Annotation of Amaya/amaya/AHTURLTools.c, revision 1.82
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 */
1.81 cvs 105: case 0x27: /* & */
106: case 0x20: /* space */
1.75 cvs 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")) ||
1.80 cvs 353: !ustrcmp (nsuffix, TEXT("jsp")) ||
1.67 cvs 354: !ustrcmp (nsuffix, TEXT("xht")) ||
355: !ustrcmp (nsuffix, TEXT("xhtm")) ||
356: !ustrcmp (nsuffix, TEXT("xhtml")))
1.39 cvs 357: return (TRUE);
1.67 cvs 358: else if (!ustrcmp (nsuffix, TEXT("gz")))
1.13 cvs 359: {
1.39 cvs 360: /* take into account compressed files */
1.13 cvs 361: ExtractSuffix (temppath, suffix);
362: /* Normalize the suffix */
363: i = 0;
1.39 cvs 364: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 365: {
1.25 cvs 366: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 367: i++;
368: }
369: nsuffix[i] = EOS;
1.67 cvs 370: if (!ustrcmp (nsuffix, TEXT("html")) ||
371: !ustrcmp (nsuffix, TEXT("htm")) ||
372: !ustrcmp (nsuffix, TEXT("shtml")) ||
1.80 cvs 373: !ustrcmp (nsuffix, TEXT("jsp")) ||
1.67 cvs 374: !ustrcmp (nsuffix, TEXT("xht")) ||
375: !ustrcmp (nsuffix, TEXT("xhtm")) ||
376: !ustrcmp (nsuffix, TEXT("xhtml")))
1.63 cvs 377:
1.39 cvs 378: return (TRUE);
379: else
1.13 cvs 380: return (FALSE);
381: }
382: else
1.39 cvs 383: return (FALSE);
1.3 cvs 384: }
385:
1.4 cvs 386: /*----------------------------------------------------------------------
1.56 cvs 387: IsXMLName
388: returns TRUE if path points to an XML resource.
389: ----------------------------------------------------------------------*/
390: #ifdef __STDC__
1.67 cvs 391: ThotBool IsXMLName (const STRING path)
1.56 cvs 392: #else /* __STDC__ */
1.67 cvs 393: ThotBool IsXMLName (path)
394: const STRING path;
1.56 cvs 395: #endif /* __STDC__ */
396: {
1.67 cvs 397: CHAR_T temppath[MAX_LENGTH];
398: CHAR_T suffix[MAX_LENGTH];
1.56 cvs 399:
400: if (!path)
401: return (FALSE);
402:
1.67 cvs 403: ustrcpy (temppath, path);
1.56 cvs 404: ExtractSuffix (temppath, suffix);
405:
1.67 cvs 406: if (!ustrcasecmp (suffix, TEXT("xml")) ||
407: !ustrcasecmp (suffix, TEXT("xht")) ||
408: !ustrcmp (suffix, TEXT("xhtm")) ||
409: !ustrcmp (suffix, TEXT("xhtml")))
1.56 cvs 410: return (TRUE);
1.67 cvs 411: else if (!ustrcmp (suffix, TEXT("gz")))
1.56 cvs 412: {
413: /* take into account compressed files */
414: ExtractSuffix (temppath, suffix);
1.67 cvs 415: if (!ustrcasecmp (suffix, TEXT("xml")) ||
416: !ustrcasecmp (suffix, TEXT("xht")) ||
417: !ustrcmp (suffix, TEXT("xhtm")) ||
418: !ustrcmp (suffix, TEXT("xhtml")))
1.60 cvs 419: return (TRUE);
420: else
421: return (FALSE);
422: }
423: else
424: return (FALSE);
425: }
426:
427: /*----------------------------------------------------------------------
428: IsCSSName
429: returns TRUE if path points to an XML resource.
430: ----------------------------------------------------------------------*/
431: #ifdef __STDC__
1.67 cvs 432: ThotBool IsCSSName (const STRING path)
1.60 cvs 433: #else /* __STDC__ */
1.67 cvs 434: ThotBool IsCSSName (path)
435: const STRING path;
1.60 cvs 436: #endif /* __STDC__ */
437: {
1.67 cvs 438: CHAR_T temppath[MAX_LENGTH];
439: CHAR_T suffix[MAX_LENGTH];
1.60 cvs 440:
441: if (!path)
442: return (FALSE);
443:
1.67 cvs 444: ustrcpy (temppath, path);
1.60 cvs 445: ExtractSuffix (temppath, suffix);
446:
1.67 cvs 447: if (!ustrcasecmp (suffix, TEXT("css")))
1.60 cvs 448: return (TRUE);
1.67 cvs 449: else if (!ustrcmp (suffix, TEXT("gz")))
1.60 cvs 450: {
451: /* take into account compressed files */
452: ExtractSuffix (temppath, suffix);
1.67 cvs 453: if (!ustrcasecmp (suffix, TEXT("css")))
1.56 cvs 454: return (TRUE);
455: else
456: return (FALSE);
457: }
458: else
459: return (FALSE);
460: }
461:
462: /*----------------------------------------------------------------------
1.9 cvs 463: IsImageName
464: returns TRUE if path points to an image resource.
1.4 cvs 465: ----------------------------------------------------------------------*/
1.3 cvs 466: #ifdef __STDC__
1.67 cvs 467: ThotBool IsImageName (const STRING path)
1.3 cvs 468: #else /* __STDC__ */
1.67 cvs 469: ThotBool IsImageName (path)
470: const STRING path;
1.3 cvs 471: #endif /* __STDC__ */
472: {
1.67 cvs 473: CHAR_T temppath[MAX_LENGTH];
474: CHAR_T suffix[MAX_LENGTH];
475: CHAR_T nsuffix[MAX_LENGTH];
1.5 cvs 476: int i;
477:
478: if (!path)
1.13 cvs 479: return (FALSE);
1.5 cvs 480:
1.67 cvs 481: ustrcpy (temppath, path);
1.5 cvs 482: ExtractSuffix (temppath, suffix);
483:
484: /* Normalize the suffix */
485: i = 0;
1.39 cvs 486: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 487: {
1.67 cvs 488: nsuffix[i] = utolower (suffix[i]);
1.13 cvs 489: i++;
490: }
1.5 cvs 491: nsuffix[i] = EOS;
1.67 cvs 492: if ((!ustrcmp (nsuffix, TEXT("gif"))) || (!ustrcmp (nsuffix, TEXT("xbm"))) ||
493: (!ustrcmp (nsuffix, TEXT("xpm"))) || (!ustrcmp (nsuffix, TEXT("jpg"))) ||
494: (!ustrcmp (nsuffix, TEXT("png"))) || (!ustrcmp (nsuffix, TEXT("au"))))
1.39 cvs 495: return (TRUE);
496: return (FALSE);
1.3 cvs 497: }
498:
1.4 cvs 499: /*----------------------------------------------------------------------
1.58 cvs 500: IsImageType
501: returns TRUE if type points to an image resource.
502: ----------------------------------------------------------------------*/
503: #ifdef __STDC__
1.67 cvs 504: ThotBool IsImageType (const STRING type)
1.58 cvs 505: #else /* __STDC__ */
1.67 cvs 506: ThotBool IsImageType (type)
507: const STRING type;
1.58 cvs 508: #endif /* __STDC__ */
509: {
1.67 cvs 510: CHAR_T temptype[MAX_LENGTH];
1.58 cvs 511: int i;
512:
513: if (!type)
514: return (FALSE);
515:
1.67 cvs 516: ustrcpy (temptype, type);
1.58 cvs 517: /* Normalize the type */
518: i = 0;
519: while (temptype[i] != EOS)
520: {
521: temptype[i] = tolower (temptype[i]);
522: i++;
523: }
1.67 cvs 524: if ((!ustrcmp (temptype, TEXT("gif"))) || (!ustrcmp (temptype, TEXT("x-xbitmap"))) ||
525: (!ustrcmp (temptype, TEXT("x-xpixmap"))) || (!ustrcmp (temptype, TEXT("jpeg"))) ||
526: (!ustrcmp (temptype, TEXT("png"))))
1.58 cvs 527: return (TRUE);
528: return (FALSE);
529: }
530:
531: /*----------------------------------------------------------------------
1.9 cvs 532: IsTextName
1.4 cvs 533: ----------------------------------------------------------------------*/
1.3 cvs 534: #ifdef __STDC__
1.67 cvs 535: ThotBool IsTextName (const STRING path)
1.3 cvs 536: #else /* __STDC__ */
1.67 cvs 537: ThotBool IsTextName (path)
538: const STRING path;
1.3 cvs 539:
540: #endif /* __STDC__ */
541: {
1.67 cvs 542: CHAR_T temppath[MAX_LENGTH];
543: CHAR_T suffix[MAX_LENGTH];
544: CHAR_T nsuffix[MAX_LENGTH];
1.5 cvs 545: int i;
546:
547: if (!path)
1.13 cvs 548: return (FALSE);
1.5 cvs 549:
1.67 cvs 550: ustrcpy (temppath, path);
1.5 cvs 551: ExtractSuffix (temppath, suffix);
552:
553: /* Normalize the suffix */
554: i = 0;
1.39 cvs 555: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 556: {
1.25 cvs 557: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 558: i++;
559: }
560: nsuffix[i] = EOS;
561:
1.67 cvs 562: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd"))))
1.13 cvs 563: return (TRUE);
1.67 cvs 564: else if (!ustrcmp (nsuffix, TEXT("gz")))
1.13 cvs 565: {
1.39 cvs 566: /* take into account compressed files */
1.13 cvs 567: ExtractSuffix (temppath, suffix);
568: /* Normalize the suffix */
569: i = 0;
1.39 cvs 570: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 571: {
1.25 cvs 572: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 573: i++;
574: }
575: nsuffix[i] = EOS;
1.67 cvs 576: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd"))))
1.13 cvs 577: return (TRUE);
578: else
579: return (FALSE);
580: }
581: else
582: return (FALSE);
1.3 cvs 583: }
584:
1.4 cvs 585: /*----------------------------------------------------------------------
1.9 cvs 586: IsHTTPPath
587: returns TRUE if path is in fact an http URL.
1.4 cvs 588: ----------------------------------------------------------------------*/
1.3 cvs 589: #ifdef __STDC__
1.67 cvs 590: ThotBool IsHTTPPath (const STRING path)
1.3 cvs 591: #else /* __STDC__ */
1.67 cvs 592: ThotBool IsHTTPPath (path)
593: const STRING path;
1.3 cvs 594: #endif /* __STDC__ */
595: {
1.5 cvs 596: if (!path)
597: return FALSE;
1.3 cvs 598:
1.67 cvs 599: if ((!ustrncmp (path, TEXT("http:"), 5) != 0)
600: || !ustrncmp (path, TEXT("internal:"), 9))
1.58 cvs 601: return TRUE;
602: return FALSE;
1.3 cvs 603: }
604:
1.4 cvs 605: /*----------------------------------------------------------------------
1.9 cvs 606: IsWithParameters
607: returns TRUE if url has a concatenated query string.
1.4 cvs 608: ----------------------------------------------------------------------*/
1.3 cvs 609: #ifdef __STDC__
1.66 cvs 610: ThotBool IsWithParameters (const char *url)
1.3 cvs 611: #else /* __STDC__ */
1.66 cvs 612: ThotBool IsWithParameters (url)
1.34 cvs 613: const char *url;
1.3 cvs 614: #endif /* __STDC__ */
615: {
1.5 cvs 616: int i;
1.3 cvs 617:
1.9 cvs 618: if ((!url) || (url[0] == EOS))
1.5 cvs 619: return FALSE;
1.3 cvs 620:
1.9 cvs 621: i = strlen (url) - 1;
622: while (i > 0 && url[i--] != '?')
1.5 cvs 623: if (i < 0)
624: return FALSE;
1.3 cvs 625:
1.5 cvs 626: /* There is a parameter */
627: return TRUE;
1.3 cvs 628: }
629:
1.4 cvs 630: /*----------------------------------------------------------------------
1.9 cvs 631: IsW3Path
632: returns TRUE if path is in fact a URL.
1.4 cvs 633: ----------------------------------------------------------------------*/
1.3 cvs 634: #ifdef __STDC__
1.82 ! cvs 635: ThotBool IsW3Path (const CharUnit* path)
1.3 cvs 636: #else /* __STDC__ */
1.67 cvs 637: ThotBool IsW3Path (path)
1.82 ! cvs 638: const CharUnit* path;
1.3 cvs 639: #endif /* __STDC__ */
640: {
1.82 ! cvs 641: if (StringNCompare (path, CUSTEXT("http:"), 5) &&
! 642: StringNCompare (path, CUSTEXT("ftp:"), 4) &&
! 643: StringNCompare (path, CUSTEXT("telnet:"), 7) &&
! 644: StringNCompare (path, CUSTEXT("wais:"), 5) &&
! 645: StringNCompare (path, CUSTEXT("news:"), 5) &&
! 646: StringNCompare (path, CUSTEXT("gopher:"), 7) &&
! 647: StringNCompare (path, CUSTEXT("mailto:"), 7) &&
! 648: StringNCompare (path, CUSTEXT("archie:"), 7))
1.72 cvs 649: return FALSE;
650: return TRUE;
1.3 cvs 651: }
652:
1.4 cvs 653: /*----------------------------------------------------------------------
1.9 cvs 654: IsValidProtocol
655: returns true if the url protocol is supported by Amaya.
1.4 cvs 656: ----------------------------------------------------------------------*/
1.3 cvs 657: #ifdef __STDC__
1.67 cvs 658: ThotBool IsValidProtocol (const STRING url)
1.3 cvs 659: #else /* __STDC__ */
1.67 cvs 660: ThotBool IsValidProtocol (url)
661: const STRING url;
1.3 cvs 662: #endif /* __STDC__ */
663: {
1.67 cvs 664: if (!ustrncmp (url, TEXT("http:"), 5)
1.69 cvs 665: || !ustrncmp (url, TEXT("internal:"), 9)
1.70 cvs 666: || !ustrncmp (url, TEXT("ftp:"), 4))
1.22 cvs 667: /* experimental */
1.58 cvs 668: /*** || !strncmp (url, "ftp:", 4) ***/
1.24 cvs 669: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 670: return (TRUE);
1.5 cvs 671: else
1.8 cvs 672: return (FALSE);
1.3 cvs 673: }
674:
1.31 cvs 675:
676: /*----------------------------------------------------------------------
677: GetBaseURL
678: normalizes orgName according to a base associated with doc, and
679: following the standard URL format rules.
680: The function returns the base used to solve relative URL and SRC:
681: - the base of the document,
682: - or the document path (without document name).
683: ----------------------------------------------------------------------*/
684: #ifdef __STDC__
1.82 ! cvs 685: CharUnit* GetBaseURL (Document doc)
1.31 cvs 686: #else /* __STDC__ */
1.82 ! cvs 687: CharUnit* GetBaseURL (doc)
1.31 cvs 688: Document doc;
689: #endif /* __STDC__ */
690: {
691: Element el;
692: ElementType elType;
693: AttributeType attrType;
694: Attribute attr;
1.67 cvs 695: STRING ptr, basename;
1.31 cvs 696: int length;
697:
1.57 cvs 698: /* @@@ irene */
699: if (!DocumentURLs[doc])
700: return NULL;
1.67 cvs 701: basename = TtaAllocString (MAX_LENGTH);
702: ustrncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1.39 cvs 703: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 704: length = MAX_LENGTH -1;
705: /* get the root element */
706: el = TtaGetMainRoot (doc);
707: /* search the BASE element */
708: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.65 cvs 709: elType.ElTypeNum = HTML_EL_HEAD;
710: el = TtaSearchTypedElement (elType, SearchForward, el);
711: if (el)
712: {
713: elType.ElTypeNum = HTML_EL_BASE;
714: el = TtaSearchTypedElement (elType, SearchInTree, el);
715: }
1.31 cvs 716: if (el)
717: {
718: /* The document has a BASE element -> Get the HREF attribute */
719: attrType.AttrSSchema = elType.ElSSchema;
720: attrType.AttrTypeNum = HTML_ATTR_HREF_;
721: attr = TtaGetAttribute (el, attrType);
722: if (attr)
723: {
724: /* Use the base path of the document */
725: TtaGiveTextAttributeValue (attr, basename, &length);
726: /* base and orgName have to be separated by a DIR_SEP */
727: length--;
1.43 cvs 728: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 729: /* verify if the base has the form "protocol://server:port" */
730: {
1.67 cvs 731: ptr = AmayaParseUrl (basename, _EMPTYSTR_, AMAYA_PARSE_ACCESS |
1.33 cvs 732: AMAYA_PARSE_HOST |
733: AMAYA_PARSE_PUNCTUATION);
1.67 cvs 734: if (ptr && !ustrcmp (ptr, basename))
1.31 cvs 735: {
1.43 cvs 736: /* it has this form, we complete it by adding a URL_STR */
1.67 cvs 737: if (ustrchr (basename, DIR_SEP))
738: ustrcat (basename, DIR_STR);
1.43 cvs 739: else
1.67 cvs 740: ustrcat (basename, URL_STR);
1.31 cvs 741: length++;
742: }
743: if (ptr)
744: TtaFreeMemory (ptr);
745: }
746: }
1.33 cvs 747: }
748:
1.31 cvs 749: /* Remove anything after the last DIR_SEP char. If no such char is found,
750: * then search for the first ":" char, hoping that what's before that is a
751: * protocol. If found, end the string there. If neither char is found,
752: * then discard the whole base element.
753: */
1.67 cvs 754: length = ustrlen (basename) - 1;
1.31 cvs 755: /* search for the last DIR_SEP char */
1.43 cvs 756: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 757: length--;
758: if (length >= 0)
759: /* found the last DIR_SEP char, end the string there */
760: basename[length + 1] = EOS;
761: else
762: /* search for the first PATH_STR char */
763: {
1.67 cvs 764: for (length = 0; basename[length] != TEXT(':') &&
1.31 cvs 765: basename[length] != EOS; length ++);
1.67 cvs 766: if (basename[length] == TEXT(':'))
1.31 cvs 767: /* found, so end the string there */
768: basename[length + 1] = EOS;
769: else
770: /* not found, discard the base */
771: basename[0] = EOS;
772: }
773: return (basename);
774: }
775:
776:
1.4 cvs 777: /*----------------------------------------------------------------------
1.40 cvs 778: GetLocalPath
779: Allocate and return the local document path associated to the url
780: ----------------------------------------------------------------------*/
781: #ifdef __STDC__
1.67 cvs 782: STRING GetLocalPath (Document doc, STRING url)
1.40 cvs 783: #else /* __STDC__ */
1.67 cvs 784: STRING GetLocalPath (doc, url)
1.40 cvs 785: Document doc;
1.67 cvs 786: STRING url;
1.40 cvs 787: #endif /* __STDC__ */
788: {
1.67 cvs 789: STRING ptr, n;
790: STRING documentname;
791: CHAR_T url_sep;
1.40 cvs 792: int len;
1.67 cvs 793: ThotBool noFile;
1.40 cvs 794:
795: if (url != NULL)
796: {
797: /* check whether the file name exists */
1.67 cvs 798: len = ustrlen (url) - 1;
1.71 cvs 799: if (IsW3Path (url))
1.67 cvs 800: url_sep = TEXT('/');
1.41 cvs 801: else
802: url_sep = DIR_SEP;
803: noFile = (url[len] == url_sep);
1.40 cvs 804: if (noFile)
805: url[len] = EOS;
1.78 cvs 806: ptr = TtaAllocString (MAX_LENGTH);
807: documentname = TtaAllocString (MAX_LENGTH);
808: TtaExtractName (url, ptr, documentname);
1.67 cvs 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: /*----------------------------------------------------------------------
1.79 cvs 839: ExtractTarget extract the target name from document nane.
840: ----------------------------------------------------------------------*/
841: #ifdef __STDC__
1.82 ! cvs 842: void ExtractTarget (CharUnit* aName, CharUnit* target)
1.79 cvs 843: #else
844: void ExtractTarget (aName, target)
1.82 ! cvs 845: CharUnit* aName;
! 846: CharUnit* target;
1.79 cvs 847: #endif
848: {
1.82 ! cvs 849: int lg, i;
! 850: CharUnit* ptr;
! 851: CharUnit* oldptr;
1.79 cvs 852:
853: if (!target || !aName)
854: /* bad target */
855: return;
856:
1.82 ! cvs 857: target[0] = CUS_EOS;
! 858: lg = StringLength (aName);
1.79 cvs 859: if (lg)
860: {
861: /* the name is not empty */
862: oldptr = ptr = &aName[0];
863: do
864: {
1.82 ! cvs 865: ptr = StrRChr (oldptr, CUSTEXT('#'));
1.79 cvs 866: if (ptr)
867: oldptr = &ptr[1];
868: }
869: while (ptr);
870:
871: i = (int) (oldptr) - (int) (aName); /* name length */
872: if (i > 1)
873: {
1.82 ! cvs 874: aName[i - 1] = CUS_EOS;
1.79 cvs 875: if (i != lg)
1.82 ! cvs 876: StringCopy (target, oldptr);
1.79 cvs 877: }
878: }
879: }
880:
881: /*----------------------------------------------------------------------
1.73 cvs 882: ConvertFileURL
883: If the URL starts with file: prefix, it removes the protocol so that we
884: can use it as a local filename
885: ----------------------------------------------------------------------*/
886: #ifdef __STDC__
1.82 ! cvs 887: void ConvertFileURL (CharUnit* url)
1.73 cvs 888: #else
889: void ConvertFileURL (url)
1.82 ! cvs 890: CharUnit* url
1.73 cvs 891: #endif /* __STDC__ */
892: {
1.82 ! cvs 893: if (!StringNCaseCompare (url, CUSTEXT("file:"), 5))
! 894: StringCopy (url, url + 5);
1.73 cvs 895: }
1.40 cvs 896:
897: /*----------------------------------------------------------------------
1.9 cvs 898: NormalizeURL
899: normalizes orgName according to a base associated with doc, and
900: following the standard URL format rules.
1.53 cvs 901: if doc is 0 and otherPath not NULL, normalizes orgName according to this
902: other path.
1.9 cvs 903: The function returns the new complete and normalized URL
1.12 cvs 904: or file name path (newName) and the name of the document (docName).
1.9 cvs 905: N.B. If the function can't find out what's the docName, it assigns
906: the name "noname.html".
1.4 cvs 907: ----------------------------------------------------------------------*/
1.3 cvs 908: #ifdef __STDC__
1.82 ! cvs 909: void NormalizeURL (CharUnit* orgName, Document doc, CharUnit* newName, CharUnit* docName, CharUnit* otherPath)
1.3 cvs 910: #else /* __STDC__ */
1.53 cvs 911: void NormalizeURL (orgName, doc, newName, docName, otherPath)
1.82 ! cvs 912: CharUnit* orgName;
1.3 cvs 913: Document doc;
1.82 ! cvs 914: CharUnit* newName;
! 915: CharUnit* docName;
! 916: CharUnit* otherPath;
1.3 cvs 917: #endif /* __STDC__ */
918: {
1.82 ! cvs 919: CharUnit* basename;
! 920: CharUnit tempOrgName[MAX_LENGTH];
! 921: CharUnit* ptr;
! 922: CharUnit used_sep;
! 923: int length;
! 924: ThotBool check;
1.5 cvs 925:
1.44 cvs 926: # ifdef _WINDOWS
927: int ndx;
928: # endif /* _WINDOWS */
929:
1.5 cvs 930: if (!newName || !docName)
931: return;
1.18 cvs 932:
1.32 cvs 933: if (doc != 0)
1.53 cvs 934: basename = GetBaseURL (doc);
935: else if (otherPath != NULL)
1.82 ! cvs 936: basename = StringDuplicate (otherPath);
1.32 cvs 937: else
1.53 cvs 938: basename = NULL;
1.32 cvs 939:
1.18 cvs 940: /*
1.31 cvs 941: * Clean orgName
942: * Make sure we have a complete orgName, without any leading or trailing
943: * white spaces, or trailinbg new lines
944: */
1.5 cvs 945: ptr = orgName;
1.18 cvs 946: /* skip leading white space and new line characters */
1.82 ! cvs 947: while ((*ptr == CUS_SPACE || *ptr == CUS_EOL) && *ptr++ != CUS_EOS);
! 948: StringNCopy (tempOrgName, ptr, MAX_LENGTH -1);
! 949: tempOrgName[MAX_LENGTH -1] = CUS_EOS;
1.18 cvs 950: /*
1.31 cvs 951: * Make orgName a complete URL
952: * If the URL does not include a protocol, then try to calculate
953: * one using the doc's base element (if it exists),
954: */
1.82 ! cvs 955: if (tempOrgName[0] == CUS_EOS)
1.53 cvs 956: {
1.82 ! cvs 957: newName[0] = CUS_EOS;
1.53 cvs 958: TtaFreeMemory (basename);
959: return;
960: }
1.49 cvs 961:
962: /* clean trailing white space */
1.82 ! cvs 963: length = StringLength (tempOrgName) - 1;
! 964: while (tempOrgName[length] == CUS_SPACE && tempOrgName[length] == CUS_EOL)
1.53 cvs 965: {
1.82 ! cvs 966: tempOrgName[length] = CUS_EOS;
1.53 cvs 967: length--;
968: }
1.50 cvs 969:
1.55 cvs 970: /* remove extra dot (which dot???) */
971: /* ugly, but faster than a strcmp */
1.82 ! cvs 972: if (tempOrgName[length] == CUSTEXT('.')
! 973: && (length == 0 || tempOrgName[length-1] != CUSTEXT('.')))
! 974: tempOrgName[length] = CUS_EOS;
1.50 cvs 975:
1.53 cvs 976: if (IsW3Path (tempOrgName))
977: {
978: /* the name is complete, go to the Sixth Step */
1.82 ! cvs 979: StringCopy (newName, tempOrgName);
1.53 cvs 980: SimplifyUrl (&newName);
981: /* verify if the URL has the form "protocol://server:port" */
1.67 cvs 982: ptr = AmayaParseUrl (newName, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.82 ! cvs 983: if (ptr && !StringCompare (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
! 984: StringConcat (newName, CUS_URL_STR);
1.49 cvs 985:
1.53 cvs 986: if (ptr)
1.50 cvs 987: TtaFreeMemory (ptr);
1.53 cvs 988: }
989: else if ( basename == NULL)
990: /* the name is complete, go to the Sixth Step */
1.82 ! cvs 991: StringCopy (newName, tempOrgName);
1.53 cvs 992: else
993: {
1.31 cvs 994: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 995: # ifdef _WINDOWS
1.53 cvs 996: if (!IsW3Path (basename))
997: {
1.67 cvs 998: length = ustrlen (tempOrgName);
1.53 cvs 999: for (ndx = 0; ndx < length; ndx++)
1.67 cvs 1000: if (tempOrgName [ndx] == TEXT('/'))
1001: tempOrgName [ndx] = TEXT('\\');
1.53 cvs 1002: }
1.44 cvs 1003: # endif /* _WINDOWS */
1.25 cvs 1004: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1005: if (ptr)
1006: {
1007: SimplifyUrl (&ptr);
1.82 ! cvs 1008: StringCopy (newName, ptr);
1.53 cvs 1009: TtaFreeMemory (ptr);
1010: }
1011: else
1.82 ! cvs 1012: newName[0] = CUS_EOS;
1.53 cvs 1013: }
1.36 cvs 1014:
1015: TtaFreeMemory (basename);
1.18 cvs 1016: /*
1.31 cvs 1017: * Prepare the docname that will refer to this ressource in the
1018: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1019: * noname.html as a default ressource name
1.18 cvs 1020: */
1.82 ! cvs 1021: if (newName[0] != CUS_EOS)
1.53 cvs 1022: {
1.82 ! cvs 1023: length = StringLength (newName) - 1;
! 1024: if (newName[length] == CUS_URL_SEP || newName[length] == CUS_DIR_SEP)
1.53 cvs 1025: {
1026: used_sep = newName[length];
1027: check = TRUE;
1028: while (check)
1029: {
1.50 cvs 1030: length--;
1031: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1032: length--;
1.82 ! cvs 1033: if (!StringNCompare (&newName[length+1], CUSTEXT(".."), 2))
1.53 cvs 1034: {
1.82 ! cvs 1035: newName[length+1] = CUS_EOS;
1.53 cvs 1036: /* remove also previous directory */
1037: length--;
1038: while (length >= 0 && newName[length] != used_sep)
1039: length--;
1.82 ! cvs 1040: if (StringNCompare (&newName[length+1], CUSTEXT("//"), 2))
1.53 cvs 1041: /* don't remove server name */
1.82 ! cvs 1042: newName[length+1] = CUS_EOS;
1.53 cvs 1043: }
1.82 ! cvs 1044: else if (!StringNCompare (&newName[length+1], CUSTEXT("."), 1))
! 1045: newName[length+1] = CUS_EOS;
1.50 cvs 1046: else
1.53 cvs 1047: check = FALSE;
1048: }
1.82 ! cvs 1049: StringCopy (docName, CUSTEXT("noname.html"));
1.53 cvs 1050: /* docname was not comprised inside the URL, so let's */
1051: /* assign the default ressource name */
1.82 ! cvs 1052: StringCopy (docName, CUSTEXT("noname.html"));
1.53 cvs 1053: }
1054: else
1055: { /* docname is comprised inside the URL */
1.82 ! cvs 1056: while (length >= 0 && newName[length] != CUS_URL_SEP && newName[length] != CUS_DIR_SEP)
1.53 cvs 1057: length--;
1058: if (length < 0)
1.82 ! cvs 1059: StringCopy (docName, newName);
1.53 cvs 1060: else
1.82 ! cvs 1061: StringCopy (docName, &newName[length+1]);
1.53 cvs 1062: }
1063: }
1064: else
1.82 ! cvs 1065: docName[0] = CUS_EOS;
1.18 cvs 1066: }
1.3 cvs 1067:
1.4 cvs 1068: /*----------------------------------------------------------------------
1.9 cvs 1069: IsSameHost
1.4 cvs 1070: ----------------------------------------------------------------------*/
1.3 cvs 1071: #ifdef __STDC__
1.67 cvs 1072: ThotBool IsSameHost (const STRING url1, const STRING url2)
1.3 cvs 1073: #else /* __STDC__ */
1.67 cvs 1074: ThotBool IsSameHost (url1, url2)
1075: const STRING url1;
1076: const STRING url2;
1.3 cvs 1077: #endif /* __STDC__ */
1078: {
1.67 cvs 1079: STRING basename_ptr1, basename_ptr2;
1080: ThotBool result;
1.3 cvs 1081:
1.67 cvs 1082: basename_ptr1 = AmayaParseUrl (url1, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1083: basename_ptr2 = AmayaParseUrl (url2, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1084:
1.67 cvs 1085: if (ustrcmp (basename_ptr1, basename_ptr2))
1.8 cvs 1086: result = FALSE;
1.5 cvs 1087: else
1.8 cvs 1088: result = TRUE;
1.3 cvs 1089:
1.25 cvs 1090: TtaFreeMemory (basename_ptr1);
1091: TtaFreeMemory (basename_ptr2);
1.5 cvs 1092: return (result);
1.3 cvs 1093: }
1094:
1095:
1.4 cvs 1096: /*----------------------------------------------------------------------
1.22 cvs 1097: HasKnownFileSuffix
1098: returns TRUE if path points to a file ending with a suffix.
1099: ----------------------------------------------------------------------*/
1100: #ifdef __STDC__
1.67 cvs 1101: ThotBool HasKnownFileSuffix (const STRING path)
1.22 cvs 1102: #else /* __STDC__ */
1.67 cvs 1103: ThotBool HasKnownFileSuffix (path)
1104: const STRING path;
1.22 cvs 1105: #endif /* __STDC__ */
1106: {
1.67 cvs 1107: STRING root;
1108: CHAR_T temppath[MAX_LENGTH];
1109: CHAR_T suffix[MAX_LENGTH];
1.22 cvs 1110:
1.67 cvs 1111: if (!path || path[0] == EOS || path[ustrlen(path)] == DIR_SEP)
1.22 cvs 1112: return (FALSE);
1113:
1.67 cvs 1114: root = AmayaParseUrl(path, _EMPTYSTR_, AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1115:
1116: if (root)
1117: {
1.67 cvs 1118: ustrcpy (temppath, root);
1.25 cvs 1119: TtaFreeMemory (root);
1.22 cvs 1120: /* Get the suffix */
1121: ExtractSuffix (temppath, suffix);
1122:
1123: if( suffix[0] == EOS)
1124: /* no suffix */
1125: return (FALSE);
1126:
1127: /* Normalize the suffix */
1128: ConvertToLowerCase (suffix);
1129:
1.67 cvs 1130: if (!ustrcmp (suffix, TEXT("gz")))
1.22 cvs 1131: /* skip the compressed suffix */
1132: {
1133: ExtractSuffix (temppath, suffix);
1134: if(suffix[0] == EOS)
1135: /* no suffix */
1136: return (FALSE);
1137: /* Normalize the suffix */
1138: ConvertToLowerCase (suffix);
1139: }
1140:
1.67 cvs 1141: if (ustrcmp (suffix, TEXT("gif")) &&
1142: ustrcmp (suffix, TEXT("xbm")) &&
1143: ustrcmp (suffix, TEXT("xpm")) &&
1144: ustrcmp (suffix, TEXT("jpg")) &&
1145: ustrcmp (suffix, TEXT("pdf")) &&
1146: ustrcmp (suffix, TEXT("png")) &&
1147: ustrcmp (suffix, TEXT("tgz")) &&
1148: ustrcmp (suffix, TEXT("xpg")) &&
1149: ustrcmp (suffix, TEXT("xpd")) &&
1150: ustrcmp (suffix, TEXT("ps")) &&
1151: ustrcmp (suffix, TEXT("au")) &&
1152: ustrcmp (suffix, TEXT("html")) &&
1153: ustrcmp (suffix, TEXT("htm")) &&
1154: ustrcmp (suffix, TEXT("shtml")) &&
1155: ustrcmp (suffix, TEXT("xht")) &&
1156: ustrcmp (suffix, TEXT("xhtm")) &&
1157: ustrcmp (suffix, TEXT("xhtml")) &&
1158: ustrcmp (suffix, TEXT("txt")) &&
1159: ustrcmp (suffix, TEXT("css")) &&
1160: ustrcmp (suffix, TEXT("eps")))
1.22 cvs 1161: return (FALSE);
1162: else
1163: return (TRUE);
1164: }
1165: else
1166: return (FALSE);
1167: }
1168:
1169:
1170: /*----------------------------------------------------------------------
1.24 cvs 1171: ChopURL
1172: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1173: If inputURL is bigger than that size, outputURL receives
1174: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1175: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1176: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1177: copied into outputURL.
1178: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1179: chars.
1180: ----------------------------------------------------------------------*/
1181: #ifdef __STDC__
1.34 cvs 1182: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1183: #else
1184: void ChopURL (outputURL, inputURL)
1.34 cvs 1185: char *outputURL;
1186: const char *inputURL;
1.24 cvs 1187: #endif
1.22 cvs 1188:
1.24 cvs 1189: {
1190: int len;
1.9 cvs 1191:
1.24 cvs 1192: len = strlen (inputURL);
1193: if (len <= MAX_PRINT_URL_LENGTH)
1.29 cvs 1194: strcpy (outputURL, inputURL);
1.24 cvs 1195: else
1196: /* make a truncated urlName on the status window */
1197: {
1198: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1199: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1200: strcat (outputURL, "...");
1201: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1202: }
1.25 cvs 1203: }
1204:
1205:
1206: /*----------------------------------------------------------------------
1207: scan
1.47 cvs 1208: Scan a filename for its constituents
1.25 cvs 1209: -----------------------------------
1210:
1211: On entry,
1212: name points to a document name which may be incomplete.
1213: On exit,
1214: absolute or relative may be nonzero (but not both).
1215: host, fragment and access may be nonzero if they were specified.
1216: Any which are nonzero point to zero terminated strings.
1217: ----------------------------------------------------------------------*/
1218: #ifdef __STDC__
1.67 cvs 1219: static void scan (STRING name, HTURI * parts)
1.25 cvs 1220: #else /* __STDC__ */
1221: static void scan (name, parts)
1.67 cvs 1222: STRING name;
1223: HTURI *parts;
1.25 cvs 1224:
1225: #endif /* __STDC__ */
1226: {
1.67 cvs 1227: STRING p;
1228: STRING after_access = name;
1.32 cvs 1229:
1.43 cvs 1230: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1231: /* Look for fragment identifier */
1.67 cvs 1232: if ((p = ustrchr(name, TEXT('#'))) != NULL)
1.28 cvs 1233: {
1.67 cvs 1234: *p++ = TEXT('\0');
1.28 cvs 1235: parts->fragment = p;
1.25 cvs 1236: }
1237:
1.28 cvs 1238: for (p=name; *p; p++)
1239: {
1.67 cvs 1240: if (*p == URL_SEP || *p == DIR_SEP || *p == TEXT('#') || *p == TEXT('?'))
1.28 cvs 1241: break;
1.67 cvs 1242: if (*p == TEXT(':'))
1.28 cvs 1243: {
1244: *p = 0;
1245: parts->access = after_access; /* Scheme has been specified */
1246:
1247: /* The combination of gcc, the "-O" flag and the HP platform is
1248: unhealthy. The following three lines is a quick & dirty fix, but is
1249: not recommended. Rather, turn off "-O". */
1250:
1251: /* after_access = p;*/
1252: /* while (*after_access == 0)*/
1253: /* after_access++;*/
1254: after_access = p+1;
1.67 cvs 1255: if (!ustrcasecmp(TEXT("URL"), parts->access))
1.28 cvs 1256: /* Ignore IETF's URL: pre-prefix */
1257: parts->access = NULL;
1258: else
1.25 cvs 1259: break;
1260: }
1261: }
1262:
1263: p = after_access;
1.43 cvs 1264: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1265: {
1.43 cvs 1266: if (p[1] == URL_SEP)
1.28 cvs 1267: {
1.25 cvs 1268: parts->host = p+2; /* host has been specified */
1.28 cvs 1269: *p = 0; /* Terminate access */
1270: /* look for end of host name if any */
1.67 cvs 1271: p = ustrchr (parts->host, URL_SEP);
1.28 cvs 1272: if (p)
1273: {
1.43 cvs 1274: *p = EOS; /* Terminate host */
1.25 cvs 1275: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1276: }
1277: }
1278: else
1279: /* Root found but no host */
1280: parts->absolute = p+1;
1281: }
1282: else
1283: {
1.25 cvs 1284: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1285: }
1.25 cvs 1286: }
1287:
1288:
1289: /*----------------------------------------------------------------------
1.28 cvs 1290: AmayaParseUrl: parse a Name relative to another name
1291:
1292: This returns those parts of a name which are given (and requested)
1293: substituting bits from the related name where necessary.
1.25 cvs 1294:
1.28 cvs 1295: On entry,
1.25 cvs 1296: aName A filename given
1297: relatedName A name relative to which aName is to be parsed. Give
1298: it an empty string if aName is absolute.
1299: wanted A mask for the bits which are wanted.
1300:
1.28 cvs 1301: On exit,
1.25 cvs 1302: returns A pointer to a malloc'd string which MUST BE FREED
1303: ----------------------------------------------------------------------*/
1304: #ifdef __STDC__
1.67 cvs 1305: STRING AmayaParseUrl (const STRING aName, STRING relatedName, int wanted)
1.25 cvs 1306: #else /* __STDC__ */
1.67 cvs 1307: STRING AmayaParseUrl (aName, relatedName, wanted)
1308: const STRING aName;
1309: STRING relatedName;
1.28 cvs 1310: int wanted;
1.25 cvs 1311:
1312: #endif /* __STDC__ */
1313: {
1.67 cvs 1314: STRING return_value;
1315: CHAR_T result[MAX_LENGTH];
1316: CHAR_T name[MAX_LENGTH];
1317: CHAR_T rel[MAX_LENGTH];
1318: STRING p, access;
1.29 cvs 1319: HTURI given, related;
1320: int len;
1.67 cvs 1321: CHAR_T used_sep;
1322: STRING used_str;
1.32 cvs 1323:
1.67 cvs 1324: if (ustrchr (aName, DIR_SEP) || ustrchr (relatedName, DIR_SEP))
1.33 cvs 1325: {
1.42 cvs 1326: used_str = DIR_STR;
1327: used_sep = DIR_SEP;
1.33 cvs 1328: }
1.32 cvs 1329: else
1.33 cvs 1330: {
1.42 cvs 1331: used_str = URL_STR;
1332: used_sep = URL_SEP;
1.33 cvs 1333: }
1.32 cvs 1334:
1.29 cvs 1335: /* Make working copies of input strings to cut up: */
1336: return_value = NULL;
1337: result[0] = 0; /* Clear string */
1.67 cvs 1338: ustrcpy (name, aName);
1.29 cvs 1339: if (relatedName != NULL)
1.67 cvs 1340: ustrcpy (rel, relatedName);
1.29 cvs 1341: else
1342: relatedName[0] = EOS;
1343:
1344: scan (name, &given);
1345: scan (rel, &related);
1346: access = given.access ? given.access : related.access;
1347: if (wanted & AMAYA_PARSE_ACCESS)
1348: if (access)
1349: {
1.67 cvs 1350: ustrcat (result, access);
1.29 cvs 1351: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1352: ustrcat (result, TEXT(":"));
1.29 cvs 1353: }
1354:
1355: if (given.access && related.access)
1356: /* If different, inherit nothing. */
1.67 cvs 1357: if (ustrcmp (given.access, related.access) != 0)
1.29 cvs 1358: {
1359: related.host = 0;
1360: related.absolute = 0;
1361: related.relative = 0;
1362: related.fragment = 0;
1363: }
1364:
1365: if (wanted & AMAYA_PARSE_HOST)
1366: if(given.host || related.host)
1367: {
1368: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1369: ustrcat (result, TEXT("//"));
1370: ustrcat (result, given.host ? given.host : related.host);
1.29 cvs 1371: }
1372:
1373: if (given.host && related.host)
1374: /* If different hosts, inherit no path. */
1.67 cvs 1375: if (ustrcmp (given.host, related.host) != 0)
1.29 cvs 1376: {
1377: related.absolute = 0;
1378: related.relative = 0;
1379: related.fragment = 0;
1380: }
1381:
1382: if (wanted & AMAYA_PARSE_PATH)
1383: {
1384: if (given.absolute)
1385: {
1386: /* All is given */
1387: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1388: ustrcat (result, used_str);
1389: ustrcat (result, given.absolute);
1.25 cvs 1390: }
1.29 cvs 1391: else if (related.absolute)
1392: {
1393: /* Adopt path not name */
1.67 cvs 1394: ustrcat (result, used_str);
1395: ustrcat (result, related.absolute);
1.29 cvs 1396: if (given.relative)
1397: {
1398: /* Search part? */
1.67 cvs 1399: p = ustrchr (result, TEXT('?'));
1.29 cvs 1400: if (!p)
1.67 cvs 1401: p=result+ustrlen(result)-1;
1.33 cvs 1402: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1403: /* Remove filename */
1404: p[1]=0;
1405: /* Add given one */
1.67 cvs 1406: ustrcat (result, given.relative);
1.25 cvs 1407: }
1408: }
1.29 cvs 1409: else if (given.relative)
1410: /* what we've got */
1.67 cvs 1411: ustrcat (result, given.relative);
1.29 cvs 1412: else if (related.relative)
1.67 cvs 1413: ustrcat (result, related.relative);
1.29 cvs 1414: else
1415: /* No inheritance */
1.67 cvs 1416: ustrcat (result, used_str);
1.25 cvs 1417: }
1.29 cvs 1418:
1419: if (wanted & AMAYA_PARSE_ANCHOR)
1420: if (given.fragment || related.fragment)
1421: {
1422: if (given.absolute && given.fragment)
1423: {
1424: /*Fixes for relURLs...*/
1425: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1426: ustrcat (result, TEXT("#"));
1427: ustrcat (result, given.fragment);
1.29 cvs 1428: }
1429: else if (!(given.absolute) && !(given.fragment))
1.67 cvs 1430: ustrcat (result, _EMPTYSTR_);
1.29 cvs 1431: else
1432: {
1433: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1434: ustrcat (result, TEXT("#"));
1435: ustrcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1436: }
1437: }
1.67 cvs 1438: len = ustrlen (result);
1439: if ((return_value = TtaAllocString (len + 1)) != NULL)
1440: ustrcpy (return_value, result);
1.29 cvs 1441: return (return_value); /* exactly the right length */
1.25 cvs 1442: }
1443:
1444: /*----------------------------------------------------------------------
1445: HTCanon
1446: Canonicalizes the URL in the following manner starting from the host
1447: pointer:
1448:
1449: 1) The host name is converted to lowercase
1450: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1451:
1452: Return: OK The position of the current path part of the URL
1453: which might be the old one or a new one.
1454:
1455: ----------------------------------------------------------------------*/
1456: #ifdef __STDC__
1.82 ! cvs 1457: static CharUnit* HTCanon (CharUnit** filename, CharUnit* host)
1.25 cvs 1458: #else /* __STDC__ */
1.67 cvs 1459: static STRING HTCanon (filename, host)
1.82 ! cvs 1460: CharUnit** filename;
! 1461: CharUnit* host;
1.25 cvs 1462: #endif /* __STDC__ */
1463: {
1.82 ! cvs 1464: CharUnit* newname = NULL;
! 1465: CharUnit used_sep;
! 1466: CharUnit* path;
! 1467: CharUnit* strptr;
! 1468: CharUnit* port;
! 1469: CharUnit* access = host-3;
1.32 cvs 1470:
1471:
1.82 ! cvs 1472: if (*filename && StrChr (*filename, CUS_URL_SEP))
1.33 cvs 1473: {
1.82 ! cvs 1474: used_sep = CUS_URL_SEP;
1.33 cvs 1475: }
1476: else
1477: {
1.82 ! cvs 1478: used_sep = CUS_DIR_SEP;
1.33 cvs 1479: }
1.32 cvs 1480:
1.82 ! cvs 1481: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1482: access--;
1.82 ! cvs 1483: if ((path = StrChr (host, used_sep)) == NULL) /* Find path */
! 1484: path = host + StringLength (host);
! 1485: if ((strptr = StrChr (host, CUSTEXT('@'))) != NULL && strptr < path) /* UserId */
! 1486: host = strptr;
! 1487: if ((port = StrChr (host, CUSTEXT(':'))) != NULL && port > path) /* Port number */
! 1488: port = NULL;
1.25 cvs 1489:
1490: strptr = host; /* Convert to lower-case */
1.82 ! cvs 1491: while (strptr < path)
1.33 cvs 1492: {
1.82 ! cvs 1493: *strptr = ToLower (*strptr);
! 1494: strptr++;
1.33 cvs 1495: }
1.25 cvs 1496:
1497: /* Does the URL contain a full domain name? This also works for a
1498: numerical host name. The domain name is already made lower-case
1499: and without a trailing dot. */
1500: {
1.82 ! cvs 1501: CharUnit* dot = port ? port : path;
! 1502: if (dot > *filename && *--dot == CUSTEXT('.'))
1.33 cvs 1503: {
1.82 ! cvs 1504: CharUnit* orig = dot;
! 1505: CharUnit* dest = dot + 1;
! 1506: while ((*orig++ = *dest++));
! 1507: if (port) port--;
1.33 cvs 1508: path--;
1.25 cvs 1509: }
1510: }
1511: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1512: if (port)
1513: {
1.82 ! cvs 1514: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1515: {
1516: if (!newname)
1517: {
1.82 ! cvs 1518: CharUnit* orig = port;
! 1519: CharUnit* dest = port + 1;
! 1520: while ((*orig++ = *dest++));
1.33 cvs 1521: }
1522: }
1.82 ! cvs 1523: else if ((!StringNCompare (access, CUSTEXT("http"), 4) &&
! 1524: (*(port + 1) == CUSTEXT('8') &&
! 1525: *(port+2) == CUSTEXT('0') &&
! 1526: (*(port+3) == used_sep || !*(port + 3)))) ||
! 1527: (!StringNCompare (access, CUSTEXT("gopher"), 6) &&
! 1528: (*(port+1) == CUSTEXT('7') &&
! 1529: *(port+2) == CUSTEXT('0') &&
! 1530: (*(port+3) == used_sep || !*(port+3)))) ||
! 1531: (!StringNCompare (access, CUSTEXT("ftp"), 3) &&
! 1532: (*(port+1) == CUSTEXT('2') &&
! 1533: *(port + 2) == CUSTEXT('1') &&
! 1534: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1535: if (!newname)
1536: {
1.82 ! cvs 1537: CharUnit* orig = port;
! 1538: CharUnit* dest = port + 3;
1.33 cvs 1539: while((*orig++ = *dest++));
1540: /* Update path position, Henry Minsky */
1541: path -= 3;
1.25 cvs 1542: }
1.33 cvs 1543: }
1544: else if (newname)
1.82 ! cvs 1545: StringNConcat (newname, port, (int) (path - port));
1.33 cvs 1546: }
1.25 cvs 1547:
1.33 cvs 1548: if (newname)
1549: {
1.82 ! cvs 1550: CharUnit* newpath = newname + StringLength (newname);
! 1551: StringConcat (newname, path);
1.25 cvs 1552: path = newpath;
1.28 cvs 1553: /* Free old copy */
1554: TtaFreeMemory(*filename);
1.25 cvs 1555: *filename = newname;
1.33 cvs 1556: }
1.25 cvs 1557: return path;
1558: }
1559:
1560:
1561: /*----------------------------------------------------------------------
1.29 cvs 1562: SimplifyUrl: simplify a URI
1.32 cvs 1563: A URI is allowed to contain the sequence xxx/../ which may be
1564: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1565: Simplification helps us recognize duplicate URIs.
1.25 cvs 1566:
1.28 cvs 1567: Thus, /etc/junk/../fred becomes /etc/fred
1568: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1569:
1.28 cvs 1570: but we should NOT change
1571: http://fred.xxx.edu/../..
1.25 cvs 1572:
1573: or ../../albert.html
1574:
1.28 cvs 1575: In order to avoid empty URLs the following URLs become:
1.25 cvs 1576:
1577: /fred/.. becomes /fred/..
1578: /fred/././.. becomes /fred/..
1579: /fred/.././junk/.././ becomes /fred/..
1580:
1.28 cvs 1581: If more than one set of `://' is found (several proxies in cascade) then
1582: only the part after the last `://' is simplified.
1.25 cvs 1583:
1.28 cvs 1584: Returns: A string which might be the old one or a new one.
1.25 cvs 1585: ----------------------------------------------------------------------*/
1586: #ifdef __STDC__
1.82 ! cvs 1587: void SimplifyUrl (CharUnit** url)
1.25 cvs 1588: #else /* __STDC__ */
1.29 cvs 1589: void SimplifyUrl (url)
1.82 ! cvs 1590: CharUnit** url;
1.25 cvs 1591: #endif /* __STDC__ */
1592: {
1.82 ! cvs 1593: CharUnit* path;
! 1594: CharUnit* access;
! 1595: CharUnit* newptr;
! 1596: STRING p;
! 1597: STRING orig, dest, end;
1.28 cvs 1598:
1.82 ! cvs 1599: CharUnit used_sep;
1.77 cvs 1600: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1601: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1602:
1603:
1.28 cvs 1604: if (!url || !*url)
1605: return;
1606:
1.82 ! cvs 1607: if (StrChr (*url, CUS_URL_SEP))
1.33 cvs 1608: {
1.82 ! cvs 1609: used_sep = CUS_URL_SEP;
1.33 cvs 1610: }
1.32 cvs 1611: else
1.33 cvs 1612: {
1.82 ! cvs 1613: used_sep = CUS_DIR_SEP;
1.33 cvs 1614: }
1.32 cvs 1615:
1.77 cvs 1616: /* should we simplify double dot? */
1617: path = *url;
1.82 ! cvs 1618: if (*path == CUSTEXT('.') && *(path + 1) == CUSTEXT('.'))
1.77 cvs 1619: ddot_simplify = FALSE;
1620: else
1621: ddot_simplify = TRUE;
1622:
1.28 cvs 1623: /* Find any scheme name */
1.82 ! cvs 1624: if ((path = StringSubstring(*url, CUSTEXT("://"))) != NULL)
1.33 cvs 1625: {
1626: /* Find host name */
1.28 cvs 1627: access = *url;
1.82 ! cvs 1628: while (access < path && (*access = ToLower (*access)))
! 1629: access++;
1.28 cvs 1630: path += 3;
1.82 ! cvs 1631: while ((newptr = StringSubstring (path, CUSTEXT ("://"))) != NULL)
! 1632: /* For proxies */
! 1633: path = newptr+3;
! 1634: /* We have a host name */
1.28 cvs 1635: path = HTCanon(url, path);
1.25 cvs 1636: }
1.67 cvs 1637: else if ((path = ustrstr(*url, TEXT(":/"))) != NULL)
1.28 cvs 1638: path += 2;
1639: else
1640: path = *url;
1.25 cvs 1641:
1.33 cvs 1642: if (*path == used_sep && *(path+1)==used_sep)
1.28 cvs 1643: /* Some URLs start //<foo> */
1644: path += 1;
1.67 cvs 1645: else if (!ustrncmp(path, TEXT("news:"), 5))
1.28 cvs 1646: {
1.67 cvs 1647: newptr = ustrchr(path+5, TEXT('@'));
1.28 cvs 1648: if (!newptr)
1649: newptr = path + 5;
1650: while (*newptr)
1651: {
1652: /* Make group or host lower case */
1653: *newptr = tolower (*newptr);
1654: newptr++;
1.25 cvs 1655: }
1.28 cvs 1656: /* Doesn't need to do any more */
1657: return;
1.25 cvs 1658: }
1.28 cvs 1659:
1660: if ((p = path))
1661: {
1.67 cvs 1662: if (!((end = ustrchr (path, TEXT(';'))) || (end = ustrchr (path, TEXT('?'))) ||
1663: (end = ustrchr (path, TEXT('#')))))
1664: end = path + ustrlen (path);
1.28 cvs 1665:
1666: /* Parse string second time to simplify */
1667: p = path;
1668: while (p < end)
1669: {
1.77 cvs 1670: /* if we're pointing to a char, it's safe to reactivate the ../ convertion */
1671: if (!ddot_simplify && *p != TEXT('.') && *p != used_sep)
1672: ddot_simplify = TRUE;
1673:
1.33 cvs 1674: if (*p==used_sep)
1.28 cvs 1675: {
1.67 cvs 1676: if (p > *url && *(p+1) == TEXT('.') && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1677: {
1678: orig = p + 1;
1.33 cvs 1679: dest = (*(p+2)!=used_sep) ? p+2 : p+3;
1.52 cvs 1680: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1681: end = orig - 1;
1682: }
1.77 cvs 1683: else if (ddot_simplify && *(p+1) == TEXT('.') && *(p+2) == TEXT('.')
1684: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1685: {
1686: newptr = p;
1.52 cvs 1687: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1688: if (*newptr == used_sep)
1689: orig = newptr + 1;
1.28 cvs 1690: else
1.52 cvs 1691: orig = newptr;
1692:
1693: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1694: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1695: end = orig-1;
1696: /* Start again with prev slash */
1697: p = newptr;
1.28 cvs 1698: }
1.33 cvs 1699: else if (*(p+1) == used_sep)
1.28 cvs 1700: {
1.33 cvs 1701: while (*(p+1) == used_sep)
1.28 cvs 1702: {
1703: orig = p;
1704: dest = p + 1;
1705: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1706: end = orig-1;
1707: }
1708: }
1709: else
1.25 cvs 1710: p++;
1.28 cvs 1711: }
1712: else
1713: p++;
1.25 cvs 1714: }
1715: }
1.51 cvs 1716:
1717: /*
1718: ** Check for host/../.. kind of things
1719: */
1.77 cvs 1720: if (*path == used_sep && *(path+1) == TEXT('.') && *(path+2) == TEXT('.')
1721: && (!*(path+3) || *(path+3) == used_sep))
1.51 cvs 1722: *(path+1) = EOS;
1723:
1.28 cvs 1724: return;
1725: }
1726:
1727:
1728: /*----------------------------------------------------------------------
1729: NormalizeFile normalizes local names.
1730: Return TRUE if target and src differ.
1731: ----------------------------------------------------------------------*/
1732: #ifdef __STDC__
1.82 ! cvs 1733: ThotBool NormalizeFile (CharUnit* src, CharUnit* target)
1.28 cvs 1734: #else
1.67 cvs 1735: ThotBool NormalizeFile (src, target)
1.82 ! cvs 1736: CharUnit* src;
! 1737: CharUnit* target;
1.28 cvs 1738:
1739: #endif
1740: {
1.82 ! cvs 1741: CharUnit* s;
! 1742: ThotBool change;
1.28 cvs 1743:
1.54 cvs 1744: change = FALSE;
1.82 ! cvs 1745: if (StringNCompare (src, CUSTEXT("file:"), 5) == 0)
1.28 cvs 1746: {
1747: /* remove the prefix file: */
1.82 ! cvs 1748: if (src[5] == CUS_EOS)
! 1749: StringCopy (target, CUS_DIR_STR);
! 1750: else if (src[0] == CUSTEXT('~'))
1.28 cvs 1751: {
1752: /* replace ~ */
1.74 cvs 1753: s = TtaGetEnvString ("HOME");
1.82 ! cvs 1754: StringCopy (target, s);
! 1755: StringConcat (target, &src[5]);
1.28 cvs 1756: }
1757: else
1.82 ! cvs 1758: StringCopy (target, &src[5]);
1.54 cvs 1759: change = TRUE;
1.28 cvs 1760: }
1.53 cvs 1761: # ifndef _WINDOWS
1.82 ! cvs 1762: else if (src[0] == CUSTEXT('~'))
1.53 cvs 1763: {
1764: /* replace ~ */
1.82 ! cvs 1765: s = TtaGetEnvString ("HOME");
! 1766: StringCopy (target, s);
! 1767: if (src[1] != CUS_DIR_SEP)
! 1768: strcat (target, CUS_DIR_STR);
! 1769: StringConcat (target, &src[1]);
1.54 cvs 1770: change = TRUE;
1.53 cvs 1771: }
1772: # endif /* _WINDOWS */
1.28 cvs 1773: else
1.82 ! cvs 1774: StringCopy (target, src);
1.28 cvs 1775:
1776: /* remove /../ and /./ */
1.29 cvs 1777: SimplifyUrl (&target);
1.54 cvs 1778: if (!change)
1.82 ! cvs 1779: change = StringCompare (src, target);
1.28 cvs 1780: return (change);
1.25 cvs 1781: }
1782:
1.28 cvs 1783:
1.25 cvs 1784: /*----------------------------------------------------------------------
1.31 cvs 1785: MakeRelativeURL: make relative name
1.25 cvs 1786:
1.28 cvs 1787: This function creates and returns a string which gives an expression of
1788: one address as related to another. Where there is no relation, an absolute
1789: address is retured.
1.25 cvs 1790:
1.28 cvs 1791: On entry,
1.25 cvs 1792: Both names must be absolute, fully qualified names of nodes
1793: (no fragment bits)
1794:
1.28 cvs 1795: On exit,
1.25 cvs 1796: The return result points to a newly allocated name which, if
1797: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1798: The caller is responsible for freeing the resulting name later.
1799: ----------------------------------------------------------------------*/
1800: #ifdef __STDC__
1.67 cvs 1801: STRING MakeRelativeURL (STRING aName, STRING relatedName)
1.25 cvs 1802: #else /* __STDC__ */
1.67 cvs 1803: STRING MakeRelativeURL (aName, relatedName)
1804: STRING aName;
1805: STRING relatedName;
1.25 cvs 1806: #endif /* __STDC__ */
1807: {
1.67 cvs 1808: STRING return_value;
1809: CHAR_T result[MAX_LENGTH];
1810: STRING p;
1811: STRING q;
1812: STRING after_access;
1813: STRING last_slash = NULL;
1.29 cvs 1814: int slashes, levels, len;
1815:
1.44 cvs 1816: # ifdef _WINDOWS
1817: int ndx;
1818: # endif /* _WINDOWS */
1819:
1.29 cvs 1820: if (aName == NULL || relatedName == NULL)
1821: return (NULL);
1822:
1823: slashes = 0;
1824: after_access = NULL;
1825: p = aName;
1826: q = relatedName;
1827: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1828: {
1829: /* Find extent of match */
1.67 cvs 1830: if (*p == TEXT(':'))
1.29 cvs 1831: after_access = p + 1;
1.28 cvs 1832: if (*p == DIR_SEP)
1.27 cvs 1833: {
1.29 cvs 1834: /* memorize the last slash position and count them */
1.27 cvs 1835: last_slash = p;
1836: slashes++;
1.25 cvs 1837: }
1838: }
1839:
1.31 cvs 1840: /* q, p point to the first non-matching character or zero */
1841: if (*q == EOS)
1842: {
1843: /* New name is a subset of the related name */
1844: /* exactly the right length */
1.67 cvs 1845: len = ustrlen (p);
1846: if ((return_value = TtaAllocString (len + 1)) != NULL)
1847: ustrcpy (return_value, p);
1.31 cvs 1848: }
1849: else if ((slashes < 2 && after_access == NULL)
1850: || (slashes < 3 && after_access != NULL))
1851: {
1852: /* Two names whitout common path */
1853: /* exactly the right length */
1.67 cvs 1854: len = ustrlen (aName);
1855: if ((return_value = TtaAllocString (len + 1)) != NULL)
1856: ustrcpy (return_value, aName);
1.31 cvs 1857: }
1858: else
1859: {
1860: /* Some path in common */
1.67 cvs 1861: if (slashes == 3 && ustrncmp (aName, TEXT("http:"), 5) == 0)
1.31 cvs 1862: /* just the same server */
1.67 cvs 1863: ustrcpy (result, last_slash);
1.31 cvs 1864: else
1865: {
1866: levels= 0;
1.67 cvs 1867: for (; *q && *q != TEXT('#') && *q != TEXT(';') && *q != TEXT('?'); q++)
1.31 cvs 1868: if (*q == DIR_SEP)
1869: levels++;
1870:
1.52 cvs 1871: result[0] = EOS;
1.31 cvs 1872: for (;levels; levels--)
1.67 cvs 1873: ustrcat (result, TEXT("../"));
1874: ustrcat (result, last_slash+1);
1.31 cvs 1875: }
1.52 cvs 1876:
1877: if (!*result)
1.67 cvs 1878: ustrcat (result, TEXT("./"));
1.52 cvs 1879:
1.31 cvs 1880: /* exactly the right length */
1.67 cvs 1881: len = ustrlen (result);
1882: if ((return_value = TtaAllocString (len + 1)) != NULL)
1883: ustrcpy (return_value, result);
1.52 cvs 1884:
1.25 cvs 1885: }
1.44 cvs 1886: # ifdef _WINDOWS
1.67 cvs 1887: len = ustrlen (return_value);
1.44 cvs 1888: for (ndx = 0; ndx < len; ndx ++)
1.67 cvs 1889: if (return_value[ndx] == TEXT('\\'))
1890: return_value[ndx] = TEXT('/') ;
1.44 cvs 1891: # endif /* _WINDOWS */
1.29 cvs 1892: return (return_value);
1.24 cvs 1893: }
1.35 cvs 1894:
1895:
Webmaster
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to AHTURLTools.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Public] / Amaya / amaya