Annotation of Amaya/amaya/AHTURLTools.c, revision 1.101
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"
1.100 kahan 23: #include "query_f.h"
1.8 cvs 24:
1.24 cvs 25: #define MAX_PRINT_URL_LENGTH 50
1.29 cvs 26: typedef struct _HTURI {
1.67 cvs 27: STRING access; /* Now known as "scheme" */
28: STRING host;
29: STRING absolute;
30: STRING relative;
31: STRING fragment;
1.29 cvs 32: } HTURI;
1.24 cvs 33:
1.28 cvs 34:
35: /*----------------------------------------------------------------------
36: ConvertToLowerCase
37: Converts a string to lowercase.
38: ----------------------------------------------------------------------*/
1.22 cvs 39: #ifdef __STDC__
1.67 cvs 40: void ConvertToLowerCase (STRING string)
1.28 cvs 41: #else /* __STDC__ */
1.38 cvs 42: void ConvertToLowerCase (string)
1.67 cvs 43: STRING string;
1.28 cvs 44:
45: #endif /* __STDC__ */
46: {
47: int i;
1.93 cvs 48:
1.28 cvs 49: if (!string)
50: return;
51:
1.87 cvs 52: for (i = 0; string[i] != WC_EOS; i++)
1.67 cvs 53: string[i] = utolower (string[i]);
1.28 cvs 54: }
1.22 cvs 55:
1.8 cvs 56: /*----------------------------------------------------------------------
1.75 cvs 57: EscapeChar
58: writes the equivalent escape code of a char in a string
59: ----------------------------------------------------------------------*/
60: #ifdef __STDC__
61: void EscapeChar (STRING string, UCHAR_T c)
62: #else
63: void EscapeChar (string, c)
64: STRING string;
65: UCHAR_T c;
66:
67: #endif
68: {
69: c &= 0xFF; /* strange behavior under solaris? */
70: usprintf (string, TEXT("%02x"), (unsigned int) c);
71: }
72:
73: /*----------------------------------------------------------------------
1.96 cvs 74: UnEscapeChar
75: writes the equivalent hex code to a %xx coded char
76: ----------------------------------------------------------------------*/
77: #ifdef __STDC__
78: static CHAR_T UnEscapeChar (CHAR_T c)
79: #else
80: static CHAR_T UnEscapeChar (c)
81: CHAR_T c;
82: #endif
83: {
84: return c >= TEXT('0') && c <= TEXT('9') ? c - TEXT('0')
85: : c >= TEXT('A') && c <= TEXT('F') ? c - TEXT('A') + 10
86: : c - TEXT('a') + 10; /* accept small letters just in case */
87: }
88:
89: /*----------------------------------------------------------------------
1.75 cvs 90: EscapeURL
91: Takes a URL and escapes all protected chars into
92: %xx sequences. Also, removes any leading white spaces
93: Returns either NULL or a new buffer, which must be freed by the caller
94: ----------------------------------------------------------------------*/
95: #ifdef __STDC__
96: STRING EscapeURL (const STRING url)
97: #else
98: STRING EscapeURL (url)
99: STRING url;
100: #endif /* __STDC__ */
101: {
102: STRING buffer;
103: int buffer_len;
104: int buffer_free_mem;
105: PCHAR_T ptr;
106: int new_chars;
107: void *status;
108:
109: if (url && *url)
110: {
1.76 cvs 111: buffer_free_mem = ustrlen (url) + 20;
112: buffer = TtaAllocString (buffer_free_mem + 1);
1.75 cvs 113: ptr = url;
114: buffer_len = 0;
115:
116: while (*ptr)
117: {
118: switch (*ptr)
119: {
120: /* put here below all the chars that need to
121: be escaped into %xx */
1.81 cvs 122: case 0x27: /* & */
123: case 0x20: /* space */
1.75 cvs 124: new_chars = 3;
125: break;
126:
127: default:
128: new_chars = 1;
129: break;
130: }
131:
132: /* see if we need extra room in the buffer */
133: if (new_chars > buffer_free_mem)
134: {
1.76 cvs 135: buffer_free_mem = 20;
1.75 cvs 136: status = TtaRealloc (buffer, sizeof (CHAR_T)
137: * (buffer_len + buffer_free_mem + 1));
138: if (status)
139: buffer = (STRING) status;
140: else {
141: /* @@ maybe we should do some other behavior here, like
142: freeing the buffer and return a void thing */
1.87 cvs 143: buffer[buffer_len] = WC_EOS;
1.75 cvs 144: break;
145: }
146: }
147: /* escape the char */
148: if (new_chars == 3)
149: {
150: buffer[buffer_len] = TEXT('%');
151: EscapeChar (&buffer[buffer_len+1], *ptr);
152: }
153: else
154: buffer[buffer_len] = *ptr;
155:
156: /* update the status */
157: buffer_len += new_chars;
158: buffer_free_mem -= new_chars;
159: /* examine the next char */
160: ptr++;
161: }
1.87 cvs 162: buffer[buffer_len] = WC_EOS;
1.75 cvs 163: }
1.76 cvs 164: else
165: buffer = NULL;
166:
1.75 cvs 167: return (buffer);
168: }
169:
170:
171: /*----------------------------------------------------------------------
1.11 cvs 172: ExplodeURL
1.8 cvs 173: ----------------------------------------------------------------------*/
174: #ifdef __STDC__
175: void ExplodeURL (char *url, char **proto, char **host, char **dir, char **file)
176: #else
177: void ExplodeURL (url, proto, host, dir, file)
178: char *url;
179: char **proto;
180: char **host;
181: char **dir;
182: char **file;
183:
184: #endif
185: {
1.33 cvs 186: char *curr, *temp;
187: char used_sep;
1.32 cvs 188:
1.33 cvs 189: if (url && strchr (url, URL_SEP))
190: used_sep = URL_SEP;
191: else
192: used_sep = DIR_SEP;
1.8 cvs 193:
194: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
195: (dir == NULL) || (file == NULL))
196: return;
197:
198: /* initialize every pointer */
199: *proto = *host = *dir = *file = NULL;
200:
201: /* skip any leading space */
202: while ((*url == SPACE) || (*url == TAB))
203: url++;
1.9 cvs 204: curr = url;
205: if (*curr == 0)
1.8 cvs 206: goto finished;
207:
208: /* go to the end of the URL */
1.68 cvs 209: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
210: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 211: curr++;
1.8 cvs 212:
213: /* mark the end of the chain */
1.9 cvs 214: *curr = EOS;
215: curr--;
216: if (curr <= url)
1.8 cvs 217: goto finished;
218:
219: /* search the next DIR_SEP indicating the beginning of the file name */
220: do
1.11 cvs 221: curr--;
1.33 cvs 222: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 223:
1.9 cvs 224: if (curr < url)
1.8 cvs 225: goto finished;
1.9 cvs 226: *file = curr + 1;
1.8 cvs 227:
228: /* mark the end of the dir */
1.9 cvs 229: *curr = EOS;
230: curr--;
231: if (curr < url)
1.8 cvs 232: goto finished;
233:
1.29 cvs 234: /* search for the DIR_STR indicating the host name start */
1.33 cvs 235: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 236: curr--;
1.8 cvs 237:
238: /* if we found it, separate the host name from the directory */
1.33 cvs 239: if ((*curr == DIR_SEP) && (*(curr + 1) == used_sep))
1.8 cvs 240: {
1.9 cvs 241: *host = temp = curr + 2;
1.33 cvs 242: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 243: temp++;
1.33 cvs 244: if (*temp == used_sep)
1.8 cvs 245: {
246: *temp = EOS;
247: *dir = temp + 1;
248: }
249: }
250: else
1.11 cvs 251: *dir = curr;
252:
1.9 cvs 253: if (curr <= url)
1.8 cvs 254: goto finished;
255:
256: /* mark the end of the proto */
1.9 cvs 257: *curr = EOS;
258: curr--;
259: if (curr < url)
1.8 cvs 260: goto finished;
261:
1.67 cvs 262: if (*curr == TEXT(':'))
1.8 cvs 263: {
1.9 cvs 264: *curr = EOS;
265: curr--;
1.8 cvs 266: }
267: else
268: goto finished;
1.11 cvs 269:
1.9 cvs 270: if (curr < url)
1.8 cvs 271: goto finished;
1.9 cvs 272: while ((curr > url) && (isalpha (*curr)))
273: curr--;
274: *proto = curr;
1.8 cvs 275:
276: finished:;
277:
278: #ifdef AMAYA_DEBUG
279: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
280: if (*proto)
281: fprintf (stderr, "proto : %s, ", *proto);
282: if (*host)
283: fprintf (stderr, "host : %s, ", *host);
284: if (*dir)
285: fprintf (stderr, "dir : %s, ", *dir);
286: if (*file)
287: fprintf (stderr, "file : %s ", *file);
288: fprintf (stderr, "\n");
289: #endif
290:
291: }
1.3 cvs 292:
1.61 cvs 293:
294: /*----------------------------------------------------------------------
295: ExtractSuffix extract suffix from document nane.
296: ----------------------------------------------------------------------*/
297: #ifdef __STDC__
1.91 cvs 298: void ExtractSuffix (CHAR_T* aName, CHAR_T* aSuffix)
1.61 cvs 299: #else
300: void ExtractSuffix (aName, aSuffix)
1.91 cvs 301: CHAR_T* aName;
302: CHAR_T* aSuffix;
1.61 cvs 303:
304: #endif
305: {
306: int lg, i;
307: STRING ptr, oldptr;
308:
309: if (!aSuffix || !aName)
310: /* bad suffix */
311: return;
312:
1.87 cvs 313: aSuffix[0] = WC_EOS;
1.61 cvs 314: lg = ustrlen (aName);
315: if (lg)
316: {
317: /* the name is not empty */
318: oldptr = ptr = &aName[0];
319: do
320: {
1.67 cvs 321: ptr = ustrrchr (oldptr, TEXT('.'));
1.61 cvs 322: if (ptr)
323: oldptr = &ptr[1];
324: }
325: while (ptr);
326:
327: i = (int) (oldptr) - (int) (aName); /* name length */
328: if (i > 1)
329: {
1.87 cvs 330: aName[i - 1] = WC_EOS;
1.61 cvs 331: if (i != lg)
332: ustrcpy (aSuffix, oldptr);
333: }
334: }
335: }
336:
1.4 cvs 337: /*----------------------------------------------------------------------
1.9 cvs 338: IsHTMLName
339: returns TRUE if path points to an HTML resource.
1.4 cvs 340: ----------------------------------------------------------------------*/
1.3 cvs 341: #ifdef __STDC__
1.84 cvs 342: ThotBool IsHTMLName (const CHAR_T* path)
1.3 cvs 343: #else /* __STDC__ */
1.67 cvs 344: ThotBool IsHTMLName (path)
1.84 cvs 345: const CHAR_T* path;
1.3 cvs 346: #endif /* __STDC__ */
347: {
1.101 ! cvs 348: CHAR_T temppath[MAX_LENGTH];
! 349: CHAR_T suffix[MAX_LENGTH];
! 350: CHAR_T nsuffix[MAX_LENGTH];
! 351: int i;
1.5 cvs 352:
1.101 ! cvs 353: if (!path)
! 354: return (FALSE);
1.5 cvs 355:
1.101 ! cvs 356: ustrcpy (temppath, path);
! 357: ExtractSuffix (temppath, suffix);
! 358: i = 0;
! 359: while (suffix[i] != WC_EOS)
! 360: {
! 361: /* Normalize the suffix */
! 362: i = 0;
! 363: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1)
! 364: {
! 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 if (!ustrcmp (nsuffix, TEXT("gz")))
! 378: {
! 379: /* take into account compressed files */
! 380: ExtractSuffix (temppath, suffix);
! 381: /* Normalize the suffix */
! 382: i = 0;
! 383: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1)
! 384: {
! 385: nsuffix[i] = utolower (suffix[i]);
! 386: i++;
! 387: }
! 388: nsuffix[i] = WC_EOS;
! 389: if (!ustrcmp (nsuffix, TEXT("html")) ||
! 390: !ustrcmp (nsuffix, TEXT("htm")) ||
! 391: !ustrcmp (nsuffix, TEXT("shtml")) ||
! 392: !ustrcmp (nsuffix, TEXT("jsp")) ||
! 393: !ustrcmp (nsuffix, TEXT("xht")) ||
! 394: !ustrcmp (nsuffix, TEXT("xhtm")) ||
! 395: !ustrcmp (nsuffix, TEXT("xhtml")))
! 396: return (TRUE);
! 397: else
! 398: return (FALSE);
! 399: }
! 400: else
! 401: /* check if there is another suffix */
! 402: ExtractSuffix (temppath, suffix);
! 403: }
1.88 cvs 404: return (FALSE);
1.3 cvs 405: }
406:
1.4 cvs 407: /*----------------------------------------------------------------------
1.56 cvs 408: IsXMLName
409: returns TRUE if path points to an XML resource.
410: ----------------------------------------------------------------------*/
411: #ifdef __STDC__
1.67 cvs 412: ThotBool IsXMLName (const STRING path)
1.56 cvs 413: #else /* __STDC__ */
1.67 cvs 414: ThotBool IsXMLName (path)
415: const STRING path;
1.56 cvs 416: #endif /* __STDC__ */
417: {
1.67 cvs 418: CHAR_T temppath[MAX_LENGTH];
419: CHAR_T suffix[MAX_LENGTH];
1.56 cvs 420:
421: if (!path)
422: return (FALSE);
423:
1.67 cvs 424: ustrcpy (temppath, path);
1.56 cvs 425: ExtractSuffix (temppath, suffix);
426:
1.67 cvs 427: if (!ustrcasecmp (suffix, TEXT("xml")) ||
428: !ustrcasecmp (suffix, TEXT("xht")) ||
429: !ustrcmp (suffix, TEXT("xhtm")) ||
430: !ustrcmp (suffix, TEXT("xhtml")))
1.56 cvs 431: return (TRUE);
1.67 cvs 432: else if (!ustrcmp (suffix, TEXT("gz")))
1.56 cvs 433: {
434: /* take into account compressed files */
435: ExtractSuffix (temppath, suffix);
1.67 cvs 436: if (!ustrcasecmp (suffix, TEXT("xml")) ||
437: !ustrcasecmp (suffix, TEXT("xht")) ||
438: !ustrcmp (suffix, TEXT("xhtm")) ||
439: !ustrcmp (suffix, TEXT("xhtml")))
1.60 cvs 440: return (TRUE);
441: else
442: return (FALSE);
443: }
444: else
445: return (FALSE);
446: }
447:
448: /*----------------------------------------------------------------------
449: IsCSSName
450: returns TRUE if path points to an XML resource.
451: ----------------------------------------------------------------------*/
452: #ifdef __STDC__
1.67 cvs 453: ThotBool IsCSSName (const STRING path)
1.60 cvs 454: #else /* __STDC__ */
1.67 cvs 455: ThotBool IsCSSName (path)
456: const STRING path;
1.60 cvs 457: #endif /* __STDC__ */
458: {
1.67 cvs 459: CHAR_T temppath[MAX_LENGTH];
460: CHAR_T suffix[MAX_LENGTH];
1.60 cvs 461:
462: if (!path)
463: return (FALSE);
464:
1.67 cvs 465: ustrcpy (temppath, path);
1.60 cvs 466: ExtractSuffix (temppath, suffix);
467:
1.67 cvs 468: if (!ustrcasecmp (suffix, TEXT("css")))
1.60 cvs 469: return (TRUE);
1.67 cvs 470: else if (!ustrcmp (suffix, TEXT("gz")))
1.60 cvs 471: {
472: /* take into account compressed files */
473: ExtractSuffix (temppath, suffix);
1.67 cvs 474: if (!ustrcasecmp (suffix, TEXT("css")))
1.56 cvs 475: return (TRUE);
476: else
477: return (FALSE);
478: }
479: else
480: return (FALSE);
481: }
482:
483: /*----------------------------------------------------------------------
1.9 cvs 484: IsImageName
485: returns TRUE if path points to an image resource.
1.4 cvs 486: ----------------------------------------------------------------------*/
1.3 cvs 487: #ifdef __STDC__
1.67 cvs 488: ThotBool IsImageName (const STRING path)
1.3 cvs 489: #else /* __STDC__ */
1.67 cvs 490: ThotBool IsImageName (path)
491: const STRING path;
1.3 cvs 492: #endif /* __STDC__ */
493: {
1.67 cvs 494: CHAR_T temppath[MAX_LENGTH];
495: CHAR_T suffix[MAX_LENGTH];
496: CHAR_T nsuffix[MAX_LENGTH];
1.5 cvs 497: int i;
498:
499: if (!path)
1.13 cvs 500: return (FALSE);
1.5 cvs 501:
1.67 cvs 502: ustrcpy (temppath, path);
1.5 cvs 503: ExtractSuffix (temppath, suffix);
504:
505: /* Normalize the suffix */
506: i = 0;
1.87 cvs 507: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1)
1.13 cvs 508: {
1.67 cvs 509: nsuffix[i] = utolower (suffix[i]);
1.13 cvs 510: i++;
511: }
1.87 cvs 512: nsuffix[i] = WC_EOS;
1.67 cvs 513: if ((!ustrcmp (nsuffix, TEXT("gif"))) || (!ustrcmp (nsuffix, TEXT("xbm"))) ||
514: (!ustrcmp (nsuffix, TEXT("xpm"))) || (!ustrcmp (nsuffix, TEXT("jpg"))) ||
515: (!ustrcmp (nsuffix, TEXT("png"))) || (!ustrcmp (nsuffix, TEXT("au"))))
1.39 cvs 516: return (TRUE);
517: return (FALSE);
1.3 cvs 518: }
519:
1.4 cvs 520: /*----------------------------------------------------------------------
1.58 cvs 521: IsImageType
522: returns TRUE if type points to an image resource.
523: ----------------------------------------------------------------------*/
524: #ifdef __STDC__
1.67 cvs 525: ThotBool IsImageType (const STRING type)
1.58 cvs 526: #else /* __STDC__ */
1.67 cvs 527: ThotBool IsImageType (type)
528: const STRING type;
1.58 cvs 529: #endif /* __STDC__ */
530: {
1.67 cvs 531: CHAR_T temptype[MAX_LENGTH];
1.58 cvs 532: int i;
533:
534: if (!type)
535: return (FALSE);
536:
1.67 cvs 537: ustrcpy (temptype, type);
1.58 cvs 538: /* Normalize the type */
539: i = 0;
1.87 cvs 540: while (temptype[i] != WC_EOS)
1.58 cvs 541: {
542: temptype[i] = tolower (temptype[i]);
543: i++;
544: }
1.67 cvs 545: if ((!ustrcmp (temptype, TEXT("gif"))) || (!ustrcmp (temptype, TEXT("x-xbitmap"))) ||
546: (!ustrcmp (temptype, TEXT("x-xpixmap"))) || (!ustrcmp (temptype, TEXT("jpeg"))) ||
547: (!ustrcmp (temptype, TEXT("png"))))
1.58 cvs 548: return (TRUE);
549: return (FALSE);
550: }
551:
552: /*----------------------------------------------------------------------
1.9 cvs 553: IsTextName
1.4 cvs 554: ----------------------------------------------------------------------*/
1.3 cvs 555: #ifdef __STDC__
1.67 cvs 556: ThotBool IsTextName (const STRING path)
1.3 cvs 557: #else /* __STDC__ */
1.67 cvs 558: ThotBool IsTextName (path)
559: const STRING path;
1.3 cvs 560:
561: #endif /* __STDC__ */
562: {
1.67 cvs 563: CHAR_T temppath[MAX_LENGTH];
564: CHAR_T suffix[MAX_LENGTH];
565: CHAR_T nsuffix[MAX_LENGTH];
1.5 cvs 566: int i;
567:
568: if (!path)
1.13 cvs 569: return (FALSE);
1.5 cvs 570:
1.67 cvs 571: ustrcpy (temppath, path);
1.5 cvs 572: ExtractSuffix (temppath, suffix);
573:
574: /* Normalize the suffix */
575: i = 0;
1.87 cvs 576: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1)
1.5 cvs 577: {
1.25 cvs 578: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 579: i++;
580: }
1.87 cvs 581: nsuffix[i] = WC_EOS;
1.5 cvs 582:
1.67 cvs 583: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd"))))
1.13 cvs 584: return (TRUE);
1.67 cvs 585: else if (!ustrcmp (nsuffix, TEXT("gz")))
1.13 cvs 586: {
1.39 cvs 587: /* take into account compressed files */
1.13 cvs 588: ExtractSuffix (temppath, suffix);
589: /* Normalize the suffix */
590: i = 0;
1.87 cvs 591: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1)
1.13 cvs 592: {
1.25 cvs 593: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 594: i++;
595: }
1.87 cvs 596: nsuffix[i] = WC_EOS;
1.67 cvs 597: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd"))))
1.13 cvs 598: return (TRUE);
599: else
600: return (FALSE);
601: }
602: else
603: return (FALSE);
1.3 cvs 604: }
605:
1.4 cvs 606: /*----------------------------------------------------------------------
1.9 cvs 607: IsHTTPPath
608: returns TRUE if path is in fact an http URL.
1.4 cvs 609: ----------------------------------------------------------------------*/
1.3 cvs 610: #ifdef __STDC__
1.67 cvs 611: ThotBool IsHTTPPath (const STRING path)
1.3 cvs 612: #else /* __STDC__ */
1.67 cvs 613: ThotBool IsHTTPPath (path)
614: const STRING path;
1.3 cvs 615: #endif /* __STDC__ */
616: {
1.5 cvs 617: if (!path)
618: return FALSE;
1.3 cvs 619:
1.67 cvs 620: if ((!ustrncmp (path, TEXT("http:"), 5) != 0)
1.100 kahan 621: || (AHTFTPURL_flag () && !ustrncmp (path, TEXT("ftp:"), 4))
1.67 cvs 622: || !ustrncmp (path, TEXT("internal:"), 9))
1.58 cvs 623: return TRUE;
624: return FALSE;
1.3 cvs 625: }
626:
1.4 cvs 627: /*----------------------------------------------------------------------
1.9 cvs 628: IsWithParameters
629: returns TRUE if url has a concatenated query string.
1.4 cvs 630: ----------------------------------------------------------------------*/
1.3 cvs 631: #ifdef __STDC__
1.66 cvs 632: ThotBool IsWithParameters (const char *url)
1.3 cvs 633: #else /* __STDC__ */
1.66 cvs 634: ThotBool IsWithParameters (url)
1.34 cvs 635: const char *url;
1.3 cvs 636: #endif /* __STDC__ */
637: {
1.5 cvs 638: int i;
1.3 cvs 639:
1.9 cvs 640: if ((!url) || (url[0] == EOS))
1.5 cvs 641: return FALSE;
1.3 cvs 642:
1.9 cvs 643: i = strlen (url) - 1;
644: while (i > 0 && url[i--] != '?')
1.5 cvs 645: if (i < 0)
646: return FALSE;
1.3 cvs 647:
1.5 cvs 648: /* There is a parameter */
649: return TRUE;
1.3 cvs 650: }
651:
1.4 cvs 652: /*----------------------------------------------------------------------
1.9 cvs 653: IsW3Path
654: returns TRUE if path is in fact a URL.
1.4 cvs 655: ----------------------------------------------------------------------*/
1.3 cvs 656: #ifdef __STDC__
1.84 cvs 657: ThotBool IsW3Path (const CHAR_T* path)
1.3 cvs 658: #else /* __STDC__ */
1.67 cvs 659: ThotBool IsW3Path (path)
1.84 cvs 660: const CHAR_T* path;
1.3 cvs 661: #endif /* __STDC__ */
662: {
1.84 cvs 663: if (ustrncmp (path, TEXT("http:"), 5) &&
664: ustrncmp (path, TEXT("ftp:"), 4) &&
665: ustrncmp (path, TEXT("telnet:"), 7) &&
666: ustrncmp (path, TEXT("wais:"), 5) &&
667: ustrncmp (path, TEXT("news:"), 5) &&
668: ustrncmp (path, TEXT("gopher:"), 7) &&
669: ustrncmp (path, TEXT("mailto:"), 7) &&
670: ustrncmp (path, TEXT("archie:"), 7))
1.72 cvs 671: return FALSE;
672: return TRUE;
1.3 cvs 673: }
674:
1.4 cvs 675: /*----------------------------------------------------------------------
1.90 cvs 676: IsFilePath
677: returns TRUE if path is in fact a URL.
678: ----------------------------------------------------------------------*/
679: #ifdef __STDC__
1.98 kahan 680: ThotBool IsFilePath (const CHAR_T* path)
1.90 cvs 681: #else /* __STDC__ */
1.98 kahan 682: ThotBool IsFilePath (path)
1.90 cvs 683: const CHAR_T* path;
684: #endif /* __STDC__ */
685: {
686: if (ustrncmp (path, TEXT("file:"), 5))
687: return FALSE;
688: return TRUE;
689: }
690:
691: /*----------------------------------------------------------------------
1.9 cvs 692: IsValidProtocol
693: returns true if the url protocol is supported by Amaya.
1.4 cvs 694: ----------------------------------------------------------------------*/
1.3 cvs 695: #ifdef __STDC__
1.67 cvs 696: ThotBool IsValidProtocol (const STRING url)
1.3 cvs 697: #else /* __STDC__ */
1.67 cvs 698: ThotBool IsValidProtocol (url)
699: const STRING url;
1.3 cvs 700: #endif /* __STDC__ */
701: {
1.67 cvs 702: if (!ustrncmp (url, TEXT("http:"), 5)
1.69 cvs 703: || !ustrncmp (url, TEXT("internal:"), 9)
1.100 kahan 704: || (AHTFTPURL_flag () && !ustrncmp (url, TEXT("ftp:"), 4)))
1.22 cvs 705: /* experimental */
1.24 cvs 706: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 707: return (TRUE);
1.5 cvs 708: else
1.8 cvs 709: return (FALSE);
1.3 cvs 710: }
711:
1.31 cvs 712:
713: /*----------------------------------------------------------------------
714: GetBaseURL
715: normalizes orgName according to a base associated with doc, and
716: following the standard URL format rules.
717: The function returns the base used to solve relative URL and SRC:
718: - the base of the document,
719: - or the document path (without document name).
720: ----------------------------------------------------------------------*/
721: #ifdef __STDC__
1.84 cvs 722: CHAR_T* GetBaseURL (Document doc)
1.31 cvs 723: #else /* __STDC__ */
1.84 cvs 724: CHAR_T* GetBaseURL (doc)
1.31 cvs 725: Document doc;
726: #endif /* __STDC__ */
727: {
728: Element el;
729: ElementType elType;
730: AttributeType attrType;
731: Attribute attr;
1.84 cvs 732: CHAR_T *ptr, *basename;
1.31 cvs 733: int length;
734:
1.57 cvs 735: /* @@@ irene */
736: if (!DocumentURLs[doc])
737: return NULL;
1.67 cvs 738: basename = TtaAllocString (MAX_LENGTH);
739: ustrncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1.84 cvs 740: basename[MAX_LENGTH-1] = WC_EOS;
1.31 cvs 741: length = MAX_LENGTH -1;
742: /* get the root element */
743: el = TtaGetMainRoot (doc);
744: /* search the BASE element */
745: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.65 cvs 746: elType.ElTypeNum = HTML_EL_HEAD;
747: el = TtaSearchTypedElement (elType, SearchForward, el);
748: if (el)
749: {
750: elType.ElTypeNum = HTML_EL_BASE;
751: el = TtaSearchTypedElement (elType, SearchInTree, el);
752: }
1.31 cvs 753: if (el)
754: {
755: /* The document has a BASE element -> Get the HREF attribute */
756: attrType.AttrSSchema = elType.ElSSchema;
757: attrType.AttrTypeNum = HTML_ATTR_HREF_;
758: attr = TtaGetAttribute (el, attrType);
759: if (attr)
760: {
761: /* Use the base path of the document */
762: TtaGiveTextAttributeValue (attr, basename, &length);
763: /* base and orgName have to be separated by a DIR_SEP */
764: length--;
1.84 cvs 765: if (basename[0] != WC_EOS && basename[length] != WC_URL_SEP && basename[length] != WC_DIR_SEP)
1.31 cvs 766: /* verify if the base has the form "protocol://server:port" */
767: {
1.84 cvs 768: ptr = AmayaParseUrl (basename, TEXT(""), AMAYA_PARSE_ACCESS |
1.33 cvs 769: AMAYA_PARSE_HOST |
770: AMAYA_PARSE_PUNCTUATION);
1.67 cvs 771: if (ptr && !ustrcmp (ptr, basename))
1.31 cvs 772: {
1.43 cvs 773: /* it has this form, we complete it by adding a URL_STR */
1.84 cvs 774: if (ustrchr (basename, WC_DIR_SEP))
775: ustrcat (basename, WC_DIR_STR);
1.43 cvs 776: else
1.84 cvs 777: ustrcat (basename, WC_URL_STR);
1.31 cvs 778: length++;
779: }
780: if (ptr)
781: TtaFreeMemory (ptr);
782: }
783: }
1.33 cvs 784: }
785:
1.31 cvs 786: /* Remove anything after the last DIR_SEP char. If no such char is found,
787: * then search for the first ":" char, hoping that what's before that is a
788: * protocol. If found, end the string there. If neither char is found,
789: * then discard the whole base element.
790: */
1.67 cvs 791: length = ustrlen (basename) - 1;
1.31 cvs 792: /* search for the last DIR_SEP char */
1.84 cvs 793: while (length >= 0 && basename[length] != WC_URL_SEP && basename[length] != WC_DIR_SEP)
1.31 cvs 794: length--;
795: if (length >= 0)
796: /* found the last DIR_SEP char, end the string there */
1.84 cvs 797: basename[length + 1] = WC_EOS;
1.31 cvs 798: else
799: /* search for the first PATH_STR char */
800: {
1.67 cvs 801: for (length = 0; basename[length] != TEXT(':') &&
1.84 cvs 802: basename[length] != WC_EOS; length ++);
1.67 cvs 803: if (basename[length] == TEXT(':'))
1.31 cvs 804: /* found, so end the string there */
1.84 cvs 805: basename[length + 1] = WC_EOS;
1.31 cvs 806: else
807: /* not found, discard the base */
1.84 cvs 808: basename[0] = WC_EOS;
1.31 cvs 809: }
810: return (basename);
811: }
812:
813:
1.4 cvs 814: /*----------------------------------------------------------------------
1.40 cvs 815: GetLocalPath
816: Allocate and return the local document path associated to the url
817: ----------------------------------------------------------------------*/
818: #ifdef __STDC__
1.84 cvs 819: CHAR_T* GetLocalPath (Document doc, CHAR_T* url)
1.40 cvs 820: #else /* __STDC__ */
1.84 cvs 821: CHAR_T* GetLocalPath (doc, url)
1.40 cvs 822: Document doc;
1.84 cvs 823: CHAR_T* url;
1.40 cvs 824: #endif /* __STDC__ */
825: {
1.84 cvs 826: CHAR_T* ptr;
827: CHAR_T* n;
828: CHAR_T* documentname;
829: CHAR_T url_sep;
1.83 cvs 830: int len;
1.67 cvs 831: ThotBool noFile;
1.40 cvs 832:
833: if (url != NULL)
834: {
835: /* check whether the file name exists */
1.84 cvs 836: len = ustrlen (url) - 1;
1.71 cvs 837: if (IsW3Path (url))
1.84 cvs 838: url_sep = TEXT('/');
1.41 cvs 839: else
1.84 cvs 840: url_sep = WC_DIR_SEP;
1.41 cvs 841: noFile = (url[len] == url_sep);
1.40 cvs 842: if (noFile)
1.84 cvs 843: url[len] = WC_EOS;
844: ptr = TtaAllocString (MAX_LENGTH);
845: documentname = TtaAllocString (MAX_LENGTH);
1.78 cvs 846: TtaExtractName (url, ptr, documentname);
1.84 cvs 847: usprintf (ptr, TEXT("%s%s%d%s"), TempFileDirectory, WC_DIR_STR, doc, WC_DIR_STR);
1.40 cvs 848: if (!TtaCheckDirectory (ptr))
849: /* directory did not exist */
1.72 cvs 850: TtaMakeDirectory (ptr);
1.47 cvs 851:
852: /* don't include the query string within document name */
1.84 cvs 853: n = ustrrchr (documentname, TEXT('?'));
1.47 cvs 854: if (n != NULL)
1.84 cvs 855: *n = WC_EOS;
1.46 cvs 856: /* don't include ':' within document name */
1.84 cvs 857: n = ustrchr (documentname, TEXT(':'));
1.46 cvs 858: if (n != NULL)
1.84 cvs 859: *n = WC_EOS;
1.69 cvs 860: /* if after all this operations document name
861: is empty, let's use noname.html instead */
1.84 cvs 862: if (documentname[0] == WC_EOS)
863: ustrcat (ptr, TEXT("noname.html"));
1.69 cvs 864: else
1.84 cvs 865: ustrcat (ptr, documentname);
1.40 cvs 866: TtaFreeMemory (documentname);
867: /* restore the url */
868: if (noFile)
1.41 cvs 869: url[len] = url_sep;
1.40 cvs 870: return (ptr);
871: }
872: else
873: return (NULL);
874: }
875:
1.73 cvs 876: /*----------------------------------------------------------------------
1.79 cvs 877: ExtractTarget extract the target name from document nane.
878: ----------------------------------------------------------------------*/
879: #ifdef __STDC__
1.84 cvs 880: void ExtractTarget (CHAR_T* aName, CHAR_T* target)
1.79 cvs 881: #else
882: void ExtractTarget (aName, target)
1.84 cvs 883: CHAR_T* aName;
884: CHAR_T* target;
1.79 cvs 885: #endif
886: {
1.82 cvs 887: int lg, i;
1.84 cvs 888: CHAR_T* ptr;
889: CHAR_T* oldptr;
1.79 cvs 890:
891: if (!target || !aName)
892: /* bad target */
893: return;
894:
1.84 cvs 895: target[0] = WC_EOS;
896: lg = ustrlen (aName);
1.79 cvs 897: if (lg)
898: {
899: /* the name is not empty */
900: oldptr = ptr = &aName[0];
901: do
902: {
1.84 cvs 903: ptr = ustrrchr (oldptr, TEXT('#'));
1.79 cvs 904: if (ptr)
905: oldptr = &ptr[1];
906: }
907: while (ptr);
908:
909: i = (int) (oldptr) - (int) (aName); /* name length */
910: if (i > 1)
911: {
1.84 cvs 912: aName[i - 1] = WC_EOS;
1.79 cvs 913: if (i != lg)
1.84 cvs 914: ustrcpy (target, oldptr);
1.79 cvs 915: }
916: }
917: }
918:
919: /*----------------------------------------------------------------------
1.90 cvs 920: RemoveNewLines (text)
921: Removes any '\n' chars that are found in text.
922: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 923: ----------------------------------------------------------------------*/
924: #ifdef __STDC__
1.90 cvs 925: ThotBool RemoveNewLines (CHAR_T *text)
1.73 cvs 926: #else
1.90 cvs 927: ThotBool RemoveNewLines (text)
928: CHAR_T *text;
929:
1.73 cvs 930: #endif /* __STDC__ */
931: {
1.90 cvs 932: ThotBool change = FALSE;
933: CHAR_T *src;
934: CHAR_T *dest;
935:
936: src = text;
937: dest = text;
938:
939: while (*src)
940: {
941: switch (*src)
942: {
943: case TEXT('\n'):
944: /* don't copy the newline */
945: change = 1;
946: break;
947: default:
948: *dest = *src;
949: dest++;
950: break;
951: }
952: src++;
953: }
954: /* copy the last EOS char */
955: *dest = *src;
956:
957: return (change);
958: }
959:
960: /*----------------------------------------------------------------------
961: CleanCopyFileURL
962: Copies a file url from a src string to destination string.
1.97 cvs 963: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
964: we want to do).
1.90 cvs 965: ----------------------------------------------------------------------*/
966: #ifdef __STDC__
1.97 cvs 967: static void CleanCopyFileURL (CHAR_T *dest, CHAR_T *src, ConvertionType convertion)
1.90 cvs 968: #else
1.97 cvs 969: static void CleanCopyFileURL (dest, src, convertion)
1.90 cvs 970: CHAR_T* dest;
971: CHAR_T* src;
1.97 cvs 972: ConvertionType convertion;
1.89 cvs 973:
1.90 cvs 974: #endif /* __STDC__ */
975: {
976: while (*src)
1.89 cvs 977: {
1.90 cvs 978: switch (*src)
1.89 cvs 979: {
980: #ifdef _WINDOWS
1.90 cvs 981: case WC_URL_SEP:
1.96 cvs 982: /* make DIR_SEP transformation */
1.97 cvs 983: if (convertion & AM_CONV_URL_SEP)
1.96 cvs 984: *dest = WC_DIR_SEP;
985: else
986: *dest = *src;
1.90 cvs 987: dest++;
1.96 cvs 988: src++;
1.90 cvs 989: break;
1.89 cvs 990: #endif /* _WINDOWS */
1.96 cvs 991:
992: case TEXT('%'):
1.97 cvs 993: if (convertion & AM_CONV_PERCENT)
1.96 cvs 994: {
1.97 cvs 995: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 996: src++;
1.97 cvs 997: if (*src != WC_EOS)
998: {
999: *dest = UnEscapeChar (*src) * 16;
1000: src++;
1001: }
1002: if (*src != WC_EOS)
1003: {
1004: *dest = *dest + UnEscapeChar (*src);
1005: src++;
1006: }
1007: dest++;
1.96 cvs 1008: }
1.97 cvs 1009: else
1.96 cvs 1010: {
1.97 cvs 1011: *dest = *src;
1012: dest++;
1.96 cvs 1013: src++;
1014: }
1015: break;
1016:
1.90 cvs 1017: default:
1018: *dest = *src;
1.89 cvs 1019: dest++;
1.96 cvs 1020: src++;
1.90 cvs 1021: break;
1.89 cvs 1022: }
1023: }
1.90 cvs 1024: /* copy the EOS char */
1025: *dest = *src;
1.73 cvs 1026: }
1.40 cvs 1027:
1028: /*----------------------------------------------------------------------
1.9 cvs 1029: NormalizeURL
1030: normalizes orgName according to a base associated with doc, and
1031: following the standard URL format rules.
1.53 cvs 1032: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1033: other path.
1.9 cvs 1034: The function returns the new complete and normalized URL
1.12 cvs 1035: or file name path (newName) and the name of the document (docName).
1.9 cvs 1036: N.B. If the function can't find out what's the docName, it assigns
1037: the name "noname.html".
1.4 cvs 1038: ----------------------------------------------------------------------*/
1.3 cvs 1039: #ifdef __STDC__
1.84 cvs 1040: void NormalizeURL (CHAR_T* orgName, Document doc, CHAR_T* newName, CHAR_T* docName, CHAR_T* otherPath)
1.3 cvs 1041: #else /* __STDC__ */
1.53 cvs 1042: void NormalizeURL (orgName, doc, newName, docName, otherPath)
1.84 cvs 1043: CHAR_T* orgName;
1.3 cvs 1044: Document doc;
1.84 cvs 1045: CHAR_T* newName;
1046: CHAR_T* docName;
1047: CHAR_T* otherPath;
1.3 cvs 1048: #endif /* __STDC__ */
1049: {
1.84 cvs 1050: CHAR_T* basename;
1051: CHAR_T tempOrgName[MAX_LENGTH];
1052: CHAR_T* ptr;
1053: CHAR_T used_sep;
1054: int length;
1055: ThotBool check;
1.5 cvs 1056:
1.44 cvs 1057: # ifdef _WINDOWS
1058: int ndx;
1059: # endif /* _WINDOWS */
1060:
1.5 cvs 1061: if (!newName || !docName)
1062: return;
1.18 cvs 1063:
1.32 cvs 1064: if (doc != 0)
1.53 cvs 1065: basename = GetBaseURL (doc);
1066: else if (otherPath != NULL)
1.84 cvs 1067: basename = TtaWCSdup (otherPath);
1.32 cvs 1068: else
1.53 cvs 1069: basename = NULL;
1.32 cvs 1070:
1.18 cvs 1071: /*
1.31 cvs 1072: * Clean orgName
1073: * Make sure we have a complete orgName, without any leading or trailing
1074: * white spaces, or trailinbg new lines
1075: */
1.5 cvs 1076: ptr = orgName;
1.18 cvs 1077: /* skip leading white space and new line characters */
1.84 cvs 1078: while ((*ptr == WC_SPACE || *ptr == WC_EOL) && *ptr++ != WC_EOS);
1079: ustrncpy (tempOrgName, ptr, MAX_LENGTH -1);
1080: tempOrgName[MAX_LENGTH -1] = WC_EOS;
1.18 cvs 1081: /*
1.31 cvs 1082: * Make orgName a complete URL
1083: * If the URL does not include a protocol, then try to calculate
1084: * one using the doc's base element (if it exists),
1085: */
1.84 cvs 1086: if (tempOrgName[0] == WC_EOS)
1.53 cvs 1087: {
1.84 cvs 1088: newName[0] = WC_EOS;
1.53 cvs 1089: TtaFreeMemory (basename);
1090: return;
1091: }
1.49 cvs 1092:
1093: /* clean trailing white space */
1.84 cvs 1094: length = ustrlen (tempOrgName) - 1;
1095: while (tempOrgName[length] == WC_SPACE && tempOrgName[length] == WC_EOL)
1.53 cvs 1096: {
1.84 cvs 1097: tempOrgName[length] = WC_EOS;
1.53 cvs 1098: length--;
1099: }
1.50 cvs 1100:
1.55 cvs 1101: /* remove extra dot (which dot???) */
1102: /* ugly, but faster than a strcmp */
1.84 cvs 1103: if (tempOrgName[length] == TEXT('.')
1104: && (length == 0 || tempOrgName[length-1] != TEXT('.')))
1105: tempOrgName[length] = WC_EOS;
1.50 cvs 1106:
1.94 cvs 1107: if (IsW3Path (tempOrgName))
1.53 cvs 1108: {
1109: /* the name is complete, go to the Sixth Step */
1.84 cvs 1110: ustrcpy (newName, tempOrgName);
1.53 cvs 1111: SimplifyUrl (&newName);
1112: /* verify if the URL has the form "protocol://server:port" */
1.84 cvs 1113: ptr = AmayaParseUrl (newName, TEXT(""), AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1114: if (ptr && !ustrcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
1115: ustrcat (newName, WC_URL_STR);
1.49 cvs 1116:
1.53 cvs 1117: if (ptr)
1.50 cvs 1118: TtaFreeMemory (ptr);
1.53 cvs 1119: }
1120: else if ( basename == NULL)
1121: /* the name is complete, go to the Sixth Step */
1.84 cvs 1122: ustrcpy (newName, tempOrgName);
1.53 cvs 1123: else
1124: {
1.31 cvs 1125: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 1126: # ifdef _WINDOWS
1.53 cvs 1127: if (!IsW3Path (basename))
1128: {
1.67 cvs 1129: length = ustrlen (tempOrgName);
1.53 cvs 1130: for (ndx = 0; ndx < length; ndx++)
1.67 cvs 1131: if (tempOrgName [ndx] == TEXT('/'))
1132: tempOrgName [ndx] = TEXT('\\');
1.53 cvs 1133: }
1.44 cvs 1134: # endif /* _WINDOWS */
1.25 cvs 1135: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1136: if (ptr)
1137: {
1138: SimplifyUrl (&ptr);
1.84 cvs 1139: ustrcpy (newName, ptr);
1.53 cvs 1140: TtaFreeMemory (ptr);
1141: }
1142: else
1.84 cvs 1143: newName[0] = WC_EOS;
1.53 cvs 1144: }
1.36 cvs 1145:
1146: TtaFreeMemory (basename);
1.18 cvs 1147: /*
1.31 cvs 1148: * Prepare the docname that will refer to this ressource in the
1149: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1150: * noname.html as a default ressource name
1.18 cvs 1151: */
1.84 cvs 1152: if (newName[0] != WC_EOS)
1.53 cvs 1153: {
1.84 cvs 1154: length = ustrlen (newName) - 1;
1155: if (newName[length] == WC_URL_SEP || newName[length] == WC_DIR_SEP)
1.53 cvs 1156: {
1157: used_sep = newName[length];
1158: check = TRUE;
1159: while (check)
1160: {
1.50 cvs 1161: length--;
1162: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1163: length--;
1.84 cvs 1164: if (!ustrncmp (&newName[length+1], TEXT(".."), 2))
1.53 cvs 1165: {
1.84 cvs 1166: newName[length+1] = WC_EOS;
1.53 cvs 1167: /* remove also previous directory */
1168: length--;
1169: while (length >= 0 && newName[length] != used_sep)
1170: length--;
1.84 cvs 1171: if (ustrncmp (&newName[length+1], TEXT("//"), 2))
1.53 cvs 1172: /* don't remove server name */
1.84 cvs 1173: newName[length+1] = WC_EOS;
1.53 cvs 1174: }
1.84 cvs 1175: else if (!ustrncmp (&newName[length+1], TEXT("."), 1))
1176: newName[length+1] = WC_EOS;
1.50 cvs 1177: else
1.53 cvs 1178: check = FALSE;
1179: }
1.84 cvs 1180: ustrcpy (docName, TEXT("noname.html"));
1.53 cvs 1181: /* docname was not comprised inside the URL, so let's */
1182: /* assign the default ressource name */
1.84 cvs 1183: ustrcpy (docName, TEXT("noname.html"));
1.53 cvs 1184: }
1185: else
1186: { /* docname is comprised inside the URL */
1.84 cvs 1187: while (length >= 0 && newName[length] != WC_URL_SEP && newName[length] != WC_DIR_SEP)
1.53 cvs 1188: length--;
1189: if (length < 0)
1.84 cvs 1190: ustrcpy (docName, newName);
1.53 cvs 1191: else
1.84 cvs 1192: ustrcpy (docName, &newName[length+1]);
1.53 cvs 1193: }
1194: }
1195: else
1.84 cvs 1196: docName[0] = WC_EOS;
1.18 cvs 1197: }
1.3 cvs 1198:
1.4 cvs 1199: /*----------------------------------------------------------------------
1.9 cvs 1200: IsSameHost
1.4 cvs 1201: ----------------------------------------------------------------------*/
1.3 cvs 1202: #ifdef __STDC__
1.67 cvs 1203: ThotBool IsSameHost (const STRING url1, const STRING url2)
1.3 cvs 1204: #else /* __STDC__ */
1.67 cvs 1205: ThotBool IsSameHost (url1, url2)
1206: const STRING url1;
1207: const STRING url2;
1.3 cvs 1208: #endif /* __STDC__ */
1209: {
1.67 cvs 1210: STRING basename_ptr1, basename_ptr2;
1211: ThotBool result;
1.3 cvs 1212:
1.84 cvs 1213: basename_ptr1 = AmayaParseUrl (url1, TEXT(""), AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.86 cvs 1214: basename_ptr2 = AmayaParseUrl (url2, TEXT(""), AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1215:
1.67 cvs 1216: if (ustrcmp (basename_ptr1, basename_ptr2))
1.8 cvs 1217: result = FALSE;
1.5 cvs 1218: else
1.8 cvs 1219: result = TRUE;
1.3 cvs 1220:
1.25 cvs 1221: TtaFreeMemory (basename_ptr1);
1222: TtaFreeMemory (basename_ptr2);
1.5 cvs 1223: return (result);
1.3 cvs 1224: }
1225:
1226:
1.4 cvs 1227: /*----------------------------------------------------------------------
1.22 cvs 1228: HasKnownFileSuffix
1229: returns TRUE if path points to a file ending with a suffix.
1230: ----------------------------------------------------------------------*/
1231: #ifdef __STDC__
1.67 cvs 1232: ThotBool HasKnownFileSuffix (const STRING path)
1.22 cvs 1233: #else /* __STDC__ */
1.67 cvs 1234: ThotBool HasKnownFileSuffix (path)
1235: const STRING path;
1.22 cvs 1236: #endif /* __STDC__ */
1237: {
1.67 cvs 1238: STRING root;
1239: CHAR_T temppath[MAX_LENGTH];
1240: CHAR_T suffix[MAX_LENGTH];
1.22 cvs 1241:
1.87 cvs 1242: if (!path || path[0] == WC_EOS || path[ustrlen(path)] == WC_DIR_SEP)
1.22 cvs 1243: return (FALSE);
1244:
1.84 cvs 1245: root = AmayaParseUrl(path, TEXT(""), AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1246:
1247: if (root)
1248: {
1.67 cvs 1249: ustrcpy (temppath, root);
1.25 cvs 1250: TtaFreeMemory (root);
1.22 cvs 1251: /* Get the suffix */
1252: ExtractSuffix (temppath, suffix);
1253:
1.87 cvs 1254: if( suffix[0] == WC_EOS)
1.22 cvs 1255: /* no suffix */
1256: return (FALSE);
1257:
1258: /* Normalize the suffix */
1259: ConvertToLowerCase (suffix);
1260:
1.67 cvs 1261: if (!ustrcmp (suffix, TEXT("gz")))
1.22 cvs 1262: /* skip the compressed suffix */
1263: {
1264: ExtractSuffix (temppath, suffix);
1.87 cvs 1265: if(suffix[0] == WC_EOS)
1.22 cvs 1266: /* no suffix */
1267: return (FALSE);
1268: /* Normalize the suffix */
1269: ConvertToLowerCase (suffix);
1270: }
1271:
1.67 cvs 1272: if (ustrcmp (suffix, TEXT("gif")) &&
1273: ustrcmp (suffix, TEXT("xbm")) &&
1274: ustrcmp (suffix, TEXT("xpm")) &&
1275: ustrcmp (suffix, TEXT("jpg")) &&
1276: ustrcmp (suffix, TEXT("pdf")) &&
1277: ustrcmp (suffix, TEXT("png")) &&
1278: ustrcmp (suffix, TEXT("tgz")) &&
1279: ustrcmp (suffix, TEXT("xpg")) &&
1280: ustrcmp (suffix, TEXT("xpd")) &&
1281: ustrcmp (suffix, TEXT("ps")) &&
1282: ustrcmp (suffix, TEXT("au")) &&
1283: ustrcmp (suffix, TEXT("html")) &&
1284: ustrcmp (suffix, TEXT("htm")) &&
1285: ustrcmp (suffix, TEXT("shtml")) &&
1286: ustrcmp (suffix, TEXT("xht")) &&
1287: ustrcmp (suffix, TEXT("xhtm")) &&
1288: ustrcmp (suffix, TEXT("xhtml")) &&
1289: ustrcmp (suffix, TEXT("txt")) &&
1290: ustrcmp (suffix, TEXT("css")) &&
1291: ustrcmp (suffix, TEXT("eps")))
1.22 cvs 1292: return (FALSE);
1293: else
1294: return (TRUE);
1295: }
1296: else
1297: return (FALSE);
1298: }
1299:
1300:
1301: /*----------------------------------------------------------------------
1.24 cvs 1302: ChopURL
1303: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1304: If inputURL is bigger than that size, outputURL receives
1305: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1306: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1307: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1308: copied into outputURL.
1309: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1310: chars.
1311: ----------------------------------------------------------------------*/
1312: #ifdef __STDC__
1.86 cvs 1313: void ChopURL (CHAR_T* outputURL, const CHAR_T* inputURL)
1.24 cvs 1314: #else
1315: void ChopURL (outputURL, inputURL)
1.86 cvs 1316: CHAR_T* outputURL;
1317: const CHAR_T* inputURL;
1.24 cvs 1318: #endif
1.22 cvs 1319:
1.24 cvs 1320: {
1321: int len;
1.9 cvs 1322:
1.86 cvs 1323: len = ustrlen (inputURL);
1.24 cvs 1324: if (len <= MAX_PRINT_URL_LENGTH)
1.86 cvs 1325: ustrcpy (outputURL, inputURL);
1.24 cvs 1326: else
1327: /* make a truncated urlName on the status window */
1328: {
1.86 cvs 1329: ustrncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1330: outputURL [MAX_PRINT_URL_LENGTH / 2] = WC_EOS;
1331: ustrcat (outputURL, TEXT("..."));
1332: ustrcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1333: }
1.25 cvs 1334: }
1335:
1336:
1337: /*----------------------------------------------------------------------
1338: scan
1.47 cvs 1339: Scan a filename for its constituents
1.25 cvs 1340: -----------------------------------
1341:
1342: On entry,
1343: name points to a document name which may be incomplete.
1344: On exit,
1345: absolute or relative may be nonzero (but not both).
1346: host, fragment and access may be nonzero if they were specified.
1347: Any which are nonzero point to zero terminated strings.
1348: ----------------------------------------------------------------------*/
1349: #ifdef __STDC__
1.67 cvs 1350: static void scan (STRING name, HTURI * parts)
1.25 cvs 1351: #else /* __STDC__ */
1352: static void scan (name, parts)
1.67 cvs 1353: STRING name;
1354: HTURI *parts;
1.25 cvs 1355:
1356: #endif /* __STDC__ */
1357: {
1.67 cvs 1358: STRING p;
1359: STRING after_access = name;
1.32 cvs 1360:
1.43 cvs 1361: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1362: /* Look for fragment identifier */
1.67 cvs 1363: if ((p = ustrchr(name, TEXT('#'))) != NULL)
1.28 cvs 1364: {
1.67 cvs 1365: *p++ = TEXT('\0');
1.28 cvs 1366: parts->fragment = p;
1.25 cvs 1367: }
1368:
1.28 cvs 1369: for (p=name; *p; p++)
1370: {
1.67 cvs 1371: if (*p == URL_SEP || *p == DIR_SEP || *p == TEXT('#') || *p == TEXT('?'))
1.28 cvs 1372: break;
1.67 cvs 1373: if (*p == TEXT(':'))
1.28 cvs 1374: {
1375: *p = 0;
1376: parts->access = after_access; /* Scheme has been specified */
1377:
1378: /* The combination of gcc, the "-O" flag and the HP platform is
1379: unhealthy. The following three lines is a quick & dirty fix, but is
1380: not recommended. Rather, turn off "-O". */
1381:
1382: /* after_access = p;*/
1383: /* while (*after_access == 0)*/
1384: /* after_access++;*/
1385: after_access = p+1;
1.67 cvs 1386: if (!ustrcasecmp(TEXT("URL"), parts->access))
1.28 cvs 1387: /* Ignore IETF's URL: pre-prefix */
1388: parts->access = NULL;
1389: else
1.25 cvs 1390: break;
1391: }
1392: }
1393:
1394: p = after_access;
1.43 cvs 1395: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1396: {
1.43 cvs 1397: if (p[1] == URL_SEP)
1.28 cvs 1398: {
1.25 cvs 1399: parts->host = p+2; /* host has been specified */
1.28 cvs 1400: *p = 0; /* Terminate access */
1401: /* look for end of host name if any */
1.67 cvs 1402: p = ustrchr (parts->host, URL_SEP);
1.28 cvs 1403: if (p)
1404: {
1.87 cvs 1405: *p = WC_EOS; /* Terminate host */
1.25 cvs 1406: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1407: }
1408: }
1409: else
1410: /* Root found but no host */
1411: parts->absolute = p+1;
1412: }
1413: else
1414: {
1.25 cvs 1415: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1416: }
1.25 cvs 1417: }
1418:
1419:
1420: /*----------------------------------------------------------------------
1.28 cvs 1421: AmayaParseUrl: parse a Name relative to another name
1422:
1423: This returns those parts of a name which are given (and requested)
1424: substituting bits from the related name where necessary.
1.25 cvs 1425:
1.28 cvs 1426: On entry,
1.25 cvs 1427: aName A filename given
1428: relatedName A name relative to which aName is to be parsed. Give
1429: it an empty string if aName is absolute.
1430: wanted A mask for the bits which are wanted.
1431:
1.28 cvs 1432: On exit,
1.25 cvs 1433: returns A pointer to a malloc'd string which MUST BE FREED
1434: ----------------------------------------------------------------------*/
1435: #ifdef __STDC__
1.84 cvs 1436: CHAR_T* AmayaParseUrl (const CHAR_T* aName, CHAR_T* relatedName, int wanted)
1.25 cvs 1437: #else /* __STDC__ */
1.84 cvs 1438: CHAR_T* AmayaParseUrl (aName, relatedName, wanted)
1439: const CHAR_T* aName;
1440: CHAR_T* relatedName;
1.28 cvs 1441: int wanted;
1.25 cvs 1442:
1443: #endif /* __STDC__ */
1444: {
1.84 cvs 1445: CHAR_T* return_value;
1.67 cvs 1446: CHAR_T result[MAX_LENGTH];
1447: CHAR_T name[MAX_LENGTH];
1448: CHAR_T rel[MAX_LENGTH];
1.84 cvs 1449: CHAR_T *p, *access;
1.29 cvs 1450: HTURI given, related;
1451: int len;
1.67 cvs 1452: CHAR_T used_sep;
1.84 cvs 1453: CHAR_T* used_str;
1.32 cvs 1454:
1.84 cvs 1455: if (ustrchr (aName, WC_DIR_SEP) || ustrchr (relatedName, WC_DIR_SEP))
1.33 cvs 1456: {
1.84 cvs 1457: used_str = WC_DIR_STR;
1458: used_sep = WC_DIR_SEP;
1.33 cvs 1459: }
1.32 cvs 1460: else
1.33 cvs 1461: {
1.84 cvs 1462: used_str = WC_URL_STR;
1463: used_sep = WC_URL_SEP;
1.33 cvs 1464: }
1.32 cvs 1465:
1.29 cvs 1466: /* Make working copies of input strings to cut up: */
1467: return_value = NULL;
1468: result[0] = 0; /* Clear string */
1.67 cvs 1469: ustrcpy (name, aName);
1.29 cvs 1470: if (relatedName != NULL)
1.67 cvs 1471: ustrcpy (rel, relatedName);
1.29 cvs 1472: else
1.84 cvs 1473: relatedName[0] = WC_EOS;
1.29 cvs 1474:
1475: scan (name, &given);
1476: scan (rel, &related);
1477: access = given.access ? given.access : related.access;
1478: if (wanted & AMAYA_PARSE_ACCESS)
1479: if (access)
1480: {
1.67 cvs 1481: ustrcat (result, access);
1.29 cvs 1482: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1483: ustrcat (result, TEXT(":"));
1.29 cvs 1484: }
1485:
1486: if (given.access && related.access)
1487: /* If different, inherit nothing. */
1.67 cvs 1488: if (ustrcmp (given.access, related.access) != 0)
1.29 cvs 1489: {
1490: related.host = 0;
1491: related.absolute = 0;
1492: related.relative = 0;
1493: related.fragment = 0;
1494: }
1495:
1496: if (wanted & AMAYA_PARSE_HOST)
1497: if(given.host || related.host)
1498: {
1499: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1500: ustrcat (result, TEXT("//"));
1501: ustrcat (result, given.host ? given.host : related.host);
1.29 cvs 1502: }
1503:
1504: if (given.host && related.host)
1505: /* If different hosts, inherit no path. */
1.67 cvs 1506: if (ustrcmp (given.host, related.host) != 0)
1.29 cvs 1507: {
1508: related.absolute = 0;
1509: related.relative = 0;
1510: related.fragment = 0;
1511: }
1512:
1513: if (wanted & AMAYA_PARSE_PATH)
1514: {
1515: if (given.absolute)
1516: {
1517: /* All is given */
1518: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1519: ustrcat (result, used_str);
1520: ustrcat (result, given.absolute);
1.25 cvs 1521: }
1.29 cvs 1522: else if (related.absolute)
1523: {
1524: /* Adopt path not name */
1.67 cvs 1525: ustrcat (result, used_str);
1526: ustrcat (result, related.absolute);
1.29 cvs 1527: if (given.relative)
1528: {
1529: /* Search part? */
1.67 cvs 1530: p = ustrchr (result, TEXT('?'));
1.29 cvs 1531: if (!p)
1.67 cvs 1532: p=result+ustrlen(result)-1;
1.33 cvs 1533: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1534: /* Remove filename */
1535: p[1]=0;
1536: /* Add given one */
1.67 cvs 1537: ustrcat (result, given.relative);
1.25 cvs 1538: }
1539: }
1.29 cvs 1540: else if (given.relative)
1541: /* what we've got */
1.67 cvs 1542: ustrcat (result, given.relative);
1.29 cvs 1543: else if (related.relative)
1.67 cvs 1544: ustrcat (result, related.relative);
1.29 cvs 1545: else
1546: /* No inheritance */
1.67 cvs 1547: ustrcat (result, used_str);
1.25 cvs 1548: }
1.29 cvs 1549:
1550: if (wanted & AMAYA_PARSE_ANCHOR)
1551: if (given.fragment || related.fragment)
1552: {
1553: if (given.absolute && given.fragment)
1554: {
1555: /*Fixes for relURLs...*/
1556: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1557: ustrcat (result, TEXT("#"));
1558: ustrcat (result, given.fragment);
1.29 cvs 1559: }
1560: else if (!(given.absolute) && !(given.fragment))
1.84 cvs 1561: ustrcat (result, TEXT(""));
1.29 cvs 1562: else
1563: {
1564: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.67 cvs 1565: ustrcat (result, TEXT("#"));
1566: ustrcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1567: }
1568: }
1.67 cvs 1569: len = ustrlen (result);
1570: if ((return_value = TtaAllocString (len + 1)) != NULL)
1571: ustrcpy (return_value, result);
1.29 cvs 1572: return (return_value); /* exactly the right length */
1.25 cvs 1573: }
1574:
1575: /*----------------------------------------------------------------------
1576: HTCanon
1577: Canonicalizes the URL in the following manner starting from the host
1578: pointer:
1579:
1580: 1) The host name is converted to lowercase
1581: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1582:
1583: Return: OK The position of the current path part of the URL
1584: which might be the old one or a new one.
1585:
1586: ----------------------------------------------------------------------*/
1587: #ifdef __STDC__
1.84 cvs 1588: static CHAR_T* HTCanon (CHAR_T** filename, CHAR_T* host)
1.25 cvs 1589: #else /* __STDC__ */
1.67 cvs 1590: static STRING HTCanon (filename, host)
1.84 cvs 1591: CHAR_T** filename;
1592: CHAR_T* host;
1.25 cvs 1593: #endif /* __STDC__ */
1594: {
1.84 cvs 1595: CHAR_T* newname = NULL;
1596: CHAR_T used_sep;
1597: CHAR_T* path;
1598: CHAR_T* strptr;
1599: CHAR_T* port;
1600: CHAR_T* access = host-3;
1.32 cvs 1601:
1602:
1.84 cvs 1603: if (*filename && ustrchr (*filename, WC_URL_SEP))
1.33 cvs 1604: {
1.84 cvs 1605: used_sep = WC_URL_SEP;
1.33 cvs 1606: }
1607: else
1608: {
1.84 cvs 1609: used_sep = WC_DIR_SEP;
1.33 cvs 1610: }
1.32 cvs 1611:
1.82 cvs 1612: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1613: access--;
1.84 cvs 1614: if ((path = ustrchr (host, used_sep)) == NULL) /* Find path */
1615: path = host + ustrlen (host);
1616: if ((strptr = ustrchr (host, TEXT('@'))) != NULL && strptr < path) /* UserId */
1.82 cvs 1617: host = strptr;
1.84 cvs 1618: if ((port = ustrchr (host, TEXT(':'))) != NULL && port > path) /* Port number */
1.82 cvs 1619: port = NULL;
1.25 cvs 1620:
1621: strptr = host; /* Convert to lower-case */
1.82 cvs 1622: while (strptr < path)
1.33 cvs 1623: {
1.84 cvs 1624: *strptr = utolower (*strptr);
1.82 cvs 1625: strptr++;
1.33 cvs 1626: }
1.25 cvs 1627:
1628: /* Does the URL contain a full domain name? This also works for a
1629: numerical host name. The domain name is already made lower-case
1630: and without a trailing dot. */
1631: {
1.84 cvs 1632: CHAR_T* dot = port ? port : path;
1633: if (dot > *filename && *--dot == TEXT('.'))
1.33 cvs 1634: {
1.84 cvs 1635: CHAR_T* orig = dot;
1636: CHAR_T* dest = dot + 1;
1.82 cvs 1637: while ((*orig++ = *dest++));
1638: if (port) port--;
1.33 cvs 1639: path--;
1.25 cvs 1640: }
1641: }
1642: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1643: if (port)
1644: {
1.82 cvs 1645: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1646: {
1647: if (!newname)
1648: {
1.84 cvs 1649: CHAR_T* orig = port;
1650: CHAR_T* dest = port + 1;
1.82 cvs 1651: while ((*orig++ = *dest++));
1.33 cvs 1652: }
1653: }
1.84 cvs 1654: else if ((!ustrncmp (access, TEXT("http"), 4) &&
1655: (*(port + 1) == TEXT('8') &&
1656: *(port+2) == TEXT('0') &&
1.82 cvs 1657: (*(port+3) == used_sep || !*(port + 3)))) ||
1.84 cvs 1658: (!ustrncmp (access, TEXT("gopher"), 6) &&
1659: (*(port+1) == TEXT('7') &&
1660: *(port+2) == TEXT('0') &&
1.82 cvs 1661: (*(port+3) == used_sep || !*(port+3)))) ||
1.84 cvs 1662: (!ustrncmp (access, TEXT("ftp"), 3) &&
1663: (*(port+1) == TEXT('2') &&
1664: *(port + 2) == TEXT('1') &&
1.82 cvs 1665: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1666: if (!newname)
1667: {
1.84 cvs 1668: CHAR_T* orig = port;
1669: CHAR_T* dest = port + 3;
1.33 cvs 1670: while((*orig++ = *dest++));
1671: /* Update path position, Henry Minsky */
1672: path -= 3;
1.25 cvs 1673: }
1.33 cvs 1674: }
1675: else if (newname)
1.84 cvs 1676: ustrncat (newname, port, (int) (path - port));
1.33 cvs 1677: }
1.25 cvs 1678:
1.33 cvs 1679: if (newname)
1680: {
1.84 cvs 1681: CHAR_T* newpath = newname + ustrlen (newname);
1682: ustrcat (newname, path);
1.25 cvs 1683: path = newpath;
1.28 cvs 1684: /* Free old copy */
1685: TtaFreeMemory(*filename);
1.25 cvs 1686: *filename = newname;
1.33 cvs 1687: }
1.25 cvs 1688: return path;
1689: }
1690:
1691:
1692: /*----------------------------------------------------------------------
1.29 cvs 1693: SimplifyUrl: simplify a URI
1.32 cvs 1694: A URI is allowed to contain the sequence xxx/../ which may be
1695: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1696: Simplification helps us recognize duplicate URIs.
1.25 cvs 1697:
1.28 cvs 1698: Thus, /etc/junk/../fred becomes /etc/fred
1699: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1700:
1.28 cvs 1701: but we should NOT change
1702: http://fred.xxx.edu/../..
1.25 cvs 1703:
1704: or ../../albert.html
1705:
1.28 cvs 1706: In order to avoid empty URLs the following URLs become:
1.25 cvs 1707:
1708: /fred/.. becomes /fred/..
1709: /fred/././.. becomes /fred/..
1710: /fred/.././junk/.././ becomes /fred/..
1711:
1.28 cvs 1712: If more than one set of `://' is found (several proxies in cascade) then
1713: only the part after the last `://' is simplified.
1.25 cvs 1714:
1.28 cvs 1715: Returns: A string which might be the old one or a new one.
1.25 cvs 1716: ----------------------------------------------------------------------*/
1717: #ifdef __STDC__
1.84 cvs 1718: void SimplifyUrl (CHAR_T** url)
1.25 cvs 1719: #else /* __STDC__ */
1.29 cvs 1720: void SimplifyUrl (url)
1.84 cvs 1721: CHAR_T** url;
1.25 cvs 1722: #endif /* __STDC__ */
1723: {
1.84 cvs 1724: CHAR_T* path;
1725: CHAR_T* access;
1726: CHAR_T* newptr;
1727: CHAR_T* p;
1728: CHAR_T *orig, *dest, *end;
1.28 cvs 1729:
1.85 cvs 1730: CHAR_T used_sep;
1.77 cvs 1731: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1732: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1733:
1734:
1.28 cvs 1735: if (!url || !*url)
1736: return;
1737:
1.84 cvs 1738: if (ustrchr (*url, WC_URL_SEP))
1.33 cvs 1739: {
1.84 cvs 1740: used_sep = WC_URL_SEP;
1.33 cvs 1741: }
1.32 cvs 1742: else
1.33 cvs 1743: {
1.84 cvs 1744: used_sep = WC_DIR_SEP;
1.33 cvs 1745: }
1.32 cvs 1746:
1.77 cvs 1747: /* should we simplify double dot? */
1748: path = *url;
1.84 cvs 1749: if (*path == TEXT('.') && *(path + 1) == TEXT('.'))
1.77 cvs 1750: ddot_simplify = FALSE;
1751: else
1752: ddot_simplify = TRUE;
1753:
1.28 cvs 1754: /* Find any scheme name */
1.84 cvs 1755: if ((path = ustrstr (*url, TEXT("://"))) != NULL)
1.33 cvs 1756: {
1757: /* Find host name */
1.28 cvs 1758: access = *url;
1.84 cvs 1759: while (access < path && (*access = utolower (*access)))
1.82 cvs 1760: access++;
1.28 cvs 1761: path += 3;
1.84 cvs 1762: while ((newptr = ustrstr (path, TEXT ("://"))) != NULL)
1.82 cvs 1763: /* For proxies */
1764: path = newptr+3;
1765: /* We have a host name */
1.84 cvs 1766: path = HTCanon (url, path);
1.25 cvs 1767: }
1.84 cvs 1768: else if ((path = ustrstr (*url, TEXT(":/"))) != NULL)
1.28 cvs 1769: path += 2;
1770: else
1771: path = *url;
1.25 cvs 1772:
1.84 cvs 1773: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1774: /* Some URLs start //<foo> */
1775: path += 1;
1.94 cvs 1776: else if (IsFilePath (path))
1777: {
1778: /* doesn't need to do anything more */
1779: return;
1780: }
1.84 cvs 1781: else if (!ustrncmp (path, TEXT("news:"), 5))
1.28 cvs 1782: {
1.84 cvs 1783: newptr = ustrchr (path+5, TEXT('@'));
1.28 cvs 1784: if (!newptr)
1785: newptr = path + 5;
1786: while (*newptr)
1787: {
1788: /* Make group or host lower case */
1.84 cvs 1789: *newptr = utolower (*newptr);
1.28 cvs 1790: newptr++;
1.25 cvs 1791: }
1.28 cvs 1792: /* Doesn't need to do any more */
1793: return;
1.25 cvs 1794: }
1.28 cvs 1795:
1796: if ((p = path))
1797: {
1.67 cvs 1798: if (!((end = ustrchr (path, TEXT(';'))) || (end = ustrchr (path, TEXT('?'))) ||
1799: (end = ustrchr (path, TEXT('#')))))
1800: end = path + ustrlen (path);
1.28 cvs 1801:
1802: /* Parse string second time to simplify */
1803: p = path;
1804: while (p < end)
1805: {
1.77 cvs 1806: /* if we're pointing to a char, it's safe to reactivate the ../ convertion */
1807: if (!ddot_simplify && *p != TEXT('.') && *p != used_sep)
1808: ddot_simplify = TRUE;
1809:
1.33 cvs 1810: if (*p==used_sep)
1.28 cvs 1811: {
1.67 cvs 1812: if (p > *url && *(p+1) == TEXT('.') && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1813: {
1814: orig = p + 1;
1.84 cvs 1815: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1816: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1817: end = orig - 1;
1818: }
1.77 cvs 1819: else if (ddot_simplify && *(p+1) == TEXT('.') && *(p+2) == TEXT('.')
1820: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1821: {
1822: newptr = p;
1.52 cvs 1823: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1824: if (*newptr == used_sep)
1825: orig = newptr + 1;
1.28 cvs 1826: else
1.52 cvs 1827: orig = newptr;
1828:
1829: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1830: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1831: end = orig-1;
1832: /* Start again with prev slash */
1833: p = newptr;
1.28 cvs 1834: }
1.33 cvs 1835: else if (*(p+1) == used_sep)
1.28 cvs 1836: {
1.33 cvs 1837: while (*(p+1) == used_sep)
1.28 cvs 1838: {
1839: orig = p;
1840: dest = p + 1;
1841: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1842: end = orig-1;
1843: }
1844: }
1845: else
1.25 cvs 1846: p++;
1.28 cvs 1847: }
1848: else
1849: p++;
1.25 cvs 1850: }
1851: }
1.51 cvs 1852:
1853: /*
1854: ** Check for host/../.. kind of things
1855: */
1.77 cvs 1856: if (*path == used_sep && *(path+1) == TEXT('.') && *(path+2) == TEXT('.')
1857: && (!*(path+3) || *(path+3) == used_sep))
1.87 cvs 1858: *(path+1) = WC_EOS;
1.51 cvs 1859:
1.28 cvs 1860: return;
1861: }
1862:
1863:
1864: /*----------------------------------------------------------------------
1.96 cvs 1865: NormalizeFile normalizes local names.
1.28 cvs 1866: Return TRUE if target and src differ.
1867: ----------------------------------------------------------------------*/
1868: #ifdef __STDC__
1.97 cvs 1869: ThotBool NormalizeFile (CHAR_T* src, CHAR_T* target, ConvertionType convertion)
1.28 cvs 1870: #else
1.97 cvs 1871: ThotBool NormalizeFile (src, target, convertion)
1.84 cvs 1872: CHAR_T* src;
1873: CHAR_T* target;
1.97 cvs 1874: ConvertionType convertion;
1.28 cvs 1875:
1876: #endif
1877: {
1.93 cvs 1878: # ifndef _WINDOWS
1.90 cvs 1879: CHAR_T *s;
1.93 cvs 1880: int i;
1881: # endif /* !_WINDOWS */
1.82 cvs 1882: ThotBool change;
1.90 cvs 1883: int start_index; /* the first char that we'll copy */
1.28 cvs 1884:
1.54 cvs 1885: change = FALSE;
1.90 cvs 1886: start_index = 0;
1887:
1888: if (!src || src[0] == WC_EOS)
1.96 cvs 1889: {
1890: target[0] = WC_EOS;
1891: return FALSE;
1892: }
1.90 cvs 1893:
1894: /* @@ do I need file: or file:/ here? */
1.84 cvs 1895: if (ustrncmp (src, TEXT("file:"), 5) == 0)
1.28 cvs 1896: {
1.90 cvs 1897: /* remove the prefix file: */
1898: start_index += 5;
1899:
1900: /* remove the localhost prefix */
1.94 cvs 1901: if (ustrncmp (&src[start_index], TEXT("//localhost/"), 12) == 0)
1902: start_index += 11;
1903:
1904: /* remove the first two slashes in / / /path */
1905: while (src[start_index] &&
1906: src[start_index] == TEXT('/')
1907: && src[start_index + 1] == TEXT('/'))
1908: start_index++;
1909:
1910: #ifdef _WINDOWS
1911: /* remove any extra slash before the drive name */
1912: if (src[start_index] == TEXT('/')
1913: &&src[start_index+2] == TEXT(':'))
1914: start_index++;
1915: #endif /* _WINDOWS */
1.90 cvs 1916:
1917: if (src[start_index] == WC_EOS)
1918: /* if there's nothing afterwards, add a DIR_STR */
1919: ustrcpy (target, WC_DIR_STR);
1920: else
1.97 cvs 1921: /* as we're inside a file: URL, we'll apply all the convertions
1922: we know */
1923: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1924:
1925: change = TRUE;
1926: }
1.97 cvs 1927: else if (convertion != AM_CONV_NONE)
1.96 cvs 1928: {
1929: /* we are following a "local" relative link, we do all the
1930: convertions except for the HOME_DIR ~ one */
1.97 cvs 1931: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1932: }
1.90 cvs 1933: #ifndef _WINDOWS
1.84 cvs 1934: else if (src[0] == TEXT('~'))
1.53 cvs 1935: {
1.96 cvs 1936: /* it must be a URL typed in a text input field */
1937: /* do the HOME_DIR ~ substitution */
1.82 cvs 1938: s = TtaGetEnvString ("HOME");
1.84 cvs 1939: ustrcpy (target, s);
1.90 cvs 1940: #if 0
1.96 cvs 1941: /* JK: invalidated this part of the code as it's simpler
1942: to add the DIR_SEP whenever we have something to add
1943: to the path rather than adding it systematically */
1.84 cvs 1944: if (src[1] != WC_DIR_SEP)
1945: ustrcat (target, WC_DIR_STR);
1.90 cvs 1946: #endif
1947: i = ustrlen (target);
1.92 cvs 1948: ustrcpy (&target[i], &src[1]);
1.54 cvs 1949: change = TRUE;
1.53 cvs 1950: }
1.90 cvs 1951: #endif /* _WINDOWS */
1.28 cvs 1952: else
1.96 cvs 1953: /* leave it as it is */
1.92 cvs 1954: ustrcpy (target, src);
1.96 cvs 1955:
1.28 cvs 1956: /* remove /../ and /./ */
1.29 cvs 1957: SimplifyUrl (&target);
1.54 cvs 1958: if (!change)
1.84 cvs 1959: change = ustrcmp (src, target);
1.28 cvs 1960: return (change);
1.25 cvs 1961: }
1962:
1.28 cvs 1963:
1.25 cvs 1964: /*----------------------------------------------------------------------
1.31 cvs 1965: MakeRelativeURL: make relative name
1.25 cvs 1966:
1.28 cvs 1967: This function creates and returns a string which gives an expression of
1968: one address as related to another. Where there is no relation, an absolute
1969: address is retured.
1.25 cvs 1970:
1.28 cvs 1971: On entry,
1.25 cvs 1972: Both names must be absolute, fully qualified names of nodes
1973: (no fragment bits)
1974:
1.28 cvs 1975: On exit,
1.25 cvs 1976: The return result points to a newly allocated name which, if
1977: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1978: The caller is responsible for freeing the resulting name later.
1979: ----------------------------------------------------------------------*/
1980: #ifdef __STDC__
1.67 cvs 1981: STRING MakeRelativeURL (STRING aName, STRING relatedName)
1.25 cvs 1982: #else /* __STDC__ */
1.67 cvs 1983: STRING MakeRelativeURL (aName, relatedName)
1984: STRING aName;
1985: STRING relatedName;
1.25 cvs 1986: #endif /* __STDC__ */
1987: {
1.87 cvs 1988: CHAR_T* return_value;
1.67 cvs 1989: CHAR_T result[MAX_LENGTH];
1.87 cvs 1990: CHAR_T* p;
1991: CHAR_T* q;
1992: CHAR_T* after_access;
1993: CHAR_T* last_slash = NULL;
1.29 cvs 1994: int slashes, levels, len;
1995:
1.44 cvs 1996: # ifdef _WINDOWS
1997: int ndx;
1998: # endif /* _WINDOWS */
1999:
1.29 cvs 2000: if (aName == NULL || relatedName == NULL)
2001: return (NULL);
2002:
2003: slashes = 0;
2004: after_access = NULL;
2005: p = aName;
2006: q = relatedName;
2007: for (; *p && (*p == *q); p++, q++)
1.27 cvs 2008: {
2009: /* Find extent of match */
1.67 cvs 2010: if (*p == TEXT(':'))
1.29 cvs 2011: after_access = p + 1;
1.28 cvs 2012: if (*p == DIR_SEP)
1.27 cvs 2013: {
1.29 cvs 2014: /* memorize the last slash position and count them */
1.27 cvs 2015: last_slash = p;
2016: slashes++;
1.25 cvs 2017: }
2018: }
2019:
1.31 cvs 2020: /* q, p point to the first non-matching character or zero */
1.87 cvs 2021: if (*q == WC_EOS)
1.31 cvs 2022: {
2023: /* New name is a subset of the related name */
2024: /* exactly the right length */
1.67 cvs 2025: len = ustrlen (p);
2026: if ((return_value = TtaAllocString (len + 1)) != NULL)
2027: ustrcpy (return_value, p);
1.31 cvs 2028: }
2029: else if ((slashes < 2 && after_access == NULL)
2030: || (slashes < 3 && after_access != NULL))
2031: {
2032: /* Two names whitout common path */
2033: /* exactly the right length */
1.67 cvs 2034: len = ustrlen (aName);
2035: if ((return_value = TtaAllocString (len + 1)) != NULL)
2036: ustrcpy (return_value, aName);
1.31 cvs 2037: }
2038: else
2039: {
2040: /* Some path in common */
1.67 cvs 2041: if (slashes == 3 && ustrncmp (aName, TEXT("http:"), 5) == 0)
1.31 cvs 2042: /* just the same server */
1.67 cvs 2043: ustrcpy (result, last_slash);
1.31 cvs 2044: else
2045: {
2046: levels= 0;
1.67 cvs 2047: for (; *q && *q != TEXT('#') && *q != TEXT(';') && *q != TEXT('?'); q++)
1.31 cvs 2048: if (*q == DIR_SEP)
2049: levels++;
2050:
1.87 cvs 2051: result[0] = WC_EOS;
1.31 cvs 2052: for (;levels; levels--)
1.67 cvs 2053: ustrcat (result, TEXT("../"));
2054: ustrcat (result, last_slash+1);
1.31 cvs 2055: }
1.52 cvs 2056:
2057: if (!*result)
1.67 cvs 2058: ustrcat (result, TEXT("./"));
1.52 cvs 2059:
1.31 cvs 2060: /* exactly the right length */
1.67 cvs 2061: len = ustrlen (result);
2062: if ((return_value = TtaAllocString (len + 1)) != NULL)
2063: ustrcpy (return_value, result);
1.52 cvs 2064:
1.25 cvs 2065: }
1.44 cvs 2066: # ifdef _WINDOWS
1.67 cvs 2067: len = ustrlen (return_value);
1.44 cvs 2068: for (ndx = 0; ndx < len; ndx ++)
1.67 cvs 2069: if (return_value[ndx] == TEXT('\\'))
2070: return_value[ndx] = TEXT('/') ;
1.44 cvs 2071: # endif /* _WINDOWS */
1.29 cvs 2072: return (return_value);
1.24 cvs 2073: }
1.35 cvs 2074:
2075:
Webmaster