Annotation of Amaya/amaya/AHTURLTools.c, revision 1.93
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
1.87 cvs 14: * R. Guetari: Windows & Unicode.
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;
1.93 ! cvs 47:
1.28 cvs 48: if (!string)
49: return;
50:
1.87 cvs 51: for (i = 0; string[i] != WC_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 */
1.87 cvs 126: buffer[buffer_len] = WC_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: }
1.87 cvs 145: buffer[buffer_len] = WC_EOS;
1.75 cvs 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__
1.91 cvs 281: void ExtractSuffix (CHAR_T* aName, CHAR_T* aSuffix)
1.61 cvs 282: #else
283: void ExtractSuffix (aName, aSuffix)
1.91 cvs 284: CHAR_T* aName;
285: CHAR_T* aSuffix;
1.61 cvs 286:
287: #endif
288: {
289: int lg, i;
290: STRING ptr, oldptr;
291:
292: if (!aSuffix || !aName)
293: /* bad suffix */
294: return;
295:
1.87 cvs 296: aSuffix[0] = WC_EOS;
1.61 cvs 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: {
1.87 cvs 313: aName[i - 1] = WC_EOS;
1.61 cvs 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__
1.84 cvs 325: ThotBool IsHTMLName (const CHAR_T* path)
1.3 cvs 326: #else /* __STDC__ */
1.67 cvs 327: ThotBool IsHTMLName (path)
1.84 cvs 328: const CHAR_T* path;
1.3 cvs 329: #endif /* __STDC__ */
330: {
1.84 cvs 331: CHAR_T temppath[MAX_LENGTH];
332: CHAR_T suffix[MAX_LENGTH];
333: CHAR_T nsuffix[MAX_LENGTH];
334: int i;
1.5 cvs 335:
336: if (!path)
1.37 cvs 337: return (FALSE);
1.5 cvs 338:
1.84 cvs 339: ustrcpy (temppath, path);
1.5 cvs 340: ExtractSuffix (temppath, suffix);
1.91 cvs 341: /* while (suffix[0] != WC_EOS) { */
342: i = 0;
343: while (suffix[i] != WC_EOS) {
344: /* Normalize the suffix */
345: i = 0;
346: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1) {
347: nsuffix[i] = utolower (suffix[i]);
348: i++;
349: }
350: nsuffix[i] = WC_EOS;
351: if (!ustrcmp (nsuffix, TEXT("html")) ||
352: !ustrcmp (nsuffix, TEXT("htm")) ||
353: !ustrcmp (nsuffix, TEXT("shtml")) ||
354: !ustrcmp (nsuffix, TEXT("jsp")) ||
355: !ustrcmp (nsuffix, TEXT("xht")) ||
356: !ustrcmp (nsuffix, TEXT("xhtm")) ||
357: !ustrcmp (nsuffix, TEXT("xhtml")))
358: return (TRUE);
359: else if (!ustrcmp (nsuffix, TEXT("gz"))) {
360: /* take into account compressed files */
361: ExtractSuffix (temppath, suffix);
362: /* Normalize the suffix */
363: i = 0;
364: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1) {
365: nsuffix[i] = utolower (suffix[i]);
366: i++;
367: }
368: nsuffix[i] = WC_EOS;
369: if (!ustrcmp (nsuffix, TEXT("html")) ||
370: !ustrcmp (nsuffix, TEXT("htm")) ||
371: !ustrcmp (nsuffix, TEXT("shtml")) ||
372: !ustrcmp (nsuffix, TEXT("jsp")) ||
373: !ustrcmp (nsuffix, TEXT("xht")) ||
374: !ustrcmp (nsuffix, TEXT("xhtm")) ||
375: !ustrcmp (nsuffix, TEXT("xhtml")))
376: return (TRUE);
377: else
378: return (FALSE);
379: } else
380: /* check if there is another suffix */
381: ExtractSuffix (temppath, suffix);
382: }
1.88 cvs 383:
384: return (FALSE);
1.3 cvs 385: }
386:
1.4 cvs 387: /*----------------------------------------------------------------------
1.56 cvs 388: IsXMLName
389: returns TRUE if path points to an XML resource.
390: ----------------------------------------------------------------------*/
391: #ifdef __STDC__
1.67 cvs 392: ThotBool IsXMLName (const STRING path)
1.56 cvs 393: #else /* __STDC__ */
1.67 cvs 394: ThotBool IsXMLName (path)
395: const STRING path;
1.56 cvs 396: #endif /* __STDC__ */
397: {
1.67 cvs 398: CHAR_T temppath[MAX_LENGTH];
399: CHAR_T suffix[MAX_LENGTH];
1.56 cvs 400:
401: if (!path)
402: return (FALSE);
403:
1.67 cvs 404: ustrcpy (temppath, path);
1.56 cvs 405: ExtractSuffix (temppath, suffix);
406:
1.67 cvs 407: if (!ustrcasecmp (suffix, TEXT("xml")) ||
408: !ustrcasecmp (suffix, TEXT("xht")) ||
409: !ustrcmp (suffix, TEXT("xhtm")) ||
410: !ustrcmp (suffix, TEXT("xhtml")))
1.56 cvs 411: return (TRUE);
1.67 cvs 412: else if (!ustrcmp (suffix, TEXT("gz")))
1.56 cvs 413: {
414: /* take into account compressed files */
415: ExtractSuffix (temppath, suffix);
1.67 cvs 416: if (!ustrcasecmp (suffix, TEXT("xml")) ||
417: !ustrcasecmp (suffix, TEXT("xht")) ||
418: !ustrcmp (suffix, TEXT("xhtm")) ||
419: !ustrcmp (suffix, TEXT("xhtml")))
1.60 cvs 420: return (TRUE);
421: else
422: return (FALSE);
423: }
424: else
425: return (FALSE);
426: }
427:
428: /*----------------------------------------------------------------------
429: IsCSSName
430: returns TRUE if path points to an XML resource.
431: ----------------------------------------------------------------------*/
432: #ifdef __STDC__
1.67 cvs 433: ThotBool IsCSSName (const STRING path)
1.60 cvs 434: #else /* __STDC__ */
1.67 cvs 435: ThotBool IsCSSName (path)
436: const STRING path;
1.60 cvs 437: #endif /* __STDC__ */
438: {
1.67 cvs 439: CHAR_T temppath[MAX_LENGTH];
440: CHAR_T suffix[MAX_LENGTH];
1.60 cvs 441:
442: if (!path)
443: return (FALSE);
444:
1.67 cvs 445: ustrcpy (temppath, path);
1.60 cvs 446: ExtractSuffix (temppath, suffix);
447:
1.67 cvs 448: if (!ustrcasecmp (suffix, TEXT("css")))
1.60 cvs 449: return (TRUE);
1.67 cvs 450: else if (!ustrcmp (suffix, TEXT("gz")))
1.60 cvs 451: {
452: /* take into account compressed files */
453: ExtractSuffix (temppath, suffix);
1.67 cvs 454: if (!ustrcasecmp (suffix, TEXT("css")))
1.56 cvs 455: return (TRUE);
456: else
457: return (FALSE);
458: }
459: else
460: return (FALSE);
461: }
462:
463: /*----------------------------------------------------------------------
1.9 cvs 464: IsImageName
465: returns TRUE if path points to an image resource.
1.4 cvs 466: ----------------------------------------------------------------------*/
1.3 cvs 467: #ifdef __STDC__
1.67 cvs 468: ThotBool IsImageName (const STRING path)
1.3 cvs 469: #else /* __STDC__ */
1.67 cvs 470: ThotBool IsImageName (path)
471: const STRING path;
1.3 cvs 472: #endif /* __STDC__ */
473: {
1.67 cvs 474: CHAR_T temppath[MAX_LENGTH];
475: CHAR_T suffix[MAX_LENGTH];
476: CHAR_T nsuffix[MAX_LENGTH];
1.5 cvs 477: int i;
478:
479: if (!path)
1.13 cvs 480: return (FALSE);
1.5 cvs 481:
1.67 cvs 482: ustrcpy (temppath, path);
1.5 cvs 483: ExtractSuffix (temppath, suffix);
484:
485: /* Normalize the suffix */
486: i = 0;
1.87 cvs 487: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1)
1.13 cvs 488: {
1.67 cvs 489: nsuffix[i] = utolower (suffix[i]);
1.13 cvs 490: i++;
491: }
1.87 cvs 492: nsuffix[i] = WC_EOS;
1.67 cvs 493: if ((!ustrcmp (nsuffix, TEXT("gif"))) || (!ustrcmp (nsuffix, TEXT("xbm"))) ||
494: (!ustrcmp (nsuffix, TEXT("xpm"))) || (!ustrcmp (nsuffix, TEXT("jpg"))) ||
495: (!ustrcmp (nsuffix, TEXT("png"))) || (!ustrcmp (nsuffix, TEXT("au"))))
1.39 cvs 496: return (TRUE);
497: return (FALSE);
1.3 cvs 498: }
499:
1.4 cvs 500: /*----------------------------------------------------------------------
1.58 cvs 501: IsImageType
502: returns TRUE if type points to an image resource.
503: ----------------------------------------------------------------------*/
504: #ifdef __STDC__
1.67 cvs 505: ThotBool IsImageType (const STRING type)
1.58 cvs 506: #else /* __STDC__ */
1.67 cvs 507: ThotBool IsImageType (type)
508: const STRING type;
1.58 cvs 509: #endif /* __STDC__ */
510: {
1.67 cvs 511: CHAR_T temptype[MAX_LENGTH];
1.58 cvs 512: int i;
513:
514: if (!type)
515: return (FALSE);
516:
1.67 cvs 517: ustrcpy (temptype, type);
1.58 cvs 518: /* Normalize the type */
519: i = 0;
1.87 cvs 520: while (temptype[i] != WC_EOS)
1.58 cvs 521: {
522: temptype[i] = tolower (temptype[i]);
523: i++;
524: }
1.67 cvs 525: if ((!ustrcmp (temptype, TEXT("gif"))) || (!ustrcmp (temptype, TEXT("x-xbitmap"))) ||
526: (!ustrcmp (temptype, TEXT("x-xpixmap"))) || (!ustrcmp (temptype, TEXT("jpeg"))) ||
527: (!ustrcmp (temptype, TEXT("png"))))
1.58 cvs 528: return (TRUE);
529: return (FALSE);
530: }
531:
532: /*----------------------------------------------------------------------
1.9 cvs 533: IsTextName
1.4 cvs 534: ----------------------------------------------------------------------*/
1.3 cvs 535: #ifdef __STDC__
1.67 cvs 536: ThotBool IsTextName (const STRING path)
1.3 cvs 537: #else /* __STDC__ */
1.67 cvs 538: ThotBool IsTextName (path)
539: const STRING path;
1.3 cvs 540:
541: #endif /* __STDC__ */
542: {
1.67 cvs 543: CHAR_T temppath[MAX_LENGTH];
544: CHAR_T suffix[MAX_LENGTH];
545: CHAR_T nsuffix[MAX_LENGTH];
1.5 cvs 546: int i;
547:
548: if (!path)
1.13 cvs 549: return (FALSE);
1.5 cvs 550:
1.67 cvs 551: ustrcpy (temppath, path);
1.5 cvs 552: ExtractSuffix (temppath, suffix);
553:
554: /* Normalize the suffix */
555: i = 0;
1.87 cvs 556: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1)
1.5 cvs 557: {
1.25 cvs 558: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 559: i++;
560: }
1.87 cvs 561: nsuffix[i] = WC_EOS;
1.5 cvs 562:
1.67 cvs 563: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd"))))
1.13 cvs 564: return (TRUE);
1.67 cvs 565: else if (!ustrcmp (nsuffix, TEXT("gz")))
1.13 cvs 566: {
1.39 cvs 567: /* take into account compressed files */
1.13 cvs 568: ExtractSuffix (temppath, suffix);
569: /* Normalize the suffix */
570: i = 0;
1.87 cvs 571: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1)
1.13 cvs 572: {
1.25 cvs 573: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 574: i++;
575: }
1.87 cvs 576: nsuffix[i] = WC_EOS;
1.67 cvs 577: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd"))))
1.13 cvs 578: return (TRUE);
579: else
580: return (FALSE);
581: }
582: else
583: return (FALSE);
1.3 cvs 584: }
585:
1.4 cvs 586: /*----------------------------------------------------------------------
1.9 cvs 587: IsHTTPPath
588: returns TRUE if path is in fact an http URL.
1.4 cvs 589: ----------------------------------------------------------------------*/
1.3 cvs 590: #ifdef __STDC__
1.67 cvs 591: ThotBool IsHTTPPath (const STRING path)
1.3 cvs 592: #else /* __STDC__ */
1.67 cvs 593: ThotBool IsHTTPPath (path)
594: const STRING path;
1.3 cvs 595: #endif /* __STDC__ */
596: {
1.5 cvs 597: if (!path)
598: return FALSE;
1.3 cvs 599:
1.67 cvs 600: if ((!ustrncmp (path, TEXT("http:"), 5) != 0)
601: || !ustrncmp (path, TEXT("internal:"), 9))
1.58 cvs 602: return TRUE;
603: return FALSE;
1.3 cvs 604: }
605:
1.4 cvs 606: /*----------------------------------------------------------------------
1.9 cvs 607: IsWithParameters
608: returns TRUE if url has a concatenated query string.
1.4 cvs 609: ----------------------------------------------------------------------*/
1.3 cvs 610: #ifdef __STDC__
1.66 cvs 611: ThotBool IsWithParameters (const char *url)
1.3 cvs 612: #else /* __STDC__ */
1.66 cvs 613: ThotBool IsWithParameters (url)
1.34 cvs 614: const char *url;
1.3 cvs 615: #endif /* __STDC__ */
616: {
1.5 cvs 617: int i;
1.3 cvs 618:
1.9 cvs 619: if ((!url) || (url[0] == EOS))
1.5 cvs 620: return FALSE;
1.3 cvs 621:
1.9 cvs 622: i = strlen (url) - 1;
623: while (i > 0 && url[i--] != '?')
1.5 cvs 624: if (i < 0)
625: return FALSE;
1.3 cvs 626:
1.5 cvs 627: /* There is a parameter */
628: return TRUE;
1.3 cvs 629: }
630:
1.4 cvs 631: /*----------------------------------------------------------------------
1.9 cvs 632: IsW3Path
633: returns TRUE if path is in fact a URL.
1.4 cvs 634: ----------------------------------------------------------------------*/
1.3 cvs 635: #ifdef __STDC__
1.84 cvs 636: ThotBool IsW3Path (const CHAR_T* path)
1.3 cvs 637: #else /* __STDC__ */
1.67 cvs 638: ThotBool IsW3Path (path)
1.84 cvs 639: const CHAR_T* path;
1.3 cvs 640: #endif /* __STDC__ */
641: {
1.84 cvs 642: if (ustrncmp (path, TEXT("http:"), 5) &&
643: ustrncmp (path, TEXT("ftp:"), 4) &&
644: ustrncmp (path, TEXT("telnet:"), 7) &&
645: ustrncmp (path, TEXT("wais:"), 5) &&
646: ustrncmp (path, TEXT("news:"), 5) &&
647: ustrncmp (path, TEXT("gopher:"), 7) &&
648: ustrncmp (path, TEXT("mailto:"), 7) &&
649: ustrncmp (path, TEXT("archie:"), 7))
1.72 cvs 650: return FALSE;
651: return TRUE;
1.3 cvs 652: }
653:
1.4 cvs 654: /*----------------------------------------------------------------------
1.90 cvs 655: IsFilePath
656: returns TRUE if path is in fact a URL.
657: ----------------------------------------------------------------------*/
658: #ifdef __STDC__
659: static ThotBool IsFilePath (const CHAR_T* path)
660: #else /* __STDC__ */
661: static ThotBool IsFilePath (path)
662: const CHAR_T* path;
663: #endif /* __STDC__ */
664: {
665: if (ustrncmp (path, TEXT("file:"), 5))
666: return FALSE;
667: return TRUE;
668: }
669:
670: /*----------------------------------------------------------------------
1.9 cvs 671: IsValidProtocol
672: returns true if the url protocol is supported by Amaya.
1.4 cvs 673: ----------------------------------------------------------------------*/
1.3 cvs 674: #ifdef __STDC__
1.67 cvs 675: ThotBool IsValidProtocol (const STRING url)
1.3 cvs 676: #else /* __STDC__ */
1.67 cvs 677: ThotBool IsValidProtocol (url)
678: const STRING url;
1.3 cvs 679: #endif /* __STDC__ */
680: {
1.67 cvs 681: if (!ustrncmp (url, TEXT("http:"), 5)
1.69 cvs 682: || !ustrncmp (url, TEXT("internal:"), 9)
1.70 cvs 683: || !ustrncmp (url, TEXT("ftp:"), 4))
1.22 cvs 684: /* experimental */
1.58 cvs 685: /*** || !strncmp (url, "ftp:", 4) ***/
1.24 cvs 686: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 687: return (TRUE);
1.5 cvs 688: else
1.8 cvs 689: return (FALSE);
1.3 cvs 690: }
691:
1.31 cvs 692:
693: /*----------------------------------------------------------------------
694: GetBaseURL
695: normalizes orgName according to a base associated with doc, and
696: following the standard URL format rules.
697: The function returns the base used to solve relative URL and SRC:
698: - the base of the document,
699: - or the document path (without document name).
700: ----------------------------------------------------------------------*/
701: #ifdef __STDC__
1.84 cvs 702: CHAR_T* GetBaseURL (Document doc)
1.31 cvs 703: #else /* __STDC__ */
1.84 cvs 704: CHAR_T* GetBaseURL (doc)
1.31 cvs 705: Document doc;
706: #endif /* __STDC__ */
707: {
708: Element el;
709: ElementType elType;
710: AttributeType attrType;
711: Attribute attr;
1.84 cvs 712: CHAR_T *ptr, *basename;
1.31 cvs 713: int length;
714:
1.57 cvs 715: /* @@@ irene */
716: if (!DocumentURLs[doc])
717: return NULL;
1.67 cvs 718: basename = TtaAllocString (MAX_LENGTH);
719: ustrncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1.84 cvs 720: basename[MAX_LENGTH-1] = WC_EOS;
1.31 cvs 721: length = MAX_LENGTH -1;
722: /* get the root element */
723: el = TtaGetMainRoot (doc);
724: /* search the BASE element */
725: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.65 cvs 726: elType.ElTypeNum = HTML_EL_HEAD;
727: el = TtaSearchTypedElement (elType, SearchForward, el);
728: if (el)
729: {
730: elType.ElTypeNum = HTML_EL_BASE;
731: el = TtaSearchTypedElement (elType, SearchInTree, el);
732: }
1.31 cvs 733: if (el)
734: {
735: /* The document has a BASE element -> Get the HREF attribute */
736: attrType.AttrSSchema = elType.ElSSchema;
737: attrType.AttrTypeNum = HTML_ATTR_HREF_;
738: attr = TtaGetAttribute (el, attrType);
739: if (attr)
740: {
741: /* Use the base path of the document */
742: TtaGiveTextAttributeValue (attr, basename, &length);
743: /* base and orgName have to be separated by a DIR_SEP */
744: length--;
1.84 cvs 745: if (basename[0] != WC_EOS && basename[length] != WC_URL_SEP && basename[length] != WC_DIR_SEP)
1.31 cvs 746: /* verify if the base has the form "protocol://server:port" */
747: {
1.84 cvs 748: ptr = AmayaParseUrl (basename, TEXT(""), AMAYA_PARSE_ACCESS |
1.33 cvs 749: AMAYA_PARSE_HOST |
750: AMAYA_PARSE_PUNCTUATION);
1.67 cvs 751: if (ptr && !ustrcmp (ptr, basename))
1.31 cvs 752: {
1.43 cvs 753: /* it has this form, we complete it by adding a URL_STR */
1.84 cvs 754: if (ustrchr (basename, WC_DIR_SEP))
755: ustrcat (basename, WC_DIR_STR);
1.43 cvs 756: else
1.84 cvs 757: ustrcat (basename, WC_URL_STR);
1.31 cvs 758: length++;
759: }
760: if (ptr)
761: TtaFreeMemory (ptr);
762: }
763: }
1.33 cvs 764: }
765:
1.31 cvs 766: /* Remove anything after the last DIR_SEP char. If no such char is found,
767: * then search for the first ":" char, hoping that what's before that is a
768: * protocol. If found, end the string there. If neither char is found,
769: * then discard the whole base element.
770: */
1.67 cvs 771: length = ustrlen (basename) - 1;
1.31 cvs 772: /* search for the last DIR_SEP char */
1.84 cvs 773: while (length >= 0 && basename[length] != WC_URL_SEP && basename[length] != WC_DIR_SEP)
1.31 cvs 774: length--;
775: if (length >= 0)
776: /* found the last DIR_SEP char, end the string there */
1.84 cvs 777: basename[length + 1] = WC_EOS;
1.31 cvs 778: else
779: /* search for the first PATH_STR char */
780: {
1.67 cvs 781: for (length = 0; basename[length] != TEXT(':') &&
1.84 cvs 782: basename[length] != WC_EOS; length ++);
1.67 cvs 783: if (basename[length] == TEXT(':'))
1.31 cvs 784: /* found, so end the string there */
1.84 cvs 785: basename[length + 1] = WC_EOS;
1.31 cvs 786: else
787: /* not found, discard the base */
1.84 cvs 788: basename[0] = WC_EOS;
1.31 cvs 789: }
790: return (basename);
791: }
792:
793:
1.4 cvs 794: /*----------------------------------------------------------------------
1.40 cvs 795: GetLocalPath
796: Allocate and return the local document path associated to the url
797: ----------------------------------------------------------------------*/
798: #ifdef __STDC__
1.84 cvs 799: CHAR_T* GetLocalPath (Document doc, CHAR_T* url)
1.40 cvs 800: #else /* __STDC__ */
1.84 cvs 801: CHAR_T* GetLocalPath (doc, url)
1.40 cvs 802: Document doc;
1.84 cvs 803: CHAR_T* url;
1.40 cvs 804: #endif /* __STDC__ */
805: {
1.84 cvs 806: CHAR_T* ptr;
807: CHAR_T* n;
808: CHAR_T* documentname;
809: CHAR_T url_sep;
1.83 cvs 810: int len;
1.67 cvs 811: ThotBool noFile;
1.40 cvs 812:
813: if (url != NULL)
814: {
815: /* check whether the file name exists */
1.84 cvs 816: len = ustrlen (url) - 1;
1.71 cvs 817: if (IsW3Path (url))
1.84 cvs 818: url_sep = TEXT('/');
1.41 cvs 819: else
1.84 cvs 820: url_sep = WC_DIR_SEP;
1.41 cvs 821: noFile = (url[len] == url_sep);
1.40 cvs 822: if (noFile)
1.84 cvs 823: url[len] = WC_EOS;
824: ptr = TtaAllocString (MAX_LENGTH);
825: documentname = TtaAllocString (MAX_LENGTH);
1.78 cvs 826: TtaExtractName (url, ptr, documentname);
1.84 cvs 827: usprintf (ptr, TEXT("%s%s%d%s"), TempFileDirectory, WC_DIR_STR, doc, WC_DIR_STR);
1.40 cvs 828: if (!TtaCheckDirectory (ptr))
829: /* directory did not exist */
1.72 cvs 830: TtaMakeDirectory (ptr);
1.47 cvs 831:
832: /* don't include the query string within document name */
1.84 cvs 833: n = ustrrchr (documentname, TEXT('?'));
1.47 cvs 834: if (n != NULL)
1.84 cvs 835: *n = WC_EOS;
1.46 cvs 836: /* don't include ':' within document name */
1.84 cvs 837: n = ustrchr (documentname, TEXT(':'));
1.46 cvs 838: if (n != NULL)
1.84 cvs 839: *n = WC_EOS;
1.69 cvs 840: /* if after all this operations document name
841: is empty, let's use noname.html instead */
1.84 cvs 842: if (documentname[0] == WC_EOS)
843: ustrcat (ptr, TEXT("noname.html"));
1.69 cvs 844: else
1.84 cvs 845: ustrcat (ptr, documentname);
1.40 cvs 846: TtaFreeMemory (documentname);
847: /* restore the url */
848: if (noFile)
1.41 cvs 849: url[len] = url_sep;
1.40 cvs 850: return (ptr);
851: }
852: else
853: return (NULL);
854: }
855:
1.73 cvs 856: /*----------------------------------------------------------------------
1.79 cvs 857: ExtractTarget extract the target name from document nane.
858: ----------------------------------------------------------------------*/
859: #ifdef __STDC__
1.84 cvs 860: void ExtractTarget (CHAR_T* aName, CHAR_T* target)
1.79 cvs 861: #else
862: void ExtractTarget (aName, target)
1.84 cvs 863: CHAR_T* aName;
864: CHAR_T* target;
1.79 cvs 865: #endif
866: {
1.82 cvs 867: int lg, i;
1.84 cvs 868: CHAR_T* ptr;
869: CHAR_T* oldptr;
1.79 cvs 870:
871: if (!target || !aName)
872: /* bad target */
873: return;
874:
1.84 cvs 875: target[0] = WC_EOS;
876: lg = ustrlen (aName);
1.79 cvs 877: if (lg)
878: {
879: /* the name is not empty */
880: oldptr = ptr = &aName[0];
881: do
882: {
1.84 cvs 883: ptr = ustrrchr (oldptr, TEXT('#'));
1.79 cvs 884: if (ptr)
885: oldptr = &ptr[1];
886: }
887: while (ptr);
888:
889: i = (int) (oldptr) - (int) (aName); /* name length */
890: if (i > 1)
891: {
1.84 cvs 892: aName[i - 1] = WC_EOS;
1.79 cvs 893: if (i != lg)
1.84 cvs 894: ustrcpy (target, oldptr);
1.79 cvs 895: }
896: }
897: }
898:
899: /*----------------------------------------------------------------------
1.90 cvs 900: RemoveNewLines (text)
901: Removes any '\n' chars that are found in text.
902: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 903: ----------------------------------------------------------------------*/
904: #ifdef __STDC__
1.90 cvs 905: ThotBool RemoveNewLines (CHAR_T *text)
1.73 cvs 906: #else
1.90 cvs 907: ThotBool RemoveNewLines (text)
908: CHAR_T *text;
909:
1.73 cvs 910: #endif /* __STDC__ */
911: {
1.90 cvs 912: ThotBool change = FALSE;
913: CHAR_T *src;
914: CHAR_T *dest;
915:
916: src = text;
917: dest = text;
918:
919: while (*src)
920: {
921: switch (*src)
922: {
923: case TEXT('\n'):
924: /* don't copy the newline */
925: change = 1;
926: break;
927: default:
928: *dest = *src;
929: dest++;
930: break;
931: }
932: src++;
933: }
934: /* copy the last EOS char */
935: *dest = *src;
936:
937: return (change);
938: }
939:
940: /*----------------------------------------------------------------------
941: CleanCopyFileURL
942: Copies a file url from a src string to destination string.
943: It changes the URL_SEP into DIR_SEPs if they are not the same
944: character.
945: ----------------------------------------------------------------------*/
946: #ifdef __STDC__
947: static void CleanCopyFileURL (CHAR_T *dest, CHAR_T *src)
948: #else
949: static void CleanCopyFileURL (dest, src)
950: CHAR_T* dest;
951: CHAR_T* src;
1.89 cvs 952:
1.90 cvs 953: #endif /* __STDC__ */
954: {
955: while (*src)
1.89 cvs 956: {
1.90 cvs 957: switch (*src)
1.89 cvs 958: {
959: #ifdef _WINDOWS
1.90 cvs 960: case WC_URL_SEP:
961: /* make the transformation */
962: *dest = WC_DIR_SEP;
963: dest++;
964: break;
1.89 cvs 965: #endif /* _WINDOWS */
1.90 cvs 966: default:
967: *dest = *src;
1.89 cvs 968: dest++;
1.90 cvs 969: break;
1.89 cvs 970: }
1.90 cvs 971: src++;
1.89 cvs 972: }
1.90 cvs 973: /* copy the EOS char */
974: *dest = *src;
1.73 cvs 975: }
1.40 cvs 976:
977: /*----------------------------------------------------------------------
1.9 cvs 978: NormalizeURL
979: normalizes orgName according to a base associated with doc, and
980: following the standard URL format rules.
1.53 cvs 981: if doc is 0 and otherPath not NULL, normalizes orgName according to this
982: other path.
1.9 cvs 983: The function returns the new complete and normalized URL
1.12 cvs 984: or file name path (newName) and the name of the document (docName).
1.9 cvs 985: N.B. If the function can't find out what's the docName, it assigns
986: the name "noname.html".
1.4 cvs 987: ----------------------------------------------------------------------*/
1.3 cvs 988: #ifdef __STDC__
1.84 cvs 989: void NormalizeURL (CHAR_T* orgName, Document doc, CHAR_T* newName, CHAR_T* docName, CHAR_T* otherPath)
1.3 cvs 990: #else /* __STDC__ */
1.53 cvs 991: void NormalizeURL (orgName, doc, newName, docName, otherPath)
1.84 cvs 992: CHAR_T* orgName;
1.3 cvs 993: Document doc;
1.84 cvs 994: CHAR_T* newName;
995: CHAR_T* docName;
996: CHAR_T* otherPath;
1.3 cvs 997: #endif /* __STDC__ */
998: {
1.84 cvs 999: CHAR_T* basename;
1000: CHAR_T tempOrgName[MAX_LENGTH];
1001: CHAR_T* ptr;
1002: CHAR_T used_sep;
1003: int length;
1004: ThotBool check;
1.5 cvs 1005:
1.44 cvs 1006: # ifdef _WINDOWS
1007: int ndx;
1008: # endif /* _WINDOWS */
1009:
1.5 cvs 1010: if (!newName || !docName)
1011: return;
1.18 cvs 1012:
1.32 cvs 1013: if (doc != 0)
1.53 cvs 1014: basename = GetBaseURL (doc);
1015: else if (otherPath != NULL)
1.84 cvs 1016: basename = TtaWCSdup (otherPath);
1.32 cvs 1017: else
1.53 cvs 1018: basename = NULL;
1.32 cvs 1019:
1.18 cvs 1020: /*
1.31 cvs 1021: * Clean orgName
1022: * Make sure we have a complete orgName, without any leading or trailing
1023: * white spaces, or trailinbg new lines
1024: */
1.5 cvs 1025: ptr = orgName;
1.18 cvs 1026: /* skip leading white space and new line characters */
1.84 cvs 1027: while ((*ptr == WC_SPACE || *ptr == WC_EOL) && *ptr++ != WC_EOS);
1028: ustrncpy (tempOrgName, ptr, MAX_LENGTH -1);
1029: tempOrgName[MAX_LENGTH -1] = WC_EOS;
1.18 cvs 1030: /*
1.31 cvs 1031: * Make orgName a complete URL
1032: * If the URL does not include a protocol, then try to calculate
1033: * one using the doc's base element (if it exists),
1034: */
1.84 cvs 1035: if (tempOrgName[0] == WC_EOS)
1.53 cvs 1036: {
1.84 cvs 1037: newName[0] = WC_EOS;
1.53 cvs 1038: TtaFreeMemory (basename);
1039: return;
1040: }
1.49 cvs 1041:
1042: /* clean trailing white space */
1.84 cvs 1043: length = ustrlen (tempOrgName) - 1;
1044: while (tempOrgName[length] == WC_SPACE && tempOrgName[length] == WC_EOL)
1.53 cvs 1045: {
1.84 cvs 1046: tempOrgName[length] = WC_EOS;
1.53 cvs 1047: length--;
1048: }
1.50 cvs 1049:
1.55 cvs 1050: /* remove extra dot (which dot???) */
1051: /* ugly, but faster than a strcmp */
1.84 cvs 1052: if (tempOrgName[length] == TEXT('.')
1053: && (length == 0 || tempOrgName[length-1] != TEXT('.')))
1054: tempOrgName[length] = WC_EOS;
1.50 cvs 1055:
1.90 cvs 1056: if (IsW3Path (tempOrgName) || IsFilePath (tempOrgName))
1.53 cvs 1057: {
1058: /* the name is complete, go to the Sixth Step */
1.84 cvs 1059: ustrcpy (newName, tempOrgName);
1.53 cvs 1060: SimplifyUrl (&newName);
1061: /* verify if the URL has the form "protocol://server:port" */
1.84 cvs 1062: ptr = AmayaParseUrl (newName, TEXT(""), AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1063: if (ptr && !ustrcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
1064: ustrcat (newName, WC_URL_STR);
1.49 cvs 1065:
1.53 cvs 1066: if (ptr)
1.50 cvs 1067: TtaFreeMemory (ptr);
1.53 cvs 1068: }
1069: else if ( basename == NULL)
1070: /* the name is complete, go to the Sixth Step */
1.84 cvs 1071: ustrcpy (newName, tempOrgName);
1.53 cvs 1072: else
1073: {
1.31 cvs 1074: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 1075: # ifdef _WINDOWS
1.53 cvs 1076: if (!IsW3Path (basename))
1077: {
1.67 cvs 1078: length = ustrlen (tempOrgName);
1.53 cvs 1079: for (ndx = 0; ndx < length; ndx++)
1.67 cvs 1080: if (tempOrgName [ndx] == TEXT('/'))
1081: tempOrgName [ndx] = TEXT('\\');
1.53 cvs 1082: }
1.44 cvs 1083: # endif /* _WINDOWS */
1.25 cvs 1084: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1085: if (ptr)
1086: {
1087: SimplifyUrl (&ptr);
1.84 cvs 1088: ustrcpy (newName, ptr);
1.53 cvs 1089: TtaFreeMemory (ptr);
1090: }
1091: else
1.84 cvs 1092: newName[0] = WC_EOS;
1.53 cvs 1093: }
1.36 cvs 1094:
1095: TtaFreeMemory (basename);
1.18 cvs 1096: /*
1.31 cvs 1097: * Prepare the docname that will refer to this ressource in the
1098: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1099: * noname.html as a default ressource name
1.18 cvs 1100: */
1.84 cvs 1101: if (newName[0] != WC_EOS)
1.53 cvs 1102: {
1.84 cvs 1103: length = ustrlen (newName) - 1;
1104: if (newName[length] == WC_URL_SEP || newName[length] == WC_DIR_SEP)
1.53 cvs 1105: {
1106: used_sep = newName[length];
1107: check = TRUE;
1108: while (check)
1109: {
1.50 cvs 1110: length--;
1111: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1112: length--;
1.84 cvs 1113: if (!ustrncmp (&newName[length+1], TEXT(".."), 2))
1.53 cvs 1114: {
1.84 cvs 1115: newName[length+1] = WC_EOS;
1.53 cvs 1116: /* remove also previous directory */
1117: length--;
1118: while (length >= 0 && newName[length] != used_sep)
1119: length--;
1.84 cvs 1120: if (ustrncmp (&newName[length+1], TEXT("//"), 2))
1.53 cvs 1121: /* don't remove server name */
1.84 cvs 1122: newName[length+1] = WC_EOS;
1.53 cvs 1123: }
1.84 cvs 1124: else if (!ustrncmp (&newName[length+1], TEXT("."), 1))
1125: newName[length+1] = WC_EOS;
1.50 cvs 1126: else
1.53 cvs 1127: check = FALSE;
1128: }
1.84 cvs 1129: ustrcpy (docName, TEXT("noname.html"));
1.53 cvs 1130: /* docname was not comprised inside the URL, so let's */
1131: /* assign the default ressource name */
1.84 cvs 1132: ustrcpy (docName, TEXT("noname.html"));
1.53 cvs 1133: }
1134: else
1135: { /* docname is comprised inside the URL */
1.84 cvs 1136: while (length >= 0 && newName[length] != WC_URL_SEP && newName[length] != WC_DIR_SEP)
1.53 cvs 1137: length--;
1138: if (length < 0)
1.84 cvs 1139: ustrcpy (docName, newName);
1.53 cvs 1140: else
1.84 cvs 1141: ustrcpy (docName, &newName[length+1]);
1.53 cvs 1142: }
1143: }
1144: else
1.84 cvs 1145: docName[0] = WC_EOS;
1.18 cvs 1146: }
1.3 cvs 1147:
1.4 cvs 1148: /*----------------------------------------------------------------------
1.9 cvs 1149: IsSameHost
1.4 cvs 1150: ----------------------------------------------------------------------*/
1.3 cvs 1151: #ifdef __STDC__
1.67 cvs 1152: ThotBool IsSameHost (const STRING url1, const STRING url2)
1.3 cvs 1153: #else /* __STDC__ */
1.67 cvs 1154: ThotBool IsSameHost (url1, url2)
1155: const STRING url1;
1156: const STRING url2;
1.3 cvs 1157: #endif /* __STDC__ */
1158: {
1.67 cvs 1159: STRING basename_ptr1, basename_ptr2;
1160: ThotBool result;
1.3 cvs 1161:
1.84 cvs 1162: basename_ptr1 = AmayaParseUrl (url1, TEXT(""), AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.86 cvs 1163: basename_ptr2 = AmayaParseUrl (url2, TEXT(""), AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1164:
1.67 cvs 1165: if (ustrcmp (basename_ptr1, basename_ptr2))
1.8 cvs 1166: result = FALSE;
1.5 cvs 1167: else
1.8 cvs 1168: result = TRUE;
1.3 cvs 1169:
1.25 cvs 1170: TtaFreeMemory (basename_ptr1);
1171: TtaFreeMemory (basename_ptr2);
1.5 cvs 1172: return (result);
1.3 cvs 1173: }
1174:
1175:
1.4 cvs 1176: /*----------------------------------------------------------------------
1.22 cvs 1177: HasKnownFileSuffix
1178: returns TRUE if path points to a file ending with a suffix.
1179: ----------------------------------------------------------------------*/
1180: #ifdef __STDC__
1.67 cvs 1181: ThotBool HasKnownFileSuffix (const STRING path)
1.22 cvs 1182: #else /* __STDC__ */
1.67 cvs 1183: ThotBool HasKnownFileSuffix (path)
1184: const STRING path;
1.22 cvs 1185: #endif /* __STDC__ */
1186: {
1.67 cvs 1187: STRING root;
1188: CHAR_T temppath[MAX_LENGTH];
1189: CHAR_T suffix[MAX_LENGTH];
1.22 cvs 1190:
1.87 cvs 1191: if (!path || path[0] == WC_EOS || path[ustrlen(path)] == WC_DIR_SEP)
1.22 cvs 1192: return (FALSE);
1193:
1.84 cvs 1194: root = AmayaParseUrl(path, TEXT(""), AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1195:
1196: if (root)
1197: {
1.67 cvs 1198: ustrcpy (temppath, root);
1.25 cvs 1199: TtaFreeMemory (root);
1.22 cvs 1200: /* Get the suffix */
1201: ExtractSuffix (temppath, suffix);
1202:
1.87 cvs 1203: if( suffix[0] == WC_EOS)
1.22 cvs 1204: /* no suffix */
1205: return (FALSE);
1206:
1207: /* Normalize the suffix */
1208: ConvertToLowerCase (suffix);
1209:
1.67 cvs 1210: if (!ustrcmp (suffix, TEXT("gz")))
1.22 cvs 1211: /* skip the compressed suffix */
1212: {
1213: ExtractSuffix (temppath, suffix);
1.87 cvs 1214: if(suffix[0] == WC_EOS)
1.22 cvs 1215: /* no suffix */
1216: return (FALSE);
1217: /* Normalize the suffix */
1218: ConvertToLowerCase (suffix);
1219: }
1220:
1.67 cvs 1221: if (ustrcmp (suffix, TEXT("gif")) &&
1222: ustrcmp (suffix, TEXT("xbm")) &&
1223: ustrcmp (suffix, TEXT("xpm")) &&
1224: ustrcmp (suffix, TEXT("jpg")) &&
1225: ustrcmp (suffix, TEXT("pdf")) &&
1226: ustrcmp (suffix, TEXT("png")) &&
1227: ustrcmp (suffix, TEXT("tgz")) &&
1228: ustrcmp (suffix, TEXT("xpg")) &&
1229: ustrcmp (suffix, TEXT("xpd")) &&
1230: ustrcmp (suffix, TEXT("ps")) &&
1231: ustrcmp (suffix, TEXT("au")) &&
1232: ustrcmp (suffix, TEXT("html")) &&
1233: ustrcmp (suffix, TEXT("htm")) &&
1234: ustrcmp (suffix, TEXT("shtml")) &&
1235: ustrcmp (suffix, TEXT("xht")) &&
1236: ustrcmp (suffix, TEXT("xhtm")) &&
1237: ustrcmp (suffix, TEXT("xhtml")) &&
1238: ustrcmp (suffix, TEXT("txt")) &&
1239: ustrcmp (suffix, TEXT("css")) &&
1240: ustrcmp (suffix, TEXT("eps")))
1.22 cvs 1241: return (FALSE);
1242: else
1243: return (TRUE);
1244: }
1245: else
1246: return (FALSE);
1247: }
1248:
1249:
1250: /*----------------------------------------------------------------------
1.24 cvs 1251: ChopURL
1252: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1253: If inputURL is bigger than that size, outputURL receives
1254: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1255: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1256: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1257: copied into outputURL.
1258: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1259: chars.
1260: ----------------------------------------------------------------------*/
1261: #ifdef __STDC__
1.86 cvs 1262: void ChopURL (CHAR_T* outputURL, const CHAR_T* inputURL)
1.24 cvs 1263: #else
1264: void ChopURL (outputURL, inputURL)
1.86 cvs 1265: CHAR_T* outputURL;
1266: const CHAR_T* inputURL;
1.24 cvs 1267: #endif
1.22 cvs 1268:
1.24 cvs 1269: {
1270: int len;
1.9 cvs 1271:
1.86 cvs 1272: len = ustrlen (inputURL);
1.24 cvs 1273: if (len <= MAX_PRINT_URL_LENGTH)
1.86 cvs 1274: ustrcpy (outputURL, inputURL);
1.24 cvs 1275: else
1276: /* make a truncated urlName on the status window */
1277: {
1.86 cvs 1278: ustrncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1279: outputURL [MAX_PRINT_URL_LENGTH / 2] = WC_EOS;
1280: ustrcat (outputURL, TEXT("..."));
1281: ustrcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1282: }
1.25 cvs 1283: }
1284:
1285:
1286: /*----------------------------------------------------------------------
1287: scan
1.47 cvs 1288: Scan a filename for its constituents
1.25 cvs 1289: -----------------------------------
1290:
1291: On entry,
1292: name points to a document name which may be incomplete.
1293: On exit,
1294: absolute or relative may be nonzero (but not both).
1295: host, fragment and access may be nonzero if they were specified.
1296: Any which are nonzero point to zero terminated strings.
1297: ----------------------------------------------------------------------*/
1298: #ifdef __STDC__
1.67 cvs 1299: static void scan (STRING name, HTURI * parts)
1.25 cvs 1300: #else /* __STDC__ */
1301: static void scan (name, parts)
1.67 cvs 1302: STRING name;
1303: HTURI *parts;
1.25 cvs 1304:
1305: #endif /* __STDC__ */
1306: {
1.67 cvs 1307: STRING p;
1308: STRING after_access = name;
1.32 cvs 1309:
1.43 cvs 1310: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1311: /* Look for fragment identifier */
1.67 cvs 1312: if ((p = ustrchr(name, TEXT('#'))) != NULL)
1.28 cvs 1313: {
1.67 cvs 1314: *p++ = TEXT('\0');
1.28 cvs 1315: parts->fragment = p;
1.25 cvs 1316: }
1317:
1.28 cvs 1318: for (p=name; *p; p++)
1319: {
1.67 cvs 1320: if (*p == URL_SEP || *p == DIR_SEP || *p == TEXT('#') || *p == TEXT('?'))
1.28 cvs 1321: break;
1.67 cvs 1322: if (*p == TEXT(':'))
1.28 cvs 1323: {
1324: *p = 0;
1325: parts->access = after_access; /* Scheme has been specified */
1326:
1327: /* The combination of gcc, the "-O" flag and the HP platform is
1328: unhealthy. The following three lines is a quick & dirty fix, but is
1329: not recommended. Rather, turn off "-O". */
1330:
1331: /* after_access = p;*/
1332: /* while (*after_access == 0)*/
1333: /* after_access++;*/
1334: after_access = p+1;
1.67 cvs 1335: if (!ustrcasecmp(TEXT("URL"), parts->access))
1.28 cvs 1336: /* Ignore IETF's URL: pre-prefix */
1337: parts->access = NULL;
1338: else
1.25 cvs 1339: break;
1340: }
1341: }
1342:
1343: p = after_access;
1.43 cvs 1344: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1345: {
1.43 cvs 1346: if (p[1] == URL_SEP)
1.28 cvs 1347: {
1.25 cvs 1348: parts->host = p+2; /* host has been specified */
1.28 cvs 1349: *p = 0; /* Terminate access */
1350: /* look for end of host name if any */
1.67 cvs 1351: p = ustrchr (parts->host, URL_SEP);
1.28 cvs 1352: if (p)
1353: {
1.87 cvs 1354: *p = WC_EOS; /* Terminate host */
1.25 cvs 1355: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1356: }
1357: }
1358: else
1359: /* Root found but no host */
1360: parts->absolute = p+1;
1361: }
1362: else
1363: {
1.25 cvs 1364: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1365: }
1.25 cvs 1366: }
1367:
1368:
1369: /*----------------------------------------------------------------------
1.28 cvs 1370: AmayaParseUrl: parse a Name relative to another name
1371:
1372: This returns those parts of a name which are given (and requested)
1373: substituting bits from the related name where necessary.
1.25 cvs 1374:
1.28 cvs 1375: On entry,
1.25 cvs 1376: aName A filename given
1377: relatedName A name relative to which aName is to be parsed. Give
1378: it an empty string if aName is absolute.
1379: wanted A mask for the bits which are wanted.
1380:
1.28 cvs 1381: On exit,
1.25 cvs 1382: returns A pointer to a malloc'd string which MUST BE FREED
1383: ----------------------------------------------------------------------*/
1384: #ifdef __STDC__
1.84 cvs 1385: CHAR_T* AmayaParseUrl (const CHAR_T* aName, CHAR_T* relatedName, int wanted)
1.25 cvs 1386: #else /* __STDC__ */
1.84 cvs 1387: CHAR_T* AmayaParseUrl (aName, relatedName, wanted)
1388: const CHAR_T* aName;
1389: CHAR_T* relatedName;
1.28 cvs 1390: int wanted;
1.25 cvs 1391:
1392: #endif /* __STDC__ */
1393: {
1.84 cvs 1394: CHAR_T* return_value;
1.67 cvs 1395: CHAR_T result[MAX_LENGTH];
1396: CHAR_T name[MAX_LENGTH];
1397: CHAR_T rel[MAX_LENGTH];
1.84 cvs 1398: CHAR_T *p, *access;
1.29 cvs 1399: HTURI given, related;
1400: int len;
1.67 cvs 1401: CHAR_T used_sep;
1.84 cvs 1402: CHAR_T* used_str;
1.32 cvs 1403:
1.84 cvs 1404: if (ustrchr (aName, WC_DIR_SEP) || ustrchr (relatedName, WC_DIR_SEP))
1.33 cvs 1405: {
1.84 cvs 1406: used_str = WC_DIR_STR;
1407: used_sep = WC_DIR_SEP;
1.33 cvs 1408: }
1.32 cvs 1409: else
1.33 cvs 1410: {
1.84 cvs 1411: used_str = WC_URL_STR;
1412: used_sep = WC_URL_SEP;
1.33 cvs 1413: }
1.32 cvs 1414:
1.29 cvs 1415: /* Make working copies of input strings to cut up: */
1416: return_value = NULL;
1417: result[0] = 0; /* Clear string */
1.67 cvs 1418: ustrcpy (name, aName);
1.29 cvs 1419: if (relatedName != NULL)
1.67 cvs 1420: ustrcpy (rel, relatedName);
1.29 cvs 1421: else
1.84 cvs 1422: relatedName[0] = WC_EOS;
1.29 cvs 1423:
1424: scan (name, &given);
1425: scan (rel, &related);
1426: access = given.access ? given.access : related.access;
1427: if (wanted & AMAYA_PARSE_ACCESS)
1428: if (access)
1429: {
1.67 cvs 1430: ustrcat (result, access);
1.29 cvs 1431: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1432: ustrcat (result, TEXT(":"));
1.29 cvs 1433: }
1434:
1435: if (given.access && related.access)
1436: /* If different, inherit nothing. */
1.67 cvs 1437: if (ustrcmp (given.access, related.access) != 0)
1.29 cvs 1438: {
1439: related.host = 0;
1440: related.absolute = 0;
1441: related.relative = 0;
1442: related.fragment = 0;
1443: }
1444:
1445: if (wanted & AMAYA_PARSE_HOST)
1446: if(given.host || related.host)
1447: {
1448: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1449: ustrcat (result, TEXT("//"));
1450: ustrcat (result, given.host ? given.host : related.host);
1.29 cvs 1451: }
1452:
1453: if (given.host && related.host)
1454: /* If different hosts, inherit no path. */
1.67 cvs 1455: if (ustrcmp (given.host, related.host) != 0)
1.29 cvs 1456: {
1457: related.absolute = 0;
1458: related.relative = 0;
1459: related.fragment = 0;
1460: }
1461:
1462: if (wanted & AMAYA_PARSE_PATH)
1463: {
1464: if (given.absolute)
1465: {
1466: /* All is given */
1467: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1468: ustrcat (result, used_str);
1469: ustrcat (result, given.absolute);
1.25 cvs 1470: }
1.29 cvs 1471: else if (related.absolute)
1472: {
1473: /* Adopt path not name */
1.67 cvs 1474: ustrcat (result, used_str);
1475: ustrcat (result, related.absolute);
1.29 cvs 1476: if (given.relative)
1477: {
1478: /* Search part? */
1.67 cvs 1479: p = ustrchr (result, TEXT('?'));
1.29 cvs 1480: if (!p)
1.67 cvs 1481: p=result+ustrlen(result)-1;
1.33 cvs 1482: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1483: /* Remove filename */
1484: p[1]=0;
1485: /* Add given one */
1.67 cvs 1486: ustrcat (result, given.relative);
1.25 cvs 1487: }
1488: }
1.29 cvs 1489: else if (given.relative)
1490: /* what we've got */
1.67 cvs 1491: ustrcat (result, given.relative);
1.29 cvs 1492: else if (related.relative)
1.67 cvs 1493: ustrcat (result, related.relative);
1.29 cvs 1494: else
1495: /* No inheritance */
1.67 cvs 1496: ustrcat (result, used_str);
1.25 cvs 1497: }
1.29 cvs 1498:
1499: if (wanted & AMAYA_PARSE_ANCHOR)
1500: if (given.fragment || related.fragment)
1501: {
1502: if (given.absolute && given.fragment)
1503: {
1504: /*Fixes for relURLs...*/
1505: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1506: ustrcat (result, TEXT("#"));
1507: ustrcat (result, given.fragment);
1.29 cvs 1508: }
1509: else if (!(given.absolute) && !(given.fragment))
1.84 cvs 1510: ustrcat (result, TEXT(""));
1.29 cvs 1511: else
1512: {
1513: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1514: ustrcat (result, TEXT("#"));
1515: ustrcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1516: }
1517: }
1.67 cvs 1518: len = ustrlen (result);
1519: if ((return_value = TtaAllocString (len + 1)) != NULL)
1520: ustrcpy (return_value, result);
1.29 cvs 1521: return (return_value); /* exactly the right length */
1.25 cvs 1522: }
1523:
1524: /*----------------------------------------------------------------------
1525: HTCanon
1526: Canonicalizes the URL in the following manner starting from the host
1527: pointer:
1528:
1529: 1) The host name is converted to lowercase
1530: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1531:
1532: Return: OK The position of the current path part of the URL
1533: which might be the old one or a new one.
1534:
1535: ----------------------------------------------------------------------*/
1536: #ifdef __STDC__
1.84 cvs 1537: static CHAR_T* HTCanon (CHAR_T** filename, CHAR_T* host)
1.25 cvs 1538: #else /* __STDC__ */
1.67 cvs 1539: static STRING HTCanon (filename, host)
1.84 cvs 1540: CHAR_T** filename;
1541: CHAR_T* host;
1.25 cvs 1542: #endif /* __STDC__ */
1543: {
1.84 cvs 1544: CHAR_T* newname = NULL;
1545: CHAR_T used_sep;
1546: CHAR_T* path;
1547: CHAR_T* strptr;
1548: CHAR_T* port;
1549: CHAR_T* access = host-3;
1.32 cvs 1550:
1551:
1.84 cvs 1552: if (*filename && ustrchr (*filename, WC_URL_SEP))
1.33 cvs 1553: {
1.84 cvs 1554: used_sep = WC_URL_SEP;
1.33 cvs 1555: }
1556: else
1557: {
1.84 cvs 1558: used_sep = WC_DIR_SEP;
1.33 cvs 1559: }
1.32 cvs 1560:
1.82 cvs 1561: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1562: access--;
1.84 cvs 1563: if ((path = ustrchr (host, used_sep)) == NULL) /* Find path */
1564: path = host + ustrlen (host);
1565: if ((strptr = ustrchr (host, TEXT('@'))) != NULL && strptr < path) /* UserId */
1.82 cvs 1566: host = strptr;
1.84 cvs 1567: if ((port = ustrchr (host, TEXT(':'))) != NULL && port > path) /* Port number */
1.82 cvs 1568: port = NULL;
1.25 cvs 1569:
1570: strptr = host; /* Convert to lower-case */
1.82 cvs 1571: while (strptr < path)
1.33 cvs 1572: {
1.84 cvs 1573: *strptr = utolower (*strptr);
1.82 cvs 1574: strptr++;
1.33 cvs 1575: }
1.25 cvs 1576:
1577: /* Does the URL contain a full domain name? This also works for a
1578: numerical host name. The domain name is already made lower-case
1579: and without a trailing dot. */
1580: {
1.84 cvs 1581: CHAR_T* dot = port ? port : path;
1582: if (dot > *filename && *--dot == TEXT('.'))
1.33 cvs 1583: {
1.84 cvs 1584: CHAR_T* orig = dot;
1585: CHAR_T* dest = dot + 1;
1.82 cvs 1586: while ((*orig++ = *dest++));
1587: if (port) port--;
1.33 cvs 1588: path--;
1.25 cvs 1589: }
1590: }
1591: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1592: if (port)
1593: {
1.82 cvs 1594: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1595: {
1596: if (!newname)
1597: {
1.84 cvs 1598: CHAR_T* orig = port;
1599: CHAR_T* dest = port + 1;
1.82 cvs 1600: while ((*orig++ = *dest++));
1.33 cvs 1601: }
1602: }
1.84 cvs 1603: else if ((!ustrncmp (access, TEXT("http"), 4) &&
1604: (*(port + 1) == TEXT('8') &&
1605: *(port+2) == TEXT('0') &&
1.82 cvs 1606: (*(port+3) == used_sep || !*(port + 3)))) ||
1.84 cvs 1607: (!ustrncmp (access, TEXT("gopher"), 6) &&
1608: (*(port+1) == TEXT('7') &&
1609: *(port+2) == TEXT('0') &&
1.82 cvs 1610: (*(port+3) == used_sep || !*(port+3)))) ||
1.84 cvs 1611: (!ustrncmp (access, TEXT("ftp"), 3) &&
1612: (*(port+1) == TEXT('2') &&
1613: *(port + 2) == TEXT('1') &&
1.82 cvs 1614: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1615: if (!newname)
1616: {
1.84 cvs 1617: CHAR_T* orig = port;
1618: CHAR_T* dest = port + 3;
1.33 cvs 1619: while((*orig++ = *dest++));
1620: /* Update path position, Henry Minsky */
1621: path -= 3;
1.25 cvs 1622: }
1.33 cvs 1623: }
1624: else if (newname)
1.84 cvs 1625: ustrncat (newname, port, (int) (path - port));
1.33 cvs 1626: }
1.25 cvs 1627:
1.33 cvs 1628: if (newname)
1629: {
1.84 cvs 1630: CHAR_T* newpath = newname + ustrlen (newname);
1631: ustrcat (newname, path);
1.25 cvs 1632: path = newpath;
1.28 cvs 1633: /* Free old copy */
1634: TtaFreeMemory(*filename);
1.25 cvs 1635: *filename = newname;
1.33 cvs 1636: }
1.25 cvs 1637: return path;
1638: }
1639:
1640:
1641: /*----------------------------------------------------------------------
1.29 cvs 1642: SimplifyUrl: simplify a URI
1.32 cvs 1643: A URI is allowed to contain the sequence xxx/../ which may be
1644: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1645: Simplification helps us recognize duplicate URIs.
1.25 cvs 1646:
1.28 cvs 1647: Thus, /etc/junk/../fred becomes /etc/fred
1648: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1649:
1.28 cvs 1650: but we should NOT change
1651: http://fred.xxx.edu/../..
1.25 cvs 1652:
1653: or ../../albert.html
1654:
1.28 cvs 1655: In order to avoid empty URLs the following URLs become:
1.25 cvs 1656:
1657: /fred/.. becomes /fred/..
1658: /fred/././.. becomes /fred/..
1659: /fred/.././junk/.././ becomes /fred/..
1660:
1.28 cvs 1661: If more than one set of `://' is found (several proxies in cascade) then
1662: only the part after the last `://' is simplified.
1.25 cvs 1663:
1.28 cvs 1664: Returns: A string which might be the old one or a new one.
1.25 cvs 1665: ----------------------------------------------------------------------*/
1666: #ifdef __STDC__
1.84 cvs 1667: void SimplifyUrl (CHAR_T** url)
1.25 cvs 1668: #else /* __STDC__ */
1.29 cvs 1669: void SimplifyUrl (url)
1.84 cvs 1670: CHAR_T** url;
1.25 cvs 1671: #endif /* __STDC__ */
1672: {
1.84 cvs 1673: CHAR_T* path;
1674: CHAR_T* access;
1675: CHAR_T* newptr;
1676: CHAR_T* p;
1677: CHAR_T *orig, *dest, *end;
1.28 cvs 1678:
1.85 cvs 1679: CHAR_T used_sep;
1.77 cvs 1680: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1681: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1682:
1683:
1.28 cvs 1684: if (!url || !*url)
1685: return;
1686:
1.84 cvs 1687: if (ustrchr (*url, WC_URL_SEP))
1.33 cvs 1688: {
1.84 cvs 1689: used_sep = WC_URL_SEP;
1.33 cvs 1690: }
1.32 cvs 1691: else
1.33 cvs 1692: {
1.84 cvs 1693: used_sep = WC_DIR_SEP;
1.33 cvs 1694: }
1.32 cvs 1695:
1.77 cvs 1696: /* should we simplify double dot? */
1697: path = *url;
1.84 cvs 1698: if (*path == TEXT('.') && *(path + 1) == TEXT('.'))
1.77 cvs 1699: ddot_simplify = FALSE;
1700: else
1701: ddot_simplify = TRUE;
1702:
1.28 cvs 1703: /* Find any scheme name */
1.84 cvs 1704: if ((path = ustrstr (*url, TEXT("://"))) != NULL)
1.33 cvs 1705: {
1706: /* Find host name */
1.28 cvs 1707: access = *url;
1.84 cvs 1708: while (access < path && (*access = utolower (*access)))
1.82 cvs 1709: access++;
1.28 cvs 1710: path += 3;
1.84 cvs 1711: while ((newptr = ustrstr (path, TEXT ("://"))) != NULL)
1.82 cvs 1712: /* For proxies */
1713: path = newptr+3;
1714: /* We have a host name */
1.84 cvs 1715: path = HTCanon (url, path);
1.25 cvs 1716: }
1.84 cvs 1717: else if ((path = ustrstr (*url, TEXT(":/"))) != NULL)
1.28 cvs 1718: path += 2;
1719: else
1720: path = *url;
1.25 cvs 1721:
1.84 cvs 1722: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1723: /* Some URLs start //<foo> */
1724: path += 1;
1.84 cvs 1725: else if (!ustrncmp (path, TEXT("news:"), 5))
1.28 cvs 1726: {
1.84 cvs 1727: newptr = ustrchr (path+5, TEXT('@'));
1.28 cvs 1728: if (!newptr)
1729: newptr = path + 5;
1730: while (*newptr)
1731: {
1732: /* Make group or host lower case */
1.84 cvs 1733: *newptr = utolower (*newptr);
1.28 cvs 1734: newptr++;
1.25 cvs 1735: }
1.28 cvs 1736: /* Doesn't need to do any more */
1737: return;
1.25 cvs 1738: }
1.28 cvs 1739:
1740: if ((p = path))
1741: {
1.67 cvs 1742: if (!((end = ustrchr (path, TEXT(';'))) || (end = ustrchr (path, TEXT('?'))) ||
1743: (end = ustrchr (path, TEXT('#')))))
1744: end = path + ustrlen (path);
1.28 cvs 1745:
1746: /* Parse string second time to simplify */
1747: p = path;
1748: while (p < end)
1749: {
1.77 cvs 1750: /* if we're pointing to a char, it's safe to reactivate the ../ convertion */
1751: if (!ddot_simplify && *p != TEXT('.') && *p != used_sep)
1752: ddot_simplify = TRUE;
1753:
1.33 cvs 1754: if (*p==used_sep)
1.28 cvs 1755: {
1.67 cvs 1756: if (p > *url && *(p+1) == TEXT('.') && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1757: {
1758: orig = p + 1;
1.84 cvs 1759: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1760: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1761: end = orig - 1;
1762: }
1.77 cvs 1763: else if (ddot_simplify && *(p+1) == TEXT('.') && *(p+2) == TEXT('.')
1764: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1765: {
1766: newptr = p;
1.52 cvs 1767: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1768: if (*newptr == used_sep)
1769: orig = newptr + 1;
1.28 cvs 1770: else
1.52 cvs 1771: orig = newptr;
1772:
1773: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1774: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1775: end = orig-1;
1776: /* Start again with prev slash */
1777: p = newptr;
1.28 cvs 1778: }
1.33 cvs 1779: else if (*(p+1) == used_sep)
1.28 cvs 1780: {
1.33 cvs 1781: while (*(p+1) == used_sep)
1.28 cvs 1782: {
1783: orig = p;
1784: dest = p + 1;
1785: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1786: end = orig-1;
1787: }
1788: }
1789: else
1.25 cvs 1790: p++;
1.28 cvs 1791: }
1792: else
1793: p++;
1.25 cvs 1794: }
1795: }
1.51 cvs 1796:
1797: /*
1798: ** Check for host/../.. kind of things
1799: */
1.77 cvs 1800: if (*path == used_sep && *(path+1) == TEXT('.') && *(path+2) == TEXT('.')
1801: && (!*(path+3) || *(path+3) == used_sep))
1.87 cvs 1802: *(path+1) = WC_EOS;
1.51 cvs 1803:
1.28 cvs 1804: return;
1805: }
1806:
1807:
1808: /*----------------------------------------------------------------------
1809: NormalizeFile normalizes local names.
1810: Return TRUE if target and src differ.
1811: ----------------------------------------------------------------------*/
1812: #ifdef __STDC__
1.90 cvs 1813: ThotBool NormalizeFile (CHAR_T* src, CHAR_T* target)
1.28 cvs 1814: #else
1.90 cvs 1815: ThotBool NormalizeFile (src, target)
1.84 cvs 1816: CHAR_T* src;
1817: CHAR_T* target;
1.28 cvs 1818:
1819: #endif
1820: {
1.93 ! cvs 1821: # ifndef _WINDOWS
1.90 cvs 1822: CHAR_T *s;
1.93 ! cvs 1823: int i;
! 1824: # endif /* !_WINDOWS */
1.82 cvs 1825: ThotBool change;
1.90 cvs 1826: int start_index; /* the first char that we'll copy */
1.28 cvs 1827:
1.54 cvs 1828: change = FALSE;
1.90 cvs 1829: start_index = 0;
1830:
1831: if (!src || src[0] == WC_EOS)
1832: return FALSE;
1833:
1834: /* @@ do I need file: or file:/ here? */
1.84 cvs 1835: if (ustrncmp (src, TEXT("file:"), 5) == 0)
1.28 cvs 1836: {
1.90 cvs 1837: /* remove the prefix file: */
1838: start_index += 5;
1839:
1840: if (ustrncmp (&src[start_index], TEXT("/localhost"), 10) == 0)
1841: /* remove the localhost prefix */
1842: start_index += 10;
1843:
1844: if (src[start_index] == WC_EOS)
1845: /* if there's nothing afterwards, add a DIR_STR */
1846: ustrcpy (target, WC_DIR_STR);
1847: #ifndef _WINDOWS
1848: else if (src[start_index] == TEXT('~'))
1849: {
1850: /* replace ~ */
1851: s = TtaGetEnvString ("HOME");
1852: ustrcpy (target, s);
1853: start_index++;
1854: i = ustrlen (target);
1855: ustrcat (&target[i], &src[start_index]);
1856: }
1857: #endif /* _WINDOWS */
1858: else
1859: {
1860: CleanCopyFileURL (target, &src[start_index]);
1861: change = TRUE;
1862: }
1.28 cvs 1863: }
1.90 cvs 1864: #ifndef _WINDOWS
1.84 cvs 1865: else if (src[0] == TEXT('~'))
1.53 cvs 1866: {
1867: /* replace ~ */
1.82 cvs 1868: s = TtaGetEnvString ("HOME");
1.84 cvs 1869: ustrcpy (target, s);
1.90 cvs 1870: #if 0
1.84 cvs 1871: if (src[1] != WC_DIR_SEP)
1872: ustrcat (target, WC_DIR_STR);
1.90 cvs 1873: #endif
1874: i = ustrlen (target);
1.92 cvs 1875: ustrcpy (&target[i], &src[1]);
1.54 cvs 1876: change = TRUE;
1.53 cvs 1877: }
1.90 cvs 1878: #endif /* _WINDOWS */
1879:
1.28 cvs 1880: else
1.92 cvs 1881: ustrcpy (target, src);
1.28 cvs 1882:
1883: /* remove /../ and /./ */
1.29 cvs 1884: SimplifyUrl (&target);
1.54 cvs 1885: if (!change)
1.84 cvs 1886: change = ustrcmp (src, target);
1.28 cvs 1887: return (change);
1.25 cvs 1888: }
1889:
1.28 cvs 1890:
1.25 cvs 1891: /*----------------------------------------------------------------------
1.31 cvs 1892: MakeRelativeURL: make relative name
1.25 cvs 1893:
1.28 cvs 1894: This function creates and returns a string which gives an expression of
1895: one address as related to another. Where there is no relation, an absolute
1896: address is retured.
1.25 cvs 1897:
1.28 cvs 1898: On entry,
1.25 cvs 1899: Both names must be absolute, fully qualified names of nodes
1900: (no fragment bits)
1901:
1.28 cvs 1902: On exit,
1.25 cvs 1903: The return result points to a newly allocated name which, if
1904: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1905: The caller is responsible for freeing the resulting name later.
1906: ----------------------------------------------------------------------*/
1907: #ifdef __STDC__
1.67 cvs 1908: STRING MakeRelativeURL (STRING aName, STRING relatedName)
1.25 cvs 1909: #else /* __STDC__ */
1.67 cvs 1910: STRING MakeRelativeURL (aName, relatedName)
1911: STRING aName;
1912: STRING relatedName;
1.25 cvs 1913: #endif /* __STDC__ */
1914: {
1.87 cvs 1915: CHAR_T* return_value;
1.67 cvs 1916: CHAR_T result[MAX_LENGTH];
1.87 cvs 1917: CHAR_T* p;
1918: CHAR_T* q;
1919: CHAR_T* after_access;
1920: CHAR_T* last_slash = NULL;
1.29 cvs 1921: int slashes, levels, len;
1922:
1.44 cvs 1923: # ifdef _WINDOWS
1924: int ndx;
1925: # endif /* _WINDOWS */
1926:
1.29 cvs 1927: if (aName == NULL || relatedName == NULL)
1928: return (NULL);
1929:
1930: slashes = 0;
1931: after_access = NULL;
1932: p = aName;
1933: q = relatedName;
1934: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1935: {
1936: /* Find extent of match */
1.67 cvs 1937: if (*p == TEXT(':'))
1.29 cvs 1938: after_access = p + 1;
1.28 cvs 1939: if (*p == DIR_SEP)
1.27 cvs 1940: {
1.29 cvs 1941: /* memorize the last slash position and count them */
1.27 cvs 1942: last_slash = p;
1943: slashes++;
1.25 cvs 1944: }
1945: }
1946:
1.31 cvs 1947: /* q, p point to the first non-matching character or zero */
1.87 cvs 1948: if (*q == WC_EOS)
1.31 cvs 1949: {
1950: /* New name is a subset of the related name */
1951: /* exactly the right length */
1.67 cvs 1952: len = ustrlen (p);
1953: if ((return_value = TtaAllocString (len + 1)) != NULL)
1954: ustrcpy (return_value, p);
1.31 cvs 1955: }
1956: else if ((slashes < 2 && after_access == NULL)
1957: || (slashes < 3 && after_access != NULL))
1958: {
1959: /* Two names whitout common path */
1960: /* exactly the right length */
1.67 cvs 1961: len = ustrlen (aName);
1962: if ((return_value = TtaAllocString (len + 1)) != NULL)
1963: ustrcpy (return_value, aName);
1.31 cvs 1964: }
1965: else
1966: {
1967: /* Some path in common */
1.67 cvs 1968: if (slashes == 3 && ustrncmp (aName, TEXT("http:"), 5) == 0)
1.31 cvs 1969: /* just the same server */
1.67 cvs 1970: ustrcpy (result, last_slash);
1.31 cvs 1971: else
1972: {
1973: levels= 0;
1.67 cvs 1974: for (; *q && *q != TEXT('#') && *q != TEXT(';') && *q != TEXT('?'); q++)
1.31 cvs 1975: if (*q == DIR_SEP)
1976: levels++;
1977:
1.87 cvs 1978: result[0] = WC_EOS;
1.31 cvs 1979: for (;levels; levels--)
1.67 cvs 1980: ustrcat (result, TEXT("../"));
1981: ustrcat (result, last_slash+1);
1.31 cvs 1982: }
1.52 cvs 1983:
1984: if (!*result)
1.67 cvs 1985: ustrcat (result, TEXT("./"));
1.52 cvs 1986:
1.31 cvs 1987: /* exactly the right length */
1.67 cvs 1988: len = ustrlen (result);
1989: if ((return_value = TtaAllocString (len + 1)) != NULL)
1990: ustrcpy (return_value, result);
1.52 cvs 1991:
1.25 cvs 1992: }
1.44 cvs 1993: # ifdef _WINDOWS
1.67 cvs 1994: len = ustrlen (return_value);
1.44 cvs 1995: for (ndx = 0; ndx < len; ndx ++)
1.67 cvs 1996: if (return_value[ndx] == TEXT('\\'))
1997: return_value[ndx] = TEXT('/') ;
1.44 cvs 1998: # endif /* _WINDOWS */
1.29 cvs 1999: return (return_value);
1.24 cvs 2000: }
1.35 cvs 2001:
2002:
Webmaster