Annotation of Amaya/amaya/AHTURLTools.c, revision 1.106
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
1.106 ! cvs 14: * R. Guetari: 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"
1.100 kahan 23: #include "query_f.h"
1.8 cvs 24:
1.24 cvs 25: #define MAX_PRINT_URL_LENGTH 50
1.106 ! cvs 26: typedef struct _HTURI
! 27: {
! 28: char *access; /* Now known as "scheme" */
! 29: char *host;
! 30: char *absolute;
! 31: char *relative;
! 32: char *fragment;
1.29 cvs 33: } HTURI;
1.24 cvs 34:
1.28 cvs 35:
36: /*----------------------------------------------------------------------
37: ConvertToLowerCase
38: Converts a string to lowercase.
39: ----------------------------------------------------------------------*/
1.106 ! cvs 40: void ConvertToLowerCase (char *string)
1.28 cvs 41: {
42: int i;
1.93 cvs 43:
1.28 cvs 44: if (!string)
45: return;
46:
1.106 ! cvs 47: for (i = 0; string[i] != EOS; i++)
1.67 cvs 48: string[i] = utolower (string[i]);
1.28 cvs 49: }
1.22 cvs 50:
1.8 cvs 51: /*----------------------------------------------------------------------
1.75 cvs 52: EscapeChar
53: writes the equivalent escape code of a char in a string
54: ----------------------------------------------------------------------*/
1.106 ! cvs 55: void EscapeChar (char *string, char c)
1.75 cvs 56: {
57: c &= 0xFF; /* strange behavior under solaris? */
1.106 ! cvs 58: sprintf (string, "%02x", (unsigned int) c);
1.75 cvs 59: }
60:
61: /*----------------------------------------------------------------------
1.96 cvs 62: UnEscapeChar
63: writes the equivalent hex code to a %xx coded char
64: ----------------------------------------------------------------------*/
1.106 ! cvs 65: static char UnEscapeChar (char c)
1.96 cvs 66: {
1.106 ! cvs 67: return c >= '0' && c <= '9' ? c - '0'
! 68: : c >= 'A' && c <= 'F' ? c - 'A' + 10
! 69: : c - 'a' + 10; /* accept small letters just in case */
1.96 cvs 70: }
71:
72: /*----------------------------------------------------------------------
1.75 cvs 73: EscapeURL
74: Takes a URL and escapes all protected chars into
75: %xx sequences. Also, removes any leading white spaces
76: Returns either NULL or a new buffer, which must be freed by the caller
77: ----------------------------------------------------------------------*/
1.106 ! cvs 78: char *EscapeURL (const char *url)
! 79: {
! 80: char *buffer;
! 81: int buffer_len;
! 82: int buffer_free_mem;
! 83: char *ptr;
! 84: int new_chars;
1.75 cvs 85: void *status;
86:
87: if (url && *url)
88: {
1.106 ! cvs 89: buffer_free_mem = strlen (url) + 20;
! 90: buffer = TtaGetMemory (buffer_free_mem + 1);
1.75 cvs 91: ptr = url;
92: buffer_len = 0;
93:
94: while (*ptr)
95: {
96: switch (*ptr)
97: {
98: /* put here below all the chars that need to
99: be escaped into %xx */
1.81 cvs 100: case 0x27: /* & */
101: case 0x20: /* space */
1.75 cvs 102: new_chars = 3;
103: break;
104:
105: default:
106: new_chars = 1;
107: break;
108: }
109:
110: /* see if we need extra room in the buffer */
111: if (new_chars > buffer_free_mem)
112: {
1.76 cvs 113: buffer_free_mem = 20;
1.106 ! cvs 114: status = TtaRealloc (buffer, sizeof (char)
1.75 cvs 115: * (buffer_len + buffer_free_mem + 1));
116: if (status)
117: buffer = (STRING) status;
1.106 ! cvs 118: else
! 119: {
! 120: /* @@ maybe we should do some other behavior here, like
! 121: freeing the buffer and return a void thing */
! 122: buffer[buffer_len] = EOS;
! 123: break;
! 124: }
1.75 cvs 125: }
126: /* escape the char */
127: if (new_chars == 3)
128: {
1.106 ! cvs 129: buffer[buffer_len] = '%';
1.75 cvs 130: EscapeChar (&buffer[buffer_len+1], *ptr);
131: }
132: else
133: buffer[buffer_len] = *ptr;
134:
135: /* update the status */
136: buffer_len += new_chars;
137: buffer_free_mem -= new_chars;
138: /* examine the next char */
139: ptr++;
140: }
1.106 ! cvs 141: buffer[buffer_len] = EOS;
1.75 cvs 142: }
1.76 cvs 143: else
144: buffer = NULL;
145:
1.75 cvs 146: return (buffer);
147: }
148:
149:
150: /*----------------------------------------------------------------------
1.11 cvs 151: ExplodeURL
1.8 cvs 152: ----------------------------------------------------------------------*/
1.106 ! cvs 153: void ExplodeURL (char *url, char **proto, char **host, char **dir,
! 154: char **file)
1.8 cvs 155: {
1.33 cvs 156: char *curr, *temp;
157: char used_sep;
1.32 cvs 158:
1.33 cvs 159: if (url && strchr (url, URL_SEP))
160: used_sep = URL_SEP;
161: else
162: used_sep = DIR_SEP;
1.8 cvs 163:
164: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
165: (dir == NULL) || (file == NULL))
166: return;
167:
168: /* initialize every pointer */
169: *proto = *host = *dir = *file = NULL;
170:
171: /* skip any leading space */
172: while ((*url == SPACE) || (*url == TAB))
173: url++;
1.9 cvs 174: curr = url;
175: if (*curr == 0)
1.8 cvs 176: goto finished;
177:
178: /* go to the end of the URL */
1.68 cvs 179: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
180: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 181: curr++;
1.8 cvs 182:
183: /* mark the end of the chain */
1.9 cvs 184: *curr = EOS;
185: curr--;
186: if (curr <= url)
1.8 cvs 187: goto finished;
188:
189: /* search the next DIR_SEP indicating the beginning of the file name */
190: do
1.11 cvs 191: curr--;
1.33 cvs 192: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 193:
1.9 cvs 194: if (curr < url)
1.8 cvs 195: goto finished;
1.9 cvs 196: *file = curr + 1;
1.8 cvs 197:
198: /* mark the end of the dir */
1.9 cvs 199: *curr = EOS;
200: curr--;
201: if (curr < url)
1.8 cvs 202: goto finished;
203:
1.29 cvs 204: /* search for the DIR_STR indicating the host name start */
1.33 cvs 205: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 206: curr--;
1.8 cvs 207:
208: /* if we found it, separate the host name from the directory */
1.102 kahan 209: if ((*curr == used_sep) && (*(curr + 1) == used_sep))
1.8 cvs 210: {
1.9 cvs 211: *host = temp = curr + 2;
1.33 cvs 212: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 213: temp++;
1.33 cvs 214: if (*temp == used_sep)
1.8 cvs 215: {
216: *temp = EOS;
217: *dir = temp + 1;
218: }
219: }
220: else
1.11 cvs 221: *dir = curr;
222:
1.9 cvs 223: if (curr <= url)
1.8 cvs 224: goto finished;
225:
226: /* mark the end of the proto */
1.9 cvs 227: *curr = EOS;
228: curr--;
229: if (curr < url)
1.8 cvs 230: goto finished;
231:
1.106 ! cvs 232: if (*curr == ':')
1.8 cvs 233: {
1.9 cvs 234: *curr = EOS;
235: curr--;
1.8 cvs 236: }
237: else
238: goto finished;
1.11 cvs 239:
1.9 cvs 240: if (curr < url)
1.8 cvs 241: goto finished;
1.9 cvs 242: while ((curr > url) && (isalpha (*curr)))
243: curr--;
244: *proto = curr;
1.8 cvs 245:
246: finished:;
247:
248: #ifdef AMAYA_DEBUG
249: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
250: if (*proto)
251: fprintf (stderr, "proto : %s, ", *proto);
252: if (*host)
253: fprintf (stderr, "host : %s, ", *host);
254: if (*dir)
255: fprintf (stderr, "dir : %s, ", *dir);
256: if (*file)
257: fprintf (stderr, "file : %s ", *file);
258: fprintf (stderr, "\n");
259: #endif
260:
261: }
1.3 cvs 262:
1.61 cvs 263:
264: /*----------------------------------------------------------------------
265: ExtractSuffix extract suffix from document nane.
266: ----------------------------------------------------------------------*/
1.106 ! cvs 267: void ExtractSuffix (char *aName, char *aSuffix)
1.61 cvs 268: {
1.106 ! cvs 269: int lg, i;
! 270: char *ptr, *oldptr;
1.61 cvs 271:
272: if (!aSuffix || !aName)
273: /* bad suffix */
274: return;
275:
1.106 ! cvs 276: aSuffix[0] = EOS;
! 277: lg = strlen (aName);
1.61 cvs 278: if (lg)
279: {
280: /* the name is not empty */
281: oldptr = ptr = &aName[0];
282: do
283: {
1.106 ! cvs 284: ptr = strrchr (oldptr, '.');
1.61 cvs 285: if (ptr)
286: oldptr = &ptr[1];
287: }
288: while (ptr);
289:
290: i = (int) (oldptr) - (int) (aName); /* name length */
291: if (i > 1)
292: {
1.106 ! cvs 293: aName[i - 1] = EOS;
1.61 cvs 294: if (i != lg)
1.106 ! cvs 295: strcpy (aSuffix, oldptr);
1.61 cvs 296: }
297: }
298: }
299:
1.4 cvs 300: /*----------------------------------------------------------------------
1.9 cvs 301: IsHTMLName
302: returns TRUE if path points to an HTML resource.
1.4 cvs 303: ----------------------------------------------------------------------*/
1.106 ! cvs 304: ThotBool IsHTMLName (const char *path)
! 305: {
! 306: char temppath[MAX_LENGTH];
! 307: char suffix[MAX_LENGTH];
! 308: char nsuffix[MAX_LENGTH];
1.101 cvs 309: int i;
1.5 cvs 310:
1.101 cvs 311: if (!path)
312: return (FALSE);
1.5 cvs 313:
1.106 ! cvs 314: strcpy (temppath, path);
1.101 cvs 315: ExtractSuffix (temppath, suffix);
316: i = 0;
1.106 ! cvs 317: while (suffix[i] != EOS)
1.101 cvs 318: {
319: /* Normalize the suffix */
320: i = 0;
1.106 ! cvs 321: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 322: {
323: nsuffix[i] = utolower (suffix[i]);
324: i++;
325: }
1.106 ! cvs 326: nsuffix[i] = EOS;
! 327: if (!strcmp (nsuffix, "html") ||
! 328: !strcmp (nsuffix, "htm") ||
! 329: !strcmp (nsuffix, "shtml") ||
! 330: !strcmp (nsuffix, "jsp") ||
! 331: !strcmp (nsuffix, "xht") ||
! 332: !strcmp (nsuffix, "xhtm") ||
! 333: !strcmp (nsuffix, "xhtml"))
1.101 cvs 334: return (TRUE);
1.106 ! cvs 335: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 336: {
337: /* take into account compressed files */
338: ExtractSuffix (temppath, suffix);
339: /* Normalize the suffix */
340: i = 0;
1.106 ! cvs 341: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 342: {
343: nsuffix[i] = utolower (suffix[i]);
344: i++;
345: }
1.106 ! cvs 346: nsuffix[i] = EOS;
! 347: if (!strcmp (nsuffix, "html") ||
! 348: !strcmp (nsuffix, "htm") ||
! 349: !strcmp (nsuffix, "shtml") ||
! 350: !strcmp (nsuffix, "jsp") ||
! 351: !strcmp (nsuffix, "xht") ||
! 352: !strcmp (nsuffix, "xhtm") ||
! 353: !strcmp (nsuffix, "xhtml"))
1.101 cvs 354: return (TRUE);
355: else
356: return (FALSE);
357: }
358: else
359: /* check if there is another suffix */
360: ExtractSuffix (temppath, suffix);
361: }
1.88 cvs 362: return (FALSE);
1.3 cvs 363: }
364:
1.4 cvs 365: /*----------------------------------------------------------------------
1.56 cvs 366: IsXMLName
367: returns TRUE if path points to an XML resource.
368: ----------------------------------------------------------------------*/
1.106 ! cvs 369: ThotBool IsXMLName (const char *path)
1.56 cvs 370: {
1.106 ! cvs 371: char temppath[MAX_LENGTH];
! 372: char suffix[MAX_LENGTH];
1.56 cvs 373:
374: if (!path)
375: return (FALSE);
376:
1.106 ! cvs 377: strcpy (temppath, path);
1.56 cvs 378: ExtractSuffix (temppath, suffix);
379:
1.106 ! cvs 380: if (!strcasecmp (suffix, "xml") ||
! 381: !strcasecmp (suffix, "xht") ||
! 382: !strcmp (suffix, "xhtm") ||
! 383: !strcmp (suffix, "xhtml"))
1.56 cvs 384: return (TRUE);
1.106 ! cvs 385: else if (!strcmp (suffix, "gz"))
1.56 cvs 386: {
387: /* take into account compressed files */
388: ExtractSuffix (temppath, suffix);
1.106 ! cvs 389: if (!strcasecmp (suffix, "xml") ||
! 390: !strcasecmp (suffix, "xht") ||
! 391: !strcmp (suffix, "xhtm") ||
! 392: !strcmp (suffix, "xhtml"))
1.60 cvs 393: return (TRUE);
394: else
395: return (FALSE);
396: }
397: else
398: return (FALSE);
399: }
400:
401: /*----------------------------------------------------------------------
1.103 cvs 402: IsMathMLName
403: returns TRUE if path points to an MathML resource.
404: ----------------------------------------------------------------------*/
1.106 ! cvs 405: ThotBool IsMathMLName (const char *path)
1.103 cvs 406: {
1.106 ! cvs 407: char temppath[MAX_LENGTH];
! 408: char suffix[MAX_LENGTH];
1.103 cvs 409:
410: if (!path)
411: return (FALSE);
412:
1.106 ! cvs 413: strcpy (temppath, path);
1.103 cvs 414: ExtractSuffix (temppath, suffix);
415:
1.106 ! cvs 416: if (!strcasecmp (suffix, "mml"))
1.103 cvs 417: return (TRUE);
1.106 ! cvs 418: else if (!strcmp (suffix, "gz"))
1.103 cvs 419: {
420: /* take into account compressed files */
421: ExtractSuffix (temppath, suffix);
1.106 ! cvs 422: if (!strcasecmp (suffix, "mml"))
1.103 cvs 423: return (TRUE);
424: else
425: return (FALSE);
426: }
427: else
428: return (FALSE);
429: }
430:
431: /*----------------------------------------------------------------------
432: IsSVGName
433: returns TRUE if path points to an MathML resource.
434: ----------------------------------------------------------------------*/
1.106 ! cvs 435: ThotBool IsSVGName (const char *path)
1.103 cvs 436: {
1.106 ! cvs 437: char temppath[MAX_LENGTH];
! 438: char suffix[MAX_LENGTH];
1.103 cvs 439:
440: if (!path)
441: return (FALSE);
442:
1.106 ! cvs 443: strcpy (temppath, path);
1.103 cvs 444: ExtractSuffix (temppath, suffix);
445:
1.106 ! cvs 446: if (!strcasecmp (suffix, "svg"))
1.103 cvs 447: return (TRUE);
1.106 ! cvs 448: else if (!strcmp (suffix, "gz"))
1.103 cvs 449: {
450: /* take into account compressed files */
451: ExtractSuffix (temppath, suffix);
1.106 ! cvs 452: if (!strcasecmp (suffix, "svg"))
1.103 cvs 453: return (TRUE);
454: else
455: return (FALSE);
456: }
457: else
458: return (FALSE);
459: }
460:
461: /*----------------------------------------------------------------------
1.60 cvs 462: IsCSSName
463: returns TRUE if path points to an XML resource.
464: ----------------------------------------------------------------------*/
1.106 ! cvs 465: ThotBool IsCSSName (const char *path)
1.60 cvs 466: {
1.106 ! cvs 467: char temppath[MAX_LENGTH];
! 468: char suffix[MAX_LENGTH];
1.60 cvs 469:
470: if (!path)
471: return (FALSE);
472:
1.106 ! cvs 473: strcpy (temppath, path);
1.60 cvs 474: ExtractSuffix (temppath, suffix);
475:
1.106 ! cvs 476: if (!strcasecmp (suffix, "css"))
1.60 cvs 477: return (TRUE);
1.106 ! cvs 478: else if (!strcmp (suffix, "gz"))
1.60 cvs 479: {
480: /* take into account compressed files */
481: ExtractSuffix (temppath, suffix);
1.106 ! cvs 482: if (!strcasecmp (suffix, "css"))
1.56 cvs 483: return (TRUE);
484: else
485: return (FALSE);
486: }
487: else
488: return (FALSE);
489: }
490:
491: /*----------------------------------------------------------------------
1.9 cvs 492: IsImageName
493: returns TRUE if path points to an image resource.
1.4 cvs 494: ----------------------------------------------------------------------*/
1.106 ! cvs 495: ThotBool IsImageName (const char *path)
! 496: {
! 497: char temppath[MAX_LENGTH];
! 498: char suffix[MAX_LENGTH];
! 499: char nsuffix[MAX_LENGTH];
1.5 cvs 500: int i;
501:
502: if (!path)
1.13 cvs 503: return (FALSE);
1.5 cvs 504:
1.106 ! cvs 505: strcpy (temppath, path);
1.5 cvs 506: ExtractSuffix (temppath, suffix);
507:
508: /* Normalize the suffix */
509: i = 0;
1.106 ! cvs 510: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 511: {
1.67 cvs 512: nsuffix[i] = utolower (suffix[i]);
1.13 cvs 513: i++;
514: }
1.106 ! cvs 515: nsuffix[i] = EOS;
! 516: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
! 517: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
! 518: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 519: return (TRUE);
520: return (FALSE);
1.3 cvs 521: }
522:
1.4 cvs 523: /*----------------------------------------------------------------------
1.58 cvs 524: IsImageType
525: returns TRUE if type points to an image resource.
526: ----------------------------------------------------------------------*/
1.106 ! cvs 527: ThotBool IsImageType (const char *type)
1.58 cvs 528: {
1.106 ! cvs 529: char temptype[MAX_LENGTH];
1.58 cvs 530: int i;
531:
532: if (!type)
533: return (FALSE);
534:
1.106 ! cvs 535: strcpy (temptype, type);
1.58 cvs 536: /* Normalize the type */
537: i = 0;
1.106 ! cvs 538: while (temptype[i] != EOS)
1.58 cvs 539: {
540: temptype[i] = tolower (temptype[i]);
541: i++;
542: }
1.106 ! cvs 543: if ((!strcmp (temptype, "gif")) || (!strcmp (temptype, "x-xbitmap")) ||
! 544: (!strcmp (temptype, "x-xpixmap")) || (!strcmp (temptype, "jpeg")) ||
! 545: (!strcmp (temptype, "png")))
1.58 cvs 546: return (TRUE);
547: return (FALSE);
548: }
549:
550: /*----------------------------------------------------------------------
1.9 cvs 551: IsTextName
1.4 cvs 552: ----------------------------------------------------------------------*/
1.106 ! cvs 553: ThotBool IsTextName (const char *path)
! 554: {
! 555: char temppath[MAX_LENGTH];
! 556: char suffix[MAX_LENGTH];
! 557: char nsuffix[MAX_LENGTH];
1.5 cvs 558: int i;
559:
560: if (!path)
1.13 cvs 561: return (FALSE);
1.5 cvs 562:
1.106 ! cvs 563: strcpy (temppath, path);
1.5 cvs 564: ExtractSuffix (temppath, suffix);
565:
566: /* Normalize the suffix */
567: i = 0;
1.106 ! cvs 568: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 569: {
1.25 cvs 570: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 571: i++;
572: }
1.106 ! cvs 573: nsuffix[i] = EOS;
1.5 cvs 574:
1.106 ! cvs 575: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 576: return (TRUE);
1.106 ! cvs 577: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 578: {
1.39 cvs 579: /* take into account compressed files */
1.13 cvs 580: ExtractSuffix (temppath, suffix);
581: /* Normalize the suffix */
582: i = 0;
1.106 ! cvs 583: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 584: {
1.25 cvs 585: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 586: i++;
587: }
1.106 ! cvs 588: nsuffix[i] = EOS;
! 589: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd")))
1.13 cvs 590: return (TRUE);
591: else
592: return (FALSE);
593: }
594: else
595: return (FALSE);
1.3 cvs 596: }
597:
1.4 cvs 598: /*----------------------------------------------------------------------
1.9 cvs 599: IsHTTPPath
600: returns TRUE if path is in fact an http URL.
1.4 cvs 601: ----------------------------------------------------------------------*/
1.106 ! cvs 602: ThotBool IsHTTPPath (const char *path)
1.3 cvs 603: {
1.5 cvs 604: if (!path)
605: return FALSE;
1.3 cvs 606:
1.106 ! cvs 607: if ((!strncmp (path, "http:", 5) != 0)
! 608: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
! 609: || !strncmp (path, "internal:", 9))
1.58 cvs 610: return TRUE;
611: return FALSE;
1.3 cvs 612: }
613:
1.4 cvs 614: /*----------------------------------------------------------------------
1.9 cvs 615: IsWithParameters
616: returns TRUE if url has a concatenated query string.
1.4 cvs 617: ----------------------------------------------------------------------*/
1.66 cvs 618: ThotBool IsWithParameters (const char *url)
1.3 cvs 619: {
1.5 cvs 620: int i;
1.3 cvs 621:
1.9 cvs 622: if ((!url) || (url[0] == EOS))
1.5 cvs 623: return FALSE;
1.3 cvs 624:
1.9 cvs 625: i = strlen (url) - 1;
626: while (i > 0 && url[i--] != '?')
1.5 cvs 627: if (i < 0)
628: return FALSE;
1.3 cvs 629:
1.5 cvs 630: /* There is a parameter */
631: return TRUE;
1.3 cvs 632: }
633:
1.4 cvs 634: /*----------------------------------------------------------------------
1.9 cvs 635: IsW3Path
636: returns TRUE if path is in fact a URL.
1.4 cvs 637: ----------------------------------------------------------------------*/
1.106 ! cvs 638: ThotBool IsW3Path (const char *path)
! 639: {
! 640: if (strncmp (path, "http:", 5) &&
! 641: strncmp (path, "ftp:", 4) &&
! 642: strncmp (path, "telnet:", 7) &&
! 643: strncmp (path, "wais:", 5) &&
! 644: strncmp (path, "news:", 5) &&
! 645: strncmp (path, "gopher:", 7) &&
! 646: strncmp (path, "mailto:", 7) &&
! 647: strncmp (path, "archie:", 7))
1.72 cvs 648: return FALSE;
649: return TRUE;
1.3 cvs 650: }
651:
1.4 cvs 652: /*----------------------------------------------------------------------
1.90 cvs 653: IsFilePath
654: returns TRUE if path is in fact a URL.
655: ----------------------------------------------------------------------*/
1.106 ! cvs 656: ThotBool IsFilePath (const char *path)
1.90 cvs 657: {
1.106 ! cvs 658: if (strncmp (path, "file:", 5))
1.90 cvs 659: return FALSE;
660: return TRUE;
661: }
662:
663: /*----------------------------------------------------------------------
1.9 cvs 664: IsValidProtocol
665: returns true if the url protocol is supported by Amaya.
1.4 cvs 666: ----------------------------------------------------------------------*/
1.106 ! cvs 667: ThotBool IsValidProtocol (const char *url)
! 668: {
! 669: if (!strncmp (url, "http:", 5)
! 670: || !strncmp (url, "internal:", 9)
! 671: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 672: /* experimental */
1.24 cvs 673: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 674: return (TRUE);
1.5 cvs 675: else
1.8 cvs 676: return (FALSE);
1.3 cvs 677: }
678:
1.31 cvs 679:
680: /*----------------------------------------------------------------------
681: GetBaseURL
682: normalizes orgName according to a base associated with doc, and
683: following the standard URL format rules.
684: The function returns the base used to solve relative URL and SRC:
685: - the base of the document,
686: - or the document path (without document name).
687: ----------------------------------------------------------------------*/
1.106 ! cvs 688: char *GetBaseURL (Document doc)
1.31 cvs 689: {
690: Element el;
691: ElementType elType;
692: AttributeType attrType;
693: Attribute attr;
1.106 ! cvs 694: char *ptr, *basename;
1.31 cvs 695: int length;
696:
1.57 cvs 697: /* @@@ irene */
698: if (!DocumentURLs[doc])
699: return NULL;
1.106 ! cvs 700: basename = TtaGetMemory (MAX_LENGTH);
! 701: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
! 702: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 703: length = MAX_LENGTH -1;
704: /* get the root element */
705: el = TtaGetMainRoot (doc);
706: /* search the BASE element */
707: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.65 cvs 708: elType.ElTypeNum = HTML_EL_HEAD;
709: el = TtaSearchTypedElement (elType, SearchForward, el);
710: if (el)
711: {
712: elType.ElTypeNum = HTML_EL_BASE;
713: el = TtaSearchTypedElement (elType, SearchInTree, el);
714: }
1.31 cvs 715: if (el)
716: {
717: /* The document has a BASE element -> Get the HREF attribute */
718: attrType.AttrSSchema = elType.ElSSchema;
719: attrType.AttrTypeNum = HTML_ATTR_HREF_;
720: attr = TtaGetAttribute (el, attrType);
721: if (attr)
722: {
723: /* Use the base path of the document */
724: TtaGiveTextAttributeValue (attr, basename, &length);
725: /* base and orgName have to be separated by a DIR_SEP */
726: length--;
1.106 ! cvs 727: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 728: /* verify if the base has the form "protocol://server:port" */
729: {
1.106 ! cvs 730: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1.33 cvs 731: AMAYA_PARSE_HOST |
732: AMAYA_PARSE_PUNCTUATION);
1.106 ! cvs 733: if (ptr && !strcmp (ptr, basename))
1.31 cvs 734: {
1.43 cvs 735: /* it has this form, we complete it by adding a URL_STR */
1.106 ! cvs 736: if (strchr (basename, DIR_SEP))
! 737: strcat (basename, DIR_STR);
1.43 cvs 738: else
1.106 ! cvs 739: strcat (basename, URL_STR);
1.31 cvs 740: length++;
741: }
742: if (ptr)
743: TtaFreeMemory (ptr);
744: }
745: }
1.33 cvs 746: }
747:
1.31 cvs 748: /* Remove anything after the last DIR_SEP char. If no such char is found,
749: * then search for the first ":" char, hoping that what's before that is a
750: * protocol. If found, end the string there. If neither char is found,
751: * then discard the whole base element.
752: */
1.106 ! cvs 753: length = strlen (basename) - 1;
1.31 cvs 754: /* search for the last DIR_SEP char */
1.106 ! cvs 755: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 756: length--;
757: if (length >= 0)
758: /* found the last DIR_SEP char, end the string there */
1.106 ! cvs 759: basename[length + 1] = EOS;
1.31 cvs 760: else
761: /* search for the first PATH_STR char */
762: {
1.106 ! cvs 763: for (length = 0; basename[length] != ':' &&
! 764: basename[length] != EOS; length ++);
! 765: if (basename[length] == ':')
1.31 cvs 766: /* found, so end the string there */
1.106 ! cvs 767: basename[length + 1] = EOS;
1.31 cvs 768: else
769: /* not found, discard the base */
1.106 ! cvs 770: basename[0] = EOS;
1.31 cvs 771: }
772: return (basename);
773: }
774:
775:
1.4 cvs 776: /*----------------------------------------------------------------------
1.40 cvs 777: GetLocalPath
778: Allocate and return the local document path associated to the url
779: ----------------------------------------------------------------------*/
1.106 ! cvs 780: char *GetLocalPath (Document doc, char *url)
! 781: {
! 782: char *ptr;
! 783: char *n;
! 784: char *documentname;
! 785: char url_sep;
1.83 cvs 786: int len;
1.67 cvs 787: ThotBool noFile;
1.40 cvs 788:
789: if (url != NULL)
790: {
791: /* check whether the file name exists */
1.106 ! cvs 792: len = strlen (url) - 1;
1.71 cvs 793: if (IsW3Path (url))
1.106 ! cvs 794: url_sep = '/';
1.41 cvs 795: else
1.106 ! cvs 796: url_sep = DIR_SEP;
1.41 cvs 797: noFile = (url[len] == url_sep);
1.40 cvs 798: if (noFile)
1.106 ! cvs 799: url[len] = EOS;
! 800: ptr = TtaGetMemory (MAX_LENGTH);
! 801: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 802: TtaExtractName (url, ptr, documentname);
1.106 ! cvs 803: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 804: if (!TtaCheckDirectory (ptr))
805: /* directory did not exist */
1.72 cvs 806: TtaMakeDirectory (ptr);
1.47 cvs 807:
808: /* don't include the query string within document name */
1.106 ! cvs 809: n = strrchr (documentname, '?');
1.47 cvs 810: if (n != NULL)
1.106 ! cvs 811: *n = EOS;
1.46 cvs 812: /* don't include ':' within document name */
1.106 ! cvs 813: n = strchr (documentname, ':');
1.46 cvs 814: if (n != NULL)
1.106 ! cvs 815: *n = EOS;
1.69 cvs 816: /* if after all this operations document name
817: is empty, let's use noname.html instead */
1.106 ! cvs 818: if (documentname[0] == EOS)
! 819: strcat (ptr, "noname.html");
1.69 cvs 820: else
1.106 ! cvs 821: strcat (ptr, documentname);
1.40 cvs 822: TtaFreeMemory (documentname);
823: /* restore the url */
824: if (noFile)
1.41 cvs 825: url[len] = url_sep;
1.40 cvs 826: return (ptr);
827: }
828: else
829: return (NULL);
830: }
831:
1.73 cvs 832: /*----------------------------------------------------------------------
1.79 cvs 833: ExtractTarget extract the target name from document nane.
834: ----------------------------------------------------------------------*/
1.106 ! cvs 835: void ExtractTarget (char *aName, char *target)
1.79 cvs 836: {
1.106 ! cvs 837: int lg, i;
! 838: char *ptr;
! 839: char *oldptr;
1.79 cvs 840:
841: if (!target || !aName)
842: /* bad target */
843: return;
844:
1.106 ! cvs 845: target[0] = EOS;
! 846: lg = strlen (aName);
1.79 cvs 847: if (lg)
848: {
849: /* the name is not empty */
850: oldptr = ptr = &aName[0];
851: do
852: {
1.106 ! cvs 853: ptr = strrchr (oldptr, '#');
1.79 cvs 854: if (ptr)
855: oldptr = &ptr[1];
856: }
857: while (ptr);
858:
859: i = (int) (oldptr) - (int) (aName); /* name length */
860: if (i > 1)
861: {
1.106 ! cvs 862: aName[i - 1] = EOS;
1.79 cvs 863: if (i != lg)
1.106 ! cvs 864: strcpy (target, oldptr);
1.79 cvs 865: }
866: }
867: }
868:
869: /*----------------------------------------------------------------------
1.90 cvs 870: RemoveNewLines (text)
871: Removes any '\n' chars that are found in text.
872: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 873: ----------------------------------------------------------------------*/
1.106 ! cvs 874: ThotBool RemoveNewLines (char *text)
! 875: {
! 876: ThotBool change = FALSE;
! 877: char *src;
! 878: char *dest;
1.90 cvs 879:
880: src = text;
881: dest = text;
882: while (*src)
883: {
884: switch (*src)
885: {
1.106 ! cvs 886: case '\n':
1.90 cvs 887: /* don't copy the newline */
888: change = 1;
889: break;
890: default:
891: *dest = *src;
892: dest++;
893: break;
894: }
895: src++;
896: }
897: /* copy the last EOS char */
898: *dest = *src;
899:
900: return (change);
901: }
902:
903: /*----------------------------------------------------------------------
904: CleanCopyFileURL
905: Copies a file url from a src string to destination string.
1.97 cvs 906: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
907: we want to do).
1.90 cvs 908: ----------------------------------------------------------------------*/
1.106 ! cvs 909: static void CleanCopyFileURL (char *dest, char *src,
! 910: ConvertionType convertion)
1.90 cvs 911: {
912: while (*src)
1.89 cvs 913: {
1.90 cvs 914: switch (*src)
1.89 cvs 915: {
916: #ifdef _WINDOWS
1.106 ! cvs 917: case URL_SEP:
1.96 cvs 918: /* make DIR_SEP transformation */
1.97 cvs 919: if (convertion & AM_CONV_URL_SEP)
1.106 ! cvs 920: *dest = DIR_SEP;
1.96 cvs 921: else
922: *dest = *src;
1.90 cvs 923: dest++;
1.96 cvs 924: src++;
1.90 cvs 925: break;
1.89 cvs 926: #endif /* _WINDOWS */
1.96 cvs 927:
1.106 ! cvs 928: case '%':
1.97 cvs 929: if (convertion & AM_CONV_PERCENT)
1.96 cvs 930: {
1.97 cvs 931: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 932: src++;
1.106 ! cvs 933: if (*src != EOS)
1.97 cvs 934: {
935: *dest = UnEscapeChar (*src) * 16;
936: src++;
937: }
1.106 ! cvs 938: if (*src != EOS)
1.97 cvs 939: {
940: *dest = *dest + UnEscapeChar (*src);
941: src++;
942: }
943: dest++;
1.96 cvs 944: }
1.97 cvs 945: else
1.96 cvs 946: {
1.97 cvs 947: *dest = *src;
948: dest++;
1.96 cvs 949: src++;
950: }
951: break;
952:
1.90 cvs 953: default:
954: *dest = *src;
1.89 cvs 955: dest++;
1.96 cvs 956: src++;
1.90 cvs 957: break;
1.89 cvs 958: }
959: }
1.90 cvs 960: /* copy the EOS char */
961: *dest = *src;
1.73 cvs 962: }
1.40 cvs 963:
964: /*----------------------------------------------------------------------
1.9 cvs 965: NormalizeURL
966: normalizes orgName according to a base associated with doc, and
967: following the standard URL format rules.
1.53 cvs 968: if doc is 0 and otherPath not NULL, normalizes orgName according to this
969: other path.
1.9 cvs 970: The function returns the new complete and normalized URL
1.12 cvs 971: or file name path (newName) and the name of the document (docName).
1.9 cvs 972: N.B. If the function can't find out what's the docName, it assigns
973: the name "noname.html".
1.4 cvs 974: ----------------------------------------------------------------------*/
1.106 ! cvs 975: void NormalizeURL (char *orgName, Document doc, char *newName,
! 976: char *docName, char *otherPath)
! 977: {
! 978: char *basename;
! 979: char tempOrgName[MAX_LENGTH];
! 980: char *ptr;
! 981: char used_sep;
1.84 cvs 982: int length;
983: ThotBool check;
1.5 cvs 984:
1.44 cvs 985: # ifdef _WINDOWS
986: int ndx;
987: # endif /* _WINDOWS */
988:
1.5 cvs 989: if (!newName || !docName)
990: return;
1.18 cvs 991:
1.32 cvs 992: if (doc != 0)
1.53 cvs 993: basename = GetBaseURL (doc);
994: else if (otherPath != NULL)
1.84 cvs 995: basename = TtaWCSdup (otherPath);
1.32 cvs 996: else
1.53 cvs 997: basename = NULL;
1.32 cvs 998:
1.18 cvs 999: /*
1.31 cvs 1000: * Clean orgName
1001: * Make sure we have a complete orgName, without any leading or trailing
1002: * white spaces, or trailinbg new lines
1003: */
1.5 cvs 1004: ptr = orgName;
1.18 cvs 1005: /* skip leading white space and new line characters */
1.106 ! cvs 1006: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
! 1007: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
! 1008: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1009: /*
1.31 cvs 1010: * Make orgName a complete URL
1011: * If the URL does not include a protocol, then try to calculate
1012: * one using the doc's base element (if it exists),
1013: */
1.106 ! cvs 1014: if (tempOrgName[0] == EOS)
1.53 cvs 1015: {
1.106 ! cvs 1016: newName[0] = EOS;
! 1017: docName[0] = EOS;
1.53 cvs 1018: TtaFreeMemory (basename);
1019: return;
1020: }
1.49 cvs 1021:
1022: /* clean trailing white space */
1.106 ! cvs 1023: length = strlen (tempOrgName) - 1;
! 1024: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1025: {
1.106 ! cvs 1026: tempOrgName[length] = EOS;
1.53 cvs 1027: length--;
1028: }
1.50 cvs 1029:
1.55 cvs 1030: /* remove extra dot (which dot???) */
1031: /* ugly, but faster than a strcmp */
1.106 ! cvs 1032: if (tempOrgName[length] == '.'
! 1033: && (length == 0 || tempOrgName[length-1] != '.'))
! 1034: tempOrgName[length] = EOS;
1.50 cvs 1035:
1.94 cvs 1036: if (IsW3Path (tempOrgName))
1.53 cvs 1037: {
1038: /* the name is complete, go to the Sixth Step */
1.106 ! cvs 1039: strcpy (newName, tempOrgName);
1.53 cvs 1040: SimplifyUrl (&newName);
1041: /* verify if the URL has the form "protocol://server:port" */
1.106 ! cvs 1042: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
! 1043: if (ptr && !strcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
! 1044: strcat (newName, URL_STR);
1.49 cvs 1045:
1.53 cvs 1046: if (ptr)
1.50 cvs 1047: TtaFreeMemory (ptr);
1.53 cvs 1048: }
1049: else if ( basename == NULL)
1050: /* the name is complete, go to the Sixth Step */
1.106 ! cvs 1051: strcpy (newName, tempOrgName);
1.53 cvs 1052: else
1053: {
1.31 cvs 1054: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 1055: # ifdef _WINDOWS
1.53 cvs 1056: if (!IsW3Path (basename))
1057: {
1.106 ! cvs 1058: length = strlen (tempOrgName);
1.53 cvs 1059: for (ndx = 0; ndx < length; ndx++)
1.106 ! cvs 1060: if (tempOrgName [ndx] == '/')
! 1061: tempOrgName [ndx] = '\\';
1.53 cvs 1062: }
1.44 cvs 1063: # endif /* _WINDOWS */
1.25 cvs 1064: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1065: if (ptr)
1066: {
1067: SimplifyUrl (&ptr);
1.106 ! cvs 1068: strcpy (newName, ptr);
1.53 cvs 1069: TtaFreeMemory (ptr);
1070: }
1071: else
1.106 ! cvs 1072: newName[0] = EOS;
1.53 cvs 1073: }
1.36 cvs 1074:
1075: TtaFreeMemory (basename);
1.18 cvs 1076: /*
1.31 cvs 1077: * Prepare the docname that will refer to this ressource in the
1078: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1079: * noname.html as a default ressource name
1.18 cvs 1080: */
1.106 ! cvs 1081: if (newName[0] != EOS)
1.53 cvs 1082: {
1.106 ! cvs 1083: length = strlen (newName) - 1;
! 1084: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1085: {
1086: used_sep = newName[length];
1087: check = TRUE;
1088: while (check)
1089: {
1.50 cvs 1090: length--;
1091: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1092: length--;
1.106 ! cvs 1093: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1094: {
1.106 ! cvs 1095: newName[length+1] = EOS;
1.53 cvs 1096: /* remove also previous directory */
1097: length--;
1098: while (length >= 0 && newName[length] != used_sep)
1099: length--;
1.106 ! cvs 1100: if (strncmp (&newName[length+1], "//", 2))
1.53 cvs 1101: /* don't remove server name */
1.106 ! cvs 1102: newName[length+1] = EOS;
1.53 cvs 1103: }
1.106 ! cvs 1104: else if (!strncmp (&newName[length+1], ".", 1))
! 1105: newName[length+1] = EOS;
1.50 cvs 1106: else
1.53 cvs 1107: check = FALSE;
1108: }
1.106 ! cvs 1109: strcpy (docName, "noname.html");
1.53 cvs 1110: /* docname was not comprised inside the URL, so let's */
1111: /* assign the default ressource name */
1.106 ! cvs 1112: strcpy (docName, "noname.html");
1.53 cvs 1113: }
1114: else
1115: { /* docname is comprised inside the URL */
1.106 ! cvs 1116: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP)
1.53 cvs 1117: length--;
1118: if (length < 0)
1.106 ! cvs 1119: strcpy (docName, newName);
1.53 cvs 1120: else
1.106 ! cvs 1121: strcpy (docName, &newName[length+1]);
1.53 cvs 1122: }
1123: }
1124: else
1.106 ! cvs 1125: docName[0] = EOS;
1.18 cvs 1126: }
1.3 cvs 1127:
1.4 cvs 1128: /*----------------------------------------------------------------------
1.9 cvs 1129: IsSameHost
1.4 cvs 1130: ----------------------------------------------------------------------*/
1.106 ! cvs 1131: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1132: {
1.106 ! cvs 1133: char *basename_ptr1, *basename_ptr2;
! 1134: ThotBool result;
1.3 cvs 1135:
1.106 ! cvs 1136: basename_ptr1 = AmayaParseUrl (url1, "",
! 1137: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
! 1138: basename_ptr2 = AmayaParseUrl (url2, "",
! 1139: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1140:
1.106 ! cvs 1141: if (strcmp (basename_ptr1, basename_ptr2))
! 1142: result = FALSE;
! 1143: else
! 1144: result = TRUE;
! 1145: TtaFreeMemory (basename_ptr1);
! 1146: TtaFreeMemory (basename_ptr2);
! 1147: return (result);
1.3 cvs 1148: }
1149:
1150:
1.4 cvs 1151: /*----------------------------------------------------------------------
1.22 cvs 1152: HasKnownFileSuffix
1153: returns TRUE if path points to a file ending with a suffix.
1154: ----------------------------------------------------------------------*/
1.106 ! cvs 1155: ThotBool HasKnownFileSuffix (const char *path)
! 1156: {
! 1157: char *root;
! 1158: char temppath[MAX_LENGTH];
! 1159: char suffix[MAX_LENGTH];
1.22 cvs 1160:
1.106 ! cvs 1161: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1162: return (FALSE);
1163:
1.106 ! cvs 1164: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1165:
1166: if (root)
1167: {
1.106 ! cvs 1168: strcpy (temppath, root);
1.25 cvs 1169: TtaFreeMemory (root);
1.22 cvs 1170: /* Get the suffix */
1171: ExtractSuffix (temppath, suffix);
1172:
1.106 ! cvs 1173: if( suffix[0] == EOS)
1.22 cvs 1174: /* no suffix */
1175: return (FALSE);
1176:
1177: /* Normalize the suffix */
1178: ConvertToLowerCase (suffix);
1179:
1.106 ! cvs 1180: if (!strcmp (suffix, "gz"))
1.22 cvs 1181: /* skip the compressed suffix */
1182: {
1183: ExtractSuffix (temppath, suffix);
1.106 ! cvs 1184: if(suffix[0] == EOS)
1.22 cvs 1185: /* no suffix */
1186: return (FALSE);
1187: /* Normalize the suffix */
1188: ConvertToLowerCase (suffix);
1189: }
1190:
1.106 ! cvs 1191: if (strcmp (suffix, "gif") &&
! 1192: strcmp (suffix, "xbm") &&
! 1193: strcmp (suffix, "xpm") &&
! 1194: strcmp (suffix, "jpg") &&
! 1195: strcmp (suffix, "pdf") &&
! 1196: strcmp (suffix, "png") &&
! 1197: strcmp (suffix, "tgz") &&
! 1198: strcmp (suffix, "xpg") &&
! 1199: strcmp (suffix, "xpd") &&
! 1200: strcmp (suffix, "ps") &&
! 1201: strcmp (suffix, "au") &&
! 1202: strcmp (suffix, "html") &&
! 1203: strcmp (suffix, "htm") &&
! 1204: strcmp (suffix, "shtml") &&
! 1205: strcmp (suffix, "xht") &&
! 1206: strcmp (suffix, "xhtm") &&
! 1207: strcmp (suffix, "xhtml") &&
! 1208: strcmp (suffix, "txt") &&
! 1209: strcmp (suffix, "css") &&
! 1210: strcmp (suffix, "eps"))
1.22 cvs 1211: return (FALSE);
1212: else
1213: return (TRUE);
1214: }
1215: else
1216: return (FALSE);
1217: }
1218:
1219:
1220: /*----------------------------------------------------------------------
1.24 cvs 1221: ChopURL
1222: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1223: If inputURL is bigger than that size, outputURL receives
1224: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1225: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1226: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1227: copied into outputURL.
1228: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1229: chars.
1230: ----------------------------------------------------------------------*/
1.106 ! cvs 1231: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1232: {
1233: int len;
1.9 cvs 1234:
1.106 ! cvs 1235: len = strlen (inputURL);
1.24 cvs 1236: if (len <= MAX_PRINT_URL_LENGTH)
1.106 ! cvs 1237: strcpy (outputURL, inputURL);
1.24 cvs 1238: else
1239: /* make a truncated urlName on the status window */
1240: {
1.106 ! cvs 1241: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
! 1242: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
! 1243: strcat (outputURL, "...");
! 1244: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1245: }
1.25 cvs 1246: }
1247:
1248:
1249: /*----------------------------------------------------------------------
1250: scan
1.47 cvs 1251: Scan a filename for its constituents
1.25 cvs 1252: -----------------------------------
1253:
1254: On entry,
1255: name points to a document name which may be incomplete.
1256: On exit,
1257: absolute or relative may be nonzero (but not both).
1258: host, fragment and access may be nonzero if they were specified.
1259: Any which are nonzero point to zero terminated strings.
1260: ----------------------------------------------------------------------*/
1.106 ! cvs 1261: static void scan (char *name, HTURI *parts)
1.25 cvs 1262: {
1.106 ! cvs 1263: char * p;
! 1264: char * after_access = name;
1.32 cvs 1265:
1.43 cvs 1266: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1267: /* Look for fragment identifier */
1.106 ! cvs 1268: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1269: {
1.106 ! cvs 1270: *p++ = '\0';
1.28 cvs 1271: parts->fragment = p;
1.25 cvs 1272: }
1273:
1.28 cvs 1274: for (p=name; *p; p++)
1275: {
1.106 ! cvs 1276: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1277: break;
1.106 ! cvs 1278: if (*p == ':')
1.28 cvs 1279: {
1280: *p = 0;
1281: parts->access = after_access; /* Scheme has been specified */
1282:
1283: /* The combination of gcc, the "-O" flag and the HP platform is
1284: unhealthy. The following three lines is a quick & dirty fix, but is
1285: not recommended. Rather, turn off "-O". */
1286:
1287: /* after_access = p;*/
1288: /* while (*after_access == 0)*/
1289: /* after_access++;*/
1290: after_access = p+1;
1.106 ! cvs 1291: if (!strcasecmp("URL", parts->access))
1.28 cvs 1292: /* Ignore IETF's URL: pre-prefix */
1293: parts->access = NULL;
1294: else
1.25 cvs 1295: break;
1296: }
1297: }
1298:
1299: p = after_access;
1.43 cvs 1300: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1301: {
1.43 cvs 1302: if (p[1] == URL_SEP)
1.28 cvs 1303: {
1.25 cvs 1304: parts->host = p+2; /* host has been specified */
1.28 cvs 1305: *p = 0; /* Terminate access */
1306: /* look for end of host name if any */
1.106 ! cvs 1307: p = strchr (parts->host, URL_SEP);
1.28 cvs 1308: if (p)
1309: {
1.106 ! cvs 1310: *p = EOS; /* Terminate host */
1.25 cvs 1311: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1312: }
1313: }
1314: else
1315: /* Root found but no host */
1316: parts->absolute = p+1;
1317: }
1318: else
1319: {
1.25 cvs 1320: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1321: }
1.25 cvs 1322: }
1323:
1324:
1325: /*----------------------------------------------------------------------
1.28 cvs 1326: AmayaParseUrl: parse a Name relative to another name
1327:
1328: This returns those parts of a name which are given (and requested)
1329: substituting bits from the related name where necessary.
1.25 cvs 1330:
1.28 cvs 1331: On entry,
1.25 cvs 1332: aName A filename given
1333: relatedName A name relative to which aName is to be parsed. Give
1334: it an empty string if aName is absolute.
1335: wanted A mask for the bits which are wanted.
1336:
1.28 cvs 1337: On exit,
1.25 cvs 1338: returns A pointer to a malloc'd string which MUST BE FREED
1339: ----------------------------------------------------------------------*/
1.106 ! cvs 1340: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
! 1341: {
! 1342: char *return_value;
! 1343: char result[MAX_LENGTH];
! 1344: char name[MAX_LENGTH];
! 1345: char rel[MAX_LENGTH];
! 1346: char *p, *access;
1.29 cvs 1347: HTURI given, related;
1348: int len;
1.106 ! cvs 1349: char used_sep;
! 1350: char *used_str;
1.32 cvs 1351:
1.106 ! cvs 1352: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1353: {
1.106 ! cvs 1354: used_str = DIR_STR;
! 1355: used_sep = DIR_SEP;
1.33 cvs 1356: }
1.32 cvs 1357: else
1.33 cvs 1358: {
1.106 ! cvs 1359: used_str = URL_STR;
! 1360: used_sep = URL_SEP;
1.33 cvs 1361: }
1.32 cvs 1362:
1.29 cvs 1363: /* Make working copies of input strings to cut up: */
1364: return_value = NULL;
1365: result[0] = 0; /* Clear string */
1.106 ! cvs 1366: strcpy (name, aName);
1.29 cvs 1367: if (relatedName != NULL)
1.106 ! cvs 1368: strcpy (rel, relatedName);
1.29 cvs 1369: else
1.106 ! cvs 1370: relatedName[0] = EOS;
1.29 cvs 1371:
1372: scan (name, &given);
1373: scan (rel, &related);
1374: access = given.access ? given.access : related.access;
1375: if (wanted & AMAYA_PARSE_ACCESS)
1376: if (access)
1377: {
1.106 ! cvs 1378: strcat (result, access);
1.29 cvs 1379: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 ! cvs 1380: strcat (result, ":");
1.29 cvs 1381: }
1382:
1383: if (given.access && related.access)
1384: /* If different, inherit nothing. */
1.106 ! cvs 1385: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1386: {
1387: related.host = 0;
1388: related.absolute = 0;
1389: related.relative = 0;
1390: related.fragment = 0;
1391: }
1392:
1393: if (wanted & AMAYA_PARSE_HOST)
1394: if(given.host || related.host)
1395: {
1396: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 ! cvs 1397: strcat (result, "//");
! 1398: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1399: }
1400:
1401: if (given.host && related.host)
1402: /* If different hosts, inherit no path. */
1.106 ! cvs 1403: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1404: {
1405: related.absolute = 0;
1406: related.relative = 0;
1407: related.fragment = 0;
1408: }
1409:
1410: if (wanted & AMAYA_PARSE_PATH)
1411: {
1412: if (given.absolute)
1413: {
1414: /* All is given */
1415: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 ! cvs 1416: strcat (result, used_str);
! 1417: strcat (result, given.absolute);
1.25 cvs 1418: }
1.29 cvs 1419: else if (related.absolute)
1420: {
1421: /* Adopt path not name */
1.106 ! cvs 1422: strcat (result, used_str);
! 1423: strcat (result, related.absolute);
1.29 cvs 1424: if (given.relative)
1425: {
1426: /* Search part? */
1.106 ! cvs 1427: p = strchr (result, '?');
1.29 cvs 1428: if (!p)
1.106 ! cvs 1429: p=result+strlen(result)-1;
1.33 cvs 1430: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1431: /* Remove filename */
1432: p[1]=0;
1433: /* Add given one */
1.106 ! cvs 1434: strcat (result, given.relative);
1.25 cvs 1435: }
1436: }
1.29 cvs 1437: else if (given.relative)
1438: /* what we've got */
1.106 ! cvs 1439: strcat (result, given.relative);
1.29 cvs 1440: else if (related.relative)
1.106 ! cvs 1441: strcat (result, related.relative);
1.29 cvs 1442: else
1443: /* No inheritance */
1.106 ! cvs 1444: strcat (result, used_str);
1.25 cvs 1445: }
1.29 cvs 1446:
1447: if (wanted & AMAYA_PARSE_ANCHOR)
1448: if (given.fragment || related.fragment)
1449: {
1450: if (given.absolute && given.fragment)
1451: {
1452: /*Fixes for relURLs...*/
1453: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 ! cvs 1454: strcat (result, "#");
! 1455: strcat (result, given.fragment);
1.29 cvs 1456: }
1457: else if (!(given.absolute) && !(given.fragment))
1.106 ! cvs 1458: strcat (result, "");
1.29 cvs 1459: else
1460: {
1461: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 ! cvs 1462: strcat (result, "#");
! 1463: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1464: }
1465: }
1.106 ! cvs 1466: len = strlen (result);
! 1467: if ((return_value = TtaGetMemory (len + 1)) != NULL)
! 1468: strcpy (return_value, result);
1.29 cvs 1469: return (return_value); /* exactly the right length */
1.25 cvs 1470: }
1471:
1472: /*----------------------------------------------------------------------
1473: HTCanon
1474: Canonicalizes the URL in the following manner starting from the host
1475: pointer:
1476:
1477: 1) The host name is converted to lowercase
1478: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1479:
1480: Return: OK The position of the current path part of the URL
1481: which might be the old one or a new one.
1482:
1483: ----------------------------------------------------------------------*/
1.106 ! cvs 1484: static char *HTCanon (char **filename, char *host)
! 1485: {
! 1486: char *newname = NULL;
! 1487: char used_sep;
! 1488: char *path;
! 1489: char *strptr;
! 1490: char *port;
! 1491: char *access = host-3;
! 1492:
! 1493: if (*filename && strchr (*filename, URL_SEP))
! 1494: used_sep = URL_SEP;
1.33 cvs 1495: else
1.106 ! cvs 1496: used_sep = DIR_SEP;
1.32 cvs 1497:
1.82 cvs 1498: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1499: access--;
1.106 ! cvs 1500: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
! 1501: path = host + strlen (host);
! 1502: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1503: host = strptr;
1.106 ! cvs 1504: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1505: port = NULL;
1.25 cvs 1506:
1507: strptr = host; /* Convert to lower-case */
1.82 cvs 1508: while (strptr < path)
1.33 cvs 1509: {
1.84 cvs 1510: *strptr = utolower (*strptr);
1.82 cvs 1511: strptr++;
1.33 cvs 1512: }
1.25 cvs 1513:
1514: /* Does the URL contain a full domain name? This also works for a
1515: numerical host name. The domain name is already made lower-case
1516: and without a trailing dot. */
1517: {
1.106 ! cvs 1518: char *dot = port ? port : path;
! 1519: if (dot > *filename && *--dot == '.')
1.33 cvs 1520: {
1.106 ! cvs 1521: char *orig = dot;
! 1522: char *dest = dot + 1;
1.82 cvs 1523: while ((*orig++ = *dest++));
1524: if (port) port--;
1.33 cvs 1525: path--;
1.25 cvs 1526: }
1527: }
1528: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1529: if (port)
1530: {
1.82 cvs 1531: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1532: {
1533: if (!newname)
1534: {
1.106 ! cvs 1535: char *orig = port;
! 1536: char *dest = port + 1;
1.82 cvs 1537: while ((*orig++ = *dest++));
1.33 cvs 1538: }
1539: }
1.106 ! cvs 1540: else if ((!strncmp (access, "http", 4) &&
! 1541: (*(port + 1) == '8' &&
! 1542: *(port+2) == '0' &&
1.82 cvs 1543: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 ! cvs 1544: (!strncmp (access, "gopher", 6) &&
! 1545: (*(port+1) == '7' &&
! 1546: *(port+2) == '0' &&
1.82 cvs 1547: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 ! cvs 1548: (!strncmp (access, "ftp", 3) &&
! 1549: (*(port+1) == '2' &&
! 1550: *(port + 2) == '1' &&
1.82 cvs 1551: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1552: if (!newname)
1553: {
1.106 ! cvs 1554: char *orig = port;
! 1555: char *dest = port + 3;
1.33 cvs 1556: while((*orig++ = *dest++));
1557: /* Update path position, Henry Minsky */
1558: path -= 3;
1.25 cvs 1559: }
1.33 cvs 1560: }
1561: else if (newname)
1.106 ! cvs 1562: strncat (newname, port, (int) (path - port));
1.33 cvs 1563: }
1.25 cvs 1564:
1.33 cvs 1565: if (newname)
1566: {
1.106 ! cvs 1567: char *newpath = newname + strlen (newname);
! 1568: strcat (newname, path);
1.25 cvs 1569: path = newpath;
1.28 cvs 1570: /* Free old copy */
1571: TtaFreeMemory(*filename);
1.25 cvs 1572: *filename = newname;
1.33 cvs 1573: }
1.25 cvs 1574: return path;
1575: }
1576:
1577:
1578: /*----------------------------------------------------------------------
1.29 cvs 1579: SimplifyUrl: simplify a URI
1.32 cvs 1580: A URI is allowed to contain the sequence xxx/../ which may be
1581: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1582: Simplification helps us recognize duplicate URIs.
1.25 cvs 1583:
1.28 cvs 1584: Thus, /etc/junk/../fred becomes /etc/fred
1585: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1586:
1.28 cvs 1587: but we should NOT change
1588: http://fred.xxx.edu/../..
1.25 cvs 1589:
1590: or ../../albert.html
1591:
1.28 cvs 1592: In order to avoid empty URLs the following URLs become:
1.25 cvs 1593:
1594: /fred/.. becomes /fred/..
1595: /fred/././.. becomes /fred/..
1596: /fred/.././junk/.././ becomes /fred/..
1597:
1.28 cvs 1598: If more than one set of `://' is found (several proxies in cascade) then
1599: only the part after the last `://' is simplified.
1.25 cvs 1600:
1.28 cvs 1601: Returns: A string which might be the old one or a new one.
1.25 cvs 1602: ----------------------------------------------------------------------*/
1.106 ! cvs 1603: void SimplifyUrl (char **url)
! 1604: {
! 1605: char *path;
! 1606: char *access;
! 1607: char *newptr;
! 1608: char *p;
! 1609: char *orig, *dest, *end;
1.28 cvs 1610:
1.106 ! cvs 1611: char used_sep;
1.77 cvs 1612: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1613: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1614:
1615:
1.28 cvs 1616: if (!url || !*url)
1617: return;
1618:
1.106 ! cvs 1619: if (strchr (*url, URL_SEP))
! 1620: used_sep = URL_SEP;
1.32 cvs 1621: else
1.106 ! cvs 1622: used_sep = DIR_SEP;
1.32 cvs 1623:
1.77 cvs 1624: /* should we simplify double dot? */
1625: path = *url;
1.106 ! cvs 1626: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1627: ddot_simplify = FALSE;
1628: else
1629: ddot_simplify = TRUE;
1630:
1.28 cvs 1631: /* Find any scheme name */
1.106 ! cvs 1632: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1633: {
1634: /* Find host name */
1.28 cvs 1635: access = *url;
1.84 cvs 1636: while (access < path && (*access = utolower (*access)))
1.82 cvs 1637: access++;
1.28 cvs 1638: path += 3;
1.106 ! cvs 1639: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1640: /* For proxies */
1.106 ! cvs 1641: path = newptr + 3;
1.82 cvs 1642: /* We have a host name */
1.84 cvs 1643: path = HTCanon (url, path);
1.25 cvs 1644: }
1.106 ! cvs 1645: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1646: path += 2;
1647: else
1648: path = *url;
1.25 cvs 1649:
1.84 cvs 1650: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1651: /* Some URLs start //<foo> */
1652: path += 1;
1.94 cvs 1653: else if (IsFilePath (path))
1654: {
1655: /* doesn't need to do anything more */
1656: return;
1657: }
1.106 ! cvs 1658: else if (!strncmp (path, "news:", 5))
1.28 cvs 1659: {
1.106 ! cvs 1660: newptr = strchr (path+5, '@');
1.28 cvs 1661: if (!newptr)
1662: newptr = path + 5;
1663: while (*newptr)
1664: {
1665: /* Make group or host lower case */
1.84 cvs 1666: *newptr = utolower (*newptr);
1.28 cvs 1667: newptr++;
1.25 cvs 1668: }
1.28 cvs 1669: /* Doesn't need to do any more */
1670: return;
1.25 cvs 1671: }
1.28 cvs 1672:
1673: if ((p = path))
1674: {
1.106 ! cvs 1675: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
! 1676: (end = strchr (path, '#'))))
! 1677: end = path + strlen (path);
1.28 cvs 1678:
1679: /* Parse string second time to simplify */
1680: p = path;
1681: while (p < end)
1682: {
1.77 cvs 1683: /* if we're pointing to a char, it's safe to reactivate the ../ convertion */
1.106 ! cvs 1684: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1685: ddot_simplify = TRUE;
1686:
1.33 cvs 1687: if (*p==used_sep)
1.28 cvs 1688: {
1.106 ! cvs 1689: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1690: {
1691: orig = p + 1;
1.84 cvs 1692: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1693: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1694: end = orig - 1;
1695: }
1.106 ! cvs 1696: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1697: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1698: {
1699: newptr = p;
1.52 cvs 1700: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1701: if (*newptr == used_sep)
1702: orig = newptr + 1;
1.28 cvs 1703: else
1.52 cvs 1704: orig = newptr;
1705:
1706: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1707: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1708: end = orig-1;
1709: /* Start again with prev slash */
1710: p = newptr;
1.28 cvs 1711: }
1.33 cvs 1712: else if (*(p+1) == used_sep)
1.28 cvs 1713: {
1.33 cvs 1714: while (*(p+1) == used_sep)
1.28 cvs 1715: {
1716: orig = p;
1717: dest = p + 1;
1718: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1719: end = orig-1;
1720: }
1721: }
1722: else
1.25 cvs 1723: p++;
1.28 cvs 1724: }
1725: else
1726: p++;
1.25 cvs 1727: }
1728: }
1.51 cvs 1729:
1730: /*
1731: ** Check for host/../.. kind of things
1732: */
1.106 ! cvs 1733: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1734: && (!*(path+3) || *(path+3) == used_sep))
1.106 ! cvs 1735: *(path+1) = EOS;
1.51 cvs 1736:
1.28 cvs 1737: return;
1738: }
1739:
1740:
1741: /*----------------------------------------------------------------------
1.96 cvs 1742: NormalizeFile normalizes local names.
1.28 cvs 1743: Return TRUE if target and src differ.
1744: ----------------------------------------------------------------------*/
1.106 ! cvs 1745: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 1746: {
1.93 cvs 1747: # ifndef _WINDOWS
1.106 ! cvs 1748: char *s;
1.93 cvs 1749: int i;
1750: # endif /* !_WINDOWS */
1.82 cvs 1751: ThotBool change;
1.90 cvs 1752: int start_index; /* the first char that we'll copy */
1.28 cvs 1753:
1.54 cvs 1754: change = FALSE;
1.90 cvs 1755: start_index = 0;
1756:
1.106 ! cvs 1757: if (!src || src[0] == EOS)
1.96 cvs 1758: {
1.106 ! cvs 1759: target[0] = EOS;
1.96 cvs 1760: return FALSE;
1761: }
1.90 cvs 1762:
1763: /* @@ do I need file: or file:/ here? */
1.106 ! cvs 1764: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1765: {
1.90 cvs 1766: /* remove the prefix file: */
1767: start_index += 5;
1768:
1769: /* remove the localhost prefix */
1.106 ! cvs 1770: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 1771: start_index += 11;
1772:
1773: /* remove the first two slashes in / / /path */
1774: while (src[start_index] &&
1.106 ! cvs 1775: src[start_index] == '/'
! 1776: && src[start_index + 1] == '/')
1.94 cvs 1777: start_index++;
1778:
1779: #ifdef _WINDOWS
1780: /* remove any extra slash before the drive name */
1.106 ! cvs 1781: if (src[start_index] == '/'
! 1782: &&src[start_index+2] == ':')
1.94 cvs 1783: start_index++;
1784: #endif /* _WINDOWS */
1.90 cvs 1785:
1.106 ! cvs 1786: if (src[start_index] == EOS)
1.90 cvs 1787: /* if there's nothing afterwards, add a DIR_STR */
1.106 ! cvs 1788: strcpy (target, DIR_STR);
1.90 cvs 1789: else
1.97 cvs 1790: /* as we're inside a file: URL, we'll apply all the convertions
1791: we know */
1792: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1793:
1794: change = TRUE;
1795: }
1.97 cvs 1796: else if (convertion != AM_CONV_NONE)
1.96 cvs 1797: {
1798: /* we are following a "local" relative link, we do all the
1799: convertions except for the HOME_DIR ~ one */
1.97 cvs 1800: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1801: }
1.90 cvs 1802: #ifndef _WINDOWS
1.106 ! cvs 1803: else if (src[0] == '~')
1.53 cvs 1804: {
1.96 cvs 1805: /* it must be a URL typed in a text input field */
1806: /* do the HOME_DIR ~ substitution */
1.82 cvs 1807: s = TtaGetEnvString ("HOME");
1.106 ! cvs 1808: strcpy (target, s);
1.90 cvs 1809: #if 0
1.96 cvs 1810: /* JK: invalidated this part of the code as it's simpler
1811: to add the DIR_SEP whenever we have something to add
1812: to the path rather than adding it systematically */
1.106 ! cvs 1813: if (src[1] != DIR_SEP)
! 1814: strcat (target, DIR_STR);
1.90 cvs 1815: #endif
1.106 ! cvs 1816: i = strlen (target);
! 1817: strcpy (&target[i], &src[1]);
1.54 cvs 1818: change = TRUE;
1.53 cvs 1819: }
1.90 cvs 1820: #endif /* _WINDOWS */
1.28 cvs 1821: else
1.96 cvs 1822: /* leave it as it is */
1.106 ! cvs 1823: strcpy (target, src);
1.96 cvs 1824:
1.28 cvs 1825: /* remove /../ and /./ */
1.29 cvs 1826: SimplifyUrl (&target);
1.54 cvs 1827: if (!change)
1.106 ! cvs 1828: change = strcmp (src, target);
1.28 cvs 1829: return (change);
1.25 cvs 1830: }
1831:
1.28 cvs 1832:
1.25 cvs 1833: /*----------------------------------------------------------------------
1.31 cvs 1834: MakeRelativeURL: make relative name
1.25 cvs 1835:
1.28 cvs 1836: This function creates and returns a string which gives an expression of
1837: one address as related to another. Where there is no relation, an absolute
1838: address is retured.
1.25 cvs 1839:
1.28 cvs 1840: On entry,
1.25 cvs 1841: Both names must be absolute, fully qualified names of nodes
1842: (no fragment bits)
1843:
1.28 cvs 1844: On exit,
1.25 cvs 1845: The return result points to a newly allocated name which, if
1846: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1847: The caller is responsible for freeing the resulting name later.
1848: ----------------------------------------------------------------------*/
1.106 ! cvs 1849: char *MakeRelativeURL (char *aName, char *relatedName)
! 1850: {
! 1851: char *return_value;
! 1852: char result[MAX_LENGTH];
! 1853: char *p;
! 1854: char *q;
! 1855: char *after_access;
! 1856: char *last_slash = NULL;
! 1857: int slashes, levels, len;
1.29 cvs 1858:
1.44 cvs 1859: # ifdef _WINDOWS
1860: int ndx;
1861: # endif /* _WINDOWS */
1862:
1.29 cvs 1863: if (aName == NULL || relatedName == NULL)
1864: return (NULL);
1865:
1866: slashes = 0;
1867: after_access = NULL;
1868: p = aName;
1869: q = relatedName;
1870: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1871: {
1872: /* Find extent of match */
1.106 ! cvs 1873: if (*p == ':')
1.29 cvs 1874: after_access = p + 1;
1.28 cvs 1875: if (*p == DIR_SEP)
1.27 cvs 1876: {
1.29 cvs 1877: /* memorize the last slash position and count them */
1.27 cvs 1878: last_slash = p;
1879: slashes++;
1.25 cvs 1880: }
1881: }
1882:
1.31 cvs 1883: /* q, p point to the first non-matching character or zero */
1.106 ! cvs 1884: if (*q == EOS)
1.31 cvs 1885: {
1886: /* New name is a subset of the related name */
1887: /* exactly the right length */
1.106 ! cvs 1888: len = strlen (p);
! 1889: if ((return_value = TtaGetMemory (len + 1)) != NULL)
! 1890: strcpy (return_value, p);
1.31 cvs 1891: }
1892: else if ((slashes < 2 && after_access == NULL)
1893: || (slashes < 3 && after_access != NULL))
1894: {
1895: /* Two names whitout common path */
1896: /* exactly the right length */
1.106 ! cvs 1897: len = strlen (aName);
! 1898: if ((return_value = TtaGetMemory (len + 1)) != NULL)
! 1899: strcpy (return_value, aName);
1.31 cvs 1900: }
1901: else
1902: {
1903: /* Some path in common */
1.106 ! cvs 1904: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 1905: /* just the same server */
1.106 ! cvs 1906: strcpy (result, last_slash);
1.31 cvs 1907: else
1908: {
1909: levels= 0;
1.106 ! cvs 1910: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 1911: if (*q == DIR_SEP)
1912: levels++;
1913:
1.106 ! cvs 1914: result[0] = EOS;
1.31 cvs 1915: for (;levels; levels--)
1.106 ! cvs 1916: strcat (result, "../");
! 1917: strcat (result, last_slash+1);
1.31 cvs 1918: }
1.52 cvs 1919:
1920: if (!*result)
1.106 ! cvs 1921: strcat (result, "./");
1.52 cvs 1922:
1.31 cvs 1923: /* exactly the right length */
1.106 ! cvs 1924: len = strlen (result);
! 1925: if ((return_value = TtaGetMemory (len + 1)) != NULL)
! 1926: strcpy (return_value, result);
1.52 cvs 1927:
1.25 cvs 1928: }
1.44 cvs 1929: # ifdef _WINDOWS
1.106 ! cvs 1930: len = strlen (return_value);
1.44 cvs 1931: for (ndx = 0; ndx < len; ndx ++)
1.106 ! cvs 1932: if (return_value[ndx] == '\\')
! 1933: return_value[ndx] = '/' ;
1.44 cvs 1934: # endif /* _WINDOWS */
1.29 cvs 1935: return (return_value);
1.24 cvs 1936: }
1.35 cvs 1937:
1.104 kahan 1938: /*----------------------------------------------------------------------
1939: AM_GetFileSize
1940: Returns TRUE and the filesize in the 2nd parameter.
1941: Otherwise, in case of a system error, returns FALSE, with a
1942: filesize of 0L.
1943: ---------------------------------------------------------------------*/
1.106 ! cvs 1944: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 1945: {
1.106 ! cvs 1946: ThotFileHandle handle = ThotFile_BADHANDLE;
! 1947: ThotFileInfo info;
1.35 cvs 1948:
1.104 kahan 1949: *file_size = 0L;
1950: if (!TtaFileExist (filename))
1951: return FALSE;
1952:
1953: handle = TtaFileOpen (filename, ThotFile_READWRITE);
1954: if (handle == ThotFile_BADHANDLE)
1955: /* ThotFile_BADHANDLE */
1956: return FALSE;
1957: if (TtaFileStat (handle, &info) == 0)
1958: /* bad stat */
1959: info.size = 0L;
1960: TtaFileClose (handle);
1961: *file_size = (unsigned long) info.size;
1962: return TRUE;
1963: }
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