Annotation of Amaya/amaya/AHTURLTools.c, revision 1.77
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
1.45 cvs 14: * R. Guetari (Stuff related to 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"
23:
1.24 cvs 24: #define MAX_PRINT_URL_LENGTH 50
1.29 cvs 25: typedef struct _HTURI {
1.67 cvs 26: STRING access; /* Now known as "scheme" */
27: STRING host;
28: STRING absolute;
29: STRING relative;
30: STRING fragment;
1.29 cvs 31: } HTURI;
1.24 cvs 32:
1.28 cvs 33:
34: /*----------------------------------------------------------------------
35: ConvertToLowerCase
36: Converts a string to lowercase.
37: ----------------------------------------------------------------------*/
1.22 cvs 38: #ifdef __STDC__
1.67 cvs 39: void ConvertToLowerCase (STRING string)
1.28 cvs 40: #else /* __STDC__ */
1.38 cvs 41: void ConvertToLowerCase (string)
1.67 cvs 42: STRING string;
1.28 cvs 43:
44: #endif /* __STDC__ */
45: {
46: int i;
47:
48: if (!string)
49: return;
50:
51: for (i = 0; string[i] != EOS; i++)
1.67 cvs 52: string[i] = utolower (string[i]);
1.28 cvs 53: }
1.22 cvs 54:
1.8 cvs 55: /*----------------------------------------------------------------------
1.75 cvs 56: EscapeChar
57: writes the equivalent escape code of a char in a string
58: ----------------------------------------------------------------------*/
59: #ifdef __STDC__
60: void EscapeChar (STRING string, UCHAR_T c)
61: #else
62: void EscapeChar (string, c)
63: STRING string;
64: UCHAR_T c;
65:
66: #endif
67: {
68: c &= 0xFF; /* strange behavior under solaris? */
69: usprintf (string, TEXT("%02x"), (unsigned int) c);
70: }
71:
72: /*----------------------------------------------------------------------
73: EscapeURL
74: Takes a URL and escapes all protected chars into
75: %xx sequences. Also, removes any leading white spaces
76: Returns either NULL or a new buffer, which must be freed by the caller
77: ----------------------------------------------------------------------*/
78: #ifdef __STDC__
79: STRING EscapeURL (const STRING url)
80: #else
81: STRING EscapeURL (url)
82: STRING url;
83: #endif /* __STDC__ */
84: {
85: STRING buffer;
86: int buffer_len;
87: int buffer_free_mem;
88: PCHAR_T ptr;
89: int new_chars;
90: void *status;
91:
92: if (url && *url)
93: {
1.76 cvs 94: buffer_free_mem = ustrlen (url) + 20;
95: buffer = TtaAllocString (buffer_free_mem + 1);
1.75 cvs 96: ptr = url;
97: buffer_len = 0;
98:
99: while (*ptr)
100: {
101: switch (*ptr)
102: {
103: /* put here below all the chars that need to
104: be escaped into %xx */
105: case TEXT(0x27): /* & */
106: case TEXT(0x20): /* space */
107: new_chars = 3;
108: break;
109:
110: default:
111: new_chars = 1;
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;
1.75 cvs 119: status = TtaRealloc (buffer, sizeof (CHAR_T)
120: * (buffer_len + buffer_free_mem + 1));
121: if (status)
122: buffer = (STRING) status;
123: else {
124: /* @@ maybe we should do some other behavior here, like
125: freeing the buffer and return a void thing */
1.76 cvs 126: buffer[buffer_len] = EOS;
1.75 cvs 127: break;
128: }
129: }
130: /* escape the char */
131: if (new_chars == 3)
132: {
133: buffer[buffer_len] = TEXT('%');
134: EscapeChar (&buffer[buffer_len+1], *ptr);
135: }
136: else
137: buffer[buffer_len] = *ptr;
138:
139: /* update the status */
140: buffer_len += new_chars;
141: buffer_free_mem -= new_chars;
142: /* examine the next char */
143: ptr++;
144: }
145: buffer[buffer_len] = EOS;
146: }
1.76 cvs 147: else
148: buffer = NULL;
149:
1.75 cvs 150: return (buffer);
151: }
152:
153:
154: /*----------------------------------------------------------------------
1.11 cvs 155: ExplodeURL
1.8 cvs 156: ----------------------------------------------------------------------*/
157: #ifdef __STDC__
158: void ExplodeURL (char *url, char **proto, char **host, char **dir, char **file)
159: #else
160: void ExplodeURL (url, proto, host, dir, file)
161: char *url;
162: char **proto;
163: char **host;
164: char **dir;
165: char **file;
166:
167: #endif
168: {
1.33 cvs 169: char *curr, *temp;
170: char used_sep;
1.32 cvs 171:
1.33 cvs 172: if (url && strchr (url, URL_SEP))
173: used_sep = URL_SEP;
174: else
175: used_sep = DIR_SEP;
1.8 cvs 176:
177: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
178: (dir == NULL) || (file == NULL))
179: return;
180:
181: /* initialize every pointer */
182: *proto = *host = *dir = *file = NULL;
183:
184: /* skip any leading space */
185: while ((*url == SPACE) || (*url == TAB))
186: url++;
1.9 cvs 187: curr = url;
188: if (*curr == 0)
1.8 cvs 189: goto finished;
190:
191: /* go to the end of the URL */
1.68 cvs 192: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
193: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 194: curr++;
1.8 cvs 195:
196: /* mark the end of the chain */
1.9 cvs 197: *curr = EOS;
198: curr--;
199: if (curr <= url)
1.8 cvs 200: goto finished;
201:
202: /* search the next DIR_SEP indicating the beginning of the file name */
203: do
1.11 cvs 204: curr--;
1.33 cvs 205: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 206:
1.9 cvs 207: if (curr < url)
1.8 cvs 208: goto finished;
1.9 cvs 209: *file = curr + 1;
1.8 cvs 210:
211: /* mark the end of the dir */
1.9 cvs 212: *curr = EOS;
213: curr--;
214: if (curr < url)
1.8 cvs 215: goto finished;
216:
1.29 cvs 217: /* search for the DIR_STR indicating the host name start */
1.33 cvs 218: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 219: curr--;
1.8 cvs 220:
221: /* if we found it, separate the host name from the directory */
1.33 cvs 222: if ((*curr == DIR_SEP) && (*(curr + 1) == used_sep))
1.8 cvs 223: {
1.9 cvs 224: *host = temp = curr + 2;
1.33 cvs 225: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 226: temp++;
1.33 cvs 227: if (*temp == used_sep)
1.8 cvs 228: {
229: *temp = EOS;
230: *dir = temp + 1;
231: }
232: }
233: else
1.11 cvs 234: *dir = curr;
235:
1.9 cvs 236: if (curr <= url)
1.8 cvs 237: goto finished;
238:
239: /* mark the end of the proto */
1.9 cvs 240: *curr = EOS;
241: curr--;
242: if (curr < url)
1.8 cvs 243: goto finished;
244:
1.67 cvs 245: if (*curr == TEXT(':'))
1.8 cvs 246: {
1.9 cvs 247: *curr = EOS;
248: curr--;
1.8 cvs 249: }
250: else
251: goto finished;
1.11 cvs 252:
1.9 cvs 253: if (curr < url)
1.8 cvs 254: goto finished;
1.9 cvs 255: while ((curr > url) && (isalpha (*curr)))
256: curr--;
257: *proto = curr;
1.8 cvs 258:
259: finished:;
260:
261: #ifdef AMAYA_DEBUG
262: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
263: if (*proto)
264: fprintf (stderr, "proto : %s, ", *proto);
265: if (*host)
266: fprintf (stderr, "host : %s, ", *host);
267: if (*dir)
268: fprintf (stderr, "dir : %s, ", *dir);
269: if (*file)
270: fprintf (stderr, "file : %s ", *file);
271: fprintf (stderr, "\n");
272: #endif
273:
274: }
1.3 cvs 275:
1.61 cvs 276:
277: /*----------------------------------------------------------------------
278: ExtractSuffix extract suffix from document nane.
279: ----------------------------------------------------------------------*/
280: #ifdef __STDC__
281: void ExtractSuffix (STRING aName, STRING aSuffix)
282: #else
283: void ExtractSuffix (aName, aSuffix)
284: STRING aName;
285: STRING aSuffix;
286:
287: #endif
288: {
289: int lg, i;
290: STRING ptr, oldptr;
291:
292: if (!aSuffix || !aName)
293: /* bad suffix */
294: return;
295:
296: aSuffix[0] = EOS;
297: lg = ustrlen (aName);
298: if (lg)
299: {
300: /* the name is not empty */
301: oldptr = ptr = &aName[0];
302: do
303: {
1.67 cvs 304: ptr = ustrrchr (oldptr, TEXT('.'));
1.61 cvs 305: if (ptr)
306: oldptr = &ptr[1];
307: }
308: while (ptr);
309:
310: i = (int) (oldptr) - (int) (aName); /* name length */
311: if (i > 1)
312: {
313: aName[i - 1] = EOS;
314: if (i != lg)
315: ustrcpy (aSuffix, oldptr);
316: }
317: }
318: }
319:
1.4 cvs 320: /*----------------------------------------------------------------------
1.9 cvs 321: IsHTMLName
322: returns TRUE if path points to an HTML resource.
1.4 cvs 323: ----------------------------------------------------------------------*/
1.3 cvs 324: #ifdef __STDC__
1.67 cvs 325: ThotBool IsHTMLName (const STRING path)
1.3 cvs 326: #else /* __STDC__ */
1.67 cvs 327: ThotBool IsHTMLName (path)
328: const STRING path;
1.3 cvs 329: #endif /* __STDC__ */
330: {
1.67 cvs 331: CHAR_T temppath[MAX_LENGTH];
332: CHAR_T suffix[MAX_LENGTH];
333: CHAR_T nsuffix[MAX_LENGTH];
1.5 cvs 334: int i;
335:
336: if (!path)
1.37 cvs 337: return (FALSE);
1.5 cvs 338:
1.67 cvs 339: ustrcpy (temppath, path);
1.5 cvs 340: ExtractSuffix (temppath, suffix);
341:
342: /* Normalize the suffix */
343: i = 0;
1.39 cvs 344: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 345: {
1.67 cvs 346: nsuffix[i] = utolower (suffix[i]);
1.13 cvs 347: i++;
348: }
1.5 cvs 349: nsuffix[i] = EOS;
1.67 cvs 350: if (!ustrcmp (nsuffix, TEXT("html")) ||
351: !ustrcmp (nsuffix, TEXT("htm")) ||
352: !ustrcmp (nsuffix, TEXT("shtml")) ||
353: !ustrcmp (nsuffix, TEXT("xht")) ||
354: !ustrcmp (nsuffix, TEXT("xhtm")) ||
355: !ustrcmp (nsuffix, TEXT("xhtml")))
1.39 cvs 356: return (TRUE);
1.67 cvs 357: else if (!ustrcmp (nsuffix, TEXT("gz")))
1.13 cvs 358: {
1.39 cvs 359: /* take into account compressed files */
1.13 cvs 360: ExtractSuffix (temppath, suffix);
361: /* Normalize the suffix */
362: i = 0;
1.39 cvs 363: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 364: {
1.25 cvs 365: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 366: i++;
367: }
368: nsuffix[i] = EOS;
1.67 cvs 369: if (!ustrcmp (nsuffix, TEXT("html")) ||
370: !ustrcmp (nsuffix, TEXT("htm")) ||
371: !ustrcmp (nsuffix, TEXT("shtml")) ||
372: !ustrcmp (nsuffix, TEXT("xht")) ||
373: !ustrcmp (nsuffix, TEXT("xhtm")) ||
374: !ustrcmp (nsuffix, TEXT("xhtml")))
1.63 cvs 375:
1.39 cvs 376: return (TRUE);
377: else
1.13 cvs 378: return (FALSE);
379: }
380: else
1.39 cvs 381: return (FALSE);
1.3 cvs 382: }
383:
1.4 cvs 384: /*----------------------------------------------------------------------
1.56 cvs 385: IsXMLName
386: returns TRUE if path points to an XML resource.
387: ----------------------------------------------------------------------*/
388: #ifdef __STDC__
1.67 cvs 389: ThotBool IsXMLName (const STRING path)
1.56 cvs 390: #else /* __STDC__ */
1.67 cvs 391: ThotBool IsXMLName (path)
392: const STRING path;
1.56 cvs 393: #endif /* __STDC__ */
394: {
1.67 cvs 395: CHAR_T temppath[MAX_LENGTH];
396: CHAR_T suffix[MAX_LENGTH];
1.56 cvs 397:
398: if (!path)
399: return (FALSE);
400:
1.67 cvs 401: ustrcpy (temppath, path);
1.56 cvs 402: ExtractSuffix (temppath, suffix);
403:
1.67 cvs 404: if (!ustrcasecmp (suffix, TEXT("xml")) ||
405: !ustrcasecmp (suffix, TEXT("xht")) ||
406: !ustrcmp (suffix, TEXT("xhtm")) ||
407: !ustrcmp (suffix, TEXT("xhtml")))
1.56 cvs 408: return (TRUE);
1.67 cvs 409: else if (!ustrcmp (suffix, TEXT("gz")))
1.56 cvs 410: {
411: /* take into account compressed files */
412: ExtractSuffix (temppath, suffix);
1.67 cvs 413: if (!ustrcasecmp (suffix, TEXT("xml")) ||
414: !ustrcasecmp (suffix, TEXT("xht")) ||
415: !ustrcmp (suffix, TEXT("xhtm")) ||
416: !ustrcmp (suffix, TEXT("xhtml")))
1.60 cvs 417: return (TRUE);
418: else
419: return (FALSE);
420: }
421: else
422: return (FALSE);
423: }
424:
425: /*----------------------------------------------------------------------
426: IsCSSName
427: returns TRUE if path points to an XML resource.
428: ----------------------------------------------------------------------*/
429: #ifdef __STDC__
1.67 cvs 430: ThotBool IsCSSName (const STRING path)
1.60 cvs 431: #else /* __STDC__ */
1.67 cvs 432: ThotBool IsCSSName (path)
433: const STRING path;
1.60 cvs 434: #endif /* __STDC__ */
435: {
1.67 cvs 436: CHAR_T temppath[MAX_LENGTH];
437: CHAR_T suffix[MAX_LENGTH];
1.60 cvs 438:
439: if (!path)
440: return (FALSE);
441:
1.67 cvs 442: ustrcpy (temppath, path);
1.60 cvs 443: ExtractSuffix (temppath, suffix);
444:
1.67 cvs 445: if (!ustrcasecmp (suffix, TEXT("css")))
1.60 cvs 446: return (TRUE);
1.67 cvs 447: else if (!ustrcmp (suffix, TEXT("gz")))
1.60 cvs 448: {
449: /* take into account compressed files */
450: ExtractSuffix (temppath, suffix);
1.67 cvs 451: if (!ustrcasecmp (suffix, TEXT("css")))
1.56 cvs 452: return (TRUE);
453: else
454: return (FALSE);
455: }
456: else
457: return (FALSE);
458: }
459:
460: /*----------------------------------------------------------------------
1.9 cvs 461: IsImageName
462: returns TRUE if path points to an image resource.
1.4 cvs 463: ----------------------------------------------------------------------*/
1.3 cvs 464: #ifdef __STDC__
1.67 cvs 465: ThotBool IsImageName (const STRING path)
1.3 cvs 466: #else /* __STDC__ */
1.67 cvs 467: ThotBool IsImageName (path)
468: const STRING path;
1.3 cvs 469: #endif /* __STDC__ */
470: {
1.67 cvs 471: CHAR_T temppath[MAX_LENGTH];
472: CHAR_T suffix[MAX_LENGTH];
473: CHAR_T nsuffix[MAX_LENGTH];
1.5 cvs 474: int i;
475:
476: if (!path)
1.13 cvs 477: return (FALSE);
1.5 cvs 478:
1.67 cvs 479: ustrcpy (temppath, path);
1.5 cvs 480: ExtractSuffix (temppath, suffix);
481:
482: /* Normalize the suffix */
483: i = 0;
1.39 cvs 484: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 485: {
1.67 cvs 486: nsuffix[i] = utolower (suffix[i]);
1.13 cvs 487: i++;
488: }
1.5 cvs 489: nsuffix[i] = EOS;
1.67 cvs 490: if ((!ustrcmp (nsuffix, TEXT("gif"))) || (!ustrcmp (nsuffix, TEXT("xbm"))) ||
491: (!ustrcmp (nsuffix, TEXT("xpm"))) || (!ustrcmp (nsuffix, TEXT("jpg"))) ||
492: (!ustrcmp (nsuffix, TEXT("png"))) || (!ustrcmp (nsuffix, TEXT("au"))))
1.39 cvs 493: return (TRUE);
494: return (FALSE);
1.3 cvs 495: }
496:
1.4 cvs 497: /*----------------------------------------------------------------------
1.58 cvs 498: IsImageType
499: returns TRUE if type points to an image resource.
500: ----------------------------------------------------------------------*/
501: #ifdef __STDC__
1.67 cvs 502: ThotBool IsImageType (const STRING type)
1.58 cvs 503: #else /* __STDC__ */
1.67 cvs 504: ThotBool IsImageType (type)
505: const STRING type;
1.58 cvs 506: #endif /* __STDC__ */
507: {
1.67 cvs 508: CHAR_T temptype[MAX_LENGTH];
1.58 cvs 509: int i;
510:
511: if (!type)
512: return (FALSE);
513:
1.67 cvs 514: ustrcpy (temptype, type);
1.58 cvs 515: /* Normalize the type */
516: i = 0;
517: while (temptype[i] != EOS)
518: {
519: temptype[i] = tolower (temptype[i]);
520: i++;
521: }
1.67 cvs 522: if ((!ustrcmp (temptype, TEXT("gif"))) || (!ustrcmp (temptype, TEXT("x-xbitmap"))) ||
523: (!ustrcmp (temptype, TEXT("x-xpixmap"))) || (!ustrcmp (temptype, TEXT("jpeg"))) ||
524: (!ustrcmp (temptype, TEXT("png"))))
1.58 cvs 525: return (TRUE);
526: return (FALSE);
527: }
528:
529: /*----------------------------------------------------------------------
1.9 cvs 530: IsTextName
1.4 cvs 531: ----------------------------------------------------------------------*/
1.3 cvs 532: #ifdef __STDC__
1.67 cvs 533: ThotBool IsTextName (const STRING path)
1.3 cvs 534: #else /* __STDC__ */
1.67 cvs 535: ThotBool IsTextName (path)
536: const STRING path;
1.3 cvs 537:
538: #endif /* __STDC__ */
539: {
1.67 cvs 540: CHAR_T temppath[MAX_LENGTH];
541: CHAR_T suffix[MAX_LENGTH];
542: CHAR_T nsuffix[MAX_LENGTH];
1.5 cvs 543: int i;
544:
545: if (!path)
1.13 cvs 546: return (FALSE);
1.5 cvs 547:
1.67 cvs 548: ustrcpy (temppath, path);
1.5 cvs 549: ExtractSuffix (temppath, suffix);
550:
551: /* Normalize the suffix */
552: i = 0;
1.39 cvs 553: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 554: {
1.25 cvs 555: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 556: i++;
557: }
558: nsuffix[i] = EOS;
559:
1.67 cvs 560: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd"))))
1.13 cvs 561: return (TRUE);
1.67 cvs 562: else if (!ustrcmp (nsuffix, TEXT("gz")))
1.13 cvs 563: {
1.39 cvs 564: /* take into account compressed files */
1.13 cvs 565: ExtractSuffix (temppath, suffix);
566: /* Normalize the suffix */
567: i = 0;
1.39 cvs 568: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 569: {
1.25 cvs 570: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 571: i++;
572: }
573: nsuffix[i] = EOS;
1.67 cvs 574: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd"))))
1.13 cvs 575: return (TRUE);
576: else
577: return (FALSE);
578: }
579: else
580: return (FALSE);
1.3 cvs 581: }
582:
1.4 cvs 583: /*----------------------------------------------------------------------
1.9 cvs 584: IsHTTPPath
585: returns TRUE if path is in fact an http URL.
1.4 cvs 586: ----------------------------------------------------------------------*/
1.3 cvs 587: #ifdef __STDC__
1.67 cvs 588: ThotBool IsHTTPPath (const STRING path)
1.3 cvs 589: #else /* __STDC__ */
1.67 cvs 590: ThotBool IsHTTPPath (path)
591: const STRING path;
1.3 cvs 592: #endif /* __STDC__ */
593: {
1.5 cvs 594: if (!path)
595: return FALSE;
1.3 cvs 596:
1.67 cvs 597: if ((!ustrncmp (path, TEXT("http:"), 5) != 0)
598: || !ustrncmp (path, TEXT("internal:"), 9))
1.58 cvs 599: return TRUE;
600: return FALSE;
1.3 cvs 601: }
602:
1.4 cvs 603: /*----------------------------------------------------------------------
1.9 cvs 604: IsWithParameters
605: returns TRUE if url has a concatenated query string.
1.4 cvs 606: ----------------------------------------------------------------------*/
1.3 cvs 607: #ifdef __STDC__
1.66 cvs 608: ThotBool IsWithParameters (const char *url)
1.3 cvs 609: #else /* __STDC__ */
1.66 cvs 610: ThotBool IsWithParameters (url)
1.34 cvs 611: const char *url;
1.3 cvs 612: #endif /* __STDC__ */
613: {
1.5 cvs 614: int i;
1.3 cvs 615:
1.9 cvs 616: if ((!url) || (url[0] == EOS))
1.5 cvs 617: return FALSE;
1.3 cvs 618:
1.9 cvs 619: i = strlen (url) - 1;
620: while (i > 0 && url[i--] != '?')
1.5 cvs 621: if (i < 0)
622: return FALSE;
1.3 cvs 623:
1.5 cvs 624: /* There is a parameter */
625: return TRUE;
1.3 cvs 626: }
627:
1.4 cvs 628: /*----------------------------------------------------------------------
1.9 cvs 629: IsW3Path
630: returns TRUE if path is in fact a URL.
1.4 cvs 631: ----------------------------------------------------------------------*/
1.3 cvs 632: #ifdef __STDC__
1.67 cvs 633: ThotBool IsW3Path (const STRING path)
1.3 cvs 634: #else /* __STDC__ */
1.67 cvs 635: ThotBool IsW3Path (path)
636: const STRING path;
1.3 cvs 637: #endif /* __STDC__ */
638: {
1.72 cvs 639: if (ustrncmp (path, TEXT("http:"), 5) && ustrncmp (path, TEXT("ftp:"), 4) &&
640: ustrncmp (path, TEXT("telnet:"), 7) && ustrncmp (path, TEXT("wais:"), 5) &&
641: ustrncmp (path, TEXT("news:"), 5) && ustrncmp (path, TEXT("gopher:"), 7) &&
642: ustrncmp (path, TEXT("mailto:"), 7) && ustrncmp (path, TEXT("archie:"), 7))
643: return FALSE;
644: return TRUE;
1.3 cvs 645: }
646:
1.4 cvs 647: /*----------------------------------------------------------------------
1.9 cvs 648: IsValidProtocol
649: returns true if the url protocol is supported by Amaya.
1.4 cvs 650: ----------------------------------------------------------------------*/
1.3 cvs 651: #ifdef __STDC__
1.67 cvs 652: ThotBool IsValidProtocol (const STRING url)
1.3 cvs 653: #else /* __STDC__ */
1.67 cvs 654: ThotBool IsValidProtocol (url)
655: const STRING url;
1.3 cvs 656: #endif /* __STDC__ */
657: {
1.67 cvs 658: if (!ustrncmp (url, TEXT("http:"), 5)
1.69 cvs 659: || !ustrncmp (url, TEXT("internal:"), 9)
1.70 cvs 660: || !ustrncmp (url, TEXT("ftp:"), 4))
1.22 cvs 661: /* experimental */
1.58 cvs 662: /*** || !strncmp (url, "ftp:", 4) ***/
1.24 cvs 663: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 664: return (TRUE);
1.5 cvs 665: else
1.8 cvs 666: return (FALSE);
1.3 cvs 667: }
668:
1.31 cvs 669:
670: /*----------------------------------------------------------------------
671: GetBaseURL
672: normalizes orgName according to a base associated with doc, and
673: following the standard URL format rules.
674: The function returns the base used to solve relative URL and SRC:
675: - the base of the document,
676: - or the document path (without document name).
677: ----------------------------------------------------------------------*/
678: #ifdef __STDC__
1.67 cvs 679: STRING GetBaseURL (Document doc)
1.31 cvs 680: #else /* __STDC__ */
1.67 cvs 681: STRING GetBaseURL (doc)
1.31 cvs 682: Document doc;
683: #endif /* __STDC__ */
684: {
685: Element el;
686: ElementType elType;
687: AttributeType attrType;
688: Attribute attr;
1.67 cvs 689: STRING ptr, basename;
1.31 cvs 690: int length;
691:
1.57 cvs 692: /* @@@ irene */
693: if (!DocumentURLs[doc])
694: return NULL;
1.67 cvs 695: basename = TtaAllocString (MAX_LENGTH);
696: ustrncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1.39 cvs 697: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 698: length = MAX_LENGTH -1;
699: /* get the root element */
700: el = TtaGetMainRoot (doc);
701: /* search the BASE element */
702: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.65 cvs 703: elType.ElTypeNum = HTML_EL_HEAD;
704: el = TtaSearchTypedElement (elType, SearchForward, el);
705: if (el)
706: {
707: elType.ElTypeNum = HTML_EL_BASE;
708: el = TtaSearchTypedElement (elType, SearchInTree, el);
709: }
1.31 cvs 710: if (el)
711: {
712: /* The document has a BASE element -> Get the HREF attribute */
713: attrType.AttrSSchema = elType.ElSSchema;
714: attrType.AttrTypeNum = HTML_ATTR_HREF_;
715: attr = TtaGetAttribute (el, attrType);
716: if (attr)
717: {
718: /* Use the base path of the document */
719: TtaGiveTextAttributeValue (attr, basename, &length);
720: /* base and orgName have to be separated by a DIR_SEP */
721: length--;
1.43 cvs 722: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 723: /* verify if the base has the form "protocol://server:port" */
724: {
1.67 cvs 725: ptr = AmayaParseUrl (basename, _EMPTYSTR_, AMAYA_PARSE_ACCESS |
1.33 cvs 726: AMAYA_PARSE_HOST |
727: AMAYA_PARSE_PUNCTUATION);
1.67 cvs 728: if (ptr && !ustrcmp (ptr, basename))
1.31 cvs 729: {
1.43 cvs 730: /* it has this form, we complete it by adding a URL_STR */
1.67 cvs 731: if (ustrchr (basename, DIR_SEP))
732: ustrcat (basename, DIR_STR);
1.43 cvs 733: else
1.67 cvs 734: ustrcat (basename, URL_STR);
1.31 cvs 735: length++;
736: }
737: if (ptr)
738: TtaFreeMemory (ptr);
739: }
740: }
1.33 cvs 741: }
742:
1.31 cvs 743: /* Remove anything after the last DIR_SEP char. If no such char is found,
744: * then search for the first ":" char, hoping that what's before that is a
745: * protocol. If found, end the string there. If neither char is found,
746: * then discard the whole base element.
747: */
1.67 cvs 748: length = ustrlen (basename) - 1;
1.31 cvs 749: /* search for the last DIR_SEP char */
1.43 cvs 750: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 751: length--;
752: if (length >= 0)
753: /* found the last DIR_SEP char, end the string there */
754: basename[length + 1] = EOS;
755: else
756: /* search for the first PATH_STR char */
757: {
1.67 cvs 758: for (length = 0; basename[length] != TEXT(':') &&
1.31 cvs 759: basename[length] != EOS; length ++);
1.67 cvs 760: if (basename[length] == TEXT(':'))
1.31 cvs 761: /* found, so end the string there */
762: basename[length + 1] = EOS;
763: else
764: /* not found, discard the base */
765: basename[0] = EOS;
766: }
767: return (basename);
768: }
769:
770:
1.4 cvs 771: /*----------------------------------------------------------------------
1.40 cvs 772: GetLocalPath
773: Allocate and return the local document path associated to the url
774: ----------------------------------------------------------------------*/
775: #ifdef __STDC__
1.67 cvs 776: STRING GetLocalPath (Document doc, STRING url)
1.40 cvs 777: #else /* __STDC__ */
1.67 cvs 778: STRING GetLocalPath (doc, url)
1.40 cvs 779: Document doc;
1.67 cvs 780: STRING url;
1.40 cvs 781: #endif /* __STDC__ */
782: {
1.67 cvs 783: STRING ptr, n;
784: STRING documentname;
785: CHAR_T url_sep;
1.40 cvs 786: int len;
1.67 cvs 787: STRING tmpDir, tmpName;
788: ThotBool noFile;
1.40 cvs 789:
790: if (url != NULL)
791: {
792: /* check whether the file name exists */
1.67 cvs 793: len = ustrlen (url) - 1;
1.71 cvs 794: if (IsW3Path (url))
1.67 cvs 795: url_sep = TEXT('/');
1.41 cvs 796: else
797: url_sep = DIR_SEP;
798: noFile = (url[len] == url_sep);
1.40 cvs 799: if (noFile)
800: url[len] = EOS;
1.67 cvs 801: /* ptr = TtaAllocString (MAX_LENGTH);
802: documentname = TtaAllocString (MAX_LENGTH); */
803: tmpDir = TtaAllocString (MAX_LENGTH );
804: tmpName = TtaAllocString(MAX_LENGTH );
805: TtaExtractName (url, tmpDir, tmpName);
806: ptr = tmpDir;
807: documentname = tmpName;
808:
809: usprintf (ptr, TEXT("%s%s%d%s"), TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 810: if (!TtaCheckDirectory (ptr))
811: /* directory did not exist */
1.72 cvs 812: TtaMakeDirectory (ptr);
1.47 cvs 813:
814: /* don't include the query string within document name */
1.67 cvs 815: n = ustrrchr(documentname, TEXT('?'));
1.47 cvs 816: if (n != NULL)
817: *n = EOS;
1.46 cvs 818: /* don't include ':' within document name */
1.67 cvs 819: n = ustrchr (documentname, TEXT(':'));
1.46 cvs 820: if (n != NULL)
821: *n = EOS;
1.69 cvs 822: /* if after all this operations document name
823: is empty, let's use noname.html instead */
824: if (documentname[0] == EOS)
825: ustrcat (ptr, TEXT("noname.html"));
826: else
827: ustrcat (ptr, documentname);
1.40 cvs 828: TtaFreeMemory (documentname);
829: /* restore the url */
830: if (noFile)
1.41 cvs 831: url[len] = url_sep;
1.40 cvs 832: return (ptr);
833: }
834: else
835: return (NULL);
836: }
837:
1.73 cvs 838: /*----------------------------------------------------------------------
839: ConvertFileURL
840: If the URL starts with file: prefix, it removes the protocol so that we
841: can use it as a local filename
842: ----------------------------------------------------------------------*/
843: #ifdef __STDC__
844: void ConvertFileURL (STRING url)
845: #else
846: void ConvertFileURL (url)
847: STRING url
848: #endif /* __STDC__ */
849: {
850: if (!ustrncasecmp (url, TEXT("file:"), 5))
851: ustrcpy (url, url + 5);
852: }
1.40 cvs 853:
854: /*----------------------------------------------------------------------
1.9 cvs 855: NormalizeURL
856: normalizes orgName according to a base associated with doc, and
857: following the standard URL format rules.
1.53 cvs 858: if doc is 0 and otherPath not NULL, normalizes orgName according to this
859: other path.
1.9 cvs 860: The function returns the new complete and normalized URL
1.12 cvs 861: or file name path (newName) and the name of the document (docName).
1.9 cvs 862: N.B. If the function can't find out what's the docName, it assigns
863: the name "noname.html".
1.4 cvs 864: ----------------------------------------------------------------------*/
1.3 cvs 865: #ifdef __STDC__
1.67 cvs 866: void NormalizeURL (STRING orgName, Document doc, STRING newName, STRING docName, STRING otherPath)
1.3 cvs 867: #else /* __STDC__ */
1.53 cvs 868: void NormalizeURL (orgName, doc, newName, docName, otherPath)
1.67 cvs 869: STRING orgName;
1.3 cvs 870: Document doc;
1.67 cvs 871: STRING newName;
872: STRING docName;
873: STRING otherPath;
1.3 cvs 874: #endif /* __STDC__ */
875: {
1.67 cvs 876: STRING basename;
877: CHAR_T tempOrgName[MAX_LENGTH];
878: STRING ptr;
879: CHAR_T used_sep;
880: int length;
1.66 cvs 881: ThotBool check;
1.5 cvs 882:
1.44 cvs 883: # ifdef _WINDOWS
884: int ndx;
885: # endif /* _WINDOWS */
886:
1.5 cvs 887: if (!newName || !docName)
888: return;
1.18 cvs 889:
1.32 cvs 890: if (doc != 0)
1.53 cvs 891: basename = GetBaseURL (doc);
892: else if (otherPath != NULL)
893: basename = TtaStrdup (otherPath);
1.32 cvs 894: else
1.53 cvs 895: basename = NULL;
1.32 cvs 896:
1.18 cvs 897: /*
1.31 cvs 898: * Clean orgName
899: * Make sure we have a complete orgName, without any leading or trailing
900: * white spaces, or trailinbg new lines
901: */
1.5 cvs 902: ptr = orgName;
1.18 cvs 903: /* skip leading white space and new line characters */
1.67 cvs 904: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
905: ustrncpy (tempOrgName, ptr, MAX_LENGTH -1);
1.39 cvs 906: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 907: /*
1.31 cvs 908: * Make orgName a complete URL
909: * If the URL does not include a protocol, then try to calculate
910: * one using the doc's base element (if it exists),
911: */
1.53 cvs 912: if (tempOrgName[0] == EOS)
913: {
914: newName[0] = EOS;
915: TtaFreeMemory (basename);
916: return;
917: }
1.49 cvs 918:
919: /* clean trailing white space */
1.67 cvs 920: length = ustrlen (tempOrgName) - 1;
1.53 cvs 921: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
922: {
923: tempOrgName[length] = EOS;
924: length--;
925: }
1.50 cvs 926:
1.55 cvs 927: /* remove extra dot (which dot???) */
928: /* ugly, but faster than a strcmp */
1.67 cvs 929: if (tempOrgName[length] == TEXT('.')
930: && (length == 0 || tempOrgName[length-1] != TEXT('.')))
1.55 cvs 931: tempOrgName[length] = EOS;
1.50 cvs 932:
1.53 cvs 933: if (IsW3Path (tempOrgName))
934: {
935: /* the name is complete, go to the Sixth Step */
1.67 cvs 936: ustrcpy (newName, tempOrgName);
1.53 cvs 937: SimplifyUrl (&newName);
938: /* verify if the URL has the form "protocol://server:port" */
1.67 cvs 939: ptr = AmayaParseUrl (newName, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
940: if (ptr && !ustrcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
941: ustrcat (newName, URL_STR);
1.49 cvs 942:
1.53 cvs 943: if (ptr)
1.50 cvs 944: TtaFreeMemory (ptr);
1.53 cvs 945: }
946: else if ( basename == NULL)
947: /* the name is complete, go to the Sixth Step */
1.67 cvs 948: ustrcpy (newName, tempOrgName);
1.53 cvs 949: else
950: {
1.31 cvs 951: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 952: # ifdef _WINDOWS
1.53 cvs 953: if (!IsW3Path (basename))
954: {
1.67 cvs 955: length = ustrlen (tempOrgName);
1.53 cvs 956: for (ndx = 0; ndx < length; ndx++)
1.67 cvs 957: if (tempOrgName [ndx] == TEXT('/'))
958: tempOrgName [ndx] = TEXT('\\');
1.53 cvs 959: }
1.44 cvs 960: # endif /* _WINDOWS */
1.25 cvs 961: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 962: if (ptr)
963: {
964: SimplifyUrl (&ptr);
1.67 cvs 965: ustrcpy (newName, ptr);
1.53 cvs 966: TtaFreeMemory (ptr);
967: }
968: else
969: newName[0] = EOS;
970: }
1.36 cvs 971:
972: TtaFreeMemory (basename);
1.18 cvs 973: /*
1.31 cvs 974: * Prepare the docname that will refer to this ressource in the
975: * .amaya directory. If the new URL finishes on DIR_SEP, then use
976: * noname.html as a default ressource name
1.18 cvs 977: */
1.53 cvs 978: if (newName[0] != EOS)
979: {
1.67 cvs 980: length = ustrlen (newName) - 1;
1.53 cvs 981: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
982: {
983: used_sep = newName[length];
984: check = TRUE;
985: while (check)
986: {
1.50 cvs 987: length--;
988: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 989: length--;
1.67 cvs 990: if (!ustrncmp (&newName[length+1], TEXT(".."), 2))
1.53 cvs 991: {
992: newName[length+1] = EOS;
993: /* remove also previous directory */
994: length--;
995: while (length >= 0 && newName[length] != used_sep)
996: length--;
1.67 cvs 997: if (ustrncmp (&newName[length+1], TEXT("//"), 2))
1.53 cvs 998: /* don't remove server name */
1.50 cvs 999: newName[length+1] = EOS;
1.53 cvs 1000: }
1.67 cvs 1001: else if (!ustrncmp (&newName[length+1], TEXT("."), 1))
1.53 cvs 1002: newName[length+1] = EOS;
1.50 cvs 1003: else
1.53 cvs 1004: check = FALSE;
1005: }
1.67 cvs 1006: ustrcpy (docName, TEXT("noname.html"));
1.53 cvs 1007: /* docname was not comprised inside the URL, so let's */
1008: /* assign the default ressource name */
1.67 cvs 1009: ustrcpy (docName, TEXT("noname.html"));
1.53 cvs 1010: }
1011: else
1012: { /* docname is comprised inside the URL */
1.50 cvs 1013: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP)
1.53 cvs 1014: length--;
1015: if (length < 0)
1.67 cvs 1016: ustrcpy (docName, newName);
1.53 cvs 1017: else
1.67 cvs 1018: ustrcpy (docName, &newName[length+1]);
1.53 cvs 1019: }
1020: }
1021: else
1022: docName[0] = EOS;
1.18 cvs 1023: }
1.3 cvs 1024:
1.4 cvs 1025: /*----------------------------------------------------------------------
1.9 cvs 1026: IsSameHost
1.4 cvs 1027: ----------------------------------------------------------------------*/
1.3 cvs 1028: #ifdef __STDC__
1.67 cvs 1029: ThotBool IsSameHost (const STRING url1, const STRING url2)
1.3 cvs 1030: #else /* __STDC__ */
1.67 cvs 1031: ThotBool IsSameHost (url1, url2)
1032: const STRING url1;
1033: const STRING url2;
1.3 cvs 1034: #endif /* __STDC__ */
1035: {
1.67 cvs 1036: STRING basename_ptr1, basename_ptr2;
1037: ThotBool result;
1.3 cvs 1038:
1.67 cvs 1039: basename_ptr1 = AmayaParseUrl (url1, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1040: basename_ptr2 = AmayaParseUrl (url2, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1041:
1.67 cvs 1042: if (ustrcmp (basename_ptr1, basename_ptr2))
1.8 cvs 1043: result = FALSE;
1.5 cvs 1044: else
1.8 cvs 1045: result = TRUE;
1.3 cvs 1046:
1.25 cvs 1047: TtaFreeMemory (basename_ptr1);
1048: TtaFreeMemory (basename_ptr2);
1.5 cvs 1049: return (result);
1.3 cvs 1050: }
1051:
1052:
1.4 cvs 1053: /*----------------------------------------------------------------------
1.22 cvs 1054: HasKnownFileSuffix
1055: returns TRUE if path points to a file ending with a suffix.
1056: ----------------------------------------------------------------------*/
1057: #ifdef __STDC__
1.67 cvs 1058: ThotBool HasKnownFileSuffix (const STRING path)
1.22 cvs 1059: #else /* __STDC__ */
1.67 cvs 1060: ThotBool HasKnownFileSuffix (path)
1061: const STRING path;
1.22 cvs 1062: #endif /* __STDC__ */
1063: {
1.67 cvs 1064: STRING root;
1065: CHAR_T temppath[MAX_LENGTH];
1066: CHAR_T suffix[MAX_LENGTH];
1.22 cvs 1067:
1.67 cvs 1068: if (!path || path[0] == EOS || path[ustrlen(path)] == DIR_SEP)
1.22 cvs 1069: return (FALSE);
1070:
1.67 cvs 1071: root = AmayaParseUrl(path, _EMPTYSTR_, AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1072:
1073: if (root)
1074: {
1.67 cvs 1075: ustrcpy (temppath, root);
1.25 cvs 1076: TtaFreeMemory (root);
1.22 cvs 1077: /* Get the suffix */
1078: ExtractSuffix (temppath, suffix);
1079:
1080: if( suffix[0] == EOS)
1081: /* no suffix */
1082: return (FALSE);
1083:
1084: /* Normalize the suffix */
1085: ConvertToLowerCase (suffix);
1086:
1.67 cvs 1087: if (!ustrcmp (suffix, TEXT("gz")))
1.22 cvs 1088: /* skip the compressed suffix */
1089: {
1090: ExtractSuffix (temppath, suffix);
1091: if(suffix[0] == EOS)
1092: /* no suffix */
1093: return (FALSE);
1094: /* Normalize the suffix */
1095: ConvertToLowerCase (suffix);
1096: }
1097:
1.67 cvs 1098: if (ustrcmp (suffix, TEXT("gif")) &&
1099: ustrcmp (suffix, TEXT("xbm")) &&
1100: ustrcmp (suffix, TEXT("xpm")) &&
1101: ustrcmp (suffix, TEXT("jpg")) &&
1102: ustrcmp (suffix, TEXT("pdf")) &&
1103: ustrcmp (suffix, TEXT("png")) &&
1104: ustrcmp (suffix, TEXT("tgz")) &&
1105: ustrcmp (suffix, TEXT("xpg")) &&
1106: ustrcmp (suffix, TEXT("xpd")) &&
1107: ustrcmp (suffix, TEXT("ps")) &&
1108: ustrcmp (suffix, TEXT("au")) &&
1109: ustrcmp (suffix, TEXT("html")) &&
1110: ustrcmp (suffix, TEXT("htm")) &&
1111: ustrcmp (suffix, TEXT("shtml")) &&
1112: ustrcmp (suffix, TEXT("xht")) &&
1113: ustrcmp (suffix, TEXT("xhtm")) &&
1114: ustrcmp (suffix, TEXT("xhtml")) &&
1115: ustrcmp (suffix, TEXT("txt")) &&
1116: ustrcmp (suffix, TEXT("css")) &&
1117: ustrcmp (suffix, TEXT("eps")))
1.22 cvs 1118: return (FALSE);
1119: else
1120: return (TRUE);
1121: }
1122: else
1123: return (FALSE);
1124: }
1125:
1126:
1127: /*----------------------------------------------------------------------
1.24 cvs 1128: ChopURL
1129: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1130: If inputURL is bigger than that size, outputURL receives
1131: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1132: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1133: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1134: copied into outputURL.
1135: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1136: chars.
1137: ----------------------------------------------------------------------*/
1138: #ifdef __STDC__
1.34 cvs 1139: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1140: #else
1141: void ChopURL (outputURL, inputURL)
1.34 cvs 1142: char *outputURL;
1143: const char *inputURL;
1.24 cvs 1144: #endif
1.22 cvs 1145:
1.24 cvs 1146: {
1147: int len;
1.9 cvs 1148:
1.24 cvs 1149: len = strlen (inputURL);
1150: if (len <= MAX_PRINT_URL_LENGTH)
1.29 cvs 1151: strcpy (outputURL, inputURL);
1.24 cvs 1152: else
1153: /* make a truncated urlName on the status window */
1154: {
1155: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1156: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1157: strcat (outputURL, "...");
1158: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1159: }
1.25 cvs 1160: }
1161:
1162:
1163: /*----------------------------------------------------------------------
1164: scan
1.47 cvs 1165: Scan a filename for its constituents
1.25 cvs 1166: -----------------------------------
1167:
1168: On entry,
1169: name points to a document name which may be incomplete.
1170: On exit,
1171: absolute or relative may be nonzero (but not both).
1172: host, fragment and access may be nonzero if they were specified.
1173: Any which are nonzero point to zero terminated strings.
1174: ----------------------------------------------------------------------*/
1175: #ifdef __STDC__
1.67 cvs 1176: static void scan (STRING name, HTURI * parts)
1.25 cvs 1177: #else /* __STDC__ */
1178: static void scan (name, parts)
1.67 cvs 1179: STRING name;
1180: HTURI *parts;
1.25 cvs 1181:
1182: #endif /* __STDC__ */
1183: {
1.67 cvs 1184: STRING p;
1185: STRING after_access = name;
1.32 cvs 1186:
1.43 cvs 1187: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1188: /* Look for fragment identifier */
1.67 cvs 1189: if ((p = ustrchr(name, TEXT('#'))) != NULL)
1.28 cvs 1190: {
1.67 cvs 1191: *p++ = TEXT('\0');
1.28 cvs 1192: parts->fragment = p;
1.25 cvs 1193: }
1194:
1.28 cvs 1195: for (p=name; *p; p++)
1196: {
1.67 cvs 1197: if (*p == URL_SEP || *p == DIR_SEP || *p == TEXT('#') || *p == TEXT('?'))
1.28 cvs 1198: break;
1.67 cvs 1199: if (*p == TEXT(':'))
1.28 cvs 1200: {
1201: *p = 0;
1202: parts->access = after_access; /* Scheme has been specified */
1203:
1204: /* The combination of gcc, the "-O" flag and the HP platform is
1205: unhealthy. The following three lines is a quick & dirty fix, but is
1206: not recommended. Rather, turn off "-O". */
1207:
1208: /* after_access = p;*/
1209: /* while (*after_access == 0)*/
1210: /* after_access++;*/
1211: after_access = p+1;
1.67 cvs 1212: if (!ustrcasecmp(TEXT("URL"), parts->access))
1.28 cvs 1213: /* Ignore IETF's URL: pre-prefix */
1214: parts->access = NULL;
1215: else
1.25 cvs 1216: break;
1217: }
1218: }
1219:
1220: p = after_access;
1.43 cvs 1221: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1222: {
1.43 cvs 1223: if (p[1] == URL_SEP)
1.28 cvs 1224: {
1.25 cvs 1225: parts->host = p+2; /* host has been specified */
1.28 cvs 1226: *p = 0; /* Terminate access */
1227: /* look for end of host name if any */
1.67 cvs 1228: p = ustrchr (parts->host, URL_SEP);
1.28 cvs 1229: if (p)
1230: {
1.43 cvs 1231: *p = EOS; /* Terminate host */
1.25 cvs 1232: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1233: }
1234: }
1235: else
1236: /* Root found but no host */
1237: parts->absolute = p+1;
1238: }
1239: else
1240: {
1.25 cvs 1241: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1242: }
1.25 cvs 1243: }
1244:
1245:
1246: /*----------------------------------------------------------------------
1.28 cvs 1247: AmayaParseUrl: parse a Name relative to another name
1248:
1249: This returns those parts of a name which are given (and requested)
1250: substituting bits from the related name where necessary.
1.25 cvs 1251:
1.28 cvs 1252: On entry,
1.25 cvs 1253: aName A filename given
1254: relatedName A name relative to which aName is to be parsed. Give
1255: it an empty string if aName is absolute.
1256: wanted A mask for the bits which are wanted.
1257:
1.28 cvs 1258: On exit,
1.25 cvs 1259: returns A pointer to a malloc'd string which MUST BE FREED
1260: ----------------------------------------------------------------------*/
1261: #ifdef __STDC__
1.67 cvs 1262: STRING AmayaParseUrl (const STRING aName, STRING relatedName, int wanted)
1.25 cvs 1263: #else /* __STDC__ */
1.67 cvs 1264: STRING AmayaParseUrl (aName, relatedName, wanted)
1265: const STRING aName;
1266: STRING relatedName;
1.28 cvs 1267: int wanted;
1.25 cvs 1268:
1269: #endif /* __STDC__ */
1270: {
1.67 cvs 1271: STRING return_value;
1272: CHAR_T result[MAX_LENGTH];
1273: CHAR_T name[MAX_LENGTH];
1274: CHAR_T rel[MAX_LENGTH];
1275: STRING p, access;
1.29 cvs 1276: HTURI given, related;
1277: int len;
1.67 cvs 1278: CHAR_T used_sep;
1279: STRING used_str;
1.32 cvs 1280:
1.67 cvs 1281: if (ustrchr (aName, DIR_SEP) || ustrchr (relatedName, DIR_SEP))
1.33 cvs 1282: {
1.42 cvs 1283: used_str = DIR_STR;
1284: used_sep = DIR_SEP;
1.33 cvs 1285: }
1.32 cvs 1286: else
1.33 cvs 1287: {
1.42 cvs 1288: used_str = URL_STR;
1289: used_sep = URL_SEP;
1.33 cvs 1290: }
1.32 cvs 1291:
1.29 cvs 1292: /* Make working copies of input strings to cut up: */
1293: return_value = NULL;
1294: result[0] = 0; /* Clear string */
1.67 cvs 1295: ustrcpy (name, aName);
1.29 cvs 1296: if (relatedName != NULL)
1.67 cvs 1297: ustrcpy (rel, relatedName);
1.29 cvs 1298: else
1299: relatedName[0] = EOS;
1300:
1301: scan (name, &given);
1302: scan (rel, &related);
1303: access = given.access ? given.access : related.access;
1304: if (wanted & AMAYA_PARSE_ACCESS)
1305: if (access)
1306: {
1.67 cvs 1307: ustrcat (result, access);
1.29 cvs 1308: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1309: ustrcat (result, TEXT(":"));
1.29 cvs 1310: }
1311:
1312: if (given.access && related.access)
1313: /* If different, inherit nothing. */
1.67 cvs 1314: if (ustrcmp (given.access, related.access) != 0)
1.29 cvs 1315: {
1316: related.host = 0;
1317: related.absolute = 0;
1318: related.relative = 0;
1319: related.fragment = 0;
1320: }
1321:
1322: if (wanted & AMAYA_PARSE_HOST)
1323: if(given.host || related.host)
1324: {
1325: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1326: ustrcat (result, TEXT("//"));
1327: ustrcat (result, given.host ? given.host : related.host);
1.29 cvs 1328: }
1329:
1330: if (given.host && related.host)
1331: /* If different hosts, inherit no path. */
1.67 cvs 1332: if (ustrcmp (given.host, related.host) != 0)
1.29 cvs 1333: {
1334: related.absolute = 0;
1335: related.relative = 0;
1336: related.fragment = 0;
1337: }
1338:
1339: if (wanted & AMAYA_PARSE_PATH)
1340: {
1341: if (given.absolute)
1342: {
1343: /* All is given */
1344: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1345: ustrcat (result, used_str);
1346: ustrcat (result, given.absolute);
1.25 cvs 1347: }
1.29 cvs 1348: else if (related.absolute)
1349: {
1350: /* Adopt path not name */
1.67 cvs 1351: ustrcat (result, used_str);
1352: ustrcat (result, related.absolute);
1.29 cvs 1353: if (given.relative)
1354: {
1355: /* Search part? */
1.67 cvs 1356: p = ustrchr (result, TEXT('?'));
1.29 cvs 1357: if (!p)
1.67 cvs 1358: p=result+ustrlen(result)-1;
1.33 cvs 1359: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1360: /* Remove filename */
1361: p[1]=0;
1362: /* Add given one */
1.67 cvs 1363: ustrcat (result, given.relative);
1.25 cvs 1364: }
1365: }
1.29 cvs 1366: else if (given.relative)
1367: /* what we've got */
1.67 cvs 1368: ustrcat (result, given.relative);
1.29 cvs 1369: else if (related.relative)
1.67 cvs 1370: ustrcat (result, related.relative);
1.29 cvs 1371: else
1372: /* No inheritance */
1.67 cvs 1373: ustrcat (result, used_str);
1.25 cvs 1374: }
1.29 cvs 1375:
1376: if (wanted & AMAYA_PARSE_ANCHOR)
1377: if (given.fragment || related.fragment)
1378: {
1379: if (given.absolute && given.fragment)
1380: {
1381: /*Fixes for relURLs...*/
1382: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1383: ustrcat (result, TEXT("#"));
1384: ustrcat (result, given.fragment);
1.29 cvs 1385: }
1386: else if (!(given.absolute) && !(given.fragment))
1.67 cvs 1387: ustrcat (result, _EMPTYSTR_);
1.29 cvs 1388: else
1389: {
1390: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1391: ustrcat (result, TEXT("#"));
1392: ustrcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1393: }
1394: }
1.67 cvs 1395: len = ustrlen (result);
1396: if ((return_value = TtaAllocString (len + 1)) != NULL)
1397: ustrcpy (return_value, result);
1.29 cvs 1398: return (return_value); /* exactly the right length */
1.25 cvs 1399: }
1400:
1401: /*----------------------------------------------------------------------
1402: HTCanon
1403: Canonicalizes the URL in the following manner starting from the host
1404: pointer:
1405:
1406: 1) The host name is converted to lowercase
1407: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1408:
1409: Return: OK The position of the current path part of the URL
1410: which might be the old one or a new one.
1411:
1412: ----------------------------------------------------------------------*/
1413: #ifdef __STDC__
1.67 cvs 1414: static STRING HTCanon (STRING* filename, STRING host)
1.25 cvs 1415: #else /* __STDC__ */
1.67 cvs 1416: static STRING HTCanon (filename, host)
1417: STRING *filename;
1418: STRING host;
1.25 cvs 1419: #endif /* __STDC__ */
1420: {
1.67 cvs 1421: STRING newname = NULL;
1422: STRING port;
1423: STRING strptr;
1424: STRING path;
1425: STRING access = host-3;
1426: CHAR_T used_sep;
1.32 cvs 1427:
1428:
1.67 cvs 1429: if (*filename && ustrchr (*filename, URL_SEP))
1.33 cvs 1430: {
1431: used_sep = URL_SEP;
1432: }
1433: else
1434: {
1435: used_sep = DIR_SEP;
1436: }
1.32 cvs 1437:
1.33 cvs 1438: while (access>*filename && *(access-1)!= used_sep) /* Find access method */
1.25 cvs 1439: access--;
1.67 cvs 1440: if ((path = ustrchr(host, used_sep)) == NULL) /* Find path */
1441: path = host + ustrlen(host);
1442: if ((strptr = ustrchr(host, TEXT('@'))) != NULL && strptr<path) /* UserId */
1.25 cvs 1443: host = strptr;
1.67 cvs 1444: if ((port = ustrchr(host, TEXT(':'))) != NULL && port>path) /* Port number */
1.25 cvs 1445: port = NULL;
1446:
1447: strptr = host; /* Convert to lower-case */
1.33 cvs 1448: while (strptr<path)
1449: {
1.25 cvs 1450: *strptr = tolower(*strptr);
1451: strptr++;
1.33 cvs 1452: }
1.25 cvs 1453:
1454: /* Does the URL contain a full domain name? This also works for a
1455: numerical host name. The domain name is already made lower-case
1456: and without a trailing dot. */
1457: {
1.67 cvs 1458: STRING dot = port ? port : path;
1459: if (dot > *filename && *--dot == TEXT('.'))
1.33 cvs 1460: {
1.67 cvs 1461: STRING orig=dot, dest=dot+1;
1.33 cvs 1462: while((*orig++ = *dest++));
1463: if (port) port--;
1464: path--;
1.25 cvs 1465: }
1466: }
1467: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1468: if (port)
1469: {
1470: if (!*(port+1) || *(port+1)==used_sep)
1471: {
1472: if (!newname)
1473: {
1.67 cvs 1474: STRING orig=port, dest=port+1;
1.25 cvs 1475: while((*orig++ = *dest++));
1.33 cvs 1476: }
1477: }
1.67 cvs 1478: else if ((!ustrncmp(access, TEXT("http"), 4) &&
1479: (*(port+1) == TEXT('8') && *(port+2) == TEXT('0') && (*(port+3) == used_sep || !*(port + 3)))) ||
1480: (!ustrncmp (access, TEXT("gopher"), 6) &&
1481: (*(port+1) == TEXT('7') && *(port+2) == TEXT('0') && (*(port+3) == used_sep || !*(port+3)))) ||
1482: (!ustrncmp(access, TEXT("ftp"), 3) &&
1483: (*(port+1) == TEXT('2') && *(port + 2) == TEXT('1') && (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1484: if (!newname)
1485: {
1.67 cvs 1486: STRING orig=port, dest=port+3;
1.33 cvs 1487: while((*orig++ = *dest++));
1488: /* Update path position, Henry Minsky */
1489: path -= 3;
1.25 cvs 1490: }
1.33 cvs 1491: }
1492: else if (newname)
1.67 cvs 1493: ustrncat(newname, port, (int) (path-port));
1.33 cvs 1494: }
1.25 cvs 1495:
1.33 cvs 1496: if (newname)
1497: {
1.67 cvs 1498: STRING newpath = newname + ustrlen (newname);
1499: ustrcat(newname, path);
1.25 cvs 1500: path = newpath;
1.28 cvs 1501: /* Free old copy */
1502: TtaFreeMemory(*filename);
1.25 cvs 1503: *filename = newname;
1.33 cvs 1504: }
1.25 cvs 1505: return path;
1506: }
1507:
1508:
1509: /*----------------------------------------------------------------------
1.29 cvs 1510: SimplifyUrl: simplify a URI
1.32 cvs 1511: A URI is allowed to contain the sequence xxx/../ which may be
1512: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1513: Simplification helps us recognize duplicate URIs.
1.25 cvs 1514:
1.28 cvs 1515: Thus, /etc/junk/../fred becomes /etc/fred
1516: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1517:
1.28 cvs 1518: but we should NOT change
1519: http://fred.xxx.edu/../..
1.25 cvs 1520:
1521: or ../../albert.html
1522:
1.28 cvs 1523: In order to avoid empty URLs the following URLs become:
1.25 cvs 1524:
1525: /fred/.. becomes /fred/..
1526: /fred/././.. becomes /fred/..
1527: /fred/.././junk/.././ becomes /fred/..
1528:
1.28 cvs 1529: If more than one set of `://' is found (several proxies in cascade) then
1530: only the part after the last `://' is simplified.
1.25 cvs 1531:
1.28 cvs 1532: Returns: A string which might be the old one or a new one.
1.25 cvs 1533: ----------------------------------------------------------------------*/
1534: #ifdef __STDC__
1.67 cvs 1535: void SimplifyUrl (STRING* url)
1.25 cvs 1536: #else /* __STDC__ */
1.29 cvs 1537: void SimplifyUrl (url)
1.67 cvs 1538: STRING *url;
1.25 cvs 1539: #endif /* __STDC__ */
1540: {
1.67 cvs 1541: STRING path, p;
1542: STRING newptr, access;
1543: STRING orig, dest, end;
1.28 cvs 1544:
1.67 cvs 1545: CHAR_T used_sep;
1.77 ! cvs 1546: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
! 1547: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1548:
1549:
1.28 cvs 1550: if (!url || !*url)
1551: return;
1552:
1.67 cvs 1553: if (ustrchr (*url, URL_SEP))
1.33 cvs 1554: {
1555: used_sep = URL_SEP;
1556: }
1.32 cvs 1557: else
1.33 cvs 1558: {
1559: used_sep = DIR_SEP;
1560: }
1.32 cvs 1561:
1.77 ! cvs 1562: /* should we simplify double dot? */
! 1563: path = *url;
! 1564: if (*path == TEXT('.') && *(path + 1) == TEXT('.'))
! 1565: ddot_simplify = FALSE;
! 1566: else
! 1567: ddot_simplify = TRUE;
! 1568:
1.28 cvs 1569: /* Find any scheme name */
1.67 cvs 1570: if ((path = ustrstr(*url, TEXT("://"))) != NULL)
1.33 cvs 1571: {
1572: /* Find host name */
1.28 cvs 1573: access = *url;
1574: while (access<path && (*access=tolower(*access)))
1575: access++;
1576: path += 3;
1.67 cvs 1577: while ((newptr = ustrstr(path, TEXT("://"))) != NULL)
1.28 cvs 1578: /* For proxies */
1579: path = newptr+3;
1580: /* We have a host name */
1581: path = HTCanon(url, path);
1.25 cvs 1582: }
1.67 cvs 1583: else if ((path = ustrstr(*url, TEXT(":/"))) != NULL)
1.28 cvs 1584: path += 2;
1585: else
1586: path = *url;
1.25 cvs 1587:
1.33 cvs 1588: if (*path == used_sep && *(path+1)==used_sep)
1.28 cvs 1589: /* Some URLs start //<foo> */
1590: path += 1;
1.67 cvs 1591: else if (!ustrncmp(path, TEXT("news:"), 5))
1.28 cvs 1592: {
1.67 cvs 1593: newptr = ustrchr(path+5, TEXT('@'));
1.28 cvs 1594: if (!newptr)
1595: newptr = path + 5;
1596: while (*newptr)
1597: {
1598: /* Make group or host lower case */
1599: *newptr = tolower (*newptr);
1600: newptr++;
1.25 cvs 1601: }
1.28 cvs 1602: /* Doesn't need to do any more */
1603: return;
1.25 cvs 1604: }
1.28 cvs 1605:
1606: if ((p = path))
1607: {
1.67 cvs 1608: if (!((end = ustrchr (path, TEXT(';'))) || (end = ustrchr (path, TEXT('?'))) ||
1609: (end = ustrchr (path, TEXT('#')))))
1610: end = path + ustrlen (path);
1.28 cvs 1611:
1612: /* Parse string second time to simplify */
1613: p = path;
1614: while (p < end)
1615: {
1.77 ! cvs 1616: /* if we're pointing to a char, it's safe to reactivate the ../ convertion */
! 1617: if (!ddot_simplify && *p != TEXT('.') && *p != used_sep)
! 1618: ddot_simplify = TRUE;
! 1619:
1.33 cvs 1620: if (*p==used_sep)
1.28 cvs 1621: {
1.67 cvs 1622: if (p > *url && *(p+1) == TEXT('.') && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1623: {
1624: orig = p + 1;
1.33 cvs 1625: dest = (*(p+2)!=used_sep) ? p+2 : p+3;
1.52 cvs 1626: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1627: end = orig - 1;
1628: }
1.77 ! cvs 1629: else if (ddot_simplify && *(p+1) == TEXT('.') && *(p+2) == TEXT('.')
! 1630: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1631: {
1632: newptr = p;
1.52 cvs 1633: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1634: if (*newptr == used_sep)
1635: orig = newptr + 1;
1.28 cvs 1636: else
1.52 cvs 1637: orig = newptr;
1638:
1639: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1640: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1641: end = orig-1;
1642: /* Start again with prev slash */
1643: p = newptr;
1.28 cvs 1644: }
1.33 cvs 1645: else if (*(p+1) == used_sep)
1.28 cvs 1646: {
1.33 cvs 1647: while (*(p+1) == used_sep)
1.28 cvs 1648: {
1649: orig = p;
1650: dest = p + 1;
1651: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1652: end = orig-1;
1653: }
1654: }
1655: else
1.25 cvs 1656: p++;
1.28 cvs 1657: }
1658: else
1659: p++;
1.25 cvs 1660: }
1661: }
1.51 cvs 1662:
1663: /*
1664: ** Check for host/../.. kind of things
1665: */
1.77 ! cvs 1666: if (*path == used_sep && *(path+1) == TEXT('.') && *(path+2) == TEXT('.')
! 1667: && (!*(path+3) || *(path+3) == used_sep))
1.51 cvs 1668: *(path+1) = EOS;
1669:
1.28 cvs 1670: return;
1671: }
1672:
1673:
1674: /*----------------------------------------------------------------------
1675: NormalizeFile normalizes local names.
1676: Return TRUE if target and src differ.
1677: ----------------------------------------------------------------------*/
1678: #ifdef __STDC__
1.67 cvs 1679: ThotBool NormalizeFile (STRING src, STRING target)
1.28 cvs 1680: #else
1.67 cvs 1681: ThotBool NormalizeFile (src, target)
1682: STRING src;
1683: STRING target;
1.28 cvs 1684:
1685: #endif
1686: {
1.67 cvs 1687: STRING s;
1.66 cvs 1688: ThotBool change;
1.28 cvs 1689:
1.54 cvs 1690: change = FALSE;
1.67 cvs 1691: if (ustrncmp (src, TEXT("file:"), 5) == 0)
1.28 cvs 1692: {
1693: /* remove the prefix file: */
1694: if (src[5] == EOS)
1.67 cvs 1695: ustrcpy (target, DIR_STR);
1696: else if (src[0] == TEXT('~'))
1.28 cvs 1697: {
1698: /* replace ~ */
1.74 cvs 1699: s = TtaGetEnvString ("HOME");
1.67 cvs 1700: ustrcpy (target, s);
1701: ustrcat (target, &src[5]);
1.28 cvs 1702: }
1703: else
1.67 cvs 1704: ustrcpy (target, &src[5]);
1.54 cvs 1705: change = TRUE;
1.28 cvs 1706: }
1.53 cvs 1707: # ifndef _WINDOWS
1.67 cvs 1708: else if (src[0] == TEXT('~'))
1.53 cvs 1709: {
1710: /* replace ~ */
1711: s = (char *) TtaGetEnvString ("HOME");
1712: strcpy (target, s);
1713: if (src[1] != DIR_SEP)
1714: strcat (target, DIR_STR);
1715: strcat (target, &src[1]);
1.54 cvs 1716: change = TRUE;
1.53 cvs 1717: }
1718: # endif /* _WINDOWS */
1.28 cvs 1719: else
1.67 cvs 1720: ustrcpy (target, src);
1.28 cvs 1721:
1722: /* remove /../ and /./ */
1.29 cvs 1723: SimplifyUrl (&target);
1.54 cvs 1724: if (!change)
1.67 cvs 1725: change = ustrcmp (src, target);
1.28 cvs 1726: return (change);
1.25 cvs 1727: }
1728:
1.28 cvs 1729:
1.25 cvs 1730: /*----------------------------------------------------------------------
1.31 cvs 1731: MakeRelativeURL: make relative name
1.25 cvs 1732:
1.28 cvs 1733: This function creates and returns a string which gives an expression of
1734: one address as related to another. Where there is no relation, an absolute
1735: address is retured.
1.25 cvs 1736:
1.28 cvs 1737: On entry,
1.25 cvs 1738: Both names must be absolute, fully qualified names of nodes
1739: (no fragment bits)
1740:
1.28 cvs 1741: On exit,
1.25 cvs 1742: The return result points to a newly allocated name which, if
1743: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1744: The caller is responsible for freeing the resulting name later.
1745: ----------------------------------------------------------------------*/
1746: #ifdef __STDC__
1.67 cvs 1747: STRING MakeRelativeURL (STRING aName, STRING relatedName)
1.25 cvs 1748: #else /* __STDC__ */
1.67 cvs 1749: STRING MakeRelativeURL (aName, relatedName)
1750: STRING aName;
1751: STRING relatedName;
1.25 cvs 1752: #endif /* __STDC__ */
1753: {
1.67 cvs 1754: STRING return_value;
1755: CHAR_T result[MAX_LENGTH];
1756: STRING p;
1757: STRING q;
1758: STRING after_access;
1759: STRING last_slash = NULL;
1.29 cvs 1760: int slashes, levels, len;
1761:
1.44 cvs 1762: # ifdef _WINDOWS
1763: int ndx;
1764: # endif /* _WINDOWS */
1765:
1.29 cvs 1766: if (aName == NULL || relatedName == NULL)
1767: return (NULL);
1768:
1769: slashes = 0;
1770: after_access = NULL;
1771: p = aName;
1772: q = relatedName;
1773: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1774: {
1775: /* Find extent of match */
1.67 cvs 1776: if (*p == TEXT(':'))
1.29 cvs 1777: after_access = p + 1;
1.28 cvs 1778: if (*p == DIR_SEP)
1.27 cvs 1779: {
1.29 cvs 1780: /* memorize the last slash position and count them */
1.27 cvs 1781: last_slash = p;
1782: slashes++;
1.25 cvs 1783: }
1784: }
1785:
1.31 cvs 1786: /* q, p point to the first non-matching character or zero */
1787: if (*q == EOS)
1788: {
1789: /* New name is a subset of the related name */
1790: /* exactly the right length */
1.67 cvs 1791: len = ustrlen (p);
1792: if ((return_value = TtaAllocString (len + 1)) != NULL)
1793: ustrcpy (return_value, p);
1.31 cvs 1794: }
1795: else if ((slashes < 2 && after_access == NULL)
1796: || (slashes < 3 && after_access != NULL))
1797: {
1798: /* Two names whitout common path */
1799: /* exactly the right length */
1.67 cvs 1800: len = ustrlen (aName);
1801: if ((return_value = TtaAllocString (len + 1)) != NULL)
1802: ustrcpy (return_value, aName);
1.31 cvs 1803: }
1804: else
1805: {
1806: /* Some path in common */
1.67 cvs 1807: if (slashes == 3 && ustrncmp (aName, TEXT("http:"), 5) == 0)
1.31 cvs 1808: /* just the same server */
1.67 cvs 1809: ustrcpy (result, last_slash);
1.31 cvs 1810: else
1811: {
1812: levels= 0;
1.67 cvs 1813: for (; *q && *q != TEXT('#') && *q != TEXT(';') && *q != TEXT('?'); q++)
1.31 cvs 1814: if (*q == DIR_SEP)
1815: levels++;
1816:
1.52 cvs 1817: result[0] = EOS;
1.31 cvs 1818: for (;levels; levels--)
1.67 cvs 1819: ustrcat (result, TEXT("../"));
1820: ustrcat (result, last_slash+1);
1.31 cvs 1821: }
1.52 cvs 1822:
1823: if (!*result)
1.67 cvs 1824: ustrcat (result, TEXT("./"));
1.52 cvs 1825:
1.31 cvs 1826: /* exactly the right length */
1.67 cvs 1827: len = ustrlen (result);
1828: if ((return_value = TtaAllocString (len + 1)) != NULL)
1829: ustrcpy (return_value, result);
1.52 cvs 1830:
1.25 cvs 1831: }
1.44 cvs 1832: # ifdef _WINDOWS
1.67 cvs 1833: len = ustrlen (return_value);
1.44 cvs 1834: for (ndx = 0; ndx < len; ndx ++)
1.67 cvs 1835: if (return_value[ndx] == TEXT('\\'))
1836: return_value[ndx] = TEXT('/') ;
1.44 cvs 1837: # endif /* _WINDOWS */
1.29 cvs 1838: return (return_value);
1.24 cvs 1839: }
1.35 cvs 1840:
1841:
Webmaster