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