Annotation of Amaya/amaya/AHTURLTools.c, revision 1.58
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 */
609: mkdir (ptr, S_IRWXU);
1.47 cvs 610:
611: /* don't include the query string within document name */
612: n = strrchr(documentname, '?');
613: if (n != NULL)
614: *n = EOS;
1.46 cvs 615: /* don't include ':' within document name */
616: n = strchr (documentname, ':');
617: if (n != NULL)
618: *n = EOS;
1.40 cvs 619: strcat (ptr, documentname);
620: TtaFreeMemory (documentname);
621: /* restore the url */
622: if (noFile)
1.41 cvs 623: url[len] = url_sep;
1.40 cvs 624: return (ptr);
625: }
626: else
627: return (NULL);
628: }
629:
630:
631: /*----------------------------------------------------------------------
1.9 cvs 632: NormalizeURL
633: normalizes orgName according to a base associated with doc, and
634: following the standard URL format rules.
1.53 cvs 635: if doc is 0 and otherPath not NULL, normalizes orgName according to this
636: other path.
1.9 cvs 637: The function returns the new complete and normalized URL
1.12 cvs 638: or file name path (newName) and the name of the document (docName).
1.9 cvs 639: N.B. If the function can't find out what's the docName, it assigns
640: the name "noname.html".
1.4 cvs 641: ----------------------------------------------------------------------*/
1.3 cvs 642: #ifdef __STDC__
1.53 cvs 643: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName, char *otherPath)
1.3 cvs 644: #else /* __STDC__ */
1.53 cvs 645: void NormalizeURL (orgName, doc, newName, docName, otherPath)
1.3 cvs 646: char *orgName;
647: Document doc;
648: char *newName;
649: char *docName;
1.53 cvs 650: char *otherPath;
1.3 cvs 651: #endif /* __STDC__ */
652: {
1.31 cvs 653: char *basename;
1.18 cvs 654: char tempOrgName[MAX_LENGTH];
1.5 cvs 655: char *ptr;
1.49 cvs 656: char used_sep;
1.5 cvs 657: int length;
1.49 cvs 658: boolean check;
1.5 cvs 659:
1.44 cvs 660: # ifdef _WINDOWS
661: int ndx;
662: # endif /* _WINDOWS */
663:
1.5 cvs 664: if (!newName || !docName)
665: return;
1.18 cvs 666:
1.32 cvs 667: if (doc != 0)
1.53 cvs 668: basename = GetBaseURL (doc);
669: else if (otherPath != NULL)
670: basename = TtaStrdup (otherPath);
1.32 cvs 671: else
1.53 cvs 672: basename = NULL;
1.32 cvs 673:
1.18 cvs 674: /*
1.31 cvs 675: * Clean orgName
676: * Make sure we have a complete orgName, without any leading or trailing
677: * white spaces, or trailinbg new lines
678: */
1.5 cvs 679: ptr = orgName;
1.18 cvs 680: /* skip leading white space and new line characters */
1.19 cvs 681: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS);
1.39 cvs 682: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
683: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 684: /*
1.31 cvs 685: * Make orgName a complete URL
686: * If the URL does not include a protocol, then try to calculate
687: * one using the doc's base element (if it exists),
688: */
1.53 cvs 689: if (tempOrgName[0] == EOS)
690: {
691: newName[0] = EOS;
692: TtaFreeMemory (basename);
693: return;
694: }
1.49 cvs 695:
696: /* clean trailing white space */
697: length = strlen (tempOrgName) - 1;
1.53 cvs 698: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
699: {
700: tempOrgName[length] = EOS;
701: length--;
702: }
1.50 cvs 703:
1.55 cvs 704: /* remove extra dot (which dot???) */
705: /* ugly, but faster than a strcmp */
706: if (tempOrgName[length] == '.'
707: && (length == 0 || tempOrgName[length-1] != '.'))
708: tempOrgName[length] = EOS;
1.50 cvs 709:
1.53 cvs 710: if (IsW3Path (tempOrgName))
711: {
712: /* the name is complete, go to the Sixth Step */
713: strcpy (newName, tempOrgName);
714: SimplifyUrl (&newName);
715: /* verify if the URL has the form "protocol://server:port" */
716: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
717: if (ptr && !strcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
1.50 cvs 718: strcat (newName, URL_STR);
1.49 cvs 719:
1.53 cvs 720: if (ptr)
1.50 cvs 721: TtaFreeMemory (ptr);
1.53 cvs 722: }
723: else if ( basename == NULL)
724: /* the name is complete, go to the Sixth Step */
1.18 cvs 725: strcpy (newName, tempOrgName);
1.53 cvs 726: else
727: {
1.31 cvs 728: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 729: # ifdef _WINDOWS
1.53 cvs 730: if (!IsW3Path (basename))
731: {
732: length = strlen (tempOrgName);
733: for (ndx = 0; ndx < length; ndx++)
734: if (tempOrgName [ndx] == '/')
735: tempOrgName [ndx] = '\\';
736: }
1.44 cvs 737: # endif /* _WINDOWS */
1.25 cvs 738: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 739: if (ptr)
740: {
741: SimplifyUrl (&ptr);
742: strcpy (newName, ptr);
743: TtaFreeMemory (ptr);
744: }
745: else
746: newName[0] = EOS;
747: }
1.36 cvs 748:
749: TtaFreeMemory (basename);
1.18 cvs 750: /*
1.31 cvs 751: * Prepare the docname that will refer to this ressource in the
752: * .amaya directory. If the new URL finishes on DIR_SEP, then use
753: * noname.html as a default ressource name
1.18 cvs 754: */
1.53 cvs 755: if (newName[0] != EOS)
756: {
757: length = strlen (newName) - 1;
758: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
759: {
760: used_sep = newName[length];
761: check = TRUE;
762: while (check)
763: {
1.50 cvs 764: length--;
765: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 766: length--;
767: if (!strncmp (&newName[length+1], "..", 2))
768: {
769: newName[length+1] = EOS;
770: /* remove also previous directory */
771: length--;
772: while (length >= 0 && newName[length] != used_sep)
773: length--;
774: if (strncmp (&newName[length+1], "//", 2))
775: /* don't remove server name */
1.50 cvs 776: newName[length+1] = EOS;
1.53 cvs 777: }
778: else if (!strncmp (&newName[length+1], ".", 1))
779: newName[length+1] = EOS;
1.50 cvs 780: else
1.53 cvs 781: check = FALSE;
782: }
783: strcpy (docName, "noname.html");
784: /* docname was not comprised inside the URL, so let's */
785: /* assign the default ressource name */
786: strcpy (docName, "noname.html");
787: }
788: else
789: { /* docname is comprised inside the URL */
1.50 cvs 790: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP)
1.53 cvs 791: length--;
792: if (length < 0)
1.50 cvs 793: strcpy (docName, newName);
1.53 cvs 794: else
795: strcpy (docName, &newName[length+1]);
796: }
797: }
798: else
799: docName[0] = EOS;
1.18 cvs 800: }
1.3 cvs 801:
1.4 cvs 802: /*----------------------------------------------------------------------
1.9 cvs 803: IsSameHost
1.4 cvs 804: ----------------------------------------------------------------------*/
1.3 cvs 805: #ifdef __STDC__
1.34 cvs 806: boolean IsSameHost (const char *url1, const char *url2)
1.3 cvs 807: #else /* __STDC__ */
808: boolean IsSameHost (url1, url2)
1.34 cvs 809: const char *url1;
810: const char *url2;
1.3 cvs 811: #endif /* __STDC__ */
812: {
1.34 cvs 813: char *basename_ptr1, *basename_ptr2;
814: boolean result;
1.3 cvs 815:
1.25 cvs 816: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
817: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 818:
1.5 cvs 819: if (strcmp (basename_ptr1, basename_ptr2))
1.8 cvs 820: result = FALSE;
1.5 cvs 821: else
1.8 cvs 822: result = TRUE;
1.3 cvs 823:
1.25 cvs 824: TtaFreeMemory (basename_ptr1);
825: TtaFreeMemory (basename_ptr2);
1.5 cvs 826: return (result);
1.3 cvs 827: }
828:
829:
1.4 cvs 830: /*----------------------------------------------------------------------
1.22 cvs 831: HasKnownFileSuffix
832: returns TRUE if path points to a file ending with a suffix.
833: ----------------------------------------------------------------------*/
834: #ifdef __STDC__
1.34 cvs 835: boolean HasKnownFileSuffix (const char *path)
1.22 cvs 836: #else /* __STDC__ */
837: boolean HasKnownFileSuffix (path)
1.34 cvs 838: const char *path;
1.22 cvs 839: #endif /* __STDC__ */
840: {
1.29 cvs 841: char *root;
842: char temppath[MAX_LENGTH];
843: char suffix[MAX_LENGTH];
1.22 cvs 844:
1.24 cvs 845: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 846: return (FALSE);
847:
1.29 cvs 848: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 849:
850: if (root)
851: {
852: strcpy (temppath, root);
1.25 cvs 853: TtaFreeMemory (root);
1.22 cvs 854: /* Get the suffix */
855: ExtractSuffix (temppath, suffix);
856:
857: if( suffix[0] == EOS)
858: /* no suffix */
859: return (FALSE);
860:
861: /* Normalize the suffix */
862: ConvertToLowerCase (suffix);
863:
1.23 cvs 864: if (!strcmp (suffix, "gz"))
1.22 cvs 865: /* skip the compressed suffix */
866: {
867: ExtractSuffix (temppath, suffix);
868: if(suffix[0] == EOS)
869: /* no suffix */
870: return (FALSE);
871: /* Normalize the suffix */
872: ConvertToLowerCase (suffix);
873: }
874:
875: if ((strcmp (suffix, "gif")) && (strcmp (suffix, "xbm")) &&
876: (strcmp (suffix, "xpm")) && (strcmp (suffix, "jpg")) &&
877: (strcmp (suffix, "pdf")) && (strcmp (suffix, "png")) &&
878: (strcmp (suffix, "tgz")) && (strcmp (suffix, "xpg")) &&
879: (strcmp (suffix, "xpd")) && (strcmp (suffix, "ps")) &&
880: (strcmp (suffix, "au")) && (strcmp (suffix, "html")) &&
881: (strcmp (suffix, "htm")) && (strcmp (suffix, "shtml")) &&
882: (strcmp (suffix, "txt")) && (strcmp (suffix, "css")) &&
883: (strcmp (suffix, "eps")))
884: return (FALSE);
885: else
886: return (TRUE);
887: }
888: else
889: return (FALSE);
890: }
891:
892:
893: /*----------------------------------------------------------------------
1.24 cvs 894: ChopURL
895: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
896: If inputURL is bigger than that size, outputURL receives
897: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
898: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
899: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
900: copied into outputURL.
901: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
902: chars.
903: ----------------------------------------------------------------------*/
904: #ifdef __STDC__
1.34 cvs 905: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 906: #else
907: void ChopURL (outputURL, inputURL)
1.34 cvs 908: char *outputURL;
909: const char *inputURL;
1.24 cvs 910: #endif
1.22 cvs 911:
1.24 cvs 912: {
913: int len;
1.9 cvs 914:
1.24 cvs 915: len = strlen (inputURL);
916: if (len <= MAX_PRINT_URL_LENGTH)
1.29 cvs 917: strcpy (outputURL, inputURL);
1.24 cvs 918: else
919: /* make a truncated urlName on the status window */
920: {
921: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
922: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
923: strcat (outputURL, "...");
924: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
925: }
1.25 cvs 926: }
927:
928:
929: /*----------------------------------------------------------------------
930: scan
1.47 cvs 931: Scan a filename for its constituents
1.25 cvs 932: -----------------------------------
933:
934: On entry,
935: name points to a document name which may be incomplete.
936: On exit,
937: absolute or relative may be nonzero (but not both).
938: host, fragment and access may be nonzero if they were specified.
939: Any which are nonzero point to zero terminated strings.
940: ----------------------------------------------------------------------*/
941: #ifdef __STDC__
942: static void scan (char * name, HTURI * parts)
943: #else /* __STDC__ */
944: static void scan (name, parts)
945: char *name;
946: HTURI *parts;
947:
948: #endif /* __STDC__ */
949: {
1.43 cvs 950: char *p;
951: char *after_access = name;
1.32 cvs 952:
1.43 cvs 953: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 954: /* Look for fragment identifier */
1.52 cvs 955: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 956: {
957: *p++ = '\0';
958: parts->fragment = p;
1.25 cvs 959: }
960:
1.28 cvs 961: for (p=name; *p; p++)
962: {
1.43 cvs 963: if (*p == URL_SEP || *p == DIR_SEP || *p=='#' || *p=='?')
1.28 cvs 964: break;
1.32 cvs 965: if (*p==':')
1.28 cvs 966: {
967: *p = 0;
968: parts->access = after_access; /* Scheme has been specified */
969:
970: /* The combination of gcc, the "-O" flag and the HP platform is
971: unhealthy. The following three lines is a quick & dirty fix, but is
972: not recommended. Rather, turn off "-O". */
973:
974: /* after_access = p;*/
975: /* while (*after_access == 0)*/
976: /* after_access++;*/
977: after_access = p+1;
1.43 cvs 978: if (!strcasecmp("URL", parts->access))
1.28 cvs 979: /* Ignore IETF's URL: pre-prefix */
980: parts->access = NULL;
981: else
1.25 cvs 982: break;
983: }
984: }
985:
986: p = after_access;
1.43 cvs 987: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 988: {
1.43 cvs 989: if (p[1] == URL_SEP)
1.28 cvs 990: {
1.25 cvs 991: parts->host = p+2; /* host has been specified */
1.28 cvs 992: *p = 0; /* Terminate access */
993: /* look for end of host name if any */
1.43 cvs 994: p = strchr (parts->host, URL_SEP);
1.28 cvs 995: if (p)
996: {
1.43 cvs 997: *p = EOS; /* Terminate host */
1.25 cvs 998: parts->absolute = p+1; /* Root has been found */
1.28 cvs 999: }
1000: }
1001: else
1002: /* Root found but no host */
1003: parts->absolute = p+1;
1004: }
1005: else
1006: {
1.25 cvs 1007: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1008: }
1.25 cvs 1009: }
1010:
1011:
1012: /*----------------------------------------------------------------------
1.28 cvs 1013: AmayaParseUrl: parse a Name relative to another name
1014:
1015: This returns those parts of a name which are given (and requested)
1016: substituting bits from the related name where necessary.
1.25 cvs 1017:
1.28 cvs 1018: On entry,
1.25 cvs 1019: aName A filename given
1020: relatedName A name relative to which aName is to be parsed. Give
1021: it an empty string if aName is absolute.
1022: wanted A mask for the bits which are wanted.
1023:
1.28 cvs 1024: On exit,
1.25 cvs 1025: returns A pointer to a malloc'd string which MUST BE FREED
1026: ----------------------------------------------------------------------*/
1027: #ifdef __STDC__
1.34 cvs 1028: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1.25 cvs 1029: #else /* __STDC__ */
1.28 cvs 1030: char *AmayaParseUrl (aName, relatedName, wanted)
1.34 cvs 1031: const char *aName;
1.28 cvs 1032: char *relatedName;
1033: int wanted;
1.25 cvs 1034:
1035: #endif /* __STDC__ */
1036: {
1.29 cvs 1037: char *return_value;
1038: char result[MAX_LENGTH];
1039: char name[MAX_LENGTH];
1040: char rel[MAX_LENGTH];
1041: char *p, *access;
1042: HTURI given, related;
1043: int len;
1.33 cvs 1044: char used_sep;
1045: char *used_str;
1.32 cvs 1046:
1.43 cvs 1047: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1048: {
1.42 cvs 1049: used_str = DIR_STR;
1050: used_sep = DIR_SEP;
1.33 cvs 1051: }
1.32 cvs 1052: else
1.33 cvs 1053: {
1.42 cvs 1054: used_str = URL_STR;
1055: used_sep = URL_SEP;
1.33 cvs 1056: }
1.32 cvs 1057:
1.29 cvs 1058: /* Make working copies of input strings to cut up: */
1059: return_value = NULL;
1060: result[0] = 0; /* Clear string */
1061: strcpy (name, aName);
1062: if (relatedName != NULL)
1063: strcpy (rel, relatedName);
1064: else
1065: relatedName[0] = EOS;
1066:
1067: scan (name, &given);
1068: scan (rel, &related);
1069: access = given.access ? given.access : related.access;
1070: if (wanted & AMAYA_PARSE_ACCESS)
1071: if (access)
1072: {
1073: strcat (result, access);
1074: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.32 cvs 1075: strcat (result, ":");
1.29 cvs 1076: }
1077:
1078: if (given.access && related.access)
1079: /* If different, inherit nothing. */
1080: if (strcmp (given.access, related.access) != 0)
1081: {
1082: related.host = 0;
1083: related.absolute = 0;
1084: related.relative = 0;
1085: related.fragment = 0;
1086: }
1087:
1088: if (wanted & AMAYA_PARSE_HOST)
1089: if(given.host || related.host)
1090: {
1091: if(wanted & AMAYA_PARSE_PUNCTUATION)
1092: strcat (result, "//");
1093: strcat (result, given.host ? given.host : related.host);
1094: }
1095:
1096: if (given.host && related.host)
1097: /* If different hosts, inherit no path. */
1098: if (strcmp(given.host, related.host) != 0)
1099: {
1100: related.absolute = 0;
1101: related.relative = 0;
1102: related.fragment = 0;
1103: }
1104:
1105: if (wanted & AMAYA_PARSE_PATH)
1106: {
1107: if (given.absolute)
1108: {
1109: /* All is given */
1110: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.33 cvs 1111: strcat (result, used_str);
1.29 cvs 1112: strcat (result, given.absolute);
1.25 cvs 1113: }
1.29 cvs 1114: else if (related.absolute)
1115: {
1116: /* Adopt path not name */
1.33 cvs 1117: strcat (result, used_str);
1.29 cvs 1118: strcat (result, related.absolute);
1119: if (given.relative)
1120: {
1121: /* Search part? */
1122: p = strchr (result, '?');
1123: if (!p)
1124: p=result+strlen(result)-1;
1.33 cvs 1125: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1126: /* Remove filename */
1127: p[1]=0;
1128: /* Add given one */
1129: strcat (result, given.relative);
1.25 cvs 1130: }
1131: }
1.29 cvs 1132: else if (given.relative)
1133: /* what we've got */
1134: strcat (result, given.relative);
1135: else if (related.relative)
1136: strcat (result, related.relative);
1137: else
1138: /* No inheritance */
1.33 cvs 1139: strcat (result, used_str);
1.25 cvs 1140: }
1.29 cvs 1141:
1142: if (wanted & AMAYA_PARSE_ANCHOR)
1143: if (given.fragment || related.fragment)
1144: {
1145: if (given.absolute && given.fragment)
1146: {
1147: /*Fixes for relURLs...*/
1148: if (wanted & AMAYA_PARSE_PUNCTUATION)
1149: strcat (result, "#");
1150: strcat (result, given.fragment);
1151: }
1152: else if (!(given.absolute) && !(given.fragment))
1153: strcat (result, "");
1154: else
1155: {
1156: if (wanted & AMAYA_PARSE_PUNCTUATION)
1157: strcat (result, "#");
1158: strcat (result, given.fragment ? given.fragment : related.fragment);
1159: }
1160: }
1161: len = strlen (result);
1162: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1163: strcpy (return_value, result);
1164: return (return_value); /* exactly the right length */
1.25 cvs 1165: }
1166:
1167: /*----------------------------------------------------------------------
1168: HTCanon
1169: Canonicalizes the URL in the following manner starting from the host
1170: pointer:
1171:
1172: 1) The host name is converted to lowercase
1173: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1174:
1175: Return: OK The position of the current path part of the URL
1176: which might be the old one or a new one.
1177:
1178: ----------------------------------------------------------------------*/
1179: #ifdef __STDC__
1.28 cvs 1180: static char *HTCanon (char ** filename, char * host)
1.25 cvs 1181: #else /* __STDC__ */
1.28 cvs 1182: static char *HTCanon (filename, host)
1183: char **filename;
1184: char *host;
1.25 cvs 1185: #endif /* __STDC__ */
1186: {
1187: char *newname = NULL;
1188: char *port;
1189: char *strptr;
1190: char *path;
1191: char *access = host-3;
1.33 cvs 1192: char used_sep;
1.32 cvs 1193:
1194:
1.33 cvs 1195: if (*filename && strchr (*filename, URL_SEP))
1196: {
1197: used_sep = URL_SEP;
1198: }
1199: else
1200: {
1201: used_sep = DIR_SEP;
1202: }
1.32 cvs 1203:
1.33 cvs 1204: while (access>*filename && *(access-1)!= used_sep) /* Find access method */
1.25 cvs 1205: access--;
1.33 cvs 1206: if ((path = strchr(host, used_sep)) == NULL) /* Find path */
1.25 cvs 1207: path = host + strlen(host);
1208: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */
1209: host = strptr;
1.32 cvs 1210: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */
1.25 cvs 1211: port = NULL;
1212:
1213: strptr = host; /* Convert to lower-case */
1.33 cvs 1214: while (strptr<path)
1215: {
1.25 cvs 1216: *strptr = tolower(*strptr);
1217: strptr++;
1.33 cvs 1218: }
1.25 cvs 1219:
1220: /* Does the URL contain a full domain name? This also works for a
1221: numerical host name. The domain name is already made lower-case
1222: and without a trailing dot. */
1223: {
1.33 cvs 1224: char *dot = port ? port : path;
1225: if (dot > *filename && *--dot=='.')
1226: {
1227: char *orig=dot, *dest=dot+1;
1228: while((*orig++ = *dest++));
1229: if (port) port--;
1230: path--;
1.25 cvs 1231: }
1232: }
1233: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1234: if (port)
1235: {
1236: if (!*(port+1) || *(port+1)==used_sep)
1237: {
1238: if (!newname)
1239: {
1.25 cvs 1240: char *orig=port, *dest=port+1;
1241: while((*orig++ = *dest++));
1.33 cvs 1242: }
1243: }
1244: else if ((!strncmp(access, "http", 4) &&
1245: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1246: (!strncmp(access, "gopher", 6) &&
1247: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1248: (!strncmp(access, "ftp", 3) &&
1249: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) {
1250: if (!newname)
1251: {
1252: char *orig=port, *dest=port+3;
1253: while((*orig++ = *dest++));
1254: /* Update path position, Henry Minsky */
1255: path -= 3;
1.25 cvs 1256: }
1.33 cvs 1257: }
1258: else if (newname)
1259: strncat(newname, port, (int) (path-port));
1260: }
1.25 cvs 1261:
1.33 cvs 1262: if (newname)
1263: {
1.25 cvs 1264: char *newpath = newname+strlen(newname);
1265: strcat(newname, path);
1266: path = newpath;
1.28 cvs 1267: /* Free old copy */
1268: TtaFreeMemory(*filename);
1.25 cvs 1269: *filename = newname;
1.33 cvs 1270: }
1.25 cvs 1271: return path;
1272: }
1273:
1274:
1275: /*----------------------------------------------------------------------
1.29 cvs 1276: SimplifyUrl: simplify a URI
1.32 cvs 1277: A URI is allowed to contain the sequence xxx/../ which may be
1278: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1279: Simplification helps us recognize duplicate URIs.
1.25 cvs 1280:
1.28 cvs 1281: Thus, /etc/junk/../fred becomes /etc/fred
1282: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1283:
1.28 cvs 1284: but we should NOT change
1285: http://fred.xxx.edu/../..
1.25 cvs 1286:
1287: or ../../albert.html
1288:
1.28 cvs 1289: In order to avoid empty URLs the following URLs become:
1.25 cvs 1290:
1291: /fred/.. becomes /fred/..
1292: /fred/././.. becomes /fred/..
1293: /fred/.././junk/.././ becomes /fred/..
1294:
1.28 cvs 1295: If more than one set of `://' is found (several proxies in cascade) then
1296: only the part after the last `://' is simplified.
1.25 cvs 1297:
1.28 cvs 1298: Returns: A string which might be the old one or a new one.
1.25 cvs 1299: ----------------------------------------------------------------------*/
1300: #ifdef __STDC__
1.29 cvs 1301: void SimplifyUrl (char ** url)
1.25 cvs 1302: #else /* __STDC__ */
1.29 cvs 1303: void SimplifyUrl (url)
1.28 cvs 1304: char **url;
1.25 cvs 1305: #endif /* __STDC__ */
1306: {
1.28 cvs 1307: char *path, *p;
1308: char *newptr, *access;
1309: char *orig, *dest, *end;
1310:
1.33 cvs 1311: char used_sep;
1.32 cvs 1312:
1313:
1.28 cvs 1314: if (!url || !*url)
1315: return;
1316:
1.33 cvs 1317: if (strchr (*url, URL_SEP))
1318: {
1319: used_sep = URL_SEP;
1320: }
1.32 cvs 1321: else
1.33 cvs 1322: {
1323: used_sep = DIR_SEP;
1324: }
1.32 cvs 1325:
1.28 cvs 1326: /* Find any scheme name */
1327: if ((path = strstr(*url, "://")) != NULL)
1.33 cvs 1328: {
1329: /* Find host name */
1.28 cvs 1330: access = *url;
1331: while (access<path && (*access=tolower(*access)))
1332: access++;
1333: path += 3;
1334: while ((newptr = strstr(path, "://")) != NULL)
1335: /* For proxies */
1336: path = newptr+3;
1337: /* We have a host name */
1338: path = HTCanon(url, path);
1.25 cvs 1339: }
1.28 cvs 1340: else if ((path = strstr(*url, ":/")) != NULL)
1341: path += 2;
1342: else
1343: path = *url;
1.25 cvs 1344:
1.33 cvs 1345: if (*path == used_sep && *(path+1)==used_sep)
1.28 cvs 1346: /* Some URLs start //<foo> */
1347: path += 1;
1348: else if (!strncmp(path, "news:", 5))
1349: {
1350: newptr = strchr(path+5, '@');
1351: if (!newptr)
1352: newptr = path + 5;
1353: while (*newptr)
1354: {
1355: /* Make group or host lower case */
1356: *newptr = tolower (*newptr);
1357: newptr++;
1.25 cvs 1358: }
1.28 cvs 1359: /* Doesn't need to do any more */
1360: return;
1.25 cvs 1361: }
1.28 cvs 1362:
1363: if ((p = path))
1364: {
1365: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1366: (end = strchr (path, '#'))))
1367: end = path + strlen (path);
1368:
1369: /* Parse string second time to simplify */
1370: p = path;
1371: while (p < end)
1372: {
1.33 cvs 1373: if (*p==used_sep)
1.28 cvs 1374: {
1.33 cvs 1375: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1376: {
1377: orig = p + 1;
1.33 cvs 1378: dest = (*(p+2)!=used_sep) ? p+2 : p+3;
1.52 cvs 1379: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1380: end = orig - 1;
1381: }
1.33 cvs 1382: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3)))
1.28 cvs 1383: {
1384: newptr = p;
1.52 cvs 1385: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1386: if (*newptr == used_sep)
1387: orig = newptr + 1;
1.28 cvs 1388: else
1.52 cvs 1389: orig = newptr;
1390:
1391: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1392: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1393: end = orig-1;
1394: /* Start again with prev slash */
1395: p = newptr;
1.28 cvs 1396: }
1.33 cvs 1397: else if (*(p+1) == used_sep)
1.28 cvs 1398: {
1.33 cvs 1399: while (*(p+1) == used_sep)
1.28 cvs 1400: {
1401: orig = p;
1402: dest = p + 1;
1403: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1404: end = orig-1;
1405: }
1406: }
1407: else
1.25 cvs 1408: p++;
1.28 cvs 1409: }
1410: else
1411: p++;
1.25 cvs 1412: }
1413: }
1.51 cvs 1414:
1415: /*
1416: ** Check for host/../.. kind of things
1417: */
1.52 cvs 1418: if (*path==used_sep && *(path+1)=='.' && *(path+2)=='.' && (!*(path+3) || *(path+3)==used_sep))
1.51 cvs 1419: *(path+1) = EOS;
1420:
1.28 cvs 1421: return;
1422: }
1423:
1424:
1425: /*----------------------------------------------------------------------
1426: NormalizeFile normalizes local names.
1427: Return TRUE if target and src differ.
1428: ----------------------------------------------------------------------*/
1429: #ifdef __STDC__
1430: boolean NormalizeFile (char *src, char *target)
1431: #else
1432: boolean NormalizeFile (src, target)
1433: char *src;
1434: char *target;
1435:
1436: #endif
1437: {
1438: char *s;
1439: boolean change;
1440:
1.54 cvs 1441: change = FALSE;
1.53 cvs 1442: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1443: {
1444: /* remove the prefix file: */
1445: if (src[5] == EOS)
1446: strcpy (target, DIR_STR);
1447: else if (src[0] == '~')
1448: {
1449: /* replace ~ */
1450: s = (char *) TtaGetEnvString ("HOME");
1451: strcpy (target, s);
1452: strcat (target, &src[5]);
1453: }
1454: else
1455: strcpy (target, &src[5]);
1.54 cvs 1456: change = TRUE;
1.28 cvs 1457: }
1.53 cvs 1458: # ifndef _WINDOWS
1459: else if (src[0] == '~')
1460: {
1461: /* replace ~ */
1462: s = (char *) TtaGetEnvString ("HOME");
1463: strcpy (target, s);
1464: if (src[1] != DIR_SEP)
1465: strcat (target, DIR_STR);
1466: strcat (target, &src[1]);
1.54 cvs 1467: change = TRUE;
1.53 cvs 1468: }
1469: # endif /* _WINDOWS */
1.28 cvs 1470: else
1471: strcpy (target, src);
1472:
1473: /* remove /../ and /./ */
1.29 cvs 1474: SimplifyUrl (&target);
1.54 cvs 1475: if (!change)
1476: change = strcmp (src, target);
1.28 cvs 1477: return (change);
1.25 cvs 1478: }
1479:
1.28 cvs 1480:
1.25 cvs 1481: /*----------------------------------------------------------------------
1.31 cvs 1482: MakeRelativeURL: make relative name
1.25 cvs 1483:
1.28 cvs 1484: This function creates and returns a string which gives an expression of
1485: one address as related to another. Where there is no relation, an absolute
1486: address is retured.
1.25 cvs 1487:
1.28 cvs 1488: On entry,
1.25 cvs 1489: Both names must be absolute, fully qualified names of nodes
1490: (no fragment bits)
1491:
1.28 cvs 1492: On exit,
1.25 cvs 1493: The return result points to a newly allocated name which, if
1494: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1495: The caller is responsible for freeing the resulting name later.
1496: ----------------------------------------------------------------------*/
1497: #ifdef __STDC__
1.31 cvs 1498: char *MakeRelativeURL (char *aName, char *relatedName)
1.25 cvs 1499: #else /* __STDC__ */
1.31 cvs 1500: char *MakeRelativeURL (aName, relatedName)
1.28 cvs 1501: char *aName;
1502: char *relatedName;
1.25 cvs 1503: #endif /* __STDC__ */
1504: {
1.39 cvs 1505: char *return_value;
1506: char result[MAX_LENGTH];
1.29 cvs 1507: char *p;
1.34 cvs 1508: char *q;
1.29 cvs 1509: char *after_access;
1510: char *last_slash = NULL;
1511: int slashes, levels, len;
1512:
1.44 cvs 1513: # ifdef _WINDOWS
1514: int ndx;
1515: # endif /* _WINDOWS */
1516:
1.29 cvs 1517: if (aName == NULL || relatedName == NULL)
1518: return (NULL);
1519:
1520: slashes = 0;
1521: after_access = NULL;
1522: p = aName;
1523: q = relatedName;
1524: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1525: {
1526: /* Find extent of match */
1.32 cvs 1527: if (*p == ':')
1.29 cvs 1528: after_access = p + 1;
1.28 cvs 1529: if (*p == DIR_SEP)
1.27 cvs 1530: {
1.29 cvs 1531: /* memorize the last slash position and count them */
1.27 cvs 1532: last_slash = p;
1533: slashes++;
1.25 cvs 1534: }
1535: }
1536:
1.31 cvs 1537: /* q, p point to the first non-matching character or zero */
1538: if (*q == EOS)
1539: {
1540: /* New name is a subset of the related name */
1541: /* exactly the right length */
1542: len = strlen (p);
1543: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1544: strcpy (return_value, p);
1545: }
1546: else if ((slashes < 2 && after_access == NULL)
1547: || (slashes < 3 && after_access != NULL))
1548: {
1549: /* Two names whitout common path */
1550: /* exactly the right length */
1551: len = strlen (aName);
1552: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1553: strcpy (return_value, aName);
1554: }
1555: else
1556: {
1557: /* Some path in common */
1558: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1559: /* just the same server */
1560: strcpy (result, last_slash);
1561: else
1562: {
1563: levels= 0;
1.52 cvs 1564: for (; *q && *q != '#' && *q!=';' && *q!='?'; q++)
1.31 cvs 1565: if (*q == DIR_SEP)
1566: levels++;
1567:
1.52 cvs 1568: result[0] = EOS;
1.31 cvs 1569: for (;levels; levels--)
1570: strcat (result, "../");
1571: strcat (result, last_slash+1);
1572: }
1.52 cvs 1573:
1574: if (!*result)
1575: strcat (result, "./");
1576:
1.31 cvs 1577: /* exactly the right length */
1578: len = strlen (result);
1579: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1580: strcpy (return_value, result);
1.52 cvs 1581:
1.25 cvs 1582: }
1.44 cvs 1583: # ifdef _WINDOWS
1584: len = strlen (return_value);
1585: for (ndx = 0; ndx < len; ndx ++)
1586: if (return_value[ndx] == '\\')
1587: return_value[ndx] = '/' ;
1588: # endif /* _WINDOWS */
1.29 cvs 1589: return (return_value);
1.24 cvs 1590: }
1.35 cvs 1591:
1592:
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