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