Annotation of Amaya/amaya/AHTURLTools.c, revision 1.64
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:
1.61 cvs 177:
178: /*----------------------------------------------------------------------
179: ExtractSuffix extract suffix from document nane.
180: ----------------------------------------------------------------------*/
181: #ifdef __STDC__
182: void ExtractSuffix (STRING aName, STRING aSuffix)
183: #else
184: void ExtractSuffix (aName, aSuffix)
185: STRING aName;
186: STRING aSuffix;
187:
188: #endif
189: {
190: int lg, i;
191: STRING ptr, oldptr;
192:
193: if (!aSuffix || !aName)
194: /* bad suffix */
195: return;
196:
197: aSuffix[0] = EOS;
198: lg = ustrlen (aName);
199: if (lg)
200: {
201: /* the name is not empty */
202: oldptr = ptr = &aName[0];
203: do
204: {
205: ptr = ustrrchr (oldptr, '.');
206: if (ptr)
207: oldptr = &ptr[1];
208: }
209: while (ptr);
210:
211: i = (int) (oldptr) - (int) (aName); /* name length */
212: if (i > 1)
213: {
214: aName[i - 1] = EOS;
215: if (i != lg)
216: ustrcpy (aSuffix, oldptr);
217: }
218: }
219: }
220:
1.4 cvs 221: /*----------------------------------------------------------------------
1.9 cvs 222: IsHTMLName
223: returns TRUE if path points to an HTML resource.
1.4 cvs 224: ----------------------------------------------------------------------*/
1.3 cvs 225: #ifdef __STDC__
1.34 cvs 226: boolean IsHTMLName (const char *path)
1.3 cvs 227: #else /* __STDC__ */
228: boolean IsHTMLName (path)
1.34 cvs 229: const char *path;
1.3 cvs 230: #endif /* __STDC__ */
231: {
1.5 cvs 232: char temppath[MAX_LENGTH];
233: char suffix[MAX_LENGTH];
234: char nsuffix[MAX_LENGTH];
235: int i;
236:
237: if (!path)
1.37 cvs 238: return (FALSE);
1.5 cvs 239:
240: strcpy (temppath, path);
241: ExtractSuffix (temppath, suffix);
242:
243: /* Normalize the suffix */
244: i = 0;
1.39 cvs 245: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 246: {
1.25 cvs 247: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 248: i++;
249: }
1.5 cvs 250: nsuffix[i] = EOS;
1.60 cvs 251: if (!strcmp (nsuffix, "html") ||
252: !strcmp (nsuffix, "htm") ||
1.63 cvs 253: !strcmp (nsuffix, "shtml") ||
254: !strcmp (nsuffix, "xht") ||
255: !strcmp (nsuffix, "xhtm") ||
256: !strcmp (nsuffix, "xhtml"))
1.39 cvs 257: return (TRUE);
1.22 cvs 258: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 259: {
1.39 cvs 260: /* take into account compressed files */
1.13 cvs 261: ExtractSuffix (temppath, suffix);
262: /* Normalize the suffix */
263: i = 0;
1.39 cvs 264: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 265: {
1.25 cvs 266: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 267: i++;
268: }
269: nsuffix[i] = EOS;
1.60 cvs 270: if (!strcmp (nsuffix, "html") ||
271: !strcmp (nsuffix, "htm") ||
1.63 cvs 272: !strcmp (nsuffix, "shtml") ||
273: !strcmp (nsuffix, "xht") ||
274: !strcmp (nsuffix, "xhtm") ||
275: !strcmp (nsuffix, "xhtml"))
276:
1.39 cvs 277: return (TRUE);
278: else
1.13 cvs 279: return (FALSE);
280: }
281: else
1.39 cvs 282: return (FALSE);
1.3 cvs 283: }
284:
1.4 cvs 285: /*----------------------------------------------------------------------
1.56 cvs 286: IsXMLName
287: returns TRUE if path points to an XML resource.
288: ----------------------------------------------------------------------*/
289: #ifdef __STDC__
290: boolean IsXMLName (const char *path)
291: #else /* __STDC__ */
292: boolean IsXMLName (path)
293: const char *path;
294: #endif /* __STDC__ */
295: {
296: char temppath[MAX_LENGTH];
297: char suffix[MAX_LENGTH];
298:
299: if (!path)
300: return (FALSE);
301:
302: strcpy (temppath, path);
303: ExtractSuffix (temppath, suffix);
304:
1.64 ! cvs 305: if (!strcasecmp (suffix, "xml") ||
! 306: !strcasecmp (suffix, "xht") ||
! 307: !strcmp (suffix, "xhtm") ||
! 308: !strcmp (suffix, "xhtml"))
1.56 cvs 309: return (TRUE);
310: else if (!strcmp (suffix, "gz"))
311: {
312: /* take into account compressed files */
313: ExtractSuffix (temppath, suffix);
1.64 ! cvs 314: if (!strcasecmp (suffix, "xml") ||
! 315: !strcasecmp (suffix, "xht") ||
! 316: !strcmp (suffix, "xhtm") ||
! 317: !strcmp (suffix, "xhtml"))
1.60 cvs 318: return (TRUE);
319: else
320: return (FALSE);
321: }
322: else
323: return (FALSE);
324: }
325:
326: /*----------------------------------------------------------------------
327: IsCSSName
328: returns TRUE if path points to an XML resource.
329: ----------------------------------------------------------------------*/
330: #ifdef __STDC__
331: boolean IsCSSName (const char *path)
332: #else /* __STDC__ */
333: boolean IsCSSName (path)
334: const char *path;
335: #endif /* __STDC__ */
336: {
337: char temppath[MAX_LENGTH];
338: char suffix[MAX_LENGTH];
339:
340: if (!path)
341: return (FALSE);
342:
343: strcpy (temppath, path);
344: ExtractSuffix (temppath, suffix);
345:
346: if (!strcasecmp (suffix, "css"))
347: return (TRUE);
348: else if (!strcmp (suffix, "gz"))
349: {
350: /* take into account compressed files */
351: ExtractSuffix (temppath, suffix);
352: if (!strcasecmp (suffix, "css"))
1.56 cvs 353: return (TRUE);
354: else
355: return (FALSE);
356: }
357: else
358: return (FALSE);
359: }
360:
361: /*----------------------------------------------------------------------
1.9 cvs 362: IsImageName
363: returns TRUE if path points to an image resource.
1.4 cvs 364: ----------------------------------------------------------------------*/
1.3 cvs 365: #ifdef __STDC__
1.34 cvs 366: boolean IsImageName (const char *path)
1.3 cvs 367: #else /* __STDC__ */
368: boolean IsImageName (path)
1.34 cvs 369: const char *path;
1.3 cvs 370: #endif /* __STDC__ */
371: {
1.5 cvs 372: char temppath[MAX_LENGTH];
373: char suffix[MAX_LENGTH];
374: char nsuffix[MAX_LENGTH];
375: int i;
376:
377: if (!path)
1.13 cvs 378: return (FALSE);
1.5 cvs 379:
380: strcpy (temppath, path);
381: ExtractSuffix (temppath, suffix);
382:
383: /* Normalize the suffix */
384: i = 0;
1.39 cvs 385: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 386: {
1.25 cvs 387: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 388: i++;
389: }
1.5 cvs 390: nsuffix[i] = EOS;
1.39 cvs 391: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
392: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
393: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
394: return (TRUE);
395: return (FALSE);
1.3 cvs 396: }
397:
1.4 cvs 398: /*----------------------------------------------------------------------
1.58 cvs 399: IsImageType
400: returns TRUE if type points to an image resource.
401: ----------------------------------------------------------------------*/
402: #ifdef __STDC__
403: boolean IsImageType (const char *type)
404: #else /* __STDC__ */
405: boolean IsImageType (type)
406: const char *type;
407: #endif /* __STDC__ */
408: {
409: char temptype[MAX_LENGTH];
410: int i;
411:
412: if (!type)
413: return (FALSE);
414:
415: strcpy (temptype, type);
416: /* Normalize the type */
417: i = 0;
418: while (temptype[i] != EOS)
419: {
420: temptype[i] = tolower (temptype[i]);
421: i++;
422: }
423: if ((!strcmp (temptype, "gif")) || (!strcmp (temptype, "x-xbitmap")) ||
424: (!strcmp (temptype, "x-xpixmap")) || (!strcmp (temptype, "jpeg")) ||
425: (!strcmp (temptype, "png")))
426: return (TRUE);
427: return (FALSE);
428: }
429:
430: /*----------------------------------------------------------------------
1.9 cvs 431: IsTextName
1.4 cvs 432: ----------------------------------------------------------------------*/
1.3 cvs 433: #ifdef __STDC__
1.34 cvs 434: boolean IsTextName (const char *path)
1.3 cvs 435: #else /* __STDC__ */
436: boolean IsTextName (path)
1.34 cvs 437: const char *path;
1.3 cvs 438:
439: #endif /* __STDC__ */
440: {
1.5 cvs 441: char temppath[MAX_LENGTH];
442: char suffix[MAX_LENGTH];
443: char nsuffix[MAX_LENGTH];
444: int i;
445:
446: if (!path)
1.13 cvs 447: return (FALSE);
1.5 cvs 448:
449: strcpy (temppath, path);
450: ExtractSuffix (temppath, suffix);
451:
452: /* Normalize the suffix */
453: i = 0;
1.39 cvs 454: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 455: {
1.25 cvs 456: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 457: i++;
458: }
459: nsuffix[i] = EOS;
460:
1.39 cvs 461: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 462: return (TRUE);
1.22 cvs 463: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 464: {
1.39 cvs 465: /* take into account compressed files */
1.13 cvs 466: ExtractSuffix (temppath, suffix);
467: /* Normalize the suffix */
468: i = 0;
1.39 cvs 469: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 470: {
1.25 cvs 471: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 472: i++;
473: }
474: nsuffix[i] = EOS;
1.39 cvs 475: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 476: return (TRUE);
477: else
478: return (FALSE);
479: }
480: else
481: return (FALSE);
1.3 cvs 482: }
483:
1.4 cvs 484: /*----------------------------------------------------------------------
1.9 cvs 485: IsHTTPPath
486: returns TRUE if path is in fact an http URL.
1.4 cvs 487: ----------------------------------------------------------------------*/
1.3 cvs 488: #ifdef __STDC__
1.34 cvs 489: boolean IsHTTPPath (const char *path)
1.3 cvs 490: #else /* __STDC__ */
491: boolean IsHTTPPath (path)
1.34 cvs 492: const char *path;
1.3 cvs 493: #endif /* __STDC__ */
494: {
1.5 cvs 495: if (!path)
496: return FALSE;
1.3 cvs 497:
1.58 cvs 498: if ((!strncmp (path, "http:", 5) != 0)
499: || !strncmp (path, "internal:", 9))
500: return TRUE;
501: return FALSE;
1.3 cvs 502: }
503:
1.4 cvs 504: /*----------------------------------------------------------------------
1.9 cvs 505: IsWithParameters
506: returns TRUE if url has a concatenated query string.
1.4 cvs 507: ----------------------------------------------------------------------*/
1.3 cvs 508: #ifdef __STDC__
1.34 cvs 509: boolean IsWithParameters (const char *url)
1.3 cvs 510: #else /* __STDC__ */
1.9 cvs 511: boolean IsWithParameters (url)
1.34 cvs 512: const char *url;
1.3 cvs 513: #endif /* __STDC__ */
514: {
1.5 cvs 515: int i;
1.3 cvs 516:
1.9 cvs 517: if ((!url) || (url[0] == EOS))
1.5 cvs 518: return FALSE;
1.3 cvs 519:
1.9 cvs 520: i = strlen (url) - 1;
521: while (i > 0 && url[i--] != '?')
1.5 cvs 522: if (i < 0)
523: return FALSE;
1.3 cvs 524:
1.5 cvs 525: /* There is a parameter */
526: return TRUE;
1.3 cvs 527: }
528:
1.4 cvs 529: /*----------------------------------------------------------------------
1.9 cvs 530: IsW3Path
531: returns TRUE if path is in fact a URL.
1.4 cvs 532: ----------------------------------------------------------------------*/
1.3 cvs 533: #ifdef __STDC__
1.34 cvs 534: boolean IsW3Path (const char *path)
1.3 cvs 535: #else /* __STDC__ */
536: boolean IsW3Path (path)
1.34 cvs 537: const char *path;
1.3 cvs 538: #endif /* __STDC__ */
539: {
1.5 cvs 540: if ((strncmp (path, "http:", 5)) && (strncmp (path, "ftp:", 4)) &&
541: (strncmp (path, "telnet:", 7)) && (strncmp (path, "wais:", 5)) &&
542: (strncmp (path, "news:", 5)) && (strncmp (path, "gopher:", 7)) &&
543: (strncmp (path, "mailto:", 7)) && (strncmp (path, "archie:", 7)))
544: return FALSE;
545: return TRUE;
1.3 cvs 546: }
547:
1.4 cvs 548: /*----------------------------------------------------------------------
1.9 cvs 549: IsValidProtocol
550: returns true if the url protocol is supported by Amaya.
1.4 cvs 551: ----------------------------------------------------------------------*/
1.3 cvs 552: #ifdef __STDC__
1.34 cvs 553: boolean IsValidProtocol (const char *url)
1.3 cvs 554: #else /* __STDC__ */
1.9 cvs 555: boolean IsValidProtocol (url)
1.34 cvs 556: const char *url;
1.3 cvs 557: #endif /* __STDC__ */
558: {
1.58 cvs 559: if (!strncmp (url, "http:", 5)
560: || !strncmp (url, "internal:", 9))
1.22 cvs 561: /* experimental */
1.58 cvs 562: /*** || !strncmp (url, "ftp:", 4) ***/
1.24 cvs 563: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 564: return (TRUE);
1.5 cvs 565: else
1.8 cvs 566: return (FALSE);
1.3 cvs 567: }
568:
1.31 cvs 569:
570: /*----------------------------------------------------------------------
571: GetBaseURL
572: normalizes orgName according to a base associated with doc, and
573: following the standard URL format rules.
574: The function returns the base used to solve relative URL and SRC:
575: - the base of the document,
576: - or the document path (without document name).
577: ----------------------------------------------------------------------*/
578: #ifdef __STDC__
579: char *GetBaseURL (Document doc)
580: #else /* __STDC__ */
581: char *GetBaseURL (doc)
582: Document doc;
583: #endif /* __STDC__ */
584: {
585: Element el;
586: ElementType elType;
587: AttributeType attrType;
588: Attribute attr;
589: char *ptr, *basename;
590: int length;
591:
1.57 cvs 592: /* @@@ irene */
593: if (!DocumentURLs[doc])
594: return NULL;
1.31 cvs 595: basename = TtaGetMemory (MAX_LENGTH);
1.39 cvs 596: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
597: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 598: length = MAX_LENGTH -1;
599: /* get the root element */
600: el = TtaGetMainRoot (doc);
601: /* search the BASE element */
602: elType.ElSSchema = TtaGetDocumentSSchema (doc);
603: elType.ElTypeNum = HTML_EL_BASE;
604: el = TtaSearchTypedElement (elType, SearchInTree, el);
605: if (el)
606: {
607: /* The document has a BASE element -> Get the HREF attribute */
608: attrType.AttrSSchema = elType.ElSSchema;
609: attrType.AttrTypeNum = HTML_ATTR_HREF_;
610: attr = TtaGetAttribute (el, attrType);
611: if (attr)
612: {
613: /* Use the base path of the document */
614: TtaGiveTextAttributeValue (attr, basename, &length);
615: /* base and orgName have to be separated by a DIR_SEP */
616: length--;
1.43 cvs 617: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 618: /* verify if the base has the form "protocol://server:port" */
619: {
1.33 cvs 620: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
621: AMAYA_PARSE_HOST |
622: AMAYA_PARSE_PUNCTUATION);
1.31 cvs 623: if (ptr && !strcmp (ptr, basename))
624: {
1.43 cvs 625: /* it has this form, we complete it by adding a URL_STR */
626: if (strchr (basename, DIR_SEP))
627: strcat (basename, DIR_STR);
628: else
629: strcat (basename, URL_STR);
1.31 cvs 630: length++;
631: }
632: if (ptr)
633: TtaFreeMemory (ptr);
634: }
635: }
1.33 cvs 636: }
637:
1.31 cvs 638: /* Remove anything after the last DIR_SEP char. If no such char is found,
639: * then search for the first ":" char, hoping that what's before that is a
640: * protocol. If found, end the string there. If neither char is found,
641: * then discard the whole base element.
642: */
643: length = strlen (basename) - 1;
644: /* search for the last DIR_SEP char */
1.43 cvs 645: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 646: length--;
647: if (length >= 0)
648: /* found the last DIR_SEP char, end the string there */
649: basename[length + 1] = EOS;
650: else
651: /* search for the first PATH_STR char */
652: {
1.32 cvs 653: for (length = 0; basename[length] != ':' &&
1.31 cvs 654: basename[length] != EOS; length ++);
1.32 cvs 655: if (basename[length] == ':')
1.31 cvs 656: /* found, so end the string there */
657: basename[length + 1] = EOS;
658: else
659: /* not found, discard the base */
660: basename[0] = EOS;
661: }
662: return (basename);
663: }
664:
665:
1.4 cvs 666: /*----------------------------------------------------------------------
1.40 cvs 667: GetLocalPath
668: Allocate and return the local document path associated to the url
669: ----------------------------------------------------------------------*/
670: #ifdef __STDC__
671: char *GetLocalPath (Document doc, char *url)
672: #else /* __STDC__ */
673: char *GetLocalPath (doc, url)
674: Document doc;
675: char *url;
676: #endif /* __STDC__ */
677: {
1.46 cvs 678: char *ptr, *n;
1.40 cvs 679: char *documentname;
1.41 cvs 680: char url_sep;
1.40 cvs 681: int len;
682: boolean noFile;
683:
684: if (url != NULL)
685: {
686: /* check whether the file name exists */
687: len = strlen (url) - 1;
1.41 cvs 688: if (IsW3Path)
689: url_sep = '/';
690: else
691: url_sep = DIR_SEP;
692: noFile = (url[len] == url_sep);
1.40 cvs 693: if (noFile)
694: url[len] = EOS;
1.47 cvs 695: ptr = TtaGetMemory (MAX_LENGTH);
696: documentname = TtaGetMemory (MAX_LENGTH);
1.40 cvs 697: TtaExtractName (url, ptr, documentname);
698: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
699: if (!TtaCheckDirectory (ptr))
700: /* directory did not exist */
1.59 cvs 701: # ifdef _WINDOWS
702: _mkdir (ptr);
703: # else /* !_WINDOWS */
1.40 cvs 704: mkdir (ptr, S_IRWXU);
1.59 cvs 705: # endif /* !_WINDOWS */
1.47 cvs 706:
707: /* don't include the query string within document name */
708: n = strrchr(documentname, '?');
709: if (n != NULL)
710: *n = EOS;
1.46 cvs 711: /* don't include ':' within document name */
712: n = strchr (documentname, ':');
713: if (n != NULL)
714: *n = EOS;
1.40 cvs 715: strcat (ptr, documentname);
716: TtaFreeMemory (documentname);
717: /* restore the url */
718: if (noFile)
1.41 cvs 719: url[len] = url_sep;
1.40 cvs 720: return (ptr);
721: }
722: else
723: return (NULL);
724: }
725:
726:
727: /*----------------------------------------------------------------------
1.9 cvs 728: NormalizeURL
729: normalizes orgName according to a base associated with doc, and
730: following the standard URL format rules.
1.53 cvs 731: if doc is 0 and otherPath not NULL, normalizes orgName according to this
732: other path.
1.9 cvs 733: The function returns the new complete and normalized URL
1.12 cvs 734: or file name path (newName) and the name of the document (docName).
1.9 cvs 735: N.B. If the function can't find out what's the docName, it assigns
736: the name "noname.html".
1.4 cvs 737: ----------------------------------------------------------------------*/
1.3 cvs 738: #ifdef __STDC__
1.53 cvs 739: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName, char *otherPath)
1.3 cvs 740: #else /* __STDC__ */
1.53 cvs 741: void NormalizeURL (orgName, doc, newName, docName, otherPath)
1.3 cvs 742: char *orgName;
743: Document doc;
744: char *newName;
745: char *docName;
1.53 cvs 746: char *otherPath;
1.3 cvs 747: #endif /* __STDC__ */
748: {
1.31 cvs 749: char *basename;
1.18 cvs 750: char tempOrgName[MAX_LENGTH];
1.5 cvs 751: char *ptr;
1.49 cvs 752: char used_sep;
1.5 cvs 753: int length;
1.49 cvs 754: boolean check;
1.5 cvs 755:
1.44 cvs 756: # ifdef _WINDOWS
757: int ndx;
758: # endif /* _WINDOWS */
759:
1.5 cvs 760: if (!newName || !docName)
761: return;
1.18 cvs 762:
1.32 cvs 763: if (doc != 0)
1.53 cvs 764: basename = GetBaseURL (doc);
765: else if (otherPath != NULL)
766: basename = TtaStrdup (otherPath);
1.32 cvs 767: else
1.53 cvs 768: basename = NULL;
1.32 cvs 769:
1.18 cvs 770: /*
1.31 cvs 771: * Clean orgName
772: * Make sure we have a complete orgName, without any leading or trailing
773: * white spaces, or trailinbg new lines
774: */
1.5 cvs 775: ptr = orgName;
1.18 cvs 776: /* skip leading white space and new line characters */
1.19 cvs 777: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS);
1.39 cvs 778: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
779: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 780: /*
1.31 cvs 781: * Make orgName a complete URL
782: * If the URL does not include a protocol, then try to calculate
783: * one using the doc's base element (if it exists),
784: */
1.53 cvs 785: if (tempOrgName[0] == EOS)
786: {
787: newName[0] = EOS;
788: TtaFreeMemory (basename);
789: return;
790: }
1.49 cvs 791:
792: /* clean trailing white space */
793: length = strlen (tempOrgName) - 1;
1.53 cvs 794: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
795: {
796: tempOrgName[length] = EOS;
797: length--;
798: }
1.50 cvs 799:
1.55 cvs 800: /* remove extra dot (which dot???) */
801: /* ugly, but faster than a strcmp */
802: if (tempOrgName[length] == '.'
803: && (length == 0 || tempOrgName[length-1] != '.'))
804: tempOrgName[length] = EOS;
1.50 cvs 805:
1.53 cvs 806: if (IsW3Path (tempOrgName))
807: {
808: /* the name is complete, go to the Sixth Step */
809: strcpy (newName, tempOrgName);
810: SimplifyUrl (&newName);
811: /* verify if the URL has the form "protocol://server:port" */
812: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
813: if (ptr && !strcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
1.50 cvs 814: strcat (newName, URL_STR);
1.49 cvs 815:
1.53 cvs 816: if (ptr)
1.50 cvs 817: TtaFreeMemory (ptr);
1.53 cvs 818: }
819: else if ( basename == NULL)
820: /* the name is complete, go to the Sixth Step */
1.18 cvs 821: strcpy (newName, tempOrgName);
1.53 cvs 822: else
823: {
1.31 cvs 824: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 825: # ifdef _WINDOWS
1.53 cvs 826: if (!IsW3Path (basename))
827: {
828: length = strlen (tempOrgName);
829: for (ndx = 0; ndx < length; ndx++)
830: if (tempOrgName [ndx] == '/')
831: tempOrgName [ndx] = '\\';
832: }
1.44 cvs 833: # endif /* _WINDOWS */
1.25 cvs 834: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 835: if (ptr)
836: {
837: SimplifyUrl (&ptr);
838: strcpy (newName, ptr);
839: TtaFreeMemory (ptr);
840: }
841: else
842: newName[0] = EOS;
843: }
1.36 cvs 844:
845: TtaFreeMemory (basename);
1.18 cvs 846: /*
1.31 cvs 847: * Prepare the docname that will refer to this ressource in the
848: * .amaya directory. If the new URL finishes on DIR_SEP, then use
849: * noname.html as a default ressource name
1.18 cvs 850: */
1.53 cvs 851: if (newName[0] != EOS)
852: {
853: length = strlen (newName) - 1;
854: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
855: {
856: used_sep = newName[length];
857: check = TRUE;
858: while (check)
859: {
1.50 cvs 860: length--;
861: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 862: length--;
863: if (!strncmp (&newName[length+1], "..", 2))
864: {
865: newName[length+1] = EOS;
866: /* remove also previous directory */
867: length--;
868: while (length >= 0 && newName[length] != used_sep)
869: length--;
870: if (strncmp (&newName[length+1], "//", 2))
871: /* don't remove server name */
1.50 cvs 872: newName[length+1] = EOS;
1.53 cvs 873: }
874: else if (!strncmp (&newName[length+1], ".", 1))
875: newName[length+1] = EOS;
1.50 cvs 876: else
1.53 cvs 877: check = FALSE;
878: }
879: strcpy (docName, "noname.html");
880: /* docname was not comprised inside the URL, so let's */
881: /* assign the default ressource name */
882: strcpy (docName, "noname.html");
883: }
884: else
885: { /* docname is comprised inside the URL */
1.50 cvs 886: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP)
1.53 cvs 887: length--;
888: if (length < 0)
1.50 cvs 889: strcpy (docName, newName);
1.53 cvs 890: else
891: strcpy (docName, &newName[length+1]);
892: }
893: }
894: else
895: docName[0] = EOS;
1.18 cvs 896: }
1.3 cvs 897:
1.4 cvs 898: /*----------------------------------------------------------------------
1.9 cvs 899: IsSameHost
1.4 cvs 900: ----------------------------------------------------------------------*/
1.3 cvs 901: #ifdef __STDC__
1.34 cvs 902: boolean IsSameHost (const char *url1, const char *url2)
1.3 cvs 903: #else /* __STDC__ */
904: boolean IsSameHost (url1, url2)
1.34 cvs 905: const char *url1;
906: const char *url2;
1.3 cvs 907: #endif /* __STDC__ */
908: {
1.34 cvs 909: char *basename_ptr1, *basename_ptr2;
910: boolean result;
1.3 cvs 911:
1.25 cvs 912: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
913: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 914:
1.5 cvs 915: if (strcmp (basename_ptr1, basename_ptr2))
1.8 cvs 916: result = FALSE;
1.5 cvs 917: else
1.8 cvs 918: result = TRUE;
1.3 cvs 919:
1.25 cvs 920: TtaFreeMemory (basename_ptr1);
921: TtaFreeMemory (basename_ptr2);
1.5 cvs 922: return (result);
1.3 cvs 923: }
924:
925:
1.4 cvs 926: /*----------------------------------------------------------------------
1.22 cvs 927: HasKnownFileSuffix
928: returns TRUE if path points to a file ending with a suffix.
929: ----------------------------------------------------------------------*/
930: #ifdef __STDC__
1.34 cvs 931: boolean HasKnownFileSuffix (const char *path)
1.22 cvs 932: #else /* __STDC__ */
933: boolean HasKnownFileSuffix (path)
1.34 cvs 934: const char *path;
1.22 cvs 935: #endif /* __STDC__ */
936: {
1.29 cvs 937: char *root;
938: char temppath[MAX_LENGTH];
939: char suffix[MAX_LENGTH];
1.22 cvs 940:
1.24 cvs 941: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 942: return (FALSE);
943:
1.29 cvs 944: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 945:
946: if (root)
947: {
948: strcpy (temppath, root);
1.25 cvs 949: TtaFreeMemory (root);
1.22 cvs 950: /* Get the suffix */
951: ExtractSuffix (temppath, suffix);
952:
953: if( suffix[0] == EOS)
954: /* no suffix */
955: return (FALSE);
956:
957: /* Normalize the suffix */
958: ConvertToLowerCase (suffix);
959:
1.23 cvs 960: if (!strcmp (suffix, "gz"))
1.22 cvs 961: /* skip the compressed suffix */
962: {
963: ExtractSuffix (temppath, suffix);
964: if(suffix[0] == EOS)
965: /* no suffix */
966: return (FALSE);
967: /* Normalize the suffix */
968: ConvertToLowerCase (suffix);
969: }
970:
1.63 cvs 971: if (strcmp (suffix, "gif") &&
972: strcmp (suffix, "xbm") &&
973: strcmp (suffix, "xpm") &&
974: strcmp (suffix, "jpg") &&
975: strcmp (suffix, "pdf") &&
976: strcmp (suffix, "png") &&
977: strcmp (suffix, "tgz") &&
978: strcmp (suffix, "xpg") &&
979: strcmp (suffix, "xpd") &&
980: strcmp (suffix, "ps") &&
981: strcmp (suffix, "au") &&
982: strcmp (suffix, "html") &&
983: strcmp (suffix, "htm") &&
984: strcmp (suffix, "shtml") &&
985: strcmp (suffix, "xht") &&
986: strcmp (suffix, "xhtm") &&
987: strcmp (suffix, "xhtml") &&
988: strcmp (suffix, "txt") &&
989: strcmp (suffix, "css") &&
990: strcmp (suffix, "eps"))
1.22 cvs 991: return (FALSE);
992: else
993: return (TRUE);
994: }
995: else
996: return (FALSE);
997: }
998:
999:
1000: /*----------------------------------------------------------------------
1.24 cvs 1001: ChopURL
1002: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1003: If inputURL is bigger than that size, outputURL receives
1004: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1005: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1006: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1007: copied into outputURL.
1008: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1009: chars.
1010: ----------------------------------------------------------------------*/
1011: #ifdef __STDC__
1.34 cvs 1012: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1013: #else
1014: void ChopURL (outputURL, inputURL)
1.34 cvs 1015: char *outputURL;
1016: const char *inputURL;
1.24 cvs 1017: #endif
1.22 cvs 1018:
1.24 cvs 1019: {
1020: int len;
1.9 cvs 1021:
1.24 cvs 1022: len = strlen (inputURL);
1023: if (len <= MAX_PRINT_URL_LENGTH)
1.29 cvs 1024: strcpy (outputURL, inputURL);
1.24 cvs 1025: else
1026: /* make a truncated urlName on the status window */
1027: {
1028: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1029: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1030: strcat (outputURL, "...");
1031: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1032: }
1.25 cvs 1033: }
1034:
1035:
1036: /*----------------------------------------------------------------------
1037: scan
1.47 cvs 1038: Scan a filename for its constituents
1.25 cvs 1039: -----------------------------------
1040:
1041: On entry,
1042: name points to a document name which may be incomplete.
1043: On exit,
1044: absolute or relative may be nonzero (but not both).
1045: host, fragment and access may be nonzero if they were specified.
1046: Any which are nonzero point to zero terminated strings.
1047: ----------------------------------------------------------------------*/
1048: #ifdef __STDC__
1049: static void scan (char * name, HTURI * parts)
1050: #else /* __STDC__ */
1051: static void scan (name, parts)
1052: char *name;
1053: HTURI *parts;
1054:
1055: #endif /* __STDC__ */
1056: {
1.43 cvs 1057: char *p;
1058: char *after_access = name;
1.32 cvs 1059:
1.43 cvs 1060: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1061: /* Look for fragment identifier */
1.52 cvs 1062: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1063: {
1064: *p++ = '\0';
1065: parts->fragment = p;
1.25 cvs 1066: }
1067:
1.28 cvs 1068: for (p=name; *p; p++)
1069: {
1.43 cvs 1070: if (*p == URL_SEP || *p == DIR_SEP || *p=='#' || *p=='?')
1.28 cvs 1071: break;
1.32 cvs 1072: if (*p==':')
1.28 cvs 1073: {
1074: *p = 0;
1075: parts->access = after_access; /* Scheme has been specified */
1076:
1077: /* The combination of gcc, the "-O" flag and the HP platform is
1078: unhealthy. The following three lines is a quick & dirty fix, but is
1079: not recommended. Rather, turn off "-O". */
1080:
1081: /* after_access = p;*/
1082: /* while (*after_access == 0)*/
1083: /* after_access++;*/
1084: after_access = p+1;
1.43 cvs 1085: if (!strcasecmp("URL", parts->access))
1.28 cvs 1086: /* Ignore IETF's URL: pre-prefix */
1087: parts->access = NULL;
1088: else
1.25 cvs 1089: break;
1090: }
1091: }
1092:
1093: p = after_access;
1.43 cvs 1094: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1095: {
1.43 cvs 1096: if (p[1] == URL_SEP)
1.28 cvs 1097: {
1.25 cvs 1098: parts->host = p+2; /* host has been specified */
1.28 cvs 1099: *p = 0; /* Terminate access */
1100: /* look for end of host name if any */
1.43 cvs 1101: p = strchr (parts->host, URL_SEP);
1.28 cvs 1102: if (p)
1103: {
1.43 cvs 1104: *p = EOS; /* Terminate host */
1.25 cvs 1105: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1106: }
1107: }
1108: else
1109: /* Root found but no host */
1110: parts->absolute = p+1;
1111: }
1112: else
1113: {
1.25 cvs 1114: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1115: }
1.25 cvs 1116: }
1117:
1118:
1119: /*----------------------------------------------------------------------
1.28 cvs 1120: AmayaParseUrl: parse a Name relative to another name
1121:
1122: This returns those parts of a name which are given (and requested)
1123: substituting bits from the related name where necessary.
1.25 cvs 1124:
1.28 cvs 1125: On entry,
1.25 cvs 1126: aName A filename given
1127: relatedName A name relative to which aName is to be parsed. Give
1128: it an empty string if aName is absolute.
1129: wanted A mask for the bits which are wanted.
1130:
1.28 cvs 1131: On exit,
1.25 cvs 1132: returns A pointer to a malloc'd string which MUST BE FREED
1133: ----------------------------------------------------------------------*/
1134: #ifdef __STDC__
1.34 cvs 1135: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1.25 cvs 1136: #else /* __STDC__ */
1.28 cvs 1137: char *AmayaParseUrl (aName, relatedName, wanted)
1.34 cvs 1138: const char *aName;
1.28 cvs 1139: char *relatedName;
1140: int wanted;
1.25 cvs 1141:
1142: #endif /* __STDC__ */
1143: {
1.29 cvs 1144: char *return_value;
1145: char result[MAX_LENGTH];
1146: char name[MAX_LENGTH];
1147: char rel[MAX_LENGTH];
1148: char *p, *access;
1149: HTURI given, related;
1150: int len;
1.33 cvs 1151: char used_sep;
1152: char *used_str;
1.32 cvs 1153:
1.43 cvs 1154: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1155: {
1.42 cvs 1156: used_str = DIR_STR;
1157: used_sep = DIR_SEP;
1.33 cvs 1158: }
1.32 cvs 1159: else
1.33 cvs 1160: {
1.42 cvs 1161: used_str = URL_STR;
1162: used_sep = URL_SEP;
1.33 cvs 1163: }
1.32 cvs 1164:
1.29 cvs 1165: /* Make working copies of input strings to cut up: */
1166: return_value = NULL;
1167: result[0] = 0; /* Clear string */
1168: strcpy (name, aName);
1169: if (relatedName != NULL)
1170: strcpy (rel, relatedName);
1171: else
1172: relatedName[0] = EOS;
1173:
1174: scan (name, &given);
1175: scan (rel, &related);
1176: access = given.access ? given.access : related.access;
1177: if (wanted & AMAYA_PARSE_ACCESS)
1178: if (access)
1179: {
1180: strcat (result, access);
1181: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.32 cvs 1182: strcat (result, ":");
1.29 cvs 1183: }
1184:
1185: if (given.access && related.access)
1186: /* If different, inherit nothing. */
1187: if (strcmp (given.access, related.access) != 0)
1188: {
1189: related.host = 0;
1190: related.absolute = 0;
1191: related.relative = 0;
1192: related.fragment = 0;
1193: }
1194:
1195: if (wanted & AMAYA_PARSE_HOST)
1196: if(given.host || related.host)
1197: {
1198: if(wanted & AMAYA_PARSE_PUNCTUATION)
1199: strcat (result, "//");
1200: strcat (result, given.host ? given.host : related.host);
1201: }
1202:
1203: if (given.host && related.host)
1204: /* If different hosts, inherit no path. */
1205: if (strcmp(given.host, related.host) != 0)
1206: {
1207: related.absolute = 0;
1208: related.relative = 0;
1209: related.fragment = 0;
1210: }
1211:
1212: if (wanted & AMAYA_PARSE_PATH)
1213: {
1214: if (given.absolute)
1215: {
1216: /* All is given */
1217: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.33 cvs 1218: strcat (result, used_str);
1.29 cvs 1219: strcat (result, given.absolute);
1.25 cvs 1220: }
1.29 cvs 1221: else if (related.absolute)
1222: {
1223: /* Adopt path not name */
1.33 cvs 1224: strcat (result, used_str);
1.29 cvs 1225: strcat (result, related.absolute);
1226: if (given.relative)
1227: {
1228: /* Search part? */
1229: p = strchr (result, '?');
1230: if (!p)
1231: p=result+strlen(result)-1;
1.33 cvs 1232: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1233: /* Remove filename */
1234: p[1]=0;
1235: /* Add given one */
1236: strcat (result, given.relative);
1.25 cvs 1237: }
1238: }
1.29 cvs 1239: else if (given.relative)
1240: /* what we've got */
1241: strcat (result, given.relative);
1242: else if (related.relative)
1243: strcat (result, related.relative);
1244: else
1245: /* No inheritance */
1.33 cvs 1246: strcat (result, used_str);
1.25 cvs 1247: }
1.29 cvs 1248:
1249: if (wanted & AMAYA_PARSE_ANCHOR)
1250: if (given.fragment || related.fragment)
1251: {
1252: if (given.absolute && given.fragment)
1253: {
1254: /*Fixes for relURLs...*/
1255: if (wanted & AMAYA_PARSE_PUNCTUATION)
1256: strcat (result, "#");
1257: strcat (result, given.fragment);
1258: }
1259: else if (!(given.absolute) && !(given.fragment))
1260: strcat (result, "");
1261: else
1262: {
1263: if (wanted & AMAYA_PARSE_PUNCTUATION)
1264: strcat (result, "#");
1265: strcat (result, given.fragment ? given.fragment : related.fragment);
1266: }
1267: }
1268: len = strlen (result);
1269: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1270: strcpy (return_value, result);
1271: return (return_value); /* exactly the right length */
1.25 cvs 1272: }
1273:
1274: /*----------------------------------------------------------------------
1275: HTCanon
1276: Canonicalizes the URL in the following manner starting from the host
1277: pointer:
1278:
1279: 1) The host name is converted to lowercase
1280: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1281:
1282: Return: OK The position of the current path part of the URL
1283: which might be the old one or a new one.
1284:
1285: ----------------------------------------------------------------------*/
1286: #ifdef __STDC__
1.28 cvs 1287: static char *HTCanon (char ** filename, char * host)
1.25 cvs 1288: #else /* __STDC__ */
1.28 cvs 1289: static char *HTCanon (filename, host)
1290: char **filename;
1291: char *host;
1.25 cvs 1292: #endif /* __STDC__ */
1293: {
1294: char *newname = NULL;
1295: char *port;
1296: char *strptr;
1297: char *path;
1298: char *access = host-3;
1.33 cvs 1299: char used_sep;
1.32 cvs 1300:
1301:
1.33 cvs 1302: if (*filename && strchr (*filename, URL_SEP))
1303: {
1304: used_sep = URL_SEP;
1305: }
1306: else
1307: {
1308: used_sep = DIR_SEP;
1309: }
1.32 cvs 1310:
1.33 cvs 1311: while (access>*filename && *(access-1)!= used_sep) /* Find access method */
1.25 cvs 1312: access--;
1.33 cvs 1313: if ((path = strchr(host, used_sep)) == NULL) /* Find path */
1.25 cvs 1314: path = host + strlen(host);
1315: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */
1316: host = strptr;
1.32 cvs 1317: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */
1.25 cvs 1318: port = NULL;
1319:
1320: strptr = host; /* Convert to lower-case */
1.33 cvs 1321: while (strptr<path)
1322: {
1.25 cvs 1323: *strptr = tolower(*strptr);
1324: strptr++;
1.33 cvs 1325: }
1.25 cvs 1326:
1327: /* Does the URL contain a full domain name? This also works for a
1328: numerical host name. The domain name is already made lower-case
1329: and without a trailing dot. */
1330: {
1.33 cvs 1331: char *dot = port ? port : path;
1332: if (dot > *filename && *--dot=='.')
1333: {
1334: char *orig=dot, *dest=dot+1;
1335: while((*orig++ = *dest++));
1336: if (port) port--;
1337: path--;
1.25 cvs 1338: }
1339: }
1340: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1341: if (port)
1342: {
1343: if (!*(port+1) || *(port+1)==used_sep)
1344: {
1345: if (!newname)
1346: {
1.25 cvs 1347: char *orig=port, *dest=port+1;
1348: while((*orig++ = *dest++));
1.33 cvs 1349: }
1350: }
1351: else if ((!strncmp(access, "http", 4) &&
1352: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1353: (!strncmp(access, "gopher", 6) &&
1354: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) ||
1355: (!strncmp(access, "ftp", 3) &&
1356: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) {
1357: if (!newname)
1358: {
1359: char *orig=port, *dest=port+3;
1360: while((*orig++ = *dest++));
1361: /* Update path position, Henry Minsky */
1362: path -= 3;
1.25 cvs 1363: }
1.33 cvs 1364: }
1365: else if (newname)
1366: strncat(newname, port, (int) (path-port));
1367: }
1.25 cvs 1368:
1.33 cvs 1369: if (newname)
1370: {
1.25 cvs 1371: char *newpath = newname+strlen(newname);
1372: strcat(newname, path);
1373: path = newpath;
1.28 cvs 1374: /* Free old copy */
1375: TtaFreeMemory(*filename);
1.25 cvs 1376: *filename = newname;
1.33 cvs 1377: }
1.25 cvs 1378: return path;
1379: }
1380:
1381:
1382: /*----------------------------------------------------------------------
1.29 cvs 1383: SimplifyUrl: simplify a URI
1.32 cvs 1384: A URI is allowed to contain the sequence xxx/../ which may be
1385: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1386: Simplification helps us recognize duplicate URIs.
1.25 cvs 1387:
1.28 cvs 1388: Thus, /etc/junk/../fred becomes /etc/fred
1389: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1390:
1.28 cvs 1391: but we should NOT change
1392: http://fred.xxx.edu/../..
1.25 cvs 1393:
1394: or ../../albert.html
1395:
1.28 cvs 1396: In order to avoid empty URLs the following URLs become:
1.25 cvs 1397:
1398: /fred/.. becomes /fred/..
1399: /fred/././.. becomes /fred/..
1400: /fred/.././junk/.././ becomes /fred/..
1401:
1.28 cvs 1402: If more than one set of `://' is found (several proxies in cascade) then
1403: only the part after the last `://' is simplified.
1.25 cvs 1404:
1.28 cvs 1405: Returns: A string which might be the old one or a new one.
1.25 cvs 1406: ----------------------------------------------------------------------*/
1407: #ifdef __STDC__
1.29 cvs 1408: void SimplifyUrl (char ** url)
1.25 cvs 1409: #else /* __STDC__ */
1.29 cvs 1410: void SimplifyUrl (url)
1.28 cvs 1411: char **url;
1.25 cvs 1412: #endif /* __STDC__ */
1413: {
1.28 cvs 1414: char *path, *p;
1415: char *newptr, *access;
1416: char *orig, *dest, *end;
1417:
1.33 cvs 1418: char used_sep;
1.32 cvs 1419:
1420:
1.28 cvs 1421: if (!url || !*url)
1422: return;
1423:
1.33 cvs 1424: if (strchr (*url, URL_SEP))
1425: {
1426: used_sep = URL_SEP;
1427: }
1.32 cvs 1428: else
1.33 cvs 1429: {
1430: used_sep = DIR_SEP;
1431: }
1.32 cvs 1432:
1.28 cvs 1433: /* Find any scheme name */
1434: if ((path = strstr(*url, "://")) != NULL)
1.33 cvs 1435: {
1436: /* Find host name */
1.28 cvs 1437: access = *url;
1438: while (access<path && (*access=tolower(*access)))
1439: access++;
1440: path += 3;
1441: while ((newptr = strstr(path, "://")) != NULL)
1442: /* For proxies */
1443: path = newptr+3;
1444: /* We have a host name */
1445: path = HTCanon(url, path);
1.25 cvs 1446: }
1.28 cvs 1447: else if ((path = strstr(*url, ":/")) != NULL)
1448: path += 2;
1449: else
1450: path = *url;
1.25 cvs 1451:
1.33 cvs 1452: if (*path == used_sep && *(path+1)==used_sep)
1.28 cvs 1453: /* Some URLs start //<foo> */
1454: path += 1;
1455: else if (!strncmp(path, "news:", 5))
1456: {
1457: newptr = strchr(path+5, '@');
1458: if (!newptr)
1459: newptr = path + 5;
1460: while (*newptr)
1461: {
1462: /* Make group or host lower case */
1463: *newptr = tolower (*newptr);
1464: newptr++;
1.25 cvs 1465: }
1.28 cvs 1466: /* Doesn't need to do any more */
1467: return;
1.25 cvs 1468: }
1.28 cvs 1469:
1470: if ((p = path))
1471: {
1472: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1473: (end = strchr (path, '#'))))
1474: end = path + strlen (path);
1475:
1476: /* Parse string second time to simplify */
1477: p = path;
1478: while (p < end)
1479: {
1.33 cvs 1480: if (*p==used_sep)
1.28 cvs 1481: {
1.33 cvs 1482: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1483: {
1484: orig = p + 1;
1.33 cvs 1485: dest = (*(p+2)!=used_sep) ? p+2 : p+3;
1.52 cvs 1486: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1487: end = orig - 1;
1488: }
1.33 cvs 1489: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3)))
1.28 cvs 1490: {
1491: newptr = p;
1.52 cvs 1492: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1493: if (*newptr == used_sep)
1494: orig = newptr + 1;
1.28 cvs 1495: else
1.52 cvs 1496: orig = newptr;
1497:
1498: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1499: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1500: end = orig-1;
1501: /* Start again with prev slash */
1502: p = newptr;
1.28 cvs 1503: }
1.33 cvs 1504: else if (*(p+1) == used_sep)
1.28 cvs 1505: {
1.33 cvs 1506: while (*(p+1) == used_sep)
1.28 cvs 1507: {
1508: orig = p;
1509: dest = p + 1;
1510: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1511: end = orig-1;
1512: }
1513: }
1514: else
1.25 cvs 1515: p++;
1.28 cvs 1516: }
1517: else
1518: p++;
1.25 cvs 1519: }
1520: }
1.51 cvs 1521:
1522: /*
1523: ** Check for host/../.. kind of things
1524: */
1.52 cvs 1525: if (*path==used_sep && *(path+1)=='.' && *(path+2)=='.' && (!*(path+3) || *(path+3)==used_sep))
1.51 cvs 1526: *(path+1) = EOS;
1527:
1.28 cvs 1528: return;
1529: }
1530:
1531:
1532: /*----------------------------------------------------------------------
1533: NormalizeFile normalizes local names.
1534: Return TRUE if target and src differ.
1535: ----------------------------------------------------------------------*/
1536: #ifdef __STDC__
1537: boolean NormalizeFile (char *src, char *target)
1538: #else
1539: boolean NormalizeFile (src, target)
1540: char *src;
1541: char *target;
1542:
1543: #endif
1544: {
1545: char *s;
1546: boolean change;
1547:
1.54 cvs 1548: change = FALSE;
1.53 cvs 1549: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1550: {
1551: /* remove the prefix file: */
1552: if (src[5] == EOS)
1553: strcpy (target, DIR_STR);
1554: else if (src[0] == '~')
1555: {
1556: /* replace ~ */
1557: s = (char *) TtaGetEnvString ("HOME");
1558: strcpy (target, s);
1559: strcat (target, &src[5]);
1560: }
1561: else
1562: strcpy (target, &src[5]);
1.54 cvs 1563: change = TRUE;
1.28 cvs 1564: }
1.53 cvs 1565: # ifndef _WINDOWS
1566: else if (src[0] == '~')
1567: {
1568: /* replace ~ */
1569: s = (char *) TtaGetEnvString ("HOME");
1570: strcpy (target, s);
1571: if (src[1] != DIR_SEP)
1572: strcat (target, DIR_STR);
1573: strcat (target, &src[1]);
1.54 cvs 1574: change = TRUE;
1.53 cvs 1575: }
1576: # endif /* _WINDOWS */
1.28 cvs 1577: else
1578: strcpy (target, src);
1579:
1580: /* remove /../ and /./ */
1.29 cvs 1581: SimplifyUrl (&target);
1.54 cvs 1582: if (!change)
1583: change = strcmp (src, target);
1.28 cvs 1584: return (change);
1.25 cvs 1585: }
1586:
1.28 cvs 1587:
1.25 cvs 1588: /*----------------------------------------------------------------------
1.31 cvs 1589: MakeRelativeURL: make relative name
1.25 cvs 1590:
1.28 cvs 1591: This function creates and returns a string which gives an expression of
1592: one address as related to another. Where there is no relation, an absolute
1593: address is retured.
1.25 cvs 1594:
1.28 cvs 1595: On entry,
1.25 cvs 1596: Both names must be absolute, fully qualified names of nodes
1597: (no fragment bits)
1598:
1.28 cvs 1599: On exit,
1.25 cvs 1600: The return result points to a newly allocated name which, if
1601: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1602: The caller is responsible for freeing the resulting name later.
1603: ----------------------------------------------------------------------*/
1604: #ifdef __STDC__
1.31 cvs 1605: char *MakeRelativeURL (char *aName, char *relatedName)
1.25 cvs 1606: #else /* __STDC__ */
1.31 cvs 1607: char *MakeRelativeURL (aName, relatedName)
1.28 cvs 1608: char *aName;
1609: char *relatedName;
1.25 cvs 1610: #endif /* __STDC__ */
1611: {
1.39 cvs 1612: char *return_value;
1613: char result[MAX_LENGTH];
1.29 cvs 1614: char *p;
1.34 cvs 1615: char *q;
1.29 cvs 1616: char *after_access;
1617: char *last_slash = NULL;
1618: int slashes, levels, len;
1619:
1.44 cvs 1620: # ifdef _WINDOWS
1621: int ndx;
1622: # endif /* _WINDOWS */
1623:
1.29 cvs 1624: if (aName == NULL || relatedName == NULL)
1625: return (NULL);
1626:
1627: slashes = 0;
1628: after_access = NULL;
1629: p = aName;
1630: q = relatedName;
1631: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1632: {
1633: /* Find extent of match */
1.32 cvs 1634: if (*p == ':')
1.29 cvs 1635: after_access = p + 1;
1.28 cvs 1636: if (*p == DIR_SEP)
1.27 cvs 1637: {
1.29 cvs 1638: /* memorize the last slash position and count them */
1.27 cvs 1639: last_slash = p;
1640: slashes++;
1.25 cvs 1641: }
1642: }
1643:
1.31 cvs 1644: /* q, p point to the first non-matching character or zero */
1645: if (*q == EOS)
1646: {
1647: /* New name is a subset of the related name */
1648: /* exactly the right length */
1649: len = strlen (p);
1650: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1651: strcpy (return_value, p);
1652: }
1653: else if ((slashes < 2 && after_access == NULL)
1654: || (slashes < 3 && after_access != NULL))
1655: {
1656: /* Two names whitout common path */
1657: /* exactly the right length */
1658: len = strlen (aName);
1659: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1660: strcpy (return_value, aName);
1661: }
1662: else
1663: {
1664: /* Some path in common */
1665: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1666: /* just the same server */
1667: strcpy (result, last_slash);
1668: else
1669: {
1670: levels= 0;
1.52 cvs 1671: for (; *q && *q != '#' && *q!=';' && *q!='?'; q++)
1.31 cvs 1672: if (*q == DIR_SEP)
1673: levels++;
1674:
1.52 cvs 1675: result[0] = EOS;
1.31 cvs 1676: for (;levels; levels--)
1677: strcat (result, "../");
1678: strcat (result, last_slash+1);
1679: }
1.52 cvs 1680:
1681: if (!*result)
1682: strcat (result, "./");
1683:
1.31 cvs 1684: /* exactly the right length */
1685: len = strlen (result);
1686: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL)
1687: strcpy (return_value, result);
1.52 cvs 1688:
1.25 cvs 1689: }
1.44 cvs 1690: # ifdef _WINDOWS
1691: len = strlen (return_value);
1692: for (ndx = 0; ndx < len; ndx ++)
1693: if (return_value[ndx] == '\\')
1694: return_value[ndx] = '/' ;
1695: # endif /* _WINDOWS */
1.29 cvs 1696: return (return_value);
1.24 cvs 1697: }
1.35 cvs 1698:
1699:
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