Annotation of Amaya/amaya/AHTURLTools.c, revision 1.63
1.7 cvs 1: /*
2: *
3: * (c) COPYRIGHT MIT and INRIA, 1996.
4: * Please first read the full copyright statement in file COPYRIGHT.
5: *
6: */
1.9 cvs 7:
1.10 cvs 8: /*
9: * AHTURLTools.c: contains all the functions for testing, manipulating,
1.25 cvs 10: * and normalizing URLs. It also contains a local copy of the libWWW
11: * URL parsing functions.
1.10 cvs 12: *
13: * Authors: J. Kahan, I. Vatton
1.45 cvs 14: * R. Guetari (Stuff related to Windows).
1.10 cvs 15: *
16: */
1.7 cvs 17:
1.15 cvs 18: #define THOT_EXPORT extern
1.3 cvs 19: #include "amaya.h"
20:
1.8 cvs 21: #include "init_f.h"
22: #include "AHTURLTools_f.h"
23:
1.24 cvs 24: #define MAX_PRINT_URL_LENGTH 50
1.29 cvs 25: typedef struct _HTURI {
26: char * access; /* Now known as "scheme" */
27: char * host;
28: char * absolute;
29: char * relative;
30: char * fragment;
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__
1.38 cvs 39: void ConvertToLowerCase (char *string)
1.28 cvs 40: #else /* __STDC__ */
1.38 cvs 41: void ConvertToLowerCase (string)
1.28 cvs 42: char *string;
43:
44: #endif /* __STDC__ */
45: {
46: int i;
47:
48: if (!string)
49: return;
50:
51: for (i = 0; string[i] != EOS; i++)
52: string[i] = tolower (string[i]);
53: }
1.22 cvs 54:
1.8 cvs 55: /*----------------------------------------------------------------------
1.11 cvs 56: ExplodeURL
1.8 cvs 57: ----------------------------------------------------------------------*/
58: #ifdef __STDC__
59: void ExplodeURL (char *url, char **proto, char **host, char **dir, char **file)
60: #else
61: void ExplodeURL (url, proto, host, dir, file)
62: char *url;
63: char **proto;
64: char **host;
65: char **dir;
66: char **file;
67:
68: #endif
69: {
1.33 cvs 70: char *curr, *temp;
71: char used_sep;
1.32 cvs 72:
1.33 cvs 73: if (url && strchr (url, URL_SEP))
74: used_sep = URL_SEP;
75: else
76: used_sep = DIR_SEP;
1.8 cvs 77:
78: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
79: (dir == NULL) || (file == NULL))
80: return;
81:
82: /* initialize every pointer */
83: *proto = *host = *dir = *file = NULL;
84:
85: /* skip any leading space */
86: while ((*url == SPACE) || (*url == TAB))
87: url++;
1.9 cvs 88: curr = url;
89: if (*curr == 0)
1.8 cvs 90: goto finished;
91:
92: /* go to the end of the URL */
1.9 cvs 93: while ((*curr != 0) && (*curr != SPACE) && (*curr != '\b') &&
94: (*curr != '\r') && (*curr != EOL))
95: curr++;
1.8 cvs 96:
97: /* mark the end of the chain */
1.9 cvs 98: *curr = EOS;
99: curr--;
100: if (curr <= url)
1.8 cvs 101: goto finished;
102:
103: /* search the next DIR_SEP indicating the beginning of the file name */
104: do
1.11 cvs 105: curr--;
1.33 cvs 106: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 107:
1.9 cvs 108: if (curr < url)
1.8 cvs 109: goto finished;
1.9 cvs 110: *file = curr + 1;
1.8 cvs 111:
112: /* mark the end of the dir */
1.9 cvs 113: *curr = EOS;
114: curr--;
115: if (curr < url)
1.8 cvs 116: goto finished;
117:
1.29 cvs 118: /* search for the DIR_STR indicating the host name start */
1.33 cvs 119: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 120: curr--;
1.8 cvs 121:
122: /* if we found it, separate the host name from the directory */
1.33 cvs 123: if ((*curr == DIR_SEP) && (*(curr + 1) == used_sep))
1.8 cvs 124: {
1.9 cvs 125: *host = temp = curr + 2;
1.33 cvs 126: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 127: temp++;
1.33 cvs 128: if (*temp == used_sep)
1.8 cvs 129: {
130: *temp = EOS;
131: *dir = temp + 1;
132: }
133: }
134: else
1.11 cvs 135: *dir = curr;
136:
1.9 cvs 137: if (curr <= url)
1.8 cvs 138: goto finished;
139:
140: /* mark the end of the proto */
1.9 cvs 141: *curr = EOS;
142: curr--;
143: if (curr < url)
1.8 cvs 144: goto finished;
145:
1.32 cvs 146: if (*curr == ':')
1.8 cvs 147: {
1.9 cvs 148: *curr = EOS;
149: curr--;
1.8 cvs 150: }
151: else
152: goto finished;
1.11 cvs 153:
1.9 cvs 154: if (curr < url)
1.8 cvs 155: goto finished;
1.9 cvs 156: while ((curr > url) && (isalpha (*curr)))
157: curr--;
158: *proto = curr;
1.8 cvs 159:
160: finished:;
161:
162: #ifdef AMAYA_DEBUG
163: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
164: if (*proto)
165: fprintf (stderr, "proto : %s, ", *proto);
166: if (*host)
167: fprintf (stderr, "host : %s, ", *host);
168: if (*dir)
169: fprintf (stderr, "dir : %s, ", *dir);
170: if (*file)
171: fprintf (stderr, "file : %s ", *file);
172: fprintf (stderr, "\n");
173: #endif
174:
175: }
1.3 cvs 176:
1.61 cvs 177:
178: /*----------------------------------------------------------------------
179: ExtractSuffix extract suffix from document nane.
180: ----------------------------------------------------------------------*/
181: #ifdef __STDC__
182: void ExtractSuffix (STRING aName, STRING aSuffix)
183: #else
184: void ExtractSuffix (aName, aSuffix)
185: STRING aName;
186: STRING aSuffix;
187:
188: #endif
189: {
190: int lg, i;
191: STRING ptr, oldptr;
192:
193: if (!aSuffix || !aName)
194: /* bad suffix */
195: return;
196:
197: aSuffix[0] = EOS;
198: lg = ustrlen (aName);
199: if (lg)
200: {
201: /* the name is not empty */
202: oldptr = ptr = &aName[0];
203: do
204: {
205: ptr = ustrrchr (oldptr, '.');
206: if (ptr)
207: oldptr = &ptr[1];
208: }
209: while (ptr);
210:
211: i = (int) (oldptr) - (int) (aName); /* name length */
212: if (i > 1)
213: {
214: aName[i - 1] = EOS;
215: if (i != lg)
216: ustrcpy (aSuffix, oldptr);
217: }
218: }
219: }
220:
1.4 cvs 221: /*----------------------------------------------------------------------
1.9 cvs 222: IsHTMLName
223: returns TRUE if path points to an HTML resource.
1.4 cvs 224: ----------------------------------------------------------------------*/
1.3 cvs 225: #ifdef __STDC__
1.34 cvs 226: boolean IsHTMLName (const char *path)
1.3 cvs 227: #else /* __STDC__ */
228: boolean IsHTMLName (path)
1.34 cvs 229: const char *path;
1.3 cvs 230: #endif /* __STDC__ */
231: {
1.5 cvs 232: char temppath[MAX_LENGTH];
233: char suffix[MAX_LENGTH];
234: char nsuffix[MAX_LENGTH];
235: int i;
236:
237: if (!path)
1.37 cvs 238: return (FALSE);
1.5 cvs 239:
240: strcpy (temppath, path);
241: ExtractSuffix (temppath, suffix);
242:
243: /* Normalize the suffix */
244: i = 0;
1.39 cvs 245: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 246: {
1.25 cvs 247: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 248: i++;
249: }
1.5 cvs 250: nsuffix[i] = EOS;
1.60 cvs 251: if (!strcmp (nsuffix, "html") ||
252: !strcmp (nsuffix, "htm") ||
1.63 ! cvs 253: !strcmp (nsuffix, "shtml") ||
! 254: !strcmp (nsuffix, "xht") ||
! 255: !strcmp (nsuffix, "xhtm") ||
! 256: !strcmp (nsuffix, "xhtml"))
1.39 cvs 257: return (TRUE);
1.22 cvs 258: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 259: {
1.39 cvs 260: /* take into account compressed files */
1.13 cvs 261: ExtractSuffix (temppath, suffix);
262: /* Normalize the suffix */
263: i = 0;
1.39 cvs 264: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 265: {
1.25 cvs 266: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 267: i++;
268: }
269: nsuffix[i] = EOS;
1.60 cvs 270: if (!strcmp (nsuffix, "html") ||
271: !strcmp (nsuffix, "htm") ||
1.63 ! cvs 272: !strcmp (nsuffix, "shtml") ||
! 273: !strcmp (nsuffix, "xht") ||
! 274: !strcmp (nsuffix, "xhtm") ||
! 275: !strcmp (nsuffix, "xhtml"))
! 276:
1.39 cvs 277: return (TRUE);
278: else
1.13 cvs 279: return (FALSE);
280: }
281: else
1.39 cvs 282: return (FALSE);
1.3 cvs 283: }
284:
1.4 cvs 285: /*----------------------------------------------------------------------
1.56 cvs 286: IsXMLName
287: returns TRUE if path points to an XML resource.
288: ----------------------------------------------------------------------*/
289: #ifdef __STDC__
290: boolean IsXMLName (const char *path)
291: #else /* __STDC__ */
292: boolean IsXMLName (path)
293: const char *path;
294: #endif /* __STDC__ */
295: {
296: char temppath[MAX_LENGTH];
297: char suffix[MAX_LENGTH];
298:
299: if (!path)
300: return (FALSE);
301:
302: strcpy (temppath, path);
303: ExtractSuffix (temppath, suffix);
304:
305: if (!strcasecmp (suffix, "xml"))
306: return (TRUE);
307: else if (!strcmp (suffix, "gz"))
308: {
309: /* take into account compressed files */
310: ExtractSuffix (temppath, suffix);
311: if (!strcasecmp (suffix, "xml"))
1.60 cvs 312: return (TRUE);
313: else
314: return (FALSE);
315: }
316: else
317: return (FALSE);
318: }
319:
320: /*----------------------------------------------------------------------
321: IsCSSName
322: returns TRUE if path points to an XML resource.
323: ----------------------------------------------------------------------*/
324: #ifdef __STDC__
325: boolean IsCSSName (const char *path)
326: #else /* __STDC__ */
327: boolean IsCSSName (path)
328: const char *path;
329: #endif /* __STDC__ */
330: {
331: char temppath[MAX_LENGTH];
332: char suffix[MAX_LENGTH];
333:
334: if (!path)
335: return (FALSE);
336:
337: strcpy (temppath, path);
338: ExtractSuffix (temppath, suffix);
339:
340: if (!strcasecmp (suffix, "css"))
341: return (TRUE);
342: else if (!strcmp (suffix, "gz"))
343: {
344: /* take into account compressed files */
345: ExtractSuffix (temppath, suffix);
346: if (!strcasecmp (suffix, "css"))
1.56 cvs 347: return (TRUE);
348: else
349: return (FALSE);
350: }
351: else
352: return (FALSE);
353: }
354:
355: /*----------------------------------------------------------------------
1.9 cvs 356: IsImageName
357: returns TRUE if path points to an image resource.
1.4 cvs 358: ----------------------------------------------------------------------*/
1.3 cvs 359: #ifdef __STDC__
1.34 cvs 360: boolean IsImageName (const char *path)
1.3 cvs 361: #else /* __STDC__ */
362: boolean IsImageName (path)
1.34 cvs 363: const char *path;
1.3 cvs 364: #endif /* __STDC__ */
365: {
1.5 cvs 366: char temppath[MAX_LENGTH];
367: char suffix[MAX_LENGTH];
368: char nsuffix[MAX_LENGTH];
369: int i;
370:
371: if (!path)
1.13 cvs 372: return (FALSE);
1.5 cvs 373:
374: strcpy (temppath, path);
375: ExtractSuffix (temppath, suffix);
376:
377: /* Normalize the suffix */
378: i = 0;
1.39 cvs 379: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 380: {
1.25 cvs 381: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 382: i++;
383: }
1.5 cvs 384: nsuffix[i] = EOS;
1.39 cvs 385: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
386: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
387: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
388: return (TRUE);
389: return (FALSE);
1.3 cvs 390: }
391:
1.4 cvs 392: /*----------------------------------------------------------------------
1.58 cvs 393: IsImageType
394: returns TRUE if type points to an image resource.
395: ----------------------------------------------------------------------*/
396: #ifdef __STDC__
397: boolean IsImageType (const char *type)
398: #else /* __STDC__ */
399: boolean IsImageType (type)
400: const char *type;
401: #endif /* __STDC__ */
402: {
403: char temptype[MAX_LENGTH];
404: int i;
405:
406: if (!type)
407: return (FALSE);
408:
409: strcpy (temptype, type);
410: /* Normalize the type */
411: i = 0;
412: while (temptype[i] != EOS)
413: {
414: temptype[i] = tolower (temptype[i]);
415: i++;
416: }
417: if ((!strcmp (temptype, "gif")) || (!strcmp (temptype, "x-xbitmap")) ||
418: (!strcmp (temptype, "x-xpixmap")) || (!strcmp (temptype, "jpeg")) ||
419: (!strcmp (temptype, "png")))
420: return (TRUE);
421: return (FALSE);
422: }
423:
424: /*----------------------------------------------------------------------
1.9 cvs 425: IsTextName
1.4 cvs 426: ----------------------------------------------------------------------*/
1.3 cvs 427: #ifdef __STDC__
1.34 cvs 428: boolean IsTextName (const char *path)
1.3 cvs 429: #else /* __STDC__ */
430: boolean IsTextName (path)
1.34 cvs 431: const char *path;
1.3 cvs 432:
433: #endif /* __STDC__ */
434: {
1.5 cvs 435: char temppath[MAX_LENGTH];
436: char suffix[MAX_LENGTH];
437: char nsuffix[MAX_LENGTH];
438: int i;
439:
440: if (!path)
1.13 cvs 441: return (FALSE);
1.5 cvs 442:
443: strcpy (temppath, path);
444: ExtractSuffix (temppath, suffix);
445:
446: /* Normalize the suffix */
447: i = 0;
1.39 cvs 448: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 449: {
1.25 cvs 450: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 451: i++;
452: }
453: nsuffix[i] = EOS;
454:
1.39 cvs 455: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 456: return (TRUE);
1.22 cvs 457: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 458: {
1.39 cvs 459: /* take into account compressed files */
1.13 cvs 460: ExtractSuffix (temppath, suffix);
461: /* Normalize the suffix */
462: i = 0;
1.39 cvs 463: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 464: {
1.25 cvs 465: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 466: i++;
467: }
468: nsuffix[i] = EOS;
1.39 cvs 469: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 470: return (TRUE);
471: else
472: return (FALSE);
473: }
474: else
475: return (FALSE);
1.3 cvs 476: }
477:
1.4 cvs 478: /*----------------------------------------------------------------------
1.9 cvs 479: IsHTTPPath
480: returns TRUE if path is in fact an http URL.
1.4 cvs 481: ----------------------------------------------------------------------*/
1.3 cvs 482: #ifdef __STDC__
1.34 cvs 483: boolean IsHTTPPath (const char *path)
1.3 cvs 484: #else /* __STDC__ */
485: boolean IsHTTPPath (path)
1.34 cvs 486: const char *path;
1.3 cvs 487: #endif /* __STDC__ */
488: {
1.5 cvs 489: if (!path)
490: return FALSE;
1.3 cvs 491:
1.58 cvs 492: if ((!strncmp (path, "http:", 5) != 0)
493: || !strncmp (path, "internal:", 9))
494: return TRUE;
495: return FALSE;
1.3 cvs 496: }
497:
1.4 cvs 498: /*----------------------------------------------------------------------
1.9 cvs 499: IsWithParameters
500: returns TRUE if url has a concatenated query string.
1.4 cvs 501: ----------------------------------------------------------------------*/
1.3 cvs 502: #ifdef __STDC__
1.34 cvs 503: boolean IsWithParameters (const char *url)
1.3 cvs 504: #else /* __STDC__ */
1.9 cvs 505: boolean IsWithParameters (url)
1.34 cvs 506: const char *url;
1.3 cvs 507: #endif /* __STDC__ */
508: {
1.5 cvs 509: int i;
1.3 cvs 510:
1.9 cvs 511: if ((!url) || (url[0] == EOS))
1.5 cvs 512: return FALSE;
1.3 cvs 513:
1.9 cvs 514: i = strlen (url) - 1;
515: while (i > 0 && url[i--] != '?')
1.5 cvs 516: if (i < 0)
517: return FALSE;
1.3 cvs 518:
1.5 cvs 519: /* There is a parameter */
520: return TRUE;
1.3 cvs 521: }
522:
1.4 cvs 523: /*----------------------------------------------------------------------
1.9 cvs 524: IsW3Path
525: returns TRUE if path is in fact a URL.
1.4 cvs 526: ----------------------------------------------------------------------*/
1.3 cvs 527: #ifdef __STDC__
1.34 cvs 528: boolean IsW3Path (const char *path)
1.3 cvs 529: #else /* __STDC__ */
530: boolean IsW3Path (path)
1.34 cvs 531: const char *path;
1.3 cvs 532: #endif /* __STDC__ */
533: {
1.5 cvs 534: if ((strncmp (path, "http:", 5)) && (strncmp (path, "ftp:", 4)) &&
535: (strncmp (path, "telnet:", 7)) && (strncmp (path, "wais:", 5)) &&
536: (strncmp (path, "news:", 5)) && (strncmp (path, "gopher:", 7)) &&
537: (strncmp (path, "mailto:", 7)) && (strncmp (path, "archie:", 7)))
538: return FALSE;
539: return TRUE;
1.3 cvs 540: }
541:
1.4 cvs 542: /*----------------------------------------------------------------------
1.9 cvs 543: IsValidProtocol
544: returns true if the url protocol is supported by Amaya.
1.4 cvs 545: ----------------------------------------------------------------------*/
1.3 cvs 546: #ifdef __STDC__
1.34 cvs 547: boolean IsValidProtocol (const char *url)
1.3 cvs 548: #else /* __STDC__ */
1.9 cvs 549: boolean IsValidProtocol (url)
1.34 cvs 550: const char *url;
1.3 cvs 551: #endif /* __STDC__ */
552: {
1.58 cvs 553: if (!strncmp (url, "http:", 5)
554: || !strncmp (url, "internal:", 9))
1.22 cvs 555: /* experimental */
1.58 cvs 556: /*** || !strncmp (url, "ftp:", 4) ***/
1.24 cvs 557: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 558: return (TRUE);
1.5 cvs 559: else
1.8 cvs 560: return (FALSE);
1.3 cvs 561: }
562:
1.31 cvs 563:
564: /*----------------------------------------------------------------------
565: GetBaseURL
566: normalizes orgName according to a base associated with doc, and
567: following the standard URL format rules.
568: The function returns the base used to solve relative URL and SRC:
569: - the base of the document,
570: - or the document path (without document name).
571: ----------------------------------------------------------------------*/
572: #ifdef __STDC__
573: char *GetBaseURL (Document doc)
574: #else /* __STDC__ */
575: char *GetBaseURL (doc)
576: Document doc;
577: #endif /* __STDC__ */
578: {
579: Element el;
580: ElementType elType;
581: AttributeType attrType;
582: Attribute attr;
583: char *ptr, *basename;
584: int length;
585:
1.57 cvs 586: /* @@@ irene */
587: if (!DocumentURLs[doc])
588: return NULL;
1.31 cvs 589: basename = TtaGetMemory (MAX_LENGTH);
1.39 cvs 590: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
591: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 592: length = MAX_LENGTH -1;
593: /* get the root element */
594: el = TtaGetMainRoot (doc);
595: /* search the BASE element */
596: elType.ElSSchema = TtaGetDocumentSSchema (doc);
597: elType.ElTypeNum = HTML_EL_BASE;
598: el = TtaSearchTypedElement (elType, SearchInTree, el);
599: if (el)
600: {
601: /* The document has a BASE element -> Get the HREF attribute */
602: attrType.AttrSSchema = elType.ElSSchema;
603: attrType.AttrTypeNum = HTML_ATTR_HREF_;
604: attr = TtaGetAttribute (el, attrType);
605: if (attr)
606: {
607: /* Use the base path of the document */
608: TtaGiveTextAttributeValue (attr, basename, &length);
609: /* base and orgName have to be separated by a DIR_SEP */
610: length--;
1.43 cvs 611: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 612: /* verify if the base has the form "protocol://server:port" */
613: {
1.33 cvs 614: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
615: AMAYA_PARSE_HOST |
616: AMAYA_PARSE_PUNCTUATION);
1.31 cvs 617: if (ptr && !strcmp (ptr, basename))
618: {
1.43 cvs 619: /* it has this form, we complete it by adding a URL_STR */
620: if (strchr (basename, DIR_SEP))
621: strcat (basename, DIR_STR);
622: else
623: strcat (basename, URL_STR);
1.31 cvs 624: length++;
625: }
626: if (ptr)
627: TtaFreeMemory (ptr);
628: }
629: }
1.33 cvs 630: }
631:
1.31 cvs 632: /* Remove anything after the last DIR_SEP char. If no such char is found,
633: * then search for the first ":" char, hoping that what's before that is a
634: * protocol. If found, end the string there. If neither char is found,
635: * then discard the whole base element.
636: */
637: length = strlen (basename) - 1;
638: /* search for the last DIR_SEP char */
1.43 cvs 639: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 640: length--;
641: if (length >= 0)
642: /* found the last DIR_SEP char, end the string there */
643: basename[length + 1] = EOS;
644: else
645: /* search for the first PATH_STR char */
646: {
1.32 cvs 647: for (length = 0; basename[length] != ':' &&
1.31 cvs 648: basename[length] != EOS; length ++);
1.32 cvs 649: if (basename[length] == ':')
1.31 cvs 650: /* found, so end the string there */
651: basename[length + 1] = EOS;
652: else
653: /* not found, discard the base */
654: basename[0] = EOS;
655: }
656: return (basename);
657: }
658:
659:
1.4 cvs 660: /*----------------------------------------------------------------------
1.40 cvs 661: GetLocalPath
662: Allocate and return the local document path associated to the url
663: ----------------------------------------------------------------------*/
664: #ifdef __STDC__
665: char *GetLocalPath (Document doc, char *url)
666: #else /* __STDC__ */
667: char *GetLocalPath (doc, url)
668: Document doc;
669: char *url;
670: #endif /* __STDC__ */
671: {
1.46 cvs 672: char *ptr, *n;
1.40 cvs 673: char *documentname;
1.41 cvs 674: char url_sep;
1.40 cvs 675: int len;
676: boolean noFile;
677:
678: if (url != NULL)
679: {
680: /* check whether the file name exists */
681: len = strlen (url) - 1;
1.41 cvs 682: if (IsW3Path)
683: url_sep = '/';
684: else
685: url_sep = DIR_SEP;
686: noFile = (url[len] == url_sep);
1.40 cvs 687: if (noFile)
688: url[len] = EOS;
1.47 cvs 689: ptr = TtaGetMemory (MAX_LENGTH);
690: documentname = TtaGetMemory (MAX_LENGTH);
1.40 cvs 691: TtaExtractName (url, ptr, documentname);
692: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
693: if (!TtaCheckDirectory (ptr))
694: /* directory did not exist */
1.59 cvs 695: # ifdef _WINDOWS
696: _mkdir (ptr);
697: # else /* !_WINDOWS */
1.40 cvs 698: mkdir (ptr, S_IRWXU);
1.59 cvs 699: # endif /* !_WINDOWS */
1.47 cvs 700:
701: /* don't include the query string within document name */
702: n = strrchr(documentname, '?');
703: if (n != NULL)
704: *n = EOS;
1.46 cvs 705: /* don't include ':' within document name */
706: n = strchr (documentname, ':');
707: if (n != NULL)
708: *n = EOS;
1.40 cvs 709: strcat (ptr, documentname);
710: TtaFreeMemory (documentname);
711: /* restore the url */
712: if (noFile)
1.41 cvs 713: url[len] = url_sep;
1.40 cvs 714: return (ptr);
715: }
716: else
717: return (NULL);
718: }
719:
720:
721: /*----------------------------------------------------------------------
1.9 cvs 722: NormalizeURL
723: normalizes orgName according to a base associated with doc, and
724: following the standard URL format rules.
1.53 cvs 725: if doc is 0 and otherPath not NULL, normalizes orgName according to this
726: other path.
1.9 cvs 727: The function returns the new complete and normalized URL
1.12 cvs 728: or file name path (newName) and the name of the document (docName).
1.9 cvs 729: N.B. If the function can't find out what's the docName, it assigns
730: the name "noname.html".
1.4 cvs 731: ----------------------------------------------------------------------*/
1.3 cvs 732: #ifdef __STDC__
1.53 cvs 733: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName, char *otherPath)
1.3 cvs 734: #else /* __STDC__ */
1.53 cvs 735: void NormalizeURL (orgName, doc, newName, docName, otherPath)
1.3 cvs 736: char *orgName;
737: Document doc;
738: char *newName;
739: char *docName;
1.53 cvs 740: char *otherPath;
1.3 cvs 741: #endif /* __STDC__ */
742: {
1.31 cvs 743: char *basename;
1.18 cvs 744: char tempOrgName[MAX_LENGTH];
1.5 cvs 745: char *ptr;
1.49 cvs 746: char used_sep;
1.5 cvs 747: int length;
1.49 cvs 748: boolean check;
1.5 cvs 749:
1.44 cvs 750: # ifdef _WINDOWS
751: int ndx;
752: # endif /* _WINDOWS */
753:
1.5 cvs 754: if (!newName || !docName)
755: return;
1.18 cvs 756:
1.32 cvs 757: if (doc != 0)
1.53 cvs 758: basename = GetBaseURL (doc);
759: else if (otherPath != NULL)
760: basename = TtaStrdup (otherPath);
1.32 cvs 761: else
1.53 cvs 762: basename = NULL;
1.32 cvs 763:
1.18 cvs 764: /*
1.31 cvs 765: * Clean orgName
766: * Make sure we have a complete orgName, without any leading or trailing
767: * white spaces, or trailinbg new lines
768: */
1.5 cvs 769: ptr = orgName;
1.18 cvs 770: /* skip leading white space and new line characters */
1.19 cvs 771: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS);
1.39 cvs 772: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
773: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 774: /*
1.31 cvs 775: * Make orgName a complete URL
776: * If the URL does not include a protocol, then try to calculate
777: * one using the doc's base element (if it exists),
778: */
1.53 cvs 779: if (tempOrgName[0] == EOS)
780: {
781: newName[0] = EOS;
782: TtaFreeMemory (basename);
783: return;
784: }
1.49 cvs 785:
786: /* clean trailing white space */
787: length = strlen (tempOrgName) - 1;
1.53 cvs 788: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
789: {
790: tempOrgName[length] = EOS;
791: length--;
792: }
1.50 cvs 793:
1.55 cvs 794: /* remove extra dot (which dot???) */
795: /* ugly, but faster than a strcmp */
796: if (tempOrgName[length] == '.'
797: && (length == 0 || tempOrgName[length-1] != '.'))
798: tempOrgName[length] = EOS;
1.50 cvs 799:
1.53 cvs 800: if (IsW3Path (tempOrgName))
801: {
802: /* the name is complete, go to the Sixth Step */
803: strcpy (newName, tempOrgName);
804: SimplifyUrl (&newName);
805: /* verify if the URL has the form "protocol://server:port" */
806: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
807: if (ptr && !strcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
1.50 cvs 808: strcat (newName, URL_STR);
1.49 cvs 809:
1.53 cvs 810: if (ptr)
1.50 cvs 811: TtaFreeMemory (ptr);
1.53 cvs 812: }
813: else if ( basename == NULL)
814: /* the name is complete, go to the Sixth Step */
1.18 cvs 815: strcpy (newName, tempOrgName);
1.53 cvs 816: else
817: {
1.31 cvs 818: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 819: # ifdef _WINDOWS
1.53 cvs 820: if (!IsW3Path (basename))
821: {
822: length = strlen (tempOrgName);
823: for (ndx = 0; ndx < length; ndx++)
824: if (tempOrgName [ndx] == '/')
825: tempOrgName [ndx] = '\\';
826: }
1.44 cvs 827: # endif /* _WINDOWS */
1.25 cvs 828: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 829: if (ptr)
830: {
831: SimplifyUrl (&ptr);
832: strcpy (newName, ptr);
833: TtaFreeMemory (ptr);
834: }
835: else
836: newName[0] = EOS;
837: }
1.36 cvs 838:
839: TtaFreeMemory (basename);
1.18 cvs 840: /*
1.31 cvs 841: * Prepare the docname that will refer to this ressource in the
842: * .amaya directory. If the new URL finishes on DIR_SEP, then use
843: * noname.html as a default ressource name
1.18 cvs 844: */
1.53 cvs 845: if (newName[0] != EOS)
846: {
847: length = strlen (newName) - 1;
848: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
849: {
850: used_sep = newName[length];
851: check = TRUE;
852: while (check)
853: {
1.50 cvs 854: length--;
855: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 856: length--;
857: if (!strncmp (&newName[length+1], "..", 2))
858: {
859: newName[length+1] = EOS;
860: /* remove also previous directory */
861: length--;
862: while (length >= 0 && newName[length] != used_sep)
863: length--;
864: if (strncmp (&newName[length+1], "//", 2))
865: /* don't remove server name */
1.50 cvs 866: newName[length+1] = EOS;
1.53 cvs 867: }
868: else if (!strncmp (&newName[length+1], ".", 1))
869: newName[length+1] = EOS;
1.50 cvs 870: else
1.53 cvs 871: check = FALSE;
872: }
873: strcpy (docName, "noname.html");
874: /* docname was not comprised inside the URL, so let's */
875: /* assign the default ressource name */
876: strcpy (docName, "noname.html");
877: }
878: else
879: { /* docname is comprised inside the URL */
1.50 cvs 880: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP)
1.53 cvs 881: length--;
882: if (length < 0)
1.50 cvs 883: strcpy (docName, newName);
1.53 cvs 884: else
885: strcpy (docName, &newName[length+1]);
886: }
887: }
888: else
889: docName[0] = EOS;
1.18 cvs 890: }
1.3 cvs 891:
1.4 cvs 892: /*----------------------------------------------------------------------
1.9 cvs 893: IsSameHost
1.4 cvs 894: ----------------------------------------------------------------------*/
1.3 cvs 895: #ifdef __STDC__
1.34 cvs 896: boolean IsSameHost (const char *url1, const char *url2)
1.3 cvs 897: #else /* __STDC__ */
898: boolean IsSameHost (url1, url2)
1.34 cvs 899: const char *url1;
900: const char *url2;
1.3 cvs 901: #endif /* __STDC__ */
902: {
1.34 cvs 903: char *basename_ptr1, *basename_ptr2;
904: boolean result;
1.3 cvs 905:
1.25 cvs 906: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
907: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 908:
1.5 cvs 909: if (strcmp (basename_ptr1, basename_ptr2))
1.8 cvs 910: result = FALSE;
1.5 cvs 911: else
1.8 cvs 912: result = TRUE;
1.3 cvs 913:
1.25 cvs 914: TtaFreeMemory (basename_ptr1);
915: TtaFreeMemory (basename_ptr2);
1.5 cvs 916: return (result);
1.3 cvs 917: }
918:
919:
1.4 cvs 920: /*----------------------------------------------------------------------
1.22 cvs 921: HasKnownFileSuffix
922: returns TRUE if path points to a file ending with a suffix.
923: ----------------------------------------------------------------------*/
924: #ifdef __STDC__
1.34 cvs 925: boolean HasKnownFileSuffix (const char *path)
1.22 cvs 926: #else /* __STDC__ */
927: boolean HasKnownFileSuffix (path)
1.34 cvs 928: const char *path;
1.22 cvs 929: #endif /* __STDC__ */
930: {
1.29 cvs 931: char *root;
932: char temppath[MAX_LENGTH];
933: char suffix[MAX_LENGTH];
1.22 cvs 934:
1.24 cvs 935: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 936: return (FALSE);
937:
1.29 cvs 938: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 939:
940: if (root)
941: {
942: strcpy (temppath, root);
1.25 cvs 943: TtaFreeMemory (root);
1.22 cvs 944: /* Get the suffix */
945: ExtractSuffix (temppath, suffix);
946:
947: if( suffix[0] == EOS)
948: /* no suffix */
949: return (FALSE);
950:
951: /* Normalize the suffix */
952: ConvertToLowerCase (suffix);
953:
1.23 cvs 954: if (!strcmp (suffix, "gz"))
1.22 cvs 955: /* skip the compressed suffix */
956: {
957: ExtractSuffix (temppath, suffix);
958: if(suffix[0] == EOS)
959: /* no suffix */
960: return (FALSE);
961: /* Normalize the suffix */
962: ConvertToLowerCase (suffix);
963: }
964:
1.63 ! cvs 965: if (strcmp (suffix, "gif") &&
! 966: strcmp (suffix, "xbm") &&
! 967: strcmp (suffix, "xpm") &&
! 968: strcmp (suffix, "jpg") &&
! 969: strcmp (suffix, "pdf") &&
! 970: strcmp (suffix, "png") &&
! 971: strcmp (suffix, "tgz") &&
! 972: strcmp (suffix, "xpg") &&
! 973: strcmp (suffix, "xpd") &&
! 974: strcmp (suffix, "ps") &&
! 975: strcmp (suffix, "au") &&
! 976: strcmp (suffix, "html") &&
! 977: strcmp (suffix, "htm") &&
! 978: strcmp (suffix, "shtml") &&
! 979: strcmp (suffix, "xht") &&
! 980: strcmp (suffix, "xhtm") &&
! 981: strcmp (suffix, "xhtml") &&
! 982: strcmp (suffix, "txt") &&
! 983: strcmp (suffix, "css") &&
! 984: strcmp (suffix, "eps"))
1.22 cvs 985: return (FALSE);
986: else
987: return (TRUE);
988: }
989: else
990: return (FALSE);
991: }
992:
993:
994: /*----------------------------------------------------------------------
1.24 cvs 995: ChopURL
996: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
997: If inputURL is bigger than that size, outputURL receives
998: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
999: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1000: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1001: copied into outputURL.
1002: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1003: chars.
1004: ----------------------------------------------------------------------*/
1005: #ifdef __STDC__
1.34 cvs 1006: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1007: #else
1008: void ChopURL (outputURL, inputURL)
1.34 cvs 1009: char *outputURL;
1010: const char *inputURL;
1.24 cvs 1011: #endif
1.22 cvs 1012:
1.24 cvs 1013: {
1014: int len;
1.9 cvs 1015:
1.24 cvs 1016: len = strlen (inputURL);
1017: if (len <= MAX_PRINT_URL_LENGTH)
1.29 cvs 1018: strcpy (outputURL, inputURL);
1.24 cvs 1019: else
1020: /* make a truncated urlName on the status window */
1021: {
1022: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1023: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1024: strcat (outputURL, "...");
1025: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1026: }
1.25 cvs 1027: }
1028:
1029:
1030: /*----------------------------------------------------------------------
1031: scan
1.47 cvs 1032: Scan a filename for its constituents
1.25 cvs 1033: -----------------------------------
1034:
1035: On entry,
1036: name points to a document name which may be incomplete.
1037: On exit,
1038: absolute or relative may be nonzero (but not both).
1039: host, fragment and access may be nonzero if they were specified.
1040: Any which are nonzero point to zero terminated strings.
1041: ----------------------------------------------------------------------*/
1042: #ifdef __STDC__
1043: static void scan (char * name, HTURI * parts)
1044: #else /* __STDC__ */
1045: static void scan (name, parts)
1046: char *name;
1047: HTURI *parts;
1048:
1049: #endif /* __STDC__ */
1050: {
1.43 cvs 1051: char *p;
1052: char *after_access = name;
1.32 cvs 1053:
1.43 cvs 1054: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1055: /* Look for fragment identifier */
1.52 cvs 1056: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1057: {
1058: *p++ = '\0';
1059: parts->fragment = p;
1.25 cvs 1060: }
1061:
1.28 cvs 1062: for (p=name; *p; p++)
1063: {
1.43 cvs 1064: if (*p == URL_SEP || *p == DIR_SEP || *p=='#' || *p=='?')
1.28 cvs 1065: break;
1.32 cvs 1066: if (*p==':')
1.28 cvs 1067: {
1068: *p = 0;
1069: parts->access = after_access; /* Scheme has been specified */
1070:
1071: /* The combination of gcc, the "-O" flag and the HP platform is
1072: unhealthy. The following three lines is a quick & dirty fix, but is
1073: not recommended. Rather, turn off "-O". */
1074:
1075: /* after_access = p;*/
1076: /* while (*after_access == 0)*/
1077: /* after_access++;*/
1078: after_access = p+1;
1.43 cvs 1079: if (!strcasecmp("URL", parts->access))
1.28 cvs 1080: /* Ignore IETF's URL: pre-prefix */
1081: parts->access = NULL;
1082: else
1.25 cvs 1083: break;
1084: }
1085: }
1086:
1087: p = after_access;
1.43 cvs 1088: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1089: {
1.43 cvs 1090: if (p[1] == URL_SEP)
1.28 cvs 1091: {
1.25 cvs 1092: parts->host = p+2; /* host has been specified */
1.28 cvs 1093: *p = 0; /* Terminate access */
1094: /* look for end of host name if any */
1.43 cvs 1095: p = strchr (parts->host, URL_SEP);
1.28 cvs 1096: if (p)
1097: {
1.43 cvs 1098: *p = EOS; /* Terminate host */
1.25 cvs 1099: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1100: }
1101: }
1102: else
1103: /* Root found but no host */
1104: parts->absolute = p+1;
1105: }
1106: else
1107: {
1.25 cvs 1108: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1109: }
1.25 cvs 1110: }
1111:
1112:
1113: /*----------------------------------------------------------------------
1.28 cvs 1114: AmayaParseUrl: parse a Name relative to another name
1115:
1116: This returns those parts of a name which are given (and requested)
1117: substituting bits from the related name where necessary.
1.25 cvs 1118:
1.28 cvs 1119: On entry,
1.25 cvs 1120: aName A filename given
1121: relatedName A name relative to which aName is to be parsed. Give
1122: it an empty string if aName is absolute.
1123: wanted A mask for the bits which are wanted.
1124:
1.28 cvs 1125: On exit,
1.25 cvs 1126: returns A pointer to a malloc'd string which MUST BE FREED
1127: ----------------------------------------------------------------------*/
1128: #ifdef __STDC__
1.34 cvs 1129: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1.25 cvs 1130: #else /* __STDC__ */
1.28 cvs 1131: char *AmayaParseUrl (aName, relatedName, wanted)
1.34 cvs 1132: const char *aName;
1.28 cvs 1133: char *relatedName;
1134: int wanted;
1.25 cvs 1135:
1136: #endif /* __STDC__ */
1137: {
1.29 cvs 1138: char *return_value;
1139: char result[MAX_LENGTH];
1140: char name[MAX_LENGTH];
1141: char rel[MAX_LENGTH];
1142: char *p, *access;
1143: HTURI given, related;
1144: int len;
1.33 cvs 1145: char used_sep;
1146: char *used_str;
1.32 cvs 1147:
1.43 cvs 1148: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1149: {
1.42 cvs 1150: used_str = DIR_STR;
1151: used_sep = DIR_SEP;
1.33 cvs 1152: }
1.32 cvs 1153: else
1.33 cvs 1154: {
1.42 cvs 1155: used_str = URL_STR;
1156: used_sep = URL_SEP;
1.33 cvs 1157: }
1.32 cvs 1158:
1.29 cvs 1159: /* Make working copies of input strings to cut up: */
1160: return_value = NULL;
1161: result[0] = 0; /* Clear string */
1162: strcpy (name, aName);
1163: if (relatedName != NULL)
1164: strcpy (rel, relatedName);
1165: else
1166: relatedName[0] = EOS;
1167:
1168: scan (name, &given);
1169: scan (rel, &related);
1170: access = given.access ? given.access : related.access;
1171: if (wanted & AMAYA_PARSE_ACCESS)
1172: if (access)
1173: {
1174: strcat (result, access);
1175: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.32 cvs 1176: strcat (result, ":");
1.29 cvs 1177: }
1178:
1179: if (given.access && related.access)
1180: /* If different, inherit nothing. */
1181: if (strcmp (given.access, related.access) != 0)
1182: {
1183: related.host = 0;
1184: related.absolute = 0;
1185: related.relative = 0;
1186: related.fragment = 0;
1187: }
1188:
1189: if (wanted & AMAYA_PARSE_HOST)
1190: if(given.host || related.host)
1191: {
1192: if(wanted & AMAYA_PARSE_PUNCTUATION)
1193: strcat (result, "//");
1194: strcat (result, given.host ? given.host : related.host);
1195: }
1196:
1197: if (given.host && related.host)
1198: /* If different hosts, inherit no path. */
1199: if (strcmp(given.host, related.host) != 0)
1200: {
1201: related.absolute = 0;
1202: related.relative = 0;
1203: related.fragment = 0;
1204: }
1205:
1206: if (wanted & AMAYA_PARSE_PATH)
1207: {
1208: if (given.absolute)
1209: {
1210: /* All is given */
1211: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.33 cvs 1212: strcat (result, used_str);
1.29 cvs 1213: strcat (result, given.absolute);
1.25 cvs 1214: }
1.29 cvs 1215: else if (related.absolute)
1216: {
1217: /* Adopt path not name */
1.33 cvs 1218: strcat (result, used_str);
1.29 cvs 1219: strcat (result, related.absolute);
1220: if (given.relative)
1221: {
1222: /* Search part? */
1223: p = strchr (result, '?');
1224: if (!p)
1225: p=result+strlen(result)-1;
1.33 cvs 1226: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1227: /* Remove filename */
1228: p[1]=0;
1229: /* Add given one */
1230: strcat (result, given.relative);
1.25 cvs 1231: }
1232: }
1.29 cvs 1233: else if (given.relative)
1234: /* what we've got */
1235: strcat (result, given.relative);
1236: else if (related.relative)
1237: strcat (result, related.relative);
1238: else
1239: /* No inheritance */
1.33 cvs 1240: strcat (result, used_str);
1.25 cvs 1241: }
1.29 cvs 1242:
1243: if (wanted & AMAYA_PARSE_ANCHOR)
1244: if (given.fragment || related.fragment)
1245: {
1246: if (given.absolute && given.fragment)
1247: {
1248: /*Fixes for relURLs...*/
1249: if (wanted & AMAYA_PARSE_PUNCTUATION)
1250: strcat (result, "#");
1251: strcat (result, given.fragment);
1252: }
1253: else if (!(given.absolute) && !(given.fragment))
1254: strcat (result, "");
1255: else
1256: {
1257: if (wanted & AMAYA_PARSE_PUNCTUATION)
1258: strcat (result, "#");
1259: strcat (result, given.fragment ? given.fragment : related.fragment);
1260: }
1261: }
1262: len = strlen (result);
1263: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1264: strcpy (return_value, result);
1265: return (return_value); /* exactly the right length */
1.25 cvs 1266: }
1267:
1268: /*----------------------------------------------------------------------
1269: HTCanon
1270: Canonicalizes the URL in the following manner starting from the host
1271: pointer:
1272:
1273: 1) The host name is converted to lowercase
1274: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1275:
1276: Return: OK The position of the current path part of the URL
1277: which might be the old one or a new one.
1278:
1279: ----------------------------------------------------------------------*/
1280: #ifdef __STDC__
1.28 cvs 1281: static char *HTCanon (char ** filename, char * host)
1.25 cvs 1282: #else /* __STDC__ */
1.28 cvs 1283: static char *HTCanon (filename, host)
1284: char **filename;
1285: char *host;
1.25 cvs 1286: #endif /* __STDC__ */
1287: {
1288: char *newname = NULL;
1289: char *port;
1290: char *strptr;
1291: char *path;
1292: char *access = host-3;
1.33 cvs 1293: char used_sep;
1.32 cvs 1294:
1295:
1.33 cvs 1296: if (*filename && strchr (*filename, URL_SEP))
1297: {
1298: used_sep = URL_SEP;
1299: }
1300: else
1301: {
1302: used_sep = DIR_SEP;
1303: }
1.32 cvs 1304:
1.33 cvs 1305: while (access>*filename && *(access-1)!= used_sep) /* Find access method */
1.25 cvs 1306: access--;
1.33 cvs 1307: if ((path = strchr(host, used_sep)) == NULL) /* Find path */
1.25 cvs 1308: path = host + strlen(host);
1309: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */
1310: host = strptr;
1.32 cvs 1311: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */
1.25 cvs 1312: port = NULL;
1313:
1314: strptr = host; /* Convert to lower-case */
1.33 cvs 1315: while (strptr<path)
1316: {
1.25 cvs 1317: *strptr = tolower(*strptr);
1318: strptr++;
1.33 cvs 1319: }
1.25 cvs 1320:
1321: /* Does the URL contain a full domain name? This also works for a
1322: numerical host name. The domain name is already made lower-case
1323: and without a trailing dot. */
1324: {
1.33 cvs 1325: char *dot = port ? port : path;
1326: if (dot > *filename && *--dot=='.')
1327: {
1328: char *orig=dot, *dest=dot+1;
1329: while((*orig++ = *dest++));
1330: if (port) port--;
1331: path--;
1.25 cvs 1332: }
1333: }
1334: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1335: if (port)
1336: {
1337: if (!*(port+1) || *(port+1)==used_sep)
1338: {
1339: if (!newname)
1340: {
1.25 cvs 1341: char *orig=port, *dest=port+1;
1342: while((*orig++ = *dest++));
1.33 cvs 1343: }
1344: }
1345: else if ((!strncmp(access, "http", 4) &&
1346: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1347: (!strncmp(access, "gopher", 6) &&
1348: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1349: (!strncmp(access, "ftp", 3) &&
1350: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) {
1351: if (!newname)
1352: {
1353: char *orig=port, *dest=port+3;
1354: while((*orig++ = *dest++));
1355: /* Update path position, Henry Minsky */
1356: path -= 3;
1.25 cvs 1357: }
1.33 cvs 1358: }
1359: else if (newname)
1360: strncat(newname, port, (int) (path-port));
1361: }
1.25 cvs 1362:
1.33 cvs 1363: if (newname)
1364: {
1.25 cvs 1365: char *newpath = newname+strlen(newname);
1366: strcat(newname, path);
1367: path = newpath;
1.28 cvs 1368: /* Free old copy */
1369: TtaFreeMemory(*filename);
1.25 cvs 1370: *filename = newname;
1.33 cvs 1371: }
1.25 cvs 1372: return path;
1373: }
1374:
1375:
1376: /*----------------------------------------------------------------------
1.29 cvs 1377: SimplifyUrl: simplify a URI
1.32 cvs 1378: A URI is allowed to contain the sequence xxx/../ which may be
1379: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1380: Simplification helps us recognize duplicate URIs.
1.25 cvs 1381:
1.28 cvs 1382: Thus, /etc/junk/../fred becomes /etc/fred
1383: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1384:
1.28 cvs 1385: but we should NOT change
1386: http://fred.xxx.edu/../..
1.25 cvs 1387:
1388: or ../../albert.html
1389:
1.28 cvs 1390: In order to avoid empty URLs the following URLs become:
1.25 cvs 1391:
1392: /fred/.. becomes /fred/..
1393: /fred/././.. becomes /fred/..
1394: /fred/.././junk/.././ becomes /fred/..
1395:
1.28 cvs 1396: If more than one set of `://' is found (several proxies in cascade) then
1397: only the part after the last `://' is simplified.
1.25 cvs 1398:
1.28 cvs 1399: Returns: A string which might be the old one or a new one.
1.25 cvs 1400: ----------------------------------------------------------------------*/
1401: #ifdef __STDC__
1.29 cvs 1402: void SimplifyUrl (char ** url)
1.25 cvs 1403: #else /* __STDC__ */
1.29 cvs 1404: void SimplifyUrl (url)
1.28 cvs 1405: char **url;
1.25 cvs 1406: #endif /* __STDC__ */
1407: {
1.28 cvs 1408: char *path, *p;
1409: char *newptr, *access;
1410: char *orig, *dest, *end;
1411:
1.33 cvs 1412: char used_sep;
1.32 cvs 1413:
1414:
1.28 cvs 1415: if (!url || !*url)
1416: return;
1417:
1.33 cvs 1418: if (strchr (*url, URL_SEP))
1419: {
1420: used_sep = URL_SEP;
1421: }
1.32 cvs 1422: else
1.33 cvs 1423: {
1424: used_sep = DIR_SEP;
1425: }
1.32 cvs 1426:
1.28 cvs 1427: /* Find any scheme name */
1428: if ((path = strstr(*url, "://")) != NULL)
1.33 cvs 1429: {
1430: /* Find host name */
1.28 cvs 1431: access = *url;
1432: while (access<path && (*access=tolower(*access)))
1433: access++;
1434: path += 3;
1435: while ((newptr = strstr(path, "://")) != NULL)
1436: /* For proxies */
1437: path = newptr+3;
1438: /* We have a host name */
1439: path = HTCanon(url, path);
1.25 cvs 1440: }
1.28 cvs 1441: else if ((path = strstr(*url, ":/")) != NULL)
1442: path += 2;
1443: else
1444: path = *url;
1.25 cvs 1445:
1.33 cvs 1446: if (*path == used_sep && *(path+1)==used_sep)
1.28 cvs 1447: /* Some URLs start //<foo> */
1448: path += 1;
1449: else if (!strncmp(path, "news:", 5))
1450: {
1451: newptr = strchr(path+5, '@');
1452: if (!newptr)
1453: newptr = path + 5;
1454: while (*newptr)
1455: {
1456: /* Make group or host lower case */
1457: *newptr = tolower (*newptr);
1458: newptr++;
1.25 cvs 1459: }
1.28 cvs 1460: /* Doesn't need to do any more */
1461: return;
1.25 cvs 1462: }
1.28 cvs 1463:
1464: if ((p = path))
1465: {
1466: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1467: (end = strchr (path, '#'))))
1468: end = path + strlen (path);
1469:
1470: /* Parse string second time to simplify */
1471: p = path;
1472: while (p < end)
1473: {
1.33 cvs 1474: if (*p==used_sep)
1.28 cvs 1475: {
1.33 cvs 1476: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1477: {
1478: orig = p + 1;
1.33 cvs 1479: dest = (*(p+2)!=used_sep) ? p+2 : p+3;
1.52 cvs 1480: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1481: end = orig - 1;
1482: }
1.33 cvs 1483: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3)))
1.28 cvs 1484: {
1485: newptr = p;
1.52 cvs 1486: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1487: if (*newptr == used_sep)
1488: orig = newptr + 1;
1.28 cvs 1489: else
1.52 cvs 1490: orig = newptr;
1491:
1492: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1493: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1494: end = orig-1;
1495: /* Start again with prev slash */
1496: p = newptr;
1.28 cvs 1497: }
1.33 cvs 1498: else if (*(p+1) == used_sep)
1.28 cvs 1499: {
1.33 cvs 1500: while (*(p+1) == used_sep)
1.28 cvs 1501: {
1502: orig = p;
1503: dest = p + 1;
1504: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1505: end = orig-1;
1506: }
1507: }
1508: else
1.25 cvs 1509: p++;
1.28 cvs 1510: }
1511: else
1512: p++;
1.25 cvs 1513: }
1514: }
1.51 cvs 1515:
1516: /*
1517: ** Check for host/../.. kind of things
1518: */
1.52 cvs 1519: if (*path==used_sep && *(path+1)=='.' && *(path+2)=='.' && (!*(path+3) || *(path+3)==used_sep))
1.51 cvs 1520: *(path+1) = EOS;
1521:
1.28 cvs 1522: return;
1523: }
1524:
1525:
1526: /*----------------------------------------------------------------------
1527: NormalizeFile normalizes local names.
1528: Return TRUE if target and src differ.
1529: ----------------------------------------------------------------------*/
1530: #ifdef __STDC__
1531: boolean NormalizeFile (char *src, char *target)
1532: #else
1533: boolean NormalizeFile (src, target)
1534: char *src;
1535: char *target;
1536:
1537: #endif
1538: {
1539: char *s;
1540: boolean change;
1541:
1.54 cvs 1542: change = FALSE;
1.53 cvs 1543: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1544: {
1545: /* remove the prefix file: */
1546: if (src[5] == EOS)
1547: strcpy (target, DIR_STR);
1548: else if (src[0] == '~')
1549: {
1550: /* replace ~ */
1551: s = (char *) TtaGetEnvString ("HOME");
1552: strcpy (target, s);
1553: strcat (target, &src[5]);
1554: }
1555: else
1556: strcpy (target, &src[5]);
1.54 cvs 1557: change = TRUE;
1.28 cvs 1558: }
1.53 cvs 1559: # ifndef _WINDOWS
1560: else if (src[0] == '~')
1561: {
1562: /* replace ~ */
1563: s = (char *) TtaGetEnvString ("HOME");
1564: strcpy (target, s);
1565: if (src[1] != DIR_SEP)
1566: strcat (target, DIR_STR);
1567: strcat (target, &src[1]);
1.54 cvs 1568: change = TRUE;
1.53 cvs 1569: }
1570: # endif /* _WINDOWS */
1.28 cvs 1571: else
1572: strcpy (target, src);
1573:
1574: /* remove /../ and /./ */
1.29 cvs 1575: SimplifyUrl (&target);
1.54 cvs 1576: if (!change)
1577: change = strcmp (src, target);
1.28 cvs 1578: return (change);
1.25 cvs 1579: }
1580:
1.28 cvs 1581:
1.25 cvs 1582: /*----------------------------------------------------------------------
1.31 cvs 1583: MakeRelativeURL: make relative name
1.25 cvs 1584:
1.28 cvs 1585: This function creates and returns a string which gives an expression of
1586: one address as related to another. Where there is no relation, an absolute
1587: address is retured.
1.25 cvs 1588:
1.28 cvs 1589: On entry,
1.25 cvs 1590: Both names must be absolute, fully qualified names of nodes
1591: (no fragment bits)
1592:
1.28 cvs 1593: On exit,
1.25 cvs 1594: The return result points to a newly allocated name which, if
1595: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1596: The caller is responsible for freeing the resulting name later.
1597: ----------------------------------------------------------------------*/
1598: #ifdef __STDC__
1.31 cvs 1599: char *MakeRelativeURL (char *aName, char *relatedName)
1.25 cvs 1600: #else /* __STDC__ */
1.31 cvs 1601: char *MakeRelativeURL (aName, relatedName)
1.28 cvs 1602: char *aName;
1603: char *relatedName;
1.25 cvs 1604: #endif /* __STDC__ */
1605: {
1.39 cvs 1606: char *return_value;
1607: char result[MAX_LENGTH];
1.29 cvs 1608: char *p;
1.34 cvs 1609: char *q;
1.29 cvs 1610: char *after_access;
1611: char *last_slash = NULL;
1612: int slashes, levels, len;
1613:
1.44 cvs 1614: # ifdef _WINDOWS
1615: int ndx;
1616: # endif /* _WINDOWS */
1617:
1.29 cvs 1618: if (aName == NULL || relatedName == NULL)
1619: return (NULL);
1620:
1621: slashes = 0;
1622: after_access = NULL;
1623: p = aName;
1624: q = relatedName;
1625: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1626: {
1627: /* Find extent of match */
1.32 cvs 1628: if (*p == ':')
1.29 cvs 1629: after_access = p + 1;
1.28 cvs 1630: if (*p == DIR_SEP)
1.27 cvs 1631: {
1.29 cvs 1632: /* memorize the last slash position and count them */
1.27 cvs 1633: last_slash = p;
1634: slashes++;
1.25 cvs 1635: }
1636: }
1637:
1.31 cvs 1638: /* q, p point to the first non-matching character or zero */
1639: if (*q == EOS)
1640: {
1641: /* New name is a subset of the related name */
1642: /* exactly the right length */
1643: len = strlen (p);
1644: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1645: strcpy (return_value, p);
1646: }
1647: else if ((slashes < 2 && after_access == NULL)
1648: || (slashes < 3 && after_access != NULL))
1649: {
1650: /* Two names whitout common path */
1651: /* exactly the right length */
1652: len = strlen (aName);
1653: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1654: strcpy (return_value, aName);
1655: }
1656: else
1657: {
1658: /* Some path in common */
1659: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1660: /* just the same server */
1661: strcpy (result, last_slash);
1662: else
1663: {
1664: levels= 0;
1.52 cvs 1665: for (; *q && *q != '#' && *q!=';' && *q!='?'; q++)
1.31 cvs 1666: if (*q == DIR_SEP)
1667: levels++;
1668:
1.52 cvs 1669: result[0] = EOS;
1.31 cvs 1670: for (;levels; levels--)
1671: strcat (result, "../");
1672: strcat (result, last_slash+1);
1673: }
1.52 cvs 1674:
1675: if (!*result)
1676: strcat (result, "./");
1677:
1.31 cvs 1678: /* exactly the right length */
1679: len = strlen (result);
1680: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1681: strcpy (return_value, result);
1.52 cvs 1682:
1.25 cvs 1683: }
1.44 cvs 1684: # ifdef _WINDOWS
1685: len = strlen (return_value);
1686: for (ndx = 0; ndx < len; ndx ++)
1687: if (return_value[ndx] == '\\')
1688: return_value[ndx] = '/' ;
1689: # endif /* _WINDOWS */
1.29 cvs 1690: return (return_value);
1.24 cvs 1691: }
1.35 cvs 1692:
1693:
Webmaster
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to AHTURLTools.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Public] / Amaya / amaya