Annotation of Amaya/amaya/AHTURLTools.c, revision 1.62
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") ||
1.62 ! cvs 253: !strcmp (nsuffix, "xhtl") ||
1.60 cvs 254: !strcmp (nsuffix, "shtml"))
1.39 cvs 255: return (TRUE);
1.22 cvs 256: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 257: {
1.39 cvs 258: /* take into account compressed files */
1.13 cvs 259: ExtractSuffix (temppath, suffix);
260: /* Normalize the suffix */
261: i = 0;
1.39 cvs 262: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 263: {
1.25 cvs 264: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 265: i++;
266: }
267: nsuffix[i] = EOS;
1.60 cvs 268: if (!strcmp (nsuffix, "html") ||
269: !strcmp (nsuffix, "htm") ||
1.62 ! cvs 270: !strcmp (nsuffix, "xhtl") ||
1.60 cvs 271: !strcmp (nsuffix, "shtml"))
1.39 cvs 272: return (TRUE);
273: else
1.13 cvs 274: return (FALSE);
275: }
276: else
1.39 cvs 277: return (FALSE);
1.3 cvs 278: }
279:
1.4 cvs 280: /*----------------------------------------------------------------------
1.56 cvs 281: IsXMLName
282: returns TRUE if path points to an XML resource.
283: ----------------------------------------------------------------------*/
284: #ifdef __STDC__
285: boolean IsXMLName (const char *path)
286: #else /* __STDC__ */
287: boolean IsXMLName (path)
288: const char *path;
289: #endif /* __STDC__ */
290: {
291: char temppath[MAX_LENGTH];
292: char suffix[MAX_LENGTH];
293:
294: if (!path)
295: return (FALSE);
296:
297: strcpy (temppath, path);
298: ExtractSuffix (temppath, suffix);
299:
300: if (!strcasecmp (suffix, "xml"))
301: return (TRUE);
302: else if (!strcmp (suffix, "gz"))
303: {
304: /* take into account compressed files */
305: ExtractSuffix (temppath, suffix);
306: if (!strcasecmp (suffix, "xml"))
1.60 cvs 307: return (TRUE);
308: else
309: return (FALSE);
310: }
311: else
312: return (FALSE);
313: }
314:
315: /*----------------------------------------------------------------------
316: IsCSSName
317: returns TRUE if path points to an XML resource.
318: ----------------------------------------------------------------------*/
319: #ifdef __STDC__
320: boolean IsCSSName (const char *path)
321: #else /* __STDC__ */
322: boolean IsCSSName (path)
323: const char *path;
324: #endif /* __STDC__ */
325: {
326: char temppath[MAX_LENGTH];
327: char suffix[MAX_LENGTH];
328:
329: if (!path)
330: return (FALSE);
331:
332: strcpy (temppath, path);
333: ExtractSuffix (temppath, suffix);
334:
335: if (!strcasecmp (suffix, "css"))
336: return (TRUE);
337: else if (!strcmp (suffix, "gz"))
338: {
339: /* take into account compressed files */
340: ExtractSuffix (temppath, suffix);
341: if (!strcasecmp (suffix, "css"))
1.56 cvs 342: return (TRUE);
343: else
344: return (FALSE);
345: }
346: else
347: return (FALSE);
348: }
349:
350: /*----------------------------------------------------------------------
1.9 cvs 351: IsImageName
352: returns TRUE if path points to an image resource.
1.4 cvs 353: ----------------------------------------------------------------------*/
1.3 cvs 354: #ifdef __STDC__
1.34 cvs 355: boolean IsImageName (const char *path)
1.3 cvs 356: #else /* __STDC__ */
357: boolean IsImageName (path)
1.34 cvs 358: const char *path;
1.3 cvs 359: #endif /* __STDC__ */
360: {
1.5 cvs 361: char temppath[MAX_LENGTH];
362: char suffix[MAX_LENGTH];
363: char nsuffix[MAX_LENGTH];
364: int i;
365:
366: if (!path)
1.13 cvs 367: return (FALSE);
1.5 cvs 368:
369: strcpy (temppath, path);
370: ExtractSuffix (temppath, suffix);
371:
372: /* Normalize the suffix */
373: i = 0;
1.39 cvs 374: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 375: {
1.25 cvs 376: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 377: i++;
378: }
1.5 cvs 379: nsuffix[i] = EOS;
1.39 cvs 380: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
381: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
382: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
383: return (TRUE);
384: return (FALSE);
1.3 cvs 385: }
386:
1.4 cvs 387: /*----------------------------------------------------------------------
1.58 cvs 388: IsImageType
389: returns TRUE if type points to an image resource.
390: ----------------------------------------------------------------------*/
391: #ifdef __STDC__
392: boolean IsImageType (const char *type)
393: #else /* __STDC__ */
394: boolean IsImageType (type)
395: const char *type;
396: #endif /* __STDC__ */
397: {
398: char temptype[MAX_LENGTH];
399: int i;
400:
401: if (!type)
402: return (FALSE);
403:
404: strcpy (temptype, type);
405: /* Normalize the type */
406: i = 0;
407: while (temptype[i] != EOS)
408: {
409: temptype[i] = tolower (temptype[i]);
410: i++;
411: }
412: if ((!strcmp (temptype, "gif")) || (!strcmp (temptype, "x-xbitmap")) ||
413: (!strcmp (temptype, "x-xpixmap")) || (!strcmp (temptype, "jpeg")) ||
414: (!strcmp (temptype, "png")))
415: return (TRUE);
416: return (FALSE);
417: }
418:
419: /*----------------------------------------------------------------------
1.9 cvs 420: IsTextName
1.4 cvs 421: ----------------------------------------------------------------------*/
1.3 cvs 422: #ifdef __STDC__
1.34 cvs 423: boolean IsTextName (const char *path)
1.3 cvs 424: #else /* __STDC__ */
425: boolean IsTextName (path)
1.34 cvs 426: const char *path;
1.3 cvs 427:
428: #endif /* __STDC__ */
429: {
1.5 cvs 430: char temppath[MAX_LENGTH];
431: char suffix[MAX_LENGTH];
432: char nsuffix[MAX_LENGTH];
433: int i;
434:
435: if (!path)
1.13 cvs 436: return (FALSE);
1.5 cvs 437:
438: strcpy (temppath, path);
439: ExtractSuffix (temppath, suffix);
440:
441: /* Normalize the suffix */
442: i = 0;
1.39 cvs 443: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 444: {
1.25 cvs 445: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 446: i++;
447: }
448: nsuffix[i] = EOS;
449:
1.39 cvs 450: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 451: return (TRUE);
1.22 cvs 452: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 453: {
1.39 cvs 454: /* take into account compressed files */
1.13 cvs 455: ExtractSuffix (temppath, suffix);
456: /* Normalize the suffix */
457: i = 0;
1.39 cvs 458: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 459: {
1.25 cvs 460: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 461: i++;
462: }
463: nsuffix[i] = EOS;
1.39 cvs 464: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 465: return (TRUE);
466: else
467: return (FALSE);
468: }
469: else
470: return (FALSE);
1.3 cvs 471: }
472:
1.4 cvs 473: /*----------------------------------------------------------------------
1.9 cvs 474: IsHTTPPath
475: returns TRUE if path is in fact an http URL.
1.4 cvs 476: ----------------------------------------------------------------------*/
1.3 cvs 477: #ifdef __STDC__
1.34 cvs 478: boolean IsHTTPPath (const char *path)
1.3 cvs 479: #else /* __STDC__ */
480: boolean IsHTTPPath (path)
1.34 cvs 481: const char *path;
1.3 cvs 482: #endif /* __STDC__ */
483: {
1.5 cvs 484: if (!path)
485: return FALSE;
1.3 cvs 486:
1.58 cvs 487: if ((!strncmp (path, "http:", 5) != 0)
488: || !strncmp (path, "internal:", 9))
489: return TRUE;
490: return FALSE;
1.3 cvs 491: }
492:
1.4 cvs 493: /*----------------------------------------------------------------------
1.9 cvs 494: IsWithParameters
495: returns TRUE if url has a concatenated query string.
1.4 cvs 496: ----------------------------------------------------------------------*/
1.3 cvs 497: #ifdef __STDC__
1.34 cvs 498: boolean IsWithParameters (const char *url)
1.3 cvs 499: #else /* __STDC__ */
1.9 cvs 500: boolean IsWithParameters (url)
1.34 cvs 501: const char *url;
1.3 cvs 502: #endif /* __STDC__ */
503: {
1.5 cvs 504: int i;
1.3 cvs 505:
1.9 cvs 506: if ((!url) || (url[0] == EOS))
1.5 cvs 507: return FALSE;
1.3 cvs 508:
1.9 cvs 509: i = strlen (url) - 1;
510: while (i > 0 && url[i--] != '?')
1.5 cvs 511: if (i < 0)
512: return FALSE;
1.3 cvs 513:
1.5 cvs 514: /* There is a parameter */
515: return TRUE;
1.3 cvs 516: }
517:
1.4 cvs 518: /*----------------------------------------------------------------------
1.9 cvs 519: IsW3Path
520: returns TRUE if path is in fact a URL.
1.4 cvs 521: ----------------------------------------------------------------------*/
1.3 cvs 522: #ifdef __STDC__
1.34 cvs 523: boolean IsW3Path (const char *path)
1.3 cvs 524: #else /* __STDC__ */
525: boolean IsW3Path (path)
1.34 cvs 526: const char *path;
1.3 cvs 527: #endif /* __STDC__ */
528: {
1.5 cvs 529: if ((strncmp (path, "http:", 5)) && (strncmp (path, "ftp:", 4)) &&
530: (strncmp (path, "telnet:", 7)) && (strncmp (path, "wais:", 5)) &&
531: (strncmp (path, "news:", 5)) && (strncmp (path, "gopher:", 7)) &&
532: (strncmp (path, "mailto:", 7)) && (strncmp (path, "archie:", 7)))
533: return FALSE;
534: return TRUE;
1.3 cvs 535: }
536:
1.4 cvs 537: /*----------------------------------------------------------------------
1.9 cvs 538: IsValidProtocol
539: returns true if the url protocol is supported by Amaya.
1.4 cvs 540: ----------------------------------------------------------------------*/
1.3 cvs 541: #ifdef __STDC__
1.34 cvs 542: boolean IsValidProtocol (const char *url)
1.3 cvs 543: #else /* __STDC__ */
1.9 cvs 544: boolean IsValidProtocol (url)
1.34 cvs 545: const char *url;
1.3 cvs 546: #endif /* __STDC__ */
547: {
1.58 cvs 548: if (!strncmp (url, "http:", 5)
549: || !strncmp (url, "internal:", 9))
1.22 cvs 550: /* experimental */
1.58 cvs 551: /*** || !strncmp (url, "ftp:", 4) ***/
1.24 cvs 552: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 553: return (TRUE);
1.5 cvs 554: else
1.8 cvs 555: return (FALSE);
1.3 cvs 556: }
557:
1.31 cvs 558:
559: /*----------------------------------------------------------------------
560: GetBaseURL
561: normalizes orgName according to a base associated with doc, and
562: following the standard URL format rules.
563: The function returns the base used to solve relative URL and SRC:
564: - the base of the document,
565: - or the document path (without document name).
566: ----------------------------------------------------------------------*/
567: #ifdef __STDC__
568: char *GetBaseURL (Document doc)
569: #else /* __STDC__ */
570: char *GetBaseURL (doc)
571: Document doc;
572: #endif /* __STDC__ */
573: {
574: Element el;
575: ElementType elType;
576: AttributeType attrType;
577: Attribute attr;
578: char *ptr, *basename;
579: int length;
580:
1.57 cvs 581: /* @@@ irene */
582: if (!DocumentURLs[doc])
583: return NULL;
1.31 cvs 584: basename = TtaGetMemory (MAX_LENGTH);
1.39 cvs 585: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
586: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 587: length = MAX_LENGTH -1;
588: /* get the root element */
589: el = TtaGetMainRoot (doc);
590: /* search the BASE element */
591: elType.ElSSchema = TtaGetDocumentSSchema (doc);
592: elType.ElTypeNum = HTML_EL_BASE;
593: el = TtaSearchTypedElement (elType, SearchInTree, el);
594: if (el)
595: {
596: /* The document has a BASE element -> Get the HREF attribute */
597: attrType.AttrSSchema = elType.ElSSchema;
598: attrType.AttrTypeNum = HTML_ATTR_HREF_;
599: attr = TtaGetAttribute (el, attrType);
600: if (attr)
601: {
602: /* Use the base path of the document */
603: TtaGiveTextAttributeValue (attr, basename, &length);
604: /* base and orgName have to be separated by a DIR_SEP */
605: length--;
1.43 cvs 606: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 607: /* verify if the base has the form "protocol://server:port" */
608: {
1.33 cvs 609: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
610: AMAYA_PARSE_HOST |
611: AMAYA_PARSE_PUNCTUATION);
1.31 cvs 612: if (ptr && !strcmp (ptr, basename))
613: {
1.43 cvs 614: /* it has this form, we complete it by adding a URL_STR */
615: if (strchr (basename, DIR_SEP))
616: strcat (basename, DIR_STR);
617: else
618: strcat (basename, URL_STR);
1.31 cvs 619: length++;
620: }
621: if (ptr)
622: TtaFreeMemory (ptr);
623: }
624: }
1.33 cvs 625: }
626:
1.31 cvs 627: /* Remove anything after the last DIR_SEP char. If no such char is found,
628: * then search for the first ":" char, hoping that what's before that is a
629: * protocol. If found, end the string there. If neither char is found,
630: * then discard the whole base element.
631: */
632: length = strlen (basename) - 1;
633: /* search for the last DIR_SEP char */
1.43 cvs 634: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 635: length--;
636: if (length >= 0)
637: /* found the last DIR_SEP char, end the string there */
638: basename[length + 1] = EOS;
639: else
640: /* search for the first PATH_STR char */
641: {
1.32 cvs 642: for (length = 0; basename[length] != ':' &&
1.31 cvs 643: basename[length] != EOS; length ++);
1.32 cvs 644: if (basename[length] == ':')
1.31 cvs 645: /* found, so end the string there */
646: basename[length + 1] = EOS;
647: else
648: /* not found, discard the base */
649: basename[0] = EOS;
650: }
651: return (basename);
652: }
653:
654:
1.4 cvs 655: /*----------------------------------------------------------------------
1.40 cvs 656: GetLocalPath
657: Allocate and return the local document path associated to the url
658: ----------------------------------------------------------------------*/
659: #ifdef __STDC__
660: char *GetLocalPath (Document doc, char *url)
661: #else /* __STDC__ */
662: char *GetLocalPath (doc, url)
663: Document doc;
664: char *url;
665: #endif /* __STDC__ */
666: {
1.46 cvs 667: char *ptr, *n;
1.40 cvs 668: char *documentname;
1.41 cvs 669: char url_sep;
1.40 cvs 670: int len;
671: boolean noFile;
672:
673: if (url != NULL)
674: {
675: /* check whether the file name exists */
676: len = strlen (url) - 1;
1.41 cvs 677: if (IsW3Path)
678: url_sep = '/';
679: else
680: url_sep = DIR_SEP;
681: noFile = (url[len] == url_sep);
1.40 cvs 682: if (noFile)
683: url[len] = EOS;
1.47 cvs 684: ptr = TtaGetMemory (MAX_LENGTH);
685: documentname = TtaGetMemory (MAX_LENGTH);
1.40 cvs 686: TtaExtractName (url, ptr, documentname);
687: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
688: if (!TtaCheckDirectory (ptr))
689: /* directory did not exist */
1.59 cvs 690: # ifdef _WINDOWS
691: _mkdir (ptr);
692: # else /* !_WINDOWS */
1.40 cvs 693: mkdir (ptr, S_IRWXU);
1.59 cvs 694: # endif /* !_WINDOWS */
1.47 cvs 695:
696: /* don't include the query string within document name */
697: n = strrchr(documentname, '?');
698: if (n != NULL)
699: *n = EOS;
1.46 cvs 700: /* don't include ':' within document name */
701: n = strchr (documentname, ':');
702: if (n != NULL)
703: *n = EOS;
1.40 cvs 704: strcat (ptr, documentname);
705: TtaFreeMemory (documentname);
706: /* restore the url */
707: if (noFile)
1.41 cvs 708: url[len] = url_sep;
1.40 cvs 709: return (ptr);
710: }
711: else
712: return (NULL);
713: }
714:
715:
716: /*----------------------------------------------------------------------
1.9 cvs 717: NormalizeURL
718: normalizes orgName according to a base associated with doc, and
719: following the standard URL format rules.
1.53 cvs 720: if doc is 0 and otherPath not NULL, normalizes orgName according to this
721: other path.
1.9 cvs 722: The function returns the new complete and normalized URL
1.12 cvs 723: or file name path (newName) and the name of the document (docName).
1.9 cvs 724: N.B. If the function can't find out what's the docName, it assigns
725: the name "noname.html".
1.4 cvs 726: ----------------------------------------------------------------------*/
1.3 cvs 727: #ifdef __STDC__
1.53 cvs 728: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName, char *otherPath)
1.3 cvs 729: #else /* __STDC__ */
1.53 cvs 730: void NormalizeURL (orgName, doc, newName, docName, otherPath)
1.3 cvs 731: char *orgName;
732: Document doc;
733: char *newName;
734: char *docName;
1.53 cvs 735: char *otherPath;
1.3 cvs 736: #endif /* __STDC__ */
737: {
1.31 cvs 738: char *basename;
1.18 cvs 739: char tempOrgName[MAX_LENGTH];
1.5 cvs 740: char *ptr;
1.49 cvs 741: char used_sep;
1.5 cvs 742: int length;
1.49 cvs 743: boolean check;
1.5 cvs 744:
1.44 cvs 745: # ifdef _WINDOWS
746: int ndx;
747: # endif /* _WINDOWS */
748:
1.5 cvs 749: if (!newName || !docName)
750: return;
1.18 cvs 751:
1.32 cvs 752: if (doc != 0)
1.53 cvs 753: basename = GetBaseURL (doc);
754: else if (otherPath != NULL)
755: basename = TtaStrdup (otherPath);
1.32 cvs 756: else
1.53 cvs 757: basename = NULL;
1.32 cvs 758:
1.18 cvs 759: /*
1.31 cvs 760: * Clean orgName
761: * Make sure we have a complete orgName, without any leading or trailing
762: * white spaces, or trailinbg new lines
763: */
1.5 cvs 764: ptr = orgName;
1.18 cvs 765: /* skip leading white space and new line characters */
1.19 cvs 766: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS);
1.39 cvs 767: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
768: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 769: /*
1.31 cvs 770: * Make orgName a complete URL
771: * If the URL does not include a protocol, then try to calculate
772: * one using the doc's base element (if it exists),
773: */
1.53 cvs 774: if (tempOrgName[0] == EOS)
775: {
776: newName[0] = EOS;
777: TtaFreeMemory (basename);
778: return;
779: }
1.49 cvs 780:
781: /* clean trailing white space */
782: length = strlen (tempOrgName) - 1;
1.53 cvs 783: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
784: {
785: tempOrgName[length] = EOS;
786: length--;
787: }
1.50 cvs 788:
1.55 cvs 789: /* remove extra dot (which dot???) */
790: /* ugly, but faster than a strcmp */
791: if (tempOrgName[length] == '.'
792: && (length == 0 || tempOrgName[length-1] != '.'))
793: tempOrgName[length] = EOS;
1.50 cvs 794:
1.53 cvs 795: if (IsW3Path (tempOrgName))
796: {
797: /* the name is complete, go to the Sixth Step */
798: strcpy (newName, tempOrgName);
799: SimplifyUrl (&newName);
800: /* verify if the URL has the form "protocol://server:port" */
801: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
802: if (ptr && !strcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
1.50 cvs 803: strcat (newName, URL_STR);
1.49 cvs 804:
1.53 cvs 805: if (ptr)
1.50 cvs 806: TtaFreeMemory (ptr);
1.53 cvs 807: }
808: else if ( basename == NULL)
809: /* the name is complete, go to the Sixth Step */
1.18 cvs 810: strcpy (newName, tempOrgName);
1.53 cvs 811: else
812: {
1.31 cvs 813: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 814: # ifdef _WINDOWS
1.53 cvs 815: if (!IsW3Path (basename))
816: {
817: length = strlen (tempOrgName);
818: for (ndx = 0; ndx < length; ndx++)
819: if (tempOrgName [ndx] == '/')
820: tempOrgName [ndx] = '\\';
821: }
1.44 cvs 822: # endif /* _WINDOWS */
1.25 cvs 823: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 824: if (ptr)
825: {
826: SimplifyUrl (&ptr);
827: strcpy (newName, ptr);
828: TtaFreeMemory (ptr);
829: }
830: else
831: newName[0] = EOS;
832: }
1.36 cvs 833:
834: TtaFreeMemory (basename);
1.18 cvs 835: /*
1.31 cvs 836: * Prepare the docname that will refer to this ressource in the
837: * .amaya directory. If the new URL finishes on DIR_SEP, then use
838: * noname.html as a default ressource name
1.18 cvs 839: */
1.53 cvs 840: if (newName[0] != EOS)
841: {
842: length = strlen (newName) - 1;
843: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
844: {
845: used_sep = newName[length];
846: check = TRUE;
847: while (check)
848: {
1.50 cvs 849: length--;
850: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 851: length--;
852: if (!strncmp (&newName[length+1], "..", 2))
853: {
854: newName[length+1] = EOS;
855: /* remove also previous directory */
856: length--;
857: while (length >= 0 && newName[length] != used_sep)
858: length--;
859: if (strncmp (&newName[length+1], "//", 2))
860: /* don't remove server name */
1.50 cvs 861: newName[length+1] = EOS;
1.53 cvs 862: }
863: else if (!strncmp (&newName[length+1], ".", 1))
864: newName[length+1] = EOS;
1.50 cvs 865: else
1.53 cvs 866: check = FALSE;
867: }
868: strcpy (docName, "noname.html");
869: /* docname was not comprised inside the URL, so let's */
870: /* assign the default ressource name */
871: strcpy (docName, "noname.html");
872: }
873: else
874: { /* docname is comprised inside the URL */
1.50 cvs 875: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP)
1.53 cvs 876: length--;
877: if (length < 0)
1.50 cvs 878: strcpy (docName, newName);
1.53 cvs 879: else
880: strcpy (docName, &newName[length+1]);
881: }
882: }
883: else
884: docName[0] = EOS;
1.18 cvs 885: }
1.3 cvs 886:
1.4 cvs 887: /*----------------------------------------------------------------------
1.9 cvs 888: IsSameHost
1.4 cvs 889: ----------------------------------------------------------------------*/
1.3 cvs 890: #ifdef __STDC__
1.34 cvs 891: boolean IsSameHost (const char *url1, const char *url2)
1.3 cvs 892: #else /* __STDC__ */
893: boolean IsSameHost (url1, url2)
1.34 cvs 894: const char *url1;
895: const char *url2;
1.3 cvs 896: #endif /* __STDC__ */
897: {
1.34 cvs 898: char *basename_ptr1, *basename_ptr2;
899: boolean result;
1.3 cvs 900:
1.25 cvs 901: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
902: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 903:
1.5 cvs 904: if (strcmp (basename_ptr1, basename_ptr2))
1.8 cvs 905: result = FALSE;
1.5 cvs 906: else
1.8 cvs 907: result = TRUE;
1.3 cvs 908:
1.25 cvs 909: TtaFreeMemory (basename_ptr1);
910: TtaFreeMemory (basename_ptr2);
1.5 cvs 911: return (result);
1.3 cvs 912: }
913:
914:
1.4 cvs 915: /*----------------------------------------------------------------------
1.22 cvs 916: HasKnownFileSuffix
917: returns TRUE if path points to a file ending with a suffix.
918: ----------------------------------------------------------------------*/
919: #ifdef __STDC__
1.34 cvs 920: boolean HasKnownFileSuffix (const char *path)
1.22 cvs 921: #else /* __STDC__ */
922: boolean HasKnownFileSuffix (path)
1.34 cvs 923: const char *path;
1.22 cvs 924: #endif /* __STDC__ */
925: {
1.29 cvs 926: char *root;
927: char temppath[MAX_LENGTH];
928: char suffix[MAX_LENGTH];
1.22 cvs 929:
1.24 cvs 930: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 931: return (FALSE);
932:
1.29 cvs 933: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 934:
935: if (root)
936: {
937: strcpy (temppath, root);
1.25 cvs 938: TtaFreeMemory (root);
1.22 cvs 939: /* Get the suffix */
940: ExtractSuffix (temppath, suffix);
941:
942: if( suffix[0] == EOS)
943: /* no suffix */
944: return (FALSE);
945:
946: /* Normalize the suffix */
947: ConvertToLowerCase (suffix);
948:
1.23 cvs 949: if (!strcmp (suffix, "gz"))
1.22 cvs 950: /* skip the compressed suffix */
951: {
952: ExtractSuffix (temppath, suffix);
953: if(suffix[0] == EOS)
954: /* no suffix */
955: return (FALSE);
956: /* Normalize the suffix */
957: ConvertToLowerCase (suffix);
958: }
959:
960: if ((strcmp (suffix, "gif")) && (strcmp (suffix, "xbm")) &&
961: (strcmp (suffix, "xpm")) && (strcmp (suffix, "jpg")) &&
962: (strcmp (suffix, "pdf")) && (strcmp (suffix, "png")) &&
963: (strcmp (suffix, "tgz")) && (strcmp (suffix, "xpg")) &&
964: (strcmp (suffix, "xpd")) && (strcmp (suffix, "ps")) &&
965: (strcmp (suffix, "au")) && (strcmp (suffix, "html")) &&
966: (strcmp (suffix, "htm")) && (strcmp (suffix, "shtml")) &&
967: (strcmp (suffix, "txt")) && (strcmp (suffix, "css")) &&
968: (strcmp (suffix, "eps")))
969: return (FALSE);
970: else
971: return (TRUE);
972: }
973: else
974: return (FALSE);
975: }
976:
977:
978: /*----------------------------------------------------------------------
1.24 cvs 979: ChopURL
980: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
981: If inputURL is bigger than that size, outputURL receives
982: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
983: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
984: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
985: copied into outputURL.
986: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
987: chars.
988: ----------------------------------------------------------------------*/
989: #ifdef __STDC__
1.34 cvs 990: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 991: #else
992: void ChopURL (outputURL, inputURL)
1.34 cvs 993: char *outputURL;
994: const char *inputURL;
1.24 cvs 995: #endif
1.22 cvs 996:
1.24 cvs 997: {
998: int len;
1.9 cvs 999:
1.24 cvs 1000: len = strlen (inputURL);
1001: if (len <= MAX_PRINT_URL_LENGTH)
1.29 cvs 1002: strcpy (outputURL, inputURL);
1.24 cvs 1003: else
1004: /* make a truncated urlName on the status window */
1005: {
1006: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1007: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1008: strcat (outputURL, "...");
1009: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1010: }
1.25 cvs 1011: }
1012:
1013:
1014: /*----------------------------------------------------------------------
1015: scan
1.47 cvs 1016: Scan a filename for its constituents
1.25 cvs 1017: -----------------------------------
1018:
1019: On entry,
1020: name points to a document name which may be incomplete.
1021: On exit,
1022: absolute or relative may be nonzero (but not both).
1023: host, fragment and access may be nonzero if they were specified.
1024: Any which are nonzero point to zero terminated strings.
1025: ----------------------------------------------------------------------*/
1026: #ifdef __STDC__
1027: static void scan (char * name, HTURI * parts)
1028: #else /* __STDC__ */
1029: static void scan (name, parts)
1030: char *name;
1031: HTURI *parts;
1032:
1033: #endif /* __STDC__ */
1034: {
1.43 cvs 1035: char *p;
1036: char *after_access = name;
1.32 cvs 1037:
1.43 cvs 1038: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1039: /* Look for fragment identifier */
1.52 cvs 1040: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1041: {
1042: *p++ = '\0';
1043: parts->fragment = p;
1.25 cvs 1044: }
1045:
1.28 cvs 1046: for (p=name; *p; p++)
1047: {
1.43 cvs 1048: if (*p == URL_SEP || *p == DIR_SEP || *p=='#' || *p=='?')
1.28 cvs 1049: break;
1.32 cvs 1050: if (*p==':')
1.28 cvs 1051: {
1052: *p = 0;
1053: parts->access = after_access; /* Scheme has been specified */
1054:
1055: /* The combination of gcc, the "-O" flag and the HP platform is
1056: unhealthy. The following three lines is a quick & dirty fix, but is
1057: not recommended. Rather, turn off "-O". */
1058:
1059: /* after_access = p;*/
1060: /* while (*after_access == 0)*/
1061: /* after_access++;*/
1062: after_access = p+1;
1.43 cvs 1063: if (!strcasecmp("URL", parts->access))
1.28 cvs 1064: /* Ignore IETF's URL: pre-prefix */
1065: parts->access = NULL;
1066: else
1.25 cvs 1067: break;
1068: }
1069: }
1070:
1071: p = after_access;
1.43 cvs 1072: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1073: {
1.43 cvs 1074: if (p[1] == URL_SEP)
1.28 cvs 1075: {
1.25 cvs 1076: parts->host = p+2; /* host has been specified */
1.28 cvs 1077: *p = 0; /* Terminate access */
1078: /* look for end of host name if any */
1.43 cvs 1079: p = strchr (parts->host, URL_SEP);
1.28 cvs 1080: if (p)
1081: {
1.43 cvs 1082: *p = EOS; /* Terminate host */
1.25 cvs 1083: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1084: }
1085: }
1086: else
1087: /* Root found but no host */
1088: parts->absolute = p+1;
1089: }
1090: else
1091: {
1.25 cvs 1092: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1093: }
1.25 cvs 1094: }
1095:
1096:
1097: /*----------------------------------------------------------------------
1.28 cvs 1098: AmayaParseUrl: parse a Name relative to another name
1099:
1100: This returns those parts of a name which are given (and requested)
1101: substituting bits from the related name where necessary.
1.25 cvs 1102:
1.28 cvs 1103: On entry,
1.25 cvs 1104: aName A filename given
1105: relatedName A name relative to which aName is to be parsed. Give
1106: it an empty string if aName is absolute.
1107: wanted A mask for the bits which are wanted.
1108:
1.28 cvs 1109: On exit,
1.25 cvs 1110: returns A pointer to a malloc'd string which MUST BE FREED
1111: ----------------------------------------------------------------------*/
1112: #ifdef __STDC__
1.34 cvs 1113: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1.25 cvs 1114: #else /* __STDC__ */
1.28 cvs 1115: char *AmayaParseUrl (aName, relatedName, wanted)
1.34 cvs 1116: const char *aName;
1.28 cvs 1117: char *relatedName;
1118: int wanted;
1.25 cvs 1119:
1120: #endif /* __STDC__ */
1121: {
1.29 cvs 1122: char *return_value;
1123: char result[MAX_LENGTH];
1124: char name[MAX_LENGTH];
1125: char rel[MAX_LENGTH];
1126: char *p, *access;
1127: HTURI given, related;
1128: int len;
1.33 cvs 1129: char used_sep;
1130: char *used_str;
1.32 cvs 1131:
1.43 cvs 1132: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1133: {
1.42 cvs 1134: used_str = DIR_STR;
1135: used_sep = DIR_SEP;
1.33 cvs 1136: }
1.32 cvs 1137: else
1.33 cvs 1138: {
1.42 cvs 1139: used_str = URL_STR;
1140: used_sep = URL_SEP;
1.33 cvs 1141: }
1.32 cvs 1142:
1.29 cvs 1143: /* Make working copies of input strings to cut up: */
1144: return_value = NULL;
1145: result[0] = 0; /* Clear string */
1146: strcpy (name, aName);
1147: if (relatedName != NULL)
1148: strcpy (rel, relatedName);
1149: else
1150: relatedName[0] = EOS;
1151:
1152: scan (name, &given);
1153: scan (rel, &related);
1154: access = given.access ? given.access : related.access;
1155: if (wanted & AMAYA_PARSE_ACCESS)
1156: if (access)
1157: {
1158: strcat (result, access);
1159: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.32 cvs 1160: strcat (result, ":");
1.29 cvs 1161: }
1162:
1163: if (given.access && related.access)
1164: /* If different, inherit nothing. */
1165: if (strcmp (given.access, related.access) != 0)
1166: {
1167: related.host = 0;
1168: related.absolute = 0;
1169: related.relative = 0;
1170: related.fragment = 0;
1171: }
1172:
1173: if (wanted & AMAYA_PARSE_HOST)
1174: if(given.host || related.host)
1175: {
1176: if(wanted & AMAYA_PARSE_PUNCTUATION)
1177: strcat (result, "//");
1178: strcat (result, given.host ? given.host : related.host);
1179: }
1180:
1181: if (given.host && related.host)
1182: /* If different hosts, inherit no path. */
1183: if (strcmp(given.host, related.host) != 0)
1184: {
1185: related.absolute = 0;
1186: related.relative = 0;
1187: related.fragment = 0;
1188: }
1189:
1190: if (wanted & AMAYA_PARSE_PATH)
1191: {
1192: if (given.absolute)
1193: {
1194: /* All is given */
1195: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.33 cvs 1196: strcat (result, used_str);
1.29 cvs 1197: strcat (result, given.absolute);
1.25 cvs 1198: }
1.29 cvs 1199: else if (related.absolute)
1200: {
1201: /* Adopt path not name */
1.33 cvs 1202: strcat (result, used_str);
1.29 cvs 1203: strcat (result, related.absolute);
1204: if (given.relative)
1205: {
1206: /* Search part? */
1207: p = strchr (result, '?');
1208: if (!p)
1209: p=result+strlen(result)-1;
1.33 cvs 1210: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1211: /* Remove filename */
1212: p[1]=0;
1213: /* Add given one */
1214: strcat (result, given.relative);
1.25 cvs 1215: }
1216: }
1.29 cvs 1217: else if (given.relative)
1218: /* what we've got */
1219: strcat (result, given.relative);
1220: else if (related.relative)
1221: strcat (result, related.relative);
1222: else
1223: /* No inheritance */
1.33 cvs 1224: strcat (result, used_str);
1.25 cvs 1225: }
1.29 cvs 1226:
1227: if (wanted & AMAYA_PARSE_ANCHOR)
1228: if (given.fragment || related.fragment)
1229: {
1230: if (given.absolute && given.fragment)
1231: {
1232: /*Fixes for relURLs...*/
1233: if (wanted & AMAYA_PARSE_PUNCTUATION)
1234: strcat (result, "#");
1235: strcat (result, given.fragment);
1236: }
1237: else if (!(given.absolute) && !(given.fragment))
1238: strcat (result, "");
1239: else
1240: {
1241: if (wanted & AMAYA_PARSE_PUNCTUATION)
1242: strcat (result, "#");
1243: strcat (result, given.fragment ? given.fragment : related.fragment);
1244: }
1245: }
1246: len = strlen (result);
1247: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1248: strcpy (return_value, result);
1249: return (return_value); /* exactly the right length */
1.25 cvs 1250: }
1251:
1252: /*----------------------------------------------------------------------
1253: HTCanon
1254: Canonicalizes the URL in the following manner starting from the host
1255: pointer:
1256:
1257: 1) The host name is converted to lowercase
1258: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1259:
1260: Return: OK The position of the current path part of the URL
1261: which might be the old one or a new one.
1262:
1263: ----------------------------------------------------------------------*/
1264: #ifdef __STDC__
1.28 cvs 1265: static char *HTCanon (char ** filename, char * host)
1.25 cvs 1266: #else /* __STDC__ */
1.28 cvs 1267: static char *HTCanon (filename, host)
1268: char **filename;
1269: char *host;
1.25 cvs 1270: #endif /* __STDC__ */
1271: {
1272: char *newname = NULL;
1273: char *port;
1274: char *strptr;
1275: char *path;
1276: char *access = host-3;
1.33 cvs 1277: char used_sep;
1.32 cvs 1278:
1279:
1.33 cvs 1280: if (*filename && strchr (*filename, URL_SEP))
1281: {
1282: used_sep = URL_SEP;
1283: }
1284: else
1285: {
1286: used_sep = DIR_SEP;
1287: }
1.32 cvs 1288:
1.33 cvs 1289: while (access>*filename && *(access-1)!= used_sep) /* Find access method */
1.25 cvs 1290: access--;
1.33 cvs 1291: if ((path = strchr(host, used_sep)) == NULL) /* Find path */
1.25 cvs 1292: path = host + strlen(host);
1293: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */
1294: host = strptr;
1.32 cvs 1295: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */
1.25 cvs 1296: port = NULL;
1297:
1298: strptr = host; /* Convert to lower-case */
1.33 cvs 1299: while (strptr<path)
1300: {
1.25 cvs 1301: *strptr = tolower(*strptr);
1302: strptr++;
1.33 cvs 1303: }
1.25 cvs 1304:
1305: /* Does the URL contain a full domain name? This also works for a
1306: numerical host name. The domain name is already made lower-case
1307: and without a trailing dot. */
1308: {
1.33 cvs 1309: char *dot = port ? port : path;
1310: if (dot > *filename && *--dot=='.')
1311: {
1312: char *orig=dot, *dest=dot+1;
1313: while((*orig++ = *dest++));
1314: if (port) port--;
1315: path--;
1.25 cvs 1316: }
1317: }
1318: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1319: if (port)
1320: {
1321: if (!*(port+1) || *(port+1)==used_sep)
1322: {
1323: if (!newname)
1324: {
1.25 cvs 1325: char *orig=port, *dest=port+1;
1326: while((*orig++ = *dest++));
1.33 cvs 1327: }
1328: }
1329: else if ((!strncmp(access, "http", 4) &&
1330: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1331: (!strncmp(access, "gopher", 6) &&
1332: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1333: (!strncmp(access, "ftp", 3) &&
1334: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) {
1335: if (!newname)
1336: {
1337: char *orig=port, *dest=port+3;
1338: while((*orig++ = *dest++));
1339: /* Update path position, Henry Minsky */
1340: path -= 3;
1.25 cvs 1341: }
1.33 cvs 1342: }
1343: else if (newname)
1344: strncat(newname, port, (int) (path-port));
1345: }
1.25 cvs 1346:
1.33 cvs 1347: if (newname)
1348: {
1.25 cvs 1349: char *newpath = newname+strlen(newname);
1350: strcat(newname, path);
1351: path = newpath;
1.28 cvs 1352: /* Free old copy */
1353: TtaFreeMemory(*filename);
1.25 cvs 1354: *filename = newname;
1.33 cvs 1355: }
1.25 cvs 1356: return path;
1357: }
1358:
1359:
1360: /*----------------------------------------------------------------------
1.29 cvs 1361: SimplifyUrl: simplify a URI
1.32 cvs 1362: A URI is allowed to contain the sequence xxx/../ which may be
1363: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1364: Simplification helps us recognize duplicate URIs.
1.25 cvs 1365:
1.28 cvs 1366: Thus, /etc/junk/../fred becomes /etc/fred
1367: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1368:
1.28 cvs 1369: but we should NOT change
1370: http://fred.xxx.edu/../..
1.25 cvs 1371:
1372: or ../../albert.html
1373:
1.28 cvs 1374: In order to avoid empty URLs the following URLs become:
1.25 cvs 1375:
1376: /fred/.. becomes /fred/..
1377: /fred/././.. becomes /fred/..
1378: /fred/.././junk/.././ becomes /fred/..
1379:
1.28 cvs 1380: If more than one set of `://' is found (several proxies in cascade) then
1381: only the part after the last `://' is simplified.
1.25 cvs 1382:
1.28 cvs 1383: Returns: A string which might be the old one or a new one.
1.25 cvs 1384: ----------------------------------------------------------------------*/
1385: #ifdef __STDC__
1.29 cvs 1386: void SimplifyUrl (char ** url)
1.25 cvs 1387: #else /* __STDC__ */
1.29 cvs 1388: void SimplifyUrl (url)
1.28 cvs 1389: char **url;
1.25 cvs 1390: #endif /* __STDC__ */
1391: {
1.28 cvs 1392: char *path, *p;
1393: char *newptr, *access;
1394: char *orig, *dest, *end;
1395:
1.33 cvs 1396: char used_sep;
1.32 cvs 1397:
1398:
1.28 cvs 1399: if (!url || !*url)
1400: return;
1401:
1.33 cvs 1402: if (strchr (*url, URL_SEP))
1403: {
1404: used_sep = URL_SEP;
1405: }
1.32 cvs 1406: else
1.33 cvs 1407: {
1408: used_sep = DIR_SEP;
1409: }
1.32 cvs 1410:
1.28 cvs 1411: /* Find any scheme name */
1412: if ((path = strstr(*url, "://")) != NULL)
1.33 cvs 1413: {
1414: /* Find host name */
1.28 cvs 1415: access = *url;
1416: while (access<path && (*access=tolower(*access)))
1417: access++;
1418: path += 3;
1419: while ((newptr = strstr(path, "://")) != NULL)
1420: /* For proxies */
1421: path = newptr+3;
1422: /* We have a host name */
1423: path = HTCanon(url, path);
1.25 cvs 1424: }
1.28 cvs 1425: else if ((path = strstr(*url, ":/")) != NULL)
1426: path += 2;
1427: else
1428: path = *url;
1.25 cvs 1429:
1.33 cvs 1430: if (*path == used_sep && *(path+1)==used_sep)
1.28 cvs 1431: /* Some URLs start //<foo> */
1432: path += 1;
1433: else if (!strncmp(path, "news:", 5))
1434: {
1435: newptr = strchr(path+5, '@');
1436: if (!newptr)
1437: newptr = path + 5;
1438: while (*newptr)
1439: {
1440: /* Make group or host lower case */
1441: *newptr = tolower (*newptr);
1442: newptr++;
1.25 cvs 1443: }
1.28 cvs 1444: /* Doesn't need to do any more */
1445: return;
1.25 cvs 1446: }
1.28 cvs 1447:
1448: if ((p = path))
1449: {
1450: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1451: (end = strchr (path, '#'))))
1452: end = path + strlen (path);
1453:
1454: /* Parse string second time to simplify */
1455: p = path;
1456: while (p < end)
1457: {
1.33 cvs 1458: if (*p==used_sep)
1.28 cvs 1459: {
1.33 cvs 1460: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1461: {
1462: orig = p + 1;
1.33 cvs 1463: dest = (*(p+2)!=used_sep) ? p+2 : p+3;
1.52 cvs 1464: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1465: end = orig - 1;
1466: }
1.33 cvs 1467: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3)))
1.28 cvs 1468: {
1469: newptr = p;
1.52 cvs 1470: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1471: if (*newptr == used_sep)
1472: orig = newptr + 1;
1.28 cvs 1473: else
1.52 cvs 1474: orig = newptr;
1475:
1476: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1477: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1478: end = orig-1;
1479: /* Start again with prev slash */
1480: p = newptr;
1.28 cvs 1481: }
1.33 cvs 1482: else if (*(p+1) == used_sep)
1.28 cvs 1483: {
1.33 cvs 1484: while (*(p+1) == used_sep)
1.28 cvs 1485: {
1486: orig = p;
1487: dest = p + 1;
1488: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1489: end = orig-1;
1490: }
1491: }
1492: else
1.25 cvs 1493: p++;
1.28 cvs 1494: }
1495: else
1496: p++;
1.25 cvs 1497: }
1498: }
1.51 cvs 1499:
1500: /*
1501: ** Check for host/../.. kind of things
1502: */
1.52 cvs 1503: if (*path==used_sep && *(path+1)=='.' && *(path+2)=='.' && (!*(path+3) || *(path+3)==used_sep))
1.51 cvs 1504: *(path+1) = EOS;
1505:
1.28 cvs 1506: return;
1507: }
1508:
1509:
1510: /*----------------------------------------------------------------------
1511: NormalizeFile normalizes local names.
1512: Return TRUE if target and src differ.
1513: ----------------------------------------------------------------------*/
1514: #ifdef __STDC__
1515: boolean NormalizeFile (char *src, char *target)
1516: #else
1517: boolean NormalizeFile (src, target)
1518: char *src;
1519: char *target;
1520:
1521: #endif
1522: {
1523: char *s;
1524: boolean change;
1525:
1.54 cvs 1526: change = FALSE;
1.53 cvs 1527: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1528: {
1529: /* remove the prefix file: */
1530: if (src[5] == EOS)
1531: strcpy (target, DIR_STR);
1532: else if (src[0] == '~')
1533: {
1534: /* replace ~ */
1535: s = (char *) TtaGetEnvString ("HOME");
1536: strcpy (target, s);
1537: strcat (target, &src[5]);
1538: }
1539: else
1540: strcpy (target, &src[5]);
1.54 cvs 1541: change = TRUE;
1.28 cvs 1542: }
1.53 cvs 1543: # ifndef _WINDOWS
1544: else if (src[0] == '~')
1545: {
1546: /* replace ~ */
1547: s = (char *) TtaGetEnvString ("HOME");
1548: strcpy (target, s);
1549: if (src[1] != DIR_SEP)
1550: strcat (target, DIR_STR);
1551: strcat (target, &src[1]);
1.54 cvs 1552: change = TRUE;
1.53 cvs 1553: }
1554: # endif /* _WINDOWS */
1.28 cvs 1555: else
1556: strcpy (target, src);
1557:
1558: /* remove /../ and /./ */
1.29 cvs 1559: SimplifyUrl (&target);
1.54 cvs 1560: if (!change)
1561: change = strcmp (src, target);
1.28 cvs 1562: return (change);
1.25 cvs 1563: }
1564:
1.28 cvs 1565:
1.25 cvs 1566: /*----------------------------------------------------------------------
1.31 cvs 1567: MakeRelativeURL: make relative name
1.25 cvs 1568:
1.28 cvs 1569: This function creates and returns a string which gives an expression of
1570: one address as related to another. Where there is no relation, an absolute
1571: address is retured.
1.25 cvs 1572:
1.28 cvs 1573: On entry,
1.25 cvs 1574: Both names must be absolute, fully qualified names of nodes
1575: (no fragment bits)
1576:
1.28 cvs 1577: On exit,
1.25 cvs 1578: The return result points to a newly allocated name which, if
1579: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1580: The caller is responsible for freeing the resulting name later.
1581: ----------------------------------------------------------------------*/
1582: #ifdef __STDC__
1.31 cvs 1583: char *MakeRelativeURL (char *aName, char *relatedName)
1.25 cvs 1584: #else /* __STDC__ */
1.31 cvs 1585: char *MakeRelativeURL (aName, relatedName)
1.28 cvs 1586: char *aName;
1587: char *relatedName;
1.25 cvs 1588: #endif /* __STDC__ */
1589: {
1.39 cvs 1590: char *return_value;
1591: char result[MAX_LENGTH];
1.29 cvs 1592: char *p;
1.34 cvs 1593: char *q;
1.29 cvs 1594: char *after_access;
1595: char *last_slash = NULL;
1596: int slashes, levels, len;
1597:
1.44 cvs 1598: # ifdef _WINDOWS
1599: int ndx;
1600: # endif /* _WINDOWS */
1601:
1.29 cvs 1602: if (aName == NULL || relatedName == NULL)
1603: return (NULL);
1604:
1605: slashes = 0;
1606: after_access = NULL;
1607: p = aName;
1608: q = relatedName;
1609: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1610: {
1611: /* Find extent of match */
1.32 cvs 1612: if (*p == ':')
1.29 cvs 1613: after_access = p + 1;
1.28 cvs 1614: if (*p == DIR_SEP)
1.27 cvs 1615: {
1.29 cvs 1616: /* memorize the last slash position and count them */
1.27 cvs 1617: last_slash = p;
1618: slashes++;
1.25 cvs 1619: }
1620: }
1621:
1.31 cvs 1622: /* q, p point to the first non-matching character or zero */
1623: if (*q == EOS)
1624: {
1625: /* New name is a subset of the related name */
1626: /* exactly the right length */
1627: len = strlen (p);
1628: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1629: strcpy (return_value, p);
1630: }
1631: else if ((slashes < 2 && after_access == NULL)
1632: || (slashes < 3 && after_access != NULL))
1633: {
1634: /* Two names whitout common path */
1635: /* exactly the right length */
1636: len = strlen (aName);
1637: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1638: strcpy (return_value, aName);
1639: }
1640: else
1641: {
1642: /* Some path in common */
1643: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1644: /* just the same server */
1645: strcpy (result, last_slash);
1646: else
1647: {
1648: levels= 0;
1.52 cvs 1649: for (; *q && *q != '#' && *q!=';' && *q!='?'; q++)
1.31 cvs 1650: if (*q == DIR_SEP)
1651: levels++;
1652:
1.52 cvs 1653: result[0] = EOS;
1.31 cvs 1654: for (;levels; levels--)
1655: strcat (result, "../");
1656: strcat (result, last_slash+1);
1657: }
1.52 cvs 1658:
1659: if (!*result)
1660: strcat (result, "./");
1661:
1.31 cvs 1662: /* exactly the right length */
1663: len = strlen (result);
1664: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1665: strcpy (return_value, result);
1.52 cvs 1666:
1.25 cvs 1667: }
1.44 cvs 1668: # ifdef _WINDOWS
1669: len = strlen (return_value);
1670: for (ndx = 0; ndx < len; ndx ++)
1671: if (return_value[ndx] == '\\')
1672: return_value[ndx] = '/' ;
1673: # endif /* _WINDOWS */
1.29 cvs 1674: return (return_value);
1.24 cvs 1675: }
1.35 cvs 1676:
1677:
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