Annotation of Amaya/amaya/AHTURLTools.c, revision 1.33
1.7 cvs 1: /*
2: *
3: * (c) COPYRIGHT MIT and INRIA, 1996.
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
14: *
15: */
1.7 cvs 16:
1.15 cvs 17: #define THOT_EXPORT extern
1.3 cvs 18: #include "amaya.h"
19:
1.8 cvs 20: #include "init_f.h"
21: #include "AHTURLTools_f.h"
22:
1.24 cvs 23: #define MAX_PRINT_URL_LENGTH 50
1.29 cvs 24: typedef struct _HTURI {
25: char * access; /* Now known as "scheme" */
26: char * host;
27: char * absolute;
28: char * relative;
29: char * fragment;
30: } HTURI;
1.24 cvs 31:
1.28 cvs 32:
33: /*----------------------------------------------------------------------
34: ConvertToLowerCase
35: Converts a string to lowercase.
36: ----------------------------------------------------------------------*/
1.22 cvs 37: #ifdef __STDC__
1.28 cvs 38: static void ConvertToLowerCase (char *string)
39: #else /* __STDC__ */
40: static void ConvertToLowerCase (string)
41: char *string;
42:
43: #endif /* __STDC__ */
44: {
45: int i;
46:
47: if (!string)
48: return;
49:
50: for (i = 0; string[i] != EOS; i++)
51: string[i] = tolower (string[i]);
52: }
1.22 cvs 53:
1.8 cvs 54: /*----------------------------------------------------------------------
1.11 cvs 55: ExplodeURL
1.8 cvs 56: ----------------------------------------------------------------------*/
57: #ifdef __STDC__
58: void ExplodeURL (char *url, char **proto, char **host, char **dir, char **file)
59: #else
60: void ExplodeURL (url, proto, host, dir, file)
61: char *url;
62: char **proto;
63: char **host;
64: char **dir;
65: char **file;
66:
67: #endif
68: {
1.33 ! cvs 69: char *curr, *temp;
! 70: char used_sep;
1.32 cvs 71:
1.33 ! cvs 72: if (url && strchr (url, URL_SEP))
! 73: used_sep = URL_SEP;
! 74: else
! 75: used_sep = DIR_SEP;
1.8 cvs 76:
77: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
78: (dir == NULL) || (file == NULL))
79: return;
80:
81: /* initialize every pointer */
82: *proto = *host = *dir = *file = NULL;
83:
84: /* skip any leading space */
85: while ((*url == SPACE) || (*url == TAB))
86: url++;
1.9 cvs 87: curr = url;
88: if (*curr == 0)
1.8 cvs 89: goto finished;
90:
91: /* go to the end of the URL */
1.9 cvs 92: while ((*curr != 0) && (*curr != SPACE) && (*curr != '\b') &&
93: (*curr != '\r') && (*curr != EOL))
94: curr++;
1.8 cvs 95:
96: /* mark the end of the chain */
1.9 cvs 97: *curr = EOS;
98: curr--;
99: if (curr <= url)
1.8 cvs 100: goto finished;
101:
102: /* search the next DIR_SEP indicating the beginning of the file name */
103: do
1.11 cvs 104: curr--;
1.33 ! cvs 105: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 106:
1.9 cvs 107: if (curr < url)
1.8 cvs 108: goto finished;
1.9 cvs 109: *file = curr + 1;
1.8 cvs 110:
111: /* mark the end of the dir */
1.9 cvs 112: *curr = EOS;
113: curr--;
114: if (curr < url)
1.8 cvs 115: goto finished;
116:
1.29 cvs 117: /* search for the DIR_STR indicating the host name start */
1.33 ! cvs 118: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 119: curr--;
1.8 cvs 120:
121: /* if we found it, separate the host name from the directory */
1.33 ! cvs 122: if ((*curr == DIR_SEP) && (*(curr + 1) == used_sep))
1.8 cvs 123: {
1.9 cvs 124: *host = temp = curr + 2;
1.33 ! cvs 125: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 126: temp++;
1.33 ! cvs 127: if (*temp == used_sep)
1.8 cvs 128: {
129: *temp = EOS;
130: *dir = temp + 1;
131: }
132: }
133: else
1.11 cvs 134: *dir = curr;
135:
1.9 cvs 136: if (curr <= url)
1.8 cvs 137: goto finished;
138:
139: /* mark the end of the proto */
1.9 cvs 140: *curr = EOS;
141: curr--;
142: if (curr < url)
1.8 cvs 143: goto finished;
144:
1.32 cvs 145: if (*curr == ':')
1.8 cvs 146: {
1.9 cvs 147: *curr = EOS;
148: curr--;
1.8 cvs 149: }
150: else
151: goto finished;
1.11 cvs 152:
1.9 cvs 153: if (curr < url)
1.8 cvs 154: goto finished;
1.9 cvs 155: while ((curr > url) && (isalpha (*curr)))
156: curr--;
157: *proto = curr;
1.8 cvs 158:
159: finished:;
160:
161: #ifdef AMAYA_DEBUG
162: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
163: if (*proto)
164: fprintf (stderr, "proto : %s, ", *proto);
165: if (*host)
166: fprintf (stderr, "host : %s, ", *host);
167: if (*dir)
168: fprintf (stderr, "dir : %s, ", *dir);
169: if (*file)
170: fprintf (stderr, "file : %s ", *file);
171: fprintf (stderr, "\n");
172: #endif
173:
174: }
1.3 cvs 175:
1.4 cvs 176: /*----------------------------------------------------------------------
1.9 cvs 177: IsHTMLName
178: returns TRUE if path points to an HTML resource.
1.4 cvs 179: ----------------------------------------------------------------------*/
1.3 cvs 180: #ifdef __STDC__
181: boolean IsHTMLName (char *path)
182: #else /* __STDC__ */
183: boolean IsHTMLName (path)
184: char *path;
185: #endif /* __STDC__ */
186: {
1.5 cvs 187: char temppath[MAX_LENGTH];
188: char suffix[MAX_LENGTH];
189: char nsuffix[MAX_LENGTH];
190: int i;
191:
192: if (!path)
1.13 cvs 193: return (FALSE);
1.5 cvs 194:
195: strcpy (temppath, path);
196: ExtractSuffix (temppath, suffix);
197:
198: /* Normalize the suffix */
199: i = 0;
200: while (suffix[i] != EOS)
1.13 cvs 201: {
1.25 cvs 202: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 203: i++;
204: }
1.5 cvs 205: nsuffix[i] = EOS;
206: if ((strcmp (nsuffix, "html")) &&
207: (strcmp (nsuffix, "htm")) &&
208: (strcmp (nsuffix, "shtml")))
1.13 cvs 209: return (FALSE);
1.22 cvs 210: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 211: {
212: /* take in account compressed files */
213: ExtractSuffix (temppath, suffix);
214: /* Normalize the suffix */
215: i = 0;
216: while (suffix[i] != EOS)
217: {
1.25 cvs 218: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 219: i++;
220: }
221: nsuffix[i] = EOS;
222: if ((strcmp (nsuffix, "html")) &&
223: (strcmp (nsuffix, "htm")) &&
224: (strcmp (nsuffix, "shtml")))
225: return (FALSE);
226: else
227: return (TRUE);
228: }
229: else
230: return (TRUE);
1.3 cvs 231: }
232:
1.4 cvs 233: /*----------------------------------------------------------------------
1.9 cvs 234: IsImageName
235: returns TRUE if path points to an image resource.
1.4 cvs 236: ----------------------------------------------------------------------*/
1.3 cvs 237: #ifdef __STDC__
238: boolean IsImageName (char *path)
239: #else /* __STDC__ */
240: boolean IsImageName (path)
241: char *path;
242: #endif /* __STDC__ */
243: {
1.5 cvs 244: char temppath[MAX_LENGTH];
245: char suffix[MAX_LENGTH];
246: char nsuffix[MAX_LENGTH];
247: int i;
248:
249: if (!path)
1.13 cvs 250: return (FALSE);
1.5 cvs 251:
252: strcpy (temppath, path);
253: ExtractSuffix (temppath, suffix);
254:
255: /* Normalize the suffix */
256: i = 0;
257: while (suffix[i] != EOS)
1.13 cvs 258: {
1.25 cvs 259: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 260: i++;
261: }
1.5 cvs 262: nsuffix[i] = EOS;
263: if ((strcmp (nsuffix, "gif")) && (strcmp (nsuffix, "xbm")) &&
264: (strcmp (nsuffix, "xpm")) && (strcmp (nsuffix, "jpg")) &&
265: (strcmp (nsuffix, "png")) && (strcmp (nsuffix, "au")))
1.13 cvs 266: return (FALSE);
267: return (TRUE);
1.3 cvs 268: }
269:
1.4 cvs 270: /*----------------------------------------------------------------------
1.9 cvs 271: IsTextName
1.4 cvs 272: ----------------------------------------------------------------------*/
1.3 cvs 273: #ifdef __STDC__
274: boolean IsTextName (char *path)
275: #else /* __STDC__ */
276: boolean IsTextName (path)
277: char *path;
278:
279: #endif /* __STDC__ */
280: {
1.5 cvs 281: char temppath[MAX_LENGTH];
282: char suffix[MAX_LENGTH];
283: char nsuffix[MAX_LENGTH];
284: int i;
285:
286: if (!path)
1.13 cvs 287: return (FALSE);
1.5 cvs 288:
289: strcpy (temppath, path);
290: ExtractSuffix (temppath, suffix);
291:
292: /* Normalize the suffix */
293: i = 0;
294: while (suffix[i] != EOS)
295: {
1.25 cvs 296: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 297: i++;
298: }
299: nsuffix[i] = EOS;
300:
301: if ((strcmp (nsuffix, "gif")) && (strcmp (nsuffix, "xbm")) &&
302: (strcmp (nsuffix, "xpm")) && (strcmp (nsuffix, "jpg")) &&
303: (strcmp (nsuffix, "pdf")) && (strcmp (nsuffix, "png")) &&
1.22 cvs 304: (strcmp (nsuffix, "tgz")) && (strcmp (nsuffix, "tar")) &&
305: (strcmp (nsuffix, "xpg")) && (strcmp (nsuffix, "xpd")) &&
306: (strcmp (nsuffix, "ps")) && (strcmp (nsuffix, "au")))
1.13 cvs 307: return (TRUE);
1.22 cvs 308: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 309: {
310: /* take in account compressed files */
311: ExtractSuffix (temppath, suffix);
312: /* Normalize the suffix */
313: i = 0;
314: while (suffix[i] != EOS)
315: {
1.25 cvs 316: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 317: i++;
318: }
319: nsuffix[i] = EOS;
320: if ((!strcmp (nsuffix, "html")) ||
321: (!strcmp (nsuffix, "htm")) ||
322: (!strcmp (nsuffix, "shtml")))
323: return (TRUE);
324: else
325: return (FALSE);
326: }
327: else
328: return (FALSE);
1.3 cvs 329: }
330:
1.4 cvs 331: /*----------------------------------------------------------------------
1.9 cvs 332: IsHTTPPath
333: returns TRUE if path is in fact an http URL.
1.4 cvs 334: ----------------------------------------------------------------------*/
1.3 cvs 335: #ifdef __STDC__
336: boolean IsHTTPPath (char *path)
337: #else /* __STDC__ */
338: boolean IsHTTPPath (path)
339: char *path;
340: #endif /* __STDC__ */
341: {
1.5 cvs 342: if (!path)
343: return FALSE;
1.3 cvs 344:
1.5 cvs 345: if (strncmp (path, "http:", 5) != 0)
346: return FALSE;
347: return TRUE;
1.3 cvs 348: }
349:
1.4 cvs 350: /*----------------------------------------------------------------------
1.9 cvs 351: IsWithParameters
352: returns TRUE if url has a concatenated query string.
1.4 cvs 353: ----------------------------------------------------------------------*/
1.3 cvs 354: #ifdef __STDC__
1.9 cvs 355: boolean IsWithParameters (char *url)
1.3 cvs 356: #else /* __STDC__ */
1.9 cvs 357: boolean IsWithParameters (url)
358: char *url;
1.3 cvs 359: #endif /* __STDC__ */
360: {
1.5 cvs 361: int i;
1.3 cvs 362:
1.9 cvs 363: if ((!url) || (url[0] == EOS))
1.5 cvs 364: return FALSE;
1.3 cvs 365:
1.9 cvs 366: i = strlen (url) - 1;
367: while (i > 0 && url[i--] != '?')
1.5 cvs 368: if (i < 0)
369: return FALSE;
1.3 cvs 370:
1.5 cvs 371: /* There is a parameter */
372: return TRUE;
1.3 cvs 373: }
374:
1.4 cvs 375: /*----------------------------------------------------------------------
1.9 cvs 376: IsW3Path
377: returns TRUE if path is in fact a URL.
1.4 cvs 378: ----------------------------------------------------------------------*/
1.3 cvs 379: #ifdef __STDC__
380: boolean IsW3Path (char *path)
381: #else /* __STDC__ */
382: boolean IsW3Path (path)
383: char *path;
384: #endif /* __STDC__ */
385: {
1.5 cvs 386: if ((strncmp (path, "http:", 5)) && (strncmp (path, "ftp:", 4)) &&
387: (strncmp (path, "telnet:", 7)) && (strncmp (path, "wais:", 5)) &&
388: (strncmp (path, "news:", 5)) && (strncmp (path, "gopher:", 7)) &&
389: (strncmp (path, "mailto:", 7)) && (strncmp (path, "archie:", 7)))
390: return FALSE;
391: return TRUE;
1.3 cvs 392: }
393:
1.4 cvs 394: /*----------------------------------------------------------------------
1.9 cvs 395: IsValidProtocol
396: returns true if the url protocol is supported by Amaya.
1.4 cvs 397: ----------------------------------------------------------------------*/
1.3 cvs 398: #ifdef __STDC__
1.9 cvs 399: boolean IsValidProtocol (char *url)
1.3 cvs 400: #else /* __STDC__ */
1.9 cvs 401: boolean IsValidProtocol (url)
402: char *url;
1.3 cvs 403: #endif /* __STDC__ */
404: {
1.26 cvs 405: if (!strncmp (url, "http:", 5))
1.22 cvs 406: /* experimental */
1.26 cvs 407: /*** || !strncmp (url, "ftp:", 4)) ***/
1.24 cvs 408: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 409: return (TRUE);
1.5 cvs 410: else
1.8 cvs 411: return (FALSE);
1.3 cvs 412: }
413:
1.31 cvs 414:
415: /*----------------------------------------------------------------------
416: GetBaseURL
417: normalizes orgName according to a base associated with doc, and
418: following the standard URL format rules.
419: The function returns the base used to solve relative URL and SRC:
420: - the base of the document,
421: - or the document path (without document name).
422: ----------------------------------------------------------------------*/
423: #ifdef __STDC__
424: char *GetBaseURL (Document doc)
425: #else /* __STDC__ */
426: char *GetBaseURL (doc)
427: Document doc;
428: #endif /* __STDC__ */
429: {
430: Element el;
431: ElementType elType;
432: AttributeType attrType;
433: Attribute attr;
1.33 ! cvs 434: char used_sep;
! 435: char *used_str;
1.31 cvs 436: char *ptr, *basename;
437: int length;
438:
439: basename = TtaGetMemory (MAX_LENGTH);
440: strcpy (basename, DocumentURLs[doc]);
441: length = MAX_LENGTH -1;
442: /* get the root element */
443: el = TtaGetMainRoot (doc);
444: /* search the BASE element */
445: elType.ElSSchema = TtaGetDocumentSSchema (doc);
446: elType.ElTypeNum = HTML_EL_BASE;
447: el = TtaSearchTypedElement (elType, SearchInTree, el);
448: if (el)
449: {
450: /* The document has a BASE element -> Get the HREF attribute */
451: attrType.AttrSSchema = elType.ElSSchema;
452: attrType.AttrTypeNum = HTML_ATTR_HREF_;
453: attr = TtaGetAttribute (el, attrType);
454: if (attr)
455: {
456: /* Use the base path of the document */
457: TtaGiveTextAttributeValue (attr, basename, &length);
458: /* base and orgName have to be separated by a DIR_SEP */
459: length--;
1.32 cvs 460:
1.33 ! cvs 461: if (basename [0] != EOS && strchr (basename, URL_SEP))
! 462: {
! 463: used_str = URL_STR;
! 464: used_sep = URL_SEP;
! 465: }
! 466: else
! 467: {
! 468: used_str = DIR_STR;
! 469: used_sep = DIR_SEP;
! 470: }
1.32 cvs 471:
1.33 ! cvs 472: if (basename[0] != EOS && basename[length] != used_sep)
1.31 cvs 473: /* verify if the base has the form "protocol://server:port" */
474: {
1.33 ! cvs 475: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
! 476: AMAYA_PARSE_HOST |
! 477: AMAYA_PARSE_PUNCTUATION);
1.31 cvs 478: if (ptr && !strcmp (ptr, basename))
479: {
480: /* it has this form, we complete it by adding a DIR_STR */
1.33 ! cvs 481: strcat (basename, used_str);
1.31 cvs 482: length++;
483: }
484: if (ptr)
485: TtaFreeMemory (ptr);
486: }
487: }
1.33 ! cvs 488: }
1.32 cvs 489: else
1.33 ! cvs 490: {
! 491: if (basename [0] != EOS && strchr (basename, URL_SEP))
! 492: {
! 493: used_str = URL_STR;
! 494: used_sep = URL_SEP;
! 495: }
! 496: else
! 497: {
! 498: used_str = DIR_STR;
! 499: used_sep = DIR_SEP;
! 500: }
! 501: }
! 502:
1.31 cvs 503: /* Remove anything after the last DIR_SEP char. If no such char is found,
504: * then search for the first ":" char, hoping that what's before that is a
505: * protocol. If found, end the string there. If neither char is found,
506: * then discard the whole base element.
507: */
508: length = strlen (basename) - 1;
509: /* search for the last DIR_SEP char */
1.33 ! cvs 510: while (length >= 0 && basename[length] != used_sep)
1.31 cvs 511: length--;
512: if (length >= 0)
513: /* found the last DIR_SEP char, end the string there */
514: basename[length + 1] = EOS;
515: else
516: /* search for the first PATH_STR char */
517: {
1.32 cvs 518: for (length = 0; basename[length] != ':' &&
1.31 cvs 519: basename[length] != EOS; length ++);
1.32 cvs 520: if (basename[length] == ':')
1.31 cvs 521: /* found, so end the string there */
522: basename[length + 1] = EOS;
523: else
524: /* not found, discard the base */
525: basename[0] = EOS;
526: }
527: return (basename);
528: }
529:
530:
1.4 cvs 531: /*----------------------------------------------------------------------
1.9 cvs 532: NormalizeURL
533: normalizes orgName according to a base associated with doc, and
534: following the standard URL format rules.
535: The function returns the new complete and normalized URL
1.12 cvs 536: or file name path (newName) and the name of the document (docName).
1.9 cvs 537: N.B. If the function can't find out what's the docName, it assigns
538: the name "noname.html".
1.4 cvs 539: ----------------------------------------------------------------------*/
1.3 cvs 540: #ifdef __STDC__
541: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName)
542: #else /* __STDC__ */
543: void NormalizeURL (orgName, doc, newName, docName)
544: char *orgName;
545: Document doc;
546: char *newName;
547: char *docName;
548: #endif /* __STDC__ */
549: {
1.31 cvs 550: char *basename;
1.18 cvs 551: char tempOrgName[MAX_LENGTH];
1.5 cvs 552: char *ptr;
553: int length;
1.33 ! cvs 554: char used_sep;
! 555: char *used_str;
1.5 cvs 556:
557: if (!newName || !docName)
558: return;
1.18 cvs 559:
1.32 cvs 560: if (doc != 0)
1.33 ! cvs 561: basename = GetBaseURL (doc);
1.32 cvs 562: else
1.33 ! cvs 563: basename = (char *) NULL;
1.32 cvs 564:
1.33 ! cvs 565: if (strchr (orgName, URL_SEP) || (basename && strchr (basename, URL_SEP)))
! 566: {
! 567: used_str = URL_STR;
! 568: used_sep = URL_SEP;
! 569: }
! 570: else
! 571: {
! 572: used_str = DIR_STR;
! 573: used_sep = DIR_SEP;
! 574: }
1.32 cvs 575:
1.18 cvs 576: /*
1.31 cvs 577: * Clean orgName
578: * Make sure we have a complete orgName, without any leading or trailing
579: * white spaces, or trailinbg new lines
580: */
1.18 cvs 581:
1.5 cvs 582: ptr = orgName;
1.18 cvs 583: /* skip leading white space and new line characters */
1.19 cvs 584: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS);
1.18 cvs 585: strcpy (tempOrgName, ptr);
586: /* clean trailing white space */
587: ptr = strchr (tempOrgName, ' ');
588: if (ptr)
589: *ptr = EOS;
590: /* clean trailing new lines */
1.19 cvs 591: ptr = strchr (tempOrgName, EOL);
1.5 cvs 592: if (ptr)
593: *ptr = EOS;
594:
1.18 cvs 595: /*
1.31 cvs 596: * Make orgName a complete URL
597: * If the URL does not include a protocol, then try to calculate
598: * one using the doc's base element (if it exists),
599: */
1.21 cvs 600: if (tempOrgName[0] == EOS)
601: {
602: newName[0] = EOS;
603: return;
604: }
605: else if (IsW3Path (tempOrgName))
606: {
607: /* the name is complete, go to the Sixth Step */
608: strcpy (newName, tempOrgName);
609: /* verify if the URL has the form "protocol://server:port" */
1.25 cvs 610: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST |
611: AMAYA_PARSE_PUNCTUATION);
1.21 cvs 612: if (ptr && !strcmp (ptr, newName))
613: {
1.29 cvs 614: /* it has this form, we complete it by adding a DIR_STR */
1.33 ! cvs 615: strcat (newName, used_str);
1.21 cvs 616: }
617: if (ptr)
1.25 cvs 618: TtaFreeMemory (ptr);
1.21 cvs 619: }
620: else if ( doc == 0)
1.19 cvs 621: /* the name is complete, go to the Sixth Step */
1.18 cvs 622: strcpy (newName, tempOrgName);
1.5 cvs 623: else
624: {
1.22 cvs 625:
1.31 cvs 626: /* Calculate the absolute URL, using the base or document URL */
1.32 cvs 627:
1.25 cvs 628: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.31 cvs 629: TtaFreeMemory (basename);
1.14 cvs 630: if (ptr)
631: {
1.29 cvs 632: SimplifyUrl (&ptr);
1.14 cvs 633: strcpy (newName, ptr);
1.25 cvs 634: TtaFreeMemory (ptr);
1.14 cvs 635: }
636: else
1.18 cvs 637: newName[0] = EOS;
1.5 cvs 638: }
639:
1.18 cvs 640: /*
1.31 cvs 641: * Prepare the docname that will refer to this ressource in the
642: * .amaya directory. If the new URL finishes on DIR_SEP, then use
643: * noname.html as a default ressource name
1.18 cvs 644: */
1.19 cvs 645: if (newName[0] != EOS)
1.5 cvs 646: {
1.19 cvs 647: length = strlen (newName) - 1;
1.33 ! cvs 648: if (newName[length] == used_sep)
1.18 cvs 649: {
650: /* docname was not comprised inside the URL, so let's */
651: /* assign the default ressource name */
652: strcpy (docName, "noname.html");
653: /* remove DIR_SEP at the end of complete path */
1.23 cvs 654: /* newName[length] = EOS; */
1.18 cvs 655: }
1.14 cvs 656: else
1.18 cvs 657: {
658: /* docname is comprised inside the URL */
1.33 ! cvs 659: while (length >= 0 && newName[length] != used_sep)
1.18 cvs 660: length--;
661: if (length < 0)
662: strcpy (docName, newName);
663: else
664: strcpy (docName, &newName[length+1]);
665: }
1.19 cvs 666:
1.5 cvs 667: }
1.18 cvs 668: else
669: docName[0] = EOS;
670: }
1.3 cvs 671:
1.4 cvs 672: /*----------------------------------------------------------------------
1.9 cvs 673: IsSameHost
1.4 cvs 674: ----------------------------------------------------------------------*/
1.3 cvs 675: #ifdef __STDC__
676: boolean IsSameHost (char *url1, char *url2)
677: #else /* __STDC__ */
678: boolean IsSameHost (url1, url2)
679: char *path;
680: #endif /* __STDC__ */
681: {
1.5 cvs 682: char *basename_ptr1, *basename_ptr2;
683: boolean result;
1.3 cvs 684:
1.25 cvs 685: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
686: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 687:
1.5 cvs 688: if (strcmp (basename_ptr1, basename_ptr2))
1.8 cvs 689: result = FALSE;
1.5 cvs 690: else
1.8 cvs 691: result = TRUE;
1.3 cvs 692:
1.25 cvs 693: TtaFreeMemory (basename_ptr1);
694: TtaFreeMemory (basename_ptr2);
1.5 cvs 695: return (result);
1.3 cvs 696: }
697:
698:
1.4 cvs 699: /*----------------------------------------------------------------------
1.22 cvs 700: HasKnownFileSuffix
701: returns TRUE if path points to a file ending with a suffix.
702: ----------------------------------------------------------------------*/
703: #ifdef __STDC__
704: boolean HasKnownFileSuffix (char *path)
705: #else /* __STDC__ */
706: boolean HasKnownFileSuffix (path)
707: char *path;
708: #endif /* __STDC__ */
709: {
1.29 cvs 710: char *root;
711: char temppath[MAX_LENGTH];
712: char suffix[MAX_LENGTH];
1.22 cvs 713:
1.24 cvs 714: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 715: return (FALSE);
716:
1.29 cvs 717: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 718:
719: if (root)
720: {
721: strcpy (temppath, root);
1.25 cvs 722: TtaFreeMemory (root);
1.22 cvs 723: /* Get the suffix */
724: ExtractSuffix (temppath, suffix);
725:
726: if( suffix[0] == EOS)
727: /* no suffix */
728: return (FALSE);
729:
730: /* Normalize the suffix */
731: ConvertToLowerCase (suffix);
732:
1.23 cvs 733: if (!strcmp (suffix, "gz"))
1.22 cvs 734: /* skip the compressed suffix */
735: {
736: ExtractSuffix (temppath, suffix);
737: if(suffix[0] == EOS)
738: /* no suffix */
739: return (FALSE);
740: /* Normalize the suffix */
741: ConvertToLowerCase (suffix);
742: }
743:
744: if ((strcmp (suffix, "gif")) && (strcmp (suffix, "xbm")) &&
745: (strcmp (suffix, "xpm")) && (strcmp (suffix, "jpg")) &&
746: (strcmp (suffix, "pdf")) && (strcmp (suffix, "png")) &&
747: (strcmp (suffix, "tgz")) && (strcmp (suffix, "xpg")) &&
748: (strcmp (suffix, "xpd")) && (strcmp (suffix, "ps")) &&
749: (strcmp (suffix, "au")) && (strcmp (suffix, "html")) &&
750: (strcmp (suffix, "htm")) && (strcmp (suffix, "shtml")) &&
751: (strcmp (suffix, "txt")) && (strcmp (suffix, "css")) &&
752: (strcmp (suffix, "eps")))
753: return (FALSE);
754: else
755: return (TRUE);
756: }
757: else
758: return (FALSE);
759: }
760:
761:
762: /*----------------------------------------------------------------------
1.24 cvs 763: ChopURL
764: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
765: If inputURL is bigger than that size, outputURL receives
766: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
767: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
768: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
769: copied into outputURL.
770: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
771: chars.
772: ----------------------------------------------------------------------*/
773: #ifdef __STDC__
774: void ChopURL (char *outputURL, char *inputURL)
775: #else
776: void ChopURL (outputURL, inputURL)
777: char *outputURL;
778: char *inputURL;
779: #endif
1.22 cvs 780:
1.24 cvs 781: {
782: int len;
1.9 cvs 783:
1.24 cvs 784: len = strlen (inputURL);
785: if (len <= MAX_PRINT_URL_LENGTH)
1.29 cvs 786: strcpy (outputURL, inputURL);
1.24 cvs 787: else
788: /* make a truncated urlName on the status window */
789: {
790: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
791: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
792: strcat (outputURL, "...");
793: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
794: }
1.25 cvs 795: }
796:
797:
798: /*----------------------------------------------------------------------
799: scan
800: Scan a filename for its consituents
801: -----------------------------------
802:
803: On entry,
804: name points to a document name which may be incomplete.
805: On exit,
806: absolute or relative may be nonzero (but not both).
807: host, fragment and access may be nonzero if they were specified.
808: Any which are nonzero point to zero terminated strings.
809: ----------------------------------------------------------------------*/
810: #ifdef __STDC__
811: static void scan (char * name, HTURI * parts)
812: #else /* __STDC__ */
813: static void scan (name, parts)
814: char *name;
815: HTURI *parts;
816:
817: #endif /* __STDC__ */
818: {
1.28 cvs 819: char * p;
820: char * after_access = name;
1.33 ! cvs 821: char used_sep;
1.32 cvs 822: char my_path_sep;
823:
1.33 ! cvs 824: if (name && strchr (name, URL_SEP))
! 825: {
! 826: used_sep = URL_SEP;
! 827: my_path_sep = ':';
! 828: }
! 829: else
! 830: {
! 831: used_sep = DIR_SEP;
! 832: my_path_sep = ':';
! 833: }
1.32 cvs 834:
1.28 cvs 835: memset(parts, '\0', sizeof(HTURI));
836: /* Look for fragment identifier */
837: if ((p = strrchr(name, '#')) != NULL)
838: {
839: *p++ = '\0';
840: parts->fragment = p;
1.25 cvs 841: }
842:
1.28 cvs 843: for (p=name; *p; p++)
844: {
1.33 ! cvs 845: if (*p==used_sep || *p=='#' || *p=='?')
1.28 cvs 846: break;
1.32 cvs 847: if (*p==':')
1.28 cvs 848: {
849: *p = 0;
850: parts->access = after_access; /* Scheme has been specified */
851:
852: /* The combination of gcc, the "-O" flag and the HP platform is
853: unhealthy. The following three lines is a quick & dirty fix, but is
854: not recommended. Rather, turn off "-O". */
855:
856: /* after_access = p;*/
857: /* while (*after_access == 0)*/
858: /* after_access++;*/
859: after_access = p+1;
860: if (0==strcasecmp("URL", parts->access))
861: /* Ignore IETF's URL: pre-prefix */
862: parts->access = NULL;
863: else
1.25 cvs 864: break;
865: }
866: }
867:
868: p = after_access;
1.33 ! cvs 869: if (*p==used_sep)
1.28 cvs 870: {
1.33 ! cvs 871: if (p[1]==used_sep)
1.28 cvs 872: {
1.25 cvs 873: parts->host = p+2; /* host has been specified */
1.28 cvs 874: *p = 0; /* Terminate access */
875: /* look for end of host name if any */
1.33 ! cvs 876: p = strchr(parts->host,used_sep);
1.28 cvs 877: if (p)
878: {
1.25 cvs 879: *p=0; /* Terminate host */
880: parts->absolute = p+1; /* Root has been found */
1.28 cvs 881: }
882: }
883: else
884: /* Root found but no host */
885: parts->absolute = p+1;
886: }
887: else
888: {
1.25 cvs 889: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 890: }
1.25 cvs 891: }
892:
893:
894: /*----------------------------------------------------------------------
1.28 cvs 895: AmayaParseUrl: parse a Name relative to another name
896:
897: This returns those parts of a name which are given (and requested)
898: substituting bits from the related name where necessary.
1.25 cvs 899:
1.28 cvs 900: On entry,
1.25 cvs 901: aName A filename given
902: relatedName A name relative to which aName is to be parsed. Give
903: it an empty string if aName is absolute.
904: wanted A mask for the bits which are wanted.
905:
1.28 cvs 906: On exit,
1.25 cvs 907: returns A pointer to a malloc'd string which MUST BE FREED
908: ----------------------------------------------------------------------*/
909: #ifdef __STDC__
1.28 cvs 910: char *AmayaParseUrl (char *aName, char *relatedName, int wanted)
1.25 cvs 911: #else /* __STDC__ */
1.28 cvs 912: char *AmayaParseUrl (aName, relatedName, wanted)
913: char *aName;
914: char *relatedName;
915: int wanted;
1.25 cvs 916:
917: #endif /* __STDC__ */
918: {
1.29 cvs 919: char *return_value;
920: char result[MAX_LENGTH];
921: char name[MAX_LENGTH];
922: char rel[MAX_LENGTH];
923: char *p, *access;
924: HTURI given, related;
925: int len;
1.33 ! cvs 926: char used_sep;
! 927: char *used_str;
1.32 cvs 928:
1.33 ! cvs 929: if (strchr (aName, URL_SEP) || strchr (relatedName, URL_SEP) )
! 930: {
! 931: used_str = URL_STR;
! 932: used_sep = URL_SEP;
! 933: }
1.32 cvs 934: else
1.33 ! cvs 935: {
! 936: used_str = DIR_STR;
! 937: used_sep = DIR_SEP;
! 938: }
1.32 cvs 939:
1.29 cvs 940: /* Make working copies of input strings to cut up: */
941: return_value = NULL;
942: result[0] = 0; /* Clear string */
943: strcpy (name, aName);
944: if (relatedName != NULL)
945: strcpy (rel, relatedName);
946: else
947: relatedName[0] = EOS;
948:
949: scan (name, &given);
950: scan (rel, &related);
951: access = given.access ? given.access : related.access;
952: if (wanted & AMAYA_PARSE_ACCESS)
953: if (access)
954: {
955: strcat (result, access);
956: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.32 cvs 957: strcat (result, ":");
1.29 cvs 958: }
959:
960: if (given.access && related.access)
961: /* If different, inherit nothing. */
962: if (strcmp (given.access, related.access) != 0)
963: {
964: related.host = 0;
965: related.absolute = 0;
966: related.relative = 0;
967: related.fragment = 0;
968: }
969:
970: if (wanted & AMAYA_PARSE_HOST)
971: if(given.host || related.host)
972: {
973: if(wanted & AMAYA_PARSE_PUNCTUATION)
974: strcat (result, "//");
975: strcat (result, given.host ? given.host : related.host);
976: }
977:
978: if (given.host && related.host)
979: /* If different hosts, inherit no path. */
980: if (strcmp(given.host, related.host) != 0)
981: {
982: related.absolute = 0;
983: related.relative = 0;
984: related.fragment = 0;
985: }
986:
987: if (wanted & AMAYA_PARSE_PATH)
988: {
989: if (given.absolute)
990: {
991: /* All is given */
992: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.33 ! cvs 993: strcat (result, used_str);
1.29 cvs 994: strcat (result, given.absolute);
1.25 cvs 995: }
1.29 cvs 996: else if (related.absolute)
997: {
998: /* Adopt path not name */
1.33 ! cvs 999: strcat (result, used_str);
1.29 cvs 1000: strcat (result, related.absolute);
1001: if (given.relative)
1002: {
1003: /* Search part? */
1004: p = strchr (result, '?');
1005: if (!p)
1006: p=result+strlen(result)-1;
1.33 ! cvs 1007: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1008: /* Remove filename */
1009: p[1]=0;
1010: /* Add given one */
1011: strcat (result, given.relative);
1012: /*SimplifyUrl (&result);*/
1.25 cvs 1013: }
1014: }
1.29 cvs 1015: else if (given.relative)
1016: /* what we've got */
1017: strcat (result, given.relative);
1018: else if (related.relative)
1019: strcat (result, related.relative);
1020: else
1021: /* No inheritance */
1.33 ! cvs 1022: strcat (result, used_str);
1.25 cvs 1023: }
1.29 cvs 1024:
1025: if (wanted & AMAYA_PARSE_ANCHOR)
1026: if (given.fragment || related.fragment)
1027: {
1028: if (given.absolute && given.fragment)
1029: {
1030: /*Fixes for relURLs...*/
1031: if (wanted & AMAYA_PARSE_PUNCTUATION)
1032: strcat (result, "#");
1033: strcat (result, given.fragment);
1034: }
1035: else if (!(given.absolute) && !(given.fragment))
1036: strcat (result, "");
1037: else
1038: {
1039: if (wanted & AMAYA_PARSE_PUNCTUATION)
1040: strcat (result, "#");
1041: strcat (result, given.fragment ? given.fragment : related.fragment);
1042: }
1043: }
1044: len = strlen (result);
1045: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1046: strcpy (return_value, result);
1047: return (return_value); /* exactly the right length */
1.25 cvs 1048: }
1049:
1050: /*----------------------------------------------------------------------
1051: HTCanon
1052: Canonicalizes the URL in the following manner starting from the host
1053: pointer:
1054:
1055: 1) The host name is converted to lowercase
1056: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1057:
1058: Return: OK The position of the current path part of the URL
1059: which might be the old one or a new one.
1060:
1061: ----------------------------------------------------------------------*/
1062: #ifdef __STDC__
1.28 cvs 1063: static char *HTCanon (char ** filename, char * host)
1.25 cvs 1064: #else /* __STDC__ */
1.28 cvs 1065: static char *HTCanon (filename, host)
1066: char **filename;
1067: char *host;
1.25 cvs 1068: #endif /* __STDC__ */
1069: {
1070: char *newname = NULL;
1071: char *port;
1072: char *strptr;
1073: char *path;
1074: char *access = host-3;
1.33 ! cvs 1075: char used_sep;
! 1076: char *used_str;
1.32 cvs 1077:
1078:
1.33 ! cvs 1079: if (*filename && strchr (*filename, URL_SEP))
! 1080: {
! 1081: used_str = URL_STR;
! 1082: used_sep = URL_SEP;
! 1083: }
! 1084: else
! 1085: {
! 1086: used_str = DIR_STR;
! 1087: used_sep = DIR_SEP;
! 1088: }
1.32 cvs 1089:
1.33 ! cvs 1090: while (access>*filename && *(access-1)!= used_sep) /* Find access method */
1.25 cvs 1091: access--;
1.33 ! cvs 1092: if ((path = strchr(host, used_sep)) == NULL) /* Find path */
1.25 cvs 1093: path = host + strlen(host);
1094: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */
1095: host = strptr;
1.32 cvs 1096: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */
1.25 cvs 1097: port = NULL;
1098:
1099: strptr = host; /* Convert to lower-case */
1.33 ! cvs 1100: while (strptr<path)
! 1101: {
1.25 cvs 1102: *strptr = tolower(*strptr);
1103: strptr++;
1.33 ! cvs 1104: }
1.25 cvs 1105:
1106: /* Does the URL contain a full domain name? This also works for a
1107: numerical host name. The domain name is already made lower-case
1108: and without a trailing dot. */
1109: {
1.33 ! cvs 1110: char *dot = port ? port : path;
! 1111: if (dot > *filename && *--dot=='.')
! 1112: {
! 1113: char *orig=dot, *dest=dot+1;
! 1114: while((*orig++ = *dest++));
! 1115: if (port) port--;
! 1116: path--;
1.25 cvs 1117: }
1118: }
1119: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 ! cvs 1120: if (port)
! 1121: {
! 1122: if (!*(port+1) || *(port+1)==used_sep)
! 1123: {
! 1124: if (!newname)
! 1125: {
1.25 cvs 1126: char *orig=port, *dest=port+1;
1127: while((*orig++ = *dest++));
1.33 ! cvs 1128: }
! 1129: }
! 1130: else if ((!strncmp(access, "http", 4) &&
! 1131: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
! 1132: (!strncmp(access, "gopher", 6) &&
! 1133: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
! 1134: (!strncmp(access, "ftp", 3) &&
! 1135: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) {
! 1136: if (!newname)
! 1137: {
! 1138: char *orig=port, *dest=port+3;
! 1139: while((*orig++ = *dest++));
! 1140: /* Update path position, Henry Minsky */
! 1141: path -= 3;
1.25 cvs 1142: }
1.33 ! cvs 1143: }
! 1144: else if (newname)
! 1145: strncat(newname, port, (int) (path-port));
! 1146: }
1.25 cvs 1147:
1.33 ! cvs 1148: if (newname)
! 1149: {
1.25 cvs 1150: char *newpath = newname+strlen(newname);
1151: strcat(newname, path);
1152: path = newpath;
1.28 cvs 1153: /* Free old copy */
1154: TtaFreeMemory(*filename);
1.25 cvs 1155: *filename = newname;
1.33 ! cvs 1156: }
1.25 cvs 1157: return path;
1158: }
1159:
1160:
1161: /*----------------------------------------------------------------------
1.29 cvs 1162: SimplifyUrl: simplify a URI
1.32 cvs 1163: A URI is allowed to contain the sequence xxx/../ which may be
1164: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1165: Simplification helps us recognize duplicate URIs.
1.25 cvs 1166:
1.28 cvs 1167: Thus, /etc/junk/../fred becomes /etc/fred
1168: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1169:
1.28 cvs 1170: but we should NOT change
1171: http://fred.xxx.edu/../..
1.25 cvs 1172:
1173: or ../../albert.html
1174:
1.28 cvs 1175: In order to avoid empty URLs the following URLs become:
1.25 cvs 1176:
1177: /fred/.. becomes /fred/..
1178: /fred/././.. becomes /fred/..
1179: /fred/.././junk/.././ becomes /fred/..
1180:
1.28 cvs 1181: If more than one set of `://' is found (several proxies in cascade) then
1182: only the part after the last `://' is simplified.
1.25 cvs 1183:
1.28 cvs 1184: Returns: A string which might be the old one or a new one.
1.25 cvs 1185: ----------------------------------------------------------------------*/
1186: #ifdef __STDC__
1.29 cvs 1187: void SimplifyUrl (char ** url)
1.25 cvs 1188: #else /* __STDC__ */
1.29 cvs 1189: void SimplifyUrl (url)
1.28 cvs 1190: char **url;
1.25 cvs 1191: #endif /* __STDC__ */
1192: {
1.28 cvs 1193: char *path, *p;
1194: char *newptr, *access;
1195: char *orig, *dest, *end;
1196:
1.33 ! cvs 1197: char used_sep;
! 1198: char *used_str;
1.32 cvs 1199:
1200:
1.28 cvs 1201: if (!url || !*url)
1202: return;
1203:
1.33 ! cvs 1204: if (strchr (*url, URL_SEP))
! 1205: {
! 1206: used_str = URL_STR;
! 1207: used_sep = URL_SEP;
! 1208: }
1.32 cvs 1209: else
1.33 ! cvs 1210: {
! 1211: used_str = DIR_STR;
! 1212: used_sep = DIR_SEP;
! 1213: }
1.32 cvs 1214:
1.28 cvs 1215: /* Find any scheme name */
1216: if ((path = strstr(*url, "://")) != NULL)
1.33 ! cvs 1217: {
! 1218: /* Find host name */
1.28 cvs 1219: access = *url;
1220: while (access<path && (*access=tolower(*access)))
1221: access++;
1222: path += 3;
1223: while ((newptr = strstr(path, "://")) != NULL)
1224: /* For proxies */
1225: path = newptr+3;
1226: /* We have a host name */
1227: path = HTCanon(url, path);
1.25 cvs 1228: }
1.28 cvs 1229: else if ((path = strstr(*url, ":/")) != NULL)
1230: path += 2;
1231: else
1232: path = *url;
1.25 cvs 1233:
1.33 ! cvs 1234: if (*path == used_sep && *(path+1)==used_sep)
1.28 cvs 1235: /* Some URLs start //<foo> */
1236: path += 1;
1237: else if (!strncmp(path, "news:", 5))
1238: {
1239: newptr = strchr(path+5, '@');
1240: if (!newptr)
1241: newptr = path + 5;
1242: while (*newptr)
1243: {
1244: /* Make group or host lower case */
1245: *newptr = tolower (*newptr);
1246: newptr++;
1.25 cvs 1247: }
1.28 cvs 1248: /* Doesn't need to do any more */
1249: return;
1.25 cvs 1250: }
1.28 cvs 1251:
1252: if ((p = path))
1253: {
1254: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1255: (end = strchr (path, '#'))))
1256: end = path + strlen (path);
1257:
1258: /* Parse string second time to simplify */
1259: p = path;
1260: while (p < end)
1261: {
1.33 ! cvs 1262: if (*p==used_sep)
1.28 cvs 1263: {
1.33 ! cvs 1264: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1265: {
1266: orig = p + 1;
1.33 ! cvs 1267: dest = (*(p+2)!=used_sep) ? p+2 : p+3;
1.28 cvs 1268: while ((*orig++ = *dest++)); /* Remove a slash and a dot */
1269: end = orig - 1;
1270: }
1.33 ! cvs 1271: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3)))
1.28 cvs 1272: {
1273: newptr = p;
1.33 ! cvs 1274: while (newptr>path && *--newptr!=used_sep); /* prev slash */
1.28 cvs 1275: if (strncmp(newptr, "/../", 4))
1276: {
1277: orig = newptr + 1;
1.33 ! cvs 1278: dest = (*(p+3)!=used_sep) ? p+3 : p+4;
1.28 cvs 1279: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1280: end = orig-1;
1281: /* Start again with prev slash */
1282: p = newptr;
1.25 cvs 1283: }
1.28 cvs 1284: else
1.25 cvs 1285: p++;
1.28 cvs 1286: }
1.33 ! cvs 1287: else if (*(p+1) == used_sep)
1.28 cvs 1288: {
1.33 ! cvs 1289: while (*(p+1) == used_sep)
1.28 cvs 1290: {
1291: orig = p;
1292: dest = p + 1;
1293: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1294: end = orig-1;
1295: }
1296: }
1297: else
1.25 cvs 1298: p++;
1.28 cvs 1299: }
1300: else
1301: p++;
1.25 cvs 1302: }
1303: }
1.28 cvs 1304: return;
1305: }
1306:
1307:
1308: /*----------------------------------------------------------------------
1309: NormalizeFile normalizes local names.
1310: Return TRUE if target and src differ.
1311: ----------------------------------------------------------------------*/
1312: #ifdef __STDC__
1313: boolean NormalizeFile (char *src, char *target)
1314: #else
1315: boolean NormalizeFile (src, target)
1316: char *src;
1317: char *target;
1318:
1319: #endif
1320: {
1321: char *s;
1322: boolean change;
1323:
1324: change = FALSE;
1325: if (src[0] == '~')
1326: {
1327: /* replace ~ */
1328: s = (char *) TtaGetEnvString ("HOME");
1329: strcpy (target, s);
1330: strcat (target, &src[1]);
1331: change = TRUE;
1332: }
1333: else if (strncmp (src, "file:", 5) == 0)
1334: {
1335: /* remove the prefix file: */
1336: if (src[5] == EOS)
1337: strcpy (target, DIR_STR);
1338: else if (src[0] == '~')
1339: {
1340: /* replace ~ */
1341: s = (char *) TtaGetEnvString ("HOME");
1342: strcpy (target, s);
1343: strcat (target, &src[5]);
1344: }
1345: else
1346: strcpy (target, &src[5]);
1347: change = TRUE;
1348: }
1349: else
1350: strcpy (target, src);
1351:
1352: /* remove /../ and /./ */
1.29 cvs 1353: SimplifyUrl (&target);
1.28 cvs 1354: return (change);
1.25 cvs 1355: }
1356:
1.28 cvs 1357:
1.25 cvs 1358: /*----------------------------------------------------------------------
1.31 cvs 1359: MakeRelativeURL: make relative name
1.25 cvs 1360:
1.28 cvs 1361: This function creates and returns a string which gives an expression of
1362: one address as related to another. Where there is no relation, an absolute
1363: address is retured.
1.25 cvs 1364:
1.28 cvs 1365: On entry,
1.25 cvs 1366: Both names must be absolute, fully qualified names of nodes
1367: (no fragment bits)
1368:
1.28 cvs 1369: On exit,
1.25 cvs 1370: The return result points to a newly allocated name which, if
1371: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1372: The caller is responsible for freeing the resulting name later.
1373: ----------------------------------------------------------------------*/
1374: #ifdef __STDC__
1.31 cvs 1375: char *MakeRelativeURL (char *aName, char *relatedName)
1.25 cvs 1376: #else /* __STDC__ */
1.31 cvs 1377: char *MakeRelativeURL (aName, relatedName)
1.28 cvs 1378: char *aName;
1379: char *relatedName;
1.25 cvs 1380: #endif /* __STDC__ */
1381: {
1.29 cvs 1382: char *return_value;
1383: char result[MAX_LENGTH];
1384: char *p;
1385: char *q = relatedName;
1386: char *after_access;
1387: char *last_slash = NULL;
1388: int slashes, levels, len;
1389:
1390: if (aName == NULL || relatedName == NULL)
1391: return (NULL);
1392:
1393: slashes = 0;
1394: after_access = NULL;
1395: p = aName;
1396: q = relatedName;
1397: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1398: {
1399: /* Find extent of match */
1.32 cvs 1400: if (*p == ':')
1.29 cvs 1401: after_access = p + 1;
1.28 cvs 1402: if (*p == DIR_SEP)
1.27 cvs 1403: {
1.29 cvs 1404: /* memorize the last slash position and count them */
1.27 cvs 1405: last_slash = p;
1406: slashes++;
1.25 cvs 1407: }
1408: }
1409:
1.31 cvs 1410: /* q, p point to the first non-matching character or zero */
1411: if (*q == EOS)
1412: {
1413: /* New name is a subset of the related name */
1414: /* exactly the right length */
1415: len = strlen (p);
1416: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1417: strcpy (return_value, p);
1418: }
1419: else if ((slashes < 2 && after_access == NULL)
1420: || (slashes < 3 && after_access != NULL))
1421: {
1422: /* Two names whitout common path */
1423: /* exactly the right length */
1424: len = strlen (aName);
1425: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1426: strcpy (return_value, aName);
1427: }
1428: else
1429: {
1430: /* Some path in common */
1431: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1432: /* just the same server */
1433: strcpy (result, last_slash);
1434: else
1435: {
1436: levels= 0;
1437: for (; *q && (*q != '#'); q++)
1438: if (*q == DIR_SEP)
1439: levels++;
1440:
1441: result[0] = 0;
1442: for (;levels; levels--)
1443: strcat (result, "../");
1444: strcat (result, last_slash+1);
1445: }
1446:
1447: /* exactly the right length */
1448: len = strlen (result);
1449: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1450: strcpy (return_value, result);
1.25 cvs 1451: }
1.29 cvs 1452: return (return_value);
1.24 cvs 1453: }
Webmaster