Annotation of Amaya/amaya/AHTURLTools.c, revision 1.43
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
14: *
15: */
1.7 cvs 16:
1.15 cvs 17: #define THOT_EXPORT extern
1.3 cvs 18: #include "amaya.h"
19:
1.8 cvs 20: #include "init_f.h"
21: #include "AHTURLTools_f.h"
22:
1.24 cvs 23: #define MAX_PRINT_URL_LENGTH 50
1.29 cvs 24: typedef struct _HTURI {
25: char * access; /* Now known as "scheme" */
26: char * host;
27: char * absolute;
28: char * relative;
29: char * fragment;
30: } HTURI;
1.24 cvs 31:
1.28 cvs 32:
33: /*----------------------------------------------------------------------
34: ConvertToLowerCase
35: Converts a string to lowercase.
36: ----------------------------------------------------------------------*/
1.22 cvs 37: #ifdef __STDC__
1.38 cvs 38: void ConvertToLowerCase (char *string)
1.28 cvs 39: #else /* __STDC__ */
1.38 cvs 40: void ConvertToLowerCase (string)
1.28 cvs 41: char *string;
42:
43: #endif /* __STDC__ */
44: {
45: int i;
46:
47: if (!string)
48: return;
49:
50: for (i = 0; string[i] != EOS; i++)
51: string[i] = tolower (string[i]);
52: }
1.22 cvs 53:
1.8 cvs 54: /*----------------------------------------------------------------------
1.11 cvs 55: ExplodeURL
1.8 cvs 56: ----------------------------------------------------------------------*/
57: #ifdef __STDC__
58: void ExplodeURL (char *url, char **proto, char **host, char **dir, char **file)
59: #else
60: void ExplodeURL (url, proto, host, dir, file)
61: char *url;
62: char **proto;
63: char **host;
64: char **dir;
65: char **file;
66:
67: #endif
68: {
1.33 cvs 69: char *curr, *temp;
70: char used_sep;
1.32 cvs 71:
1.33 cvs 72: if (url && strchr (url, URL_SEP))
73: used_sep = URL_SEP;
74: else
75: used_sep = DIR_SEP;
1.8 cvs 76:
77: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
78: (dir == NULL) || (file == NULL))
79: return;
80:
81: /* initialize every pointer */
82: *proto = *host = *dir = *file = NULL;
83:
84: /* skip any leading space */
85: while ((*url == SPACE) || (*url == TAB))
86: url++;
1.9 cvs 87: curr = url;
88: if (*curr == 0)
1.8 cvs 89: goto finished;
90:
91: /* go to the end of the URL */
1.9 cvs 92: while ((*curr != 0) && (*curr != SPACE) && (*curr != '\b') &&
93: (*curr != '\r') && (*curr != EOL))
94: curr++;
1.8 cvs 95:
96: /* mark the end of the chain */
1.9 cvs 97: *curr = EOS;
98: curr--;
99: if (curr <= url)
1.8 cvs 100: goto finished;
101:
102: /* search the next DIR_SEP indicating the beginning of the file name */
103: do
1.11 cvs 104: curr--;
1.33 cvs 105: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 106:
1.9 cvs 107: if (curr < url)
1.8 cvs 108: goto finished;
1.9 cvs 109: *file = curr + 1;
1.8 cvs 110:
111: /* mark the end of the dir */
1.9 cvs 112: *curr = EOS;
113: curr--;
114: if (curr < url)
1.8 cvs 115: goto finished;
116:
1.29 cvs 117: /* search for the DIR_STR indicating the host name start */
1.33 cvs 118: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 119: curr--;
1.8 cvs 120:
121: /* if we found it, separate the host name from the directory */
1.33 cvs 122: if ((*curr == DIR_SEP) && (*(curr + 1) == used_sep))
1.8 cvs 123: {
1.9 cvs 124: *host = temp = curr + 2;
1.33 cvs 125: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 126: temp++;
1.33 cvs 127: if (*temp == used_sep)
1.8 cvs 128: {
129: *temp = EOS;
130: *dir = temp + 1;
131: }
132: }
133: else
1.11 cvs 134: *dir = curr;
135:
1.9 cvs 136: if (curr <= url)
1.8 cvs 137: goto finished;
138:
139: /* mark the end of the proto */
1.9 cvs 140: *curr = EOS;
141: curr--;
142: if (curr < url)
1.8 cvs 143: goto finished;
144:
1.32 cvs 145: if (*curr == ':')
1.8 cvs 146: {
1.9 cvs 147: *curr = EOS;
148: curr--;
1.8 cvs 149: }
150: else
151: goto finished;
1.11 cvs 152:
1.9 cvs 153: if (curr < url)
1.8 cvs 154: goto finished;
1.9 cvs 155: while ((curr > url) && (isalpha (*curr)))
156: curr--;
157: *proto = curr;
1.8 cvs 158:
159: finished:;
160:
161: #ifdef AMAYA_DEBUG
162: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
163: if (*proto)
164: fprintf (stderr, "proto : %s, ", *proto);
165: if (*host)
166: fprintf (stderr, "host : %s, ", *host);
167: if (*dir)
168: fprintf (stderr, "dir : %s, ", *dir);
169: if (*file)
170: fprintf (stderr, "file : %s ", *file);
171: fprintf (stderr, "\n");
172: #endif
173:
174: }
1.3 cvs 175:
1.4 cvs 176: /*----------------------------------------------------------------------
1.9 cvs 177: IsHTMLName
178: returns TRUE if path points to an HTML resource.
1.4 cvs 179: ----------------------------------------------------------------------*/
1.3 cvs 180: #ifdef __STDC__
1.34 cvs 181: boolean IsHTMLName (const char *path)
1.3 cvs 182: #else /* __STDC__ */
183: boolean IsHTMLName (path)
1.34 cvs 184: const char *path;
1.3 cvs 185: #endif /* __STDC__ */
186: {
1.5 cvs 187: char temppath[MAX_LENGTH];
188: char suffix[MAX_LENGTH];
189: char nsuffix[MAX_LENGTH];
190: int i;
191:
192: if (!path)
1.37 cvs 193: return (FALSE);
1.5 cvs 194:
195: strcpy (temppath, path);
196: ExtractSuffix (temppath, suffix);
197:
198: /* Normalize the suffix */
199: i = 0;
1.39 cvs 200: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 201: {
1.25 cvs 202: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 203: i++;
204: }
1.5 cvs 205: nsuffix[i] = EOS;
1.39 cvs 206: if ((!strcmp (nsuffix, "html")) ||
207: (!strcmp (nsuffix, "htm")) ||
208: (!strcmp (nsuffix, "shtml")))
209: return (TRUE);
1.22 cvs 210: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 211: {
1.39 cvs 212: /* take into account compressed files */
1.13 cvs 213: ExtractSuffix (temppath, suffix);
214: /* Normalize the suffix */
215: i = 0;
1.39 cvs 216: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 217: {
1.25 cvs 218: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 219: i++;
220: }
221: nsuffix[i] = EOS;
1.39 cvs 222: if ((!strcmp (nsuffix, "html")) ||
223: (!strcmp (nsuffix, "htm")) ||
224: (!strcmp (nsuffix, "shtml")))
225: return (TRUE);
226: else
1.13 cvs 227: return (FALSE);
228: }
229: else
1.39 cvs 230: return (FALSE);
1.3 cvs 231: }
232:
1.4 cvs 233: /*----------------------------------------------------------------------
1.9 cvs 234: IsImageName
235: returns TRUE if path points to an image resource.
1.4 cvs 236: ----------------------------------------------------------------------*/
1.3 cvs 237: #ifdef __STDC__
1.34 cvs 238: boolean IsImageName (const char *path)
1.3 cvs 239: #else /* __STDC__ */
240: boolean IsImageName (path)
1.34 cvs 241: const char *path;
1.3 cvs 242: #endif /* __STDC__ */
243: {
1.5 cvs 244: char temppath[MAX_LENGTH];
245: char suffix[MAX_LENGTH];
246: char nsuffix[MAX_LENGTH];
247: int i;
248:
249: if (!path)
1.13 cvs 250: return (FALSE);
1.5 cvs 251:
252: strcpy (temppath, path);
253: ExtractSuffix (temppath, suffix);
254:
255: /* Normalize the suffix */
256: i = 0;
1.39 cvs 257: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 258: {
1.25 cvs 259: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 260: i++;
261: }
1.5 cvs 262: nsuffix[i] = EOS;
1.39 cvs 263: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
264: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
265: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
266: return (TRUE);
267: return (FALSE);
1.3 cvs 268: }
269:
1.4 cvs 270: /*----------------------------------------------------------------------
1.9 cvs 271: IsTextName
1.4 cvs 272: ----------------------------------------------------------------------*/
1.3 cvs 273: #ifdef __STDC__
1.34 cvs 274: boolean IsTextName (const char *path)
1.3 cvs 275: #else /* __STDC__ */
276: boolean IsTextName (path)
1.34 cvs 277: const char *path;
1.3 cvs 278:
279: #endif /* __STDC__ */
280: {
1.5 cvs 281: char temppath[MAX_LENGTH];
282: char suffix[MAX_LENGTH];
283: char nsuffix[MAX_LENGTH];
284: int i;
285:
286: if (!path)
1.13 cvs 287: return (FALSE);
1.5 cvs 288:
289: strcpy (temppath, path);
290: ExtractSuffix (temppath, suffix);
291:
292: /* Normalize the suffix */
293: i = 0;
1.39 cvs 294: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 295: {
1.25 cvs 296: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 297: i++;
298: }
299: nsuffix[i] = EOS;
300:
1.39 cvs 301: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 302: return (TRUE);
1.22 cvs 303: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 304: {
1.39 cvs 305: /* take into account compressed files */
1.13 cvs 306: ExtractSuffix (temppath, suffix);
307: /* Normalize the suffix */
308: i = 0;
1.39 cvs 309: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 310: {
1.25 cvs 311: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 312: i++;
313: }
314: nsuffix[i] = EOS;
1.39 cvs 315: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 316: return (TRUE);
317: else
318: return (FALSE);
319: }
320: else
321: return (FALSE);
1.3 cvs 322: }
323:
1.4 cvs 324: /*----------------------------------------------------------------------
1.9 cvs 325: IsHTTPPath
326: returns TRUE if path is in fact an http URL.
1.4 cvs 327: ----------------------------------------------------------------------*/
1.3 cvs 328: #ifdef __STDC__
1.34 cvs 329: boolean IsHTTPPath (const char *path)
1.3 cvs 330: #else /* __STDC__ */
331: boolean IsHTTPPath (path)
1.34 cvs 332: const char *path;
1.3 cvs 333: #endif /* __STDC__ */
334: {
1.5 cvs 335: if (!path)
336: return FALSE;
1.3 cvs 337:
1.5 cvs 338: if (strncmp (path, "http:", 5) != 0)
339: return FALSE;
340: return TRUE;
1.3 cvs 341: }
342:
1.4 cvs 343: /*----------------------------------------------------------------------
1.9 cvs 344: IsWithParameters
345: returns TRUE if url has a concatenated query string.
1.4 cvs 346: ----------------------------------------------------------------------*/
1.3 cvs 347: #ifdef __STDC__
1.34 cvs 348: boolean IsWithParameters (const char *url)
1.3 cvs 349: #else /* __STDC__ */
1.9 cvs 350: boolean IsWithParameters (url)
1.34 cvs 351: const char *url;
1.3 cvs 352: #endif /* __STDC__ */
353: {
1.5 cvs 354: int i;
1.3 cvs 355:
1.9 cvs 356: if ((!url) || (url[0] == EOS))
1.5 cvs 357: return FALSE;
1.3 cvs 358:
1.9 cvs 359: i = strlen (url) - 1;
360: while (i > 0 && url[i--] != '?')
1.5 cvs 361: if (i < 0)
362: return FALSE;
1.3 cvs 363:
1.5 cvs 364: /* There is a parameter */
365: return TRUE;
1.3 cvs 366: }
367:
1.4 cvs 368: /*----------------------------------------------------------------------
1.9 cvs 369: IsW3Path
370: returns TRUE if path is in fact a URL.
1.4 cvs 371: ----------------------------------------------------------------------*/
1.3 cvs 372: #ifdef __STDC__
1.34 cvs 373: boolean IsW3Path (const char *path)
1.3 cvs 374: #else /* __STDC__ */
375: boolean IsW3Path (path)
1.34 cvs 376: const char *path;
1.3 cvs 377: #endif /* __STDC__ */
378: {
1.5 cvs 379: if ((strncmp (path, "http:", 5)) && (strncmp (path, "ftp:", 4)) &&
380: (strncmp (path, "telnet:", 7)) && (strncmp (path, "wais:", 5)) &&
381: (strncmp (path, "news:", 5)) && (strncmp (path, "gopher:", 7)) &&
382: (strncmp (path, "mailto:", 7)) && (strncmp (path, "archie:", 7)))
383: return FALSE;
384: return TRUE;
1.3 cvs 385: }
386:
1.4 cvs 387: /*----------------------------------------------------------------------
1.9 cvs 388: IsValidProtocol
389: returns true if the url protocol is supported by Amaya.
1.4 cvs 390: ----------------------------------------------------------------------*/
1.3 cvs 391: #ifdef __STDC__
1.34 cvs 392: boolean IsValidProtocol (const char *url)
1.3 cvs 393: #else /* __STDC__ */
1.9 cvs 394: boolean IsValidProtocol (url)
1.34 cvs 395: const char *url;
1.3 cvs 396: #endif /* __STDC__ */
397: {
1.26 cvs 398: if (!strncmp (url, "http:", 5))
1.22 cvs 399: /* experimental */
1.26 cvs 400: /*** || !strncmp (url, "ftp:", 4)) ***/
1.24 cvs 401: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 402: return (TRUE);
1.5 cvs 403: else
1.8 cvs 404: return (FALSE);
1.3 cvs 405: }
406:
1.31 cvs 407:
408: /*----------------------------------------------------------------------
409: GetBaseURL
410: normalizes orgName according to a base associated with doc, and
411: following the standard URL format rules.
412: The function returns the base used to solve relative URL and SRC:
413: - the base of the document,
414: - or the document path (without document name).
415: ----------------------------------------------------------------------*/
416: #ifdef __STDC__
417: char *GetBaseURL (Document doc)
418: #else /* __STDC__ */
419: char *GetBaseURL (doc)
420: Document doc;
421: #endif /* __STDC__ */
422: {
423: Element el;
424: ElementType elType;
425: AttributeType attrType;
426: Attribute attr;
427: char *ptr, *basename;
428: int length;
429:
430: basename = TtaGetMemory (MAX_LENGTH);
1.39 cvs 431: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
432: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 433: length = MAX_LENGTH -1;
434: /* get the root element */
435: el = TtaGetMainRoot (doc);
436: /* search the BASE element */
437: elType.ElSSchema = TtaGetDocumentSSchema (doc);
438: elType.ElTypeNum = HTML_EL_BASE;
439: el = TtaSearchTypedElement (elType, SearchInTree, el);
440: if (el)
441: {
442: /* The document has a BASE element -> Get the HREF attribute */
443: attrType.AttrSSchema = elType.ElSSchema;
444: attrType.AttrTypeNum = HTML_ATTR_HREF_;
445: attr = TtaGetAttribute (el, attrType);
446: if (attr)
447: {
448: /* Use the base path of the document */
449: TtaGiveTextAttributeValue (attr, basename, &length);
450: /* base and orgName have to be separated by a DIR_SEP */
451: length--;
1.43 ! cvs 452: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 453: /* verify if the base has the form "protocol://server:port" */
454: {
1.33 cvs 455: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
456: AMAYA_PARSE_HOST |
457: AMAYA_PARSE_PUNCTUATION);
1.31 cvs 458: if (ptr && !strcmp (ptr, basename))
459: {
1.43 ! cvs 460: /* it has this form, we complete it by adding a URL_STR */
! 461: if (strchr (basename, DIR_SEP))
! 462: strcat (basename, DIR_STR);
! 463: else
! 464: strcat (basename, URL_STR);
1.31 cvs 465: length++;
466: }
467: if (ptr)
468: TtaFreeMemory (ptr);
469: }
470: }
1.33 cvs 471: }
472:
1.31 cvs 473: /* Remove anything after the last DIR_SEP char. If no such char is found,
474: * then search for the first ":" char, hoping that what's before that is a
475: * protocol. If found, end the string there. If neither char is found,
476: * then discard the whole base element.
477: */
478: length = strlen (basename) - 1;
479: /* search for the last DIR_SEP char */
1.43 ! cvs 480: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 481: length--;
482: if (length >= 0)
483: /* found the last DIR_SEP char, end the string there */
484: basename[length + 1] = EOS;
485: else
486: /* search for the first PATH_STR char */
487: {
1.32 cvs 488: for (length = 0; basename[length] != ':' &&
1.31 cvs 489: basename[length] != EOS; length ++);
1.32 cvs 490: if (basename[length] == ':')
1.31 cvs 491: /* found, so end the string there */
492: basename[length + 1] = EOS;
493: else
494: /* not found, discard the base */
495: basename[0] = EOS;
496: }
497: return (basename);
498: }
499:
500:
1.4 cvs 501: /*----------------------------------------------------------------------
1.40 cvs 502: GetLocalPath
503: Allocate and return the local document path associated to the url
504: ----------------------------------------------------------------------*/
505: #ifdef __STDC__
506: char *GetLocalPath (Document doc, char *url)
507: #else /* __STDC__ */
508: char *GetLocalPath (doc, url)
509: Document doc;
510: char *url;
511: #endif /* __STDC__ */
512: {
513: char *ptr;
514: char *documentname;
1.41 cvs 515: char url_sep;
1.40 cvs 516: int len;
517: boolean noFile;
518:
519: if (url != NULL)
520: {
521: ptr = TtaGetMemory (MAX_LENGTH);
522: documentname = TtaGetMemory (MAX_LENGTH);
523: /* check whether the file name exists */
524: len = strlen (url) - 1;
1.41 cvs 525: if (IsW3Path)
526: url_sep = '/';
527: else
528: url_sep = DIR_SEP;
529: noFile = (url[len] == url_sep);
1.40 cvs 530: if (noFile)
531: url[len] = EOS;
532: TtaExtractName (url, ptr, documentname);
533: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
534: if (!TtaCheckDirectory (ptr))
535: /* directory did not exist */
536: mkdir (ptr, S_IRWXU);
537: strcat (ptr, documentname);
538: TtaFreeMemory (documentname);
539: /* restore the url */
540: if (noFile)
1.41 cvs 541: url[len] = url_sep;
1.40 cvs 542: return (ptr);
543: }
544: else
545: return (NULL);
546: }
547:
548:
549: /*----------------------------------------------------------------------
1.9 cvs 550: NormalizeURL
551: normalizes orgName according to a base associated with doc, and
552: following the standard URL format rules.
553: The function returns the new complete and normalized URL
1.12 cvs 554: or file name path (newName) and the name of the document (docName).
1.9 cvs 555: N.B. If the function can't find out what's the docName, it assigns
556: the name "noname.html".
1.4 cvs 557: ----------------------------------------------------------------------*/
1.3 cvs 558: #ifdef __STDC__
559: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName)
560: #else /* __STDC__ */
561: void NormalizeURL (orgName, doc, newName, docName)
562: char *orgName;
563: Document doc;
564: char *newName;
565: char *docName;
566: #endif /* __STDC__ */
567: {
1.31 cvs 568: char *basename;
1.18 cvs 569: char tempOrgName[MAX_LENGTH];
1.5 cvs 570: char *ptr;
571: int length;
572:
573: if (!newName || !docName)
574: return;
1.18 cvs 575:
1.32 cvs 576: if (doc != 0)
1.33 cvs 577: basename = GetBaseURL (doc);
1.32 cvs 578: else
1.33 cvs 579: basename = (char *) NULL;
1.32 cvs 580:
1.18 cvs 581: /*
1.31 cvs 582: * Clean orgName
583: * Make sure we have a complete orgName, without any leading or trailing
584: * white spaces, or trailinbg new lines
585: */
1.5 cvs 586: ptr = orgName;
1.18 cvs 587: /* skip leading white space and new line characters */
1.19 cvs 588: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS);
1.39 cvs 589: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
590: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 591: /* clean trailing white space */
592: ptr = strchr (tempOrgName, ' ');
593: if (ptr)
594: *ptr = EOS;
595: /* clean trailing new lines */
1.19 cvs 596: ptr = strchr (tempOrgName, EOL);
1.5 cvs 597: if (ptr)
598: *ptr = EOS;
599:
1.18 cvs 600: /*
1.31 cvs 601: * Make orgName a complete URL
602: * If the URL does not include a protocol, then try to calculate
603: * one using the doc's base element (if it exists),
604: */
1.21 cvs 605: if (tempOrgName[0] == EOS)
606: {
607: newName[0] = EOS;
1.36 cvs 608: TtaFreeMemory (basename);
1.21 cvs 609: return;
610: }
611: else if (IsW3Path (tempOrgName))
612: {
613: /* the name is complete, go to the Sixth Step */
614: strcpy (newName, tempOrgName);
615: /* verify if the URL has the form "protocol://server:port" */
1.25 cvs 616: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST |
617: AMAYA_PARSE_PUNCTUATION);
1.21 cvs 618: if (ptr && !strcmp (ptr, newName))
1.43 ! cvs 619: /* it has this form, we complete it by adding a DIR_STR */
! 620: strcat (newName, URL_STR);
! 621:
1.21 cvs 622: if (ptr)
1.25 cvs 623: TtaFreeMemory (ptr);
1.21 cvs 624: }
625: else if ( doc == 0)
1.19 cvs 626: /* the name is complete, go to the Sixth Step */
1.18 cvs 627: strcpy (newName, tempOrgName);
1.5 cvs 628: else
629: {
1.31 cvs 630: /* Calculate the absolute URL, using the base or document URL */
1.25 cvs 631: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.14 cvs 632: if (ptr)
633: {
1.29 cvs 634: SimplifyUrl (&ptr);
1.14 cvs 635: strcpy (newName, ptr);
1.25 cvs 636: TtaFreeMemory (ptr);
1.14 cvs 637: }
638: else
1.18 cvs 639: newName[0] = EOS;
1.5 cvs 640: }
1.36 cvs 641:
642: TtaFreeMemory (basename);
1.18 cvs 643: /*
1.31 cvs 644: * Prepare the docname that will refer to this ressource in the
645: * .amaya directory. If the new URL finishes on DIR_SEP, then use
646: * noname.html as a default ressource name
1.18 cvs 647: */
1.19 cvs 648: if (newName[0] != EOS)
1.5 cvs 649: {
1.19 cvs 650: length = strlen (newName) - 1;
1.43 ! cvs 651: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.18 cvs 652: {
653: /* docname was not comprised inside the URL, so let's */
654: /* assign the default ressource name */
655: strcpy (docName, "noname.html");
656: /* remove DIR_SEP at the end of complete path */
1.23 cvs 657: /* newName[length] = EOS; */
1.18 cvs 658: }
1.14 cvs 659: else
1.18 cvs 660: {
661: /* docname is comprised inside the URL */
1.43 ! cvs 662: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP)
1.18 cvs 663: length--;
664: if (length < 0)
665: strcpy (docName, newName);
666: else
667: strcpy (docName, &newName[length+1]);
668: }
1.5 cvs 669: }
1.18 cvs 670: else
671: docName[0] = EOS;
672: }
1.3 cvs 673:
1.4 cvs 674: /*----------------------------------------------------------------------
1.9 cvs 675: IsSameHost
1.4 cvs 676: ----------------------------------------------------------------------*/
1.3 cvs 677: #ifdef __STDC__
1.34 cvs 678: boolean IsSameHost (const char *url1, const char *url2)
1.3 cvs 679: #else /* __STDC__ */
680: boolean IsSameHost (url1, url2)
1.34 cvs 681: const char *url1;
682: const char *url2;
1.3 cvs 683: #endif /* __STDC__ */
684: {
1.34 cvs 685: char *basename_ptr1, *basename_ptr2;
686: boolean result;
1.3 cvs 687:
1.25 cvs 688: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
689: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 690:
1.5 cvs 691: if (strcmp (basename_ptr1, basename_ptr2))
1.8 cvs 692: result = FALSE;
1.5 cvs 693: else
1.8 cvs 694: result = TRUE;
1.3 cvs 695:
1.25 cvs 696: TtaFreeMemory (basename_ptr1);
697: TtaFreeMemory (basename_ptr2);
1.5 cvs 698: return (result);
1.3 cvs 699: }
700:
701:
1.4 cvs 702: /*----------------------------------------------------------------------
1.22 cvs 703: HasKnownFileSuffix
704: returns TRUE if path points to a file ending with a suffix.
705: ----------------------------------------------------------------------*/
706: #ifdef __STDC__
1.34 cvs 707: boolean HasKnownFileSuffix (const char *path)
1.22 cvs 708: #else /* __STDC__ */
709: boolean HasKnownFileSuffix (path)
1.34 cvs 710: const char *path;
1.22 cvs 711: #endif /* __STDC__ */
712: {
1.29 cvs 713: char *root;
714: char temppath[MAX_LENGTH];
715: char suffix[MAX_LENGTH];
1.22 cvs 716:
1.24 cvs 717: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 718: return (FALSE);
719:
1.29 cvs 720: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 721:
722: if (root)
723: {
724: strcpy (temppath, root);
1.25 cvs 725: TtaFreeMemory (root);
1.22 cvs 726: /* Get the suffix */
727: ExtractSuffix (temppath, suffix);
728:
729: if( suffix[0] == EOS)
730: /* no suffix */
731: return (FALSE);
732:
733: /* Normalize the suffix */
734: ConvertToLowerCase (suffix);
735:
1.23 cvs 736: if (!strcmp (suffix, "gz"))
1.22 cvs 737: /* skip the compressed suffix */
738: {
739: ExtractSuffix (temppath, suffix);
740: if(suffix[0] == EOS)
741: /* no suffix */
742: return (FALSE);
743: /* Normalize the suffix */
744: ConvertToLowerCase (suffix);
745: }
746:
747: if ((strcmp (suffix, "gif")) && (strcmp (suffix, "xbm")) &&
748: (strcmp (suffix, "xpm")) && (strcmp (suffix, "jpg")) &&
749: (strcmp (suffix, "pdf")) && (strcmp (suffix, "png")) &&
750: (strcmp (suffix, "tgz")) && (strcmp (suffix, "xpg")) &&
751: (strcmp (suffix, "xpd")) && (strcmp (suffix, "ps")) &&
752: (strcmp (suffix, "au")) && (strcmp (suffix, "html")) &&
753: (strcmp (suffix, "htm")) && (strcmp (suffix, "shtml")) &&
754: (strcmp (suffix, "txt")) && (strcmp (suffix, "css")) &&
755: (strcmp (suffix, "eps")))
756: return (FALSE);
757: else
758: return (TRUE);
759: }
760: else
761: return (FALSE);
762: }
763:
764:
765: /*----------------------------------------------------------------------
1.24 cvs 766: ChopURL
767: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
768: If inputURL is bigger than that size, outputURL receives
769: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
770: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
771: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
772: copied into outputURL.
773: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
774: chars.
775: ----------------------------------------------------------------------*/
776: #ifdef __STDC__
1.34 cvs 777: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 778: #else
779: void ChopURL (outputURL, inputURL)
1.34 cvs 780: char *outputURL;
781: const char *inputURL;
1.24 cvs 782: #endif
1.22 cvs 783:
1.24 cvs 784: {
785: int len;
1.9 cvs 786:
1.24 cvs 787: len = strlen (inputURL);
788: if (len <= MAX_PRINT_URL_LENGTH)
1.29 cvs 789: strcpy (outputURL, inputURL);
1.24 cvs 790: else
791: /* make a truncated urlName on the status window */
792: {
793: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
794: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
795: strcat (outputURL, "...");
796: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
797: }
1.25 cvs 798: }
799:
800:
801: /*----------------------------------------------------------------------
802: scan
803: Scan a filename for its consituents
804: -----------------------------------
805:
806: On entry,
807: name points to a document name which may be incomplete.
808: On exit,
809: absolute or relative may be nonzero (but not both).
810: host, fragment and access may be nonzero if they were specified.
811: Any which are nonzero point to zero terminated strings.
812: ----------------------------------------------------------------------*/
813: #ifdef __STDC__
814: static void scan (char * name, HTURI * parts)
815: #else /* __STDC__ */
816: static void scan (name, parts)
817: char *name;
818: HTURI *parts;
819:
820: #endif /* __STDC__ */
821: {
1.43 ! cvs 822: char *p;
! 823: char *after_access = name;
1.32 cvs 824:
1.43 ! cvs 825: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 826: /* Look for fragment identifier */
827: if ((p = strrchr(name, '#')) != NULL)
828: {
829: *p++ = '\0';
830: parts->fragment = p;
1.25 cvs 831: }
832:
1.28 cvs 833: for (p=name; *p; p++)
834: {
1.43 ! cvs 835: if (*p == URL_SEP || *p == DIR_SEP || *p=='#' || *p=='?')
1.28 cvs 836: break;
1.32 cvs 837: if (*p==':')
1.28 cvs 838: {
839: *p = 0;
840: parts->access = after_access; /* Scheme has been specified */
841:
842: /* The combination of gcc, the "-O" flag and the HP platform is
843: unhealthy. The following three lines is a quick & dirty fix, but is
844: not recommended. Rather, turn off "-O". */
845:
846: /* after_access = p;*/
847: /* while (*after_access == 0)*/
848: /* after_access++;*/
849: after_access = p+1;
1.43 ! cvs 850: if (!strcasecmp("URL", parts->access))
1.28 cvs 851: /* Ignore IETF's URL: pre-prefix */
852: parts->access = NULL;
853: else
1.25 cvs 854: break;
855: }
856: }
857:
858: p = after_access;
1.43 ! cvs 859: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 860: {
1.43 ! cvs 861: if (p[1] == URL_SEP)
1.28 cvs 862: {
1.25 cvs 863: parts->host = p+2; /* host has been specified */
1.28 cvs 864: *p = 0; /* Terminate access */
865: /* look for end of host name if any */
1.43 ! cvs 866: p = strchr (parts->host, URL_SEP);
1.28 cvs 867: if (p)
868: {
1.43 ! cvs 869: *p = EOS; /* Terminate host */
1.25 cvs 870: parts->absolute = p+1; /* Root has been found */
1.28 cvs 871: }
872: }
873: else
874: /* Root found but no host */
875: parts->absolute = p+1;
876: }
877: else
878: {
1.25 cvs 879: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 880: }
1.25 cvs 881: }
882:
883:
884: /*----------------------------------------------------------------------
1.28 cvs 885: AmayaParseUrl: parse a Name relative to another name
886:
887: This returns those parts of a name which are given (and requested)
888: substituting bits from the related name where necessary.
1.25 cvs 889:
1.28 cvs 890: On entry,
1.25 cvs 891: aName A filename given
892: relatedName A name relative to which aName is to be parsed. Give
893: it an empty string if aName is absolute.
894: wanted A mask for the bits which are wanted.
895:
1.28 cvs 896: On exit,
1.25 cvs 897: returns A pointer to a malloc'd string which MUST BE FREED
898: ----------------------------------------------------------------------*/
899: #ifdef __STDC__
1.34 cvs 900: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1.25 cvs 901: #else /* __STDC__ */
1.28 cvs 902: char *AmayaParseUrl (aName, relatedName, wanted)
1.34 cvs 903: const char *aName;
1.28 cvs 904: char *relatedName;
905: int wanted;
1.25 cvs 906:
907: #endif /* __STDC__ */
908: {
1.29 cvs 909: char *return_value;
910: char result[MAX_LENGTH];
911: char name[MAX_LENGTH];
912: char rel[MAX_LENGTH];
913: char *p, *access;
914: HTURI given, related;
915: int len;
1.33 cvs 916: char used_sep;
917: char *used_str;
1.32 cvs 918:
1.43 ! cvs 919: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 920: {
1.42 cvs 921: used_str = DIR_STR;
922: used_sep = DIR_SEP;
1.33 cvs 923: }
1.32 cvs 924: else
1.33 cvs 925: {
1.42 cvs 926: used_str = URL_STR;
927: used_sep = URL_SEP;
1.33 cvs 928: }
1.32 cvs 929:
1.29 cvs 930: /* Make working copies of input strings to cut up: */
931: return_value = NULL;
932: result[0] = 0; /* Clear string */
933: strcpy (name, aName);
934: if (relatedName != NULL)
935: strcpy (rel, relatedName);
936: else
937: relatedName[0] = EOS;
938:
939: scan (name, &given);
940: scan (rel, &related);
941: access = given.access ? given.access : related.access;
942: if (wanted & AMAYA_PARSE_ACCESS)
943: if (access)
944: {
945: strcat (result, access);
946: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.32 cvs 947: strcat (result, ":");
1.29 cvs 948: }
949:
950: if (given.access && related.access)
951: /* If different, inherit nothing. */
952: if (strcmp (given.access, related.access) != 0)
953: {
954: related.host = 0;
955: related.absolute = 0;
956: related.relative = 0;
957: related.fragment = 0;
958: }
959:
960: if (wanted & AMAYA_PARSE_HOST)
961: if(given.host || related.host)
962: {
963: if(wanted & AMAYA_PARSE_PUNCTUATION)
964: strcat (result, "//");
965: strcat (result, given.host ? given.host : related.host);
966: }
967:
968: if (given.host && related.host)
969: /* If different hosts, inherit no path. */
970: if (strcmp(given.host, related.host) != 0)
971: {
972: related.absolute = 0;
973: related.relative = 0;
974: related.fragment = 0;
975: }
976:
977: if (wanted & AMAYA_PARSE_PATH)
978: {
979: if (given.absolute)
980: {
981: /* All is given */
982: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.33 cvs 983: strcat (result, used_str);
1.29 cvs 984: strcat (result, given.absolute);
1.25 cvs 985: }
1.29 cvs 986: else if (related.absolute)
987: {
988: /* Adopt path not name */
1.33 cvs 989: strcat (result, used_str);
1.29 cvs 990: strcat (result, related.absolute);
991: if (given.relative)
992: {
993: /* Search part? */
994: p = strchr (result, '?');
995: if (!p)
996: p=result+strlen(result)-1;
1.33 cvs 997: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 998: /* Remove filename */
999: p[1]=0;
1000: /* Add given one */
1001: strcat (result, given.relative);
1002: /*SimplifyUrl (&result);*/
1.25 cvs 1003: }
1004: }
1.29 cvs 1005: else if (given.relative)
1006: /* what we've got */
1007: strcat (result, given.relative);
1008: else if (related.relative)
1009: strcat (result, related.relative);
1010: else
1011: /* No inheritance */
1.33 cvs 1012: strcat (result, used_str);
1.25 cvs 1013: }
1.29 cvs 1014:
1015: if (wanted & AMAYA_PARSE_ANCHOR)
1016: if (given.fragment || related.fragment)
1017: {
1018: if (given.absolute && given.fragment)
1019: {
1020: /*Fixes for relURLs...*/
1021: if (wanted & AMAYA_PARSE_PUNCTUATION)
1022: strcat (result, "#");
1023: strcat (result, given.fragment);
1024: }
1025: else if (!(given.absolute) && !(given.fragment))
1026: strcat (result, "");
1027: else
1028: {
1029: if (wanted & AMAYA_PARSE_PUNCTUATION)
1030: strcat (result, "#");
1031: strcat (result, given.fragment ? given.fragment : related.fragment);
1032: }
1033: }
1034: len = strlen (result);
1035: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1036: strcpy (return_value, result);
1037: return (return_value); /* exactly the right length */
1.25 cvs 1038: }
1039:
1040: /*----------------------------------------------------------------------
1041: HTCanon
1042: Canonicalizes the URL in the following manner starting from the host
1043: pointer:
1044:
1045: 1) The host name is converted to lowercase
1046: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1047:
1048: Return: OK The position of the current path part of the URL
1049: which might be the old one or a new one.
1050:
1051: ----------------------------------------------------------------------*/
1052: #ifdef __STDC__
1.28 cvs 1053: static char *HTCanon (char ** filename, char * host)
1.25 cvs 1054: #else /* __STDC__ */
1.28 cvs 1055: static char *HTCanon (filename, host)
1056: char **filename;
1057: char *host;
1.25 cvs 1058: #endif /* __STDC__ */
1059: {
1060: char *newname = NULL;
1061: char *port;
1062: char *strptr;
1063: char *path;
1064: char *access = host-3;
1.33 cvs 1065: char used_sep;
1.32 cvs 1066:
1067:
1.33 cvs 1068: if (*filename && strchr (*filename, URL_SEP))
1069: {
1070: used_sep = URL_SEP;
1071: }
1072: else
1073: {
1074: used_sep = DIR_SEP;
1075: }
1.32 cvs 1076:
1.33 cvs 1077: while (access>*filename && *(access-1)!= used_sep) /* Find access method */
1.25 cvs 1078: access--;
1.33 cvs 1079: if ((path = strchr(host, used_sep)) == NULL) /* Find path */
1.25 cvs 1080: path = host + strlen(host);
1081: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */
1082: host = strptr;
1.32 cvs 1083: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */
1.25 cvs 1084: port = NULL;
1085:
1086: strptr = host; /* Convert to lower-case */
1.33 cvs 1087: while (strptr<path)
1088: {
1.25 cvs 1089: *strptr = tolower(*strptr);
1090: strptr++;
1.33 cvs 1091: }
1.25 cvs 1092:
1093: /* Does the URL contain a full domain name? This also works for a
1094: numerical host name. The domain name is already made lower-case
1095: and without a trailing dot. */
1096: {
1.33 cvs 1097: char *dot = port ? port : path;
1098: if (dot > *filename && *--dot=='.')
1099: {
1100: char *orig=dot, *dest=dot+1;
1101: while((*orig++ = *dest++));
1102: if (port) port--;
1103: path--;
1.25 cvs 1104: }
1105: }
1106: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1107: if (port)
1108: {
1109: if (!*(port+1) || *(port+1)==used_sep)
1110: {
1111: if (!newname)
1112: {
1.25 cvs 1113: char *orig=port, *dest=port+1;
1114: while((*orig++ = *dest++));
1.33 cvs 1115: }
1116: }
1117: else if ((!strncmp(access, "http", 4) &&
1118: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1119: (!strncmp(access, "gopher", 6) &&
1120: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1121: (!strncmp(access, "ftp", 3) &&
1122: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) {
1123: if (!newname)
1124: {
1125: char *orig=port, *dest=port+3;
1126: while((*orig++ = *dest++));
1127: /* Update path position, Henry Minsky */
1128: path -= 3;
1.25 cvs 1129: }
1.33 cvs 1130: }
1131: else if (newname)
1132: strncat(newname, port, (int) (path-port));
1133: }
1.25 cvs 1134:
1.33 cvs 1135: if (newname)
1136: {
1.25 cvs 1137: char *newpath = newname+strlen(newname);
1138: strcat(newname, path);
1139: path = newpath;
1.28 cvs 1140: /* Free old copy */
1141: TtaFreeMemory(*filename);
1.25 cvs 1142: *filename = newname;
1.33 cvs 1143: }
1.25 cvs 1144: return path;
1145: }
1146:
1147:
1148: /*----------------------------------------------------------------------
1.29 cvs 1149: SimplifyUrl: simplify a URI
1.32 cvs 1150: A URI is allowed to contain the sequence xxx/../ which may be
1151: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1152: Simplification helps us recognize duplicate URIs.
1.25 cvs 1153:
1.28 cvs 1154: Thus, /etc/junk/../fred becomes /etc/fred
1155: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1156:
1.28 cvs 1157: but we should NOT change
1158: http://fred.xxx.edu/../..
1.25 cvs 1159:
1160: or ../../albert.html
1161:
1.28 cvs 1162: In order to avoid empty URLs the following URLs become:
1.25 cvs 1163:
1164: /fred/.. becomes /fred/..
1165: /fred/././.. becomes /fred/..
1166: /fred/.././junk/.././ becomes /fred/..
1167:
1.28 cvs 1168: If more than one set of `://' is found (several proxies in cascade) then
1169: only the part after the last `://' is simplified.
1.25 cvs 1170:
1.28 cvs 1171: Returns: A string which might be the old one or a new one.
1.25 cvs 1172: ----------------------------------------------------------------------*/
1173: #ifdef __STDC__
1.29 cvs 1174: void SimplifyUrl (char ** url)
1.25 cvs 1175: #else /* __STDC__ */
1.29 cvs 1176: void SimplifyUrl (url)
1.28 cvs 1177: char **url;
1.25 cvs 1178: #endif /* __STDC__ */
1179: {
1.28 cvs 1180: char *path, *p;
1181: char *newptr, *access;
1182: char *orig, *dest, *end;
1183:
1.33 cvs 1184: char used_sep;
1.32 cvs 1185:
1186:
1.28 cvs 1187: if (!url || !*url)
1188: return;
1189:
1.33 cvs 1190: if (strchr (*url, URL_SEP))
1191: {
1192: used_sep = URL_SEP;
1193: }
1.32 cvs 1194: else
1.33 cvs 1195: {
1196: used_sep = DIR_SEP;
1197: }
1.32 cvs 1198:
1.28 cvs 1199: /* Find any scheme name */
1200: if ((path = strstr(*url, "://")) != NULL)
1.33 cvs 1201: {
1202: /* Find host name */
1.28 cvs 1203: access = *url;
1204: while (access<path && (*access=tolower(*access)))
1205: access++;
1206: path += 3;
1207: while ((newptr = strstr(path, "://")) != NULL)
1208: /* For proxies */
1209: path = newptr+3;
1210: /* We have a host name */
1211: path = HTCanon(url, path);
1.25 cvs 1212: }
1.28 cvs 1213: else if ((path = strstr(*url, ":/")) != NULL)
1214: path += 2;
1215: else
1216: path = *url;
1.25 cvs 1217:
1.33 cvs 1218: if (*path == used_sep && *(path+1)==used_sep)
1.28 cvs 1219: /* Some URLs start //<foo> */
1220: path += 1;
1221: else if (!strncmp(path, "news:", 5))
1222: {
1223: newptr = strchr(path+5, '@');
1224: if (!newptr)
1225: newptr = path + 5;
1226: while (*newptr)
1227: {
1228: /* Make group or host lower case */
1229: *newptr = tolower (*newptr);
1230: newptr++;
1.25 cvs 1231: }
1.28 cvs 1232: /* Doesn't need to do any more */
1233: return;
1.25 cvs 1234: }
1.28 cvs 1235:
1236: if ((p = path))
1237: {
1238: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1239: (end = strchr (path, '#'))))
1240: end = path + strlen (path);
1241:
1242: /* Parse string second time to simplify */
1243: p = path;
1244: while (p < end)
1245: {
1.33 cvs 1246: if (*p==used_sep)
1.28 cvs 1247: {
1.33 cvs 1248: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1249: {
1250: orig = p + 1;
1.33 cvs 1251: dest = (*(p+2)!=used_sep) ? p+2 : p+3;
1.28 cvs 1252: while ((*orig++ = *dest++)); /* Remove a slash and a dot */
1253: end = orig - 1;
1254: }
1.33 cvs 1255: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3)))
1.28 cvs 1256: {
1257: newptr = p;
1.33 cvs 1258: while (newptr>path && *--newptr!=used_sep); /* prev slash */
1.28 cvs 1259: if (strncmp(newptr, "/../", 4))
1260: {
1261: orig = newptr + 1;
1.33 cvs 1262: dest = (*(p+3)!=used_sep) ? p+3 : p+4;
1.28 cvs 1263: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1264: end = orig-1;
1265: /* Start again with prev slash */
1266: p = newptr;
1.25 cvs 1267: }
1.28 cvs 1268: else
1.25 cvs 1269: p++;
1.28 cvs 1270: }
1.33 cvs 1271: else if (*(p+1) == used_sep)
1.28 cvs 1272: {
1.33 cvs 1273: while (*(p+1) == used_sep)
1.28 cvs 1274: {
1275: orig = p;
1276: dest = p + 1;
1277: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1278: end = orig-1;
1279: }
1280: }
1281: else
1.25 cvs 1282: p++;
1.28 cvs 1283: }
1284: else
1285: p++;
1.25 cvs 1286: }
1287: }
1.28 cvs 1288: return;
1289: }
1290:
1291:
1292: /*----------------------------------------------------------------------
1293: NormalizeFile normalizes local names.
1294: Return TRUE if target and src differ.
1295: ----------------------------------------------------------------------*/
1296: #ifdef __STDC__
1297: boolean NormalizeFile (char *src, char *target)
1298: #else
1299: boolean NormalizeFile (src, target)
1300: char *src;
1301: char *target;
1302:
1303: #endif
1304: {
1305: char *s;
1306: boolean change;
1307:
1308: change = FALSE;
1309: if (src[0] == '~')
1310: {
1311: /* replace ~ */
1312: s = (char *) TtaGetEnvString ("HOME");
1313: strcpy (target, s);
1314: strcat (target, &src[1]);
1315: change = TRUE;
1316: }
1317: else if (strncmp (src, "file:", 5) == 0)
1318: {
1319: /* remove the prefix file: */
1320: if (src[5] == EOS)
1321: strcpy (target, DIR_STR);
1322: else if (src[0] == '~')
1323: {
1324: /* replace ~ */
1325: s = (char *) TtaGetEnvString ("HOME");
1326: strcpy (target, s);
1327: strcat (target, &src[5]);
1328: }
1329: else
1330: strcpy (target, &src[5]);
1331: change = TRUE;
1332: }
1333: else
1334: strcpy (target, src);
1335:
1336: /* remove /../ and /./ */
1.29 cvs 1337: SimplifyUrl (&target);
1.28 cvs 1338: return (change);
1.25 cvs 1339: }
1340:
1.28 cvs 1341:
1.25 cvs 1342: /*----------------------------------------------------------------------
1.31 cvs 1343: MakeRelativeURL: make relative name
1.25 cvs 1344:
1.28 cvs 1345: This function creates and returns a string which gives an expression of
1346: one address as related to another. Where there is no relation, an absolute
1347: address is retured.
1.25 cvs 1348:
1.28 cvs 1349: On entry,
1.25 cvs 1350: Both names must be absolute, fully qualified names of nodes
1351: (no fragment bits)
1352:
1.28 cvs 1353: On exit,
1.25 cvs 1354: The return result points to a newly allocated name which, if
1355: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1356: The caller is responsible for freeing the resulting name later.
1357: ----------------------------------------------------------------------*/
1358: #ifdef __STDC__
1.31 cvs 1359: char *MakeRelativeURL (char *aName, char *relatedName)
1.25 cvs 1360: #else /* __STDC__ */
1.31 cvs 1361: char *MakeRelativeURL (aName, relatedName)
1.28 cvs 1362: char *aName;
1363: char *relatedName;
1.25 cvs 1364: #endif /* __STDC__ */
1365: {
1.39 cvs 1366: char *return_value;
1367: char result[MAX_LENGTH];
1.29 cvs 1368: char *p;
1.34 cvs 1369: char *q;
1.29 cvs 1370: char *after_access;
1371: char *last_slash = NULL;
1372: int slashes, levels, len;
1373:
1374: if (aName == NULL || relatedName == NULL)
1375: return (NULL);
1376:
1377: slashes = 0;
1378: after_access = NULL;
1379: p = aName;
1380: q = relatedName;
1381: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1382: {
1383: /* Find extent of match */
1.32 cvs 1384: if (*p == ':')
1.29 cvs 1385: after_access = p + 1;
1.28 cvs 1386: if (*p == DIR_SEP)
1.27 cvs 1387: {
1.29 cvs 1388: /* memorize the last slash position and count them */
1.27 cvs 1389: last_slash = p;
1390: slashes++;
1.25 cvs 1391: }
1392: }
1393:
1.31 cvs 1394: /* q, p point to the first non-matching character or zero */
1395: if (*q == EOS)
1396: {
1397: /* New name is a subset of the related name */
1398: /* exactly the right length */
1399: len = strlen (p);
1400: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1401: strcpy (return_value, p);
1402: }
1403: else if ((slashes < 2 && after_access == NULL)
1404: || (slashes < 3 && after_access != NULL))
1405: {
1406: /* Two names whitout common path */
1407: /* exactly the right length */
1408: len = strlen (aName);
1409: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1410: strcpy (return_value, aName);
1411: }
1412: else
1413: {
1414: /* Some path in common */
1415: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1416: /* just the same server */
1417: strcpy (result, last_slash);
1418: else
1419: {
1420: levels= 0;
1421: for (; *q && (*q != '#'); q++)
1422: if (*q == DIR_SEP)
1423: levels++;
1424:
1425: result[0] = 0;
1426: for (;levels; levels--)
1427: strcat (result, "../");
1428: strcat (result, last_slash+1);
1429: }
1430:
1431: /* exactly the right length */
1432: len = strlen (result);
1433: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1434: strcpy (return_value, result);
1.25 cvs 1435: }
1.29 cvs 1436: return (return_value);
1.24 cvs 1437: }
1.35 cvs 1438:
1439:
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