Annotation of Amaya/amaya/AHTURLTools.c, revision 1.133
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") ||
! 542: !strcasecmp (suffix, "tar") ||
! 543: !strcasecmp (suffix, "tgz") ||
! 544: !strcasecmp (suffix, "ddl") ||
! 545: !strcasecmp (suffix, "o"))
! 546: return (TRUE);
! 547: else if (!strcmp (suffix, "gz"))
! 548: {
! 549: /* take into account compressed files */
! 550: TtaExtractSuffix (temppath, suffix);
! 551: if (!strcasecmp (suffix, "exe") ||
! 552: !strcasecmp (suffix, "pdf") ||
! 553: !strcasecmp (suffix, "tar") ||
! 554: !strcasecmp (suffix, "ddl") ||
! 555: !strcasecmp (suffix, "o"))
! 556: return (TRUE);
! 557: else
! 558: return (FALSE);
! 559: }
! 560: else
! 561: return (FALSE);
! 562: }
! 563:
! 564: /*----------------------------------------------------------------------
1.103 cvs 565: IsMathMLName
566: returns TRUE if path points to an MathML resource.
567: ----------------------------------------------------------------------*/
1.111 cvs 568: ThotBool IsMathMLName (const char *path)
1.103 cvs 569: {
1.106 cvs 570: char temppath[MAX_LENGTH];
571: char suffix[MAX_LENGTH];
1.103 cvs 572:
573: if (!path)
574: return (FALSE);
575:
1.106 cvs 576: strcpy (temppath, path);
1.124 vatton 577: TtaExtractSuffix (temppath, suffix);
1.103 cvs 578:
1.106 cvs 579: if (!strcasecmp (suffix, "mml"))
1.103 cvs 580: return (TRUE);
1.106 cvs 581: else if (!strcmp (suffix, "gz"))
1.103 cvs 582: {
583: /* take into account compressed files */
1.124 vatton 584: TtaExtractSuffix (temppath, suffix);
1.106 cvs 585: if (!strcasecmp (suffix, "mml"))
1.103 cvs 586: return (TRUE);
587: else
588: return (FALSE);
589: }
590: else
591: return (FALSE);
592: }
593:
594: /*----------------------------------------------------------------------
595: IsSVGName
1.119 kahan 596: returns TRUE if path points to an SVG resource.
1.103 cvs 597: ----------------------------------------------------------------------*/
1.111 cvs 598: ThotBool IsSVGName (const char *path)
1.103 cvs 599: {
1.106 cvs 600: char temppath[MAX_LENGTH];
601: char suffix[MAX_LENGTH];
1.103 cvs 602:
603: if (!path)
604: return (FALSE);
605:
1.106 cvs 606: strcpy (temppath, path);
1.124 vatton 607: TtaExtractSuffix (temppath, suffix);
1.103 cvs 608:
1.106 cvs 609: if (!strcasecmp (suffix, "svg"))
1.103 cvs 610: return (TRUE);
1.106 cvs 611: else if (!strcmp (suffix, "gz"))
1.103 cvs 612: {
613: /* take into account compressed files */
1.124 vatton 614: TtaExtractSuffix (temppath, suffix);
1.106 cvs 615: if (!strcasecmp (suffix, "svg"))
1.103 cvs 616: return (TRUE);
617: else
618: return (FALSE);
619: }
620: else
621: return (FALSE);
622: }
623:
624: /*----------------------------------------------------------------------
1.60 cvs 625: IsCSSName
626: returns TRUE if path points to an XML resource.
627: ----------------------------------------------------------------------*/
1.111 cvs 628: ThotBool IsCSSName (const char *path)
1.60 cvs 629: {
1.106 cvs 630: char temppath[MAX_LENGTH];
631: char suffix[MAX_LENGTH];
1.60 cvs 632:
633: if (!path)
634: return (FALSE);
635:
1.106 cvs 636: strcpy (temppath, path);
1.124 vatton 637: TtaExtractSuffix (temppath, suffix);
1.60 cvs 638:
1.106 cvs 639: if (!strcasecmp (suffix, "css"))
1.60 cvs 640: return (TRUE);
1.106 cvs 641: else if (!strcmp (suffix, "gz"))
1.60 cvs 642: {
643: /* take into account compressed files */
1.124 vatton 644: TtaExtractSuffix (temppath, suffix);
1.106 cvs 645: if (!strcasecmp (suffix, "css"))
1.56 cvs 646: return (TRUE);
647: else
648: return (FALSE);
649: }
650: else
651: return (FALSE);
652: }
653:
654: /*----------------------------------------------------------------------
1.9 cvs 655: IsImageName
656: returns TRUE if path points to an image resource.
1.4 cvs 657: ----------------------------------------------------------------------*/
1.111 cvs 658: ThotBool IsImageName (const char *path)
1.106 cvs 659: {
660: char temppath[MAX_LENGTH];
661: char suffix[MAX_LENGTH];
662: char nsuffix[MAX_LENGTH];
1.5 cvs 663: int i;
664:
665: if (!path)
1.13 cvs 666: return (FALSE);
1.5 cvs 667:
1.106 cvs 668: strcpy (temppath, path);
1.124 vatton 669: TtaExtractSuffix (temppath, suffix);
1.5 cvs 670:
671: /* Normalize the suffix */
672: i = 0;
1.106 cvs 673: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 674: {
1.123 vatton 675: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 676: i++;
677: }
1.106 cvs 678: nsuffix[i] = EOS;
679: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
680: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
681: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 682: return (TRUE);
683: return (FALSE);
1.3 cvs 684: }
685:
1.4 cvs 686: /*----------------------------------------------------------------------
1.58 cvs 687: IsImageType
688: returns TRUE if type points to an image resource.
689: ----------------------------------------------------------------------*/
1.111 cvs 690: ThotBool IsImageType (const char *type)
1.58 cvs 691: {
1.106 cvs 692: char temptype[MAX_LENGTH];
1.58 cvs 693: int i;
694:
695: if (!type)
696: return (FALSE);
697:
1.106 cvs 698: strcpy (temptype, type);
1.58 cvs 699: /* Normalize the type */
700: i = 0;
1.106 cvs 701: while (temptype[i] != EOS)
1.58 cvs 702: {
703: temptype[i] = tolower (temptype[i]);
704: i++;
705: }
1.111 cvs 706: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
707: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
708: !strcmp (temptype, "png"))
1.58 cvs 709: return (TRUE);
710: return (FALSE);
711: }
712:
713: /*----------------------------------------------------------------------
1.9 cvs 714: IsTextName
1.4 cvs 715: ----------------------------------------------------------------------*/
1.111 cvs 716: ThotBool IsTextName (const char *path)
1.106 cvs 717: {
718: char temppath[MAX_LENGTH];
719: char suffix[MAX_LENGTH];
720: char nsuffix[MAX_LENGTH];
1.5 cvs 721: int i;
722:
723: if (!path)
1.13 cvs 724: return (FALSE);
1.5 cvs 725:
1.106 cvs 726: strcpy (temppath, path);
1.124 vatton 727: TtaExtractSuffix (temppath, suffix);
1.5 cvs 728:
729: /* Normalize the suffix */
730: i = 0;
1.106 cvs 731: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 732: {
1.25 cvs 733: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 734: i++;
735: }
1.106 cvs 736: nsuffix[i] = EOS;
1.5 cvs 737:
1.111 cvs 738: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 739: return (TRUE);
1.106 cvs 740: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 741: {
1.39 cvs 742: /* take into account compressed files */
1.124 vatton 743: TtaExtractSuffix (temppath, suffix);
1.13 cvs 744: /* Normalize the suffix */
745: i = 0;
1.106 cvs 746: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 747: {
1.25 cvs 748: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 749: i++;
750: }
1.106 cvs 751: nsuffix[i] = EOS;
1.111 cvs 752: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 753: return (TRUE);
754: else
755: return (FALSE);
756: }
757: else
758: return (FALSE);
1.3 cvs 759: }
760:
1.4 cvs 761: /*----------------------------------------------------------------------
1.9 cvs 762: IsHTTPPath
763: returns TRUE if path is in fact an http URL.
1.4 cvs 764: ----------------------------------------------------------------------*/
1.112 cvs 765: ThotBool IsHTTPPath (const char *path)
1.3 cvs 766: {
1.5 cvs 767: if (!path)
768: return FALSE;
1.3 cvs 769:
1.106 cvs 770: if ((!strncmp (path, "http:", 5) != 0)
771: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
772: || !strncmp (path, "internal:", 9))
1.58 cvs 773: return TRUE;
774: return FALSE;
1.3 cvs 775: }
776:
1.4 cvs 777: /*----------------------------------------------------------------------
1.9 cvs 778: IsWithParameters
779: returns TRUE if url has a concatenated query string.
1.4 cvs 780: ----------------------------------------------------------------------*/
1.133 ! vatton 781: ThotBool IsWithParameters (const char *url)
1.3 cvs 782: {
1.5 cvs 783: int i;
1.3 cvs 784:
1.9 cvs 785: if ((!url) || (url[0] == EOS))
1.5 cvs 786: return FALSE;
1.3 cvs 787:
1.9 cvs 788: i = strlen (url) - 1;
789: while (i > 0 && url[i--] != '?')
1.5 cvs 790: if (i < 0)
791: return FALSE;
1.3 cvs 792:
1.5 cvs 793: /* There is a parameter */
794: return TRUE;
1.3 cvs 795: }
796:
1.4 cvs 797: /*----------------------------------------------------------------------
1.9 cvs 798: IsW3Path
799: returns TRUE if path is in fact a URL.
1.4 cvs 800: ----------------------------------------------------------------------*/
1.133 ! vatton 801: ThotBool IsW3Path (const char *path)
1.106 cvs 802: {
803: if (strncmp (path, "http:", 5) &&
804: strncmp (path, "ftp:", 4) &&
805: strncmp (path, "telnet:", 7) &&
806: strncmp (path, "wais:", 5) &&
807: strncmp (path, "news:", 5) &&
808: strncmp (path, "gopher:", 7) &&
809: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 810: strncmp (path, "archie:", 7) &&
811: strncmp (path, "https:", 6))
1.72 cvs 812: return FALSE;
813: return TRUE;
1.3 cvs 814: }
815:
1.4 cvs 816: /*----------------------------------------------------------------------
1.90 cvs 817: IsFilePath
818: returns TRUE if path is in fact a URL.
819: ----------------------------------------------------------------------*/
1.133 ! vatton 820: ThotBool IsFilePath (const char *path)
1.90 cvs 821: {
1.106 cvs 822: if (strncmp (path, "file:", 5))
1.90 cvs 823: return FALSE;
824: return TRUE;
825: }
826:
827: /*----------------------------------------------------------------------
1.9 cvs 828: IsValidProtocol
829: returns true if the url protocol is supported by Amaya.
1.4 cvs 830: ----------------------------------------------------------------------*/
1.133 ! vatton 831: ThotBool IsValidProtocol (const char *url)
1.106 cvs 832: {
833: if (!strncmp (url, "http:", 5)
834: || !strncmp (url, "internal:", 9)
835: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 836: /* experimental */
1.24 cvs 837: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 838: return (TRUE);
1.5 cvs 839: else
1.8 cvs 840: return (FALSE);
1.3 cvs 841: }
842:
1.31 cvs 843:
844: /*----------------------------------------------------------------------
845: GetBaseURL
846: normalizes orgName according to a base associated with doc, and
847: following the standard URL format rules.
848: The function returns the base used to solve relative URL and SRC:
849: - the base of the document,
850: - or the document path (without document name).
851: ----------------------------------------------------------------------*/
1.106 cvs 852: char *GetBaseURL (Document doc)
1.31 cvs 853: {
854: Element el;
855: ElementType elType;
856: AttributeType attrType;
857: Attribute attr;
1.106 cvs 858: char *ptr, *basename;
1.31 cvs 859: int length;
860:
1.113 cvs 861: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 862: return NULL;
1.106 cvs 863: basename = TtaGetMemory (MAX_LENGTH);
864: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
865: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 866: length = MAX_LENGTH -1;
1.113 cvs 867: /* is it a HTML document ? */
1.31 cvs 868: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.113 cvs 869: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
870: /* it's a HTML document */
1.65 cvs 871: {
1.113 cvs 872: /* get the document element */
873: el = TtaGetMainRoot (doc);
874: /* search the BASE element */
875: elType.ElTypeNum = HTML_EL_HEAD;
876: el = TtaSearchTypedElement (elType, SearchForward, el);
877: if (el)
878: /* there is a HEAD element */
879: {
880: /* look for a BASE element within the HEAD */
881: elType.ElTypeNum = HTML_EL_BASE;
882: el = TtaSearchTypedElement (elType, SearchInTree, el);
883: }
884: if (el)
1.31 cvs 885: {
1.113 cvs 886: /* The document has a BASE element. Get the HREF attribute of the
887: BASE element */
888: attrType.AttrSSchema = elType.ElSSchema;
889: attrType.AttrTypeNum = HTML_ATTR_HREF_;
890: attr = TtaGetAttribute (el, attrType);
891: if (attr)
1.31 cvs 892: {
1.113 cvs 893: /* Use the base path of the document */
894: TtaGiveTextAttributeValue (attr, basename, &length);
895: /* base and orgName have to be separated by a DIR_SEP */
896: length--;
897: if (basename[0] != EOS && basename[length] != URL_SEP &&
898: basename[length] != DIR_SEP)
899: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 900: {
1.113 cvs 901: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
902: AMAYA_PARSE_HOST |
903: AMAYA_PARSE_PUNCTUATION);
904: if (ptr && !strcmp (ptr, basename))
905: {
906: /* it has this form, complete it by adding a URL_STR */
907: if (strchr (basename, DIR_SEP))
908: strcat (basename, DIR_STR);
909: else
910: strcat (basename, URL_STR);
911: length++;
912: }
913: if (ptr)
914: TtaFreeMemory (ptr);
1.31 cvs 915: }
916: }
917: }
1.113 cvs 918: }
919:
1.31 cvs 920: /* Remove anything after the last DIR_SEP char. If no such char is found,
921: * then search for the first ":" char, hoping that what's before that is a
922: * protocol. If found, end the string there. If neither char is found,
923: * then discard the whole base element.
924: */
1.106 cvs 925: length = strlen (basename) - 1;
1.31 cvs 926: /* search for the last DIR_SEP char */
1.106 cvs 927: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 928: length--;
929: if (length >= 0)
930: /* found the last DIR_SEP char, end the string there */
1.106 cvs 931: basename[length + 1] = EOS;
1.31 cvs 932: else
933: /* search for the first PATH_STR char */
934: {
1.106 cvs 935: for (length = 0; basename[length] != ':' &&
936: basename[length] != EOS; length ++);
937: if (basename[length] == ':')
1.31 cvs 938: /* found, so end the string there */
1.106 cvs 939: basename[length + 1] = EOS;
1.31 cvs 940: else
941: /* not found, discard the base */
1.106 cvs 942: basename[0] = EOS;
1.31 cvs 943: }
944: return (basename);
945: }
946:
947:
1.4 cvs 948: /*----------------------------------------------------------------------
1.40 cvs 949: GetLocalPath
950: Allocate and return the local document path associated to the url
951: ----------------------------------------------------------------------*/
1.106 cvs 952: char *GetLocalPath (Document doc, char *url)
953: {
954: char *ptr;
955: char *n;
956: char *documentname;
957: char url_sep;
1.83 cvs 958: int len;
1.67 cvs 959: ThotBool noFile;
1.40 cvs 960:
961: if (url != NULL)
962: {
963: /* check whether the file name exists */
1.106 cvs 964: len = strlen (url) - 1;
1.71 cvs 965: if (IsW3Path (url))
1.106 cvs 966: url_sep = '/';
1.41 cvs 967: else
1.106 cvs 968: url_sep = DIR_SEP;
1.41 cvs 969: noFile = (url[len] == url_sep);
1.40 cvs 970: if (noFile)
1.106 cvs 971: url[len] = EOS;
972: ptr = TtaGetMemory (MAX_LENGTH);
973: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 974: TtaExtractName (url, ptr, documentname);
1.106 cvs 975: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 976: if (!TtaCheckDirectory (ptr))
977: /* directory did not exist */
1.72 cvs 978: TtaMakeDirectory (ptr);
1.47 cvs 979:
980: /* don't include the query string within document name */
1.106 cvs 981: n = strrchr (documentname, '?');
1.47 cvs 982: if (n != NULL)
1.106 cvs 983: *n = EOS;
1.46 cvs 984: /* don't include ':' within document name */
1.106 cvs 985: n = strchr (documentname, ':');
1.46 cvs 986: if (n != NULL)
1.106 cvs 987: *n = EOS;
1.69 cvs 988: /* if after all this operations document name
989: is empty, let's use noname.html instead */
1.106 cvs 990: if (documentname[0] == EOS)
991: strcat (ptr, "noname.html");
1.69 cvs 992: else
1.106 cvs 993: strcat (ptr, documentname);
1.40 cvs 994: TtaFreeMemory (documentname);
995: /* restore the url */
996: if (noFile)
1.41 cvs 997: url[len] = url_sep;
1.40 cvs 998: return (ptr);
999: }
1000: else
1001: return (NULL);
1002: }
1003:
1.73 cvs 1004: /*----------------------------------------------------------------------
1.79 cvs 1005: ExtractTarget extract the target name from document nane.
1006: ----------------------------------------------------------------------*/
1.106 cvs 1007: void ExtractTarget (char *aName, char *target)
1.79 cvs 1008: {
1.106 cvs 1009: int lg, i;
1010: char *ptr;
1011: char *oldptr;
1.79 cvs 1012:
1013: if (!target || !aName)
1014: /* bad target */
1015: return;
1016:
1.106 cvs 1017: target[0] = EOS;
1018: lg = strlen (aName);
1.79 cvs 1019: if (lg)
1020: {
1021: /* the name is not empty */
1022: oldptr = ptr = &aName[0];
1023: do
1024: {
1.106 cvs 1025: ptr = strrchr (oldptr, '#');
1.79 cvs 1026: if (ptr)
1027: oldptr = &ptr[1];
1028: }
1029: while (ptr);
1030:
1031: i = (int) (oldptr) - (int) (aName); /* name length */
1032: if (i > 1)
1033: {
1.106 cvs 1034: aName[i - 1] = EOS;
1.79 cvs 1035: if (i != lg)
1.106 cvs 1036: strcpy (target, oldptr);
1.79 cvs 1037: }
1038: }
1039: }
1040:
1041: /*----------------------------------------------------------------------
1.90 cvs 1042: RemoveNewLines (text)
1043: Removes any '\n' chars that are found in text.
1044: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1045: ----------------------------------------------------------------------*/
1.106 cvs 1046: ThotBool RemoveNewLines (char *text)
1047: {
1048: ThotBool change = FALSE;
1049: char *src;
1050: char *dest;
1.90 cvs 1051:
1052: src = text;
1053: dest = text;
1.115 kahan 1054:
1055: /* remove any preceding whitespace */
1056: while (*src && *src == ' ')
1057: {
1058: src++;
1059: change = 1;
1060: }
1061:
1.90 cvs 1062: while (*src)
1063: {
1064: switch (*src)
1065: {
1.106 cvs 1066: case '\n':
1.90 cvs 1067: /* don't copy the newline */
1068: change = 1;
1069: break;
1070: default:
1071: *dest = *src;
1072: dest++;
1073: break;
1074: }
1075: src++;
1076: }
1077: /* copy the last EOS char */
1078: *dest = *src;
1079:
1080: return (change);
1081: }
1082:
1083: /*----------------------------------------------------------------------
1084: CleanCopyFileURL
1085: Copies a file url from a src string to destination string.
1.97 cvs 1086: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1087: we want to do).
1.90 cvs 1088: ----------------------------------------------------------------------*/
1.106 cvs 1089: static void CleanCopyFileURL (char *dest, char *src,
1090: ConvertionType convertion)
1.90 cvs 1091: {
1092: while (*src)
1.89 cvs 1093: {
1.90 cvs 1094: switch (*src)
1.89 cvs 1095: {
1096: #ifdef _WINDOWS
1.106 cvs 1097: case URL_SEP:
1.96 cvs 1098: /* make DIR_SEP transformation */
1.97 cvs 1099: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1100: *dest = DIR_SEP;
1.96 cvs 1101: else
1102: *dest = *src;
1.90 cvs 1103: dest++;
1.96 cvs 1104: src++;
1.90 cvs 1105: break;
1.89 cvs 1106: #endif /* _WINDOWS */
1.96 cvs 1107:
1.106 cvs 1108: case '%':
1.97 cvs 1109: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1110: {
1.97 cvs 1111: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1112: src++;
1.106 cvs 1113: if (*src != EOS)
1.97 cvs 1114: {
1115: *dest = UnEscapeChar (*src) * 16;
1116: src++;
1117: }
1.106 cvs 1118: if (*src != EOS)
1.97 cvs 1119: {
1120: *dest = *dest + UnEscapeChar (*src);
1121: src++;
1122: }
1123: dest++;
1.96 cvs 1124: }
1.97 cvs 1125: else
1.96 cvs 1126: {
1.97 cvs 1127: *dest = *src;
1128: dest++;
1.96 cvs 1129: src++;
1130: }
1131: break;
1132:
1.90 cvs 1133: default:
1134: *dest = *src;
1.89 cvs 1135: dest++;
1.96 cvs 1136: src++;
1.90 cvs 1137: break;
1.89 cvs 1138: }
1139: }
1.90 cvs 1140: /* copy the EOS char */
1141: *dest = *src;
1.73 cvs 1142: }
1.40 cvs 1143:
1144: /*----------------------------------------------------------------------
1.9 cvs 1145: NormalizeURL
1146: normalizes orgName according to a base associated with doc, and
1147: following the standard URL format rules.
1.113 cvs 1148: if doc is < 0, use as a base the URL of the document that contains
1149: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1150: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1151: other path.
1.9 cvs 1152: The function returns the new complete and normalized URL
1.12 cvs 1153: or file name path (newName) and the name of the document (docName).
1.9 cvs 1154: N.B. If the function can't find out what's the docName, it assigns
1155: the name "noname.html".
1.4 cvs 1156: ----------------------------------------------------------------------*/
1.106 cvs 1157: void NormalizeURL (char *orgName, Document doc, char *newName,
1158: char *docName, char *otherPath)
1159: {
1160: char *basename;
1161: char tempOrgName[MAX_LENGTH];
1162: char *ptr;
1163: char used_sep;
1.84 cvs 1164: int length;
1165: ThotBool check;
1.5 cvs 1166:
1.110 cvs 1167: #ifdef _WINDOWS
1.44 cvs 1168: int ndx;
1.110 cvs 1169: #endif /* _WINDOWS */
1.44 cvs 1170:
1.5 cvs 1171: if (!newName || !docName)
1172: return;
1.18 cvs 1173:
1.113 cvs 1174: if (doc < 0)
1175: basename = TtaStrdup (SavedDocumentURL);
1176: else if (doc > 0)
1.53 cvs 1177: basename = GetBaseURL (doc);
1178: else if (otherPath != NULL)
1.108 cvs 1179: basename = TtaStrdup (otherPath);
1.32 cvs 1180: else
1.53 cvs 1181: basename = NULL;
1.32 cvs 1182:
1.18 cvs 1183: /*
1.31 cvs 1184: * Clean orgName
1185: * Make sure we have a complete orgName, without any leading or trailing
1186: * white spaces, or trailinbg new lines
1187: */
1.5 cvs 1188: ptr = orgName;
1.18 cvs 1189: /* skip leading white space and new line characters */
1.106 cvs 1190: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1191: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1192: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1193: /*
1.31 cvs 1194: * Make orgName a complete URL
1195: * If the URL does not include a protocol, then try to calculate
1196: * one using the doc's base element (if it exists),
1197: */
1.106 cvs 1198: if (tempOrgName[0] == EOS)
1.53 cvs 1199: {
1.106 cvs 1200: newName[0] = EOS;
1201: docName[0] = EOS;
1.53 cvs 1202: TtaFreeMemory (basename);
1203: return;
1204: }
1.49 cvs 1205:
1206: /* clean trailing white space */
1.106 cvs 1207: length = strlen (tempOrgName) - 1;
1208: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1209: {
1.106 cvs 1210: tempOrgName[length] = EOS;
1.53 cvs 1211: length--;
1212: }
1.50 cvs 1213:
1.55 cvs 1214: /* remove extra dot (which dot???) */
1215: /* ugly, but faster than a strcmp */
1.106 cvs 1216: if (tempOrgName[length] == '.'
1217: && (length == 0 || tempOrgName[length-1] != '.'))
1218: tempOrgName[length] = EOS;
1.50 cvs 1219:
1.94 cvs 1220: if (IsW3Path (tempOrgName))
1.53 cvs 1221: {
1222: /* the name is complete, go to the Sixth Step */
1.106 cvs 1223: strcpy (newName, tempOrgName);
1.53 cvs 1224: SimplifyUrl (&newName);
1225: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1226: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1227: AMAYA_PARSE_HOST |
1228: AMAYA_PARSE_PUNCTUATION);
1229: if (ptr && !strcmp (ptr, newName))
1230: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1231: strcat (newName, URL_STR);
1.49 cvs 1232:
1.53 cvs 1233: if (ptr)
1.50 cvs 1234: TtaFreeMemory (ptr);
1.53 cvs 1235: }
1.113 cvs 1236: else if (basename == NULL)
1.53 cvs 1237: /* the name is complete, go to the Sixth Step */
1.106 cvs 1238: strcpy (newName, tempOrgName);
1.53 cvs 1239: else
1240: {
1.31 cvs 1241: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1242: #ifdef _WINDOWS
1.53 cvs 1243: if (!IsW3Path (basename))
1244: {
1.106 cvs 1245: length = strlen (tempOrgName);
1.53 cvs 1246: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1247: if (tempOrgName [ndx] == '/')
1248: tempOrgName [ndx] = '\\';
1.53 cvs 1249: }
1.110 cvs 1250: #endif /* _WINDOWS */
1.25 cvs 1251: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1252: if (ptr)
1253: {
1254: SimplifyUrl (&ptr);
1.106 cvs 1255: strcpy (newName, ptr);
1.53 cvs 1256: TtaFreeMemory (ptr);
1257: }
1258: else
1.106 cvs 1259: newName[0] = EOS;
1.53 cvs 1260: }
1.36 cvs 1261:
1262: TtaFreeMemory (basename);
1.18 cvs 1263: /*
1.31 cvs 1264: * Prepare the docname that will refer to this ressource in the
1265: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1266: * noname.html as a default ressource name
1.18 cvs 1267: */
1.106 cvs 1268: if (newName[0] != EOS)
1.53 cvs 1269: {
1.106 cvs 1270: length = strlen (newName) - 1;
1271: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1272: {
1273: used_sep = newName[length];
1274: check = TRUE;
1275: while (check)
1276: {
1.50 cvs 1277: length--;
1278: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1279: length--;
1.106 cvs 1280: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1281: {
1.106 cvs 1282: newName[length+1] = EOS;
1.53 cvs 1283: /* remove also previous directory */
1284: length--;
1285: while (length >= 0 && newName[length] != used_sep)
1286: length--;
1.106 cvs 1287: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1288: /* don't remove server name */
1.106 cvs 1289: newName[length+1] = EOS;
1.53 cvs 1290: }
1.106 cvs 1291: else if (!strncmp (&newName[length+1], ".", 1))
1292: newName[length+1] = EOS;
1.50 cvs 1293: else
1.53 cvs 1294: check = FALSE;
1295: }
1296: /* docname was not comprised inside the URL, so let's */
1297: /* assign the default ressource name */
1.106 cvs 1298: strcpy (docName, "noname.html");
1.53 cvs 1299: }
1300: else
1301: { /* docname is comprised inside the URL */
1.110 cvs 1302: while (length >= 0 && newName[length] != URL_SEP &&
1303: newName[length] != DIR_SEP)
1.53 cvs 1304: length--;
1305: if (length < 0)
1.106 cvs 1306: strcpy (docName, newName);
1.53 cvs 1307: else
1.106 cvs 1308: strcpy (docName, &newName[length+1]);
1.53 cvs 1309: }
1310: }
1311: else
1.106 cvs 1312: docName[0] = EOS;
1.18 cvs 1313: }
1.3 cvs 1314:
1.4 cvs 1315: /*----------------------------------------------------------------------
1.9 cvs 1316: IsSameHost
1.4 cvs 1317: ----------------------------------------------------------------------*/
1.106 cvs 1318: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1319: {
1.106 cvs 1320: char *basename_ptr1, *basename_ptr2;
1321: ThotBool result;
1.3 cvs 1322:
1.106 cvs 1323: basename_ptr1 = AmayaParseUrl (url1, "",
1324: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1325: basename_ptr2 = AmayaParseUrl (url2, "",
1326: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1327:
1.106 cvs 1328: if (strcmp (basename_ptr1, basename_ptr2))
1329: result = FALSE;
1330: else
1331: result = TRUE;
1332: TtaFreeMemory (basename_ptr1);
1333: TtaFreeMemory (basename_ptr2);
1334: return (result);
1.3 cvs 1335: }
1336:
1337:
1.4 cvs 1338: /*----------------------------------------------------------------------
1.22 cvs 1339: HasKnownFileSuffix
1340: returns TRUE if path points to a file ending with a suffix.
1341: ----------------------------------------------------------------------*/
1.106 cvs 1342: ThotBool HasKnownFileSuffix (const char *path)
1343: {
1344: char *root;
1345: char temppath[MAX_LENGTH];
1346: char suffix[MAX_LENGTH];
1.22 cvs 1347:
1.106 cvs 1348: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1349: return (FALSE);
1350:
1.106 cvs 1351: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1352:
1353: if (root)
1354: {
1.106 cvs 1355: strcpy (temppath, root);
1.25 cvs 1356: TtaFreeMemory (root);
1.22 cvs 1357: /* Get the suffix */
1.124 vatton 1358: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1359:
1.106 cvs 1360: if( suffix[0] == EOS)
1.22 cvs 1361: /* no suffix */
1362: return (FALSE);
1363:
1364: /* Normalize the suffix */
1365: ConvertToLowerCase (suffix);
1366:
1.106 cvs 1367: if (!strcmp (suffix, "gz"))
1.22 cvs 1368: /* skip the compressed suffix */
1369: {
1.124 vatton 1370: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1371: if(suffix[0] == EOS)
1.22 cvs 1372: /* no suffix */
1373: return (FALSE);
1374: /* Normalize the suffix */
1375: ConvertToLowerCase (suffix);
1376: }
1377:
1.106 cvs 1378: if (strcmp (suffix, "gif") &&
1379: strcmp (suffix, "xbm") &&
1380: strcmp (suffix, "xpm") &&
1381: strcmp (suffix, "jpg") &&
1382: strcmp (suffix, "pdf") &&
1383: strcmp (suffix, "png") &&
1384: strcmp (suffix, "tgz") &&
1385: strcmp (suffix, "xpg") &&
1386: strcmp (suffix, "xpd") &&
1387: strcmp (suffix, "ps") &&
1388: strcmp (suffix, "au") &&
1389: strcmp (suffix, "html") &&
1390: strcmp (suffix, "htm") &&
1391: strcmp (suffix, "shtml") &&
1392: strcmp (suffix, "xht") &&
1393: strcmp (suffix, "xhtm") &&
1394: strcmp (suffix, "xhtml") &&
1395: strcmp (suffix, "txt") &&
1396: strcmp (suffix, "css") &&
1397: strcmp (suffix, "eps"))
1.22 cvs 1398: return (FALSE);
1399: else
1400: return (TRUE);
1401: }
1402: else
1403: return (FALSE);
1404: }
1405:
1406:
1407: /*----------------------------------------------------------------------
1.24 cvs 1408: ChopURL
1409: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1410: If inputURL is bigger than that size, outputURL receives
1411: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1412: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1413: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1414: copied into outputURL.
1415: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1416: chars.
1417: ----------------------------------------------------------------------*/
1.106 cvs 1418: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1419: {
1420: int len;
1.9 cvs 1421:
1.106 cvs 1422: len = strlen (inputURL);
1.24 cvs 1423: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1424: strcpy (outputURL, inputURL);
1.24 cvs 1425: else
1426: /* make a truncated urlName on the status window */
1427: {
1.106 cvs 1428: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1429: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1430: strcat (outputURL, "...");
1431: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1432: }
1.25 cvs 1433: }
1434:
1435:
1436: /*----------------------------------------------------------------------
1437: scan
1.47 cvs 1438: Scan a filename for its constituents
1.25 cvs 1439: -----------------------------------
1440:
1441: On entry,
1442: name points to a document name which may be incomplete.
1443: On exit,
1444: absolute or relative may be nonzero (but not both).
1445: host, fragment and access may be nonzero if they were specified.
1446: Any which are nonzero point to zero terminated strings.
1447: ----------------------------------------------------------------------*/
1.106 cvs 1448: static void scan (char *name, HTURI *parts)
1.25 cvs 1449: {
1.106 cvs 1450: char * p;
1451: char * after_access = name;
1.32 cvs 1452:
1.43 cvs 1453: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1454: /* Look for fragment identifier */
1.106 cvs 1455: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1456: {
1.106 cvs 1457: *p++ = '\0';
1.28 cvs 1458: parts->fragment = p;
1.25 cvs 1459: }
1460:
1.28 cvs 1461: for (p=name; *p; p++)
1462: {
1.106 cvs 1463: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1464: break;
1.106 cvs 1465: if (*p == ':')
1.28 cvs 1466: {
1467: *p = 0;
1468: parts->access = after_access; /* Scheme has been specified */
1469:
1470: /* The combination of gcc, the "-O" flag and the HP platform is
1471: unhealthy. The following three lines is a quick & dirty fix, but is
1472: not recommended. Rather, turn off "-O". */
1473:
1474: /* after_access = p;*/
1475: /* while (*after_access == 0)*/
1476: /* after_access++;*/
1477: after_access = p+1;
1.106 cvs 1478: if (!strcasecmp("URL", parts->access))
1.28 cvs 1479: /* Ignore IETF's URL: pre-prefix */
1480: parts->access = NULL;
1481: else
1.25 cvs 1482: break;
1483: }
1484: }
1485:
1486: p = after_access;
1.43 cvs 1487: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1488: {
1.43 cvs 1489: if (p[1] == URL_SEP)
1.28 cvs 1490: {
1.25 cvs 1491: parts->host = p+2; /* host has been specified */
1.28 cvs 1492: *p = 0; /* Terminate access */
1493: /* look for end of host name if any */
1.106 cvs 1494: p = strchr (parts->host, URL_SEP);
1.28 cvs 1495: if (p)
1496: {
1.106 cvs 1497: *p = EOS; /* Terminate host */
1.25 cvs 1498: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1499: }
1500: }
1501: else
1502: /* Root found but no host */
1503: parts->absolute = p+1;
1504: }
1505: else
1506: {
1.25 cvs 1507: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1508: }
1.25 cvs 1509: }
1510:
1511:
1512: /*----------------------------------------------------------------------
1.28 cvs 1513: AmayaParseUrl: parse a Name relative to another name
1514:
1515: This returns those parts of a name which are given (and requested)
1516: substituting bits from the related name where necessary.
1.25 cvs 1517:
1.28 cvs 1518: On entry,
1.25 cvs 1519: aName A filename given
1520: relatedName A name relative to which aName is to be parsed. Give
1521: it an empty string if aName is absolute.
1522: wanted A mask for the bits which are wanted.
1523:
1.28 cvs 1524: On exit,
1.25 cvs 1525: returns A pointer to a malloc'd string which MUST BE FREED
1526: ----------------------------------------------------------------------*/
1.106 cvs 1527: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1528: {
1529: char *return_value;
1530: char result[MAX_LENGTH];
1531: char name[MAX_LENGTH];
1532: char rel[MAX_LENGTH];
1533: char *p, *access;
1.29 cvs 1534: HTURI given, related;
1535: int len;
1.106 cvs 1536: char used_sep;
1537: char *used_str;
1.32 cvs 1538:
1.106 cvs 1539: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1540: {
1.106 cvs 1541: used_str = DIR_STR;
1542: used_sep = DIR_SEP;
1.33 cvs 1543: }
1.32 cvs 1544: else
1.33 cvs 1545: {
1.106 cvs 1546: used_str = URL_STR;
1547: used_sep = URL_SEP;
1.33 cvs 1548: }
1.32 cvs 1549:
1.29 cvs 1550: /* Make working copies of input strings to cut up: */
1551: return_value = NULL;
1552: result[0] = 0; /* Clear string */
1.106 cvs 1553: strcpy (name, aName);
1.29 cvs 1554: if (relatedName != NULL)
1.106 cvs 1555: strcpy (rel, relatedName);
1.29 cvs 1556: else
1.106 cvs 1557: relatedName[0] = EOS;
1.29 cvs 1558:
1559: scan (name, &given);
1560: scan (rel, &related);
1561: access = given.access ? given.access : related.access;
1562: if (wanted & AMAYA_PARSE_ACCESS)
1563: if (access)
1564: {
1.106 cvs 1565: strcat (result, access);
1.29 cvs 1566: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1567: strcat (result, ":");
1.29 cvs 1568: }
1569:
1570: if (given.access && related.access)
1571: /* If different, inherit nothing. */
1.106 cvs 1572: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1573: {
1574: related.host = 0;
1575: related.absolute = 0;
1576: related.relative = 0;
1577: related.fragment = 0;
1578: }
1579:
1580: if (wanted & AMAYA_PARSE_HOST)
1581: if(given.host || related.host)
1582: {
1583: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1584: strcat (result, "//");
1585: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1586: }
1587:
1588: if (given.host && related.host)
1589: /* If different hosts, inherit no path. */
1.106 cvs 1590: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1591: {
1592: related.absolute = 0;
1593: related.relative = 0;
1594: related.fragment = 0;
1595: }
1596:
1597: if (wanted & AMAYA_PARSE_PATH)
1598: {
1599: if (given.absolute)
1600: {
1601: /* All is given */
1602: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1603: strcat (result, used_str);
1604: strcat (result, given.absolute);
1.25 cvs 1605: }
1.29 cvs 1606: else if (related.absolute)
1607: {
1608: /* Adopt path not name */
1.106 cvs 1609: strcat (result, used_str);
1610: strcat (result, related.absolute);
1.29 cvs 1611: if (given.relative)
1612: {
1613: /* Search part? */
1.106 cvs 1614: p = strchr (result, '?');
1.29 cvs 1615: if (!p)
1.106 cvs 1616: p=result+strlen(result)-1;
1.33 cvs 1617: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1618: /* Remove filename */
1619: p[1]=0;
1620: /* Add given one */
1.106 cvs 1621: strcat (result, given.relative);
1.25 cvs 1622: }
1623: }
1.29 cvs 1624: else if (given.relative)
1625: /* what we've got */
1.106 cvs 1626: strcat (result, given.relative);
1.29 cvs 1627: else if (related.relative)
1.106 cvs 1628: strcat (result, related.relative);
1.29 cvs 1629: else
1630: /* No inheritance */
1.106 cvs 1631: strcat (result, used_str);
1.25 cvs 1632: }
1.29 cvs 1633:
1634: if (wanted & AMAYA_PARSE_ANCHOR)
1635: if (given.fragment || related.fragment)
1636: {
1637: if (given.absolute && given.fragment)
1638: {
1639: /*Fixes for relURLs...*/
1640: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1641: strcat (result, "#");
1642: strcat (result, given.fragment);
1.29 cvs 1643: }
1644: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1645: strcat (result, "");
1.29 cvs 1646: else
1647: {
1.110 cvs 1648: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1649: strcat (result, "#");
1.110 cvs 1650: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1651: }
1652: }
1.106 cvs 1653: len = strlen (result);
1654: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1655: strcpy (return_value, result);
1.29 cvs 1656: return (return_value); /* exactly the right length */
1.25 cvs 1657: }
1658:
1659: /*----------------------------------------------------------------------
1660: HTCanon
1661: Canonicalizes the URL in the following manner starting from the host
1662: pointer:
1663:
1664: 1) The host name is converted to lowercase
1665: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1666:
1667: Return: OK The position of the current path part of the URL
1668: which might be the old one or a new one.
1669:
1670: ----------------------------------------------------------------------*/
1.106 cvs 1671: static char *HTCanon (char **filename, char *host)
1672: {
1673: char *newname = NULL;
1674: char used_sep;
1675: char *path;
1676: char *strptr;
1677: char *port;
1678: char *access = host-3;
1679:
1680: if (*filename && strchr (*filename, URL_SEP))
1681: used_sep = URL_SEP;
1.33 cvs 1682: else
1.106 cvs 1683: used_sep = DIR_SEP;
1.32 cvs 1684:
1.110 cvs 1685: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1686: access--;
1.110 cvs 1687: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1688: path = host + strlen (host);
1689: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1690: host = strptr;
1.110 cvs 1691: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1692: port = NULL;
1.25 cvs 1693:
1694: strptr = host; /* Convert to lower-case */
1.82 cvs 1695: while (strptr < path)
1.33 cvs 1696: {
1.123 vatton 1697: *strptr = tolower (*strptr);
1.82 cvs 1698: strptr++;
1.33 cvs 1699: }
1.25 cvs 1700:
1701: /* Does the URL contain a full domain name? This also works for a
1702: numerical host name. The domain name is already made lower-case
1703: and without a trailing dot. */
1704: {
1.106 cvs 1705: char *dot = port ? port : path;
1706: if (dot > *filename && *--dot == '.')
1.33 cvs 1707: {
1.106 cvs 1708: char *orig = dot;
1709: char *dest = dot + 1;
1.82 cvs 1710: while ((*orig++ = *dest++));
1711: if (port) port--;
1.33 cvs 1712: path--;
1.25 cvs 1713: }
1714: }
1715: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1716: if (port)
1717: {
1.82 cvs 1718: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1719: {
1720: if (!newname)
1721: {
1.106 cvs 1722: char *orig = port;
1723: char *dest = port + 1;
1.82 cvs 1724: while ((*orig++ = *dest++));
1.33 cvs 1725: }
1726: }
1.106 cvs 1727: else if ((!strncmp (access, "http", 4) &&
1728: (*(port + 1) == '8' &&
1729: *(port+2) == '0' &&
1.82 cvs 1730: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1731: (!strncmp (access, "gopher", 6) &&
1732: (*(port+1) == '7' &&
1733: *(port+2) == '0' &&
1.82 cvs 1734: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1735: (!strncmp (access, "ftp", 3) &&
1736: (*(port+1) == '2' &&
1737: *(port + 2) == '1' &&
1.82 cvs 1738: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1739: if (!newname)
1740: {
1.106 cvs 1741: char *orig = port;
1742: char *dest = port + 3;
1.33 cvs 1743: while((*orig++ = *dest++));
1744: /* Update path position, Henry Minsky */
1745: path -= 3;
1.25 cvs 1746: }
1.33 cvs 1747: }
1748: else if (newname)
1.106 cvs 1749: strncat (newname, port, (int) (path - port));
1.33 cvs 1750: }
1.25 cvs 1751:
1.33 cvs 1752: if (newname)
1753: {
1.106 cvs 1754: char *newpath = newname + strlen (newname);
1755: strcat (newname, path);
1.25 cvs 1756: path = newpath;
1.28 cvs 1757: /* Free old copy */
1758: TtaFreeMemory(*filename);
1.25 cvs 1759: *filename = newname;
1.33 cvs 1760: }
1.25 cvs 1761: return path;
1762: }
1763:
1764:
1765: /*----------------------------------------------------------------------
1.29 cvs 1766: SimplifyUrl: simplify a URI
1.32 cvs 1767: A URI is allowed to contain the sequence xxx/../ which may be
1768: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1769: Simplification helps us recognize duplicate URIs.
1.25 cvs 1770:
1.28 cvs 1771: Thus, /etc/junk/../fred becomes /etc/fred
1772: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1773:
1.28 cvs 1774: but we should NOT change
1775: http://fred.xxx.edu/../..
1.25 cvs 1776:
1777: or ../../albert.html
1778:
1.28 cvs 1779: In order to avoid empty URLs the following URLs become:
1.25 cvs 1780:
1781: /fred/.. becomes /fred/..
1782: /fred/././.. becomes /fred/..
1783: /fred/.././junk/.././ becomes /fred/..
1784:
1.28 cvs 1785: If more than one set of `://' is found (several proxies in cascade) then
1786: only the part after the last `://' is simplified.
1.25 cvs 1787:
1.28 cvs 1788: Returns: A string which might be the old one or a new one.
1.25 cvs 1789: ----------------------------------------------------------------------*/
1.106 cvs 1790: void SimplifyUrl (char **url)
1791: {
1792: char *path;
1793: char *access;
1794: char *newptr;
1795: char *p;
1796: char *orig, *dest, *end;
1.28 cvs 1797:
1.106 cvs 1798: char used_sep;
1.77 cvs 1799: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1800: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1801:
1.28 cvs 1802: if (!url || !*url)
1803: return;
1804:
1.106 cvs 1805: if (strchr (*url, URL_SEP))
1806: used_sep = URL_SEP;
1.32 cvs 1807: else
1.106 cvs 1808: used_sep = DIR_SEP;
1.32 cvs 1809:
1.77 cvs 1810: /* should we simplify double dot? */
1811: path = *url;
1.106 cvs 1812: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1813: ddot_simplify = FALSE;
1814: else
1815: ddot_simplify = TRUE;
1816:
1.28 cvs 1817: /* Find any scheme name */
1.106 cvs 1818: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1819: {
1820: /* Find host name */
1.28 cvs 1821: access = *url;
1.123 vatton 1822: while (access < path && (*access = tolower (*access)))
1.82 cvs 1823: access++;
1.28 cvs 1824: path += 3;
1.106 cvs 1825: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1826: /* For proxies */
1.106 cvs 1827: path = newptr + 3;
1.82 cvs 1828: /* We have a host name */
1.84 cvs 1829: path = HTCanon (url, path);
1.25 cvs 1830: }
1.106 cvs 1831: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1832: path += 2;
1833: else
1834: path = *url;
1.84 cvs 1835: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1836: /* Some URLs start //<foo> */
1837: path += 1;
1.94 cvs 1838: else if (IsFilePath (path))
1839: {
1840: /* doesn't need to do anything more */
1841: return;
1842: }
1.106 cvs 1843: else if (!strncmp (path, "news:", 5))
1.28 cvs 1844: {
1.106 cvs 1845: newptr = strchr (path+5, '@');
1.28 cvs 1846: if (!newptr)
1847: newptr = path + 5;
1848: while (*newptr)
1849: {
1850: /* Make group or host lower case */
1.123 vatton 1851: *newptr = tolower (*newptr);
1.28 cvs 1852: newptr++;
1.25 cvs 1853: }
1.28 cvs 1854: /* Doesn't need to do any more */
1855: return;
1.25 cvs 1856: }
1.130 cheyroul 1857:
1.126 cheyroul 1858:
1.28 cvs 1859: if ((p = path))
1860: {
1.106 cvs 1861: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1862: (end = strchr (path, '#'))))
1863: end = path + strlen (path);
1.28 cvs 1864:
1865: /* Parse string second time to simplify */
1866: p = path;
1867: while (p < end)
1868: {
1.110 cvs 1869: /* if we're pointing to a char, it's safe to reactivate the
1870: ../ convertion */
1.106 cvs 1871: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1872: ddot_simplify = TRUE;
1873:
1.33 cvs 1874: if (*p==used_sep)
1.28 cvs 1875: {
1.106 cvs 1876: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1877: {
1878: orig = p + 1;
1.84 cvs 1879: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1880: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1881: end = orig - 1;
1882: }
1.106 cvs 1883: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1884: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1885: {
1886: newptr = p;
1.52 cvs 1887: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1888: if (*newptr == used_sep)
1889: orig = newptr + 1;
1.28 cvs 1890: else
1.52 cvs 1891: orig = newptr;
1892:
1893: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1894: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1895: end = orig-1;
1896: /* Start again with prev slash */
1897: p = newptr;
1.28 cvs 1898: }
1.33 cvs 1899: else if (*(p+1) == used_sep)
1.28 cvs 1900: {
1.33 cvs 1901: while (*(p+1) == used_sep)
1.28 cvs 1902: {
1903: orig = p;
1904: dest = p + 1;
1905: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1906: end = orig-1;
1907: }
1908: }
1909: else
1.25 cvs 1910: p++;
1.28 cvs 1911: }
1912: else
1913: p++;
1.25 cvs 1914: }
1915: }
1.51 cvs 1916: /*
1917: ** Check for host/../.. kind of things
1918: */
1.106 cvs 1919: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1920: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 1921: *(path+1) = EOS;
1.28 cvs 1922: return;
1923: }
1924:
1925:
1926: /*----------------------------------------------------------------------
1.96 cvs 1927: NormalizeFile normalizes local names.
1.28 cvs 1928: Return TRUE if target and src differ.
1929: ----------------------------------------------------------------------*/
1.106 cvs 1930: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 1931: {
1.110 cvs 1932: #ifndef _WINDOWS
1.106 cvs 1933: char *s;
1.93 cvs 1934: int i;
1.110 cvs 1935: #endif /* !_WINDOWS */
1.82 cvs 1936: ThotBool change;
1.90 cvs 1937: int start_index; /* the first char that we'll copy */
1.28 cvs 1938:
1.54 cvs 1939: change = FALSE;
1.90 cvs 1940: start_index = 0;
1941:
1.106 cvs 1942: if (!src || src[0] == EOS)
1.96 cvs 1943: {
1.106 cvs 1944: target[0] = EOS;
1.96 cvs 1945: return FALSE;
1946: }
1.90 cvs 1947:
1948: /* @@ do I need file: or file:/ here? */
1.106 cvs 1949: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1950: {
1.90 cvs 1951: /* remove the prefix file: */
1952: start_index += 5;
1953:
1954: /* remove the localhost prefix */
1.106 cvs 1955: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 1956: start_index += 11;
1957:
1958: /* remove the first two slashes in / / /path */
1959: while (src[start_index] &&
1.106 cvs 1960: src[start_index] == '/'
1961: && src[start_index + 1] == '/')
1.94 cvs 1962: start_index++;
1963:
1964: #ifdef _WINDOWS
1965: /* remove any extra slash before the drive name */
1.106 cvs 1966: if (src[start_index] == '/'
1967: &&src[start_index+2] == ':')
1.94 cvs 1968: start_index++;
1969: #endif /* _WINDOWS */
1.90 cvs 1970:
1.106 cvs 1971: if (src[start_index] == EOS)
1.90 cvs 1972: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 1973: strcpy (target, DIR_STR);
1.90 cvs 1974: else
1.97 cvs 1975: /* as we're inside a file: URL, we'll apply all the convertions
1976: we know */
1977: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1978:
1979: change = TRUE;
1980: }
1.97 cvs 1981: else if (convertion != AM_CONV_NONE)
1.96 cvs 1982: {
1983: /* we are following a "local" relative link, we do all the
1984: convertions except for the HOME_DIR ~ one */
1.97 cvs 1985: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1986: }
1.90 cvs 1987: #ifndef _WINDOWS
1.106 cvs 1988: else if (src[0] == '~')
1.53 cvs 1989: {
1.96 cvs 1990: /* it must be a URL typed in a text input field */
1991: /* do the HOME_DIR ~ substitution */
1.82 cvs 1992: s = TtaGetEnvString ("HOME");
1.106 cvs 1993: strcpy (target, s);
1.90 cvs 1994: #if 0
1.96 cvs 1995: /* JK: invalidated this part of the code as it's simpler
1996: to add the DIR_SEP whenever we have something to add
1997: to the path rather than adding it systematically */
1.106 cvs 1998: if (src[1] != DIR_SEP)
1999: strcat (target, DIR_STR);
1.90 cvs 2000: #endif
1.106 cvs 2001: i = strlen (target);
2002: strcpy (&target[i], &src[1]);
1.54 cvs 2003: change = TRUE;
1.53 cvs 2004: }
1.90 cvs 2005: #endif /* _WINDOWS */
1.28 cvs 2006: else
1.96 cvs 2007: /* leave it as it is */
1.106 cvs 2008: strcpy (target, src);
1.96 cvs 2009:
1.28 cvs 2010: /* remove /../ and /./ */
1.29 cvs 2011: SimplifyUrl (&target);
1.54 cvs 2012: if (!change)
1.106 cvs 2013: change = strcmp (src, target);
1.28 cvs 2014: return (change);
1.25 cvs 2015: }
2016:
1.28 cvs 2017:
1.25 cvs 2018: /*----------------------------------------------------------------------
1.31 cvs 2019: MakeRelativeURL: make relative name
1.25 cvs 2020:
1.28 cvs 2021: This function creates and returns a string which gives an expression of
2022: one address as related to another. Where there is no relation, an absolute
2023: address is retured.
1.25 cvs 2024:
1.28 cvs 2025: On entry,
1.25 cvs 2026: Both names must be absolute, fully qualified names of nodes
2027: (no fragment bits)
2028:
1.28 cvs 2029: On exit,
1.25 cvs 2030: The return result points to a newly allocated name which, if
2031: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2032: The caller is responsible for freeing the resulting name later.
2033: ----------------------------------------------------------------------*/
1.106 cvs 2034: char *MakeRelativeURL (char *aName, char *relatedName)
2035: {
2036: char *return_value;
2037: char result[MAX_LENGTH];
2038: char *p;
2039: char *q;
2040: char *after_access;
2041: char *last_slash = NULL;
2042: int slashes, levels, len;
1.110 cvs 2043: #ifdef _WINDOWS
1.44 cvs 2044: int ndx;
1.110 cvs 2045: #endif /* _WINDOWS */
1.44 cvs 2046:
1.29 cvs 2047: if (aName == NULL || relatedName == NULL)
2048: return (NULL);
2049:
2050: slashes = 0;
2051: after_access = NULL;
2052: p = aName;
2053: q = relatedName;
2054: for (; *p && (*p == *q); p++, q++)
1.27 cvs 2055: {
2056: /* Find extent of match */
1.106 cvs 2057: if (*p == ':')
1.29 cvs 2058: after_access = p + 1;
1.28 cvs 2059: if (*p == DIR_SEP)
1.27 cvs 2060: {
1.29 cvs 2061: /* memorize the last slash position and count them */
1.27 cvs 2062: last_slash = p;
2063: slashes++;
1.25 cvs 2064: }
2065: }
2066:
1.31 cvs 2067: /* q, p point to the first non-matching character or zero */
1.106 cvs 2068: if (*q == EOS)
1.31 cvs 2069: {
2070: /* New name is a subset of the related name */
2071: /* exactly the right length */
1.106 cvs 2072: len = strlen (p);
2073: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2074: strcpy (return_value, p);
1.31 cvs 2075: }
2076: else if ((slashes < 2 && after_access == NULL)
2077: || (slashes < 3 && after_access != NULL))
2078: {
2079: /* Two names whitout common path */
2080: /* exactly the right length */
1.106 cvs 2081: len = strlen (aName);
2082: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2083: strcpy (return_value, aName);
1.31 cvs 2084: }
2085: else
2086: {
2087: /* Some path in common */
1.106 cvs 2088: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2089: /* just the same server */
1.106 cvs 2090: strcpy (result, last_slash);
1.31 cvs 2091: else
2092: {
2093: levels= 0;
1.106 cvs 2094: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2095: if (*q == DIR_SEP)
2096: levels++;
2097:
1.106 cvs 2098: result[0] = EOS;
1.31 cvs 2099: for (;levels; levels--)
1.106 cvs 2100: strcat (result, "../");
2101: strcat (result, last_slash+1);
1.31 cvs 2102: }
1.52 cvs 2103:
2104: if (!*result)
1.106 cvs 2105: strcat (result, "./");
1.52 cvs 2106:
1.31 cvs 2107: /* exactly the right length */
1.106 cvs 2108: len = strlen (result);
2109: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2110: strcpy (return_value, result);
1.52 cvs 2111:
1.25 cvs 2112: }
1.110 cvs 2113: #ifdef _WINDOWS
1.106 cvs 2114: len = strlen (return_value);
1.44 cvs 2115: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2116: if (return_value[ndx] == '\\')
2117: return_value[ndx] = '/' ;
1.110 cvs 2118: #endif /* _WINDOWS */
1.29 cvs 2119: return (return_value);
1.24 cvs 2120: }
1.35 cvs 2121:
1.104 kahan 2122: /*----------------------------------------------------------------------
2123: AM_GetFileSize
2124: Returns TRUE and the filesize in the 2nd parameter.
2125: Otherwise, in case of a system error, returns FALSE, with a
2126: filesize of 0L.
2127: ---------------------------------------------------------------------*/
1.106 cvs 2128: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2129: {
1.106 cvs 2130: ThotFileHandle handle = ThotFile_BADHANDLE;
2131: ThotFileInfo info;
1.35 cvs 2132:
1.104 kahan 2133: *file_size = 0L;
2134: if (!TtaFileExist (filename))
2135: return FALSE;
2136:
2137: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2138: if (handle == ThotFile_BADHANDLE)
2139: /* ThotFile_BADHANDLE */
2140: return FALSE;
2141: if (TtaFileStat (handle, &info) == 0)
2142: /* bad stat */
2143: info.size = 0L;
2144: TtaFileClose (handle);
2145: *file_size = (unsigned long) info.size;
2146: return TRUE;
2147: }
Webmaster