Annotation of Amaya/amaya/AHTURLTools.c, revision 1.221
1.7 cvs 1: /*
2: *
1.204 vatton 3: * (c) COPYRIGHT INRIA and W3C, 1996-2008
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: /*----------------------------------------------------------------------
1.136 cvs 852: IsUndisplayedName
853: returns TRUE if path points to an undisplayed resource.
1.103 cvs 854: ----------------------------------------------------------------------*/
1.136 cvs 855: ThotBool IsUndisplayedName (const char *path)
1.103 cvs 856: {
1.195 vatton 857: char *temppath;
858: char *suffix;
859: ThotBool ret;
860:
861: if (!path)
862: return (FALSE);
863:
864: temppath = TtaStrdup ((char *)path);
865: suffix = (char *)TtaGetMemory (strlen (path) + 1);
866: TtaExtractSuffix (temppath, suffix);
867:
868: if (!strcasecmp (suffix, "exe") ||
869: !strcasecmp (suffix, "zip") ||
870: !strcasecmp (suffix, "ppt") ||
871: !strcasecmp (suffix, "pdf") ||
872: !strcasecmp (suffix, "ps") ||
873: !strcasecmp (suffix, "eps") ||
874: !strcasecmp (suffix, "tar") ||
875: !strcasecmp (suffix, "tgz") ||
876: !strcasecmp (suffix, "ddl") ||
1.196 vatton 877: !strcasecmp (suffix, "deb") ||
878: !strcasecmp (suffix, "dmg") ||
1.195 vatton 879: !strcasecmp (suffix, "mpg") ||
880: !strcasecmp (suffix, "mpeg") ||
1.196 vatton 881: !strcasecmp (suffix, "rpm") ||
1.195 vatton 882: !strcasecmp (suffix, "wmv") ||
883: !strcasecmp (suffix, "wma") ||
884: !strcasecmp (suffix, "o"))
885: ret = TRUE;
886: else if (!strcmp (suffix, "gz"))
887: {
888: /* take into account compressed files */
889: TtaExtractSuffix (temppath, suffix);
890: if (!strcasecmp (suffix, "exe") ||
891: !strcasecmp (suffix, "zip") ||
892: !strcasecmp (suffix, "ppt") ||
893: !strcasecmp (suffix, "pdf") ||
894: !strcasecmp (suffix, "ps") ||
895: !strcasecmp (suffix, "eps") ||
896: !strcasecmp (suffix, "tar") ||
897: !strcasecmp (suffix, "ddl") ||
1.196 vatton 898: !strcasecmp (suffix, "dmg") ||
1.195 vatton 899: !strcasecmp (suffix, "mpg") ||
900: !strcasecmp (suffix, "mpeg") ||
901: !strcasecmp (suffix, "wmv") ||
902: !strcasecmp (suffix, "wma") ||
903: !strcasecmp (suffix, "o"))
904: ret = TRUE;
905: else
906: ret = FALSE;
907: }
908: else
909: ret = FALSE;
910:
911: TtaFreeMemory (temppath);
912: TtaFreeMemory (suffix);
913: return (ret);
1.103 cvs 914: }
915:
916: /*----------------------------------------------------------------------
1.60 cvs 917: IsCSSName
918: returns TRUE if path points to an XML resource.
919: ----------------------------------------------------------------------*/
1.111 cvs 920: ThotBool IsCSSName (const char *path)
1.60 cvs 921: {
1.195 vatton 922: char *temppath;
923: char *suffix;
924: ThotBool ret;
925:
926: if (!path)
927: return (FALSE);
928:
929: temppath = TtaStrdup ((char *)path);
930: suffix = (char *)TtaGetMemory (strlen (path) + 1);
931: TtaExtractSuffix (temppath, suffix);
932:
933: if (!strcasecmp (suffix, "css"))
934: ret = TRUE;
935: else if (!strcmp (suffix, "gz"))
936: {
937: /* take into account compressed files */
938: TtaExtractSuffix (temppath, suffix);
939: if (!strcasecmp (suffix, "css"))
940: ret = TRUE;
941: else
942: ret = FALSE;
943: }
944: else
945: ret = FALSE;
946:
947: TtaFreeMemory (temppath);
948: TtaFreeMemory (suffix);
949: return (ret);
1.56 cvs 950: }
951:
952: /*----------------------------------------------------------------------
1.177 kahan 953: MultipleBookmarks
954: returns the value of the multiple bookmarks environment variable.
955: (TRUE if enabled).
956: ----------------------------------------------------------------------*/
957: ThotBool MultipleBookmarks (void)
958: {
959: ThotBool multiple_bm;
960:
1.180 kahan 961: TtaGetEnvBoolean ("DISABLE_MULTIPLE_BM", &multiple_bm);
1.177 kahan 962:
1.180 kahan 963: return (multiple_bm == FALSE);
1.177 kahan 964: }
965:
966: /*----------------------------------------------------------------------
1.172 kahan 967: IsRDFName
968: returns TRUE if path points to an RDF resource.
969: ----------------------------------------------------------------------*/
970: ThotBool IsRDFName (const char *path)
971: {
1.195 vatton 972: char *temppath;
973: char *suffix;
974: ThotBool ret;
975:
976: /* temporarily disabling this function */
977: if (!MultipleBookmarks ())
978: return (FALSE);
979:
980: if (!path)
981: return (FALSE);
982:
983: temppath = TtaStrdup ((char *)path);
984: suffix = (char *)TtaGetMemory (strlen (path) + 1);
985: TtaExtractSuffix (temppath, suffix);
986:
987: if (!strcasecmp (suffix, "rdf"))
988: ret = TRUE;
989: else if (!strcmp (suffix, "gz"))
990: {
991: /* take into account compressed files */
992: TtaExtractSuffix (temppath, suffix);
993: if (!strcasecmp (suffix, "rdf"))
994: ret = TRUE;
995: else
996: ret = FALSE;
997: }
998: else
999: ret = FALSE;
1.183 quint 1000:
1.195 vatton 1001: TtaFreeMemory (temppath);
1002: TtaFreeMemory (suffix);
1003: return (ret);
1.172 kahan 1004: }
1005:
1006: /*----------------------------------------------------------------------
1.9 cvs 1007: IsImageName
1008: returns TRUE if path points to an image resource.
1.4 cvs 1009: ----------------------------------------------------------------------*/
1.111 cvs 1010: ThotBool IsImageName (const char *path)
1.106 cvs 1011: {
1.195 vatton 1012: char *temppath;
1013: char *suffix;
1014: char nsuffix[MAX_LENGTH];
1015: int i;
1016: ThotBool ret;
1017:
1018: if (!path)
1019: return (FALSE);
1020:
1021: temppath = TtaStrdup ((char *)path);
1022: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1023: TtaExtractSuffix (temppath, suffix);
1024:
1025: /* Normalize the suffix */
1026: i = 0;
1027: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1028: {
1029: nsuffix[i] = tolower (suffix[i]);
1030: i++;
1031: }
1032: nsuffix[i] = EOS;
1033: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
1034: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
1035: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1036: ret = TRUE;
1037: else
1038: ret = FALSE;
1039:
1040: TtaFreeMemory (temppath);
1041: TtaFreeMemory (suffix);
1042: return (ret);
1.3 cvs 1043: }
1044:
1.4 cvs 1045: /*----------------------------------------------------------------------
1.58 cvs 1046: IsImageType
1047: returns TRUE if type points to an image resource.
1048: ----------------------------------------------------------------------*/
1.111 cvs 1049: ThotBool IsImageType (const char *type)
1.58 cvs 1050: {
1.195 vatton 1051: char *temptype;
1052: int i;
1053: ThotBool ret;
1054:
1055: if (!type)
1056: return (FALSE);
1057:
1058: temptype = TtaStrdup ((char *)type);
1059: /* Normalize the type */
1060: i = 0;
1061: while (temptype[i] != EOS)
1062: {
1063: temptype[i] = tolower (temptype[i]);
1064: i++;
1065: }
1.166 vatton 1066: if (!strncmp (temptype, "image/", sizeof ("image/") - 1))
1.195 vatton 1067: i = sizeof ("image/") - 1;
1068: else
1069: i = 0;
1070: if (!strcmp (&temptype[i], "gif") ||
1071: !strcmp (&temptype[i], "x-xbitmap") ||
1072: !strcmp (&temptype[i], "x-xpixmap") ||
1073: !strcmp (&temptype[i], "jpeg") ||
1074: !strcmp (&temptype[i], "png"))
1075: ret = TRUE;
1076: else
1077: ret = FALSE;
1078: TtaFreeMemory (temptype);
1079: return (ret);
1.58 cvs 1080: }
1081:
1082: /*----------------------------------------------------------------------
1.9 cvs 1083: IsTextName
1.4 cvs 1084: ----------------------------------------------------------------------*/
1.111 cvs 1085: ThotBool IsTextName (const char *path)
1.106 cvs 1086: {
1.195 vatton 1087: char *temppath;
1088: char *suffix;
1089: char nsuffix[MAX_LENGTH];
1090: int i;
1091: ThotBool ret;
1092:
1093: if (!path)
1094: return (FALSE);
1095:
1096: temppath = TtaStrdup ((char *)path);
1097: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1098: TtaExtractSuffix (temppath, suffix);
1099:
1100: /* Normalize the suffix */
1101: i = 0;
1102: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1103: {
1104: nsuffix[i] = tolower (suffix[i]);
1105: i++;
1106: }
1107: nsuffix[i] = EOS;
1108:
1109: if (!strcmp (nsuffix, "txt") ||
1110: !strcmp (nsuffix, "dtd"))
1111: ret = TRUE;
1112: else if (!strcmp (nsuffix, "gz"))
1113: {
1114: /* take into account compressed files */
1115: TtaExtractSuffix (temppath, suffix);
1116: /* Normalize the suffix */
1117: i = 0;
1118: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1119: {
1120: nsuffix[i] = tolower (suffix[i]);
1121: i++;
1122: }
1123: nsuffix[i] = EOS;
1124: if (!strcmp (nsuffix, "txt") ||
1125: !strcmp (nsuffix, "dtd"))
1126: ret = TRUE;
1127: else
1128: ret = FALSE;
1129: }
1130: else
1131: ret = FALSE;
1132:
1133: TtaFreeMemory (temppath);
1134: TtaFreeMemory (suffix);
1135: return (ret);
1.3 cvs 1136: }
1137:
1.4 cvs 1138: /*----------------------------------------------------------------------
1.9 cvs 1139: IsHTTPPath
1140: returns TRUE if path is in fact an http URL.
1.4 cvs 1141: ----------------------------------------------------------------------*/
1.112 cvs 1142: ThotBool IsHTTPPath (const char *path)
1.3 cvs 1143: {
1.195 vatton 1144: if (!path)
1145: return FALSE;
1.3 cvs 1146:
1.195 vatton 1147: if ((!strncmp (path, "http:", 5) != 0)
1.204 vatton 1148: || !strncmp (path, "ftp:", 4)
1.195 vatton 1149: || !strncmp (path, "internal:", 9))
1150: return TRUE;
1151: return FALSE;
1.3 cvs 1152: }
1153:
1.4 cvs 1154: /*----------------------------------------------------------------------
1.9 cvs 1155: IsWithParameters
1156: returns TRUE if url has a concatenated query string.
1.4 cvs 1157: ----------------------------------------------------------------------*/
1.133 vatton 1158: ThotBool IsWithParameters (const char *url)
1.3 cvs 1159: {
1.195 vatton 1160: int i;
1161:
1162: if ((!url) || (url[0] == EOS))
1163: return FALSE;
1.3 cvs 1164:
1.195 vatton 1165: i = strlen (url) - 1;
1166: while (i > 0 && url[i--] != '?')
1167: if (i < 0)
1.5 cvs 1168: return FALSE;
1.3 cvs 1169:
1.195 vatton 1170: /* There is a parameter */
1171: return TRUE;
1.3 cvs 1172: }
1173:
1.4 cvs 1174: /*----------------------------------------------------------------------
1.9 cvs 1175: IsW3Path
1176: returns TRUE if path is in fact a URL.
1.4 cvs 1177: ----------------------------------------------------------------------*/
1.133 vatton 1178: ThotBool IsW3Path (const char *path)
1.106 cvs 1179: {
1.170 quint 1180: if (path == NULL)
1181: return FALSE;
1.106 cvs 1182: if (strncmp (path, "http:", 5) &&
1183: strncmp (path, "ftp:", 4) &&
1184: strncmp (path, "telnet:", 7) &&
1185: strncmp (path, "wais:", 5) &&
1186: strncmp (path, "news:", 5) &&
1187: strncmp (path, "gopher:", 7) &&
1188: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 1189: strncmp (path, "archie:", 7) &&
1190: strncmp (path, "https:", 6))
1.72 cvs 1191: return FALSE;
1192: return TRUE;
1.3 cvs 1193: }
1194:
1.4 cvs 1195: /*----------------------------------------------------------------------
1.90 cvs 1196: IsFilePath
1197: returns TRUE if path is in fact a URL.
1198: ----------------------------------------------------------------------*/
1.133 vatton 1199: ThotBool IsFilePath (const char *path)
1.90 cvs 1200: {
1.106 cvs 1201: if (strncmp (path, "file:", 5))
1.90 cvs 1202: return FALSE;
1203: return TRUE;
1204: }
1205:
1206: /*----------------------------------------------------------------------
1.9 cvs 1207: IsValidProtocol
1208: returns true if the url protocol is supported by Amaya.
1.4 cvs 1209: ----------------------------------------------------------------------*/
1.133 vatton 1210: ThotBool IsValidProtocol (const char *url)
1.106 cvs 1211: {
1.195 vatton 1212: if (!strncmp (url, "http:", 5)
1.106 cvs 1213: || !strncmp (url, "internal:", 9)
1.204 vatton 1214: || !strncmp (url, "ftp:", 4))
1.195 vatton 1215: /* experimental */
1216: /*** || !strncmp (path, "news:", 5)***/
1217: return (TRUE);
1218: else
1219: return (FALSE);
1.3 cvs 1220: }
1221:
1.31 cvs 1222:
1223: /*----------------------------------------------------------------------
1.195 vatton 1224: GetBaseURL
1225: normalizes orgName according to a base associated with doc, and
1226: following the standard URL format rules.
1227: The function returns the base used to solve relative URL and SRC:
1228: - the base of the document,
1229: - or the document path (without document name).
1.31 cvs 1230: ----------------------------------------------------------------------*/
1.106 cvs 1231: char *GetBaseURL (Document doc)
1.31 cvs 1232: {
1233: Element el;
1234: ElementType elType;
1235: AttributeType attrType;
1236: Attribute attr;
1.176 vatton 1237: char *ptr, *basename, *utf8path;
1.217 vatton 1238: char documentname[MAX_LENGTH];
1.31 cvs 1239: int length;
1.151 kahan 1240: ThotBool hasDocBase;
1.31 cvs 1241:
1.113 cvs 1242: if (doc == 0 || !DocumentURLs[doc])
1.195 vatton 1243: return NULL;
1.148 kahan 1244: /* the other functions expect basename to have no more than MAX_LENGTH chars */
1.171 gully 1245: basename = (char *)TtaGetMemory (MAX_LENGTH);
1.148 kahan 1246: basename[0] = EOS;
1.31 cvs 1247: length = MAX_LENGTH -1;
1.151 kahan 1248: hasDocBase = FALSE;
1249:
1250: /* If the document has a base URL, it has a priority over the headers. */
1251: /* @@ We need to do this too when we support XML:base */
1252:
1253: /* is it a HTML document ? */
1254: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1255: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
1256: /* it's a HTML document */
1257: {
1258: /* get the document element */
1259: el = TtaGetMainRoot (doc);
1260: /* search the BASE element */
1261: elType.ElTypeNum = HTML_EL_HEAD;
1262: el = TtaSearchTypedElement (elType, SearchForward, el);
1263: if (el)
1.195 vatton 1264: /* there is a HEAD element */
1265: {
1266: /* look for a BASE element within the HEAD */
1267: elType.ElTypeNum = HTML_EL_BASE;
1268: el = TtaSearchTypedElement (elType, SearchInTree, el);
1269: }
1.151 kahan 1270: if (el)
1.195 vatton 1271: {
1272: /* The document has a BASE element. Get the HREF attribute of the
1273: BASE element */
1274: hasDocBase = TRUE;
1275: attrType.AttrSSchema = elType.ElSSchema;
1276: attrType.AttrTypeNum = HTML_ATTR_HREF_;
1277: attr = TtaGetAttribute (el, attrType);
1278: if (attr)
1279: {
1280: /* Use the base path of the document */
1281: TtaGiveTextAttributeValue (attr, basename, &length);
1282: utf8path = (char *)TtaConvertMbsToByte ((unsigned char *)basename,
1283: TtaGetDefaultCharset ());
1.217 vatton 1284: if (IsW3Path (DocumentURLs[doc]) && utf8path[0] == '/')
1285: NormalizeURL (utf8path, 0, basename, documentname, DocumentURLs[doc]);
1286: else
1287: strncpy (basename, utf8path, MAX_LENGTH-1);
1.195 vatton 1288: TtaFreeMemory (utf8path);
1289: }
1290: }
1.151 kahan 1291: }
1292:
1293: /* there was no BASE. Do we have a location header? */
1294: if (!hasDocBase && DocumentMeta[doc] && DocumentMeta[doc]->full_content_location
1.148 kahan 1295: && DocumentMeta[doc]->full_content_location[0] != EOS)
1.65 cvs 1296: {
1.148 kahan 1297: strncpy (basename, DocumentMeta[doc]->full_content_location, MAX_LENGTH-1);
1298: basename[MAX_LENGTH-1] = EOS;
1299: length = strlen (basename);
1300: }
1301:
1302: if (basename[0] != EOS)
1303: {
1304: /* base and orgName have to be separated by a DIR_SEP */
1305: length--;
1306: if (basename[0] != EOS && basename[length] != URL_SEP &&
1.195 vatton 1307: basename[length] != DIR_SEP)
1308: /* verify if the base has the form "protocol://server:port" */
1309: {
1310: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1311: AMAYA_PARSE_HOST |
1312: AMAYA_PARSE_PUNCTUATION);
1313: if (ptr && !strcmp (ptr, basename))
1314: {
1315: /* it has this form, complete it by adding a URL_STR */
1316: if (strchr (basename, DIR_SEP))
1317: strcat (basename, DIR_STR);
1318: else
1319: strcat (basename, URL_STR);
1320: length++;
1321: }
1322: else if (!ptr || ptr[0] == EOS)
1323: {
1324: /* no host was detected, we may have a relative URL. We test
1325: if it begins with a URL_SEP, DIR_SEP or period. If yes, it's
1326: relative. */
1327: if (! (basename[0] == '.' || basename[0] == URL_SEP
1328: || basename[0] == DIR_SEP))
1329: basename[0] = EOS;
1330: }
1331: if (ptr)
1332: TtaFreeMemory (ptr);
1333: }
1.113 cvs 1334: }
1335:
1.148 kahan 1336: /* there was no base element and no location header, we use the DocumentURL */
1337: if (basename[0] == EOS)
1338: {
1339: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1340: basename[MAX_LENGTH-1] = EOS;
1341: }
1342:
1.31 cvs 1343: /* Remove anything after the last DIR_SEP char. If no such char is found,
1344: * then search for the first ":" char, hoping that what's before that is a
1345: * protocol. If found, end the string there. If neither char is found,
1346: * then discard the whole base element.
1347: */
1.106 cvs 1348: length = strlen (basename) - 1;
1.31 cvs 1349: /* search for the last DIR_SEP char */
1.220 vatton 1350: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 1351: length--;
1.221 ! vatton 1352: if (length > 1 && basename[length-1] == URL_SEP && basename[length-2] == ':')
1.219 vatton 1353: return (basename);
1.31 cvs 1354: if (length >= 0)
1355: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1356: basename[length + 1] = EOS;
1.31 cvs 1357: else
1358: /* search for the first PATH_STR char */
1359: {
1.106 cvs 1360: for (length = 0; basename[length] != ':' &&
1.195 vatton 1361: basename[length] != EOS; length ++);
1.106 cvs 1362: if (basename[length] == ':')
1.195 vatton 1363: /* found, so end the string there */
1364: basename[length + 1] = EOS;
1.31 cvs 1365: else
1.195 vatton 1366: /* not found, discard the base */
1367: basename[0] = EOS;
1.31 cvs 1368: }
1369: return (basename);
1370: }
1371:
1372:
1.4 cvs 1373: /*----------------------------------------------------------------------
1.195 vatton 1374: GetLocalPath
1375: Allocate and return the local document path associated to the url
1.40 cvs 1376: ----------------------------------------------------------------------*/
1.150 vatton 1377: char *GetLocalPath (Document doc, char *url)
1.106 cvs 1378: {
1379: char *ptr;
1380: char *n;
1381: char *documentname;
1382: char url_sep;
1.83 cvs 1383: int len;
1.67 cvs 1384: ThotBool noFile;
1.40 cvs 1385:
1.153 vatton 1386: if (url)
1.40 cvs 1387: {
1388: /* check whether the file name exists */
1.106 cvs 1389: len = strlen (url) - 1;
1.71 cvs 1390: if (IsW3Path (url))
1.195 vatton 1391: url_sep = URL_SEP;
1.41 cvs 1392: else
1.195 vatton 1393: url_sep = DIR_SEP;
1.41 cvs 1394: noFile = (url[len] == url_sep);
1.40 cvs 1395: if (noFile)
1.195 vatton 1396: url[len] = EOS;
1.171 gully 1397: ptr = (char *)TtaGetMemory (MAX_LENGTH);
1398: documentname = (char *)TtaGetMemory (MAX_LENGTH);
1.78 cvs 1399: TtaExtractName (url, ptr, documentname);
1.106 cvs 1400: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1401: if (!TtaCheckDirectory (ptr))
1.195 vatton 1402: /* directory did not exist */
1403: TtaMakeDirectory (ptr);
1.47 cvs 1404:
1.153 vatton 1405: if (doc == 0)
1.195 vatton 1406: {
1407: n = strrchr (documentname, '.');
1408: if (n)
1409: *n = EOS;
1410: if (documentname[0] == EOS)
1411: strcpy (documentname, "noname");
1412: n = GetTempName (ptr, documentname);
1413: TtaFreeMemory (ptr);
1414: ptr = n;
1415: }
1.69 cvs 1416: else
1.195 vatton 1417: {
1418: /* don't include the query string within document name */
1419: n = strrchr (documentname, '?');
1420: if (n)
1421: *n = EOS;
1422: /* don't include ':' within document name */
1423: n = strchr (documentname, ':');
1424: if (n)
1425: *n = EOS;
1426: /* if after all this operations document name
1427: is empty, let's use noname.html instead */
1428: if (documentname[0] == EOS)
1429: strcat (ptr, "noname.html");
1430: else
1431: strcat (ptr, documentname);
1432: }
1.40 cvs 1433: TtaFreeMemory (documentname);
1.157 kahan 1434: /* substitute invalid chars in file names by a _ */
1.188 gully 1435: n = strrchr(ptr,DIR_SEP);
1.157 kahan 1436: while (*n)
1.195 vatton 1437: {
1438: if (*n == '*' || *n == ',')
1439: *n = '_';
1440: if ((unsigned char)*n >= 0x80) /* avoid non-ASCII */
1441: *n = 'A' + ((unsigned char)*n % 26);
1442: n++;
1443: }
1.40 cvs 1444: /* restore the url */
1445: if (noFile)
1.195 vatton 1446: url[len] = url_sep;
1.40 cvs 1447: return (ptr);
1448: }
1449: else
1450: return (NULL);
1451: }
1452:
1.73 cvs 1453: /*----------------------------------------------------------------------
1.195 vatton 1454: ExtractTarget extract the target name from document nane.
1.79 cvs 1455: ----------------------------------------------------------------------*/
1.150 vatton 1456: void ExtractTarget (char *aName, char *target)
1.79 cvs 1457: {
1.195 vatton 1458: long int lg, i;
1459: char *ptr;
1460: char *oldptr;
1461:
1462: if (!target || !aName)
1463: /* bad target */
1464: return;
1465:
1466: target[0] = EOS;
1467: lg = strlen (aName);
1468: if (lg)
1469: {
1470: /* the name is not empty */
1471: oldptr = ptr = &aName[0];
1472: do
1473: {
1474: ptr = strrchr (oldptr, '#');
1475: if (ptr)
1476: oldptr = &ptr[1];
1477: }
1478: while (ptr);
1479:
1480: i = (long int) (oldptr) - (long int) (aName); /* name length */
1481: if (i > 1)
1482: {
1483: aName[i - 1] = EOS;
1484: if (i != lg)
1485: strcpy (target, oldptr);
1486: }
1487: }
1.79 cvs 1488: }
1489:
1490: /*----------------------------------------------------------------------
1.195 vatton 1491: RemoveNewLines (text)
1492: Removes any '\n' chars that are found in text.
1493: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1494: ----------------------------------------------------------------------*/
1.106 cvs 1495: ThotBool RemoveNewLines (char *text)
1496: {
1497: ThotBool change = FALSE;
1498: char *src;
1499: char *dest;
1.90 cvs 1500:
1501: src = text;
1502: dest = text;
1.115 kahan 1503:
1504: /* remove any preceding whitespace */
1505: while (*src && *src == ' ')
1506: {
1507: src++;
1508: change = 1;
1509: }
1510:
1.90 cvs 1511: while (*src)
1512: {
1513: switch (*src)
1.195 vatton 1514: {
1515: case '\n':
1516: /* don't copy the newline */
1517: change = 1;
1518: break;
1519: default:
1520: *dest = *src;
1521: dest++;
1522: break;
1523: }
1.90 cvs 1524: src++;
1525: }
1526: /* copy the last EOS char */
1527: *dest = *src;
1528:
1529: return (change);
1530: }
1531:
1532: /*----------------------------------------------------------------------
1.195 vatton 1533: CleanCopyFileURL
1534: Copies a file url from a src string to destination string.
1535: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1536: we want to do).
1.90 cvs 1537: ----------------------------------------------------------------------*/
1.106 cvs 1538: static void CleanCopyFileURL (char *dest, char *src,
1.195 vatton 1539: ConvertionType convertion)
1.90 cvs 1540: {
1541: while (*src)
1.89 cvs 1542: {
1.90 cvs 1543: switch (*src)
1.195 vatton 1544: {
1.184 gully 1545: #ifdef _WINDOWS
1.195 vatton 1546: case URL_SEP:
1547: /* make DIR_SEP transformation */
1548: if (convertion & AM_CONV_URL_SEP)
1549: *dest = DIR_SEP;
1550: else
1551: *dest = *src;
1552: dest++;
1553: src++;
1554: break;
1.184 gully 1555: #endif /* _WINDOWS */
1.96 cvs 1556:
1.195 vatton 1557: case '%':
1558: if (convertion & AM_CONV_PERCENT)
1559: {
1560: /* (code adapted from libwww's HTUnEscape function */
1561: src++;
1562: if (*src != EOS)
1563: {
1564: *dest = UnEscapeChar (*src) * 16;
1565: src++;
1566: }
1567: if (*src != EOS)
1568: {
1569: *dest = *dest + UnEscapeChar (*src);
1570: src++;
1571: }
1572: dest++;
1573: }
1574: else
1575: {
1576: *dest = *src;
1577: dest++;
1578: src++;
1579: }
1580: break;
1581:
1582: default:
1583: *dest = *src;
1584: dest++;
1585: src++;
1586: break;
1587: }
1.89 cvs 1588: }
1.90 cvs 1589: /* copy the EOS char */
1590: *dest = *src;
1.73 cvs 1591: }
1.40 cvs 1592:
1593: /*----------------------------------------------------------------------
1.195 vatton 1594: NormalizeURL
1595: normalizes orgName according to a base associated with doc, and
1596: following the standard URL format rules.
1597: if doc is < 0, use as a base the URL of the document that contains
1598: (or contained) the elements that are now in the copy/cut buffer.
1599: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1600: other path.
1601: The function returns the new complete and normalized URL
1602: or file name path (newName) and the name of the document (docName).
1603: N.B. If the function can't find out what's the docName, it assigns
1604: the name "noname.html".
1.4 cvs 1605: ----------------------------------------------------------------------*/
1.106 cvs 1606: void NormalizeURL (char *orgName, Document doc, char *newName,
1.206 kia 1607: char *docName, const char *otherPath)
1.106 cvs 1608: {
1.195 vatton 1609: char *basename;
1610: char tempOrgName[MAX_LENGTH];
1611: char *ptr;
1612: char used_sep;
1613: int length;
1614: ThotBool check;
1.5 cvs 1615:
1.184 gully 1616: #ifdef _WINDOWS
1.195 vatton 1617: int ndx;
1.184 gully 1618: #endif /* _WINDOWS */
1.44 cvs 1619:
1.195 vatton 1620: if (!newName || !docName)
1621: return;
1622:
1623: if (doc < 0)
1624: basename = TtaStrdup (SavedDocumentURL);
1625: else if (doc > 0)
1626: basename = GetBaseURL (doc);
1.217 vatton 1627: else if (otherPath)
1.195 vatton 1628: basename = TtaStrdup (otherPath);
1629: else
1630: basename = NULL;
1631:
1632: /*
1633: * Clean orgName
1634: * Make sure we have a complete orgName, without any leading or trailing
1635: * white spaces, or trailinbg new lines
1636: */
1637: ptr = orgName;
1638: /* skip leading white space and new line characters */
1639: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1640: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1641: tempOrgName[MAX_LENGTH -1] = EOS;
1642: /*
1643: * Make orgName a complete URL
1644: * If the URL does not include a protocol, then try to calculate
1645: * one using the doc's base element (if it exists),
1646: */
1647: if (tempOrgName[0] == EOS)
1648: {
1649: newName[0] = EOS;
1650: docName[0] = EOS;
1651: TtaFreeMemory (basename);
1.5 cvs 1652: return;
1.195 vatton 1653: }
1654:
1655: /* clean trailing white space */
1656: length = strlen (tempOrgName) - 1;
1657: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1658: {
1659: tempOrgName[length] = EOS;
1660: length--;
1661: }
1.18 cvs 1662:
1.195 vatton 1663: /* remove extra dot (which dot???) */
1664: /* ugly, but faster than a strcmp */
1665: if (tempOrgName[length] == '.'
1666: && (length == 0 || tempOrgName[length-1] != '.'))
1667: tempOrgName[length] = EOS;
1668:
1669: if (IsW3Path (tempOrgName))
1670: {
1671: /* the name is complete, go to the Sixth Step */
1672: strcpy (newName, tempOrgName);
1673: SimplifyUrl (&newName);
1674: /* verify if the URL has the form "protocol://server:port" */
1675: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1676: AMAYA_PARSE_HOST |
1677: AMAYA_PARSE_PUNCTUATION);
1678: if (ptr && !strcmp (ptr, newName))
1679: /* it has this form, we complete it by adding a DIR_STR */
1680: strcat (newName, URL_STR);
1681:
1682: if (ptr)
1683: TtaFreeMemory (ptr);
1684: }
1685: else if (basename == NULL)
1686: /* the name is complete, go to the Sixth Step */
1687: strcpy (newName, tempOrgName);
1688: else
1689: {
1690: /* Calculate the absolute URL, using the base or document URL */
1.184 gully 1691: #ifdef _WINDOWS
1.195 vatton 1692: if (!IsW3Path (basename))
1693: {
1694: length = strlen (tempOrgName);
1695: for (ndx = 0; ndx < length; ndx++)
1.211 vatton 1696: if (tempOrgName[ndx] == '/')
1697: tempOrgName[ndx] = '\\';
1.195 vatton 1698: }
1.184 gully 1699: #endif /* _WINDOWS */
1.195 vatton 1700: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1701: if (ptr)
1702: {
1703: SimplifyUrl (&ptr);
1704: strcpy (newName, ptr);
1705: TtaFreeMemory (ptr);
1706: }
1707: else
1708: newName[0] = EOS;
1709: }
1710:
1711: TtaFreeMemory (basename);
1712: /*
1713: * Prepare the docname that will refer to this ressource in the
1714: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1715: * noname.html as a default ressource name
1.18 cvs 1716: */
1.195 vatton 1717: if (newName[0] != EOS)
1718: {
1719: length = strlen (newName) - 1;
1720: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1721: {
1722: used_sep = newName[length];
1723: check = TRUE;
1724: while (check)
1725: {
1726: length--;
1727: while (length >= 0 && newName[length] != used_sep)
1728: length--;
1729: if (!strncmp (&newName[length+1], "..", 2))
1730: {
1731: newName[length+1] = EOS;
1732: /* remove also previous directory */
1733: length--;
1734: while (length >= 0 && newName[length] != used_sep)
1735: length--;
1736: if (strncmp (&newName[length+1], "//", 2))
1737: /* don't remove server name */
1738: newName[length+1] = EOS;
1739: }
1740: else if (!strncmp (&newName[length+1], ".", 1))
1741: newName[length+1] = EOS;
1742: else
1743: check = FALSE;
1744: }
1745: /* docname was not comprised inside the URL, so let's */
1746: /* assign the default ressource name */
1747: strcpy (docName, "noname.html");
1748: }
1749: else
1750: { /* docname is comprised inside the URL */
1751: while (length >= 0 && newName[length] != URL_SEP &&
1752: newName[length] != DIR_SEP)
1753: length--;
1754: if (length < 0)
1755: strcpy (docName, newName);
1756: else
1757: strcpy (docName, &newName[length+1]);
1758: }
1759: }
1760: else
1761: docName[0] = EOS;
1.18 cvs 1762: }
1.3 cvs 1763:
1.4 cvs 1764: /*----------------------------------------------------------------------
1.9 cvs 1765: IsSameHost
1.4 cvs 1766: ----------------------------------------------------------------------*/
1.106 cvs 1767: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1768: {
1.106 cvs 1769: char *basename_ptr1, *basename_ptr2;
1770: ThotBool result;
1.3 cvs 1771:
1.106 cvs 1772: basename_ptr1 = AmayaParseUrl (url1, "",
1.195 vatton 1773: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.106 cvs 1774: basename_ptr2 = AmayaParseUrl (url2, "",
1.195 vatton 1775: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1776:
1.106 cvs 1777: if (strcmp (basename_ptr1, basename_ptr2))
1778: result = FALSE;
1779: else
1780: result = TRUE;
1781: TtaFreeMemory (basename_ptr1);
1782: TtaFreeMemory (basename_ptr2);
1783: return (result);
1.3 cvs 1784: }
1785:
1786:
1.4 cvs 1787: /*----------------------------------------------------------------------
1.22 cvs 1788: HasKnownFileSuffix
1789: returns TRUE if path points to a file ending with a suffix.
1790: ----------------------------------------------------------------------*/
1.153 vatton 1791: ThotBool HasKnownFileSuffix (const char *path)
1.106 cvs 1792: {
1.195 vatton 1793: char *root;
1794: char temppath[MAX_LENGTH];
1795: char suffix[MAX_LENGTH];
1796:
1797: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1798: return (FALSE);
1799:
1800: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1801:
1802: if (root)
1803: {
1804: strcpy (temppath, root);
1805: TtaFreeMemory (root);
1806: /* Get the suffix */
1807: TtaExtractSuffix (temppath, suffix);
1808:
1809: if( suffix[0] == EOS)
1810: /* no suffix */
1811: return (FALSE);
1812:
1813: /* Normalize the suffix */
1814: ConvertToLowerCase (suffix);
1815:
1816: if (!strcmp (suffix, "gz"))
1817: /* skip the compressed suffix */
1818: {
1819: TtaExtractSuffix (temppath, suffix);
1820: if(suffix[0] == EOS)
1821: /* no suffix */
1822: return (FALSE);
1823: /* Normalize the suffix */
1824: ConvertToLowerCase (suffix);
1825: }
1826:
1827: if (strcmp (suffix, "gif") &&
1828: strcmp (suffix, "xbm") &&
1829: strcmp (suffix, "xpm") &&
1830: strcmp (suffix, "jpg") &&
1831: strcmp (suffix, "pdf") &&
1832: strcmp (suffix, "png") &&
1833: strcmp (suffix, "tgz") &&
1834: strcmp (suffix, "xpg") &&
1835: strcmp (suffix, "xpd") &&
1836: strcmp (suffix, "ps") &&
1837: strcmp (suffix, "au") &&
1838: strcmp (suffix, "html") &&
1839: strcmp (suffix, "htm") &&
1840: strcmp (suffix, "shtml") &&
1841: strcmp (suffix, "xht") &&
1842: strcmp (suffix, "xhtm") &&
1843: strcmp (suffix, "xhtml") &&
1844: strcmp (suffix, "txt") &&
1845: strcmp (suffix, "css") &&
1846: strcmp (suffix, "eps"))
1847: return (FALSE);
1848: else
1849: return (TRUE);
1850: }
1851: else
1852: return (FALSE);
1.22 cvs 1853: }
1854:
1855:
1856: /*----------------------------------------------------------------------
1.24 cvs 1857: ChopURL
1858: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1859: If inputURL is bigger than that size, outputURL receives
1860: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1861: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1862: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1863: copied into outputURL.
1864: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1865: chars.
1866: ----------------------------------------------------------------------*/
1.106 cvs 1867: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1868: {
1869: int len;
1.9 cvs 1870:
1.106 cvs 1871: len = strlen (inputURL);
1.24 cvs 1872: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1873: strcpy (outputURL, inputURL);
1.24 cvs 1874: else
1875: /* make a truncated urlName on the status window */
1876: {
1.106 cvs 1877: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1878: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1879: strcat (outputURL, "...");
1880: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1881: }
1.25 cvs 1882: }
1883:
1884:
1885: /*----------------------------------------------------------------------
1.195 vatton 1886: scan
1887: Scan a filename for its constituents
1888: -----------------------------------
1889:
1890: On entry,
1891: name points to a document name which may be incomplete.
1892: On exit,
1893: absolute or relative may be nonzero (but not both).
1894: host, fragment and access may be nonzero if they were specified.
1895: Any which are nonzero point to zero terminated strings.
1.25 cvs 1896: ----------------------------------------------------------------------*/
1.106 cvs 1897: static void scan (char *name, HTURI *parts)
1.25 cvs 1898: {
1.106 cvs 1899: char * p;
1900: char * after_access = name;
1.32 cvs 1901:
1.211 vatton 1902: memset (parts, 0, sizeof (HTURI));
1.28 cvs 1903: /* Look for fragment identifier */
1.106 cvs 1904: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1905: {
1.211 vatton 1906: *p++ = EOS;
1.28 cvs 1907: parts->fragment = p;
1.25 cvs 1908: }
1909:
1.28 cvs 1910: for (p=name; *p; p++)
1911: {
1.106 cvs 1912: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.195 vatton 1913: break;
1.106 cvs 1914: if (*p == ':')
1.195 vatton 1915: {
1916: *p = 0;
1917: parts->access = after_access; /* Scheme has been specified */
1918:
1919: /* The combination of gcc, the "-O" flag and the HP platform is
1920: unhealthy. The following three lines is a quick & dirty fix, but is
1921: not recommended. Rather, turn off "-O". */
1922:
1923: /* after_access = p;*/
1924: /* while (*after_access == 0)*/
1925: /* after_access++;*/
1926: after_access = p+1;
1927: if (!strcasecmp("URL", parts->access))
1928: /* Ignore IETF's URL: pre-prefix */
1929: parts->access = NULL;
1930: else
1931: break;
1932: }
1.25 cvs 1933: }
1934:
1.195 vatton 1935: p = after_access;
1936: if (*p == URL_SEP || *p == DIR_SEP)
1937: {
1938: if (p[1] == URL_SEP)
1939: {
1940: parts->host = p+2; /* host has been specified */
1941: *p = 0; /* Terminate access */
1942: /* look for end of host name if any */
1943: p = strchr (parts->host, URL_SEP);
1944: if (p)
1945: {
1946: *p = EOS; /* Terminate host */
1947: parts->absolute = p+1; /* Root has been found */
1948: }
1949: }
1950: else
1951: /* Root found but no host */
1952: parts->absolute = p+1;
1953: }
1954: else
1955: {
1956: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1957: }
1.25 cvs 1958: }
1959:
1960:
1961: /*----------------------------------------------------------------------
1.28 cvs 1962: AmayaParseUrl: parse a Name relative to another name
1963:
1964: This returns those parts of a name which are given (and requested)
1965: substituting bits from the related name where necessary.
1.25 cvs 1966:
1.28 cvs 1967: On entry,
1.195 vatton 1968: aName A filename given
1969: relatedName A name relative to which aName is to be parsed. Give
1970: it an empty string if aName is absolute.
1971: wanted A mask for the bits which are wanted.
1.25 cvs 1972:
1.28 cvs 1973: On exit,
1.195 vatton 1974: returns A pointer to a malloc'd string which MUST BE FREED
1.25 cvs 1975: ----------------------------------------------------------------------*/
1.206 kia 1976: char *AmayaParseUrl (const char *aName, const char *relatedName, int wanted)
1.106 cvs 1977: {
1978: char *return_value;
1979: char result[MAX_LENGTH];
1980: char name[MAX_LENGTH];
1981: char rel[MAX_LENGTH];
1982: char *p, *access;
1.29 cvs 1983: HTURI given, related;
1.205 vatton 1984: int len, l;
1.106 cvs 1985: char used_sep;
1.206 kia 1986: const char*used_str;
1.32 cvs 1987:
1.106 cvs 1988: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1989: {
1.106 cvs 1990: used_str = DIR_STR;
1991: used_sep = DIR_SEP;
1.33 cvs 1992: }
1.32 cvs 1993: else
1.33 cvs 1994: {
1.106 cvs 1995: used_str = URL_STR;
1996: used_sep = URL_SEP;
1.33 cvs 1997: }
1.32 cvs 1998:
1.29 cvs 1999: /* Make working copies of input strings to cut up: */
2000: return_value = NULL;
2001: result[0] = 0; /* Clear string */
1.169 quint 2002: rel[0] = EOS;
2003: strncpy (name, aName, MAX_LENGTH - 1);
2004: name[MAX_LENGTH - 1] = EOS;
2005: if (relatedName != NULL)
2006: {
2007: strncpy (rel, relatedName, MAX_LENGTH - 1);
2008: rel[MAX_LENGTH - 1] = EOS;
2009: }
1.29 cvs 2010:
2011: scan (name, &given);
2012: scan (rel, &related);
2013: access = given.access ? given.access : related.access;
2014: if (wanted & AMAYA_PARSE_ACCESS)
2015: if (access)
2016: {
1.195 vatton 2017: strcat (result, access);
1.211 vatton 2018: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.195 vatton 2019: strcat (result, ":");
1.29 cvs 2020: }
2021:
2022: if (given.access && related.access)
2023: /* If different, inherit nothing. */
1.106 cvs 2024: if (strcmp (given.access, related.access) != 0)
1.29 cvs 2025: {
1.195 vatton 2026: related.host = 0;
2027: related.absolute = 0;
2028: related.relative = 0;
2029: related.fragment = 0;
1.29 cvs 2030: }
2031:
2032: if (wanted & AMAYA_PARSE_HOST)
1.211 vatton 2033: if (given.host || related.host)
1.29 cvs 2034: {
1.211 vatton 2035: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.195 vatton 2036: strcat (result, "//");
2037: strcat (result, given.host ? given.host : related.host);
1.29 cvs 2038: }
2039:
2040: if (given.host && related.host)
2041: /* If different hosts, inherit no path. */
1.106 cvs 2042: if (strcmp (given.host, related.host) != 0)
1.29 cvs 2043: {
1.195 vatton 2044: related.absolute = 0;
2045: related.relative = 0;
2046: related.fragment = 0;
1.29 cvs 2047: }
2048:
1.205 vatton 2049: len = MAX_LENGTH - 1 - strlen (result);
1.29 cvs 2050: if (wanted & AMAYA_PARSE_PATH)
2051: {
2052: if (given.absolute)
1.195 vatton 2053: {
2054: /* All is given */
2055: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.205 vatton 2056: {
2057: strcat (result, used_str);
2058: len--;
2059: }
2060: l = strlen (given.absolute);
2061: if (l <= len)
2062: {
2063: strcat (result, given.absolute);
2064: len -= l;
2065: }
1.195 vatton 2066: }
1.29 cvs 2067: else if (related.absolute)
1.195 vatton 2068: {
2069: /* Adopt path not name */
2070: strcat (result, used_str);
1.205 vatton 2071: len--;
2072: l = strlen (related.absolute);
2073: if (l <= len)
2074: {
2075: strcat (result, related.absolute);
2076: len -= l;
2077: }
1.195 vatton 2078: if (given.relative)
2079: {
2080: /* Search part? */
2081: p = strchr (result, '?');
2082: if (!p)
1.205 vatton 2083: p = result + strlen(result) - 1;
2084: for (; *p != used_sep; p--); /* last / */
1.195 vatton 2085: /* Remove filename */
1.205 vatton 2086: p[1] = EOS;
1.195 vatton 2087: /* Add given one */
1.205 vatton 2088: l = strlen (given.relative);
2089: if (l <= len)
2090: {
2091: strcat (result, given.relative);
2092: len -= l;
2093: }
2094: }
2095: }
2096: else if (given.relative)
2097: {
2098: /* what we've got */
2099: l = strlen (given.relative);
2100: if (l <= len)
2101: {
1.195 vatton 2102: strcat (result, given.relative);
1.205 vatton 2103: len -= l;
1.195 vatton 2104: }
2105: }
1.29 cvs 2106: else if (related.relative)
1.205 vatton 2107: {
2108: l = strlen (related.relative);
2109: if (l <= len)
2110: {
2111: strcat (result, related.relative);
2112: len -= l;
2113: }
2114: }
1.29 cvs 2115: else
1.205 vatton 2116: {
2117: /* No inheritance */
2118: strcat (result, used_str);
2119: len--;
2120: }
1.25 cvs 2121: }
1.29 cvs 2122:
2123: if (wanted & AMAYA_PARSE_ANCHOR)
1.205 vatton 2124: if (len && (given.fragment || related.fragment))
1.29 cvs 2125: {
1.195 vatton 2126: if (given.absolute && given.fragment)
2127: {
2128: /*Fixes for relURLs...*/
2129: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.205 vatton 2130: {
2131: strcat (result, "#");
2132: len --;
2133: }
2134: l = strlen (given.fragment);
2135: if (l <= len)
2136: strcat (result, given.fragment);
1.195 vatton 2137: }
1.205 vatton 2138: else if (given.absolute || given.fragment)
1.195 vatton 2139: {
2140: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.205 vatton 2141: {
2142: strcat (result, "#");
2143: len--;
2144: }
2145: if (given.fragment)
2146: {
2147: l = strlen (given.fragment);
2148: if (l <= len)
2149: strcat (result, given.fragment);
2150: }
2151: else
2152: {
2153: l = strlen (given.fragment);
2154: if (l <= len)
2155: strcat (result, related.fragment);
2156: }
1.195 vatton 2157: }
1.29 cvs 2158: }
1.205 vatton 2159: return_value = TtaStrdup (result);
1.29 cvs 2160: return (return_value); /* exactly the right length */
1.25 cvs 2161: }
2162:
2163: /*----------------------------------------------------------------------
1.195 vatton 2164: HTCanon
2165: Canonicalizes the URL in the following manner starting from the host
2166: pointer:
1.25 cvs 2167:
1.195 vatton 2168: 1) The host name is converted to lowercase
2169: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1.25 cvs 2170:
1.195 vatton 2171: Return: OK The position of the current path part of the URL
2172: which might be the old one or a new one.
1.25 cvs 2173:
2174: ----------------------------------------------------------------------*/
1.106 cvs 2175: static char *HTCanon (char **filename, char *host)
2176: {
1.195 vatton 2177: char *newname = NULL;
2178: char used_sep;
2179: char *path;
2180: char *strptr;
2181: char *port;
2182: char *access = host-3;
2183:
2184: if (*filename && strchr (*filename, URL_SEP))
2185: used_sep = URL_SEP;
2186: else
2187: used_sep = DIR_SEP;
2188:
2189: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
2190: access--;
2191: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
2192: path = host + strlen (host);
2193: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
2194: host = strptr;
2195: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
2196: port = NULL;
1.25 cvs 2197:
1.195 vatton 2198: strptr = host; /* Convert to lower-case */
2199: while (strptr < path)
2200: {
2201: *strptr = tolower (*strptr);
2202: strptr++;
2203: }
2204:
2205: /* Does the URL contain a full domain name? This also works for a
2206: numerical host name. The domain name is already made lower-case
2207: and without a trailing dot. */
2208: {
2209: char *dot = port ? port : path;
2210: if (dot > *filename && *--dot == '.')
1.33 cvs 2211: {
1.195 vatton 2212: char *orig = dot;
2213: char *dest = dot + 1;
2214: while ((*orig++ = *dest++));
2215: if (port) port--;
2216: path--;
2217: }
2218: }
2219: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
2220: if (port)
2221: {
2222: if (!*(port+1) || *(port+1) == used_sep)
2223: {
2224: if (!newname)
2225: {
2226: char *orig = port;
2227: char *dest = port + 1;
2228: while ((*orig++ = *dest++));
2229: }
2230: }
2231: else if ((!strncmp (access, "http", 4) &&
2232: (*(port + 1) == '8' &&
2233: *(port+2) == '0' &&
2234: (*(port+3) == used_sep || !*(port + 3)))) ||
2235: (!strncmp (access, "gopher", 6) &&
2236: (*(port+1) == '7' &&
2237: *(port+2) == '0' &&
2238: (*(port+3) == used_sep || !*(port+3)))) ||
2239: (!strncmp (access, "ftp", 3) &&
2240: (*(port+1) == '2' &&
2241: *(port + 2) == '1' &&
2242: (*(port+3) == used_sep || !*(port+3))))) {
2243: if (!newname)
2244: {
2245: char *orig = port;
2246: char *dest = port + 3;
2247: while((*orig++ = *dest++));
2248: /* Update path position, Henry Minsky */
2249: path -= 3;
2250: }
1.33 cvs 2251: }
1.195 vatton 2252: else if (newname)
2253: strncat (newname, port, (int) (path - port));
1.25 cvs 2254: }
2255:
1.195 vatton 2256: if (newname)
2257: {
2258: char *newpath = newname + strlen (newname);
2259: strcat (newname, path);
2260: path = newpath;
2261: /* Free old copy */
2262: TtaFreeMemory(*filename);
2263: *filename = newname;
2264: }
2265: return path;
1.25 cvs 2266: }
2267:
2268:
2269: /*----------------------------------------------------------------------
1.29 cvs 2270: SimplifyUrl: simplify a URI
1.32 cvs 2271: A URI is allowed to contain the sequence xxx/../ which may be
2272: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 2273: Simplification helps us recognize duplicate URIs.
1.25 cvs 2274:
1.28 cvs 2275: Thus, /etc/junk/../fred becomes /etc/fred
1.195 vatton 2276: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 2277:
1.28 cvs 2278: but we should NOT change
1.195 vatton 2279: http://fred.xxx.edu/../..
1.25 cvs 2280:
1.195 vatton 2281: or ../../albert.html
1.25 cvs 2282:
1.28 cvs 2283: In order to avoid empty URLs the following URLs become:
1.25 cvs 2284:
1.195 vatton 2285: /fred/.. becomes /fred/..
2286: /fred/././.. becomes /fred/..
2287: /fred/.././junk/.././ becomes /fred/..
1.25 cvs 2288:
1.28 cvs 2289: If more than one set of `://' is found (several proxies in cascade) then
2290: only the part after the last `://' is simplified.
1.25 cvs 2291:
1.28 cvs 2292: Returns: A string which might be the old one or a new one.
1.25 cvs 2293: ----------------------------------------------------------------------*/
1.106 cvs 2294: void SimplifyUrl (char **url)
2295: {
2296: char *path;
2297: char *access;
2298: char *newptr;
2299: char *p;
2300: char *orig, *dest, *end;
1.28 cvs 2301:
1.106 cvs 2302: char used_sep;
1.77 cvs 2303: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1.195 vatton 2304: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 2305:
1.28 cvs 2306: if (!url || !*url)
2307: return;
2308:
1.106 cvs 2309: if (strchr (*url, URL_SEP))
1.195 vatton 2310: used_sep = URL_SEP;
1.32 cvs 2311: else
1.195 vatton 2312: used_sep = DIR_SEP;
1.32 cvs 2313:
1.77 cvs 2314: /* should we simplify double dot? */
2315: path = *url;
1.106 cvs 2316: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 2317: ddot_simplify = FALSE;
2318: else
2319: ddot_simplify = TRUE;
2320:
1.28 cvs 2321: /* Find any scheme name */
1.106 cvs 2322: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 2323: {
2324: /* Find host name */
1.28 cvs 2325: access = *url;
1.123 vatton 2326: while (access < path && (*access = tolower (*access)))
1.195 vatton 2327: access++;
1.28 cvs 2328: path += 3;
1.106 cvs 2329: while ((newptr = strstr (path, "://")) != NULL)
1.195 vatton 2330: /* For proxies */
2331: path = newptr + 3;
2332: /* We have a host name */
1.84 cvs 2333: path = HTCanon (url, path);
1.25 cvs 2334: }
1.106 cvs 2335: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 2336: path += 2;
2337: else
2338: path = *url;
1.84 cvs 2339: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 2340: /* Some URLs start //<foo> */
2341: path += 1;
1.94 cvs 2342: else if (IsFilePath (path))
2343: {
2344: /* doesn't need to do anything more */
2345: return;
2346: }
1.106 cvs 2347: else if (!strncmp (path, "news:", 5))
1.28 cvs 2348: {
1.106 cvs 2349: newptr = strchr (path+5, '@');
1.28 cvs 2350: if (!newptr)
1.195 vatton 2351: newptr = path + 5;
1.28 cvs 2352: while (*newptr)
1.195 vatton 2353: {
2354: /* Make group or host lower case */
2355: *newptr = tolower (*newptr);
2356: newptr++;
2357: }
1.28 cvs 2358: /* Doesn't need to do any more */
2359: return;
1.25 cvs 2360: }
1.126 cheyroul 2361:
1.189 vatton 2362: /* remove initial multiple /'s */
2363: p = path - 1;
2364: if (path != *url && *p == used_sep && *path == used_sep)
2365: {
2366: while (*path == used_sep)
1.195 vatton 2367: {
2368: orig = path;
2369: dest = path + 1;
2370: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2371: end = orig-1;
2372: }
1.189 vatton 2373: }
2374:
2375: if (path)
1.28 cvs 2376: {
1.106 cvs 2377: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1.195 vatton 2378: (end = strchr (path, '#'))))
2379: end = path + strlen (path);
1.28 cvs 2380:
2381: /* Parse string second time to simplify */
2382: p = path;
2383: while (p < end)
1.195 vatton 2384: {
2385: /* if we're pointing to a char, it's safe to reactivate the
2386: ../ convertion */
2387: if (!ddot_simplify && *p != '.' && *p != used_sep)
2388: ddot_simplify = TRUE;
2389:
2390: if (*p==used_sep)
2391: {
2392: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
2393: {
2394: orig = p + 1;
2395: dest = (*(p+2) != used_sep) ? p+2 : p+3;
2396: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
2397: end = orig - 1;
2398: }
2399: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
2400: && (*(p+3) == used_sep || !*(p+3)))
2401: {
2402: newptr = p;
2403: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
2404: if (*newptr == used_sep)
2405: orig = newptr + 1;
2406: else
2407: orig = newptr;
2408:
2409: dest = (*(p+3) != used_sep) ? p+3 : p+4;
2410: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
2411: end = orig-1;
2412: /* Start again with prev slash */
2413: p = newptr;
2414: }
2415: else if (*(p+1) == used_sep)
2416: {
2417: while (*(p+1) == used_sep)
2418: {
2419: orig = p;
2420: dest = p + 1;
2421: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2422: end = orig-1;
2423: }
2424: }
2425: else
2426: p++;
2427: }
2428: else
2429: p++;
2430: }
2431: }
2432: /*
2433: ** Check for host/../.. kind of things
2434: */
2435: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
2436: && (!*(path+3) || *(path+3) == used_sep))
2437: *(path+1) = EOS;
1.28 cvs 2438: return;
2439: }
2440:
2441:
2442: /*----------------------------------------------------------------------
1.202 vatton 2443: NormalizeFile normalizes local names.
2444: convertion is AM_CONV_NONE or AM_CONV_ALL
1.195 vatton 2445: Return TRUE if target and src differ.
1.28 cvs 2446: ----------------------------------------------------------------------*/
1.106 cvs 2447: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2448: {
1.214 vatton 2449: #ifndef _WINDOWS
1.195 vatton 2450: char *s;
2451: int i;
1.214 vatton 2452: #endif /* _WINDOWS */
1.198 cvs 2453: int start_index; /* the first char that we'll copy */
1.195 vatton 2454: ThotBool change;
1.28 cvs 2455:
1.195 vatton 2456: change = FALSE;
2457: start_index = 0;
1.90 cvs 2458:
1.195 vatton 2459: if (!src || src[0] == EOS)
2460: {
2461: target[0] = EOS;
2462: return FALSE;
2463: }
2464:
2465: /* @@ do I need file: or file:/ here? */
2466: if (strncmp (src, "file:", 5) == 0)
2467: {
2468: /* remove the prefix file: */
2469: start_index += 5;
1.90 cvs 2470:
1.195 vatton 2471: /* remove the localhost prefix */
2472: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
2473: start_index += 11;
1.94 cvs 2474:
1.211 vatton 2475: #ifdef _IV
1.195 vatton 2476: /* remove the first two slashes in / / /path */
2477: while (src[start_index] &&
2478: src[start_index] == '/'
2479: && src[start_index + 1] == '/')
2480: start_index++;
1.211 vatton 2481: #endif /* IV */
1.94 cvs 2482:
1.211 vatton 2483: #ifdef _IV
1.195 vatton 2484: /* remove any extra slash before the drive name */
1.211 vatton 2485: if (src[start_index] == '/' &&src[start_index+2] == ':')
1.195 vatton 2486: start_index++;
1.184 gully 2487: #endif /* _WINDOWS */
1.90 cvs 2488:
1.195 vatton 2489: if (src[start_index] == EOS)
2490: /* if there's nothing afterwards, add a DIR_STR */
2491: strcpy (target, DIR_STR);
2492: else
2493: /* as we're inside a file: URL, we'll apply all the convertions
2494: we know */
2495: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
2496:
2497: change = TRUE;
2498: }
2499: else if (convertion != AM_CONV_NONE)
2500: {
2501: /* we are following a "local" relative link, we do all the
2502: convertions except for the HOME_DIR ~ one */
2503: CleanCopyFileURL (target, src, convertion);
2504: }
1.198 cvs 2505: #ifdef _WINDOWS
1.211 vatton 2506: #ifdef IV
1.198 cvs 2507: else if (src[0] == DIR_SEP && src[1] == DIR_SEP)
2508: {
2509: s = getenv ("HOMEDRIVE");
2510: strcpy (target, s);
2511: i = strlen (target);
2512: strcpy (&target[i], &src[1]);
2513: change = TRUE;
2514: }
1.211 vatton 2515: #endif /* IV */
1.198 cvs 2516: #else /* _WINDOWS */
1.195 vatton 2517: else if (src[0] == '~')
2518: {
2519: /* it must be a URL typed in a text input field */
2520: /* do the HOME_DIR ~ substitution */
2521: s = TtaGetEnvString ("HOME");
2522: strcpy (target, s);
2523: i = strlen (target);
2524: strcpy (&target[i], &src[1]);
2525: change = TRUE;
2526: }
1.184 gully 2527: #endif /* _WINDOWS */
1.195 vatton 2528: else
2529: /* leave it as it is */
2530: strcpy (target, src);
1.96 cvs 2531:
1.195 vatton 2532: /* remove /../ and /./ */
2533: SimplifyUrl (&target);
2534: if (!change)
2535: change = (strcmp (src, target) != 0);
2536: return (change);
1.25 cvs 2537: }
2538:
1.28 cvs 2539:
1.25 cvs 2540: /*----------------------------------------------------------------------
1.31 cvs 2541: MakeRelativeURL: make relative name
1.25 cvs 2542:
1.28 cvs 2543: This function creates and returns a string which gives an expression of
2544: one address as related to another. Where there is no relation, an absolute
2545: address is retured.
1.25 cvs 2546:
1.28 cvs 2547: On entry,
1.195 vatton 2548: Both names must be absolute, fully qualified names of nodes
2549: (no fragment bits)
1.25 cvs 2550:
1.28 cvs 2551: On exit,
1.195 vatton 2552: The return result points to a newly allocated name which, if
2553: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2554: The caller is responsible for freeing the resulting name later.
1.25 cvs 2555: ----------------------------------------------------------------------*/
1.206 kia 2556: char *MakeRelativeURL (const char *aName, const char *relatedName)
1.106 cvs 2557: {
1.206 kia 2558: char *return_value;
2559: char result[MAX_LENGTH];
2560: const char *p;
2561: const char *q;
2562: const char *after_access;
2563: const char *last_slash = NULL;
2564: int slashes, levels, len;
1.184 gully 2565: #ifdef _WINDOWS
1.44 cvs 2566: int ndx;
1.184 gully 2567: #endif /* _WINDOWS */
1.44 cvs 2568:
1.29 cvs 2569: if (aName == NULL || relatedName == NULL)
2570: return (NULL);
2571:
2572: slashes = 0;
2573: after_access = NULL;
2574: p = aName;
2575: q = relatedName;
1.147 vatton 2576: len = 0;
2577: for (; *p && !strncasecmp (p, q, 1); p++, q++, len++)
1.27 cvs 2578: {
2579: /* Find extent of match */
1.106 cvs 2580: if (*p == ':')
1.195 vatton 2581: {
2582: after_access = p + 1;
1.184 gully 2583: #ifdef _WINDOWS
1.195 vatton 2584: if (len == 1)
2585: {
2586: /* it's a local Windows path like c:... */
2587: slashes+=2;
2588: }
1.184 gully 2589: #endif /* _WINDOWS */
1.195 vatton 2590: }
1.168 cvs 2591: if (*p == DIR_SEP)
1.195 vatton 2592: {
2593: /* memorize the last slash position and count them */
2594: last_slash = p;
2595: slashes++;
2596: }
1.25 cvs 2597: }
2598:
1.31 cvs 2599: /* q, p point to the first non-matching character or zero */
1.106 cvs 2600: if (*q == EOS)
1.31 cvs 2601: {
2602: /* New name is a subset of the related name */
2603: /* exactly the right length */
1.106 cvs 2604: len = strlen (p);
1.171 gully 2605: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.195 vatton 2606: strcpy (return_value, p);
1.31 cvs 2607: }
2608: else if ((slashes < 2 && after_access == NULL)
1.195 vatton 2609: || (slashes < 3 && after_access != NULL))
2610: {
1.31 cvs 2611: /* Two names whitout common path */
2612: /* exactly the right length */
1.106 cvs 2613: len = strlen (aName);
1.171 gully 2614: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.195 vatton 2615: strcpy (return_value, aName);
1.31 cvs 2616: }
2617: else
2618: {
2619: /* Some path in common */
1.218 vatton 2620: #ifdef IV
1.106 cvs 2621: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.195 vatton 2622: /* just the same server */
2623: strcpy (result, last_slash);
1.31 cvs 2624: else
1.218 vatton 2625: #endif
1.195 vatton 2626: {
2627: levels= 0;
2628: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
2629: if (*q == DIR_SEP)
2630: levels++;
1.31 cvs 2631:
1.195 vatton 2632: result[0] = EOS;
2633: for (;levels; levels--)
2634: strcat (result, "../");
2635: strcat (result, last_slash+1);
2636: }
1.52 cvs 2637:
2638: if (!*result)
1.195 vatton 2639: strcat (result, "./");
1.52 cvs 2640:
1.31 cvs 2641: /* exactly the right length */
1.106 cvs 2642: len = strlen (result);
1.171 gully 2643: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.195 vatton 2644: strcpy (return_value, result);
1.52 cvs 2645:
1.25 cvs 2646: }
1.184 gully 2647: #ifdef _WINDOWS
1.106 cvs 2648: len = strlen (return_value);
1.195 vatton 2649: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2650: if (return_value[ndx] == '\\')
1.195 vatton 2651: return_value[ndx] = '/' ;
1.184 gully 2652: #endif /* _WINDOWS */
1.29 cvs 2653: return (return_value);
1.24 cvs 2654: }
1.35 cvs 2655:
1.104 kahan 2656: /*----------------------------------------------------------------------
2657: AM_GetFileSize
2658: Returns TRUE and the filesize in the 2nd parameter.
2659: Otherwise, in case of a system error, returns FALSE, with a
2660: filesize of 0L.
2661: ---------------------------------------------------------------------*/
1.206 kia 2662: ThotBool AM_GetFileSize (const char *filename, unsigned long *file_size)
1.104 kahan 2663: {
2664: if (!TtaFileExist (filename))
2665: return FALSE;
1.185 vatton 2666: *file_size = TtaGetFileSize (filename);
2667: return TRUE;
1.104 kahan 2668: }
1.139 kahan 2669:
2670: /*----------------------------------------------------------------------
2671: AM_UseXHTMLMimeType
2672: Returns TRUE if the user has configured Amaya to use this MIME type,
2673: FALSE otherwise.
2674: ---------------------------------------------------------------------*/
2675: ThotBool AM_UseXHTMLMimeType (void)
2676: {
2677: ThotBool xhtml_mimetype;
2678:
2679: /* does the user wants to use the new MIME type? */
2680: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2681:
2682: return (xhtml_mimetype);
1.152 kahan 2683: }
2684:
1.154 kahan 2685:
2686: /********************************************
2687: The following routines were adapted from the GNU libc functions
2688: for generating a tmpnam.
2689: *********************************************/
2690:
2691: /* These are the characters used in temporary filenames. */
2692: static const char letters[] =
2693: "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
2694:
2695: /* Generate a temporary file name based on TMPL. TMPL must match the
2696: rules for mk[s]temp (i.e. end in "XXXXXX"). The name constructed
2697: does not exist at the time of the call to __gen_tempname. TMPL is
2698: overwritten with the result.
2699:
2700: We use a clever algorithm to get hard-to-predict names. */
2701: void
2702: AM_gen_tempname (char *tmpl)
2703: {
2704: int len;
2705: char *XXXXXX;
2706: static uint64_t value;
2707: uint64_t random_time_bits;
2708: unsigned int count;
2709: int save_errno = errno;
2710: struct stat st;
2711:
2712: /* A lower bound on the number of temporary files to attempt to
2713: generate. The maximum total number of temporary file names that
2714: can exist for a given template is 62**6. It should never be
2715: necessary to try all these combinations. Instead if a reasonable
2716: number of names is tried (we define reasonable as 62**3) fail to
2717: give the system administrator the chance to remove the problems. */
2718: unsigned int attempts_min = 62 * 62 * 62;
2719:
2720: /* The number of times to attempt to generate a temporary file. To
2721: conform to POSIX, this must be no smaller than TMP_MAX. */
2722: unsigned int attempts = attempts_min < TMP_MAX ? TMP_MAX : attempts_min;
2723:
2724: len = strlen (tmpl);
2725: if (len < 6 || strcmp (&tmpl[len - 6], "XXXXXX"))
2726: {
2727: /* @@ JK ? */
2728: errno = EINVAL;
2729: return;
2730: }
2731:
2732: /* This is where the Xs start. */
2733: XXXXXX = &tmpl[len - 6];
2734:
2735: /* Get some more or less random data. */
2736: #ifdef RANDOM_BITS
2737: RANDOM_BITS (random_time_bits);
2738: #else
2739: # if HAVE_GETTIMEOFDAY || _LIBC
2740: {
2741: struct timeval tv;
2742: gettimeofday (&tv, NULL);
2743: random_time_bits = ((uint64_t) tv.tv_usec << 16) ^ tv.tv_sec;
2744: }
2745: # else
2746: random_time_bits = time (NULL);
2747: # endif
2748: #endif
2749: value += random_time_bits ^ getpid ();
2750:
2751: for (count = 0; count < attempts; value += 7777, ++count)
2752: {
2753: uint64_t v = value;
2754:
2755: /* Fill in the random bits. */
2756: XXXXXX[0] = letters[v % 62];
2757: v /= 62;
2758: XXXXXX[1] = letters[v % 62];
2759: v /= 62;
2760: XXXXXX[2] = letters[v % 62];
2761: v /= 62;
2762: XXXXXX[3] = letters[v % 62];
2763: v /= 62;
2764: XXXXXX[4] = letters[v % 62];
2765: v /= 62;
2766: XXXXXX[5] = letters[v % 62];
2767:
2768: /* This case is backward from the other three. AM_gen_tempname
1.195 vatton 2769: succeeds if __xstat fails because the name does not exist.
2770: Note the continue to bypass the common logic at the bottom
2771: of the loop. */
1.154 kahan 2772: if (stat (tmpl, &st) < 0)
1.195 vatton 2773: break;
1.154 kahan 2774:
2775: continue;
2776: }
2777:
2778: if (count == attempts || errno != ENOENT)
2779: tmpl[0] = EOS;
2780: else
2781: errno = save_errno;
2782:
2783: return;
2784: }
2785:
2786: #define JOSE 1
2787:
1.152 kahan 2788: /*-----------------------------------------------------------------------
2789: GetTempName
2790: Front end to the Unix tempnam function, which is independent of the
2791: value of the TMPDIR env value
2792: Returns a dynamically allocated string with a tempname. The user
2793: must free this memory.
2794: -----------------------------------------------------------------------*/
2795: char *GetTempName (const char *dir, const char *prefix)
2796: {
1.154 kahan 2797: #ifdef JOSE
2798:
1.162 kahan 2799: static char tmpbufmem[PATH_MAX + 1];
1.154 kahan 2800: int len;
2801: int i;
2802:
1.155 cvs 2803: if (!dir || *dir == EOS || !TtaDirExists (dir))
1.154 kahan 2804: return NULL;
2805:
1.162 kahan 2806: /* make sure that the name is no bigger than PATH_MAX + the 6 tempname chars we
1.195 vatton 2807: will add */
1.154 kahan 2808:
1.156 cvs 2809: len = strlen (dir);
1.162 kahan 2810: if (len + 6 > PATH_MAX)
1.154 kahan 2811: return NULL;
2812:
2813: /* copy the dir name, and add a DIR_SEP if it's missing */
2814: if (dir[strlen (dir) - 1] == DIR_SEP)
2815: strcpy (tmpbufmem, dir);
2816: else
1.195 vatton 2817: {
2818: sprintf (tmpbufmem, "%s%c", dir, DIR_SEP);
2819: len++;
2820: }
1.154 kahan 2821:
1.161 kahan 2822: /* copy the prefix (no more than L_tmpnam chars, to respect POSIX). Save
1.156 cvs 2823: space for the 6 X and EOS chars that will become the random bits */
2824: if (prefix)
1.195 vatton 2825: {
1.156 cvs 2826: i = 0;
1.195 vatton 2827: while (prefix[i] != EOS && i < L_tmpnam - 8)
2828: tmpbufmem[len++] = prefix[i++];
2829: tmpbufmem[len] = EOS;
2830: }
1.156 cvs 2831:
2832: /* Add the 6 X chars */
2833: len = strlen (tmpbufmem);
2834: i = 0;
2835: while (i < 6)
1.195 vatton 2836: {
2837: tmpbufmem[len++] = 'X';
2838: i++;
2839: }
1.156 cvs 2840: tmpbufmem[len] = EOS;
1.154 kahan 2841:
2842: AM_gen_tempname (tmpbufmem);
2843:
2844: if (tmpbufmem[0] == EOS)
2845: return NULL;
2846: else
2847: return (TtaStrdup (tmpbufmem));
2848:
2849: #else
1.152 kahan 2850: char *tmpdir;
1.181 vatton 2851: char *tmp = NULL;
1.152 kahan 2852: char *name = NULL;
2853:
2854: /* save the value of TMPDIR */
1.181 vatton 2855: tmpdir = getenv ("TMPDIR");
1.152 kahan 2856: if (tmpdir)
2857: {
1.181 vatton 2858: /* remove TMPDIR from the environment */
2859: tmp = TtaGetMemory (strlen (tmpdir) + 20);
2860: sprintf (tmp, "TMPDIR=");
1.184 gully 2861: #ifdef _WINDOWS
1.152 kahan 2862: _putenv (tmp);
2863: #else
2864: putenv (tmp);
1.184 gully 2865: #endif /* _WINDOWS */
1.181 vatton 2866: /* prepare the string to restore the value of TMPDIR */
2867: strrcat (tmp, tmpdir);
1.152 kahan 2868: }
2869:
2870: /* create the tempname */
1.184 gully 2871: #ifdef _WINDOWS
1.152 kahan 2872: /* Under Windows, _tempnam returns the same name until the file is created */
2873: {
2874: char *altprefix;
2875: name = tmpnam (NULL); /* get a possibly unique string */
1.181 vatton 2876: altprefix = TtaGetMemory (strlen (prefix) + strlen(name) + 1);
1.152 kahan 2877: sprintf (altprefix, "%s%s", prefix, name + strlen(_P_tmpdir));
2878: name = _tempnam (dir, altprefix); /* get a name that isn't yet in use */
2879: TtaFreeMemory (altprefix);
2880: }
2881: #else
2882: name = tempnam (dir, prefix);
1.184 gully 2883: #endif /* _WINDOWS */
1.152 kahan 2884:
2885: if (tmpdir)
2886: {
1.181 vatton 2887: /* restore the value of TMPDIR */
1.184 gully 2888: #ifdef _WINDOWS
1.152 kahan 2889: _putenv (tmpdir);
2890: #else
2891: putenv (tmpdir);
1.184 gully 2892: #endif /* _WINDOWS */
1.181 vatton 2893: TtaFreeMemory (tmpdir);
1.152 kahan 2894: }
2895: return (name);
1.154 kahan 2896: #endif
1.139 kahan 2897: }
Webmaster