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