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