quick & dirty fix, but is
1980: not recommended. Rather, turn off "-O". */
1981:
1982: /* after_access = p;*/
1983: /* while (*after_access == 0)*/
1984: /* after_access++;*/
1985: after_access = p+1;
1986: if (!strcasecmp("URL", parts->access))
1987: /* Ignore IETF's URL: pre-prefix */
1988: parts->access = NULL;
1989: else
1990: break;
1991: }
1.25 cvs 1992: }
1993:
1.195 vatton 1994: p = after_access;
1995: if (*p == URL_SEP || *p == DIR_SEP)
1996: {
1997: if (p[1] == URL_SEP)
1998: {
1999: parts->host = p+2; /* host has been specified */
2000: *p = 0; /* Terminate access */
2001: /* look for end of host name if any */
2002: p = strchr (parts->host, URL_SEP);
2003: if (p)
2004: {
2005: *p = EOS; /* Terminate host */
2006: parts->absolute = p+1; /* Root has been found */
2007: }
2008: }
2009: else
2010: /* Root found but no host */
2011: parts->absolute = p+1;
2012: }
2013: else
2014: {
2015: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
2016: }
1.25 cvs 2017: }
2018:
2019:
2020: /*----------------------------------------------------------------------
1.28 cvs 2021: AmayaParseUrl: parse a Name relative to another name
2022:
2023: This returns those parts of a name which are given (and requested)
2024: substituting bits from the related name where necessary.
1.25 cvs 2025:
1.28 cvs 2026: On entry,
1.195 vatton 2027: aName A filename given
2028: relatedName A name relative to which aName is to be parsed. Give
2029: it an empty string if aName is absolute.
2030: wanted A mask for the bits which are wanted.
1.25 cvs 2031:
1.28 cvs 2032: On exit,
1.195 vatton 2033: returns A pointer to a malloc'd string which MUST BE FREED
1.25 cvs 2034: ----------------------------------------------------------------------*/
1.206 kia 2035: char *AmayaParseUrl (const char *aName, const char *relatedName, int wanted)
1.106 cvs 2036: {
2037: char *return_value;
2038: char result[MAX_LENGTH];
2039: char name[MAX_LENGTH];
2040: char rel[MAX_LENGTH];
2041: char *p, *access;
1.29 cvs 2042: HTURI given, related;
1.205 vatton 2043: int len, l;
1.106 cvs 2044: char used_sep;
1.206 kia 2045: const char*used_str;
1.32 cvs 2046:
1.106 cvs 2047: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 2048: {
1.106 cvs 2049: used_str = DIR_STR;
2050: used_sep = DIR_SEP;
1.33 cvs 2051: }
1.32 cvs 2052: else
1.33 cvs 2053: {
1.106 cvs 2054: used_str = URL_STR;
2055: used_sep = URL_SEP;
1.33 cvs 2056: }
1.32 cvs 2057:
1.29 cvs 2058: /* Make working copies of input strings to cut up: */
2059: return_value = NULL;
2060: result[0] = 0; /* Clear string */
1.169 quint 2061: rel[0] = EOS;
2062: strncpy (name, aName, MAX_LENGTH - 1);
2063: name[MAX_LENGTH - 1] = EOS;
2064: if (relatedName != NULL)
2065: {
2066: strncpy (rel, relatedName, MAX_LENGTH - 1);
2067: rel[MAX_LENGTH - 1] = EOS;
2068: }
1.29 cvs 2069:
2070: scan (name, &given);
2071: scan (rel, &related);
2072: access = given.access ? given.access : related.access;
2073: if (wanted & AMAYA_PARSE_ACCESS)
2074: if (access)
2075: {
1.195 vatton 2076: strcat (result, access);
1.211 vatton 2077: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.195 vatton 2078: strcat (result, ":");
1.29 cvs 2079: }
2080:
2081: if (given.access && related.access)
2082: /* If different, inherit nothing. */
1.106 cvs 2083: if (strcmp (given.access, related.access) != 0)
1.29 cvs 2084: {
1.195 vatton 2085: related.host = 0;
2086: related.absolute = 0;
2087: related.relative = 0;
2088: related.fragment = 0;
1.29 cvs 2089: }
2090:
2091: if (wanted & AMAYA_PARSE_HOST)
1.211 vatton 2092: if (given.host || related.host)
1.29 cvs 2093: {
1.211 vatton 2094: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.195 vatton 2095: strcat (result, "//");
2096: strcat (result, given.host ? given.host : related.host);
1.29 cvs 2097: }
2098:
2099: if (given.host && related.host)
2100: /* If different hosts, inherit no path. */
1.106 cvs 2101: if (strcmp (given.host, related.host) != 0)
1.29 cvs 2102: {
1.195 vatton 2103: related.absolute = 0;
2104: related.relative = 0;
2105: related.fragment = 0;
1.29 cvs 2106: }
2107:
1.205 vatton 2108: len = MAX_LENGTH - 1 - strlen (result);
1.29 cvs 2109: if (wanted & AMAYA_PARSE_PATH)
2110: {
2111: if (given.absolute)
1.195 vatton 2112: {
2113: /* All is given */
2114: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.205 vatton 2115: {
2116: strcat (result, used_str);
2117: len--;
2118: }
2119: l = strlen (given.absolute);
2120: if (l <= len)
2121: {
2122: strcat (result, given.absolute);
2123: len -= l;
2124: }
1.195 vatton 2125: }
1.29 cvs 2126: else if (related.absolute)
1.195 vatton 2127: {
2128: /* Adopt path not name */
2129: strcat (result, used_str);
1.205 vatton 2130: len--;
2131: l = strlen (related.absolute);
2132: if (l <= len)
2133: {
2134: strcat (result, related.absolute);
2135: len -= l;
2136: }
1.195 vatton 2137: if (given.relative)
2138: {
2139: /* Search part? */
2140: p = strchr (result, '?');
2141: if (!p)
1.205 vatton 2142: p = result + strlen(result) - 1;
2143: for (; *p != used_sep; p--); /* last / */
1.195 vatton 2144: /* Remove filename */
1.205 vatton 2145: p[1] = EOS;
1.195 vatton 2146: /* Add given one */
1.205 vatton 2147: l = strlen (given.relative);
2148: if (l <= len)
2149: {
2150: strcat (result, given.relative);
2151: len -= l;
2152: }
2153: }
2154: }
2155: else if (given.relative)
2156: {
2157: /* what we've got */
2158: l = strlen (given.relative);
2159: if (l <= len)
2160: {
1.195 vatton 2161: strcat (result, given.relative);
1.205 vatton 2162: len -= l;
1.195 vatton 2163: }
2164: }
1.29 cvs 2165: else if (related.relative)
1.205 vatton 2166: {
2167: l = strlen (related.relative);
2168: if (l <= len)
2169: {
2170: strcat (result, related.relative);
2171: len -= l;
2172: }
2173: }
1.29 cvs 2174: else
1.205 vatton 2175: {
2176: /* No inheritance */
2177: strcat (result, used_str);
2178: len--;
2179: }
1.25 cvs 2180: }
1.29 cvs 2181:
2182: if (wanted & AMAYA_PARSE_ANCHOR)
1.205 vatton 2183: if (len && (given.fragment || related.fragment))
1.29 cvs 2184: {
1.195 vatton 2185: if (given.absolute && given.fragment)
2186: {
2187: /*Fixes for relURLs...*/
2188: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.205 vatton 2189: {
2190: strcat (result, "#");
2191: len --;
2192: }
2193: l = strlen (given.fragment);
2194: if (l <= len)
2195: strcat (result, given.fragment);
1.195 vatton 2196: }
1.205 vatton 2197: else if (given.absolute || given.fragment)
1.195 vatton 2198: {
2199: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.205 vatton 2200: {
2201: strcat (result, "#");
2202: len--;
2203: }
2204: if (given.fragment)
2205: {
2206: l = strlen (given.fragment);
2207: if (l <= len)
2208: strcat (result, given.fragment);
2209: }
2210: else
2211: {
2212: l = strlen (given.fragment);
2213: if (l <= len)
2214: strcat (result, related.fragment);
2215: }
1.195 vatton 2216: }
1.29 cvs 2217: }
1.205 vatton 2218: return_value = TtaStrdup (result);
1.29 cvs 2219: return (return_value); /* exactly the right length */
1.25 cvs 2220: }
2221:
2222: /*----------------------------------------------------------------------
1.195 vatton 2223: HTCanon
2224: Canonicalizes the URL in the following manner starting from the host
2225: pointer:
1.25 cvs 2226:
1.195 vatton 2227: 1) The host name is converted to lowercase
2228: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1.25 cvs 2229:
1.195 vatton 2230: Return: OK The position of the current path part of the URL
2231: which might be the old one or a new one.
1.25 cvs 2232:
2233: ----------------------------------------------------------------------*/
1.223 vatton 2234: static char *HTCanon (char **filename, char *host)
1.106 cvs 2235: {
1.195 vatton 2236: char *newname = NULL;
2237: char used_sep;
2238: char *path;
2239: char *strptr;
2240: char *port;
2241: char *access = host-3;
2242:
2243: if (*filename && strchr (*filename, URL_SEP))
2244: used_sep = URL_SEP;
2245: else
2246: used_sep = DIR_SEP;
2247:
2248: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
2249: access--;
2250: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
2251: path = host + strlen (host);
2252: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
2253: host = strptr;
2254: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
2255: port = NULL;
1.25 cvs 2256:
1.195 vatton 2257: strptr = host; /* Convert to lower-case */
2258: while (strptr < path)
2259: {
2260: *strptr = tolower (*strptr);
2261: strptr++;
2262: }
2263:
2264: /* Does the URL contain a full domain name? This also works for a
2265: numerical host name. The domain name is already made lower-case
2266: and without a trailing dot. */
2267: {
2268: char *dot = port ? port : path;
2269: if (dot > *filename && *--dot == '.')
1.33 cvs 2270: {
1.195 vatton 2271: char *orig = dot;
2272: char *dest = dot + 1;
2273: while ((*orig++ = *dest++));
2274: if (port) port--;
2275: path--;
2276: }
2277: }
2278: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
2279: if (port)
2280: {
2281: if (!*(port+1) || *(port+1) == used_sep)
2282: {
2283: if (!newname)
2284: {
2285: char *orig = port;
2286: char *dest = port + 1;
2287: while ((*orig++ = *dest++));
2288: }
2289: }
2290: else if ((!strncmp (access, "http", 4) &&
2291: (*(port + 1) == '8' &&
2292: *(port+2) == '0' &&
2293: (*(port+3) == used_sep || !*(port + 3)))) ||
2294: (!strncmp (access, "gopher", 6) &&
2295: (*(port+1) == '7' &&
2296: *(port+2) == '0' &&
2297: (*(port+3) == used_sep || !*(port+3)))) ||
2298: (!strncmp (access, "ftp", 3) &&
2299: (*(port+1) == '2' &&
2300: *(port + 2) == '1' &&
2301: (*(port+3) == used_sep || !*(port+3))))) {
2302: if (!newname)
2303: {
2304: char *orig = port;
2305: char *dest = port + 3;
2306: while((*orig++ = *dest++));
2307: /* Update path position, Henry Minsky */
2308: path -= 3;
2309: }
1.33 cvs 2310: }
1.195 vatton 2311: else if (newname)
2312: strncat (newname, port, (int) (path - port));
1.25 cvs 2313: }
2314:
1.195 vatton 2315: if (newname)
2316: {
2317: char *newpath = newname + strlen (newname);
2318: strcat (newname, path);
2319: path = newpath;
2320: /* Free old copy */
2321: TtaFreeMemory(*filename);
2322: *filename = newname;
2323: }
2324: return path;
1.25 cvs 2325: }
2326:
2327:
2328: /*----------------------------------------------------------------------
1.29 cvs 2329: SimplifyUrl: simplify a URI
1.32 cvs 2330: A URI is allowed to contain the sequence xxx/../ which may be
2331: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 2332: Simplification helps us recognize duplicate URIs.
1.25 cvs 2333:
1.28 cvs 2334: Thus, /etc/junk/../fred becomes /etc/fred
1.195 vatton 2335: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 2336:
1.28 cvs 2337: but we should NOT change
1.195 vatton 2338: http://fred.xxx.edu/../..
1.25 cvs 2339:
1.195 vatton 2340: or ../../albert.html
1.25 cvs 2341:
1.28 cvs 2342: In order to avoid empty URLs the following URLs become:
1.25 cvs 2343:
1.195 vatton 2344: /fred/.. becomes /fred/..
2345: /fred/././.. becomes /fred/..
2346: /fred/.././junk/.././ becomes /fred/..
1.25 cvs 2347:
1.28 cvs 2348: If more than one set of `://' is found (several proxies in cascade) then
2349: only the part after the last `://' is simplified.
1.25 cvs 2350:
1.28 cvs 2351: Returns: A string which might be the old one or a new one.
1.25 cvs 2352: ----------------------------------------------------------------------*/
1.223 vatton 2353: void SimplifyUrl (char **url)
1.106 cvs 2354: {
2355: char *path;
2356: char *access;
2357: char *newptr;
2358: char *p;
2359: char *orig, *dest, *end;
1.28 cvs 2360:
1.106 cvs 2361: char used_sep;
1.77 cvs 2362: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1.195 vatton 2363: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 2364:
1.28 cvs 2365: if (!url || !*url)
2366: return;
2367:
1.106 cvs 2368: if (strchr (*url, URL_SEP))
1.195 vatton 2369: used_sep = URL_SEP;
1.32 cvs 2370: else
1.195 vatton 2371: used_sep = DIR_SEP;
1.32 cvs 2372:
1.77 cvs 2373: /* should we simplify double dot? */
2374: path = *url;
1.106 cvs 2375: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 2376: ddot_simplify = FALSE;
2377: else
2378: ddot_simplify = TRUE;
2379:
1.28 cvs 2380: /* Find any scheme name */
1.106 cvs 2381: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 2382: {
2383: /* Find host name */
1.28 cvs 2384: access = *url;
1.123 vatton 2385: while (access < path && (*access = tolower (*access)))
1.195 vatton 2386: access++;
1.28 cvs 2387: path += 3;
1.106 cvs 2388: while ((newptr = strstr (path, "://")) != NULL)
1.195 vatton 2389: /* For proxies */
2390: path = newptr + 3;
2391: /* We have a host name */
1.84 cvs 2392: path = HTCanon (url, path);
1.25 cvs 2393: }
1.106 cvs 2394: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 2395: path += 2;
2396: else
2397: path = *url;
1.222 vatton 2398: #ifndef _WINDOWS
1.84 cvs 2399: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 2400: /* Some URLs start //<foo> */
2401: path += 1;
1.222 vatton 2402: else
2403: #endif /* _WINDOWS */
2404: if (IsFilePath (path))
1.94 cvs 2405: {
2406: /* doesn't need to do anything more */
2407: return;
2408: }
1.106 cvs 2409: else if (!strncmp (path, "news:", 5))
1.28 cvs 2410: {
1.106 cvs 2411: newptr = strchr (path+5, '@');
1.28 cvs 2412: if (!newptr)
1.195 vatton 2413: newptr = path + 5;
1.28 cvs 2414: while (*newptr)
1.195 vatton 2415: {
2416: /* Make group or host lower case */
2417: *newptr = tolower (*newptr);
2418: newptr++;
2419: }
1.28 cvs 2420: /* Doesn't need to do any more */
2421: return;
1.25 cvs 2422: }
1.126 cheyroul 2423:
1.189 vatton 2424: /* remove initial multiple /'s */
2425: p = path - 1;
2426: if (path != *url && *p == used_sep && *path == used_sep)
2427: {
2428: while (*path == used_sep)
1.195 vatton 2429: {
2430: orig = path;
2431: dest = path + 1;
2432: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2433: end = orig-1;
2434: }
1.189 vatton 2435: }
2436:
2437: if (path)
1.28 cvs 2438: {
1.106 cvs 2439: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1.195 vatton 2440: (end = strchr (path, '#'))))
2441: end = path + strlen (path);
1.28 cvs 2442:
2443: /* Parse string second time to simplify */
2444: p = path;
2445: while (p < end)
1.195 vatton 2446: {
2447: /* if we're pointing to a char, it's safe to reactivate the
2448: ../ convertion */
2449: if (!ddot_simplify && *p != '.' && *p != used_sep)
2450: ddot_simplify = TRUE;
2451:
1.222 vatton 2452: if (*p == used_sep)
1.195 vatton 2453: {
2454: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
2455: {
2456: orig = p + 1;
2457: dest = (*(p+2) != used_sep) ? p+2 : p+3;
2458: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
2459: end = orig - 1;
2460: }
2461: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
2462: && (*(p+3) == used_sep || !*(p+3)))
2463: {
2464: newptr = p;
2465: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
2466: if (*newptr == used_sep)
2467: orig = newptr + 1;
2468: else
2469: orig = newptr;
2470:
2471: dest = (*(p+3) != used_sep) ? p+3 : p+4;
2472: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
2473: end = orig-1;
2474: /* Start again with prev slash */
2475: p = newptr;
2476: }
1.222 vatton 2477: else if (p != path && *(p+1) == used_sep)
1.195 vatton 2478: {
2479: while (*(p+1) == used_sep)
2480: {
2481: orig = p;
2482: dest = p + 1;
2483: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2484: end = orig-1;
2485: }
2486: }
2487: else
2488: p++;
2489: }
2490: else
2491: p++;
2492: }
2493: }
2494: /*
2495: ** Check for host/../.. kind of things
2496: */
2497: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
2498: && (!*(path+3) || *(path+3) == used_sep))
2499: *(path+1) = EOS;
1.28 cvs 2500: return;
2501: }
2502:
2503:
2504: /*----------------------------------------------------------------------
1.202 vatton 2505: NormalizeFile normalizes local names.
2506: convertion is AM_CONV_NONE or AM_CONV_ALL
1.195 vatton 2507: Return TRUE if target and src differ.
1.28 cvs 2508: ----------------------------------------------------------------------*/
1.106 cvs 2509: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2510: {
1.214 vatton 2511: #ifndef _WINDOWS
1.195 vatton 2512: char *s;
2513: int i;
1.214 vatton 2514: #endif /* _WINDOWS */
1.198 cvs 2515: int start_index; /* the first char that we'll copy */
1.195 vatton 2516: ThotBool change;
1.28 cvs 2517:
1.195 vatton 2518: change = FALSE;
2519: start_index = 0;
1.90 cvs 2520:
1.195 vatton 2521: if (!src || src[0] == EOS)
2522: {
2523: target[0] = EOS;
2524: return FALSE;
2525: }
2526:
2527: /* @@ do I need file: or file:/ here? */
2528: if (strncmp (src, "file:", 5) == 0)
2529: {
2530: /* remove the prefix file: */
2531: start_index += 5;
1.90 cvs 2532:
1.195 vatton 2533: /* remove the localhost prefix */
2534: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
2535: start_index += 11;
1.94 cvs 2536:
1.195 vatton 2537: /* remove the first two slashes in / / /path */
1.222 vatton 2538: if (src[start_index] == '/' && src[start_index + 1] == '/')
2539: start_index +=2;
2540: #ifdef _WINDOWS
1.195 vatton 2541: /* remove any extra slash before the drive name */
1.222 vatton 2542: if (src[start_index] == '/' && src[start_index+2] == ':')
1.195 vatton 2543: start_index++;
1.184 gully 2544: #endif /* _WINDOWS */
1.90 cvs 2545:
1.195 vatton 2546: if (src[start_index] == EOS)
2547: /* if there's nothing afterwards, add a DIR_STR */
2548: strcpy (target, DIR_STR);
2549: else
2550: /* as we're inside a file: URL, we'll apply all the convertions
2551: we know */
2552: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
2553:
2554: change = TRUE;
2555: }
2556: else if (convertion != AM_CONV_NONE)
2557: {
2558: /* we are following a "local" relative link, we do all the
2559: convertions except for the HOME_DIR ~ one */
2560: CleanCopyFileURL (target, src, convertion);
2561: }
1.198 cvs 2562: #ifdef _WINDOWS
1.211 vatton 2563: #ifdef IV
1.198 cvs 2564: else if (src[0] == DIR_SEP && src[1] == DIR_SEP)
2565: {
2566: s = getenv ("HOMEDRIVE");
2567: strcpy (target, s);
2568: i = strlen (target);
2569: strcpy (&target[i], &src[1]);
2570: change = TRUE;
2571: }
1.211 vatton 2572: #endif /* IV */
1.198 cvs 2573: #else /* _WINDOWS */
1.195 vatton 2574: else if (src[0] == '~')
2575: {
2576: /* it must be a URL typed in a text input field */
2577: /* do the HOME_DIR ~ substitution */
2578: s = TtaGetEnvString ("HOME");
2579: strcpy (target, s);
2580: i = strlen (target);
2581: strcpy (&target[i], &src[1]);
2582: change = TRUE;
2583: }
1.184 gully 2584: #endif /* _WINDOWS */
1.195 vatton 2585: else
2586: /* leave it as it is */
2587: strcpy (target, src);
1.96 cvs 2588:
1.195 vatton 2589: /* remove /../ and /./ */
2590: SimplifyUrl (&target);
2591: if (!change)
2592: change = (strcmp (src, target) != 0);
2593: return (change);
1.25 cvs 2594: }
2595:
1.28 cvs 2596:
1.25 cvs 2597: /*----------------------------------------------------------------------
1.31 cvs 2598: MakeRelativeURL: make relative name
1.25 cvs 2599:
1.28 cvs 2600: This function creates and returns a string which gives an expression of
2601: one address as related to another. Where there is no relation, an absolute
2602: address is retured.
1.25 cvs 2603:
1.28 cvs 2604: On entry,
1.195 vatton 2605: Both names must be absolute, fully qualified names of nodes
2606: (no fragment bits)
1.25 cvs 2607:
1.28 cvs 2608: On exit,
1.195 vatton 2609: The return result points to a newly allocated name which, if
2610: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2611: The caller is responsible for freeing the resulting name later.
1.25 cvs 2612: ----------------------------------------------------------------------*/
1.206 kia 2613: char *MakeRelativeURL (const char *aName, const char *relatedName)
1.106 cvs 2614: {
1.206 kia 2615: char *return_value;
2616: char result[MAX_LENGTH];
2617: const char *p;
2618: const char *q;
2619: const char *after_access;
2620: const char *last_slash = NULL;
2621: int slashes, levels, len;
1.184 gully 2622: #ifdef _WINDOWS
1.44 cvs 2623: int ndx;
1.184 gully 2624: #endif /* _WINDOWS */
1.44 cvs 2625:
1.29 cvs 2626: if (aName == NULL || relatedName == NULL)
2627: return (NULL);
2628:
2629: slashes = 0;
2630: after_access = NULL;
2631: p = aName;
2632: q = relatedName;
1.147 vatton 2633: len = 0;
2634: for (; *p && !strncasecmp (p, q, 1); p++, q++, len++)
1.27 cvs 2635: {
2636: /* Find extent of match */
1.106 cvs 2637: if (*p == ':')
1.195 vatton 2638: {
2639: after_access = p + 1;
1.184 gully 2640: #ifdef _WINDOWS
1.195 vatton 2641: if (len == 1)
2642: {
2643: /* it's a local Windows path like c:... */
2644: slashes+=2;
2645: }
1.184 gully 2646: #endif /* _WINDOWS */
1.195 vatton 2647: }
1.168 cvs 2648: if (*p == DIR_SEP)
1.195 vatton 2649: {
2650: /* memorize the last slash position and count them */
2651: last_slash = p;
2652: slashes++;
2653: }
1.25 cvs 2654: }
2655:
1.31 cvs 2656: /* q, p point to the first non-matching character or zero */
1.106 cvs 2657: if (*q == EOS)
1.31 cvs 2658: {
2659: /* New name is a subset of the related name */
2660: /* exactly the right length */
1.106 cvs 2661: len = strlen (p);
1.171 gully 2662: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.195 vatton 2663: strcpy (return_value, p);
1.31 cvs 2664: }
2665: else if ((slashes < 2 && after_access == NULL)
1.195 vatton 2666: || (slashes < 3 && after_access != NULL))
2667: {
1.31 cvs 2668: /* Two names whitout common path */
2669: /* exactly the right length */
1.106 cvs 2670: len = strlen (aName);
1.171 gully 2671: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.195 vatton 2672: strcpy (return_value, aName);
1.31 cvs 2673: }
2674: else
2675: {
2676: /* Some path in common */
1.218 vatton 2677: #ifdef IV
1.106 cvs 2678: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.195 vatton 2679: /* just the same server */
2680: strcpy (result, last_slash);
1.31 cvs 2681: else
1.218 vatton 2682: #endif
1.195 vatton 2683: {
2684: levels= 0;
2685: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
2686: if (*q == DIR_SEP)
2687: levels++;
1.31 cvs 2688:
1.195 vatton 2689: result[0] = EOS;
2690: for (;levels; levels--)
2691: strcat (result, "../");
2692: strcat (result, last_slash+1);
2693: }
1.52 cvs 2694:
2695: if (!*result)
1.195 vatton 2696: strcat (result, "./");
1.52 cvs 2697:
1.31 cvs 2698: /* exactly the right length */
1.106 cvs 2699: len = strlen (result);
1.171 gully 2700: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.195 vatton 2701: strcpy (return_value, result);
1.52 cvs 2702:
1.25 cvs 2703: }
1.184 gully 2704: #ifdef _WINDOWS
1.106 cvs 2705: len = strlen (return_value);
1.195 vatton 2706: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2707: if (return_value[ndx] == '\\')
1.195 vatton 2708: return_value[ndx] = '/' ;
1.184 gully 2709: #endif /* _WINDOWS */
1.29 cvs 2710: return (return_value);
1.24 cvs 2711: }
1.35 cvs 2712:
1.104 kahan 2713: /*----------------------------------------------------------------------
2714: AM_GetFileSize
2715: Returns TRUE and the filesize in the 2nd parameter.
2716: Otherwise, in case of a system error, returns FALSE, with a
2717: filesize of 0L.
2718: ---------------------------------------------------------------------*/
1.206 kia 2719: ThotBool AM_GetFileSize (const char *filename, unsigned long *file_size)
1.104 kahan 2720: {
2721: if (!TtaFileExist (filename))
2722: return FALSE;
1.185 vatton 2723: *file_size = TtaGetFileSize (filename);
2724: return TRUE;
1.104 kahan 2725: }
1.139 kahan 2726:
2727: /*----------------------------------------------------------------------
2728: AM_UseXHTMLMimeType
2729: Returns TRUE if the user has configured Amaya to use this MIME type,
2730: FALSE otherwise.
2731: ---------------------------------------------------------------------*/
2732: ThotBool AM_UseXHTMLMimeType (void)
2733: {
2734: ThotBool xhtml_mimetype;
2735:
2736: /* does the user wants to use the new MIME type? */
2737: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2738:
2739: return (xhtml_mimetype);
1.152 kahan 2740: }
2741:
1.154 kahan 2742:
2743: /********************************************
2744: The following routines were adapted from the GNU libc functions
2745: for generating a tmpnam.
2746: *********************************************/
2747:
2748: /* These are the characters used in temporary filenames. */
2749: static const char letters[] =
2750: "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
2751:
2752: /* Generate a temporary file name based on TMPL. TMPL must match the
2753: rules for mk[s]temp (i.e. end in "XXXXXX"). The name constructed
2754: does not exist at the time of the call to __gen_tempname. TMPL is
2755: overwritten with the result.
2756:
2757: We use a clever algorithm to get hard-to-predict names. */
2758: void
2759: AM_gen_tempname (char *tmpl)
2760: {
2761: int len;
2762: char *XXXXXX;
2763: static uint64_t value;
2764: uint64_t random_time_bits;
2765: unsigned int count;
2766: int save_errno = errno;
2767: struct stat st;
2768:
2769: /* A lower bound on the number of temporary files to attempt to
2770: generate. The maximum total number of temporary file names that
2771: can exist for a given template is 62**6. It should never be
2772: necessary to try all these combinations. Instead if a reasonable
2773: number of names is tried (we define reasonable as 62**3) fail to
2774: give the system administrator the chance to remove the problems. */
2775: unsigned int attempts_min = 62 * 62 * 62;
2776:
2777: /* The number of times to attempt to generate a temporary file. To
2778: conform to POSIX, this must be no smaller than TMP_MAX. */
2779: unsigned int attempts = attempts_min < TMP_MAX ? TMP_MAX : attempts_min;
2780:
2781: len = strlen (tmpl);
2782: if (len < 6 || strcmp (&tmpl[len - 6], "XXXXXX"))
2783: {
2784: /* @@ JK ? */
2785: errno = EINVAL;
2786: return;
2787: }
2788:
2789: /* This is where the Xs start. */
2790: XXXXXX = &tmpl[len - 6];
2791:
2792: /* Get some more or less random data. */
2793: #ifdef RANDOM_BITS
2794: RANDOM_BITS (random_time_bits);
2795: #else
2796: # if HAVE_GETTIMEOFDAY || _LIBC
2797: {
2798: struct timeval tv;
2799: gettimeofday (&tv, NULL);
2800: random_time_bits = ((uint64_t) tv.tv_usec << 16) ^ tv.tv_sec;
2801: }
2802: # else
2803: random_time_bits = time (NULL);
2804: # endif
2805: #endif
2806: value += random_time_bits ^ getpid ();
2807:
2808: for (count = 0; count < attempts; value += 7777, ++count)
2809: {
2810: uint64_t v = value;
2811:
2812: /* Fill in the random bits. */
2813: XXXXXX[0] = letters[v % 62];
2814: v /= 62;
2815: XXXXXX[1] = letters[v % 62];
2816: v /= 62;
2817: XXXXXX[2] = letters[v % 62];
2818: v /= 62;
2819: XXXXXX[3] = letters[v % 62];
2820: v /= 62;
2821: XXXXXX[4] = letters[v % 62];
2822: v /= 62;
2823: XXXXXX[5] = letters[v % 62];
2824:
2825: /* This case is backward from the other three. AM_gen_tempname
1.195 vatton 2826: succeeds if __xstat fails because the name does not exist.
2827: Note the continue to bypass the common logic at the bottom
2828: of the loop. */
1.154 kahan 2829: if (stat (tmpl, &st) < 0)
1.195 vatton 2830: break;
1.154 kahan 2831:
2832: continue;
2833: }
2834:
2835: if (count == attempts || errno != ENOENT)
2836: tmpl[0] = EOS;
2837: else
2838: errno = save_errno;
2839:
2840: return;
2841: }
2842:
2843: #define JOSE 1
2844:
1.152 kahan 2845: /*-----------------------------------------------------------------------
2846: GetTempName
2847: Front end to the Unix tempnam function, which is independent of the
2848: value of the TMPDIR env value
2849: Returns a dynamically allocated string with a tempname. The user
2850: must free this memory.
2851: -----------------------------------------------------------------------*/
2852: char *GetTempName (const char *dir, const char *prefix)
2853: {
1.154 kahan 2854: #ifdef JOSE
2855:
1.162 kahan 2856: static char tmpbufmem[PATH_MAX + 1];
1.154 kahan 2857: int len;
2858: int i;
2859:
1.155 cvs 2860: if (!dir || *dir == EOS || !TtaDirExists (dir))
1.154 kahan 2861: return NULL;
2862:
1.162 kahan 2863: /* make sure that the name is no bigger than PATH_MAX + the 6 tempname chars we
1.195 vatton 2864: will add */
1.154 kahan 2865:
1.156 cvs 2866: len = strlen (dir);
1.162 kahan 2867: if (len + 6 > PATH_MAX)
1.154 kahan 2868: return NULL;
2869:
2870: /* copy the dir name, and add a DIR_SEP if it's missing */
2871: if (dir[strlen (dir) - 1] == DIR_SEP)
2872: strcpy (tmpbufmem, dir);
2873: else
1.195 vatton 2874: {
2875: sprintf (tmpbufmem, "%s%c", dir, DIR_SEP);
2876: len++;
2877: }
1.154 kahan 2878:
1.161 kahan 2879: /* copy the prefix (no more than L_tmpnam chars, to respect POSIX). Save
1.156 cvs 2880: space for the 6 X and EOS chars that will become the random bits */
2881: if (prefix)
1.195 vatton 2882: {
1.156 cvs 2883: i = 0;
1.195 vatton 2884: while (prefix[i] != EOS && i < L_tmpnam - 8)
2885: tmpbufmem[len++] = prefix[i++];
2886: tmpbufmem[len] = EOS;
2887: }
1.156 cvs 2888:
2889: /* Add the 6 X chars */
2890: len = strlen (tmpbufmem);
2891: i = 0;
2892: while (i < 6)
1.195 vatton 2893: {
2894: tmpbufmem[len++] = 'X';
2895: i++;
2896: }
1.156 cvs 2897: tmpbufmem[len] = EOS;
1.154 kahan 2898:
2899: AM_gen_tempname (tmpbufmem);
2900:
2901: if (tmpbufmem[0] == EOS)
2902: return NULL;
2903: else
2904: return (TtaStrdup (tmpbufmem));
2905:
2906: #else
1.152 kahan 2907: char *tmpdir;
1.181 vatton 2908: char *tmp = NULL;
1.152 kahan 2909: char *name = NULL;
2910:
2911: /* save the value of TMPDIR */
1.181 vatton 2912: tmpdir = getenv ("TMPDIR");
1.152 kahan 2913: if (tmpdir)
2914: {
1.181 vatton 2915: /* remove TMPDIR from the environment */
2916: tmp = TtaGetMemory (strlen (tmpdir) + 20);
2917: sprintf (tmp, "TMPDIR=");
1.184 gully 2918: #ifdef _WINDOWS
1.152 kahan 2919: _putenv (tmp);
2920: #else
2921: putenv (tmp);
1.184 gully 2922: #endif /* _WINDOWS */
1.181 vatton 2923: /* prepare the string to restore the value of TMPDIR */
2924: strrcat (tmp, tmpdir);
1.152 kahan 2925: }
2926:
2927: /* create the tempname */
1.184 gully 2928: #ifdef _WINDOWS
1.152 kahan 2929: /* Under Windows, _tempnam returns the same name until the file is created */
2930: {
2931: char *altprefix;
2932: name = tmpnam (NULL); /* get a possibly unique string */
1.181 vatton 2933: altprefix = TtaGetMemory (strlen (prefix) + strlen(name) + 1);
1.152 kahan 2934: sprintf (altprefix, "%s%s", prefix, name + strlen(_P_tmpdir));
2935: name = _tempnam (dir, altprefix); /* get a name that isn't yet in use */
2936: TtaFreeMemory (altprefix);
2937: }
2938: #else
2939: name = tempnam (dir, prefix);
1.184 gully 2940: #endif /* _WINDOWS */
1.152 kahan 2941:
2942: if (tmpdir)
2943: {
1.181 vatton 2944: /* restore the value of TMPDIR */
1.184 gully 2945: #ifdef _WINDOWS
1.152 kahan 2946: _putenv (tmpdir);
2947: #else
2948: putenv (tmpdir);
1.184 gully 2949: #endif /* _WINDOWS */
1.181 vatton 2950: TtaFreeMemory (tmpdir);
1.152 kahan 2951: }
2952: return (name);
1.154 kahan 2953: #endif
1.139 kahan 2954: }
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