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