Annotation of Amaya/amaya/AHTURLTools.c, revision 1.127
1.7 cvs 1: /*
2: *
1.110 cvs 3: * (c) COPYRIGHT MIT and INRIA, 1996-2001
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: /*----------------------------------------------------------------------
1.103 cvs 526: IsMathMLName
527: returns TRUE if path points to an MathML resource.
528: ----------------------------------------------------------------------*/
1.111 cvs 529: ThotBool IsMathMLName (const char *path)
1.103 cvs 530: {
1.106 cvs 531: char temppath[MAX_LENGTH];
532: char suffix[MAX_LENGTH];
1.103 cvs 533:
534: if (!path)
535: return (FALSE);
536:
1.106 cvs 537: strcpy (temppath, path);
1.124 vatton 538: TtaExtractSuffix (temppath, suffix);
1.103 cvs 539:
1.106 cvs 540: if (!strcasecmp (suffix, "mml"))
1.103 cvs 541: return (TRUE);
1.106 cvs 542: else if (!strcmp (suffix, "gz"))
1.103 cvs 543: {
544: /* take into account compressed files */
1.124 vatton 545: TtaExtractSuffix (temppath, suffix);
1.106 cvs 546: if (!strcasecmp (suffix, "mml"))
1.103 cvs 547: return (TRUE);
548: else
549: return (FALSE);
550: }
551: else
552: return (FALSE);
553: }
554:
555: /*----------------------------------------------------------------------
556: IsSVGName
1.119 kahan 557: returns TRUE if path points to an SVG resource.
1.103 cvs 558: ----------------------------------------------------------------------*/
1.111 cvs 559: ThotBool IsSVGName (const char *path)
1.103 cvs 560: {
1.106 cvs 561: char temppath[MAX_LENGTH];
562: char suffix[MAX_LENGTH];
1.103 cvs 563:
564: if (!path)
565: return (FALSE);
566:
1.106 cvs 567: strcpy (temppath, path);
1.124 vatton 568: TtaExtractSuffix (temppath, suffix);
1.103 cvs 569:
1.106 cvs 570: if (!strcasecmp (suffix, "svg"))
1.103 cvs 571: return (TRUE);
1.106 cvs 572: else if (!strcmp (suffix, "gz"))
1.103 cvs 573: {
574: /* take into account compressed files */
1.124 vatton 575: TtaExtractSuffix (temppath, suffix);
1.106 cvs 576: if (!strcasecmp (suffix, "svg"))
1.103 cvs 577: return (TRUE);
578: else
579: return (FALSE);
580: }
581: else
582: return (FALSE);
583: }
584:
585: /*----------------------------------------------------------------------
1.60 cvs 586: IsCSSName
587: returns TRUE if path points to an XML resource.
588: ----------------------------------------------------------------------*/
1.111 cvs 589: ThotBool IsCSSName (const char *path)
1.60 cvs 590: {
1.106 cvs 591: char temppath[MAX_LENGTH];
592: char suffix[MAX_LENGTH];
1.60 cvs 593:
594: if (!path)
595: return (FALSE);
596:
1.106 cvs 597: strcpy (temppath, path);
1.124 vatton 598: TtaExtractSuffix (temppath, suffix);
1.60 cvs 599:
1.106 cvs 600: if (!strcasecmp (suffix, "css"))
1.60 cvs 601: return (TRUE);
1.106 cvs 602: else if (!strcmp (suffix, "gz"))
1.60 cvs 603: {
604: /* take into account compressed files */
1.124 vatton 605: TtaExtractSuffix (temppath, suffix);
1.106 cvs 606: if (!strcasecmp (suffix, "css"))
1.56 cvs 607: return (TRUE);
608: else
609: return (FALSE);
610: }
611: else
612: return (FALSE);
613: }
614:
615: /*----------------------------------------------------------------------
1.9 cvs 616: IsImageName
617: returns TRUE if path points to an image resource.
1.4 cvs 618: ----------------------------------------------------------------------*/
1.111 cvs 619: ThotBool IsImageName (const char *path)
1.106 cvs 620: {
621: char temppath[MAX_LENGTH];
622: char suffix[MAX_LENGTH];
623: char nsuffix[MAX_LENGTH];
1.5 cvs 624: int i;
625:
626: if (!path)
1.13 cvs 627: return (FALSE);
1.5 cvs 628:
1.106 cvs 629: strcpy (temppath, path);
1.124 vatton 630: TtaExtractSuffix (temppath, suffix);
1.5 cvs 631:
632: /* Normalize the suffix */
633: i = 0;
1.106 cvs 634: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 635: {
1.123 vatton 636: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 637: i++;
638: }
1.106 cvs 639: nsuffix[i] = EOS;
640: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
641: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
642: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 643: return (TRUE);
644: return (FALSE);
1.3 cvs 645: }
646:
1.4 cvs 647: /*----------------------------------------------------------------------
1.58 cvs 648: IsImageType
649: returns TRUE if type points to an image resource.
650: ----------------------------------------------------------------------*/
1.111 cvs 651: ThotBool IsImageType (const char *type)
1.58 cvs 652: {
1.106 cvs 653: char temptype[MAX_LENGTH];
1.58 cvs 654: int i;
655:
656: if (!type)
657: return (FALSE);
658:
1.106 cvs 659: strcpy (temptype, type);
1.58 cvs 660: /* Normalize the type */
661: i = 0;
1.106 cvs 662: while (temptype[i] != EOS)
1.58 cvs 663: {
664: temptype[i] = tolower (temptype[i]);
665: i++;
666: }
1.111 cvs 667: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
668: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
669: !strcmp (temptype, "png"))
1.58 cvs 670: return (TRUE);
671: return (FALSE);
672: }
673:
674: /*----------------------------------------------------------------------
1.9 cvs 675: IsTextName
1.4 cvs 676: ----------------------------------------------------------------------*/
1.111 cvs 677: ThotBool IsTextName (const char *path)
1.106 cvs 678: {
679: char temppath[MAX_LENGTH];
680: char suffix[MAX_LENGTH];
681: char nsuffix[MAX_LENGTH];
1.5 cvs 682: int i;
683:
684: if (!path)
1.13 cvs 685: return (FALSE);
1.5 cvs 686:
1.106 cvs 687: strcpy (temppath, path);
1.124 vatton 688: TtaExtractSuffix (temppath, suffix);
1.5 cvs 689:
690: /* Normalize the suffix */
691: i = 0;
1.106 cvs 692: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 693: {
1.25 cvs 694: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 695: i++;
696: }
1.106 cvs 697: nsuffix[i] = EOS;
1.5 cvs 698:
1.111 cvs 699: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 700: return (TRUE);
1.106 cvs 701: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 702: {
1.39 cvs 703: /* take into account compressed files */
1.124 vatton 704: TtaExtractSuffix (temppath, suffix);
1.13 cvs 705: /* Normalize the suffix */
706: i = 0;
1.106 cvs 707: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 708: {
1.25 cvs 709: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 710: i++;
711: }
1.106 cvs 712: nsuffix[i] = EOS;
1.111 cvs 713: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 714: return (TRUE);
715: else
716: return (FALSE);
717: }
718: else
719: return (FALSE);
1.3 cvs 720: }
721:
1.4 cvs 722: /*----------------------------------------------------------------------
1.9 cvs 723: IsHTTPPath
724: returns TRUE if path is in fact an http URL.
1.4 cvs 725: ----------------------------------------------------------------------*/
1.112 cvs 726: ThotBool IsHTTPPath (const char *path)
1.3 cvs 727: {
1.5 cvs 728: if (!path)
729: return FALSE;
1.3 cvs 730:
1.106 cvs 731: if ((!strncmp (path, "http:", 5) != 0)
732: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
733: || !strncmp (path, "internal:", 9))
1.58 cvs 734: return TRUE;
735: return FALSE;
1.3 cvs 736: }
737:
1.4 cvs 738: /*----------------------------------------------------------------------
1.9 cvs 739: IsWithParameters
740: returns TRUE if url has a concatenated query string.
1.4 cvs 741: ----------------------------------------------------------------------*/
1.66 cvs 742: ThotBool IsWithParameters (const char *url)
1.3 cvs 743: {
1.5 cvs 744: int i;
1.3 cvs 745:
1.9 cvs 746: if ((!url) || (url[0] == EOS))
1.5 cvs 747: return FALSE;
1.3 cvs 748:
1.9 cvs 749: i = strlen (url) - 1;
750: while (i > 0 && url[i--] != '?')
1.5 cvs 751: if (i < 0)
752: return FALSE;
1.3 cvs 753:
1.5 cvs 754: /* There is a parameter */
755: return TRUE;
1.3 cvs 756: }
757:
1.4 cvs 758: /*----------------------------------------------------------------------
1.9 cvs 759: IsW3Path
760: returns TRUE if path is in fact a URL.
1.4 cvs 761: ----------------------------------------------------------------------*/
1.106 cvs 762: ThotBool IsW3Path (const char *path)
763: {
764: if (strncmp (path, "http:", 5) &&
765: strncmp (path, "ftp:", 4) &&
766: strncmp (path, "telnet:", 7) &&
767: strncmp (path, "wais:", 5) &&
768: strncmp (path, "news:", 5) &&
769: strncmp (path, "gopher:", 7) &&
770: strncmp (path, "mailto:", 7) &&
771: strncmp (path, "archie:", 7))
1.72 cvs 772: return FALSE;
773: return TRUE;
1.3 cvs 774: }
775:
1.4 cvs 776: /*----------------------------------------------------------------------
1.90 cvs 777: IsFilePath
778: returns TRUE if path is in fact a URL.
779: ----------------------------------------------------------------------*/
1.106 cvs 780: ThotBool IsFilePath (const char *path)
1.90 cvs 781: {
1.106 cvs 782: if (strncmp (path, "file:", 5))
1.90 cvs 783: return FALSE;
784: return TRUE;
785: }
786:
787: /*----------------------------------------------------------------------
1.9 cvs 788: IsValidProtocol
789: returns true if the url protocol is supported by Amaya.
1.4 cvs 790: ----------------------------------------------------------------------*/
1.106 cvs 791: ThotBool IsValidProtocol (const char *url)
792: {
793: if (!strncmp (url, "http:", 5)
794: || !strncmp (url, "internal:", 9)
795: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 796: /* experimental */
1.24 cvs 797: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 798: return (TRUE);
1.5 cvs 799: else
1.8 cvs 800: return (FALSE);
1.3 cvs 801: }
802:
1.31 cvs 803:
804: /*----------------------------------------------------------------------
805: GetBaseURL
806: normalizes orgName according to a base associated with doc, and
807: following the standard URL format rules.
808: The function returns the base used to solve relative URL and SRC:
809: - the base of the document,
810: - or the document path (without document name).
811: ----------------------------------------------------------------------*/
1.106 cvs 812: char *GetBaseURL (Document doc)
1.31 cvs 813: {
814: Element el;
815: ElementType elType;
816: AttributeType attrType;
817: Attribute attr;
1.106 cvs 818: char *ptr, *basename;
1.31 cvs 819: int length;
820:
1.113 cvs 821: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 822: return NULL;
1.106 cvs 823: basename = TtaGetMemory (MAX_LENGTH);
824: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
825: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 826: length = MAX_LENGTH -1;
1.113 cvs 827: /* is it a HTML document ? */
1.31 cvs 828: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.113 cvs 829: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
830: /* it's a HTML document */
1.65 cvs 831: {
1.113 cvs 832: /* get the document element */
833: el = TtaGetMainRoot (doc);
834: /* search the BASE element */
835: elType.ElTypeNum = HTML_EL_HEAD;
836: el = TtaSearchTypedElement (elType, SearchForward, el);
837: if (el)
838: /* there is a HEAD element */
839: {
840: /* look for a BASE element within the HEAD */
841: elType.ElTypeNum = HTML_EL_BASE;
842: el = TtaSearchTypedElement (elType, SearchInTree, el);
843: }
844: if (el)
1.31 cvs 845: {
1.113 cvs 846: /* The document has a BASE element. Get the HREF attribute of the
847: BASE element */
848: attrType.AttrSSchema = elType.ElSSchema;
849: attrType.AttrTypeNum = HTML_ATTR_HREF_;
850: attr = TtaGetAttribute (el, attrType);
851: if (attr)
1.31 cvs 852: {
1.113 cvs 853: /* Use the base path of the document */
854: TtaGiveTextAttributeValue (attr, basename, &length);
855: /* base and orgName have to be separated by a DIR_SEP */
856: length--;
857: if (basename[0] != EOS && basename[length] != URL_SEP &&
858: basename[length] != DIR_SEP)
859: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 860: {
1.113 cvs 861: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
862: AMAYA_PARSE_HOST |
863: AMAYA_PARSE_PUNCTUATION);
864: if (ptr && !strcmp (ptr, basename))
865: {
866: /* it has this form, complete it by adding a URL_STR */
867: if (strchr (basename, DIR_SEP))
868: strcat (basename, DIR_STR);
869: else
870: strcat (basename, URL_STR);
871: length++;
872: }
873: if (ptr)
874: TtaFreeMemory (ptr);
1.31 cvs 875: }
876: }
877: }
1.113 cvs 878: }
879:
1.31 cvs 880: /* Remove anything after the last DIR_SEP char. If no such char is found,
881: * then search for the first ":" char, hoping that what's before that is a
882: * protocol. If found, end the string there. If neither char is found,
883: * then discard the whole base element.
884: */
1.106 cvs 885: length = strlen (basename) - 1;
1.31 cvs 886: /* search for the last DIR_SEP char */
1.106 cvs 887: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 888: length--;
889: if (length >= 0)
890: /* found the last DIR_SEP char, end the string there */
1.106 cvs 891: basename[length + 1] = EOS;
1.31 cvs 892: else
893: /* search for the first PATH_STR char */
894: {
1.106 cvs 895: for (length = 0; basename[length] != ':' &&
896: basename[length] != EOS; length ++);
897: if (basename[length] == ':')
1.31 cvs 898: /* found, so end the string there */
1.106 cvs 899: basename[length + 1] = EOS;
1.31 cvs 900: else
901: /* not found, discard the base */
1.106 cvs 902: basename[0] = EOS;
1.31 cvs 903: }
904: return (basename);
905: }
906:
907:
1.4 cvs 908: /*----------------------------------------------------------------------
1.40 cvs 909: GetLocalPath
910: Allocate and return the local document path associated to the url
911: ----------------------------------------------------------------------*/
1.106 cvs 912: char *GetLocalPath (Document doc, char *url)
913: {
914: char *ptr;
915: char *n;
916: char *documentname;
917: char url_sep;
1.83 cvs 918: int len;
1.67 cvs 919: ThotBool noFile;
1.40 cvs 920:
921: if (url != NULL)
922: {
923: /* check whether the file name exists */
1.106 cvs 924: len = strlen (url) - 1;
1.71 cvs 925: if (IsW3Path (url))
1.106 cvs 926: url_sep = '/';
1.41 cvs 927: else
1.106 cvs 928: url_sep = DIR_SEP;
1.41 cvs 929: noFile = (url[len] == url_sep);
1.40 cvs 930: if (noFile)
1.106 cvs 931: url[len] = EOS;
932: ptr = TtaGetMemory (MAX_LENGTH);
933: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 934: TtaExtractName (url, ptr, documentname);
1.106 cvs 935: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 936: if (!TtaCheckDirectory (ptr))
937: /* directory did not exist */
1.72 cvs 938: TtaMakeDirectory (ptr);
1.47 cvs 939:
940: /* don't include the query string within document name */
1.106 cvs 941: n = strrchr (documentname, '?');
1.47 cvs 942: if (n != NULL)
1.106 cvs 943: *n = EOS;
1.46 cvs 944: /* don't include ':' within document name */
1.106 cvs 945: n = strchr (documentname, ':');
1.46 cvs 946: if (n != NULL)
1.106 cvs 947: *n = EOS;
1.69 cvs 948: /* if after all this operations document name
949: is empty, let's use noname.html instead */
1.106 cvs 950: if (documentname[0] == EOS)
951: strcat (ptr, "noname.html");
1.69 cvs 952: else
1.106 cvs 953: strcat (ptr, documentname);
1.40 cvs 954: TtaFreeMemory (documentname);
955: /* restore the url */
956: if (noFile)
1.41 cvs 957: url[len] = url_sep;
1.40 cvs 958: return (ptr);
959: }
960: else
961: return (NULL);
962: }
963:
1.73 cvs 964: /*----------------------------------------------------------------------
1.79 cvs 965: ExtractTarget extract the target name from document nane.
966: ----------------------------------------------------------------------*/
1.106 cvs 967: void ExtractTarget (char *aName, char *target)
1.79 cvs 968: {
1.106 cvs 969: int lg, i;
970: char *ptr;
971: char *oldptr;
1.79 cvs 972:
973: if (!target || !aName)
974: /* bad target */
975: return;
976:
1.106 cvs 977: target[0] = EOS;
978: lg = strlen (aName);
1.79 cvs 979: if (lg)
980: {
981: /* the name is not empty */
982: oldptr = ptr = &aName[0];
983: do
984: {
1.106 cvs 985: ptr = strrchr (oldptr, '#');
1.79 cvs 986: if (ptr)
987: oldptr = &ptr[1];
988: }
989: while (ptr);
990:
991: i = (int) (oldptr) - (int) (aName); /* name length */
992: if (i > 1)
993: {
1.106 cvs 994: aName[i - 1] = EOS;
1.79 cvs 995: if (i != lg)
1.106 cvs 996: strcpy (target, oldptr);
1.79 cvs 997: }
998: }
999: }
1000:
1001: /*----------------------------------------------------------------------
1.90 cvs 1002: RemoveNewLines (text)
1003: Removes any '\n' chars that are found in text.
1004: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1005: ----------------------------------------------------------------------*/
1.106 cvs 1006: ThotBool RemoveNewLines (char *text)
1007: {
1008: ThotBool change = FALSE;
1009: char *src;
1010: char *dest;
1.90 cvs 1011:
1012: src = text;
1013: dest = text;
1.115 kahan 1014:
1015: /* remove any preceding whitespace */
1016: while (*src && *src == ' ')
1017: {
1018: src++;
1019: change = 1;
1020: }
1021:
1.90 cvs 1022: while (*src)
1023: {
1024: switch (*src)
1025: {
1.106 cvs 1026: case '\n':
1.90 cvs 1027: /* don't copy the newline */
1028: change = 1;
1029: break;
1030: default:
1031: *dest = *src;
1032: dest++;
1033: break;
1034: }
1035: src++;
1036: }
1037: /* copy the last EOS char */
1038: *dest = *src;
1039:
1040: return (change);
1041: }
1042:
1043: /*----------------------------------------------------------------------
1044: CleanCopyFileURL
1045: Copies a file url from a src string to destination string.
1.97 cvs 1046: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1047: we want to do).
1.90 cvs 1048: ----------------------------------------------------------------------*/
1.106 cvs 1049: static void CleanCopyFileURL (char *dest, char *src,
1050: ConvertionType convertion)
1.90 cvs 1051: {
1052: while (*src)
1.89 cvs 1053: {
1.90 cvs 1054: switch (*src)
1.89 cvs 1055: {
1056: #ifdef _WINDOWS
1.106 cvs 1057: case URL_SEP:
1.96 cvs 1058: /* make DIR_SEP transformation */
1.97 cvs 1059: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1060: *dest = DIR_SEP;
1.96 cvs 1061: else
1062: *dest = *src;
1.90 cvs 1063: dest++;
1.96 cvs 1064: src++;
1.90 cvs 1065: break;
1.89 cvs 1066: #endif /* _WINDOWS */
1.96 cvs 1067:
1.106 cvs 1068: case '%':
1.97 cvs 1069: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1070: {
1.97 cvs 1071: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1072: src++;
1.106 cvs 1073: if (*src != EOS)
1.97 cvs 1074: {
1075: *dest = UnEscapeChar (*src) * 16;
1076: src++;
1077: }
1.106 cvs 1078: if (*src != EOS)
1.97 cvs 1079: {
1080: *dest = *dest + UnEscapeChar (*src);
1081: src++;
1082: }
1083: dest++;
1.96 cvs 1084: }
1.97 cvs 1085: else
1.96 cvs 1086: {
1.97 cvs 1087: *dest = *src;
1088: dest++;
1.96 cvs 1089: src++;
1090: }
1091: break;
1092:
1.90 cvs 1093: default:
1094: *dest = *src;
1.89 cvs 1095: dest++;
1.96 cvs 1096: src++;
1.90 cvs 1097: break;
1.89 cvs 1098: }
1099: }
1.90 cvs 1100: /* copy the EOS char */
1101: *dest = *src;
1.73 cvs 1102: }
1.40 cvs 1103:
1104: /*----------------------------------------------------------------------
1.9 cvs 1105: NormalizeURL
1106: normalizes orgName according to a base associated with doc, and
1107: following the standard URL format rules.
1.113 cvs 1108: if doc is < 0, use as a base the URL of the document that contains
1109: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1110: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1111: other path.
1.9 cvs 1112: The function returns the new complete and normalized URL
1.12 cvs 1113: or file name path (newName) and the name of the document (docName).
1.9 cvs 1114: N.B. If the function can't find out what's the docName, it assigns
1115: the name "noname.html".
1.4 cvs 1116: ----------------------------------------------------------------------*/
1.106 cvs 1117: void NormalizeURL (char *orgName, Document doc, char *newName,
1118: char *docName, char *otherPath)
1119: {
1120: char *basename;
1121: char tempOrgName[MAX_LENGTH];
1122: char *ptr;
1123: char used_sep;
1.84 cvs 1124: int length;
1125: ThotBool check;
1.5 cvs 1126:
1.110 cvs 1127: #ifdef _WINDOWS
1.44 cvs 1128: int ndx;
1.110 cvs 1129: #endif /* _WINDOWS */
1.44 cvs 1130:
1.5 cvs 1131: if (!newName || !docName)
1132: return;
1.18 cvs 1133:
1.113 cvs 1134: if (doc < 0)
1135: basename = TtaStrdup (SavedDocumentURL);
1136: else if (doc > 0)
1.53 cvs 1137: basename = GetBaseURL (doc);
1138: else if (otherPath != NULL)
1.108 cvs 1139: basename = TtaStrdup (otherPath);
1.32 cvs 1140: else
1.53 cvs 1141: basename = NULL;
1.32 cvs 1142:
1.18 cvs 1143: /*
1.31 cvs 1144: * Clean orgName
1145: * Make sure we have a complete orgName, without any leading or trailing
1146: * white spaces, or trailinbg new lines
1147: */
1.5 cvs 1148: ptr = orgName;
1.18 cvs 1149: /* skip leading white space and new line characters */
1.106 cvs 1150: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1151: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1152: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1153: /*
1.31 cvs 1154: * Make orgName a complete URL
1155: * If the URL does not include a protocol, then try to calculate
1156: * one using the doc's base element (if it exists),
1157: */
1.106 cvs 1158: if (tempOrgName[0] == EOS)
1.53 cvs 1159: {
1.106 cvs 1160: newName[0] = EOS;
1161: docName[0] = EOS;
1.53 cvs 1162: TtaFreeMemory (basename);
1163: return;
1164: }
1.49 cvs 1165:
1166: /* clean trailing white space */
1.106 cvs 1167: length = strlen (tempOrgName) - 1;
1168: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1169: {
1.106 cvs 1170: tempOrgName[length] = EOS;
1.53 cvs 1171: length--;
1172: }
1.50 cvs 1173:
1.55 cvs 1174: /* remove extra dot (which dot???) */
1175: /* ugly, but faster than a strcmp */
1.106 cvs 1176: if (tempOrgName[length] == '.'
1177: && (length == 0 || tempOrgName[length-1] != '.'))
1178: tempOrgName[length] = EOS;
1.50 cvs 1179:
1.94 cvs 1180: if (IsW3Path (tempOrgName))
1.53 cvs 1181: {
1182: /* the name is complete, go to the Sixth Step */
1.106 cvs 1183: strcpy (newName, tempOrgName);
1.53 cvs 1184: SimplifyUrl (&newName);
1185: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1186: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1187: AMAYA_PARSE_HOST |
1188: AMAYA_PARSE_PUNCTUATION);
1189: if (ptr && !strcmp (ptr, newName))
1190: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1191: strcat (newName, URL_STR);
1.49 cvs 1192:
1.53 cvs 1193: if (ptr)
1.50 cvs 1194: TtaFreeMemory (ptr);
1.53 cvs 1195: }
1.113 cvs 1196: else if (basename == NULL)
1.53 cvs 1197: /* the name is complete, go to the Sixth Step */
1.106 cvs 1198: strcpy (newName, tempOrgName);
1.53 cvs 1199: else
1200: {
1.31 cvs 1201: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1202: #ifdef _WINDOWS
1.53 cvs 1203: if (!IsW3Path (basename))
1204: {
1.106 cvs 1205: length = strlen (tempOrgName);
1.53 cvs 1206: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1207: if (tempOrgName [ndx] == '/')
1208: tempOrgName [ndx] = '\\';
1.53 cvs 1209: }
1.110 cvs 1210: #endif /* _WINDOWS */
1.25 cvs 1211: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1212: if (ptr)
1213: {
1214: SimplifyUrl (&ptr);
1.106 cvs 1215: strcpy (newName, ptr);
1.53 cvs 1216: TtaFreeMemory (ptr);
1217: }
1218: else
1.106 cvs 1219: newName[0] = EOS;
1.53 cvs 1220: }
1.36 cvs 1221:
1222: TtaFreeMemory (basename);
1.18 cvs 1223: /*
1.31 cvs 1224: * Prepare the docname that will refer to this ressource in the
1225: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1226: * noname.html as a default ressource name
1.18 cvs 1227: */
1.106 cvs 1228: if (newName[0] != EOS)
1.53 cvs 1229: {
1.106 cvs 1230: length = strlen (newName) - 1;
1231: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1232: {
1233: used_sep = newName[length];
1234: check = TRUE;
1235: while (check)
1236: {
1.50 cvs 1237: length--;
1238: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1239: length--;
1.106 cvs 1240: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1241: {
1.106 cvs 1242: newName[length+1] = EOS;
1.53 cvs 1243: /* remove also previous directory */
1244: length--;
1245: while (length >= 0 && newName[length] != used_sep)
1246: length--;
1.106 cvs 1247: if (strncmp (&newName[length+1], "//", 2))
1.53 cvs 1248: /* don't remove server name */
1.106 cvs 1249: newName[length+1] = EOS;
1.53 cvs 1250: }
1.106 cvs 1251: else if (!strncmp (&newName[length+1], ".", 1))
1252: newName[length+1] = EOS;
1.50 cvs 1253: else
1.53 cvs 1254: check = FALSE;
1255: }
1256: /* docname was not comprised inside the URL, so let's */
1257: /* assign the default ressource name */
1.106 cvs 1258: strcpy (docName, "noname.html");
1.53 cvs 1259: }
1260: else
1261: { /* docname is comprised inside the URL */
1.110 cvs 1262: while (length >= 0 && newName[length] != URL_SEP &&
1263: newName[length] != DIR_SEP)
1.53 cvs 1264: length--;
1265: if (length < 0)
1.106 cvs 1266: strcpy (docName, newName);
1.53 cvs 1267: else
1.106 cvs 1268: strcpy (docName, &newName[length+1]);
1.53 cvs 1269: }
1270: }
1271: else
1.106 cvs 1272: docName[0] = EOS;
1.18 cvs 1273: }
1.3 cvs 1274:
1.4 cvs 1275: /*----------------------------------------------------------------------
1.9 cvs 1276: IsSameHost
1.4 cvs 1277: ----------------------------------------------------------------------*/
1.106 cvs 1278: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1279: {
1.106 cvs 1280: char *basename_ptr1, *basename_ptr2;
1281: ThotBool result;
1.3 cvs 1282:
1.106 cvs 1283: basename_ptr1 = AmayaParseUrl (url1, "",
1284: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1285: basename_ptr2 = AmayaParseUrl (url2, "",
1286: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1287:
1.106 cvs 1288: if (strcmp (basename_ptr1, basename_ptr2))
1289: result = FALSE;
1290: else
1291: result = TRUE;
1292: TtaFreeMemory (basename_ptr1);
1293: TtaFreeMemory (basename_ptr2);
1294: return (result);
1.3 cvs 1295: }
1296:
1297:
1.4 cvs 1298: /*----------------------------------------------------------------------
1.22 cvs 1299: HasKnownFileSuffix
1300: returns TRUE if path points to a file ending with a suffix.
1301: ----------------------------------------------------------------------*/
1.106 cvs 1302: ThotBool HasKnownFileSuffix (const char *path)
1303: {
1304: char *root;
1305: char temppath[MAX_LENGTH];
1306: char suffix[MAX_LENGTH];
1.22 cvs 1307:
1.106 cvs 1308: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1309: return (FALSE);
1310:
1.106 cvs 1311: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1312:
1313: if (root)
1314: {
1.106 cvs 1315: strcpy (temppath, root);
1.25 cvs 1316: TtaFreeMemory (root);
1.22 cvs 1317: /* Get the suffix */
1.124 vatton 1318: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1319:
1.106 cvs 1320: if( suffix[0] == EOS)
1.22 cvs 1321: /* no suffix */
1322: return (FALSE);
1323:
1324: /* Normalize the suffix */
1325: ConvertToLowerCase (suffix);
1326:
1.106 cvs 1327: if (!strcmp (suffix, "gz"))
1.22 cvs 1328: /* skip the compressed suffix */
1329: {
1.124 vatton 1330: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1331: if(suffix[0] == EOS)
1.22 cvs 1332: /* no suffix */
1333: return (FALSE);
1334: /* Normalize the suffix */
1335: ConvertToLowerCase (suffix);
1336: }
1337:
1.106 cvs 1338: if (strcmp (suffix, "gif") &&
1339: strcmp (suffix, "xbm") &&
1340: strcmp (suffix, "xpm") &&
1341: strcmp (suffix, "jpg") &&
1342: strcmp (suffix, "pdf") &&
1343: strcmp (suffix, "png") &&
1344: strcmp (suffix, "tgz") &&
1345: strcmp (suffix, "xpg") &&
1346: strcmp (suffix, "xpd") &&
1347: strcmp (suffix, "ps") &&
1348: strcmp (suffix, "au") &&
1349: strcmp (suffix, "html") &&
1350: strcmp (suffix, "htm") &&
1351: strcmp (suffix, "shtml") &&
1352: strcmp (suffix, "xht") &&
1353: strcmp (suffix, "xhtm") &&
1354: strcmp (suffix, "xhtml") &&
1355: strcmp (suffix, "txt") &&
1356: strcmp (suffix, "css") &&
1357: strcmp (suffix, "eps"))
1.22 cvs 1358: return (FALSE);
1359: else
1360: return (TRUE);
1361: }
1362: else
1363: return (FALSE);
1364: }
1365:
1366:
1367: /*----------------------------------------------------------------------
1.24 cvs 1368: ChopURL
1369: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1370: If inputURL is bigger than that size, outputURL receives
1371: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1372: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1373: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1374: copied into outputURL.
1375: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1376: chars.
1377: ----------------------------------------------------------------------*/
1.106 cvs 1378: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1379: {
1380: int len;
1.9 cvs 1381:
1.106 cvs 1382: len = strlen (inputURL);
1.24 cvs 1383: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1384: strcpy (outputURL, inputURL);
1.24 cvs 1385: else
1386: /* make a truncated urlName on the status window */
1387: {
1.106 cvs 1388: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1389: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1390: strcat (outputURL, "...");
1391: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1392: }
1.25 cvs 1393: }
1394:
1395:
1396: /*----------------------------------------------------------------------
1397: scan
1.47 cvs 1398: Scan a filename for its constituents
1.25 cvs 1399: -----------------------------------
1400:
1401: On entry,
1402: name points to a document name which may be incomplete.
1403: On exit,
1404: absolute or relative may be nonzero (but not both).
1405: host, fragment and access may be nonzero if they were specified.
1406: Any which are nonzero point to zero terminated strings.
1407: ----------------------------------------------------------------------*/
1.106 cvs 1408: static void scan (char *name, HTURI *parts)
1.25 cvs 1409: {
1.106 cvs 1410: char * p;
1411: char * after_access = name;
1.32 cvs 1412:
1.43 cvs 1413: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1414: /* Look for fragment identifier */
1.106 cvs 1415: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1416: {
1.106 cvs 1417: *p++ = '\0';
1.28 cvs 1418: parts->fragment = p;
1.25 cvs 1419: }
1420:
1.28 cvs 1421: for (p=name; *p; p++)
1422: {
1.106 cvs 1423: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1424: break;
1.106 cvs 1425: if (*p == ':')
1.28 cvs 1426: {
1427: *p = 0;
1428: parts->access = after_access; /* Scheme has been specified */
1429:
1430: /* The combination of gcc, the "-O" flag and the HP platform is
1431: unhealthy. The following three lines is a quick & dirty fix, but is
1432: not recommended. Rather, turn off "-O". */
1433:
1434: /* after_access = p;*/
1435: /* while (*after_access == 0)*/
1436: /* after_access++;*/
1437: after_access = p+1;
1.106 cvs 1438: if (!strcasecmp("URL", parts->access))
1.28 cvs 1439: /* Ignore IETF's URL: pre-prefix */
1440: parts->access = NULL;
1441: else
1.25 cvs 1442: break;
1443: }
1444: }
1445:
1446: p = after_access;
1.43 cvs 1447: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1448: {
1.43 cvs 1449: if (p[1] == URL_SEP)
1.28 cvs 1450: {
1.25 cvs 1451: parts->host = p+2; /* host has been specified */
1.28 cvs 1452: *p = 0; /* Terminate access */
1453: /* look for end of host name if any */
1.106 cvs 1454: p = strchr (parts->host, URL_SEP);
1.28 cvs 1455: if (p)
1456: {
1.106 cvs 1457: *p = EOS; /* Terminate host */
1.25 cvs 1458: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1459: }
1460: }
1461: else
1462: /* Root found but no host */
1463: parts->absolute = p+1;
1464: }
1465: else
1466: {
1.25 cvs 1467: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1468: }
1.25 cvs 1469: }
1470:
1471:
1472: /*----------------------------------------------------------------------
1.28 cvs 1473: AmayaParseUrl: parse a Name relative to another name
1474:
1475: This returns those parts of a name which are given (and requested)
1476: substituting bits from the related name where necessary.
1.25 cvs 1477:
1.28 cvs 1478: On entry,
1.25 cvs 1479: aName A filename given
1480: relatedName A name relative to which aName is to be parsed. Give
1481: it an empty string if aName is absolute.
1482: wanted A mask for the bits which are wanted.
1483:
1.28 cvs 1484: On exit,
1.25 cvs 1485: returns A pointer to a malloc'd string which MUST BE FREED
1486: ----------------------------------------------------------------------*/
1.106 cvs 1487: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1488: {
1489: char *return_value;
1490: char result[MAX_LENGTH];
1491: char name[MAX_LENGTH];
1492: char rel[MAX_LENGTH];
1493: char *p, *access;
1.29 cvs 1494: HTURI given, related;
1495: int len;
1.106 cvs 1496: char used_sep;
1497: char *used_str;
1.32 cvs 1498:
1.106 cvs 1499: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1500: {
1.106 cvs 1501: used_str = DIR_STR;
1502: used_sep = DIR_SEP;
1.33 cvs 1503: }
1.32 cvs 1504: else
1.33 cvs 1505: {
1.106 cvs 1506: used_str = URL_STR;
1507: used_sep = URL_SEP;
1.33 cvs 1508: }
1.32 cvs 1509:
1.29 cvs 1510: /* Make working copies of input strings to cut up: */
1511: return_value = NULL;
1512: result[0] = 0; /* Clear string */
1.106 cvs 1513: strcpy (name, aName);
1.29 cvs 1514: if (relatedName != NULL)
1.106 cvs 1515: strcpy (rel, relatedName);
1.29 cvs 1516: else
1.106 cvs 1517: relatedName[0] = EOS;
1.29 cvs 1518:
1519: scan (name, &given);
1520: scan (rel, &related);
1521: access = given.access ? given.access : related.access;
1522: if (wanted & AMAYA_PARSE_ACCESS)
1523: if (access)
1524: {
1.106 cvs 1525: strcat (result, access);
1.29 cvs 1526: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1527: strcat (result, ":");
1.29 cvs 1528: }
1529:
1530: if (given.access && related.access)
1531: /* If different, inherit nothing. */
1.106 cvs 1532: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1533: {
1534: related.host = 0;
1535: related.absolute = 0;
1536: related.relative = 0;
1537: related.fragment = 0;
1538: }
1539:
1540: if (wanted & AMAYA_PARSE_HOST)
1541: if(given.host || related.host)
1542: {
1543: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1544: strcat (result, "//");
1545: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1546: }
1547:
1548: if (given.host && related.host)
1549: /* If different hosts, inherit no path. */
1.106 cvs 1550: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1551: {
1552: related.absolute = 0;
1553: related.relative = 0;
1554: related.fragment = 0;
1555: }
1556:
1557: if (wanted & AMAYA_PARSE_PATH)
1558: {
1559: if (given.absolute)
1560: {
1561: /* All is given */
1562: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1563: strcat (result, used_str);
1564: strcat (result, given.absolute);
1.25 cvs 1565: }
1.29 cvs 1566: else if (related.absolute)
1567: {
1568: /* Adopt path not name */
1.106 cvs 1569: strcat (result, used_str);
1570: strcat (result, related.absolute);
1.29 cvs 1571: if (given.relative)
1572: {
1573: /* Search part? */
1.106 cvs 1574: p = strchr (result, '?');
1.29 cvs 1575: if (!p)
1.106 cvs 1576: p=result+strlen(result)-1;
1.33 cvs 1577: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1578: /* Remove filename */
1579: p[1]=0;
1580: /* Add given one */
1.106 cvs 1581: strcat (result, given.relative);
1.25 cvs 1582: }
1583: }
1.29 cvs 1584: else if (given.relative)
1585: /* what we've got */
1.106 cvs 1586: strcat (result, given.relative);
1.29 cvs 1587: else if (related.relative)
1.106 cvs 1588: strcat (result, related.relative);
1.29 cvs 1589: else
1590: /* No inheritance */
1.106 cvs 1591: strcat (result, used_str);
1.25 cvs 1592: }
1.29 cvs 1593:
1594: if (wanted & AMAYA_PARSE_ANCHOR)
1595: if (given.fragment || related.fragment)
1596: {
1597: if (given.absolute && given.fragment)
1598: {
1599: /*Fixes for relURLs...*/
1600: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1601: strcat (result, "#");
1602: strcat (result, given.fragment);
1.29 cvs 1603: }
1604: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1605: strcat (result, "");
1.29 cvs 1606: else
1607: {
1.110 cvs 1608: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1609: strcat (result, "#");
1.110 cvs 1610: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1611: }
1612: }
1.106 cvs 1613: len = strlen (result);
1614: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1615: strcpy (return_value, result);
1.29 cvs 1616: return (return_value); /* exactly the right length */
1.25 cvs 1617: }
1618:
1619: /*----------------------------------------------------------------------
1620: HTCanon
1621: Canonicalizes the URL in the following manner starting from the host
1622: pointer:
1623:
1624: 1) The host name is converted to lowercase
1625: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1626:
1627: Return: OK The position of the current path part of the URL
1628: which might be the old one or a new one.
1629:
1630: ----------------------------------------------------------------------*/
1.106 cvs 1631: static char *HTCanon (char **filename, char *host)
1632: {
1633: char *newname = NULL;
1634: char used_sep;
1635: char *path;
1636: char *strptr;
1637: char *port;
1638: char *access = host-3;
1639:
1640: if (*filename && strchr (*filename, URL_SEP))
1641: used_sep = URL_SEP;
1.33 cvs 1642: else
1.106 cvs 1643: used_sep = DIR_SEP;
1.32 cvs 1644:
1.110 cvs 1645: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1646: access--;
1.110 cvs 1647: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1648: path = host + strlen (host);
1649: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1650: host = strptr;
1.110 cvs 1651: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1652: port = NULL;
1.25 cvs 1653:
1654: strptr = host; /* Convert to lower-case */
1.82 cvs 1655: while (strptr < path)
1.33 cvs 1656: {
1.123 vatton 1657: *strptr = tolower (*strptr);
1.82 cvs 1658: strptr++;
1.33 cvs 1659: }
1.25 cvs 1660:
1661: /* Does the URL contain a full domain name? This also works for a
1662: numerical host name. The domain name is already made lower-case
1663: and without a trailing dot. */
1664: {
1.106 cvs 1665: char *dot = port ? port : path;
1666: if (dot > *filename && *--dot == '.')
1.33 cvs 1667: {
1.106 cvs 1668: char *orig = dot;
1669: char *dest = dot + 1;
1.82 cvs 1670: while ((*orig++ = *dest++));
1671: if (port) port--;
1.33 cvs 1672: path--;
1.25 cvs 1673: }
1674: }
1675: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1676: if (port)
1677: {
1.82 cvs 1678: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1679: {
1680: if (!newname)
1681: {
1.106 cvs 1682: char *orig = port;
1683: char *dest = port + 1;
1.82 cvs 1684: while ((*orig++ = *dest++));
1.33 cvs 1685: }
1686: }
1.106 cvs 1687: else if ((!strncmp (access, "http", 4) &&
1688: (*(port + 1) == '8' &&
1689: *(port+2) == '0' &&
1.82 cvs 1690: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1691: (!strncmp (access, "gopher", 6) &&
1692: (*(port+1) == '7' &&
1693: *(port+2) == '0' &&
1.82 cvs 1694: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1695: (!strncmp (access, "ftp", 3) &&
1696: (*(port+1) == '2' &&
1697: *(port + 2) == '1' &&
1.82 cvs 1698: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1699: if (!newname)
1700: {
1.106 cvs 1701: char *orig = port;
1702: char *dest = port + 3;
1.33 cvs 1703: while((*orig++ = *dest++));
1704: /* Update path position, Henry Minsky */
1705: path -= 3;
1.25 cvs 1706: }
1.33 cvs 1707: }
1708: else if (newname)
1.106 cvs 1709: strncat (newname, port, (int) (path - port));
1.33 cvs 1710: }
1.25 cvs 1711:
1.33 cvs 1712: if (newname)
1713: {
1.106 cvs 1714: char *newpath = newname + strlen (newname);
1715: strcat (newname, path);
1.25 cvs 1716: path = newpath;
1.28 cvs 1717: /* Free old copy */
1718: TtaFreeMemory(*filename);
1.25 cvs 1719: *filename = newname;
1.33 cvs 1720: }
1.25 cvs 1721: return path;
1722: }
1723:
1724:
1725: /*----------------------------------------------------------------------
1.29 cvs 1726: SimplifyUrl: simplify a URI
1.32 cvs 1727: A URI is allowed to contain the sequence xxx/../ which may be
1728: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1729: Simplification helps us recognize duplicate URIs.
1.25 cvs 1730:
1.28 cvs 1731: Thus, /etc/junk/../fred becomes /etc/fred
1732: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1733:
1.28 cvs 1734: but we should NOT change
1735: http://fred.xxx.edu/../..
1.25 cvs 1736:
1737: or ../../albert.html
1738:
1.28 cvs 1739: In order to avoid empty URLs the following URLs become:
1.25 cvs 1740:
1741: /fred/.. becomes /fred/..
1742: /fred/././.. becomes /fred/..
1743: /fred/.././junk/.././ becomes /fred/..
1744:
1.28 cvs 1745: If more than one set of `://' is found (several proxies in cascade) then
1746: only the part after the last `://' is simplified.
1.25 cvs 1747:
1.28 cvs 1748: Returns: A string which might be the old one or a new one.
1.25 cvs 1749: ----------------------------------------------------------------------*/
1.106 cvs 1750: void SimplifyUrl (char **url)
1751: {
1752: char *path;
1753: char *access;
1754: char *newptr;
1755: char *p;
1756: char *orig, *dest, *end;
1.28 cvs 1757:
1.106 cvs 1758: char used_sep;
1.77 cvs 1759: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1760: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1761:
1762:
1.28 cvs 1763: if (!url || !*url)
1764: return;
1765:
1.106 cvs 1766: if (strchr (*url, URL_SEP))
1767: used_sep = URL_SEP;
1.32 cvs 1768: else
1.106 cvs 1769: used_sep = DIR_SEP;
1.32 cvs 1770:
1.77 cvs 1771: /* should we simplify double dot? */
1772: path = *url;
1.106 cvs 1773: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1774: ddot_simplify = FALSE;
1775: else
1776: ddot_simplify = TRUE;
1777:
1.28 cvs 1778: /* Find any scheme name */
1.106 cvs 1779: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1780: {
1781: /* Find host name */
1.28 cvs 1782: access = *url;
1.123 vatton 1783: while (access < path && (*access = tolower (*access)))
1.82 cvs 1784: access++;
1.28 cvs 1785: path += 3;
1.106 cvs 1786: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1787: /* For proxies */
1.106 cvs 1788: path = newptr + 3;
1.82 cvs 1789: /* We have a host name */
1.84 cvs 1790: path = HTCanon (url, path);
1.25 cvs 1791: }
1.106 cvs 1792: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1793: path += 2;
1794: else
1795: path = *url;
1.25 cvs 1796:
1.84 cvs 1797: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1798: /* Some URLs start //<foo> */
1799: path += 1;
1.94 cvs 1800: else if (IsFilePath (path))
1801: {
1802: /* doesn't need to do anything more */
1803: return;
1804: }
1.106 cvs 1805: else if (!strncmp (path, "news:", 5))
1.28 cvs 1806: {
1.106 cvs 1807: newptr = strchr (path+5, '@');
1.28 cvs 1808: if (!newptr)
1809: newptr = path + 5;
1810: while (*newptr)
1811: {
1812: /* Make group or host lower case */
1.123 vatton 1813: *newptr = tolower (*newptr);
1.28 cvs 1814: newptr++;
1.25 cvs 1815: }
1.28 cvs 1816: /* Doesn't need to do any more */
1817: return;
1.25 cvs 1818: }
1.126 cheyroul 1819: else if (**url != DIR_SEP
1820: && **url != '~'
1.127 ! cvs 1821: #ifdef _WINDOWS
! 1822: && (*url)[1] != ':'
! 1823: #endif /* _WINDOWS */
1.126 cheyroul 1824: && !IsW3Path(*url)
1825: /* && TtaFileExist (*url) == 0) */
1.127 ! cvs 1826: && (strlen (*url) + 8) < MAX_LENGTH)
1.126 cheyroul 1827: {
1828: /* In case of a user typed url without protocol specification
1829: and filepath like url (the ~ or / url beginning),
1830: we add the http:// (more conveniant when you often type urls)
1831: so that you can now enter w3.org directly */
1.127 ! cvs 1832: newptr = TtaGetMemory (strlen (path) + 8);
1.126 cheyroul 1833: *newptr = EOS;
1.125 cheyroul 1834: strcat (newptr, "http://");
1835: strcat (newptr, *url);
1.126 cheyroul 1836: **url = EOS;
1837: strcpy (*url, newptr);
1838: TtaFreeMemory (newptr);
1.125 cheyroul 1839: return;
1.126 cheyroul 1840: }
1841:
1.28 cvs 1842: if ((p = path))
1843: {
1.106 cvs 1844: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1845: (end = strchr (path, '#'))))
1846: end = path + strlen (path);
1.28 cvs 1847:
1848: /* Parse string second time to simplify */
1849: p = path;
1850: while (p < end)
1851: {
1.110 cvs 1852: /* if we're pointing to a char, it's safe to reactivate the
1853: ../ convertion */
1.106 cvs 1854: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1855: ddot_simplify = TRUE;
1856:
1.33 cvs 1857: if (*p==used_sep)
1.28 cvs 1858: {
1.106 cvs 1859: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1860: {
1861: orig = p + 1;
1.84 cvs 1862: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1863: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1864: end = orig - 1;
1865: }
1.106 cvs 1866: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1867: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1868: {
1869: newptr = p;
1.52 cvs 1870: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1871: if (*newptr == used_sep)
1872: orig = newptr + 1;
1.28 cvs 1873: else
1.52 cvs 1874: orig = newptr;
1875:
1876: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1877: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1878: end = orig-1;
1879: /* Start again with prev slash */
1880: p = newptr;
1.28 cvs 1881: }
1.33 cvs 1882: else if (*(p+1) == used_sep)
1.28 cvs 1883: {
1.33 cvs 1884: while (*(p+1) == used_sep)
1.28 cvs 1885: {
1886: orig = p;
1887: dest = p + 1;
1888: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1889: end = orig-1;
1890: }
1891: }
1892: else
1.25 cvs 1893: p++;
1.28 cvs 1894: }
1895: else
1896: p++;
1.25 cvs 1897: }
1898: }
1.51 cvs 1899:
1900: /*
1901: ** Check for host/../.. kind of things
1902: */
1.106 cvs 1903: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1904: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 1905: *(path+1) = EOS;
1.51 cvs 1906:
1.28 cvs 1907: return;
1908: }
1909:
1910:
1911: /*----------------------------------------------------------------------
1.96 cvs 1912: NormalizeFile normalizes local names.
1.28 cvs 1913: Return TRUE if target and src differ.
1914: ----------------------------------------------------------------------*/
1.106 cvs 1915: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 1916: {
1.110 cvs 1917: #ifndef _WINDOWS
1.106 cvs 1918: char *s;
1.93 cvs 1919: int i;
1.110 cvs 1920: #endif /* !_WINDOWS */
1.82 cvs 1921: ThotBool change;
1.90 cvs 1922: int start_index; /* the first char that we'll copy */
1.28 cvs 1923:
1.54 cvs 1924: change = FALSE;
1.90 cvs 1925: start_index = 0;
1926:
1.106 cvs 1927: if (!src || src[0] == EOS)
1.96 cvs 1928: {
1.106 cvs 1929: target[0] = EOS;
1.96 cvs 1930: return FALSE;
1931: }
1.90 cvs 1932:
1933: /* @@ do I need file: or file:/ here? */
1.106 cvs 1934: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1935: {
1.90 cvs 1936: /* remove the prefix file: */
1937: start_index += 5;
1938:
1939: /* remove the localhost prefix */
1.106 cvs 1940: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 1941: start_index += 11;
1942:
1943: /* remove the first two slashes in / / /path */
1944: while (src[start_index] &&
1.106 cvs 1945: src[start_index] == '/'
1946: && src[start_index + 1] == '/')
1.94 cvs 1947: start_index++;
1948:
1949: #ifdef _WINDOWS
1950: /* remove any extra slash before the drive name */
1.106 cvs 1951: if (src[start_index] == '/'
1952: &&src[start_index+2] == ':')
1.94 cvs 1953: start_index++;
1954: #endif /* _WINDOWS */
1.90 cvs 1955:
1.106 cvs 1956: if (src[start_index] == EOS)
1.90 cvs 1957: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 1958: strcpy (target, DIR_STR);
1.90 cvs 1959: else
1.97 cvs 1960: /* as we're inside a file: URL, we'll apply all the convertions
1961: we know */
1962: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1963:
1964: change = TRUE;
1965: }
1.97 cvs 1966: else if (convertion != AM_CONV_NONE)
1.96 cvs 1967: {
1968: /* we are following a "local" relative link, we do all the
1969: convertions except for the HOME_DIR ~ one */
1.97 cvs 1970: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1971: }
1.90 cvs 1972: #ifndef _WINDOWS
1.106 cvs 1973: else if (src[0] == '~')
1.53 cvs 1974: {
1.96 cvs 1975: /* it must be a URL typed in a text input field */
1976: /* do the HOME_DIR ~ substitution */
1.82 cvs 1977: s = TtaGetEnvString ("HOME");
1.106 cvs 1978: strcpy (target, s);
1.90 cvs 1979: #if 0
1.96 cvs 1980: /* JK: invalidated this part of the code as it's simpler
1981: to add the DIR_SEP whenever we have something to add
1982: to the path rather than adding it systematically */
1.106 cvs 1983: if (src[1] != DIR_SEP)
1984: strcat (target, DIR_STR);
1.90 cvs 1985: #endif
1.106 cvs 1986: i = strlen (target);
1987: strcpy (&target[i], &src[1]);
1.54 cvs 1988: change = TRUE;
1.53 cvs 1989: }
1.90 cvs 1990: #endif /* _WINDOWS */
1.28 cvs 1991: else
1.96 cvs 1992: /* leave it as it is */
1.106 cvs 1993: strcpy (target, src);
1.96 cvs 1994:
1.28 cvs 1995: /* remove /../ and /./ */
1.29 cvs 1996: SimplifyUrl (&target);
1.54 cvs 1997: if (!change)
1.106 cvs 1998: change = strcmp (src, target);
1.28 cvs 1999: return (change);
1.25 cvs 2000: }
2001:
1.28 cvs 2002:
1.25 cvs 2003: /*----------------------------------------------------------------------
1.31 cvs 2004: MakeRelativeURL: make relative name
1.25 cvs 2005:
1.28 cvs 2006: This function creates and returns a string which gives an expression of
2007: one address as related to another. Where there is no relation, an absolute
2008: address is retured.
1.25 cvs 2009:
1.28 cvs 2010: On entry,
1.25 cvs 2011: Both names must be absolute, fully qualified names of nodes
2012: (no fragment bits)
2013:
1.28 cvs 2014: On exit,
1.25 cvs 2015: The return result points to a newly allocated name which, if
2016: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2017: The caller is responsible for freeing the resulting name later.
2018: ----------------------------------------------------------------------*/
1.106 cvs 2019: char *MakeRelativeURL (char *aName, char *relatedName)
2020: {
2021: char *return_value;
2022: char result[MAX_LENGTH];
2023: char *p;
2024: char *q;
2025: char *after_access;
2026: char *last_slash = NULL;
2027: int slashes, levels, len;
1.110 cvs 2028: #ifdef _WINDOWS
1.44 cvs 2029: int ndx;
1.110 cvs 2030: #endif /* _WINDOWS */
1.44 cvs 2031:
1.29 cvs 2032: if (aName == NULL || relatedName == NULL)
2033: return (NULL);
2034:
2035: slashes = 0;
2036: after_access = NULL;
2037: p = aName;
2038: q = relatedName;
2039: for (; *p && (*p == *q); p++, q++)
1.27 cvs 2040: {
2041: /* Find extent of match */
1.106 cvs 2042: if (*p == ':')
1.29 cvs 2043: after_access = p + 1;
1.28 cvs 2044: if (*p == DIR_SEP)
1.27 cvs 2045: {
1.29 cvs 2046: /* memorize the last slash position and count them */
1.27 cvs 2047: last_slash = p;
2048: slashes++;
1.25 cvs 2049: }
2050: }
2051:
1.31 cvs 2052: /* q, p point to the first non-matching character or zero */
1.106 cvs 2053: if (*q == EOS)
1.31 cvs 2054: {
2055: /* New name is a subset of the related name */
2056: /* exactly the right length */
1.106 cvs 2057: len = strlen (p);
2058: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2059: strcpy (return_value, p);
1.31 cvs 2060: }
2061: else if ((slashes < 2 && after_access == NULL)
2062: || (slashes < 3 && after_access != NULL))
2063: {
2064: /* Two names whitout common path */
2065: /* exactly the right length */
1.106 cvs 2066: len = strlen (aName);
2067: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2068: strcpy (return_value, aName);
1.31 cvs 2069: }
2070: else
2071: {
2072: /* Some path in common */
1.106 cvs 2073: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2074: /* just the same server */
1.106 cvs 2075: strcpy (result, last_slash);
1.31 cvs 2076: else
2077: {
2078: levels= 0;
1.106 cvs 2079: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2080: if (*q == DIR_SEP)
2081: levels++;
2082:
1.106 cvs 2083: result[0] = EOS;
1.31 cvs 2084: for (;levels; levels--)
1.106 cvs 2085: strcat (result, "../");
2086: strcat (result, last_slash+1);
1.31 cvs 2087: }
1.52 cvs 2088:
2089: if (!*result)
1.106 cvs 2090: strcat (result, "./");
1.52 cvs 2091:
1.31 cvs 2092: /* exactly the right length */
1.106 cvs 2093: len = strlen (result);
2094: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2095: strcpy (return_value, result);
1.52 cvs 2096:
1.25 cvs 2097: }
1.110 cvs 2098: #ifdef _WINDOWS
1.106 cvs 2099: len = strlen (return_value);
1.44 cvs 2100: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2101: if (return_value[ndx] == '\\')
2102: return_value[ndx] = '/' ;
1.110 cvs 2103: #endif /* _WINDOWS */
1.29 cvs 2104: return (return_value);
1.24 cvs 2105: }
1.35 cvs 2106:
1.104 kahan 2107: /*----------------------------------------------------------------------
2108: AM_GetFileSize
2109: Returns TRUE and the filesize in the 2nd parameter.
2110: Otherwise, in case of a system error, returns FALSE, with a
2111: filesize of 0L.
2112: ---------------------------------------------------------------------*/
1.106 cvs 2113: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2114: {
1.106 cvs 2115: ThotFileHandle handle = ThotFile_BADHANDLE;
2116: ThotFileInfo info;
1.35 cvs 2117:
1.104 kahan 2118: *file_size = 0L;
2119: if (!TtaFileExist (filename))
2120: return FALSE;
2121:
2122: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2123: if (handle == ThotFile_BADHANDLE)
2124: /* ThotFile_BADHANDLE */
2125: return FALSE;
2126: if (TtaFileStat (handle, &info) == 0)
2127: /* bad stat */
2128: info.size = 0L;
2129: TtaFileClose (handle);
2130: *file_size = (unsigned long) info.size;
2131: return TRUE;
2132: }
Webmaster
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to AHTURLTools.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Public] / Amaya / amaya