Annotation of Amaya/amaya/AHTURLTools.c, revision 1.47
1.7 cvs 1: /*
2: *
3: * (c) COPYRIGHT MIT and INRIA, 1996.
4: * Please first read the full copyright statement in file COPYRIGHT.
5: *
6: */
1.9 cvs 7:
1.10 cvs 8: /*
9: * AHTURLTools.c: contains all the functions for testing, manipulating,
1.25 cvs 10: * and normalizing URLs. It also contains a local copy of the libWWW
11: * URL parsing functions.
1.10 cvs 12: *
13: * Authors: J. Kahan, I. Vatton
1.45 cvs 14: * R. Guetari (Stuff related to Windows).
1.10 cvs 15: *
16: */
1.7 cvs 17:
1.15 cvs 18: #define THOT_EXPORT extern
1.3 cvs 19: #include "amaya.h"
20:
1.8 cvs 21: #include "init_f.h"
22: #include "AHTURLTools_f.h"
23:
1.24 cvs 24: #define MAX_PRINT_URL_LENGTH 50
1.29 cvs 25: typedef struct _HTURI {
26: char * access; /* Now known as "scheme" */
27: char * host;
28: char * absolute;
29: char * relative;
30: char * fragment;
31: } HTURI;
1.24 cvs 32:
1.28 cvs 33:
34: /*----------------------------------------------------------------------
35: ConvertToLowerCase
36: Converts a string to lowercase.
37: ----------------------------------------------------------------------*/
1.22 cvs 38: #ifdef __STDC__
1.38 cvs 39: void ConvertToLowerCase (char *string)
1.28 cvs 40: #else /* __STDC__ */
1.38 cvs 41: void ConvertToLowerCase (string)
1.28 cvs 42: char *string;
43:
44: #endif /* __STDC__ */
45: {
46: int i;
47:
48: if (!string)
49: return;
50:
51: for (i = 0; string[i] != EOS; i++)
52: string[i] = tolower (string[i]);
53: }
1.22 cvs 54:
1.8 cvs 55: /*----------------------------------------------------------------------
1.11 cvs 56: ExplodeURL
1.8 cvs 57: ----------------------------------------------------------------------*/
58: #ifdef __STDC__
59: void ExplodeURL (char *url, char **proto, char **host, char **dir, char **file)
60: #else
61: void ExplodeURL (url, proto, host, dir, file)
62: char *url;
63: char **proto;
64: char **host;
65: char **dir;
66: char **file;
67:
68: #endif
69: {
1.33 cvs 70: char *curr, *temp;
71: char used_sep;
1.32 cvs 72:
1.33 cvs 73: if (url && strchr (url, URL_SEP))
74: used_sep = URL_SEP;
75: else
76: used_sep = DIR_SEP;
1.8 cvs 77:
78: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
79: (dir == NULL) || (file == NULL))
80: return;
81:
82: /* initialize every pointer */
83: *proto = *host = *dir = *file = NULL;
84:
85: /* skip any leading space */
86: while ((*url == SPACE) || (*url == TAB))
87: url++;
1.9 cvs 88: curr = url;
89: if (*curr == 0)
1.8 cvs 90: goto finished;
91:
92: /* go to the end of the URL */
1.9 cvs 93: while ((*curr != 0) && (*curr != SPACE) && (*curr != '\b') &&
94: (*curr != '\r') && (*curr != EOL))
95: curr++;
1.8 cvs 96:
97: /* mark the end of the chain */
1.9 cvs 98: *curr = EOS;
99: curr--;
100: if (curr <= url)
1.8 cvs 101: goto finished;
102:
103: /* search the next DIR_SEP indicating the beginning of the file name */
104: do
1.11 cvs 105: curr--;
1.33 cvs 106: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 107:
1.9 cvs 108: if (curr < url)
1.8 cvs 109: goto finished;
1.9 cvs 110: *file = curr + 1;
1.8 cvs 111:
112: /* mark the end of the dir */
1.9 cvs 113: *curr = EOS;
114: curr--;
115: if (curr < url)
1.8 cvs 116: goto finished;
117:
1.29 cvs 118: /* search for the DIR_STR indicating the host name start */
1.33 cvs 119: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 120: curr--;
1.8 cvs 121:
122: /* if we found it, separate the host name from the directory */
1.33 cvs 123: if ((*curr == DIR_SEP) && (*(curr + 1) == used_sep))
1.8 cvs 124: {
1.9 cvs 125: *host = temp = curr + 2;
1.33 cvs 126: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 127: temp++;
1.33 cvs 128: if (*temp == used_sep)
1.8 cvs 129: {
130: *temp = EOS;
131: *dir = temp + 1;
132: }
133: }
134: else
1.11 cvs 135: *dir = curr;
136:
1.9 cvs 137: if (curr <= url)
1.8 cvs 138: goto finished;
139:
140: /* mark the end of the proto */
1.9 cvs 141: *curr = EOS;
142: curr--;
143: if (curr < url)
1.8 cvs 144: goto finished;
145:
1.32 cvs 146: if (*curr == ':')
1.8 cvs 147: {
1.9 cvs 148: *curr = EOS;
149: curr--;
1.8 cvs 150: }
151: else
152: goto finished;
1.11 cvs 153:
1.9 cvs 154: if (curr < url)
1.8 cvs 155: goto finished;
1.9 cvs 156: while ((curr > url) && (isalpha (*curr)))
157: curr--;
158: *proto = curr;
1.8 cvs 159:
160: finished:;
161:
162: #ifdef AMAYA_DEBUG
163: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
164: if (*proto)
165: fprintf (stderr, "proto : %s, ", *proto);
166: if (*host)
167: fprintf (stderr, "host : %s, ", *host);
168: if (*dir)
169: fprintf (stderr, "dir : %s, ", *dir);
170: if (*file)
171: fprintf (stderr, "file : %s ", *file);
172: fprintf (stderr, "\n");
173: #endif
174:
175: }
1.3 cvs 176:
1.4 cvs 177: /*----------------------------------------------------------------------
1.9 cvs 178: IsHTMLName
179: returns TRUE if path points to an HTML resource.
1.4 cvs 180: ----------------------------------------------------------------------*/
1.3 cvs 181: #ifdef __STDC__
1.34 cvs 182: boolean IsHTMLName (const char *path)
1.3 cvs 183: #else /* __STDC__ */
184: boolean IsHTMLName (path)
1.34 cvs 185: const char *path;
1.3 cvs 186: #endif /* __STDC__ */
187: {
1.5 cvs 188: char temppath[MAX_LENGTH];
189: char suffix[MAX_LENGTH];
190: char nsuffix[MAX_LENGTH];
191: int i;
192:
193: if (!path)
1.37 cvs 194: return (FALSE);
1.5 cvs 195:
196: strcpy (temppath, path);
197: ExtractSuffix (temppath, suffix);
198:
199: /* Normalize the suffix */
200: i = 0;
1.39 cvs 201: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 202: {
1.25 cvs 203: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 204: i++;
205: }
1.5 cvs 206: nsuffix[i] = EOS;
1.39 cvs 207: if ((!strcmp (nsuffix, "html")) ||
208: (!strcmp (nsuffix, "htm")) ||
209: (!strcmp (nsuffix, "shtml")))
210: return (TRUE);
1.22 cvs 211: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 212: {
1.39 cvs 213: /* take into account compressed files */
1.13 cvs 214: ExtractSuffix (temppath, suffix);
215: /* Normalize the suffix */
216: i = 0;
1.39 cvs 217: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 218: {
1.25 cvs 219: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 220: i++;
221: }
222: nsuffix[i] = EOS;
1.39 cvs 223: if ((!strcmp (nsuffix, "html")) ||
224: (!strcmp (nsuffix, "htm")) ||
225: (!strcmp (nsuffix, "shtml")))
226: return (TRUE);
227: else
1.13 cvs 228: return (FALSE);
229: }
230: else
1.39 cvs 231: return (FALSE);
1.3 cvs 232: }
233:
1.4 cvs 234: /*----------------------------------------------------------------------
1.9 cvs 235: IsImageName
236: returns TRUE if path points to an image resource.
1.4 cvs 237: ----------------------------------------------------------------------*/
1.3 cvs 238: #ifdef __STDC__
1.34 cvs 239: boolean IsImageName (const char *path)
1.3 cvs 240: #else /* __STDC__ */
241: boolean IsImageName (path)
1.34 cvs 242: const char *path;
1.3 cvs 243: #endif /* __STDC__ */
244: {
1.5 cvs 245: char temppath[MAX_LENGTH];
246: char suffix[MAX_LENGTH];
247: char nsuffix[MAX_LENGTH];
248: int i;
249:
250: if (!path)
1.13 cvs 251: return (FALSE);
1.5 cvs 252:
253: strcpy (temppath, path);
254: ExtractSuffix (temppath, suffix);
255:
256: /* Normalize the suffix */
257: i = 0;
1.39 cvs 258: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 259: {
1.25 cvs 260: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 261: i++;
262: }
1.5 cvs 263: nsuffix[i] = EOS;
1.39 cvs 264: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
265: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
266: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
267: return (TRUE);
268: return (FALSE);
1.3 cvs 269: }
270:
1.4 cvs 271: /*----------------------------------------------------------------------
1.9 cvs 272: IsTextName
1.4 cvs 273: ----------------------------------------------------------------------*/
1.3 cvs 274: #ifdef __STDC__
1.34 cvs 275: boolean IsTextName (const char *path)
1.3 cvs 276: #else /* __STDC__ */
277: boolean IsTextName (path)
1.34 cvs 278: const char *path;
1.3 cvs 279:
280: #endif /* __STDC__ */
281: {
1.5 cvs 282: char temppath[MAX_LENGTH];
283: char suffix[MAX_LENGTH];
284: char nsuffix[MAX_LENGTH];
285: int i;
286:
287: if (!path)
1.13 cvs 288: return (FALSE);
1.5 cvs 289:
290: strcpy (temppath, path);
291: ExtractSuffix (temppath, suffix);
292:
293: /* Normalize the suffix */
294: i = 0;
1.39 cvs 295: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 296: {
1.25 cvs 297: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 298: i++;
299: }
300: nsuffix[i] = EOS;
301:
1.39 cvs 302: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 303: return (TRUE);
1.22 cvs 304: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 305: {
1.39 cvs 306: /* take into account compressed files */
1.13 cvs 307: ExtractSuffix (temppath, suffix);
308: /* Normalize the suffix */
309: i = 0;
1.39 cvs 310: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 311: {
1.25 cvs 312: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 313: i++;
314: }
315: nsuffix[i] = EOS;
1.39 cvs 316: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 317: return (TRUE);
318: else
319: return (FALSE);
320: }
321: else
322: return (FALSE);
1.3 cvs 323: }
324:
1.4 cvs 325: /*----------------------------------------------------------------------
1.9 cvs 326: IsHTTPPath
327: returns TRUE if path is in fact an http URL.
1.4 cvs 328: ----------------------------------------------------------------------*/
1.3 cvs 329: #ifdef __STDC__
1.34 cvs 330: boolean IsHTTPPath (const char *path)
1.3 cvs 331: #else /* __STDC__ */
332: boolean IsHTTPPath (path)
1.34 cvs 333: const char *path;
1.3 cvs 334: #endif /* __STDC__ */
335: {
1.5 cvs 336: if (!path)
337: return FALSE;
1.3 cvs 338:
1.5 cvs 339: if (strncmp (path, "http:", 5) != 0)
340: return FALSE;
341: return TRUE;
1.3 cvs 342: }
343:
1.4 cvs 344: /*----------------------------------------------------------------------
1.9 cvs 345: IsWithParameters
346: returns TRUE if url has a concatenated query string.
1.4 cvs 347: ----------------------------------------------------------------------*/
1.3 cvs 348: #ifdef __STDC__
1.34 cvs 349: boolean IsWithParameters (const char *url)
1.3 cvs 350: #else /* __STDC__ */
1.9 cvs 351: boolean IsWithParameters (url)
1.34 cvs 352: const char *url;
1.3 cvs 353: #endif /* __STDC__ */
354: {
1.5 cvs 355: int i;
1.3 cvs 356:
1.9 cvs 357: if ((!url) || (url[0] == EOS))
1.5 cvs 358: return FALSE;
1.3 cvs 359:
1.9 cvs 360: i = strlen (url) - 1;
361: while (i > 0 && url[i--] != '?')
1.5 cvs 362: if (i < 0)
363: return FALSE;
1.3 cvs 364:
1.5 cvs 365: /* There is a parameter */
366: return TRUE;
1.3 cvs 367: }
368:
1.4 cvs 369: /*----------------------------------------------------------------------
1.9 cvs 370: IsW3Path
371: returns TRUE if path is in fact a URL.
1.4 cvs 372: ----------------------------------------------------------------------*/
1.3 cvs 373: #ifdef __STDC__
1.34 cvs 374: boolean IsW3Path (const char *path)
1.3 cvs 375: #else /* __STDC__ */
376: boolean IsW3Path (path)
1.34 cvs 377: const char *path;
1.3 cvs 378: #endif /* __STDC__ */
379: {
1.5 cvs 380: if ((strncmp (path, "http:", 5)) && (strncmp (path, "ftp:", 4)) &&
381: (strncmp (path, "telnet:", 7)) && (strncmp (path, "wais:", 5)) &&
382: (strncmp (path, "news:", 5)) && (strncmp (path, "gopher:", 7)) &&
383: (strncmp (path, "mailto:", 7)) && (strncmp (path, "archie:", 7)))
384: return FALSE;
385: return TRUE;
1.3 cvs 386: }
387:
1.4 cvs 388: /*----------------------------------------------------------------------
1.9 cvs 389: IsValidProtocol
390: returns true if the url protocol is supported by Amaya.
1.4 cvs 391: ----------------------------------------------------------------------*/
1.3 cvs 392: #ifdef __STDC__
1.34 cvs 393: boolean IsValidProtocol (const char *url)
1.3 cvs 394: #else /* __STDC__ */
1.9 cvs 395: boolean IsValidProtocol (url)
1.34 cvs 396: const char *url;
1.3 cvs 397: #endif /* __STDC__ */
398: {
1.26 cvs 399: if (!strncmp (url, "http:", 5))
1.22 cvs 400: /* experimental */
1.26 cvs 401: /*** || !strncmp (url, "ftp:", 4)) ***/
1.24 cvs 402: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 403: return (TRUE);
1.5 cvs 404: else
1.8 cvs 405: return (FALSE);
1.3 cvs 406: }
407:
1.31 cvs 408:
409: /*----------------------------------------------------------------------
410: GetBaseURL
411: normalizes orgName according to a base associated with doc, and
412: following the standard URL format rules.
413: The function returns the base used to solve relative URL and SRC:
414: - the base of the document,
415: - or the document path (without document name).
416: ----------------------------------------------------------------------*/
417: #ifdef __STDC__
418: char *GetBaseURL (Document doc)
419: #else /* __STDC__ */
420: char *GetBaseURL (doc)
421: Document doc;
422: #endif /* __STDC__ */
423: {
424: Element el;
425: ElementType elType;
426: AttributeType attrType;
427: Attribute attr;
428: char *ptr, *basename;
429: int length;
430:
431: basename = TtaGetMemory (MAX_LENGTH);
1.39 cvs 432: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
433: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 434: length = MAX_LENGTH -1;
435: /* get the root element */
436: el = TtaGetMainRoot (doc);
437: /* search the BASE element */
438: elType.ElSSchema = TtaGetDocumentSSchema (doc);
439: elType.ElTypeNum = HTML_EL_BASE;
440: el = TtaSearchTypedElement (elType, SearchInTree, el);
441: if (el)
442: {
443: /* The document has a BASE element -> Get the HREF attribute */
444: attrType.AttrSSchema = elType.ElSSchema;
445: attrType.AttrTypeNum = HTML_ATTR_HREF_;
446: attr = TtaGetAttribute (el, attrType);
447: if (attr)
448: {
449: /* Use the base path of the document */
450: TtaGiveTextAttributeValue (attr, basename, &length);
451: /* base and orgName have to be separated by a DIR_SEP */
452: length--;
1.43 cvs 453: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 454: /* verify if the base has the form "protocol://server:port" */
455: {
1.33 cvs 456: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
457: AMAYA_PARSE_HOST |
458: AMAYA_PARSE_PUNCTUATION);
1.31 cvs 459: if (ptr && !strcmp (ptr, basename))
460: {
1.43 cvs 461: /* it has this form, we complete it by adding a URL_STR */
462: if (strchr (basename, DIR_SEP))
463: strcat (basename, DIR_STR);
464: else
465: strcat (basename, URL_STR);
1.31 cvs 466: length++;
467: }
468: if (ptr)
469: TtaFreeMemory (ptr);
470: }
471: }
1.33 cvs 472: }
473:
1.31 cvs 474: /* Remove anything after the last DIR_SEP char. If no such char is found,
475: * then search for the first ":" char, hoping that what's before that is a
476: * protocol. If found, end the string there. If neither char is found,
477: * then discard the whole base element.
478: */
479: length = strlen (basename) - 1;
480: /* search for the last DIR_SEP char */
1.43 cvs 481: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 482: length--;
483: if (length >= 0)
484: /* found the last DIR_SEP char, end the string there */
485: basename[length + 1] = EOS;
486: else
487: /* search for the first PATH_STR char */
488: {
1.32 cvs 489: for (length = 0; basename[length] != ':' &&
1.31 cvs 490: basename[length] != EOS; length ++);
1.32 cvs 491: if (basename[length] == ':')
1.31 cvs 492: /* found, so end the string there */
493: basename[length + 1] = EOS;
494: else
495: /* not found, discard the base */
496: basename[0] = EOS;
497: }
498: return (basename);
499: }
500:
501:
1.4 cvs 502: /*----------------------------------------------------------------------
1.40 cvs 503: GetLocalPath
504: Allocate and return the local document path associated to the url
505: ----------------------------------------------------------------------*/
506: #ifdef __STDC__
507: char *GetLocalPath (Document doc, char *url)
508: #else /* __STDC__ */
509: char *GetLocalPath (doc, url)
510: Document doc;
511: char *url;
512: #endif /* __STDC__ */
513: {
1.46 cvs 514: char *ptr, *n;
1.40 cvs 515: char *documentname;
1.41 cvs 516: char url_sep;
1.40 cvs 517: int len;
518: boolean noFile;
519:
520: if (url != NULL)
521: {
522: /* check whether the file name exists */
523: len = strlen (url) - 1;
1.41 cvs 524: if (IsW3Path)
525: url_sep = '/';
526: else
527: url_sep = DIR_SEP;
528: noFile = (url[len] == url_sep);
1.40 cvs 529: if (noFile)
530: url[len] = EOS;
1.47 ! cvs 531: ptr = TtaGetMemory (MAX_LENGTH);
! 532: documentname = TtaGetMemory (MAX_LENGTH);
1.40 cvs 533: TtaExtractName (url, ptr, documentname);
534: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
535: if (!TtaCheckDirectory (ptr))
536: /* directory did not exist */
537: mkdir (ptr, S_IRWXU);
1.47 ! cvs 538:
! 539: /* don't include the query string within document name */
! 540: n = strrchr(documentname, '?');
! 541: if (n != NULL)
! 542: *n = EOS;
1.46 cvs 543: /* don't include ':' within document name */
544: n = strchr (documentname, ':');
545: if (n != NULL)
546: *n = EOS;
1.40 cvs 547: strcat (ptr, documentname);
548: TtaFreeMemory (documentname);
549: /* restore the url */
550: if (noFile)
1.41 cvs 551: url[len] = url_sep;
1.40 cvs 552: return (ptr);
553: }
554: else
555: return (NULL);
556: }
557:
558:
559: /*----------------------------------------------------------------------
1.9 cvs 560: NormalizeURL
561: normalizes orgName according to a base associated with doc, and
562: following the standard URL format rules.
563: The function returns the new complete and normalized URL
1.12 cvs 564: or file name path (newName) and the name of the document (docName).
1.9 cvs 565: N.B. If the function can't find out what's the docName, it assigns
566: the name "noname.html".
1.4 cvs 567: ----------------------------------------------------------------------*/
1.3 cvs 568: #ifdef __STDC__
569: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName)
570: #else /* __STDC__ */
571: void NormalizeURL (orgName, doc, newName, docName)
572: char *orgName;
573: Document doc;
574: char *newName;
575: char *docName;
576: #endif /* __STDC__ */
577: {
1.31 cvs 578: char *basename;
1.18 cvs 579: char tempOrgName[MAX_LENGTH];
1.5 cvs 580: char *ptr;
581: int length;
582:
1.44 cvs 583: # ifdef _WINDOWS
584: int ndx;
585: # endif /* _WINDOWS */
586:
1.5 cvs 587: if (!newName || !docName)
588: return;
1.18 cvs 589:
1.32 cvs 590: if (doc != 0)
1.33 cvs 591: basename = GetBaseURL (doc);
1.32 cvs 592: else
1.33 cvs 593: basename = (char *) NULL;
1.32 cvs 594:
1.18 cvs 595: /*
1.31 cvs 596: * Clean orgName
597: * Make sure we have a complete orgName, without any leading or trailing
598: * white spaces, or trailinbg new lines
599: */
1.5 cvs 600: ptr = orgName;
1.18 cvs 601: /* skip leading white space and new line characters */
1.19 cvs 602: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS);
1.39 cvs 603: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
604: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 605: /* clean trailing white space */
606: ptr = strchr (tempOrgName, ' ');
607: if (ptr)
608: *ptr = EOS;
609: /* clean trailing new lines */
1.19 cvs 610: ptr = strchr (tempOrgName, EOL);
1.5 cvs 611: if (ptr)
612: *ptr = EOS;
613:
1.18 cvs 614: /*
1.31 cvs 615: * Make orgName a complete URL
616: * If the URL does not include a protocol, then try to calculate
617: * one using the doc's base element (if it exists),
618: */
1.21 cvs 619: if (tempOrgName[0] == EOS)
620: {
621: newName[0] = EOS;
1.36 cvs 622: TtaFreeMemory (basename);
1.21 cvs 623: return;
624: }
625: else if (IsW3Path (tempOrgName))
626: {
627: /* the name is complete, go to the Sixth Step */
628: strcpy (newName, tempOrgName);
629: /* verify if the URL has the form "protocol://server:port" */
1.25 cvs 630: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST |
631: AMAYA_PARSE_PUNCTUATION);
1.21 cvs 632: if (ptr && !strcmp (ptr, newName))
1.43 cvs 633: /* it has this form, we complete it by adding a DIR_STR */
634: strcat (newName, URL_STR);
635:
1.21 cvs 636: if (ptr)
1.25 cvs 637: TtaFreeMemory (ptr);
1.21 cvs 638: }
639: else if ( doc == 0)
1.19 cvs 640: /* the name is complete, go to the Sixth Step */
1.18 cvs 641: strcpy (newName, tempOrgName);
1.5 cvs 642: else
643: {
1.31 cvs 644: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 645: # ifdef _WINDOWS
1.45 cvs 646: if (!IsW3Path (basename)) {
647: length = strlen (tempOrgName);
648: for (ndx = 0; ndx < length; ndx++)
649: if (tempOrgName [ndx] == '/')
650: tempOrgName [ndx] = '\\';
651: }
1.44 cvs 652: # endif /* _WINDOWS */
1.25 cvs 653: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.14 cvs 654: if (ptr)
655: {
1.29 cvs 656: SimplifyUrl (&ptr);
1.14 cvs 657: strcpy (newName, ptr);
1.25 cvs 658: TtaFreeMemory (ptr);
1.14 cvs 659: }
660: else
1.18 cvs 661: newName[0] = EOS;
1.5 cvs 662: }
1.36 cvs 663:
664: TtaFreeMemory (basename);
1.18 cvs 665: /*
1.31 cvs 666: * Prepare the docname that will refer to this ressource in the
667: * .amaya directory. If the new URL finishes on DIR_SEP, then use
668: * noname.html as a default ressource name
1.18 cvs 669: */
1.19 cvs 670: if (newName[0] != EOS)
1.5 cvs 671: {
1.19 cvs 672: length = strlen (newName) - 1;
1.43 cvs 673: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.18 cvs 674: {
675: /* docname was not comprised inside the URL, so let's */
676: /* assign the default ressource name */
677: strcpy (docName, "noname.html");
678: /* remove DIR_SEP at the end of complete path */
1.23 cvs 679: /* newName[length] = EOS; */
1.18 cvs 680: }
1.14 cvs 681: else
1.18 cvs 682: {
683: /* docname is comprised inside the URL */
1.43 cvs 684: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP)
1.18 cvs 685: length--;
686: if (length < 0)
687: strcpy (docName, newName);
688: else
689: strcpy (docName, &newName[length+1]);
690: }
1.5 cvs 691: }
1.18 cvs 692: else
693: docName[0] = EOS;
694: }
1.3 cvs 695:
1.4 cvs 696: /*----------------------------------------------------------------------
1.9 cvs 697: IsSameHost
1.4 cvs 698: ----------------------------------------------------------------------*/
1.3 cvs 699: #ifdef __STDC__
1.34 cvs 700: boolean IsSameHost (const char *url1, const char *url2)
1.3 cvs 701: #else /* __STDC__ */
702: boolean IsSameHost (url1, url2)
1.34 cvs 703: const char *url1;
704: const char *url2;
1.3 cvs 705: #endif /* __STDC__ */
706: {
1.34 cvs 707: char *basename_ptr1, *basename_ptr2;
708: boolean result;
1.3 cvs 709:
1.25 cvs 710: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
711: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 712:
1.5 cvs 713: if (strcmp (basename_ptr1, basename_ptr2))
1.8 cvs 714: result = FALSE;
1.5 cvs 715: else
1.8 cvs 716: result = TRUE;
1.3 cvs 717:
1.25 cvs 718: TtaFreeMemory (basename_ptr1);
719: TtaFreeMemory (basename_ptr2);
1.5 cvs 720: return (result);
1.3 cvs 721: }
722:
723:
1.4 cvs 724: /*----------------------------------------------------------------------
1.22 cvs 725: HasKnownFileSuffix
726: returns TRUE if path points to a file ending with a suffix.
727: ----------------------------------------------------------------------*/
728: #ifdef __STDC__
1.34 cvs 729: boolean HasKnownFileSuffix (const char *path)
1.22 cvs 730: #else /* __STDC__ */
731: boolean HasKnownFileSuffix (path)
1.34 cvs 732: const char *path;
1.22 cvs 733: #endif /* __STDC__ */
734: {
1.29 cvs 735: char *root;
736: char temppath[MAX_LENGTH];
737: char suffix[MAX_LENGTH];
1.22 cvs 738:
1.24 cvs 739: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 740: return (FALSE);
741:
1.29 cvs 742: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 743:
744: if (root)
745: {
746: strcpy (temppath, root);
1.25 cvs 747: TtaFreeMemory (root);
1.22 cvs 748: /* Get the suffix */
749: ExtractSuffix (temppath, suffix);
750:
751: if( suffix[0] == EOS)
752: /* no suffix */
753: return (FALSE);
754:
755: /* Normalize the suffix */
756: ConvertToLowerCase (suffix);
757:
1.23 cvs 758: if (!strcmp (suffix, "gz"))
1.22 cvs 759: /* skip the compressed suffix */
760: {
761: ExtractSuffix (temppath, suffix);
762: if(suffix[0] == EOS)
763: /* no suffix */
764: return (FALSE);
765: /* Normalize the suffix */
766: ConvertToLowerCase (suffix);
767: }
768:
769: if ((strcmp (suffix, "gif")) && (strcmp (suffix, "xbm")) &&
770: (strcmp (suffix, "xpm")) && (strcmp (suffix, "jpg")) &&
771: (strcmp (suffix, "pdf")) && (strcmp (suffix, "png")) &&
772: (strcmp (suffix, "tgz")) && (strcmp (suffix, "xpg")) &&
773: (strcmp (suffix, "xpd")) && (strcmp (suffix, "ps")) &&
774: (strcmp (suffix, "au")) && (strcmp (suffix, "html")) &&
775: (strcmp (suffix, "htm")) && (strcmp (suffix, "shtml")) &&
776: (strcmp (suffix, "txt")) && (strcmp (suffix, "css")) &&
777: (strcmp (suffix, "eps")))
778: return (FALSE);
779: else
780: return (TRUE);
781: }
782: else
783: return (FALSE);
784: }
785:
786:
787: /*----------------------------------------------------------------------
1.24 cvs 788: ChopURL
789: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
790: If inputURL is bigger than that size, outputURL receives
791: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
792: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
793: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
794: copied into outputURL.
795: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
796: chars.
797: ----------------------------------------------------------------------*/
798: #ifdef __STDC__
1.34 cvs 799: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 800: #else
801: void ChopURL (outputURL, inputURL)
1.34 cvs 802: char *outputURL;
803: const char *inputURL;
1.24 cvs 804: #endif
1.22 cvs 805:
1.24 cvs 806: {
807: int len;
1.9 cvs 808:
1.24 cvs 809: len = strlen (inputURL);
810: if (len <= MAX_PRINT_URL_LENGTH)
1.29 cvs 811: strcpy (outputURL, inputURL);
1.24 cvs 812: else
813: /* make a truncated urlName on the status window */
814: {
815: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
816: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
817: strcat (outputURL, "...");
818: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
819: }
1.25 cvs 820: }
821:
822:
823: /*----------------------------------------------------------------------
824: scan
1.47 ! cvs 825: Scan a filename for its constituents
1.25 cvs 826: -----------------------------------
827:
828: On entry,
829: name points to a document name which may be incomplete.
830: On exit,
831: absolute or relative may be nonzero (but not both).
832: host, fragment and access may be nonzero if they were specified.
833: Any which are nonzero point to zero terminated strings.
834: ----------------------------------------------------------------------*/
835: #ifdef __STDC__
836: static void scan (char * name, HTURI * parts)
837: #else /* __STDC__ */
838: static void scan (name, parts)
839: char *name;
840: HTURI *parts;
841:
842: #endif /* __STDC__ */
843: {
1.43 cvs 844: char *p;
845: char *after_access = name;
1.32 cvs 846:
1.43 cvs 847: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 848: /* Look for fragment identifier */
849: if ((p = strrchr(name, '#')) != NULL)
850: {
851: *p++ = '\0';
852: parts->fragment = p;
1.25 cvs 853: }
854:
1.28 cvs 855: for (p=name; *p; p++)
856: {
1.43 cvs 857: if (*p == URL_SEP || *p == DIR_SEP || *p=='#' || *p=='?')
1.28 cvs 858: break;
1.32 cvs 859: if (*p==':')
1.28 cvs 860: {
861: *p = 0;
862: parts->access = after_access; /* Scheme has been specified */
863:
864: /* The combination of gcc, the "-O" flag and the HP platform is
865: unhealthy. The following three lines is a quick & dirty fix, but is
866: not recommended. Rather, turn off "-O". */
867:
868: /* after_access = p;*/
869: /* while (*after_access == 0)*/
870: /* after_access++;*/
871: after_access = p+1;
1.43 cvs 872: if (!strcasecmp("URL", parts->access))
1.28 cvs 873: /* Ignore IETF's URL: pre-prefix */
874: parts->access = NULL;
875: else
1.25 cvs 876: break;
877: }
878: }
879:
880: p = after_access;
1.43 cvs 881: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 882: {
1.43 cvs 883: if (p[1] == URL_SEP)
1.28 cvs 884: {
1.25 cvs 885: parts->host = p+2; /* host has been specified */
1.28 cvs 886: *p = 0; /* Terminate access */
887: /* look for end of host name if any */
1.43 cvs 888: p = strchr (parts->host, URL_SEP);
1.28 cvs 889: if (p)
890: {
1.43 cvs 891: *p = EOS; /* Terminate host */
1.25 cvs 892: parts->absolute = p+1; /* Root has been found */
1.28 cvs 893: }
894: }
895: else
896: /* Root found but no host */
897: parts->absolute = p+1;
898: }
899: else
900: {
1.25 cvs 901: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 902: }
1.25 cvs 903: }
904:
905:
906: /*----------------------------------------------------------------------
1.28 cvs 907: AmayaParseUrl: parse a Name relative to another name
908:
909: This returns those parts of a name which are given (and requested)
910: substituting bits from the related name where necessary.
1.25 cvs 911:
1.28 cvs 912: On entry,
1.25 cvs 913: aName A filename given
914: relatedName A name relative to which aName is to be parsed. Give
915: it an empty string if aName is absolute.
916: wanted A mask for the bits which are wanted.
917:
1.28 cvs 918: On exit,
1.25 cvs 919: returns A pointer to a malloc'd string which MUST BE FREED
920: ----------------------------------------------------------------------*/
921: #ifdef __STDC__
1.34 cvs 922: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1.25 cvs 923: #else /* __STDC__ */
1.28 cvs 924: char *AmayaParseUrl (aName, relatedName, wanted)
1.34 cvs 925: const char *aName;
1.28 cvs 926: char *relatedName;
927: int wanted;
1.25 cvs 928:
929: #endif /* __STDC__ */
930: {
1.29 cvs 931: char *return_value;
932: char result[MAX_LENGTH];
933: char name[MAX_LENGTH];
934: char rel[MAX_LENGTH];
935: char *p, *access;
936: HTURI given, related;
937: int len;
1.33 cvs 938: char used_sep;
939: char *used_str;
1.32 cvs 940:
1.43 cvs 941: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 942: {
1.42 cvs 943: used_str = DIR_STR;
944: used_sep = DIR_SEP;
1.33 cvs 945: }
1.32 cvs 946: else
1.33 cvs 947: {
1.42 cvs 948: used_str = URL_STR;
949: used_sep = URL_SEP;
1.33 cvs 950: }
1.32 cvs 951:
1.29 cvs 952: /* Make working copies of input strings to cut up: */
953: return_value = NULL;
954: result[0] = 0; /* Clear string */
955: strcpy (name, aName);
956: if (relatedName != NULL)
957: strcpy (rel, relatedName);
958: else
959: relatedName[0] = EOS;
960:
961: scan (name, &given);
962: scan (rel, &related);
963: access = given.access ? given.access : related.access;
964: if (wanted & AMAYA_PARSE_ACCESS)
965: if (access)
966: {
967: strcat (result, access);
968: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.32 cvs 969: strcat (result, ":");
1.29 cvs 970: }
971:
972: if (given.access && related.access)
973: /* If different, inherit nothing. */
974: if (strcmp (given.access, related.access) != 0)
975: {
976: related.host = 0;
977: related.absolute = 0;
978: related.relative = 0;
979: related.fragment = 0;
980: }
981:
982: if (wanted & AMAYA_PARSE_HOST)
983: if(given.host || related.host)
984: {
985: if(wanted & AMAYA_PARSE_PUNCTUATION)
986: strcat (result, "//");
987: strcat (result, given.host ? given.host : related.host);
988: }
989:
990: if (given.host && related.host)
991: /* If different hosts, inherit no path. */
992: if (strcmp(given.host, related.host) != 0)
993: {
994: related.absolute = 0;
995: related.relative = 0;
996: related.fragment = 0;
997: }
998:
999: if (wanted & AMAYA_PARSE_PATH)
1000: {
1001: if (given.absolute)
1002: {
1003: /* All is given */
1004: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.33 cvs 1005: strcat (result, used_str);
1.29 cvs 1006: strcat (result, given.absolute);
1.25 cvs 1007: }
1.29 cvs 1008: else if (related.absolute)
1009: {
1010: /* Adopt path not name */
1.33 cvs 1011: strcat (result, used_str);
1.29 cvs 1012: strcat (result, related.absolute);
1013: if (given.relative)
1014: {
1015: /* Search part? */
1016: p = strchr (result, '?');
1017: if (!p)
1018: p=result+strlen(result)-1;
1.33 cvs 1019: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1020: /* Remove filename */
1021: p[1]=0;
1022: /* Add given one */
1023: strcat (result, given.relative);
1024: /*SimplifyUrl (&result);*/
1.25 cvs 1025: }
1026: }
1.29 cvs 1027: else if (given.relative)
1028: /* what we've got */
1029: strcat (result, given.relative);
1030: else if (related.relative)
1031: strcat (result, related.relative);
1032: else
1033: /* No inheritance */
1.33 cvs 1034: strcat (result, used_str);
1.25 cvs 1035: }
1.29 cvs 1036:
1037: if (wanted & AMAYA_PARSE_ANCHOR)
1038: if (given.fragment || related.fragment)
1039: {
1040: if (given.absolute && given.fragment)
1041: {
1042: /*Fixes for relURLs...*/
1043: if (wanted & AMAYA_PARSE_PUNCTUATION)
1044: strcat (result, "#");
1045: strcat (result, given.fragment);
1046: }
1047: else if (!(given.absolute) && !(given.fragment))
1048: strcat (result, "");
1049: else
1050: {
1051: if (wanted & AMAYA_PARSE_PUNCTUATION)
1052: strcat (result, "#");
1053: strcat (result, given.fragment ? given.fragment : related.fragment);
1054: }
1055: }
1056: len = strlen (result);
1057: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1058: strcpy (return_value, result);
1059: return (return_value); /* exactly the right length */
1.25 cvs 1060: }
1061:
1062: /*----------------------------------------------------------------------
1063: HTCanon
1064: Canonicalizes the URL in the following manner starting from the host
1065: pointer:
1066:
1067: 1) The host name is converted to lowercase
1068: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1069:
1070: Return: OK The position of the current path part of the URL
1071: which might be the old one or a new one.
1072:
1073: ----------------------------------------------------------------------*/
1074: #ifdef __STDC__
1.28 cvs 1075: static char *HTCanon (char ** filename, char * host)
1.25 cvs 1076: #else /* __STDC__ */
1.28 cvs 1077: static char *HTCanon (filename, host)
1078: char **filename;
1079: char *host;
1.25 cvs 1080: #endif /* __STDC__ */
1081: {
1082: char *newname = NULL;
1083: char *port;
1084: char *strptr;
1085: char *path;
1086: char *access = host-3;
1.33 cvs 1087: char used_sep;
1.32 cvs 1088:
1089:
1.33 cvs 1090: if (*filename && strchr (*filename, URL_SEP))
1091: {
1092: used_sep = URL_SEP;
1093: }
1094: else
1095: {
1096: used_sep = DIR_SEP;
1097: }
1.32 cvs 1098:
1.33 cvs 1099: while (access>*filename && *(access-1)!= used_sep) /* Find access method */
1.25 cvs 1100: access--;
1.33 cvs 1101: if ((path = strchr(host, used_sep)) == NULL) /* Find path */
1.25 cvs 1102: path = host + strlen(host);
1103: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */
1104: host = strptr;
1.32 cvs 1105: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */
1.25 cvs 1106: port = NULL;
1107:
1108: strptr = host; /* Convert to lower-case */
1.33 cvs 1109: while (strptr<path)
1110: {
1.25 cvs 1111: *strptr = tolower(*strptr);
1112: strptr++;
1.33 cvs 1113: }
1.25 cvs 1114:
1115: /* Does the URL contain a full domain name? This also works for a
1116: numerical host name. The domain name is already made lower-case
1117: and without a trailing dot. */
1118: {
1.33 cvs 1119: char *dot = port ? port : path;
1120: if (dot > *filename && *--dot=='.')
1121: {
1122: char *orig=dot, *dest=dot+1;
1123: while((*orig++ = *dest++));
1124: if (port) port--;
1125: path--;
1.25 cvs 1126: }
1127: }
1128: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1129: if (port)
1130: {
1131: if (!*(port+1) || *(port+1)==used_sep)
1132: {
1133: if (!newname)
1134: {
1.25 cvs 1135: char *orig=port, *dest=port+1;
1136: while((*orig++ = *dest++));
1.33 cvs 1137: }
1138: }
1139: else if ((!strncmp(access, "http", 4) &&
1140: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1141: (!strncmp(access, "gopher", 6) &&
1142: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1143: (!strncmp(access, "ftp", 3) &&
1144: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) {
1145: if (!newname)
1146: {
1147: char *orig=port, *dest=port+3;
1148: while((*orig++ = *dest++));
1149: /* Update path position, Henry Minsky */
1150: path -= 3;
1.25 cvs 1151: }
1.33 cvs 1152: }
1153: else if (newname)
1154: strncat(newname, port, (int) (path-port));
1155: }
1.25 cvs 1156:
1.33 cvs 1157: if (newname)
1158: {
1.25 cvs 1159: char *newpath = newname+strlen(newname);
1160: strcat(newname, path);
1161: path = newpath;
1.28 cvs 1162: /* Free old copy */
1163: TtaFreeMemory(*filename);
1.25 cvs 1164: *filename = newname;
1.33 cvs 1165: }
1.25 cvs 1166: return path;
1167: }
1168:
1169:
1170: /*----------------------------------------------------------------------
1.29 cvs 1171: SimplifyUrl: simplify a URI
1.32 cvs 1172: A URI is allowed to contain the sequence xxx/../ which may be
1173: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1174: Simplification helps us recognize duplicate URIs.
1.25 cvs 1175:
1.28 cvs 1176: Thus, /etc/junk/../fred becomes /etc/fred
1177: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1178:
1.28 cvs 1179: but we should NOT change
1180: http://fred.xxx.edu/../..
1.25 cvs 1181:
1182: or ../../albert.html
1183:
1.28 cvs 1184: In order to avoid empty URLs the following URLs become:
1.25 cvs 1185:
1186: /fred/.. becomes /fred/..
1187: /fred/././.. becomes /fred/..
1188: /fred/.././junk/.././ becomes /fred/..
1189:
1.28 cvs 1190: If more than one set of `://' is found (several proxies in cascade) then
1191: only the part after the last `://' is simplified.
1.25 cvs 1192:
1.28 cvs 1193: Returns: A string which might be the old one or a new one.
1.25 cvs 1194: ----------------------------------------------------------------------*/
1195: #ifdef __STDC__
1.29 cvs 1196: void SimplifyUrl (char ** url)
1.25 cvs 1197: #else /* __STDC__ */
1.29 cvs 1198: void SimplifyUrl (url)
1.28 cvs 1199: char **url;
1.25 cvs 1200: #endif /* __STDC__ */
1201: {
1.28 cvs 1202: char *path, *p;
1203: char *newptr, *access;
1204: char *orig, *dest, *end;
1205:
1.33 cvs 1206: char used_sep;
1.32 cvs 1207:
1208:
1.28 cvs 1209: if (!url || !*url)
1210: return;
1211:
1.33 cvs 1212: if (strchr (*url, URL_SEP))
1213: {
1214: used_sep = URL_SEP;
1215: }
1.32 cvs 1216: else
1.33 cvs 1217: {
1218: used_sep = DIR_SEP;
1219: }
1.32 cvs 1220:
1.28 cvs 1221: /* Find any scheme name */
1222: if ((path = strstr(*url, "://")) != NULL)
1.33 cvs 1223: {
1224: /* Find host name */
1.28 cvs 1225: access = *url;
1226: while (access<path && (*access=tolower(*access)))
1227: access++;
1228: path += 3;
1229: while ((newptr = strstr(path, "://")) != NULL)
1230: /* For proxies */
1231: path = newptr+3;
1232: /* We have a host name */
1233: path = HTCanon(url, path);
1.25 cvs 1234: }
1.28 cvs 1235: else if ((path = strstr(*url, ":/")) != NULL)
1236: path += 2;
1237: else
1238: path = *url;
1.25 cvs 1239:
1.33 cvs 1240: if (*path == used_sep && *(path+1)==used_sep)
1.28 cvs 1241: /* Some URLs start //<foo> */
1242: path += 1;
1243: else if (!strncmp(path, "news:", 5))
1244: {
1245: newptr = strchr(path+5, '@');
1246: if (!newptr)
1247: newptr = path + 5;
1248: while (*newptr)
1249: {
1250: /* Make group or host lower case */
1251: *newptr = tolower (*newptr);
1252: newptr++;
1.25 cvs 1253: }
1.28 cvs 1254: /* Doesn't need to do any more */
1255: return;
1.25 cvs 1256: }
1.28 cvs 1257:
1258: if ((p = path))
1259: {
1260: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1261: (end = strchr (path, '#'))))
1262: end = path + strlen (path);
1263:
1264: /* Parse string second time to simplify */
1265: p = path;
1266: while (p < end)
1267: {
1.33 cvs 1268: if (*p==used_sep)
1.28 cvs 1269: {
1.33 cvs 1270: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1271: {
1272: orig = p + 1;
1.33 cvs 1273: dest = (*(p+2)!=used_sep) ? p+2 : p+3;
1.28 cvs 1274: while ((*orig++ = *dest++)); /* Remove a slash and a dot */
1275: end = orig - 1;
1276: }
1.33 cvs 1277: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3)))
1.28 cvs 1278: {
1279: newptr = p;
1.33 cvs 1280: while (newptr>path && *--newptr!=used_sep); /* prev slash */
1.28 cvs 1281: if (strncmp(newptr, "/../", 4))
1282: {
1283: orig = newptr + 1;
1.33 cvs 1284: dest = (*(p+3)!=used_sep) ? p+3 : p+4;
1.28 cvs 1285: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1286: end = orig-1;
1287: /* Start again with prev slash */
1288: p = newptr;
1.25 cvs 1289: }
1.28 cvs 1290: else
1.25 cvs 1291: p++;
1.28 cvs 1292: }
1.33 cvs 1293: else if (*(p+1) == used_sep)
1.28 cvs 1294: {
1.33 cvs 1295: while (*(p+1) == used_sep)
1.28 cvs 1296: {
1297: orig = p;
1298: dest = p + 1;
1299: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1300: end = orig-1;
1301: }
1302: }
1303: else
1.25 cvs 1304: p++;
1.28 cvs 1305: }
1306: else
1307: p++;
1.25 cvs 1308: }
1309: }
1.28 cvs 1310: return;
1311: }
1312:
1313:
1314: /*----------------------------------------------------------------------
1315: NormalizeFile normalizes local names.
1316: Return TRUE if target and src differ.
1317: ----------------------------------------------------------------------*/
1318: #ifdef __STDC__
1319: boolean NormalizeFile (char *src, char *target)
1320: #else
1321: boolean NormalizeFile (src, target)
1322: char *src;
1323: char *target;
1324:
1325: #endif
1326: {
1327: char *s;
1328: boolean change;
1329:
1330: change = FALSE;
1331: if (src[0] == '~')
1332: {
1333: /* replace ~ */
1334: s = (char *) TtaGetEnvString ("HOME");
1335: strcpy (target, s);
1336: strcat (target, &src[1]);
1337: change = TRUE;
1338: }
1339: else if (strncmp (src, "file:", 5) == 0)
1340: {
1341: /* remove the prefix file: */
1342: if (src[5] == EOS)
1343: strcpy (target, DIR_STR);
1344: else if (src[0] == '~')
1345: {
1346: /* replace ~ */
1347: s = (char *) TtaGetEnvString ("HOME");
1348: strcpy (target, s);
1349: strcat (target, &src[5]);
1350: }
1351: else
1352: strcpy (target, &src[5]);
1353: change = TRUE;
1354: }
1355: else
1356: strcpy (target, src);
1357:
1358: /* remove /../ and /./ */
1.29 cvs 1359: SimplifyUrl (&target);
1.28 cvs 1360: return (change);
1.25 cvs 1361: }
1362:
1.28 cvs 1363:
1.25 cvs 1364: /*----------------------------------------------------------------------
1.31 cvs 1365: MakeRelativeURL: make relative name
1.25 cvs 1366:
1.28 cvs 1367: This function creates and returns a string which gives an expression of
1368: one address as related to another. Where there is no relation, an absolute
1369: address is retured.
1.25 cvs 1370:
1.28 cvs 1371: On entry,
1.25 cvs 1372: Both names must be absolute, fully qualified names of nodes
1373: (no fragment bits)
1374:
1.28 cvs 1375: On exit,
1.25 cvs 1376: The return result points to a newly allocated name which, if
1377: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1378: The caller is responsible for freeing the resulting name later.
1379: ----------------------------------------------------------------------*/
1380: #ifdef __STDC__
1.31 cvs 1381: char *MakeRelativeURL (char *aName, char *relatedName)
1.25 cvs 1382: #else /* __STDC__ */
1.31 cvs 1383: char *MakeRelativeURL (aName, relatedName)
1.28 cvs 1384: char *aName;
1385: char *relatedName;
1.25 cvs 1386: #endif /* __STDC__ */
1387: {
1.39 cvs 1388: char *return_value;
1389: char result[MAX_LENGTH];
1.29 cvs 1390: char *p;
1.34 cvs 1391: char *q;
1.29 cvs 1392: char *after_access;
1393: char *last_slash = NULL;
1394: int slashes, levels, len;
1395:
1.44 cvs 1396: # ifdef _WINDOWS
1397: int ndx;
1398: # endif /* _WINDOWS */
1399:
1.29 cvs 1400: if (aName == NULL || relatedName == NULL)
1401: return (NULL);
1402:
1403: slashes = 0;
1404: after_access = NULL;
1405: p = aName;
1406: q = relatedName;
1407: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1408: {
1409: /* Find extent of match */
1.32 cvs 1410: if (*p == ':')
1.29 cvs 1411: after_access = p + 1;
1.28 cvs 1412: if (*p == DIR_SEP)
1.27 cvs 1413: {
1.29 cvs 1414: /* memorize the last slash position and count them */
1.27 cvs 1415: last_slash = p;
1416: slashes++;
1.25 cvs 1417: }
1418: }
1419:
1.31 cvs 1420: /* q, p point to the first non-matching character or zero */
1421: if (*q == EOS)
1422: {
1423: /* New name is a subset of the related name */
1424: /* exactly the right length */
1425: len = strlen (p);
1426: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1427: strcpy (return_value, p);
1428: }
1429: else if ((slashes < 2 && after_access == NULL)
1430: || (slashes < 3 && after_access != NULL))
1431: {
1432: /* Two names whitout common path */
1433: /* exactly the right length */
1434: len = strlen (aName);
1435: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1436: strcpy (return_value, aName);
1437: }
1438: else
1439: {
1440: /* Some path in common */
1441: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1442: /* just the same server */
1443: strcpy (result, last_slash);
1444: else
1445: {
1446: levels= 0;
1447: for (; *q && (*q != '#'); q++)
1448: if (*q == DIR_SEP)
1449: levels++;
1450:
1451: result[0] = 0;
1452: for (;levels; levels--)
1453: strcat (result, "../");
1454: strcat (result, last_slash+1);
1455: }
1456:
1457: /* exactly the right length */
1458: len = strlen (result);
1459: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1460: strcpy (return_value, result);
1.25 cvs 1461: }
1.44 cvs 1462: # ifdef _WINDOWS
1463: len = strlen (return_value);
1464: for (ndx = 0; ndx < len; ndx ++)
1465: if (return_value[ndx] == '\\')
1466: return_value[ndx] = '/' ;
1467: # endif /* _WINDOWS */
1.29 cvs 1468: return (return_value);
1.24 cvs 1469: }
1.35 cvs 1470:
1471:
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