Annotation of Amaya/amaya/AHTURLTools.c, revision 1.134
1.7 cvs 1: /*
2: *
1.133 vatton 3: * (c) COPYRIGHT MIT and INRIA, 1996-2002
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: */
1.7 cvs 17:
1.15 cvs 18: #define THOT_EXPORT extern
1.3 cvs 19: #include "amaya.h"
20:
1.8 cvs 21: #include "init_f.h"
22: #include "AHTURLTools_f.h"
1.100 kahan 23: #include "query_f.h"
1.8 cvs 24:
1.24 cvs 25: #define MAX_PRINT_URL_LENGTH 50
1.106 cvs 26: typedef struct _HTURI
27: {
28: char *access; /* Now known as "scheme" */
29: char *host;
30: char *absolute;
31: char *relative;
32: char *fragment;
1.29 cvs 33: } HTURI;
1.24 cvs 34:
1.28 cvs 35:
36: /*----------------------------------------------------------------------
37: ConvertToLowerCase
38: Converts a string to lowercase.
39: ----------------------------------------------------------------------*/
1.124 vatton 40: void ConvertToLowerCase (char *string)
1.28 cvs 41: {
42: int i;
1.93 cvs 43:
1.28 cvs 44: if (!string)
45: return;
46:
1.106 cvs 47: for (i = 0; string[i] != EOS; i++)
1.123 vatton 48: string[i] = tolower (string[i]);
1.28 cvs 49: }
1.22 cvs 50:
1.8 cvs 51: /*----------------------------------------------------------------------
1.75 cvs 52: EscapeChar
53: writes the equivalent escape code of a char in a string
54: ----------------------------------------------------------------------*/
1.109 cvs 55: void EscapeChar (char *string, char c)
1.75 cvs 56: {
1.109 cvs 57: unsigned int i;
58:
59: i = (unsigned char) c & 0xFF;
60: sprintf (string, "%02x", i);
1.75 cvs 61: }
62:
63: /*----------------------------------------------------------------------
1.96 cvs 64: UnEscapeChar
65: writes the equivalent hex code to a %xx coded char
66: ----------------------------------------------------------------------*/
1.109 cvs 67: static char UnEscapeChar (char c)
1.96 cvs 68: {
1.106 cvs 69: return c >= '0' && c <= '9' ? c - '0'
70: : c >= 'A' && c <= 'F' ? c - 'A' + 10
71: : c - 'a' + 10; /* accept small letters just in case */
1.96 cvs 72: }
73:
74: /*----------------------------------------------------------------------
1.75 cvs 75: EscapeURL
76: Takes a URL and escapes all protected chars into
77: %xx sequences. Also, removes any leading white spaces
78: Returns either NULL or a new buffer, which must be freed by the caller
79: ----------------------------------------------------------------------*/
1.106 cvs 80: char *EscapeURL (const char *url)
81: {
82: char *buffer;
83: int buffer_len;
84: int buffer_free_mem;
85: char *ptr;
86: int new_chars;
1.75 cvs 87: void *status;
88:
89: if (url && *url)
90: {
1.106 cvs 91: buffer_free_mem = strlen (url) + 20;
92: buffer = TtaGetMemory (buffer_free_mem + 1);
1.107 kahan 93: ptr = (char *) url;
1.75 cvs 94: buffer_len = 0;
95:
96: while (*ptr)
97: {
98: switch (*ptr)
99: {
100: /* put here below all the chars that need to
101: be escaped into %xx */
1.81 cvs 102: case 0x27: /* & */
103: case 0x20: /* space */
1.75 cvs 104: new_chars = 3;
105: break;
106:
107: default:
1.122 kahan 108: if ((unsigned char )*ptr > 127)
109: new_chars = 3;
110: else
111: new_chars = 1;
1.75 cvs 112: break;
113: }
114:
115: /* see if we need extra room in the buffer */
116: if (new_chars > buffer_free_mem)
117: {
1.76 cvs 118: buffer_free_mem = 20;
1.106 cvs 119: status = TtaRealloc (buffer, sizeof (char)
1.75 cvs 120: * (buffer_len + buffer_free_mem + 1));
121: if (status)
1.114 cvs 122: buffer = (char *) status;
1.106 cvs 123: else
124: {
125: /* @@ maybe we should do some other behavior here, like
126: freeing the buffer and return a void thing */
127: buffer[buffer_len] = EOS;
128: break;
129: }
1.75 cvs 130: }
131: /* escape the char */
132: if (new_chars == 3)
133: {
1.106 cvs 134: buffer[buffer_len] = '%';
1.75 cvs 135: EscapeChar (&buffer[buffer_len+1], *ptr);
136: }
137: else
138: buffer[buffer_len] = *ptr;
139:
140: /* update the status */
141: buffer_len += new_chars;
142: buffer_free_mem -= new_chars;
143: /* examine the next char */
144: ptr++;
145: }
1.106 cvs 146: buffer[buffer_len] = EOS;
1.75 cvs 147: }
1.76 cvs 148: else
149: buffer = NULL;
150:
1.75 cvs 151: return (buffer);
1.122 kahan 152: }
153:
154: /*----------------------------------------------------------------------
155: URLToUTF8
156: If such convertion is needed, returns a converted string that the
157: user must free.
158: ----------------------------------------------------------------------*/
159: char *URLToUTF8 (char *url, Document doc)
160: {
161: unsigned char *tmp;
162:
163: if (!url || *url == EOS)
164: return NULL;
165:
166: /* does the URL contain chars > 127 ? */
167: tmp = url;
168: while (*tmp)
169: {
170: if (*tmp > 127)
171: break;
172: tmp++;
173: }
174:
175: if (*tmp == EOS)
176: return NULL; /* no such chars found */
177:
178: tmp = TtaConvertIsoToMbs (url, TtaGetDocumentCharset (doc));
179:
180: return tmp;
1.75 cvs 181: }
182:
183:
184: /*----------------------------------------------------------------------
1.11 cvs 185: ExplodeURL
1.8 cvs 186: ----------------------------------------------------------------------*/
1.106 cvs 187: void ExplodeURL (char *url, char **proto, char **host, char **dir,
188: char **file)
1.8 cvs 189: {
1.33 cvs 190: char *curr, *temp;
191: char used_sep;
1.32 cvs 192:
1.33 cvs 193: if (url && strchr (url, URL_SEP))
194: used_sep = URL_SEP;
195: else
196: used_sep = DIR_SEP;
1.8 cvs 197:
198: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
199: (dir == NULL) || (file == NULL))
200: return;
201:
202: /* initialize every pointer */
203: *proto = *host = *dir = *file = NULL;
204:
205: /* skip any leading space */
206: while ((*url == SPACE) || (*url == TAB))
207: url++;
1.9 cvs 208: curr = url;
209: if (*curr == 0)
1.8 cvs 210: goto finished;
211:
212: /* go to the end of the URL */
1.68 cvs 213: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
214: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 215: curr++;
1.8 cvs 216:
217: /* mark the end of the chain */
1.9 cvs 218: *curr = EOS;
219: curr--;
220: if (curr <= url)
1.8 cvs 221: goto finished;
222:
223: /* search the next DIR_SEP indicating the beginning of the file name */
224: do
1.11 cvs 225: curr--;
1.33 cvs 226: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 227:
1.9 cvs 228: if (curr < url)
1.8 cvs 229: goto finished;
1.9 cvs 230: *file = curr + 1;
1.8 cvs 231:
232: /* mark the end of the dir */
1.9 cvs 233: *curr = EOS;
234: curr--;
235: if (curr < url)
1.8 cvs 236: goto finished;
237:
1.29 cvs 238: /* search for the DIR_STR indicating the host name start */
1.33 cvs 239: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 240: curr--;
1.8 cvs 241:
242: /* if we found it, separate the host name from the directory */
1.102 kahan 243: if ((*curr == used_sep) && (*(curr + 1) == used_sep))
1.8 cvs 244: {
1.9 cvs 245: *host = temp = curr + 2;
1.33 cvs 246: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 247: temp++;
1.33 cvs 248: if (*temp == used_sep)
1.8 cvs 249: {
250: *temp = EOS;
251: *dir = temp + 1;
252: }
253: }
254: else
1.11 cvs 255: *dir = curr;
256:
1.9 cvs 257: if (curr <= url)
1.8 cvs 258: goto finished;
259:
260: /* mark the end of the proto */
1.9 cvs 261: *curr = EOS;
262: curr--;
263: if (curr < url)
1.8 cvs 264: goto finished;
265:
1.106 cvs 266: if (*curr == ':')
1.8 cvs 267: {
1.9 cvs 268: *curr = EOS;
269: curr--;
1.8 cvs 270: }
271: else
272: goto finished;
1.11 cvs 273:
1.9 cvs 274: if (curr < url)
1.8 cvs 275: goto finished;
1.9 cvs 276: while ((curr > url) && (isalpha (*curr)))
277: curr--;
278: *proto = curr;
1.8 cvs 279:
280: finished:;
281:
282: #ifdef AMAYA_DEBUG
283: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
284: if (*proto)
285: fprintf (stderr, "proto : %s, ", *proto);
286: if (*host)
287: fprintf (stderr, "host : %s, ", *host);
288: if (*dir)
289: fprintf (stderr, "dir : %s, ", *dir);
290: if (*file)
291: fprintf (stderr, "file : %s ", *file);
292: fprintf (stderr, "\n");
293: #endif
294:
295: }
1.3 cvs 296:
1.116 kahan 297: /*----------------------------------------------------------------------
298: PicTypeToMime
299: Converts a Thot PicType into the equivalent MIME type. If no convertion
300: is possible, it returns NULL.
301: ----------------------------------------------------------------------*/
302: char *PicTypeToMIME (PicType contentType)
303: {
304: char *mime_type;
305:
306: switch (contentType)
307: {
308: case xbm_type:
309: mime_type ="image/x-xbitmap";
310: break;
311: case eps_type:
312: mime_type ="application/postscript";
313: break;
314: case xpm_type:
315: mime_type ="image/x-xpicmap";
316: break;
317: case gif_type:
318: mime_type ="image/gif";
319: break;
320: case jpeg_type:
321: mime_type ="image/jpeg";
322: break;
323: case png_type:
324: mime_type ="image/png";
325: break;
326: case svg_type:
1.118 kahan 327: mime_type ="image/svg+xml";
1.116 kahan 328: break;
329: case unknown_type:
330: default:
331: mime_type = NULL;
332: }
333:
334: return mime_type;
335: }
1.61 cvs 336:
337: /*----------------------------------------------------------------------
1.117 kahan 338: ImageElement
339: Returns the element (image parameter) and URL (url parameter) of an
340: image in a docImage document. The user must free the memory associated
1.120 kahan 341: with the url parameter if the function is succesful.
342: If the url parameter is NULL, we won't initialize it.
1.117 kahan 343: Returns TRUE if succesful, FALSE otherwise.
344: ----------------------------------------------------------------------*/
345: ThotBool ImageElement (Document doc, char **url, Element *image)
346: {
347: Element el, imgEl;
348: Attribute attr, srcAttr;
349: AttributeType attrType;
350: int length;
351: char *value;
352:
353: if (DocumentTypes[doc] != docImage)
354: return FALSE;
355:
356: /* find the value of the src attribute */
357: attrType.AttrSSchema = TtaGetSSchema ("HTML", doc);
358: attrType.AttrTypeNum = HTML_ATTR_SRC;
359: el = TtaGetRootElement (doc);
360: TtaSearchAttribute (attrType, SearchInTree, el, &imgEl, &srcAttr);
361:
362: if (!imgEl)
363: return FALSE;
364: *image = imgEl;
365:
1.120 kahan 366: if (url)
367: {
368: attr = TtaGetAttribute (imgEl, attrType);
369: length = TtaGetTextAttributeLength (srcAttr) + 1;
370: value = TtaGetMemory (length);
371: TtaGiveTextAttributeValue (srcAttr, value, &length);
372: *url = value;
373: }
1.117 kahan 374: return TRUE;
375: }
376:
377: /*----------------------------------------------------------------------
378: DocImageMimeType
379: Returns the MIME type of a docImage document.
380: ----------------------------------------------------------------------*/
381: char *DocImageMimeType (Document doc)
382: {
383: char *mime_type;
384: LoadedImageDesc *pImage;
385: PicType type;
386: Element image;
387:
388: if (DocumentTypes[doc] != docImage)
389: return NULL;
390:
391: mime_type = NULL;
392: if (!IsHTTPPath (DocumentURLs[doc]))
393: {
394: /* it is a local image */
1.120 kahan 395: if (ImageElement (doc, NULL, &image))
1.117 kahan 396: {
397: type = TtaGetPictureType (image);
398: mime_type = PicTypeToMIME (type);
399: }
400: }
401: else
402: {
403: /* find the value of the src attribute */
404: pImage = ImageURLs;
405: while (pImage != NULL)
406: {
407: if (pImage->document == doc)
408: {
409: if (pImage->content_type)
410: mime_type = pImage->content_type;
411: else if (pImage->elImage && pImage->elImage->currentElement)
412: {
413: type = TtaGetPictureType (pImage->elImage->currentElement);
414: mime_type = PicTypeToMIME (type);
415: }
416: break;
417: }
418: }
419: }
420: return (mime_type);
421: }
422:
1.61 cvs 423:
1.4 cvs 424: /*----------------------------------------------------------------------
1.9 cvs 425: IsHTMLName
426: returns TRUE if path points to an HTML resource.
1.4 cvs 427: ----------------------------------------------------------------------*/
1.109 cvs 428: ThotBool IsHTMLName (const char *path)
1.106 cvs 429: {
430: char temppath[MAX_LENGTH];
431: char suffix[MAX_LENGTH];
432: char nsuffix[MAX_LENGTH];
1.101 cvs 433: int i;
1.5 cvs 434:
1.101 cvs 435: if (!path)
436: return (FALSE);
1.5 cvs 437:
1.106 cvs 438: strcpy (temppath, path);
1.124 vatton 439: TtaExtractSuffix (temppath, suffix);
1.101 cvs 440: i = 0;
1.106 cvs 441: while (suffix[i] != EOS)
1.101 cvs 442: {
443: /* Normalize the suffix */
444: i = 0;
1.106 cvs 445: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 446: {
1.123 vatton 447: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 448: i++;
449: }
1.106 cvs 450: nsuffix[i] = EOS;
451: if (!strcmp (nsuffix, "html") ||
452: !strcmp (nsuffix, "htm") ||
453: !strcmp (nsuffix, "shtml") ||
454: !strcmp (nsuffix, "jsp") ||
455: !strcmp (nsuffix, "xht") ||
456: !strcmp (nsuffix, "xhtm") ||
457: !strcmp (nsuffix, "xhtml"))
1.101 cvs 458: return (TRUE);
1.106 cvs 459: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 460: {
461: /* take into account compressed files */
1.124 vatton 462: TtaExtractSuffix (temppath, suffix);
1.101 cvs 463: /* Normalize the suffix */
464: i = 0;
1.106 cvs 465: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 466: {
1.123 vatton 467: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 468: i++;
469: }
1.106 cvs 470: nsuffix[i] = EOS;
471: if (!strcmp (nsuffix, "html") ||
472: !strcmp (nsuffix, "htm") ||
473: !strcmp (nsuffix, "shtml") ||
474: !strcmp (nsuffix, "jsp") ||
475: !strcmp (nsuffix, "xht") ||
476: !strcmp (nsuffix, "xhtm") ||
477: !strcmp (nsuffix, "xhtml"))
1.101 cvs 478: return (TRUE);
479: else
480: return (FALSE);
481: }
482: else
483: /* check if there is another suffix */
1.124 vatton 484: TtaExtractSuffix (temppath, suffix);
1.101 cvs 485: }
1.88 cvs 486: return (FALSE);
1.3 cvs 487: }
488:
1.4 cvs 489: /*----------------------------------------------------------------------
1.56 cvs 490: IsXMLName
491: returns TRUE if path points to an XML resource.
492: ----------------------------------------------------------------------*/
1.111 cvs 493: ThotBool IsXMLName (const char *path)
1.56 cvs 494: {
1.106 cvs 495: char temppath[MAX_LENGTH];
496: char suffix[MAX_LENGTH];
1.56 cvs 497:
498: if (!path)
499: return (FALSE);
500:
1.106 cvs 501: strcpy (temppath, path);
1.124 vatton 502: TtaExtractSuffix (temppath, suffix);
1.56 cvs 503:
1.106 cvs 504: if (!strcasecmp (suffix, "xml") ||
505: !strcasecmp (suffix, "xht") ||
506: !strcmp (suffix, "xhtm") ||
507: !strcmp (suffix, "xhtml"))
1.56 cvs 508: return (TRUE);
1.106 cvs 509: else if (!strcmp (suffix, "gz"))
1.56 cvs 510: {
511: /* take into account compressed files */
1.124 vatton 512: TtaExtractSuffix (temppath, suffix);
1.106 cvs 513: if (!strcasecmp (suffix, "xml") ||
514: !strcasecmp (suffix, "xht") ||
515: !strcmp (suffix, "xhtm") ||
516: !strcmp (suffix, "xhtml"))
1.60 cvs 517: return (TRUE);
518: else
519: return (FALSE);
520: }
521: else
522: return (FALSE);
523: }
524:
525: /*----------------------------------------------------------------------
1.133 vatton 526: IsUndisplayedName
527: returns TRUE if path points to an undisplayed resource.
528: ----------------------------------------------------------------------*/
529: ThotBool IsUndisplayedName (const char *path)
530: {
531: char temppath[MAX_LENGTH];
532: char suffix[MAX_LENGTH];
533:
534: if (!path)
535: return (FALSE);
536:
537: strcpy (temppath, path);
538: TtaExtractSuffix (temppath, suffix);
539:
540: if (!strcasecmp (suffix, "exe") ||
541: !strcasecmp (suffix, "pdf") ||
1.134 ! vatton 542: !strcasecmp (suffix, "ps") ||
! 543: !strcasecmp (suffix, "eps") ||
1.133 vatton 544: !strcasecmp (suffix, "tar") ||
545: !strcasecmp (suffix, "tgz") ||
546: !strcasecmp (suffix, "ddl") ||
547: !strcasecmp (suffix, "o"))
548: return (TRUE);
549: else if (!strcmp (suffix, "gz"))
550: {
551: /* take into account compressed files */
552: TtaExtractSuffix (temppath, suffix);
553: if (!strcasecmp (suffix, "exe") ||
554: !strcasecmp (suffix, "pdf") ||
1.134 ! vatton 555: !strcasecmp (suffix, "ps") ||
! 556: !strcasecmp (suffix, "eps") ||
1.133 vatton 557: !strcasecmp (suffix, "tar") ||
558: !strcasecmp (suffix, "ddl") ||
559: !strcasecmp (suffix, "o"))
560: return (TRUE);
561: else
562: return (FALSE);
563: }
564: else
565: return (FALSE);
566: }
567:
568: /*----------------------------------------------------------------------
1.103 cvs 569: IsMathMLName
570: returns TRUE if path points to an MathML resource.
571: ----------------------------------------------------------------------*/
1.111 cvs 572: ThotBool IsMathMLName (const char *path)
1.103 cvs 573: {
1.106 cvs 574: char temppath[MAX_LENGTH];
575: char suffix[MAX_LENGTH];
1.103 cvs 576:
577: if (!path)
578: return (FALSE);
579:
1.106 cvs 580: strcpy (temppath, path);
1.124 vatton 581: TtaExtractSuffix (temppath, suffix);
1.103 cvs 582:
1.106 cvs 583: if (!strcasecmp (suffix, "mml"))
1.103 cvs 584: return (TRUE);
1.106 cvs 585: else if (!strcmp (suffix, "gz"))
1.103 cvs 586: {
587: /* take into account compressed files */
1.124 vatton 588: TtaExtractSuffix (temppath, suffix);
1.106 cvs 589: if (!strcasecmp (suffix, "mml"))
1.103 cvs 590: return (TRUE);
591: else
592: return (FALSE);
593: }
594: else
595: return (FALSE);
596: }
597:
598: /*----------------------------------------------------------------------
599: IsSVGName
1.119 kahan 600: returns TRUE if path points to an SVG resource.
1.103 cvs 601: ----------------------------------------------------------------------*/
1.111 cvs 602: ThotBool IsSVGName (const char *path)
1.103 cvs 603: {
1.106 cvs 604: char temppath[MAX_LENGTH];
605: char suffix[MAX_LENGTH];
1.103 cvs 606:
607: if (!path)
608: return (FALSE);
609:
1.106 cvs 610: strcpy (temppath, path);
1.124 vatton 611: TtaExtractSuffix (temppath, suffix);
1.103 cvs 612:
1.106 cvs 613: if (!strcasecmp (suffix, "svg"))
1.103 cvs 614: return (TRUE);
1.106 cvs 615: else if (!strcmp (suffix, "gz"))
1.103 cvs 616: {
617: /* take into account compressed files */
1.124 vatton 618: TtaExtractSuffix (temppath, suffix);
1.106 cvs 619: if (!strcasecmp (suffix, "svg"))
1.103 cvs 620: return (TRUE);
621: else
622: return (FALSE);
623: }
624: else
625: return (FALSE);
626: }
627:
628: /*----------------------------------------------------------------------
1.60 cvs 629: IsCSSName
630: returns TRUE if path points to an XML resource.
631: ----------------------------------------------------------------------*/
1.111 cvs 632: ThotBool IsCSSName (const char *path)
1.60 cvs 633: {
1.106 cvs 634: char temppath[MAX_LENGTH];
635: char suffix[MAX_LENGTH];
1.60 cvs 636:
637: if (!path)
638: return (FALSE);
639:
1.106 cvs 640: strcpy (temppath, path);
1.124 vatton 641: TtaExtractSuffix (temppath, suffix);
1.60 cvs 642:
1.106 cvs 643: if (!strcasecmp (suffix, "css"))
1.60 cvs 644: return (TRUE);
1.106 cvs 645: else if (!strcmp (suffix, "gz"))
1.60 cvs 646: {
647: /* take into account compressed files */
1.124 vatton 648: TtaExtractSuffix (temppath, suffix);
1.106 cvs 649: if (!strcasecmp (suffix, "css"))
1.56 cvs 650: return (TRUE);
651: else
652: return (FALSE);
653: }
654: else
655: return (FALSE);
656: }
657:
658: /*----------------------------------------------------------------------
1.9 cvs 659: IsImageName
660: returns TRUE if path points to an image resource.
1.4 cvs 661: ----------------------------------------------------------------------*/
1.111 cvs 662: ThotBool IsImageName (const char *path)
1.106 cvs 663: {
664: char temppath[MAX_LENGTH];
665: char suffix[MAX_LENGTH];
666: char nsuffix[MAX_LENGTH];
1.5 cvs 667: int i;
668:
669: if (!path)
1.13 cvs 670: return (FALSE);
1.5 cvs 671:
1.106 cvs 672: strcpy (temppath, path);
1.124 vatton 673: TtaExtractSuffix (temppath, suffix);
1.5 cvs 674:
675: /* Normalize the suffix */
676: i = 0;
1.106 cvs 677: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 678: {
1.123 vatton 679: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 680: i++;
681: }
1.106 cvs 682: nsuffix[i] = EOS;
683: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
684: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
685: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 686: return (TRUE);
687: return (FALSE);
1.3 cvs 688: }
689:
1.4 cvs 690: /*----------------------------------------------------------------------
1.58 cvs 691: IsImageType
692: returns TRUE if type points to an image resource.
693: ----------------------------------------------------------------------*/
1.111 cvs 694: ThotBool IsImageType (const char *type)
1.58 cvs 695: {
1.106 cvs 696: char temptype[MAX_LENGTH];
1.58 cvs 697: int i;
698:
699: if (!type)
700: return (FALSE);
701:
1.106 cvs 702: strcpy (temptype, type);
1.58 cvs 703: /* Normalize the type */
704: i = 0;
1.106 cvs 705: while (temptype[i] != EOS)
1.58 cvs 706: {
707: temptype[i] = tolower (temptype[i]);
708: i++;
709: }
1.111 cvs 710: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
711: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
712: !strcmp (temptype, "png"))
1.58 cvs 713: return (TRUE);
714: return (FALSE);
715: }
716:
717: /*----------------------------------------------------------------------
1.9 cvs 718: IsTextName
1.4 cvs 719: ----------------------------------------------------------------------*/
1.111 cvs 720: ThotBool IsTextName (const char *path)
1.106 cvs 721: {
722: char temppath[MAX_LENGTH];
723: char suffix[MAX_LENGTH];
724: char nsuffix[MAX_LENGTH];
1.5 cvs 725: int i;
726:
727: if (!path)
1.13 cvs 728: return (FALSE);
1.5 cvs 729:
1.106 cvs 730: strcpy (temppath, path);
1.124 vatton 731: TtaExtractSuffix (temppath, suffix);
1.5 cvs 732:
733: /* Normalize the suffix */
734: i = 0;
1.106 cvs 735: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 736: {
1.25 cvs 737: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 738: i++;
739: }
1.106 cvs 740: nsuffix[i] = EOS;
1.5 cvs 741:
1.111 cvs 742: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 743: return (TRUE);
1.106 cvs 744: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 745: {
1.39 cvs 746: /* take into account compressed files */
1.124 vatton 747: TtaExtractSuffix (temppath, suffix);
1.13 cvs 748: /* Normalize the suffix */
749: i = 0;
1.106 cvs 750: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 751: {
1.25 cvs 752: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 753: i++;
754: }
1.106 cvs 755: nsuffix[i] = EOS;
1.111 cvs 756: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 757: return (TRUE);
758: else
759: return (FALSE);
760: }
761: else
762: return (FALSE);
1.3 cvs 763: }
764:
1.4 cvs 765: /*----------------------------------------------------------------------
1.9 cvs 766: IsHTTPPath
767: returns TRUE if path is in fact an http URL.
1.4 cvs 768: ----------------------------------------------------------------------*/
1.112 cvs 769: ThotBool IsHTTPPath (const char *path)
1.3 cvs 770: {
1.5 cvs 771: if (!path)
772: return FALSE;
1.3 cvs 773:
1.106 cvs 774: if ((!strncmp (path, "http:", 5) != 0)
775: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
776: || !strncmp (path, "internal:", 9))
1.58 cvs 777: return TRUE;
778: return FALSE;
1.3 cvs 779: }
780:
1.4 cvs 781: /*----------------------------------------------------------------------
1.9 cvs 782: IsWithParameters
783: returns TRUE if url has a concatenated query string.
1.4 cvs 784: ----------------------------------------------------------------------*/
1.133 vatton 785: ThotBool IsWithParameters (const char *url)
1.3 cvs 786: {
1.5 cvs 787: int i;
1.3 cvs 788:
1.9 cvs 789: if ((!url) || (url[0] == EOS))
1.5 cvs 790: return FALSE;
1.3 cvs 791:
1.9 cvs 792: i = strlen (url) - 1;
793: while (i > 0 && url[i--] != '?')
1.5 cvs 794: if (i < 0)
795: return FALSE;
1.3 cvs 796:
1.5 cvs 797: /* There is a parameter */
798: return TRUE;
1.3 cvs 799: }
800:
1.4 cvs 801: /*----------------------------------------------------------------------
1.9 cvs 802: IsW3Path
803: returns TRUE if path is in fact a URL.
1.4 cvs 804: ----------------------------------------------------------------------*/
1.133 vatton 805: ThotBool IsW3Path (const char *path)
1.106 cvs 806: {
807: if (strncmp (path, "http:", 5) &&
808: strncmp (path, "ftp:", 4) &&
809: strncmp (path, "telnet:", 7) &&
810: strncmp (path, "wais:", 5) &&
811: strncmp (path, "news:", 5) &&
812: strncmp (path, "gopher:", 7) &&
813: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 814: strncmp (path, "archie:", 7) &&
815: strncmp (path, "https:", 6))
1.72 cvs 816: return FALSE;
817: return TRUE;
1.3 cvs 818: }
819:
1.4 cvs 820: /*----------------------------------------------------------------------
1.90 cvs 821: IsFilePath
822: returns TRUE if path is in fact a URL.
823: ----------------------------------------------------------------------*/
1.133 vatton 824: ThotBool IsFilePath (const char *path)
1.90 cvs 825: {
1.106 cvs 826: if (strncmp (path, "file:", 5))
1.90 cvs 827: return FALSE;
828: return TRUE;
829: }
830:
831: /*----------------------------------------------------------------------
1.9 cvs 832: IsValidProtocol
833: returns true if the url protocol is supported by Amaya.
1.4 cvs 834: ----------------------------------------------------------------------*/
1.133 vatton 835: ThotBool IsValidProtocol (const char *url)
1.106 cvs 836: {
837: if (!strncmp (url, "http:", 5)
838: || !strncmp (url, "internal:", 9)
839: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 840: /* experimental */
1.24 cvs 841: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 842: return (TRUE);
1.5 cvs 843: else
1.8 cvs 844: return (FALSE);
1.3 cvs 845: }
846:
1.31 cvs 847:
848: /*----------------------------------------------------------------------
849: GetBaseURL
850: normalizes orgName according to a base associated with doc, and
851: following the standard URL format rules.
852: The function returns the base used to solve relative URL and SRC:
853: - the base of the document,
854: - or the document path (without document name).
855: ----------------------------------------------------------------------*/
1.106 cvs 856: char *GetBaseURL (Document doc)
1.31 cvs 857: {
858: Element el;
859: ElementType elType;
860: AttributeType attrType;
861: Attribute attr;
1.106 cvs 862: char *ptr, *basename;
1.31 cvs 863: int length;
864:
1.113 cvs 865: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 866: return NULL;
1.106 cvs 867: basename = TtaGetMemory (MAX_LENGTH);
868: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
869: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 870: length = MAX_LENGTH -1;
1.113 cvs 871: /* is it a HTML document ? */
1.31 cvs 872: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.113 cvs 873: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
874: /* it's a HTML document */
1.65 cvs 875: {
1.113 cvs 876: /* get the document element */
877: el = TtaGetMainRoot (doc);
878: /* search the BASE element */
879: elType.ElTypeNum = HTML_EL_HEAD;
880: el = TtaSearchTypedElement (elType, SearchForward, el);
881: if (el)
882: /* there is a HEAD element */
883: {
884: /* look for a BASE element within the HEAD */
885: elType.ElTypeNum = HTML_EL_BASE;
886: el = TtaSearchTypedElement (elType, SearchInTree, el);
887: }
888: if (el)
1.31 cvs 889: {
1.113 cvs 890: /* The document has a BASE element. Get the HREF attribute of the
891: BASE element */
892: attrType.AttrSSchema = elType.ElSSchema;
893: attrType.AttrTypeNum = HTML_ATTR_HREF_;
894: attr = TtaGetAttribute (el, attrType);
895: if (attr)
1.31 cvs 896: {
1.113 cvs 897: /* Use the base path of the document */
898: TtaGiveTextAttributeValue (attr, basename, &length);
899: /* base and orgName have to be separated by a DIR_SEP */
900: length--;
901: if (basename[0] != EOS && basename[length] != URL_SEP &&
902: basename[length] != DIR_SEP)
903: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 904: {
1.113 cvs 905: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
906: AMAYA_PARSE_HOST |
907: AMAYA_PARSE_PUNCTUATION);
908: if (ptr && !strcmp (ptr, basename))
909: {
910: /* it has this form, complete it by adding a URL_STR */
911: if (strchr (basename, DIR_SEP))
912: strcat (basename, DIR_STR);
913: else
914: strcat (basename, URL_STR);
915: length++;
916: }
917: if (ptr)
918: TtaFreeMemory (ptr);
1.31 cvs 919: }
920: }
921: }
1.113 cvs 922: }
923:
1.31 cvs 924: /* Remove anything after the last DIR_SEP char. If no such char is found,
925: * then search for the first ":" char, hoping that what's before that is a
926: * protocol. If found, end the string there. If neither char is found,
927: * then discard the whole base element.
928: */
1.106 cvs 929: length = strlen (basename) - 1;
1.31 cvs 930: /* search for the last DIR_SEP char */
1.106 cvs 931: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 932: length--;
933: if (length >= 0)
934: /* found the last DIR_SEP char, end the string there */
1.106 cvs 935: basename[length + 1] = EOS;
1.31 cvs 936: else
937: /* search for the first PATH_STR char */
938: {
1.106 cvs 939: for (length = 0; basename[length] != ':' &&
940: basename[length] != EOS; length ++);
941: if (basename[length] == ':')
1.31 cvs 942: /* found, so end the string there */
1.106 cvs 943: basename[length + 1] = EOS;
1.31 cvs 944: else
945: /* not found, discard the base */
1.106 cvs 946: basename[0] = EOS;
1.31 cvs 947: }
948: return (basename);
949: }
950:
951:
1.4 cvs 952: /*----------------------------------------------------------------------
1.40 cvs 953: GetLocalPath
954: Allocate and return the local document path associated to the url
955: ----------------------------------------------------------------------*/
1.106 cvs 956: char *GetLocalPath (Document doc, char *url)
957: {
958: char *ptr;
959: char *n;
960: char *documentname;
961: char url_sep;
1.83 cvs 962: int len;
1.67 cvs 963: ThotBool noFile;
1.40 cvs 964:
965: if (url != NULL)
966: {
967: /* check whether the file name exists */
1.106 cvs 968: len = strlen (url) - 1;
1.71 cvs 969: if (IsW3Path (url))
1.106 cvs 970: url_sep = '/';
1.41 cvs 971: else
1.106 cvs 972: url_sep = DIR_SEP;
1.41 cvs 973: noFile = (url[len] == url_sep);
1.40 cvs 974: if (noFile)
1.106 cvs 975: url[len] = EOS;
976: ptr = TtaGetMemory (MAX_LENGTH);
977: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 978: TtaExtractName (url, ptr, documentname);
1.106 cvs 979: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 980: if (!TtaCheckDirectory (ptr))
981: /* directory did not exist */
1.72 cvs 982: TtaMakeDirectory (ptr);
1.47 cvs 983:
984: /* don't include the query string within document name */
1.106 cvs 985: n = strrchr (documentname, '?');
1.47 cvs 986: if (n != NULL)
1.106 cvs 987: *n = EOS;
1.46 cvs 988: /* don't include ':' within document name */
1.106 cvs 989: n = strchr (documentname, ':');
1.46 cvs 990: if (n != NULL)
1.106 cvs 991: *n = EOS;
1.69 cvs 992: /* if after all this operations document name
993: is empty, let's use noname.html instead */
1.106 cvs 994: if (documentname[0] == EOS)
995: strcat (ptr, "noname.html");
1.69 cvs 996: else
1.106 cvs 997: strcat (ptr, documentname);
1.40 cvs 998: TtaFreeMemory (documentname);
999: /* restore the url */
1000: if (noFile)
1.41 cvs 1001: url[len] = url_sep;
1.40 cvs 1002: return (ptr);
1003: }
1004: else
1005: return (NULL);
1006: }
1007:
1.73 cvs 1008: /*----------------------------------------------------------------------
1.79 cvs 1009: ExtractTarget extract the target name from document nane.
1010: ----------------------------------------------------------------------*/
1.106 cvs 1011: void ExtractTarget (char *aName, char *target)
1.79 cvs 1012: {
1.106 cvs 1013: int lg, i;
1014: char *ptr;
1015: char *oldptr;
1.79 cvs 1016:
1017: if (!target || !aName)
1018: /* bad target */
1019: return;
1020:
1.106 cvs 1021: target[0] = EOS;
1022: lg = strlen (aName);
1.79 cvs 1023: if (lg)
1024: {
1025: /* the name is not empty */
1026: oldptr = ptr = &aName[0];
1027: do
1028: {
1.106 cvs 1029: ptr = strrchr (oldptr, '#');
1.79 cvs 1030: if (ptr)
1031: oldptr = &ptr[1];
1032: }
1033: while (ptr);
1034:
1035: i = (int) (oldptr) - (int) (aName); /* name length */
1036: if (i > 1)
1037: {
1.106 cvs 1038: aName[i - 1] = EOS;
1.79 cvs 1039: if (i != lg)
1.106 cvs 1040: strcpy (target, oldptr);
1.79 cvs 1041: }
1042: }
1043: }
1044:
1045: /*----------------------------------------------------------------------
1.90 cvs 1046: RemoveNewLines (text)
1047: Removes any '\n' chars that are found in text.
1048: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1049: ----------------------------------------------------------------------*/
1.106 cvs 1050: ThotBool RemoveNewLines (char *text)
1051: {
1052: ThotBool change = FALSE;
1053: char *src;
1054: char *dest;
1.90 cvs 1055:
1056: src = text;
1057: dest = text;
1.115 kahan 1058:
1059: /* remove any preceding whitespace */
1060: while (*src && *src == ' ')
1061: {
1062: src++;
1063: change = 1;
1064: }
1065:
1.90 cvs 1066: while (*src)
1067: {
1068: switch (*src)
1069: {
1.106 cvs 1070: case '\n':
1.90 cvs 1071: /* don't copy the newline */
1072: change = 1;
1073: break;
1074: default:
1075: *dest = *src;
1076: dest++;
1077: break;
1078: }
1079: src++;
1080: }
1081: /* copy the last EOS char */
1082: *dest = *src;
1083:
1084: return (change);
1085: }
1086:
1087: /*----------------------------------------------------------------------
1088: CleanCopyFileURL
1089: Copies a file url from a src string to destination string.
1.97 cvs 1090: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1091: we want to do).
1.90 cvs 1092: ----------------------------------------------------------------------*/
1.106 cvs 1093: static void CleanCopyFileURL (char *dest, char *src,
1094: ConvertionType convertion)
1.90 cvs 1095: {
1096: while (*src)
1.89 cvs 1097: {
1.90 cvs 1098: switch (*src)
1.89 cvs 1099: {
1100: #ifdef _WINDOWS
1.106 cvs 1101: case URL_SEP:
1.96 cvs 1102: /* make DIR_SEP transformation */
1.97 cvs 1103: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1104: *dest = DIR_SEP;
1.96 cvs 1105: else
1106: *dest = *src;
1.90 cvs 1107: dest++;
1.96 cvs 1108: src++;
1.90 cvs 1109: break;
1.89 cvs 1110: #endif /* _WINDOWS */
1.96 cvs 1111:
1.106 cvs 1112: case '%':
1.97 cvs 1113: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1114: {
1.97 cvs 1115: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1116: src++;
1.106 cvs 1117: if (*src != EOS)
1.97 cvs 1118: {
1119: *dest = UnEscapeChar (*src) * 16;
1120: src++;
1121: }
1.106 cvs 1122: if (*src != EOS)
1.97 cvs 1123: {
1124: *dest = *dest + UnEscapeChar (*src);
1125: src++;
1126: }
1127: dest++;
1.96 cvs 1128: }
1.97 cvs 1129: else
1.96 cvs 1130: {
1.97 cvs 1131: *dest = *src;
1132: dest++;
1.96 cvs 1133: src++;
1134: }
1135: break;
1136:
1.90 cvs 1137: default:
1138: *dest = *src;
1.89 cvs 1139: dest++;
1.96 cvs 1140: src++;
1.90 cvs 1141: break;
1.89 cvs 1142: }
1143: }
1.90 cvs 1144: /* copy the EOS char */
1145: *dest = *src;
1.73 cvs 1146: }
1.40 cvs 1147:
1148: /*----------------------------------------------------------------------
1.9 cvs 1149: NormalizeURL
1150: normalizes orgName according to a base associated with doc, and
1151: following the standard URL format rules.
1.113 cvs 1152: if doc is < 0, use as a base the URL of the document that contains
1153: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1154: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1155: other path.
1.9 cvs 1156: The function returns the new complete and normalized URL
1.12 cvs 1157: or file name path (newName) and the name of the document (docName).
1.9 cvs 1158: N.B. If the function can't find out what's the docName, it assigns
1159: the name "noname.html".
1.4 cvs 1160: ----------------------------------------------------------------------*/
1.106 cvs 1161: void NormalizeURL (char *orgName, Document doc, char *newName,
1162: char *docName, char *otherPath)
1163: {
1164: char *basename;
1165: char tempOrgName[MAX_LENGTH];
1166: char *ptr;
1167: char used_sep;
1.84 cvs 1168: int length;
1169: ThotBool check;
1.5 cvs 1170:
1.110 cvs 1171: #ifdef _WINDOWS
1.44 cvs 1172: int ndx;
1.110 cvs 1173: #endif /* _WINDOWS */
1.44 cvs 1174:
1.5 cvs 1175: if (!newName || !docName)
1176: return;
1.18 cvs 1177:
1.113 cvs 1178: if (doc < 0)
1179: basename = TtaStrdup (SavedDocumentURL);
1180: else if (doc > 0)
1.53 cvs 1181: basename = GetBaseURL (doc);
1182: else if (otherPath != NULL)
1.108 cvs 1183: basename = TtaStrdup (otherPath);
1.32 cvs 1184: else
1.53 cvs 1185: basename = NULL;
1.32 cvs 1186:
1.18 cvs 1187: /*
1.31 cvs 1188: * Clean orgName
1189: * Make sure we have a complete orgName, without any leading or trailing
1190: * white spaces, or trailinbg new lines
1191: */
1.5 cvs 1192: ptr = orgName;
1.18 cvs 1193: /* skip leading white space and new line characters */
1.106 cvs 1194: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1195: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1196: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1197: /*
1.31 cvs 1198: * Make orgName a complete URL
1199: * If the URL does not include a protocol, then try to calculate
1200: * one using the doc's base element (if it exists),
1201: */
1.106 cvs 1202: if (tempOrgName[0] == EOS)
1.53 cvs 1203: {
1.106 cvs 1204: newName[0] = EOS;
1205: docName[0] = EOS;
1.53 cvs 1206: TtaFreeMemory (basename);
1207: return;
1208: }
1.49 cvs 1209:
1210: /* clean trailing white space */
1.106 cvs 1211: length = strlen (tempOrgName) - 1;
1212: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1213: {
1.106 cvs 1214: tempOrgName[length] = EOS;
1.53 cvs 1215: length--;
1216: }
1.50 cvs 1217:
1.55 cvs 1218: /* remove extra dot (which dot???) */
1219: /* ugly, but faster than a strcmp */
1.106 cvs 1220: if (tempOrgName[length] == '.'
1221: && (length == 0 || tempOrgName[length-1] != '.'))
1222: tempOrgName[length] = EOS;
1.50 cvs 1223:
1.94 cvs 1224: if (IsW3Path (tempOrgName))
1.53 cvs 1225: {
1226: /* the name is complete, go to the Sixth Step */
1.106 cvs 1227: strcpy (newName, tempOrgName);
1.53 cvs 1228: SimplifyUrl (&newName);
1229: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1230: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1231: AMAYA_PARSE_HOST |
1232: AMAYA_PARSE_PUNCTUATION);
1233: if (ptr && !strcmp (ptr, newName))
1234: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1235: strcat (newName, URL_STR);
1.49 cvs 1236:
1.53 cvs 1237: if (ptr)
1.50 cvs 1238: TtaFreeMemory (ptr);
1.53 cvs 1239: }
1.113 cvs 1240: else if (basename == NULL)
1.53 cvs 1241: /* the name is complete, go to the Sixth Step */
1.106 cvs 1242: strcpy (newName, tempOrgName);
1.53 cvs 1243: else
1244: {
1.31 cvs 1245: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1246: #ifdef _WINDOWS
1.53 cvs 1247: if (!IsW3Path (basename))
1248: {
1.106 cvs 1249: length = strlen (tempOrgName);
1.53 cvs 1250: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1251: if (tempOrgName [ndx] == '/')
1252: tempOrgName [ndx] = '\\';
1.53 cvs 1253: }
1.110 cvs 1254: #endif /* _WINDOWS */
1.25 cvs 1255: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1256: if (ptr)
1257: {
1258: SimplifyUrl (&ptr);
1.106 cvs 1259: strcpy (newName, ptr);
1.53 cvs 1260: TtaFreeMemory (ptr);
1261: }
1262: else
1.106 cvs 1263: newName[0] = EOS;
1.53 cvs 1264: }
1.36 cvs 1265:
1266: TtaFreeMemory (basename);
1.18 cvs 1267: /*
1.31 cvs 1268: * Prepare the docname that will refer to this ressource in the
1269: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1270: * noname.html as a default ressource name
1.18 cvs 1271: */
1.106 cvs 1272: if (newName[0] != EOS)
1.53 cvs 1273: {
1.106 cvs 1274: length = strlen (newName) - 1;
1275: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1276: {
1277: used_sep = newName[length];
1278: check = TRUE;
1279: while (check)
1280: {
1.50 cvs 1281: length--;
1282: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1283: length--;
1.106 cvs 1284: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1285: {
1.106 cvs 1286: newName[length+1] = EOS;
1.53 cvs 1287: /* remove also previous directory */
1288: length--;
1289: while (length >= 0 && newName[length] != used_sep)
1290: length--;
1.106 cvs 1291: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1292: /* don't remove server name */
1.106 cvs 1293: newName[length+1] = EOS;
1.53 cvs 1294: }
1.106 cvs 1295: else if (!strncmp (&newName[length+1], ".", 1))
1296: newName[length+1] = EOS;
1.50 cvs 1297: else
1.53 cvs 1298: check = FALSE;
1299: }
1300: /* docname was not comprised inside the URL, so let's */
1301: /* assign the default ressource name */
1.106 cvs 1302: strcpy (docName, "noname.html");
1.53 cvs 1303: }
1304: else
1305: { /* docname is comprised inside the URL */
1.110 cvs 1306: while (length >= 0 && newName[length] != URL_SEP &&
1307: newName[length] != DIR_SEP)
1.53 cvs 1308: length--;
1309: if (length < 0)
1.106 cvs 1310: strcpy (docName, newName);
1.53 cvs 1311: else
1.106 cvs 1312: strcpy (docName, &newName[length+1]);
1.53 cvs 1313: }
1314: }
1315: else
1.106 cvs 1316: docName[0] = EOS;
1.18 cvs 1317: }
1.3 cvs 1318:
1.4 cvs 1319: /*----------------------------------------------------------------------
1.9 cvs 1320: IsSameHost
1.4 cvs 1321: ----------------------------------------------------------------------*/
1.106 cvs 1322: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1323: {
1.106 cvs 1324: char *basename_ptr1, *basename_ptr2;
1325: ThotBool result;
1.3 cvs 1326:
1.106 cvs 1327: basename_ptr1 = AmayaParseUrl (url1, "",
1328: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1329: basename_ptr2 = AmayaParseUrl (url2, "",
1330: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1331:
1.106 cvs 1332: if (strcmp (basename_ptr1, basename_ptr2))
1333: result = FALSE;
1334: else
1335: result = TRUE;
1336: TtaFreeMemory (basename_ptr1);
1337: TtaFreeMemory (basename_ptr2);
1338: return (result);
1.3 cvs 1339: }
1340:
1341:
1.4 cvs 1342: /*----------------------------------------------------------------------
1.22 cvs 1343: HasKnownFileSuffix
1344: returns TRUE if path points to a file ending with a suffix.
1345: ----------------------------------------------------------------------*/
1.106 cvs 1346: ThotBool HasKnownFileSuffix (const char *path)
1347: {
1348: char *root;
1349: char temppath[MAX_LENGTH];
1350: char suffix[MAX_LENGTH];
1.22 cvs 1351:
1.106 cvs 1352: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1353: return (FALSE);
1354:
1.106 cvs 1355: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1356:
1357: if (root)
1358: {
1.106 cvs 1359: strcpy (temppath, root);
1.25 cvs 1360: TtaFreeMemory (root);
1.22 cvs 1361: /* Get the suffix */
1.124 vatton 1362: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1363:
1.106 cvs 1364: if( suffix[0] == EOS)
1.22 cvs 1365: /* no suffix */
1366: return (FALSE);
1367:
1368: /* Normalize the suffix */
1369: ConvertToLowerCase (suffix);
1370:
1.106 cvs 1371: if (!strcmp (suffix, "gz"))
1.22 cvs 1372: /* skip the compressed suffix */
1373: {
1.124 vatton 1374: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1375: if(suffix[0] == EOS)
1.22 cvs 1376: /* no suffix */
1377: return (FALSE);
1378: /* Normalize the suffix */
1379: ConvertToLowerCase (suffix);
1380: }
1381:
1.106 cvs 1382: if (strcmp (suffix, "gif") &&
1383: strcmp (suffix, "xbm") &&
1384: strcmp (suffix, "xpm") &&
1385: strcmp (suffix, "jpg") &&
1386: strcmp (suffix, "pdf") &&
1387: strcmp (suffix, "png") &&
1388: strcmp (suffix, "tgz") &&
1389: strcmp (suffix, "xpg") &&
1390: strcmp (suffix, "xpd") &&
1391: strcmp (suffix, "ps") &&
1392: strcmp (suffix, "au") &&
1393: strcmp (suffix, "html") &&
1394: strcmp (suffix, "htm") &&
1395: strcmp (suffix, "shtml") &&
1396: strcmp (suffix, "xht") &&
1397: strcmp (suffix, "xhtm") &&
1398: strcmp (suffix, "xhtml") &&
1399: strcmp (suffix, "txt") &&
1400: strcmp (suffix, "css") &&
1401: strcmp (suffix, "eps"))
1.22 cvs 1402: return (FALSE);
1403: else
1404: return (TRUE);
1405: }
1406: else
1407: return (FALSE);
1408: }
1409:
1410:
1411: /*----------------------------------------------------------------------
1.24 cvs 1412: ChopURL
1413: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1414: If inputURL is bigger than that size, outputURL receives
1415: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1416: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1417: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1418: copied into outputURL.
1419: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1420: chars.
1421: ----------------------------------------------------------------------*/
1.106 cvs 1422: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1423: {
1424: int len;
1.9 cvs 1425:
1.106 cvs 1426: len = strlen (inputURL);
1.24 cvs 1427: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1428: strcpy (outputURL, inputURL);
1.24 cvs 1429: else
1430: /* make a truncated urlName on the status window */
1431: {
1.106 cvs 1432: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1433: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1434: strcat (outputURL, "...");
1435: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1436: }
1.25 cvs 1437: }
1438:
1439:
1440: /*----------------------------------------------------------------------
1441: scan
1.47 cvs 1442: Scan a filename for its constituents
1.25 cvs 1443: -----------------------------------
1444:
1445: On entry,
1446: name points to a document name which may be incomplete.
1447: On exit,
1448: absolute or relative may be nonzero (but not both).
1449: host, fragment and access may be nonzero if they were specified.
1450: Any which are nonzero point to zero terminated strings.
1451: ----------------------------------------------------------------------*/
1.106 cvs 1452: static void scan (char *name, HTURI *parts)
1.25 cvs 1453: {
1.106 cvs 1454: char * p;
1455: char * after_access = name;
1.32 cvs 1456:
1.43 cvs 1457: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1458: /* Look for fragment identifier */
1.106 cvs 1459: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1460: {
1.106 cvs 1461: *p++ = '\0';
1.28 cvs 1462: parts->fragment = p;
1.25 cvs 1463: }
1464:
1.28 cvs 1465: for (p=name; *p; p++)
1466: {
1.106 cvs 1467: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1468: break;
1.106 cvs 1469: if (*p == ':')
1.28 cvs 1470: {
1471: *p = 0;
1472: parts->access = after_access; /* Scheme has been specified */
1473:
1474: /* The combination of gcc, the "-O" flag and the HP platform is
1475: unhealthy. The following three lines is a quick & dirty fix, but is
1476: not recommended. Rather, turn off "-O". */
1477:
1478: /* after_access = p;*/
1479: /* while (*after_access == 0)*/
1480: /* after_access++;*/
1481: after_access = p+1;
1.106 cvs 1482: if (!strcasecmp("URL", parts->access))
1.28 cvs 1483: /* Ignore IETF's URL: pre-prefix */
1484: parts->access = NULL;
1485: else
1.25 cvs 1486: break;
1487: }
1488: }
1489:
1490: p = after_access;
1.43 cvs 1491: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1492: {
1.43 cvs 1493: if (p[1] == URL_SEP)
1.28 cvs 1494: {
1.25 cvs 1495: parts->host = p+2; /* host has been specified */
1.28 cvs 1496: *p = 0; /* Terminate access */
1497: /* look for end of host name if any */
1.106 cvs 1498: p = strchr (parts->host, URL_SEP);
1.28 cvs 1499: if (p)
1500: {
1.106 cvs 1501: *p = EOS; /* Terminate host */
1.25 cvs 1502: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1503: }
1504: }
1505: else
1506: /* Root found but no host */
1507: parts->absolute = p+1;
1508: }
1509: else
1510: {
1.25 cvs 1511: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1512: }
1.25 cvs 1513: }
1514:
1515:
1516: /*----------------------------------------------------------------------
1.28 cvs 1517: AmayaParseUrl: parse a Name relative to another name
1518:
1519: This returns those parts of a name which are given (and requested)
1520: substituting bits from the related name where necessary.
1.25 cvs 1521:
1.28 cvs 1522: On entry,
1.25 cvs 1523: aName A filename given
1524: relatedName A name relative to which aName is to be parsed. Give
1525: it an empty string if aName is absolute.
1526: wanted A mask for the bits which are wanted.
1527:
1.28 cvs 1528: On exit,
1.25 cvs 1529: returns A pointer to a malloc'd string which MUST BE FREED
1530: ----------------------------------------------------------------------*/
1.106 cvs 1531: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1532: {
1533: char *return_value;
1534: char result[MAX_LENGTH];
1535: char name[MAX_LENGTH];
1536: char rel[MAX_LENGTH];
1537: char *p, *access;
1.29 cvs 1538: HTURI given, related;
1539: int len;
1.106 cvs 1540: char used_sep;
1541: char *used_str;
1.32 cvs 1542:
1.106 cvs 1543: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1544: {
1.106 cvs 1545: used_str = DIR_STR;
1546: used_sep = DIR_SEP;
1.33 cvs 1547: }
1.32 cvs 1548: else
1.33 cvs 1549: {
1.106 cvs 1550: used_str = URL_STR;
1551: used_sep = URL_SEP;
1.33 cvs 1552: }
1.32 cvs 1553:
1.29 cvs 1554: /* Make working copies of input strings to cut up: */
1555: return_value = NULL;
1556: result[0] = 0; /* Clear string */
1.106 cvs 1557: strcpy (name, aName);
1.29 cvs 1558: if (relatedName != NULL)
1.106 cvs 1559: strcpy (rel, relatedName);
1.29 cvs 1560: else
1.106 cvs 1561: relatedName[0] = EOS;
1.29 cvs 1562:
1563: scan (name, &given);
1564: scan (rel, &related);
1565: access = given.access ? given.access : related.access;
1566: if (wanted & AMAYA_PARSE_ACCESS)
1567: if (access)
1568: {
1.106 cvs 1569: strcat (result, access);
1.29 cvs 1570: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1571: strcat (result, ":");
1.29 cvs 1572: }
1573:
1574: if (given.access && related.access)
1575: /* If different, inherit nothing. */
1.106 cvs 1576: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1577: {
1578: related.host = 0;
1579: related.absolute = 0;
1580: related.relative = 0;
1581: related.fragment = 0;
1582: }
1583:
1584: if (wanted & AMAYA_PARSE_HOST)
1585: if(given.host || related.host)
1586: {
1587: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1588: strcat (result, "//");
1589: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1590: }
1591:
1592: if (given.host && related.host)
1593: /* If different hosts, inherit no path. */
1.106 cvs 1594: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1595: {
1596: related.absolute = 0;
1597: related.relative = 0;
1598: related.fragment = 0;
1599: }
1600:
1601: if (wanted & AMAYA_PARSE_PATH)
1602: {
1603: if (given.absolute)
1604: {
1605: /* All is given */
1606: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1607: strcat (result, used_str);
1608: strcat (result, given.absolute);
1.25 cvs 1609: }
1.29 cvs 1610: else if (related.absolute)
1611: {
1612: /* Adopt path not name */
1.106 cvs 1613: strcat (result, used_str);
1614: strcat (result, related.absolute);
1.29 cvs 1615: if (given.relative)
1616: {
1617: /* Search part? */
1.106 cvs 1618: p = strchr (result, '?');
1.29 cvs 1619: if (!p)
1.106 cvs 1620: p=result+strlen(result)-1;
1.33 cvs 1621: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1622: /* Remove filename */
1623: p[1]=0;
1624: /* Add given one */
1.106 cvs 1625: strcat (result, given.relative);
1.25 cvs 1626: }
1627: }
1.29 cvs 1628: else if (given.relative)
1629: /* what we've got */
1.106 cvs 1630: strcat (result, given.relative);
1.29 cvs 1631: else if (related.relative)
1.106 cvs 1632: strcat (result, related.relative);
1.29 cvs 1633: else
1634: /* No inheritance */
1.106 cvs 1635: strcat (result, used_str);
1.25 cvs 1636: }
1.29 cvs 1637:
1638: if (wanted & AMAYA_PARSE_ANCHOR)
1639: if (given.fragment || related.fragment)
1640: {
1641: if (given.absolute && given.fragment)
1642: {
1643: /*Fixes for relURLs...*/
1644: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1645: strcat (result, "#");
1646: strcat (result, given.fragment);
1.29 cvs 1647: }
1648: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1649: strcat (result, "");
1.29 cvs 1650: else
1651: {
1.110 cvs 1652: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1653: strcat (result, "#");
1.110 cvs 1654: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1655: }
1656: }
1.106 cvs 1657: len = strlen (result);
1658: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1659: strcpy (return_value, result);
1.29 cvs 1660: return (return_value); /* exactly the right length */
1.25 cvs 1661: }
1662:
1663: /*----------------------------------------------------------------------
1664: HTCanon
1665: Canonicalizes the URL in the following manner starting from the host
1666: pointer:
1667:
1668: 1) The host name is converted to lowercase
1669: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1670:
1671: Return: OK The position of the current path part of the URL
1672: which might be the old one or a new one.
1673:
1674: ----------------------------------------------------------------------*/
1.106 cvs 1675: static char *HTCanon (char **filename, char *host)
1676: {
1677: char *newname = NULL;
1678: char used_sep;
1679: char *path;
1680: char *strptr;
1681: char *port;
1682: char *access = host-3;
1683:
1684: if (*filename && strchr (*filename, URL_SEP))
1685: used_sep = URL_SEP;
1.33 cvs 1686: else
1.106 cvs 1687: used_sep = DIR_SEP;
1.32 cvs 1688:
1.110 cvs 1689: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1690: access--;
1.110 cvs 1691: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1692: path = host + strlen (host);
1693: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1694: host = strptr;
1.110 cvs 1695: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1696: port = NULL;
1.25 cvs 1697:
1698: strptr = host; /* Convert to lower-case */
1.82 cvs 1699: while (strptr < path)
1.33 cvs 1700: {
1.123 vatton 1701: *strptr = tolower (*strptr);
1.82 cvs 1702: strptr++;
1.33 cvs 1703: }
1.25 cvs 1704:
1705: /* Does the URL contain a full domain name? This also works for a
1706: numerical host name. The domain name is already made lower-case
1707: and without a trailing dot. */
1708: {
1.106 cvs 1709: char *dot = port ? port : path;
1710: if (dot > *filename && *--dot == '.')
1.33 cvs 1711: {
1.106 cvs 1712: char *orig = dot;
1713: char *dest = dot + 1;
1.82 cvs 1714: while ((*orig++ = *dest++));
1715: if (port) port--;
1.33 cvs 1716: path--;
1.25 cvs 1717: }
1718: }
1719: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1720: if (port)
1721: {
1.82 cvs 1722: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1723: {
1724: if (!newname)
1725: {
1.106 cvs 1726: char *orig = port;
1727: char *dest = port + 1;
1.82 cvs 1728: while ((*orig++ = *dest++));
1.33 cvs 1729: }
1730: }
1.106 cvs 1731: else if ((!strncmp (access, "http", 4) &&
1732: (*(port + 1) == '8' &&
1733: *(port+2) == '0' &&
1.82 cvs 1734: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1735: (!strncmp (access, "gopher", 6) &&
1736: (*(port+1) == '7' &&
1737: *(port+2) == '0' &&
1.82 cvs 1738: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1739: (!strncmp (access, "ftp", 3) &&
1740: (*(port+1) == '2' &&
1741: *(port + 2) == '1' &&
1.82 cvs 1742: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1743: if (!newname)
1744: {
1.106 cvs 1745: char *orig = port;
1746: char *dest = port + 3;
1.33 cvs 1747: while((*orig++ = *dest++));
1748: /* Update path position, Henry Minsky */
1749: path -= 3;
1.25 cvs 1750: }
1.33 cvs 1751: }
1752: else if (newname)
1.106 cvs 1753: strncat (newname, port, (int) (path - port));
1.33 cvs 1754: }
1.25 cvs 1755:
1.33 cvs 1756: if (newname)
1757: {
1.106 cvs 1758: char *newpath = newname + strlen (newname);
1759: strcat (newname, path);
1.25 cvs 1760: path = newpath;
1.28 cvs 1761: /* Free old copy */
1762: TtaFreeMemory(*filename);
1.25 cvs 1763: *filename = newname;
1.33 cvs 1764: }
1.25 cvs 1765: return path;
1766: }
1767:
1768:
1769: /*----------------------------------------------------------------------
1.29 cvs 1770: SimplifyUrl: simplify a URI
1.32 cvs 1771: A URI is allowed to contain the sequence xxx/../ which may be
1772: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1773: Simplification helps us recognize duplicate URIs.
1.25 cvs 1774:
1.28 cvs 1775: Thus, /etc/junk/../fred becomes /etc/fred
1776: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1777:
1.28 cvs 1778: but we should NOT change
1779: http://fred.xxx.edu/../..
1.25 cvs 1780:
1781: or ../../albert.html
1782:
1.28 cvs 1783: In order to avoid empty URLs the following URLs become:
1.25 cvs 1784:
1785: /fred/.. becomes /fred/..
1786: /fred/././.. becomes /fred/..
1787: /fred/.././junk/.././ becomes /fred/..
1788:
1.28 cvs 1789: If more than one set of `://' is found (several proxies in cascade) then
1790: only the part after the last `://' is simplified.
1.25 cvs 1791:
1.28 cvs 1792: Returns: A string which might be the old one or a new one.
1.25 cvs 1793: ----------------------------------------------------------------------*/
1.106 cvs 1794: void SimplifyUrl (char **url)
1795: {
1796: char *path;
1797: char *access;
1798: char *newptr;
1799: char *p;
1800: char *orig, *dest, *end;
1.28 cvs 1801:
1.106 cvs 1802: char used_sep;
1.77 cvs 1803: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1804: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1805:
1.28 cvs 1806: if (!url || !*url)
1807: return;
1808:
1.106 cvs 1809: if (strchr (*url, URL_SEP))
1810: used_sep = URL_SEP;
1.32 cvs 1811: else
1.106 cvs 1812: used_sep = DIR_SEP;
1.32 cvs 1813:
1.77 cvs 1814: /* should we simplify double dot? */
1815: path = *url;
1.106 cvs 1816: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1817: ddot_simplify = FALSE;
1818: else
1819: ddot_simplify = TRUE;
1820:
1.28 cvs 1821: /* Find any scheme name */
1.106 cvs 1822: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1823: {
1824: /* Find host name */
1.28 cvs 1825: access = *url;
1.123 vatton 1826: while (access < path && (*access = tolower (*access)))
1.82 cvs 1827: access++;
1.28 cvs 1828: path += 3;
1.106 cvs 1829: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1830: /* For proxies */
1.106 cvs 1831: path = newptr + 3;
1.82 cvs 1832: /* We have a host name */
1.84 cvs 1833: path = HTCanon (url, path);
1.25 cvs 1834: }
1.106 cvs 1835: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1836: path += 2;
1837: else
1838: path = *url;
1.84 cvs 1839: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1840: /* Some URLs start //<foo> */
1841: path += 1;
1.94 cvs 1842: else if (IsFilePath (path))
1843: {
1844: /* doesn't need to do anything more */
1845: return;
1846: }
1.106 cvs 1847: else if (!strncmp (path, "news:", 5))
1.28 cvs 1848: {
1.106 cvs 1849: newptr = strchr (path+5, '@');
1.28 cvs 1850: if (!newptr)
1851: newptr = path + 5;
1852: while (*newptr)
1853: {
1854: /* Make group or host lower case */
1.123 vatton 1855: *newptr = tolower (*newptr);
1.28 cvs 1856: newptr++;
1.25 cvs 1857: }
1.28 cvs 1858: /* Doesn't need to do any more */
1859: return;
1.25 cvs 1860: }
1.130 cheyroul 1861:
1.126 cheyroul 1862:
1.28 cvs 1863: if ((p = path))
1864: {
1.106 cvs 1865: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1866: (end = strchr (path, '#'))))
1867: end = path + strlen (path);
1.28 cvs 1868:
1869: /* Parse string second time to simplify */
1870: p = path;
1871: while (p < end)
1872: {
1.110 cvs 1873: /* if we're pointing to a char, it's safe to reactivate the
1874: ../ convertion */
1.106 cvs 1875: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1876: ddot_simplify = TRUE;
1877:
1.33 cvs 1878: if (*p==used_sep)
1.28 cvs 1879: {
1.106 cvs 1880: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1881: {
1882: orig = p + 1;
1.84 cvs 1883: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1884: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1885: end = orig - 1;
1886: }
1.106 cvs 1887: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1888: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1889: {
1890: newptr = p;
1.52 cvs 1891: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1892: if (*newptr == used_sep)
1893: orig = newptr + 1;
1.28 cvs 1894: else
1.52 cvs 1895: orig = newptr;
1896:
1897: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1898: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1899: end = orig-1;
1900: /* Start again with prev slash */
1901: p = newptr;
1.28 cvs 1902: }
1.33 cvs 1903: else if (*(p+1) == used_sep)
1.28 cvs 1904: {
1.33 cvs 1905: while (*(p+1) == used_sep)
1.28 cvs 1906: {
1907: orig = p;
1908: dest = p + 1;
1909: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1910: end = orig-1;
1911: }
1912: }
1913: else
1.25 cvs 1914: p++;
1.28 cvs 1915: }
1916: else
1917: p++;
1.25 cvs 1918: }
1919: }
1.51 cvs 1920: /*
1921: ** Check for host/../.. kind of things
1922: */
1.106 cvs 1923: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1924: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 1925: *(path+1) = EOS;
1.28 cvs 1926: return;
1927: }
1928:
1929:
1930: /*----------------------------------------------------------------------
1.96 cvs 1931: NormalizeFile normalizes local names.
1.28 cvs 1932: Return TRUE if target and src differ.
1933: ----------------------------------------------------------------------*/
1.106 cvs 1934: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 1935: {
1.110 cvs 1936: #ifndef _WINDOWS
1.106 cvs 1937: char *s;
1.93 cvs 1938: int i;
1.110 cvs 1939: #endif /* !_WINDOWS */
1.82 cvs 1940: ThotBool change;
1.90 cvs 1941: int start_index; /* the first char that we'll copy */
1.28 cvs 1942:
1.54 cvs 1943: change = FALSE;
1.90 cvs 1944: start_index = 0;
1945:
1.106 cvs 1946: if (!src || src[0] == EOS)
1.96 cvs 1947: {
1.106 cvs 1948: target[0] = EOS;
1.96 cvs 1949: return FALSE;
1950: }
1.90 cvs 1951:
1952: /* @@ do I need file: or file:/ here? */
1.106 cvs 1953: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1954: {
1.90 cvs 1955: /* remove the prefix file: */
1956: start_index += 5;
1957:
1958: /* remove the localhost prefix */
1.106 cvs 1959: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 1960: start_index += 11;
1961:
1962: /* remove the first two slashes in / / /path */
1963: while (src[start_index] &&
1.106 cvs 1964: src[start_index] == '/'
1965: && src[start_index + 1] == '/')
1.94 cvs 1966: start_index++;
1967:
1968: #ifdef _WINDOWS
1969: /* remove any extra slash before the drive name */
1.106 cvs 1970: if (src[start_index] == '/'
1971: &&src[start_index+2] == ':')
1.94 cvs 1972: start_index++;
1973: #endif /* _WINDOWS */
1.90 cvs 1974:
1.106 cvs 1975: if (src[start_index] == EOS)
1.90 cvs 1976: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 1977: strcpy (target, DIR_STR);
1.90 cvs 1978: else
1.97 cvs 1979: /* as we're inside a file: URL, we'll apply all the convertions
1980: we know */
1981: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1982:
1983: change = TRUE;
1984: }
1.97 cvs 1985: else if (convertion != AM_CONV_NONE)
1.96 cvs 1986: {
1987: /* we are following a "local" relative link, we do all the
1988: convertions except for the HOME_DIR ~ one */
1.97 cvs 1989: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1990: }
1.90 cvs 1991: #ifndef _WINDOWS
1.106 cvs 1992: else if (src[0] == '~')
1.53 cvs 1993: {
1.96 cvs 1994: /* it must be a URL typed in a text input field */
1995: /* do the HOME_DIR ~ substitution */
1.82 cvs 1996: s = TtaGetEnvString ("HOME");
1.106 cvs 1997: strcpy (target, s);
1.90 cvs 1998: #if 0
1.96 cvs 1999: /* JK: invalidated this part of the code as it's simpler
2000: to add the DIR_SEP whenever we have something to add
2001: to the path rather than adding it systematically */
1.106 cvs 2002: if (src[1] != DIR_SEP)
2003: strcat (target, DIR_STR);
1.90 cvs 2004: #endif
1.106 cvs 2005: i = strlen (target);
2006: strcpy (&target[i], &src[1]);
1.54 cvs 2007: change = TRUE;
1.53 cvs 2008: }
1.90 cvs 2009: #endif /* _WINDOWS */
1.28 cvs 2010: else
1.96 cvs 2011: /* leave it as it is */
1.106 cvs 2012: strcpy (target, src);
1.96 cvs 2013:
1.28 cvs 2014: /* remove /../ and /./ */
1.29 cvs 2015: SimplifyUrl (&target);
1.54 cvs 2016: if (!change)
1.106 cvs 2017: change = strcmp (src, target);
1.28 cvs 2018: return (change);
1.25 cvs 2019: }
2020:
1.28 cvs 2021:
1.25 cvs 2022: /*----------------------------------------------------------------------
1.31 cvs 2023: MakeRelativeURL: make relative name
1.25 cvs 2024:
1.28 cvs 2025: This function creates and returns a string which gives an expression of
2026: one address as related to another. Where there is no relation, an absolute
2027: address is retured.
1.25 cvs 2028:
1.28 cvs 2029: On entry,
1.25 cvs 2030: Both names must be absolute, fully qualified names of nodes
2031: (no fragment bits)
2032:
1.28 cvs 2033: On exit,
1.25 cvs 2034: The return result points to a newly allocated name which, if
2035: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2036: The caller is responsible for freeing the resulting name later.
2037: ----------------------------------------------------------------------*/
1.106 cvs 2038: char *MakeRelativeURL (char *aName, char *relatedName)
2039: {
2040: char *return_value;
2041: char result[MAX_LENGTH];
2042: char *p;
2043: char *q;
2044: char *after_access;
2045: char *last_slash = NULL;
2046: int slashes, levels, len;
1.110 cvs 2047: #ifdef _WINDOWS
1.44 cvs 2048: int ndx;
1.110 cvs 2049: #endif /* _WINDOWS */
1.44 cvs 2050:
1.29 cvs 2051: if (aName == NULL || relatedName == NULL)
2052: return (NULL);
2053:
2054: slashes = 0;
2055: after_access = NULL;
2056: p = aName;
2057: q = relatedName;
2058: for (; *p && (*p == *q); p++, q++)
1.27 cvs 2059: {
2060: /* Find extent of match */
1.106 cvs 2061: if (*p == ':')
1.29 cvs 2062: after_access = p + 1;
1.28 cvs 2063: if (*p == DIR_SEP)
1.27 cvs 2064: {
1.29 cvs 2065: /* memorize the last slash position and count them */
1.27 cvs 2066: last_slash = p;
2067: slashes++;
1.25 cvs 2068: }
2069: }
2070:
1.31 cvs 2071: /* q, p point to the first non-matching character or zero */
1.106 cvs 2072: if (*q == EOS)
1.31 cvs 2073: {
2074: /* New name is a subset of the related name */
2075: /* exactly the right length */
1.106 cvs 2076: len = strlen (p);
2077: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2078: strcpy (return_value, p);
1.31 cvs 2079: }
2080: else if ((slashes < 2 && after_access == NULL)
2081: || (slashes < 3 && after_access != NULL))
2082: {
2083: /* Two names whitout common path */
2084: /* exactly the right length */
1.106 cvs 2085: len = strlen (aName);
2086: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2087: strcpy (return_value, aName);
1.31 cvs 2088: }
2089: else
2090: {
2091: /* Some path in common */
1.106 cvs 2092: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2093: /* just the same server */
1.106 cvs 2094: strcpy (result, last_slash);
1.31 cvs 2095: else
2096: {
2097: levels= 0;
1.106 cvs 2098: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2099: if (*q == DIR_SEP)
2100: levels++;
2101:
1.106 cvs 2102: result[0] = EOS;
1.31 cvs 2103: for (;levels; levels--)
1.106 cvs 2104: strcat (result, "../");
2105: strcat (result, last_slash+1);
1.31 cvs 2106: }
1.52 cvs 2107:
2108: if (!*result)
1.106 cvs 2109: strcat (result, "./");
1.52 cvs 2110:
1.31 cvs 2111: /* exactly the right length */
1.106 cvs 2112: len = strlen (result);
2113: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2114: strcpy (return_value, result);
1.52 cvs 2115:
1.25 cvs 2116: }
1.110 cvs 2117: #ifdef _WINDOWS
1.106 cvs 2118: len = strlen (return_value);
1.44 cvs 2119: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2120: if (return_value[ndx] == '\\')
2121: return_value[ndx] = '/' ;
1.110 cvs 2122: #endif /* _WINDOWS */
1.29 cvs 2123: return (return_value);
1.24 cvs 2124: }
1.35 cvs 2125:
1.104 kahan 2126: /*----------------------------------------------------------------------
2127: AM_GetFileSize
2128: Returns TRUE and the filesize in the 2nd parameter.
2129: Otherwise, in case of a system error, returns FALSE, with a
2130: filesize of 0L.
2131: ---------------------------------------------------------------------*/
1.106 cvs 2132: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2133: {
1.106 cvs 2134: ThotFileHandle handle = ThotFile_BADHANDLE;
2135: ThotFileInfo info;
1.35 cvs 2136:
1.104 kahan 2137: *file_size = 0L;
2138: if (!TtaFileExist (filename))
2139: return FALSE;
2140:
2141: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2142: if (handle == ThotFile_BADHANDLE)
2143: /* ThotFile_BADHANDLE */
2144: return FALSE;
2145: if (TtaFileStat (handle, &info) == 0)
2146: /* bad stat */
2147: info.size = 0L;
2148: TtaFileClose (handle);
2149: *file_size = (unsigned long) info.size;
2150: return TRUE;
2151: }
Webmaster