Annotation of Amaya/amaya/AHTURLTools.c, revision 1.136
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:
1.4 cvs 423: /*----------------------------------------------------------------------
1.9 cvs 424: IsHTMLName
425: returns TRUE if path points to an HTML resource.
1.4 cvs 426: ----------------------------------------------------------------------*/
1.109 cvs 427: ThotBool IsHTMLName (const char *path)
1.106 cvs 428: {
1.136 ! cvs 429: char temppath[MAX_LENGTH];
! 430: char suffix[MAX_LENGTH];
! 431: char nsuffix[MAX_LENGTH];
! 432: int i;
1.5 cvs 433:
1.101 cvs 434: if (!path)
435: return (FALSE);
1.5 cvs 436:
1.106 cvs 437: strcpy (temppath, path);
1.124 vatton 438: TtaExtractSuffix (temppath, suffix);
1.101 cvs 439: i = 0;
1.106 cvs 440: while (suffix[i] != EOS)
1.101 cvs 441: {
442: /* Normalize the suffix */
443: i = 0;
1.106 cvs 444: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 445: {
1.123 vatton 446: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 447: i++;
448: }
1.106 cvs 449: nsuffix[i] = EOS;
450: if (!strcmp (nsuffix, "html") ||
451: !strcmp (nsuffix, "htm") ||
452: !strcmp (nsuffix, "shtml") ||
453: !strcmp (nsuffix, "jsp") ||
454: !strcmp (nsuffix, "xht") ||
455: !strcmp (nsuffix, "xhtm") ||
456: !strcmp (nsuffix, "xhtml"))
1.101 cvs 457: return (TRUE);
1.106 cvs 458: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 459: {
460: /* take into account compressed files */
1.124 vatton 461: TtaExtractSuffix (temppath, suffix);
1.101 cvs 462: /* Normalize the suffix */
463: i = 0;
1.106 cvs 464: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 465: {
1.123 vatton 466: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 467: i++;
468: }
1.106 cvs 469: nsuffix[i] = EOS;
470: if (!strcmp (nsuffix, "html") ||
471: !strcmp (nsuffix, "htm") ||
472: !strcmp (nsuffix, "shtml") ||
473: !strcmp (nsuffix, "jsp") ||
474: !strcmp (nsuffix, "xht") ||
475: !strcmp (nsuffix, "xhtm") ||
476: !strcmp (nsuffix, "xhtml"))
1.101 cvs 477: return (TRUE);
478: else
479: return (FALSE);
480: }
481: else
482: /* check if there is another suffix */
1.124 vatton 483: TtaExtractSuffix (temppath, suffix);
1.101 cvs 484: }
1.88 cvs 485: return (FALSE);
1.3 cvs 486: }
487:
1.4 cvs 488: /*----------------------------------------------------------------------
1.136 ! cvs 489: IsMathMLName
! 490: returns TRUE if path points to an MathML resource.
1.56 cvs 491: ----------------------------------------------------------------------*/
1.136 ! cvs 492: ThotBool IsMathMLName (const char *path)
1.56 cvs 493: {
1.136 ! cvs 494: char temppath[MAX_LENGTH];
! 495: char suffix[MAX_LENGTH];
1.56 cvs 496:
497: if (!path)
498: return (FALSE);
499:
1.106 cvs 500: strcpy (temppath, path);
1.124 vatton 501: TtaExtractSuffix (temppath, suffix);
1.56 cvs 502:
1.136 ! cvs 503: if (!strcasecmp (suffix, "mml"))
1.56 cvs 504: return (TRUE);
1.106 cvs 505: else if (!strcmp (suffix, "gz"))
1.56 cvs 506: {
507: /* take into account compressed files */
1.124 vatton 508: TtaExtractSuffix (temppath, suffix);
1.136 ! cvs 509: if (!strcasecmp (suffix, "mml"))
1.60 cvs 510: return (TRUE);
511: else
512: return (FALSE);
513: }
514: else
515: return (FALSE);
516: }
517:
518: /*----------------------------------------------------------------------
1.136 ! cvs 519: IsSVGName
! 520: returns TRUE if path points to an SVG resource.
1.133 vatton 521: ----------------------------------------------------------------------*/
1.136 ! cvs 522: ThotBool IsSVGName (const char *path)
1.133 vatton 523: {
1.136 ! cvs 524: char temppath[MAX_LENGTH];
! 525: char suffix[MAX_LENGTH];
1.133 vatton 526:
527: if (!path)
528: return (FALSE);
529:
530: strcpy (temppath, path);
531: TtaExtractSuffix (temppath, suffix);
532:
1.136 ! cvs 533: if (!strcasecmp (suffix, "svg"))
1.133 vatton 534: return (TRUE);
535: else if (!strcmp (suffix, "gz"))
536: {
537: /* take into account compressed files */
538: TtaExtractSuffix (temppath, suffix);
1.136 ! cvs 539: if (!strcasecmp (suffix, "svg"))
1.133 vatton 540: return (TRUE);
541: else
542: return (FALSE);
543: }
544: else
545: return (FALSE);
546: }
547:
548: /*----------------------------------------------------------------------
1.136 ! cvs 549: IsXMLName
! 550: returns TRUE if path points to an XML resource.
1.103 cvs 551: ----------------------------------------------------------------------*/
1.136 ! cvs 552: ThotBool IsXMLName (const char *path)
1.103 cvs 553: {
1.136 ! cvs 554: char temppath[MAX_LENGTH];
! 555: char suffix[MAX_LENGTH];
1.103 cvs 556:
557: if (!path)
558: return (FALSE);
559:
1.106 cvs 560: strcpy (temppath, path);
1.124 vatton 561: TtaExtractSuffix (temppath, suffix);
1.103 cvs 562:
1.136 ! cvs 563: if (!strcasecmp (suffix, "xml") ||
! 564: !strcasecmp (suffix, "xht") ||
! 565: !strcmp (suffix, "xhtm") ||
! 566: !strcmp (suffix, "xhtml"))
1.103 cvs 567: return (TRUE);
1.106 cvs 568: else if (!strcmp (suffix, "gz"))
1.103 cvs 569: {
570: /* take into account compressed files */
1.124 vatton 571: TtaExtractSuffix (temppath, suffix);
1.136 ! cvs 572: if (!strcasecmp (suffix, "xml") ||
! 573: !strcasecmp (suffix, "xht") ||
! 574: !strcmp (suffix, "xhtm") ||
! 575: !strcmp (suffix, "xhtml"))
1.103 cvs 576: return (TRUE);
577: else
578: return (FALSE);
579: }
580: else
581: return (FALSE);
582: }
583:
584: /*----------------------------------------------------------------------
1.136 ! cvs 585: IsUndisplayedName
! 586: returns TRUE if path points to an undisplayed resource.
1.103 cvs 587: ----------------------------------------------------------------------*/
1.136 ! cvs 588: ThotBool IsUndisplayedName (const char *path)
1.103 cvs 589: {
1.106 cvs 590: char temppath[MAX_LENGTH];
591: char suffix[MAX_LENGTH];
1.103 cvs 592:
593: if (!path)
594: return (FALSE);
595:
1.106 cvs 596: strcpy (temppath, path);
1.124 vatton 597: TtaExtractSuffix (temppath, suffix);
1.103 cvs 598:
1.136 ! cvs 599: if (!strcasecmp (suffix, "exe") ||
! 600: !strcasecmp (suffix, "zip") ||
! 601: !strcasecmp (suffix, "ppt") ||
! 602: !strcasecmp (suffix, "pdf") ||
! 603: !strcasecmp (suffix, "ps") ||
! 604: !strcasecmp (suffix, "eps") ||
! 605: !strcasecmp (suffix, "tar") ||
! 606: !strcasecmp (suffix, "tgz") ||
! 607: !strcasecmp (suffix, "ddl") ||
! 608: !strcasecmp (suffix, "o"))
1.103 cvs 609: return (TRUE);
1.106 cvs 610: else if (!strcmp (suffix, "gz"))
1.103 cvs 611: {
612: /* take into account compressed files */
1.124 vatton 613: TtaExtractSuffix (temppath, suffix);
1.136 ! cvs 614: if (!strcasecmp (suffix, "exe") ||
! 615: !strcasecmp (suffix, "zip") ||
! 616: !strcasecmp (suffix, "ppt") ||
! 617: !strcasecmp (suffix, "pdf") ||
! 618: !strcasecmp (suffix, "ps") ||
! 619: !strcasecmp (suffix, "eps") ||
! 620: !strcasecmp (suffix, "tar") ||
! 621: !strcasecmp (suffix, "ddl") ||
! 622: !strcasecmp (suffix, "o"))
1.103 cvs 623: return (TRUE);
624: else
625: return (FALSE);
626: }
627: else
628: return (FALSE);
629: }
630:
631: /*----------------------------------------------------------------------
1.60 cvs 632: IsCSSName
633: returns TRUE if path points to an XML resource.
634: ----------------------------------------------------------------------*/
1.111 cvs 635: ThotBool IsCSSName (const char *path)
1.60 cvs 636: {
1.106 cvs 637: char temppath[MAX_LENGTH];
638: char suffix[MAX_LENGTH];
1.60 cvs 639:
640: if (!path)
641: return (FALSE);
642:
1.106 cvs 643: strcpy (temppath, path);
1.124 vatton 644: TtaExtractSuffix (temppath, suffix);
1.60 cvs 645:
1.106 cvs 646: if (!strcasecmp (suffix, "css"))
1.60 cvs 647: return (TRUE);
1.106 cvs 648: else if (!strcmp (suffix, "gz"))
1.60 cvs 649: {
650: /* take into account compressed files */
1.124 vatton 651: TtaExtractSuffix (temppath, suffix);
1.106 cvs 652: if (!strcasecmp (suffix, "css"))
1.56 cvs 653: return (TRUE);
654: else
655: return (FALSE);
656: }
657: else
658: return (FALSE);
659: }
660:
661: /*----------------------------------------------------------------------
1.9 cvs 662: IsImageName
663: returns TRUE if path points to an image resource.
1.4 cvs 664: ----------------------------------------------------------------------*/
1.111 cvs 665: ThotBool IsImageName (const char *path)
1.106 cvs 666: {
667: char temppath[MAX_LENGTH];
668: char suffix[MAX_LENGTH];
669: char nsuffix[MAX_LENGTH];
1.5 cvs 670: int i;
671:
672: if (!path)
1.13 cvs 673: return (FALSE);
1.5 cvs 674:
1.106 cvs 675: strcpy (temppath, path);
1.124 vatton 676: TtaExtractSuffix (temppath, suffix);
1.5 cvs 677:
678: /* Normalize the suffix */
679: i = 0;
1.106 cvs 680: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 681: {
1.123 vatton 682: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 683: i++;
684: }
1.106 cvs 685: nsuffix[i] = EOS;
686: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
687: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
688: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 689: return (TRUE);
690: return (FALSE);
1.3 cvs 691: }
692:
1.4 cvs 693: /*----------------------------------------------------------------------
1.58 cvs 694: IsImageType
695: returns TRUE if type points to an image resource.
696: ----------------------------------------------------------------------*/
1.111 cvs 697: ThotBool IsImageType (const char *type)
1.58 cvs 698: {
1.106 cvs 699: char temptype[MAX_LENGTH];
1.58 cvs 700: int i;
701:
702: if (!type)
703: return (FALSE);
704:
1.106 cvs 705: strcpy (temptype, type);
1.58 cvs 706: /* Normalize the type */
707: i = 0;
1.106 cvs 708: while (temptype[i] != EOS)
1.58 cvs 709: {
710: temptype[i] = tolower (temptype[i]);
711: i++;
712: }
1.111 cvs 713: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
714: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
715: !strcmp (temptype, "png"))
1.58 cvs 716: return (TRUE);
717: return (FALSE);
718: }
719:
720: /*----------------------------------------------------------------------
1.9 cvs 721: IsTextName
1.4 cvs 722: ----------------------------------------------------------------------*/
1.111 cvs 723: ThotBool IsTextName (const char *path)
1.106 cvs 724: {
725: char temppath[MAX_LENGTH];
726: char suffix[MAX_LENGTH];
727: char nsuffix[MAX_LENGTH];
1.5 cvs 728: int i;
729:
730: if (!path)
1.13 cvs 731: return (FALSE);
1.5 cvs 732:
1.106 cvs 733: strcpy (temppath, path);
1.124 vatton 734: TtaExtractSuffix (temppath, suffix);
1.5 cvs 735:
736: /* Normalize the suffix */
737: i = 0;
1.106 cvs 738: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 739: {
1.25 cvs 740: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 741: i++;
742: }
1.106 cvs 743: nsuffix[i] = EOS;
1.5 cvs 744:
1.111 cvs 745: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 746: return (TRUE);
1.106 cvs 747: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 748: {
1.39 cvs 749: /* take into account compressed files */
1.124 vatton 750: TtaExtractSuffix (temppath, suffix);
1.13 cvs 751: /* Normalize the suffix */
752: i = 0;
1.106 cvs 753: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 754: {
1.25 cvs 755: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 756: i++;
757: }
1.106 cvs 758: nsuffix[i] = EOS;
1.111 cvs 759: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 760: return (TRUE);
761: else
762: return (FALSE);
763: }
764: else
765: return (FALSE);
1.3 cvs 766: }
767:
1.4 cvs 768: /*----------------------------------------------------------------------
1.9 cvs 769: IsHTTPPath
770: returns TRUE if path is in fact an http URL.
1.4 cvs 771: ----------------------------------------------------------------------*/
1.112 cvs 772: ThotBool IsHTTPPath (const char *path)
1.3 cvs 773: {
1.5 cvs 774: if (!path)
775: return FALSE;
1.3 cvs 776:
1.106 cvs 777: if ((!strncmp (path, "http:", 5) != 0)
778: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
779: || !strncmp (path, "internal:", 9))
1.58 cvs 780: return TRUE;
781: return FALSE;
1.3 cvs 782: }
783:
1.4 cvs 784: /*----------------------------------------------------------------------
1.9 cvs 785: IsWithParameters
786: returns TRUE if url has a concatenated query string.
1.4 cvs 787: ----------------------------------------------------------------------*/
1.133 vatton 788: ThotBool IsWithParameters (const char *url)
1.3 cvs 789: {
1.5 cvs 790: int i;
1.3 cvs 791:
1.9 cvs 792: if ((!url) || (url[0] == EOS))
1.5 cvs 793: return FALSE;
1.3 cvs 794:
1.9 cvs 795: i = strlen (url) - 1;
796: while (i > 0 && url[i--] != '?')
1.5 cvs 797: if (i < 0)
798: return FALSE;
1.3 cvs 799:
1.5 cvs 800: /* There is a parameter */
801: return TRUE;
1.3 cvs 802: }
803:
1.4 cvs 804: /*----------------------------------------------------------------------
1.9 cvs 805: IsW3Path
806: returns TRUE if path is in fact a URL.
1.4 cvs 807: ----------------------------------------------------------------------*/
1.133 vatton 808: ThotBool IsW3Path (const char *path)
1.106 cvs 809: {
810: if (strncmp (path, "http:", 5) &&
811: strncmp (path, "ftp:", 4) &&
812: strncmp (path, "telnet:", 7) &&
813: strncmp (path, "wais:", 5) &&
814: strncmp (path, "news:", 5) &&
815: strncmp (path, "gopher:", 7) &&
816: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 817: strncmp (path, "archie:", 7) &&
818: strncmp (path, "https:", 6))
1.72 cvs 819: return FALSE;
820: return TRUE;
1.3 cvs 821: }
822:
1.4 cvs 823: /*----------------------------------------------------------------------
1.90 cvs 824: IsFilePath
825: returns TRUE if path is in fact a URL.
826: ----------------------------------------------------------------------*/
1.133 vatton 827: ThotBool IsFilePath (const char *path)
1.90 cvs 828: {
1.106 cvs 829: if (strncmp (path, "file:", 5))
1.90 cvs 830: return FALSE;
831: return TRUE;
832: }
833:
834: /*----------------------------------------------------------------------
1.9 cvs 835: IsValidProtocol
836: returns true if the url protocol is supported by Amaya.
1.4 cvs 837: ----------------------------------------------------------------------*/
1.133 vatton 838: ThotBool IsValidProtocol (const char *url)
1.106 cvs 839: {
840: if (!strncmp (url, "http:", 5)
841: || !strncmp (url, "internal:", 9)
842: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 843: /* experimental */
1.24 cvs 844: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 845: return (TRUE);
1.5 cvs 846: else
1.8 cvs 847: return (FALSE);
1.3 cvs 848: }
849:
1.31 cvs 850:
851: /*----------------------------------------------------------------------
852: GetBaseURL
853: normalizes orgName according to a base associated with doc, and
854: following the standard URL format rules.
855: The function returns the base used to solve relative URL and SRC:
856: - the base of the document,
857: - or the document path (without document name).
858: ----------------------------------------------------------------------*/
1.106 cvs 859: char *GetBaseURL (Document doc)
1.31 cvs 860: {
861: Element el;
862: ElementType elType;
863: AttributeType attrType;
864: Attribute attr;
1.106 cvs 865: char *ptr, *basename;
1.31 cvs 866: int length;
867:
1.113 cvs 868: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 869: return NULL;
1.106 cvs 870: basename = TtaGetMemory (MAX_LENGTH);
871: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
872: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 873: length = MAX_LENGTH -1;
1.113 cvs 874: /* is it a HTML document ? */
1.31 cvs 875: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.113 cvs 876: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
877: /* it's a HTML document */
1.65 cvs 878: {
1.113 cvs 879: /* get the document element */
880: el = TtaGetMainRoot (doc);
881: /* search the BASE element */
882: elType.ElTypeNum = HTML_EL_HEAD;
883: el = TtaSearchTypedElement (elType, SearchForward, el);
884: if (el)
885: /* there is a HEAD element */
886: {
887: /* look for a BASE element within the HEAD */
888: elType.ElTypeNum = HTML_EL_BASE;
889: el = TtaSearchTypedElement (elType, SearchInTree, el);
890: }
891: if (el)
1.31 cvs 892: {
1.113 cvs 893: /* The document has a BASE element. Get the HREF attribute of the
894: BASE element */
895: attrType.AttrSSchema = elType.ElSSchema;
896: attrType.AttrTypeNum = HTML_ATTR_HREF_;
897: attr = TtaGetAttribute (el, attrType);
898: if (attr)
1.31 cvs 899: {
1.113 cvs 900: /* Use the base path of the document */
901: TtaGiveTextAttributeValue (attr, basename, &length);
902: /* base and orgName have to be separated by a DIR_SEP */
903: length--;
904: if (basename[0] != EOS && basename[length] != URL_SEP &&
905: basename[length] != DIR_SEP)
906: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 907: {
1.113 cvs 908: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
909: AMAYA_PARSE_HOST |
910: AMAYA_PARSE_PUNCTUATION);
911: if (ptr && !strcmp (ptr, basename))
912: {
913: /* it has this form, complete it by adding a URL_STR */
914: if (strchr (basename, DIR_SEP))
915: strcat (basename, DIR_STR);
916: else
917: strcat (basename, URL_STR);
918: length++;
919: }
920: if (ptr)
921: TtaFreeMemory (ptr);
1.31 cvs 922: }
923: }
924: }
1.113 cvs 925: }
926:
1.31 cvs 927: /* Remove anything after the last DIR_SEP char. If no such char is found,
928: * then search for the first ":" char, hoping that what's before that is a
929: * protocol. If found, end the string there. If neither char is found,
930: * then discard the whole base element.
931: */
1.106 cvs 932: length = strlen (basename) - 1;
1.31 cvs 933: /* search for the last DIR_SEP char */
1.106 cvs 934: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 935: length--;
936: if (length >= 0)
937: /* found the last DIR_SEP char, end the string there */
1.106 cvs 938: basename[length + 1] = EOS;
1.31 cvs 939: else
940: /* search for the first PATH_STR char */
941: {
1.106 cvs 942: for (length = 0; basename[length] != ':' &&
943: basename[length] != EOS; length ++);
944: if (basename[length] == ':')
1.31 cvs 945: /* found, so end the string there */
1.106 cvs 946: basename[length + 1] = EOS;
1.31 cvs 947: else
948: /* not found, discard the base */
1.106 cvs 949: basename[0] = EOS;
1.31 cvs 950: }
951: return (basename);
952: }
953:
954:
1.4 cvs 955: /*----------------------------------------------------------------------
1.40 cvs 956: GetLocalPath
957: Allocate and return the local document path associated to the url
958: ----------------------------------------------------------------------*/
1.106 cvs 959: char *GetLocalPath (Document doc, char *url)
960: {
961: char *ptr;
962: char *n;
963: char *documentname;
964: char url_sep;
1.83 cvs 965: int len;
1.67 cvs 966: ThotBool noFile;
1.40 cvs 967:
968: if (url != NULL)
969: {
970: /* check whether the file name exists */
1.106 cvs 971: len = strlen (url) - 1;
1.71 cvs 972: if (IsW3Path (url))
1.106 cvs 973: url_sep = '/';
1.41 cvs 974: else
1.106 cvs 975: url_sep = DIR_SEP;
1.41 cvs 976: noFile = (url[len] == url_sep);
1.40 cvs 977: if (noFile)
1.106 cvs 978: url[len] = EOS;
979: ptr = TtaGetMemory (MAX_LENGTH);
980: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 981: TtaExtractName (url, ptr, documentname);
1.106 cvs 982: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 983: if (!TtaCheckDirectory (ptr))
984: /* directory did not exist */
1.72 cvs 985: TtaMakeDirectory (ptr);
1.47 cvs 986:
987: /* don't include the query string within document name */
1.106 cvs 988: n = strrchr (documentname, '?');
1.47 cvs 989: if (n != NULL)
1.106 cvs 990: *n = EOS;
1.46 cvs 991: /* don't include ':' within document name */
1.106 cvs 992: n = strchr (documentname, ':');
1.46 cvs 993: if (n != NULL)
1.106 cvs 994: *n = EOS;
1.69 cvs 995: /* if after all this operations document name
996: is empty, let's use noname.html instead */
1.106 cvs 997: if (documentname[0] == EOS)
998: strcat (ptr, "noname.html");
1.69 cvs 999: else
1.106 cvs 1000: strcat (ptr, documentname);
1.40 cvs 1001: TtaFreeMemory (documentname);
1002: /* restore the url */
1003: if (noFile)
1.41 cvs 1004: url[len] = url_sep;
1.40 cvs 1005: return (ptr);
1006: }
1007: else
1008: return (NULL);
1009: }
1010:
1.73 cvs 1011: /*----------------------------------------------------------------------
1.79 cvs 1012: ExtractTarget extract the target name from document nane.
1013: ----------------------------------------------------------------------*/
1.106 cvs 1014: void ExtractTarget (char *aName, char *target)
1.79 cvs 1015: {
1.106 cvs 1016: int lg, i;
1017: char *ptr;
1018: char *oldptr;
1.79 cvs 1019:
1020: if (!target || !aName)
1021: /* bad target */
1022: return;
1023:
1.106 cvs 1024: target[0] = EOS;
1025: lg = strlen (aName);
1.79 cvs 1026: if (lg)
1027: {
1028: /* the name is not empty */
1029: oldptr = ptr = &aName[0];
1030: do
1031: {
1.106 cvs 1032: ptr = strrchr (oldptr, '#');
1.79 cvs 1033: if (ptr)
1034: oldptr = &ptr[1];
1035: }
1036: while (ptr);
1037:
1038: i = (int) (oldptr) - (int) (aName); /* name length */
1039: if (i > 1)
1040: {
1.106 cvs 1041: aName[i - 1] = EOS;
1.79 cvs 1042: if (i != lg)
1.106 cvs 1043: strcpy (target, oldptr);
1.79 cvs 1044: }
1045: }
1046: }
1047:
1048: /*----------------------------------------------------------------------
1.90 cvs 1049: RemoveNewLines (text)
1050: Removes any '\n' chars that are found in text.
1051: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1052: ----------------------------------------------------------------------*/
1.106 cvs 1053: ThotBool RemoveNewLines (char *text)
1054: {
1055: ThotBool change = FALSE;
1056: char *src;
1057: char *dest;
1.90 cvs 1058:
1059: src = text;
1060: dest = text;
1.115 kahan 1061:
1062: /* remove any preceding whitespace */
1063: while (*src && *src == ' ')
1064: {
1065: src++;
1066: change = 1;
1067: }
1068:
1.90 cvs 1069: while (*src)
1070: {
1071: switch (*src)
1072: {
1.106 cvs 1073: case '\n':
1.90 cvs 1074: /* don't copy the newline */
1075: change = 1;
1076: break;
1077: default:
1078: *dest = *src;
1079: dest++;
1080: break;
1081: }
1082: src++;
1083: }
1084: /* copy the last EOS char */
1085: *dest = *src;
1086:
1087: return (change);
1088: }
1089:
1090: /*----------------------------------------------------------------------
1091: CleanCopyFileURL
1092: Copies a file url from a src string to destination string.
1.97 cvs 1093: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1094: we want to do).
1.90 cvs 1095: ----------------------------------------------------------------------*/
1.106 cvs 1096: static void CleanCopyFileURL (char *dest, char *src,
1097: ConvertionType convertion)
1.90 cvs 1098: {
1099: while (*src)
1.89 cvs 1100: {
1.90 cvs 1101: switch (*src)
1.89 cvs 1102: {
1103: #ifdef _WINDOWS
1.106 cvs 1104: case URL_SEP:
1.96 cvs 1105: /* make DIR_SEP transformation */
1.97 cvs 1106: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1107: *dest = DIR_SEP;
1.96 cvs 1108: else
1109: *dest = *src;
1.90 cvs 1110: dest++;
1.96 cvs 1111: src++;
1.90 cvs 1112: break;
1.89 cvs 1113: #endif /* _WINDOWS */
1.96 cvs 1114:
1.106 cvs 1115: case '%':
1.97 cvs 1116: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1117: {
1.97 cvs 1118: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1119: src++;
1.106 cvs 1120: if (*src != EOS)
1.97 cvs 1121: {
1122: *dest = UnEscapeChar (*src) * 16;
1123: src++;
1124: }
1.106 cvs 1125: if (*src != EOS)
1.97 cvs 1126: {
1127: *dest = *dest + UnEscapeChar (*src);
1128: src++;
1129: }
1130: dest++;
1.96 cvs 1131: }
1.97 cvs 1132: else
1.96 cvs 1133: {
1.97 cvs 1134: *dest = *src;
1135: dest++;
1.96 cvs 1136: src++;
1137: }
1138: break;
1139:
1.90 cvs 1140: default:
1141: *dest = *src;
1.89 cvs 1142: dest++;
1.96 cvs 1143: src++;
1.90 cvs 1144: break;
1.89 cvs 1145: }
1146: }
1.90 cvs 1147: /* copy the EOS char */
1148: *dest = *src;
1.73 cvs 1149: }
1.40 cvs 1150:
1151: /*----------------------------------------------------------------------
1.9 cvs 1152: NormalizeURL
1153: normalizes orgName according to a base associated with doc, and
1154: following the standard URL format rules.
1.113 cvs 1155: if doc is < 0, use as a base the URL of the document that contains
1156: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1157: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1158: other path.
1.9 cvs 1159: The function returns the new complete and normalized URL
1.12 cvs 1160: or file name path (newName) and the name of the document (docName).
1.9 cvs 1161: N.B. If the function can't find out what's the docName, it assigns
1162: the name "noname.html".
1.4 cvs 1163: ----------------------------------------------------------------------*/
1.106 cvs 1164: void NormalizeURL (char *orgName, Document doc, char *newName,
1165: char *docName, char *otherPath)
1166: {
1167: char *basename;
1168: char tempOrgName[MAX_LENGTH];
1169: char *ptr;
1170: char used_sep;
1.84 cvs 1171: int length;
1172: ThotBool check;
1.5 cvs 1173:
1.110 cvs 1174: #ifdef _WINDOWS
1.44 cvs 1175: int ndx;
1.110 cvs 1176: #endif /* _WINDOWS */
1.44 cvs 1177:
1.5 cvs 1178: if (!newName || !docName)
1179: return;
1.18 cvs 1180:
1.113 cvs 1181: if (doc < 0)
1182: basename = TtaStrdup (SavedDocumentURL);
1183: else if (doc > 0)
1.53 cvs 1184: basename = GetBaseURL (doc);
1185: else if (otherPath != NULL)
1.108 cvs 1186: basename = TtaStrdup (otherPath);
1.32 cvs 1187: else
1.53 cvs 1188: basename = NULL;
1.32 cvs 1189:
1.18 cvs 1190: /*
1.31 cvs 1191: * Clean orgName
1192: * Make sure we have a complete orgName, without any leading or trailing
1193: * white spaces, or trailinbg new lines
1194: */
1.5 cvs 1195: ptr = orgName;
1.18 cvs 1196: /* skip leading white space and new line characters */
1.106 cvs 1197: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1198: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1199: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1200: /*
1.31 cvs 1201: * Make orgName a complete URL
1202: * If the URL does not include a protocol, then try to calculate
1203: * one using the doc's base element (if it exists),
1204: */
1.106 cvs 1205: if (tempOrgName[0] == EOS)
1.53 cvs 1206: {
1.106 cvs 1207: newName[0] = EOS;
1208: docName[0] = EOS;
1.53 cvs 1209: TtaFreeMemory (basename);
1210: return;
1211: }
1.49 cvs 1212:
1213: /* clean trailing white space */
1.106 cvs 1214: length = strlen (tempOrgName) - 1;
1215: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1216: {
1.106 cvs 1217: tempOrgName[length] = EOS;
1.53 cvs 1218: length--;
1219: }
1.50 cvs 1220:
1.55 cvs 1221: /* remove extra dot (which dot???) */
1222: /* ugly, but faster than a strcmp */
1.106 cvs 1223: if (tempOrgName[length] == '.'
1224: && (length == 0 || tempOrgName[length-1] != '.'))
1225: tempOrgName[length] = EOS;
1.50 cvs 1226:
1.94 cvs 1227: if (IsW3Path (tempOrgName))
1.53 cvs 1228: {
1229: /* the name is complete, go to the Sixth Step */
1.106 cvs 1230: strcpy (newName, tempOrgName);
1.53 cvs 1231: SimplifyUrl (&newName);
1232: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1233: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1234: AMAYA_PARSE_HOST |
1235: AMAYA_PARSE_PUNCTUATION);
1236: if (ptr && !strcmp (ptr, newName))
1237: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1238: strcat (newName, URL_STR);
1.49 cvs 1239:
1.53 cvs 1240: if (ptr)
1.50 cvs 1241: TtaFreeMemory (ptr);
1.53 cvs 1242: }
1.113 cvs 1243: else if (basename == NULL)
1.53 cvs 1244: /* the name is complete, go to the Sixth Step */
1.106 cvs 1245: strcpy (newName, tempOrgName);
1.53 cvs 1246: else
1247: {
1.31 cvs 1248: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1249: #ifdef _WINDOWS
1.53 cvs 1250: if (!IsW3Path (basename))
1251: {
1.106 cvs 1252: length = strlen (tempOrgName);
1.53 cvs 1253: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1254: if (tempOrgName [ndx] == '/')
1255: tempOrgName [ndx] = '\\';
1.53 cvs 1256: }
1.110 cvs 1257: #endif /* _WINDOWS */
1.25 cvs 1258: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1259: if (ptr)
1260: {
1261: SimplifyUrl (&ptr);
1.106 cvs 1262: strcpy (newName, ptr);
1.53 cvs 1263: TtaFreeMemory (ptr);
1264: }
1265: else
1.106 cvs 1266: newName[0] = EOS;
1.53 cvs 1267: }
1.36 cvs 1268:
1269: TtaFreeMemory (basename);
1.18 cvs 1270: /*
1.31 cvs 1271: * Prepare the docname that will refer to this ressource in the
1272: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1273: * noname.html as a default ressource name
1.18 cvs 1274: */
1.106 cvs 1275: if (newName[0] != EOS)
1.53 cvs 1276: {
1.106 cvs 1277: length = strlen (newName) - 1;
1278: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1279: {
1280: used_sep = newName[length];
1281: check = TRUE;
1282: while (check)
1283: {
1.50 cvs 1284: length--;
1285: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1286: length--;
1.106 cvs 1287: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1288: {
1.106 cvs 1289: newName[length+1] = EOS;
1.53 cvs 1290: /* remove also previous directory */
1291: length--;
1292: while (length >= 0 && newName[length] != used_sep)
1293: length--;
1.106 cvs 1294: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1295: /* don't remove server name */
1.106 cvs 1296: newName[length+1] = EOS;
1.53 cvs 1297: }
1.106 cvs 1298: else if (!strncmp (&newName[length+1], ".", 1))
1299: newName[length+1] = EOS;
1.50 cvs 1300: else
1.53 cvs 1301: check = FALSE;
1302: }
1303: /* docname was not comprised inside the URL, so let's */
1304: /* assign the default ressource name */
1.106 cvs 1305: strcpy (docName, "noname.html");
1.53 cvs 1306: }
1307: else
1308: { /* docname is comprised inside the URL */
1.110 cvs 1309: while (length >= 0 && newName[length] != URL_SEP &&
1310: newName[length] != DIR_SEP)
1.53 cvs 1311: length--;
1312: if (length < 0)
1.106 cvs 1313: strcpy (docName, newName);
1.53 cvs 1314: else
1.106 cvs 1315: strcpy (docName, &newName[length+1]);
1.53 cvs 1316: }
1317: }
1318: else
1.106 cvs 1319: docName[0] = EOS;
1.18 cvs 1320: }
1.3 cvs 1321:
1.4 cvs 1322: /*----------------------------------------------------------------------
1.9 cvs 1323: IsSameHost
1.4 cvs 1324: ----------------------------------------------------------------------*/
1.106 cvs 1325: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1326: {
1.106 cvs 1327: char *basename_ptr1, *basename_ptr2;
1328: ThotBool result;
1.3 cvs 1329:
1.106 cvs 1330: basename_ptr1 = AmayaParseUrl (url1, "",
1331: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1332: basename_ptr2 = AmayaParseUrl (url2, "",
1333: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1334:
1.106 cvs 1335: if (strcmp (basename_ptr1, basename_ptr2))
1336: result = FALSE;
1337: else
1338: result = TRUE;
1339: TtaFreeMemory (basename_ptr1);
1340: TtaFreeMemory (basename_ptr2);
1341: return (result);
1.3 cvs 1342: }
1343:
1344:
1.4 cvs 1345: /*----------------------------------------------------------------------
1.22 cvs 1346: HasKnownFileSuffix
1347: returns TRUE if path points to a file ending with a suffix.
1348: ----------------------------------------------------------------------*/
1.106 cvs 1349: ThotBool HasKnownFileSuffix (const char *path)
1350: {
1351: char *root;
1352: char temppath[MAX_LENGTH];
1353: char suffix[MAX_LENGTH];
1.22 cvs 1354:
1.106 cvs 1355: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1356: return (FALSE);
1357:
1.106 cvs 1358: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1359:
1360: if (root)
1361: {
1.106 cvs 1362: strcpy (temppath, root);
1.25 cvs 1363: TtaFreeMemory (root);
1.22 cvs 1364: /* Get the suffix */
1.124 vatton 1365: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1366:
1.106 cvs 1367: if( suffix[0] == EOS)
1.22 cvs 1368: /* no suffix */
1369: return (FALSE);
1370:
1371: /* Normalize the suffix */
1372: ConvertToLowerCase (suffix);
1373:
1.106 cvs 1374: if (!strcmp (suffix, "gz"))
1.22 cvs 1375: /* skip the compressed suffix */
1376: {
1.124 vatton 1377: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1378: if(suffix[0] == EOS)
1.22 cvs 1379: /* no suffix */
1380: return (FALSE);
1381: /* Normalize the suffix */
1382: ConvertToLowerCase (suffix);
1383: }
1384:
1.106 cvs 1385: if (strcmp (suffix, "gif") &&
1386: strcmp (suffix, "xbm") &&
1387: strcmp (suffix, "xpm") &&
1388: strcmp (suffix, "jpg") &&
1389: strcmp (suffix, "pdf") &&
1390: strcmp (suffix, "png") &&
1391: strcmp (suffix, "tgz") &&
1392: strcmp (suffix, "xpg") &&
1393: strcmp (suffix, "xpd") &&
1394: strcmp (suffix, "ps") &&
1395: strcmp (suffix, "au") &&
1396: strcmp (suffix, "html") &&
1397: strcmp (suffix, "htm") &&
1398: strcmp (suffix, "shtml") &&
1399: strcmp (suffix, "xht") &&
1400: strcmp (suffix, "xhtm") &&
1401: strcmp (suffix, "xhtml") &&
1402: strcmp (suffix, "txt") &&
1403: strcmp (suffix, "css") &&
1404: strcmp (suffix, "eps"))
1.22 cvs 1405: return (FALSE);
1406: else
1407: return (TRUE);
1408: }
1409: else
1410: return (FALSE);
1411: }
1412:
1413:
1414: /*----------------------------------------------------------------------
1.24 cvs 1415: ChopURL
1416: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1417: If inputURL is bigger than that size, outputURL receives
1418: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1419: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1420: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1421: copied into outputURL.
1422: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1423: chars.
1424: ----------------------------------------------------------------------*/
1.106 cvs 1425: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1426: {
1427: int len;
1.9 cvs 1428:
1.106 cvs 1429: len = strlen (inputURL);
1.24 cvs 1430: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1431: strcpy (outputURL, inputURL);
1.24 cvs 1432: else
1433: /* make a truncated urlName on the status window */
1434: {
1.106 cvs 1435: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1436: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1437: strcat (outputURL, "...");
1438: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1439: }
1.25 cvs 1440: }
1441:
1442:
1443: /*----------------------------------------------------------------------
1444: scan
1.47 cvs 1445: Scan a filename for its constituents
1.25 cvs 1446: -----------------------------------
1447:
1448: On entry,
1449: name points to a document name which may be incomplete.
1450: On exit,
1451: absolute or relative may be nonzero (but not both).
1452: host, fragment and access may be nonzero if they were specified.
1453: Any which are nonzero point to zero terminated strings.
1454: ----------------------------------------------------------------------*/
1.106 cvs 1455: static void scan (char *name, HTURI *parts)
1.25 cvs 1456: {
1.106 cvs 1457: char * p;
1458: char * after_access = name;
1.32 cvs 1459:
1.43 cvs 1460: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1461: /* Look for fragment identifier */
1.106 cvs 1462: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1463: {
1.106 cvs 1464: *p++ = '\0';
1.28 cvs 1465: parts->fragment = p;
1.25 cvs 1466: }
1467:
1.28 cvs 1468: for (p=name; *p; p++)
1469: {
1.106 cvs 1470: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1471: break;
1.106 cvs 1472: if (*p == ':')
1.28 cvs 1473: {
1474: *p = 0;
1475: parts->access = after_access; /* Scheme has been specified */
1476:
1477: /* The combination of gcc, the "-O" flag and the HP platform is
1478: unhealthy. The following three lines is a quick & dirty fix, but is
1479: not recommended. Rather, turn off "-O". */
1480:
1481: /* after_access = p;*/
1482: /* while (*after_access == 0)*/
1483: /* after_access++;*/
1484: after_access = p+1;
1.106 cvs 1485: if (!strcasecmp("URL", parts->access))
1.28 cvs 1486: /* Ignore IETF's URL: pre-prefix */
1487: parts->access = NULL;
1488: else
1.25 cvs 1489: break;
1490: }
1491: }
1492:
1493: p = after_access;
1.43 cvs 1494: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1495: {
1.43 cvs 1496: if (p[1] == URL_SEP)
1.28 cvs 1497: {
1.25 cvs 1498: parts->host = p+2; /* host has been specified */
1.28 cvs 1499: *p = 0; /* Terminate access */
1500: /* look for end of host name if any */
1.106 cvs 1501: p = strchr (parts->host, URL_SEP);
1.28 cvs 1502: if (p)
1503: {
1.106 cvs 1504: *p = EOS; /* Terminate host */
1.25 cvs 1505: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1506: }
1507: }
1508: else
1509: /* Root found but no host */
1510: parts->absolute = p+1;
1511: }
1512: else
1513: {
1.25 cvs 1514: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1515: }
1.25 cvs 1516: }
1517:
1518:
1519: /*----------------------------------------------------------------------
1.28 cvs 1520: AmayaParseUrl: parse a Name relative to another name
1521:
1522: This returns those parts of a name which are given (and requested)
1523: substituting bits from the related name where necessary.
1.25 cvs 1524:
1.28 cvs 1525: On entry,
1.25 cvs 1526: aName A filename given
1527: relatedName A name relative to which aName is to be parsed. Give
1528: it an empty string if aName is absolute.
1529: wanted A mask for the bits which are wanted.
1530:
1.28 cvs 1531: On exit,
1.25 cvs 1532: returns A pointer to a malloc'd string which MUST BE FREED
1533: ----------------------------------------------------------------------*/
1.106 cvs 1534: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1535: {
1536: char *return_value;
1537: char result[MAX_LENGTH];
1538: char name[MAX_LENGTH];
1539: char rel[MAX_LENGTH];
1540: char *p, *access;
1.29 cvs 1541: HTURI given, related;
1542: int len;
1.106 cvs 1543: char used_sep;
1544: char *used_str;
1.32 cvs 1545:
1.106 cvs 1546: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1547: {
1.106 cvs 1548: used_str = DIR_STR;
1549: used_sep = DIR_SEP;
1.33 cvs 1550: }
1.32 cvs 1551: else
1.33 cvs 1552: {
1.106 cvs 1553: used_str = URL_STR;
1554: used_sep = URL_SEP;
1.33 cvs 1555: }
1.32 cvs 1556:
1.29 cvs 1557: /* Make working copies of input strings to cut up: */
1558: return_value = NULL;
1559: result[0] = 0; /* Clear string */
1.106 cvs 1560: strcpy (name, aName);
1.29 cvs 1561: if (relatedName != NULL)
1.106 cvs 1562: strcpy (rel, relatedName);
1.29 cvs 1563: else
1.106 cvs 1564: relatedName[0] = EOS;
1.29 cvs 1565:
1566: scan (name, &given);
1567: scan (rel, &related);
1568: access = given.access ? given.access : related.access;
1569: if (wanted & AMAYA_PARSE_ACCESS)
1570: if (access)
1571: {
1.106 cvs 1572: strcat (result, access);
1.29 cvs 1573: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1574: strcat (result, ":");
1.29 cvs 1575: }
1576:
1577: if (given.access && related.access)
1578: /* If different, inherit nothing. */
1.106 cvs 1579: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1580: {
1581: related.host = 0;
1582: related.absolute = 0;
1583: related.relative = 0;
1584: related.fragment = 0;
1585: }
1586:
1587: if (wanted & AMAYA_PARSE_HOST)
1588: if(given.host || related.host)
1589: {
1590: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1591: strcat (result, "//");
1592: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1593: }
1594:
1595: if (given.host && related.host)
1596: /* If different hosts, inherit no path. */
1.106 cvs 1597: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1598: {
1599: related.absolute = 0;
1600: related.relative = 0;
1601: related.fragment = 0;
1602: }
1603:
1604: if (wanted & AMAYA_PARSE_PATH)
1605: {
1606: if (given.absolute)
1607: {
1608: /* All is given */
1609: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1610: strcat (result, used_str);
1611: strcat (result, given.absolute);
1.25 cvs 1612: }
1.29 cvs 1613: else if (related.absolute)
1614: {
1615: /* Adopt path not name */
1.106 cvs 1616: strcat (result, used_str);
1617: strcat (result, related.absolute);
1.29 cvs 1618: if (given.relative)
1619: {
1620: /* Search part? */
1.106 cvs 1621: p = strchr (result, '?');
1.29 cvs 1622: if (!p)
1.106 cvs 1623: p=result+strlen(result)-1;
1.33 cvs 1624: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1625: /* Remove filename */
1626: p[1]=0;
1627: /* Add given one */
1.106 cvs 1628: strcat (result, given.relative);
1.25 cvs 1629: }
1630: }
1.29 cvs 1631: else if (given.relative)
1632: /* what we've got */
1.106 cvs 1633: strcat (result, given.relative);
1.29 cvs 1634: else if (related.relative)
1.106 cvs 1635: strcat (result, related.relative);
1.29 cvs 1636: else
1637: /* No inheritance */
1.106 cvs 1638: strcat (result, used_str);
1.25 cvs 1639: }
1.29 cvs 1640:
1641: if (wanted & AMAYA_PARSE_ANCHOR)
1642: if (given.fragment || related.fragment)
1643: {
1644: if (given.absolute && given.fragment)
1645: {
1646: /*Fixes for relURLs...*/
1647: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1648: strcat (result, "#");
1649: strcat (result, given.fragment);
1.29 cvs 1650: }
1651: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1652: strcat (result, "");
1.29 cvs 1653: else
1654: {
1.110 cvs 1655: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1656: strcat (result, "#");
1.110 cvs 1657: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1658: }
1659: }
1.106 cvs 1660: len = strlen (result);
1661: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1662: strcpy (return_value, result);
1.29 cvs 1663: return (return_value); /* exactly the right length */
1.25 cvs 1664: }
1665:
1666: /*----------------------------------------------------------------------
1667: HTCanon
1668: Canonicalizes the URL in the following manner starting from the host
1669: pointer:
1670:
1671: 1) The host name is converted to lowercase
1672: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1673:
1674: Return: OK The position of the current path part of the URL
1675: which might be the old one or a new one.
1676:
1677: ----------------------------------------------------------------------*/
1.106 cvs 1678: static char *HTCanon (char **filename, char *host)
1679: {
1680: char *newname = NULL;
1681: char used_sep;
1682: char *path;
1683: char *strptr;
1684: char *port;
1685: char *access = host-3;
1686:
1687: if (*filename && strchr (*filename, URL_SEP))
1688: used_sep = URL_SEP;
1.33 cvs 1689: else
1.106 cvs 1690: used_sep = DIR_SEP;
1.32 cvs 1691:
1.110 cvs 1692: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1693: access--;
1.110 cvs 1694: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1695: path = host + strlen (host);
1696: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1697: host = strptr;
1.110 cvs 1698: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1699: port = NULL;
1.25 cvs 1700:
1701: strptr = host; /* Convert to lower-case */
1.82 cvs 1702: while (strptr < path)
1.33 cvs 1703: {
1.123 vatton 1704: *strptr = tolower (*strptr);
1.82 cvs 1705: strptr++;
1.33 cvs 1706: }
1.25 cvs 1707:
1708: /* Does the URL contain a full domain name? This also works for a
1709: numerical host name. The domain name is already made lower-case
1710: and without a trailing dot. */
1711: {
1.106 cvs 1712: char *dot = port ? port : path;
1713: if (dot > *filename && *--dot == '.')
1.33 cvs 1714: {
1.106 cvs 1715: char *orig = dot;
1716: char *dest = dot + 1;
1.82 cvs 1717: while ((*orig++ = *dest++));
1718: if (port) port--;
1.33 cvs 1719: path--;
1.25 cvs 1720: }
1721: }
1722: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1723: if (port)
1724: {
1.82 cvs 1725: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1726: {
1727: if (!newname)
1728: {
1.106 cvs 1729: char *orig = port;
1730: char *dest = port + 1;
1.82 cvs 1731: while ((*orig++ = *dest++));
1.33 cvs 1732: }
1733: }
1.106 cvs 1734: else if ((!strncmp (access, "http", 4) &&
1735: (*(port + 1) == '8' &&
1736: *(port+2) == '0' &&
1.82 cvs 1737: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1738: (!strncmp (access, "gopher", 6) &&
1739: (*(port+1) == '7' &&
1740: *(port+2) == '0' &&
1.82 cvs 1741: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1742: (!strncmp (access, "ftp", 3) &&
1743: (*(port+1) == '2' &&
1744: *(port + 2) == '1' &&
1.82 cvs 1745: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1746: if (!newname)
1747: {
1.106 cvs 1748: char *orig = port;
1749: char *dest = port + 3;
1.33 cvs 1750: while((*orig++ = *dest++));
1751: /* Update path position, Henry Minsky */
1752: path -= 3;
1.25 cvs 1753: }
1.33 cvs 1754: }
1755: else if (newname)
1.106 cvs 1756: strncat (newname, port, (int) (path - port));
1.33 cvs 1757: }
1.25 cvs 1758:
1.33 cvs 1759: if (newname)
1760: {
1.106 cvs 1761: char *newpath = newname + strlen (newname);
1762: strcat (newname, path);
1.25 cvs 1763: path = newpath;
1.28 cvs 1764: /* Free old copy */
1765: TtaFreeMemory(*filename);
1.25 cvs 1766: *filename = newname;
1.33 cvs 1767: }
1.25 cvs 1768: return path;
1769: }
1770:
1771:
1772: /*----------------------------------------------------------------------
1.29 cvs 1773: SimplifyUrl: simplify a URI
1.32 cvs 1774: A URI is allowed to contain the sequence xxx/../ which may be
1775: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1776: Simplification helps us recognize duplicate URIs.
1.25 cvs 1777:
1.28 cvs 1778: Thus, /etc/junk/../fred becomes /etc/fred
1779: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1780:
1.28 cvs 1781: but we should NOT change
1782: http://fred.xxx.edu/../..
1.25 cvs 1783:
1784: or ../../albert.html
1785:
1.28 cvs 1786: In order to avoid empty URLs the following URLs become:
1.25 cvs 1787:
1788: /fred/.. becomes /fred/..
1789: /fred/././.. becomes /fred/..
1790: /fred/.././junk/.././ becomes /fred/..
1791:
1.28 cvs 1792: If more than one set of `://' is found (several proxies in cascade) then
1793: only the part after the last `://' is simplified.
1.25 cvs 1794:
1.28 cvs 1795: Returns: A string which might be the old one or a new one.
1.25 cvs 1796: ----------------------------------------------------------------------*/
1.106 cvs 1797: void SimplifyUrl (char **url)
1798: {
1799: char *path;
1800: char *access;
1801: char *newptr;
1802: char *p;
1803: char *orig, *dest, *end;
1.28 cvs 1804:
1.106 cvs 1805: char used_sep;
1.77 cvs 1806: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1807: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1808:
1.28 cvs 1809: if (!url || !*url)
1810: return;
1811:
1.106 cvs 1812: if (strchr (*url, URL_SEP))
1813: used_sep = URL_SEP;
1.32 cvs 1814: else
1.106 cvs 1815: used_sep = DIR_SEP;
1.32 cvs 1816:
1.77 cvs 1817: /* should we simplify double dot? */
1818: path = *url;
1.106 cvs 1819: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1820: ddot_simplify = FALSE;
1821: else
1822: ddot_simplify = TRUE;
1823:
1.28 cvs 1824: /* Find any scheme name */
1.106 cvs 1825: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1826: {
1827: /* Find host name */
1.28 cvs 1828: access = *url;
1.123 vatton 1829: while (access < path && (*access = tolower (*access)))
1.82 cvs 1830: access++;
1.28 cvs 1831: path += 3;
1.106 cvs 1832: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1833: /* For proxies */
1.106 cvs 1834: path = newptr + 3;
1.82 cvs 1835: /* We have a host name */
1.84 cvs 1836: path = HTCanon (url, path);
1.25 cvs 1837: }
1.106 cvs 1838: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1839: path += 2;
1840: else
1841: path = *url;
1.84 cvs 1842: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1843: /* Some URLs start //<foo> */
1844: path += 1;
1.94 cvs 1845: else if (IsFilePath (path))
1846: {
1847: /* doesn't need to do anything more */
1848: return;
1849: }
1.106 cvs 1850: else if (!strncmp (path, "news:", 5))
1.28 cvs 1851: {
1.106 cvs 1852: newptr = strchr (path+5, '@');
1.28 cvs 1853: if (!newptr)
1854: newptr = path + 5;
1855: while (*newptr)
1856: {
1857: /* Make group or host lower case */
1.123 vatton 1858: *newptr = tolower (*newptr);
1.28 cvs 1859: newptr++;
1.25 cvs 1860: }
1.28 cvs 1861: /* Doesn't need to do any more */
1862: return;
1.25 cvs 1863: }
1.130 cheyroul 1864:
1.126 cheyroul 1865:
1.28 cvs 1866: if ((p = path))
1867: {
1.106 cvs 1868: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1869: (end = strchr (path, '#'))))
1870: end = path + strlen (path);
1.28 cvs 1871:
1872: /* Parse string second time to simplify */
1873: p = path;
1874: while (p < end)
1875: {
1.110 cvs 1876: /* if we're pointing to a char, it's safe to reactivate the
1877: ../ convertion */
1.106 cvs 1878: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1879: ddot_simplify = TRUE;
1880:
1.33 cvs 1881: if (*p==used_sep)
1.28 cvs 1882: {
1.106 cvs 1883: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1884: {
1885: orig = p + 1;
1.84 cvs 1886: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1887: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1888: end = orig - 1;
1889: }
1.106 cvs 1890: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1891: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1892: {
1893: newptr = p;
1.52 cvs 1894: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1895: if (*newptr == used_sep)
1896: orig = newptr + 1;
1.28 cvs 1897: else
1.52 cvs 1898: orig = newptr;
1899:
1900: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1901: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1902: end = orig-1;
1903: /* Start again with prev slash */
1904: p = newptr;
1.28 cvs 1905: }
1.33 cvs 1906: else if (*(p+1) == used_sep)
1.28 cvs 1907: {
1.33 cvs 1908: while (*(p+1) == used_sep)
1.28 cvs 1909: {
1910: orig = p;
1911: dest = p + 1;
1912: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1913: end = orig-1;
1914: }
1915: }
1916: else
1.25 cvs 1917: p++;
1.28 cvs 1918: }
1919: else
1920: p++;
1.25 cvs 1921: }
1922: }
1.51 cvs 1923: /*
1924: ** Check for host/../.. kind of things
1925: */
1.106 cvs 1926: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1927: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 1928: *(path+1) = EOS;
1.28 cvs 1929: return;
1930: }
1931:
1932:
1933: /*----------------------------------------------------------------------
1.96 cvs 1934: NormalizeFile normalizes local names.
1.28 cvs 1935: Return TRUE if target and src differ.
1936: ----------------------------------------------------------------------*/
1.106 cvs 1937: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 1938: {
1.110 cvs 1939: #ifndef _WINDOWS
1.106 cvs 1940: char *s;
1.93 cvs 1941: int i;
1.110 cvs 1942: #endif /* !_WINDOWS */
1.82 cvs 1943: ThotBool change;
1.90 cvs 1944: int start_index; /* the first char that we'll copy */
1.28 cvs 1945:
1.54 cvs 1946: change = FALSE;
1.90 cvs 1947: start_index = 0;
1948:
1.106 cvs 1949: if (!src || src[0] == EOS)
1.96 cvs 1950: {
1.106 cvs 1951: target[0] = EOS;
1.96 cvs 1952: return FALSE;
1953: }
1.90 cvs 1954:
1955: /* @@ do I need file: or file:/ here? */
1.106 cvs 1956: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1957: {
1.90 cvs 1958: /* remove the prefix file: */
1959: start_index += 5;
1960:
1961: /* remove the localhost prefix */
1.106 cvs 1962: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 1963: start_index += 11;
1964:
1965: /* remove the first two slashes in / / /path */
1966: while (src[start_index] &&
1.106 cvs 1967: src[start_index] == '/'
1968: && src[start_index + 1] == '/')
1.94 cvs 1969: start_index++;
1970:
1971: #ifdef _WINDOWS
1972: /* remove any extra slash before the drive name */
1.106 cvs 1973: if (src[start_index] == '/'
1974: &&src[start_index+2] == ':')
1.94 cvs 1975: start_index++;
1976: #endif /* _WINDOWS */
1.90 cvs 1977:
1.106 cvs 1978: if (src[start_index] == EOS)
1.90 cvs 1979: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 1980: strcpy (target, DIR_STR);
1.90 cvs 1981: else
1.97 cvs 1982: /* as we're inside a file: URL, we'll apply all the convertions
1983: we know */
1984: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1985:
1986: change = TRUE;
1987: }
1.97 cvs 1988: else if (convertion != AM_CONV_NONE)
1.96 cvs 1989: {
1990: /* we are following a "local" relative link, we do all the
1991: convertions except for the HOME_DIR ~ one */
1.97 cvs 1992: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1993: }
1.90 cvs 1994: #ifndef _WINDOWS
1.106 cvs 1995: else if (src[0] == '~')
1.53 cvs 1996: {
1.96 cvs 1997: /* it must be a URL typed in a text input field */
1998: /* do the HOME_DIR ~ substitution */
1.82 cvs 1999: s = TtaGetEnvString ("HOME");
1.106 cvs 2000: strcpy (target, s);
1.90 cvs 2001: #if 0
1.96 cvs 2002: /* JK: invalidated this part of the code as it's simpler
2003: to add the DIR_SEP whenever we have something to add
2004: to the path rather than adding it systematically */
1.106 cvs 2005: if (src[1] != DIR_SEP)
2006: strcat (target, DIR_STR);
1.90 cvs 2007: #endif
1.106 cvs 2008: i = strlen (target);
2009: strcpy (&target[i], &src[1]);
1.54 cvs 2010: change = TRUE;
1.53 cvs 2011: }
1.90 cvs 2012: #endif /* _WINDOWS */
1.28 cvs 2013: else
1.96 cvs 2014: /* leave it as it is */
1.106 cvs 2015: strcpy (target, src);
1.96 cvs 2016:
1.28 cvs 2017: /* remove /../ and /./ */
1.29 cvs 2018: SimplifyUrl (&target);
1.54 cvs 2019: if (!change)
1.106 cvs 2020: change = strcmp (src, target);
1.28 cvs 2021: return (change);
1.25 cvs 2022: }
2023:
1.28 cvs 2024:
1.25 cvs 2025: /*----------------------------------------------------------------------
1.31 cvs 2026: MakeRelativeURL: make relative name
1.25 cvs 2027:
1.28 cvs 2028: This function creates and returns a string which gives an expression of
2029: one address as related to another. Where there is no relation, an absolute
2030: address is retured.
1.25 cvs 2031:
1.28 cvs 2032: On entry,
1.25 cvs 2033: Both names must be absolute, fully qualified names of nodes
2034: (no fragment bits)
2035:
1.28 cvs 2036: On exit,
1.25 cvs 2037: The return result points to a newly allocated name which, if
2038: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2039: The caller is responsible for freeing the resulting name later.
2040: ----------------------------------------------------------------------*/
1.106 cvs 2041: char *MakeRelativeURL (char *aName, char *relatedName)
2042: {
2043: char *return_value;
2044: char result[MAX_LENGTH];
2045: char *p;
2046: char *q;
2047: char *after_access;
2048: char *last_slash = NULL;
2049: int slashes, levels, len;
1.110 cvs 2050: #ifdef _WINDOWS
1.44 cvs 2051: int ndx;
1.110 cvs 2052: #endif /* _WINDOWS */
1.44 cvs 2053:
1.29 cvs 2054: if (aName == NULL || relatedName == NULL)
2055: return (NULL);
2056:
2057: slashes = 0;
2058: after_access = NULL;
2059: p = aName;
2060: q = relatedName;
2061: for (; *p && (*p == *q); p++, q++)
1.27 cvs 2062: {
2063: /* Find extent of match */
1.106 cvs 2064: if (*p == ':')
1.29 cvs 2065: after_access = p + 1;
1.28 cvs 2066: if (*p == DIR_SEP)
1.27 cvs 2067: {
1.29 cvs 2068: /* memorize the last slash position and count them */
1.27 cvs 2069: last_slash = p;
2070: slashes++;
1.25 cvs 2071: }
2072: }
2073:
1.31 cvs 2074: /* q, p point to the first non-matching character or zero */
1.106 cvs 2075: if (*q == EOS)
1.31 cvs 2076: {
2077: /* New name is a subset of the related name */
2078: /* exactly the right length */
1.106 cvs 2079: len = strlen (p);
2080: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2081: strcpy (return_value, p);
1.31 cvs 2082: }
2083: else if ((slashes < 2 && after_access == NULL)
2084: || (slashes < 3 && after_access != NULL))
2085: {
2086: /* Two names whitout common path */
2087: /* exactly the right length */
1.106 cvs 2088: len = strlen (aName);
2089: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2090: strcpy (return_value, aName);
1.31 cvs 2091: }
2092: else
2093: {
2094: /* Some path in common */
1.106 cvs 2095: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2096: /* just the same server */
1.106 cvs 2097: strcpy (result, last_slash);
1.31 cvs 2098: else
2099: {
2100: levels= 0;
1.106 cvs 2101: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2102: if (*q == DIR_SEP)
2103: levels++;
2104:
1.106 cvs 2105: result[0] = EOS;
1.31 cvs 2106: for (;levels; levels--)
1.106 cvs 2107: strcat (result, "../");
2108: strcat (result, last_slash+1);
1.31 cvs 2109: }
1.52 cvs 2110:
2111: if (!*result)
1.106 cvs 2112: strcat (result, "./");
1.52 cvs 2113:
1.31 cvs 2114: /* exactly the right length */
1.106 cvs 2115: len = strlen (result);
2116: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2117: strcpy (return_value, result);
1.52 cvs 2118:
1.25 cvs 2119: }
1.110 cvs 2120: #ifdef _WINDOWS
1.106 cvs 2121: len = strlen (return_value);
1.44 cvs 2122: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2123: if (return_value[ndx] == '\\')
2124: return_value[ndx] = '/' ;
1.110 cvs 2125: #endif /* _WINDOWS */
1.29 cvs 2126: return (return_value);
1.24 cvs 2127: }
1.35 cvs 2128:
1.104 kahan 2129: /*----------------------------------------------------------------------
2130: AM_GetFileSize
2131: Returns TRUE and the filesize in the 2nd parameter.
2132: Otherwise, in case of a system error, returns FALSE, with a
2133: filesize of 0L.
2134: ---------------------------------------------------------------------*/
1.106 cvs 2135: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2136: {
1.106 cvs 2137: ThotFileHandle handle = ThotFile_BADHANDLE;
2138: ThotFileInfo info;
1.35 cvs 2139:
1.104 kahan 2140: *file_size = 0L;
2141: if (!TtaFileExist (filename))
2142: return FALSE;
2143:
2144: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2145: if (handle == ThotFile_BADHANDLE)
2146: /* ThotFile_BADHANDLE */
2147: return FALSE;
2148: if (TtaFileStat (handle, &info) == 0)
2149: /* bad stat */
2150: info.size = 0L;
2151: TtaFileClose (handle);
2152: *file_size = (unsigned long) info.size;
2153: return TRUE;
2154: }
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