Annotation of Amaya/amaya/AHTURLTools.c, revision 1.111
1.7 cvs 1: /*
2: *
1.110 cvs 3: * (c) COPYRIGHT MIT and INRIA, 1996-2001
1.7 cvs 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: ----------------------------------------------------------------------*/
1.109 cvs 55: void EscapeChar (char *string, char c)
1.75 cvs 56: {
1.109 cvs 57: unsigned int i;
58:
59: i = (unsigned char) c & 0xFF;
60: sprintf (string, "%02x", i);
1.75 cvs 61: }
62:
63: /*----------------------------------------------------------------------
1.96 cvs 64: UnEscapeChar
65: writes the equivalent hex code to a %xx coded char
66: ----------------------------------------------------------------------*/
1.109 cvs 67: static char UnEscapeChar (char c)
1.96 cvs 68: {
1.106 cvs 69: return c >= '0' && c <= '9' ? c - '0'
70: : c >= 'A' && c <= 'F' ? c - 'A' + 10
71: : c - 'a' + 10; /* accept small letters just in case */
1.96 cvs 72: }
73:
74: /*----------------------------------------------------------------------
1.75 cvs 75: EscapeURL
76: Takes a URL and escapes all protected chars into
77: %xx sequences. Also, removes any leading white spaces
78: Returns either NULL or a new buffer, which must be freed by the caller
79: ----------------------------------------------------------------------*/
1.106 cvs 80: char *EscapeURL (const char *url)
81: {
82: char *buffer;
83: int buffer_len;
84: int buffer_free_mem;
85: char *ptr;
86: int new_chars;
1.75 cvs 87: void *status;
88:
89: if (url && *url)
90: {
1.106 cvs 91: buffer_free_mem = strlen (url) + 20;
92: buffer = TtaGetMemory (buffer_free_mem + 1);
1.107 kahan 93: ptr = (char *) url;
1.75 cvs 94: buffer_len = 0;
95:
96: while (*ptr)
97: {
98: switch (*ptr)
99: {
100: /* put here below all the chars that need to
101: be escaped into %xx */
1.81 cvs 102: case 0x27: /* & */
103: case 0x20: /* space */
1.75 cvs 104: new_chars = 3;
105: break;
106:
107: default:
108: new_chars = 1;
109: break;
110: }
111:
112: /* see if we need extra room in the buffer */
113: if (new_chars > buffer_free_mem)
114: {
1.76 cvs 115: buffer_free_mem = 20;
1.106 cvs 116: status = TtaRealloc (buffer, sizeof (char)
1.75 cvs 117: * (buffer_len + buffer_free_mem + 1));
118: if (status)
119: buffer = (STRING) status;
1.106 cvs 120: else
121: {
122: /* @@ maybe we should do some other behavior here, like
123: freeing the buffer and return a void thing */
124: buffer[buffer_len] = EOS;
125: break;
126: }
1.75 cvs 127: }
128: /* escape the char */
129: if (new_chars == 3)
130: {
1.106 cvs 131: buffer[buffer_len] = '%';
1.75 cvs 132: EscapeChar (&buffer[buffer_len+1], *ptr);
133: }
134: else
135: buffer[buffer_len] = *ptr;
136:
137: /* update the status */
138: buffer_len += new_chars;
139: buffer_free_mem -= new_chars;
140: /* examine the next char */
141: ptr++;
142: }
1.106 cvs 143: buffer[buffer_len] = EOS;
1.75 cvs 144: }
1.76 cvs 145: else
146: buffer = NULL;
147:
1.75 cvs 148: return (buffer);
149: }
150:
151:
152: /*----------------------------------------------------------------------
1.11 cvs 153: ExplodeURL
1.8 cvs 154: ----------------------------------------------------------------------*/
1.106 cvs 155: void ExplodeURL (char *url, char **proto, char **host, char **dir,
156: char **file)
1.8 cvs 157: {
1.33 cvs 158: char *curr, *temp;
159: char used_sep;
1.32 cvs 160:
1.33 cvs 161: if (url && strchr (url, URL_SEP))
162: used_sep = URL_SEP;
163: else
164: used_sep = DIR_SEP;
1.8 cvs 165:
166: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
167: (dir == NULL) || (file == NULL))
168: return;
169:
170: /* initialize every pointer */
171: *proto = *host = *dir = *file = NULL;
172:
173: /* skip any leading space */
174: while ((*url == SPACE) || (*url == TAB))
175: url++;
1.9 cvs 176: curr = url;
177: if (*curr == 0)
1.8 cvs 178: goto finished;
179:
180: /* go to the end of the URL */
1.68 cvs 181: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
182: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 183: curr++;
1.8 cvs 184:
185: /* mark the end of the chain */
1.9 cvs 186: *curr = EOS;
187: curr--;
188: if (curr <= url)
1.8 cvs 189: goto finished;
190:
191: /* search the next DIR_SEP indicating the beginning of the file name */
192: do
1.11 cvs 193: curr--;
1.33 cvs 194: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 195:
1.9 cvs 196: if (curr < url)
1.8 cvs 197: goto finished;
1.9 cvs 198: *file = curr + 1;
1.8 cvs 199:
200: /* mark the end of the dir */
1.9 cvs 201: *curr = EOS;
202: curr--;
203: if (curr < url)
1.8 cvs 204: goto finished;
205:
1.29 cvs 206: /* search for the DIR_STR indicating the host name start */
1.33 cvs 207: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 208: curr--;
1.8 cvs 209:
210: /* if we found it, separate the host name from the directory */
1.102 kahan 211: if ((*curr == used_sep) && (*(curr + 1) == used_sep))
1.8 cvs 212: {
1.9 cvs 213: *host = temp = curr + 2;
1.33 cvs 214: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 215: temp++;
1.33 cvs 216: if (*temp == used_sep)
1.8 cvs 217: {
218: *temp = EOS;
219: *dir = temp + 1;
220: }
221: }
222: else
1.11 cvs 223: *dir = curr;
224:
1.9 cvs 225: if (curr <= url)
1.8 cvs 226: goto finished;
227:
228: /* mark the end of the proto */
1.9 cvs 229: *curr = EOS;
230: curr--;
231: if (curr < url)
1.8 cvs 232: goto finished;
233:
1.106 cvs 234: if (*curr == ':')
1.8 cvs 235: {
1.9 cvs 236: *curr = EOS;
237: curr--;
1.8 cvs 238: }
239: else
240: goto finished;
1.11 cvs 241:
1.9 cvs 242: if (curr < url)
1.8 cvs 243: goto finished;
1.9 cvs 244: while ((curr > url) && (isalpha (*curr)))
245: curr--;
246: *proto = curr;
1.8 cvs 247:
248: finished:;
249:
250: #ifdef AMAYA_DEBUG
251: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
252: if (*proto)
253: fprintf (stderr, "proto : %s, ", *proto);
254: if (*host)
255: fprintf (stderr, "host : %s, ", *host);
256: if (*dir)
257: fprintf (stderr, "dir : %s, ", *dir);
258: if (*file)
259: fprintf (stderr, "file : %s ", *file);
260: fprintf (stderr, "\n");
261: #endif
262:
263: }
1.3 cvs 264:
1.61 cvs 265:
266: /*----------------------------------------------------------------------
267: ExtractSuffix extract suffix from document nane.
268: ----------------------------------------------------------------------*/
1.106 cvs 269: void ExtractSuffix (char *aName, char *aSuffix)
1.61 cvs 270: {
1.106 cvs 271: int lg, i;
272: char *ptr, *oldptr;
1.61 cvs 273:
274: if (!aSuffix || !aName)
275: /* bad suffix */
276: return;
277:
1.106 cvs 278: aSuffix[0] = EOS;
279: lg = strlen (aName);
1.61 cvs 280: if (lg)
281: {
282: /* the name is not empty */
283: oldptr = ptr = &aName[0];
284: do
285: {
1.106 cvs 286: ptr = strrchr (oldptr, '.');
1.61 cvs 287: if (ptr)
288: oldptr = &ptr[1];
289: }
290: while (ptr);
291:
292: i = (int) (oldptr) - (int) (aName); /* name length */
293: if (i > 1)
294: {
1.106 cvs 295: aName[i - 1] = EOS;
1.61 cvs 296: if (i != lg)
1.106 cvs 297: strcpy (aSuffix, oldptr);
1.61 cvs 298: }
299: }
300: }
301:
1.4 cvs 302: /*----------------------------------------------------------------------
1.9 cvs 303: IsHTMLName
304: returns TRUE if path points to an HTML resource.
1.4 cvs 305: ----------------------------------------------------------------------*/
1.109 cvs 306: ThotBool IsHTMLName (const char *path)
1.106 cvs 307: {
308: char temppath[MAX_LENGTH];
309: char suffix[MAX_LENGTH];
310: char nsuffix[MAX_LENGTH];
1.101 cvs 311: int i;
1.5 cvs 312:
1.101 cvs 313: if (!path)
314: return (FALSE);
1.5 cvs 315:
1.106 cvs 316: strcpy (temppath, path);
1.101 cvs 317: ExtractSuffix (temppath, suffix);
318: i = 0;
1.106 cvs 319: while (suffix[i] != EOS)
1.101 cvs 320: {
321: /* Normalize the suffix */
322: i = 0;
1.106 cvs 323: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 324: {
325: nsuffix[i] = utolower (suffix[i]);
326: i++;
327: }
1.106 cvs 328: nsuffix[i] = EOS;
329: if (!strcmp (nsuffix, "html") ||
330: !strcmp (nsuffix, "htm") ||
331: !strcmp (nsuffix, "shtml") ||
332: !strcmp (nsuffix, "jsp") ||
333: !strcmp (nsuffix, "xht") ||
334: !strcmp (nsuffix, "xhtm") ||
335: !strcmp (nsuffix, "xhtml"))
1.101 cvs 336: return (TRUE);
1.106 cvs 337: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 338: {
339: /* take into account compressed files */
340: ExtractSuffix (temppath, suffix);
341: /* Normalize the suffix */
342: i = 0;
1.106 cvs 343: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 344: {
345: nsuffix[i] = utolower (suffix[i]);
346: i++;
347: }
1.106 cvs 348: nsuffix[i] = EOS;
349: if (!strcmp (nsuffix, "html") ||
350: !strcmp (nsuffix, "htm") ||
351: !strcmp (nsuffix, "shtml") ||
352: !strcmp (nsuffix, "jsp") ||
353: !strcmp (nsuffix, "xht") ||
354: !strcmp (nsuffix, "xhtm") ||
355: !strcmp (nsuffix, "xhtml"))
1.101 cvs 356: return (TRUE);
357: else
358: return (FALSE);
359: }
360: else
361: /* check if there is another suffix */
362: ExtractSuffix (temppath, suffix);
363: }
1.88 cvs 364: return (FALSE);
1.3 cvs 365: }
366:
1.4 cvs 367: /*----------------------------------------------------------------------
1.56 cvs 368: IsXMLName
369: returns TRUE if path points to an XML resource.
370: ----------------------------------------------------------------------*/
1.111 ! cvs 371: ThotBool IsXMLName (const char *path)
1.56 cvs 372: {
1.106 cvs 373: char temppath[MAX_LENGTH];
374: char suffix[MAX_LENGTH];
1.56 cvs 375:
376: if (!path)
377: return (FALSE);
378:
1.106 cvs 379: strcpy (temppath, path);
1.56 cvs 380: ExtractSuffix (temppath, suffix);
381:
1.106 cvs 382: if (!strcasecmp (suffix, "xml") ||
383: !strcasecmp (suffix, "xht") ||
384: !strcmp (suffix, "xhtm") ||
385: !strcmp (suffix, "xhtml"))
1.56 cvs 386: return (TRUE);
1.106 cvs 387: else if (!strcmp (suffix, "gz"))
1.56 cvs 388: {
389: /* take into account compressed files */
390: ExtractSuffix (temppath, suffix);
1.106 cvs 391: if (!strcasecmp (suffix, "xml") ||
392: !strcasecmp (suffix, "xht") ||
393: !strcmp (suffix, "xhtm") ||
394: !strcmp (suffix, "xhtml"))
1.60 cvs 395: return (TRUE);
396: else
397: return (FALSE);
398: }
399: else
400: return (FALSE);
401: }
402:
403: /*----------------------------------------------------------------------
1.103 cvs 404: IsMathMLName
405: returns TRUE if path points to an MathML resource.
406: ----------------------------------------------------------------------*/
1.111 ! cvs 407: ThotBool IsMathMLName (const char *path)
1.103 cvs 408: {
1.106 cvs 409: char temppath[MAX_LENGTH];
410: char suffix[MAX_LENGTH];
1.103 cvs 411:
412: if (!path)
413: return (FALSE);
414:
1.106 cvs 415: strcpy (temppath, path);
1.103 cvs 416: ExtractSuffix (temppath, suffix);
417:
1.106 cvs 418: if (!strcasecmp (suffix, "mml"))
1.103 cvs 419: return (TRUE);
1.106 cvs 420: else if (!strcmp (suffix, "gz"))
1.103 cvs 421: {
422: /* take into account compressed files */
423: ExtractSuffix (temppath, suffix);
1.106 cvs 424: if (!strcasecmp (suffix, "mml"))
1.103 cvs 425: return (TRUE);
426: else
427: return (FALSE);
428: }
429: else
430: return (FALSE);
431: }
432:
433: /*----------------------------------------------------------------------
434: IsSVGName
435: returns TRUE if path points to an MathML resource.
436: ----------------------------------------------------------------------*/
1.111 ! cvs 437: ThotBool IsSVGName (const char *path)
1.103 cvs 438: {
1.106 cvs 439: char temppath[MAX_LENGTH];
440: char suffix[MAX_LENGTH];
1.103 cvs 441:
442: if (!path)
443: return (FALSE);
444:
1.106 cvs 445: strcpy (temppath, path);
1.103 cvs 446: ExtractSuffix (temppath, suffix);
447:
1.106 cvs 448: if (!strcasecmp (suffix, "svg"))
1.103 cvs 449: return (TRUE);
1.106 cvs 450: else if (!strcmp (suffix, "gz"))
1.103 cvs 451: {
452: /* take into account compressed files */
453: ExtractSuffix (temppath, suffix);
1.106 cvs 454: if (!strcasecmp (suffix, "svg"))
1.103 cvs 455: return (TRUE);
456: else
457: return (FALSE);
458: }
459: else
460: return (FALSE);
461: }
462:
463: /*----------------------------------------------------------------------
1.60 cvs 464: IsCSSName
465: returns TRUE if path points to an XML resource.
466: ----------------------------------------------------------------------*/
1.111 ! cvs 467: ThotBool IsCSSName (const char *path)
1.60 cvs 468: {
1.106 cvs 469: char temppath[MAX_LENGTH];
470: char suffix[MAX_LENGTH];
1.60 cvs 471:
472: if (!path)
473: return (FALSE);
474:
1.106 cvs 475: strcpy (temppath, path);
1.60 cvs 476: ExtractSuffix (temppath, suffix);
477:
1.106 cvs 478: if (!strcasecmp (suffix, "css"))
1.60 cvs 479: return (TRUE);
1.106 cvs 480: else if (!strcmp (suffix, "gz"))
1.60 cvs 481: {
482: /* take into account compressed files */
483: ExtractSuffix (temppath, suffix);
1.106 cvs 484: if (!strcasecmp (suffix, "css"))
1.56 cvs 485: return (TRUE);
486: else
487: return (FALSE);
488: }
489: else
490: return (FALSE);
491: }
492:
493: /*----------------------------------------------------------------------
1.9 cvs 494: IsImageName
495: returns TRUE if path points to an image resource.
1.4 cvs 496: ----------------------------------------------------------------------*/
1.111 ! cvs 497: ThotBool IsImageName (const char *path)
1.106 cvs 498: {
499: char temppath[MAX_LENGTH];
500: char suffix[MAX_LENGTH];
501: char nsuffix[MAX_LENGTH];
1.5 cvs 502: int i;
503:
504: if (!path)
1.13 cvs 505: return (FALSE);
1.5 cvs 506:
1.106 cvs 507: strcpy (temppath, path);
1.5 cvs 508: ExtractSuffix (temppath, suffix);
509:
510: /* Normalize the suffix */
511: i = 0;
1.106 cvs 512: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 513: {
1.67 cvs 514: nsuffix[i] = utolower (suffix[i]);
1.13 cvs 515: i++;
516: }
1.106 cvs 517: nsuffix[i] = EOS;
518: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
519: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
520: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 521: return (TRUE);
522: return (FALSE);
1.3 cvs 523: }
524:
1.4 cvs 525: /*----------------------------------------------------------------------
1.58 cvs 526: IsImageType
527: returns TRUE if type points to an image resource.
528: ----------------------------------------------------------------------*/
1.111 ! cvs 529: ThotBool IsImageType (const char *type)
1.58 cvs 530: {
1.106 cvs 531: char temptype[MAX_LENGTH];
1.58 cvs 532: int i;
533:
534: if (!type)
535: return (FALSE);
536:
1.106 cvs 537: strcpy (temptype, type);
1.58 cvs 538: /* Normalize the type */
539: i = 0;
1.106 cvs 540: while (temptype[i] != EOS)
1.58 cvs 541: {
542: temptype[i] = tolower (temptype[i]);
543: i++;
544: }
1.111 ! cvs 545: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
! 546: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
! 547: !strcmp (temptype, "png"))
1.58 cvs 548: return (TRUE);
549: return (FALSE);
550: }
551:
552: /*----------------------------------------------------------------------
1.9 cvs 553: IsTextName
1.4 cvs 554: ----------------------------------------------------------------------*/
1.111 ! cvs 555: ThotBool IsTextName (const char *path)
1.106 cvs 556: {
557: char temppath[MAX_LENGTH];
558: char suffix[MAX_LENGTH];
559: char nsuffix[MAX_LENGTH];
1.5 cvs 560: int i;
561:
562: if (!path)
1.13 cvs 563: return (FALSE);
1.5 cvs 564:
1.106 cvs 565: strcpy (temppath, path);
1.5 cvs 566: ExtractSuffix (temppath, suffix);
567:
568: /* Normalize the suffix */
569: i = 0;
1.106 cvs 570: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 571: {
1.25 cvs 572: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 573: i++;
574: }
1.106 cvs 575: nsuffix[i] = EOS;
1.5 cvs 576:
1.111 ! cvs 577: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 578: return (TRUE);
1.106 cvs 579: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 580: {
1.39 cvs 581: /* take into account compressed files */
1.13 cvs 582: ExtractSuffix (temppath, suffix);
583: /* Normalize the suffix */
584: i = 0;
1.106 cvs 585: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 586: {
1.25 cvs 587: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 588: i++;
589: }
1.106 cvs 590: nsuffix[i] = EOS;
1.111 ! cvs 591: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 592: return (TRUE);
593: else
594: return (FALSE);
595: }
596: else
597: return (FALSE);
1.3 cvs 598: }
599:
1.4 cvs 600: /*----------------------------------------------------------------------
1.9 cvs 601: IsHTTPPath
602: returns TRUE if path is in fact an http URL.
1.4 cvs 603: ----------------------------------------------------------------------*/
1.106 cvs 604: ThotBool IsHTTPPath (const char *path)
1.3 cvs 605: {
1.5 cvs 606: if (!path)
607: return FALSE;
1.3 cvs 608:
1.106 cvs 609: if ((!strncmp (path, "http:", 5) != 0)
610: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
611: || !strncmp (path, "internal:", 9))
1.58 cvs 612: return TRUE;
613: return FALSE;
1.3 cvs 614: }
615:
1.4 cvs 616: /*----------------------------------------------------------------------
1.9 cvs 617: IsWithParameters
618: returns TRUE if url has a concatenated query string.
1.4 cvs 619: ----------------------------------------------------------------------*/
1.66 cvs 620: ThotBool IsWithParameters (const char *url)
1.3 cvs 621: {
1.5 cvs 622: int i;
1.3 cvs 623:
1.9 cvs 624: if ((!url) || (url[0] == EOS))
1.5 cvs 625: return FALSE;
1.3 cvs 626:
1.9 cvs 627: i = strlen (url) - 1;
628: while (i > 0 && url[i--] != '?')
1.5 cvs 629: if (i < 0)
630: return FALSE;
1.3 cvs 631:
1.5 cvs 632: /* There is a parameter */
633: return TRUE;
1.3 cvs 634: }
635:
1.4 cvs 636: /*----------------------------------------------------------------------
1.9 cvs 637: IsW3Path
638: returns TRUE if path is in fact a URL.
1.4 cvs 639: ----------------------------------------------------------------------*/
1.106 cvs 640: ThotBool IsW3Path (const char *path)
641: {
642: if (strncmp (path, "http:", 5) &&
643: strncmp (path, "ftp:", 4) &&
644: strncmp (path, "telnet:", 7) &&
645: strncmp (path, "wais:", 5) &&
646: strncmp (path, "news:", 5) &&
647: strncmp (path, "gopher:", 7) &&
648: strncmp (path, "mailto:", 7) &&
649: strncmp (path, "archie:", 7))
1.72 cvs 650: return FALSE;
651: return TRUE;
1.3 cvs 652: }
653:
1.4 cvs 654: /*----------------------------------------------------------------------
1.90 cvs 655: IsFilePath
656: returns TRUE if path is in fact a URL.
657: ----------------------------------------------------------------------*/
1.106 cvs 658: ThotBool IsFilePath (const char *path)
1.90 cvs 659: {
1.106 cvs 660: if (strncmp (path, "file:", 5))
1.90 cvs 661: return FALSE;
662: return TRUE;
663: }
664:
665: /*----------------------------------------------------------------------
1.9 cvs 666: IsValidProtocol
667: returns true if the url protocol is supported by Amaya.
1.4 cvs 668: ----------------------------------------------------------------------*/
1.106 cvs 669: ThotBool IsValidProtocol (const char *url)
670: {
671: if (!strncmp (url, "http:", 5)
672: || !strncmp (url, "internal:", 9)
673: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 674: /* experimental */
1.24 cvs 675: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 676: return (TRUE);
1.5 cvs 677: else
1.8 cvs 678: return (FALSE);
1.3 cvs 679: }
680:
1.31 cvs 681:
682: /*----------------------------------------------------------------------
683: GetBaseURL
684: normalizes orgName according to a base associated with doc, and
685: following the standard URL format rules.
686: The function returns the base used to solve relative URL and SRC:
687: - the base of the document,
688: - or the document path (without document name).
689: ----------------------------------------------------------------------*/
1.106 cvs 690: char *GetBaseURL (Document doc)
1.31 cvs 691: {
692: Element el;
693: ElementType elType;
694: AttributeType attrType;
695: Attribute attr;
1.106 cvs 696: char *ptr, *basename;
1.31 cvs 697: int length;
698:
1.57 cvs 699: /* @@@ irene */
700: if (!DocumentURLs[doc])
1.110 cvs 701: return NULL;
1.106 cvs 702: basename = TtaGetMemory (MAX_LENGTH);
703: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
704: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 705: length = MAX_LENGTH -1;
1.110 cvs 706: /* get the document element */
1.31 cvs 707: el = TtaGetMainRoot (doc);
708: /* search the BASE element */
709: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.65 cvs 710: elType.ElTypeNum = HTML_EL_HEAD;
711: el = TtaSearchTypedElement (elType, SearchForward, el);
712: if (el)
713: {
714: elType.ElTypeNum = HTML_EL_BASE;
715: el = TtaSearchTypedElement (elType, SearchInTree, el);
716: }
1.31 cvs 717: if (el)
718: {
719: /* The document has a BASE element -> Get the HREF attribute */
720: attrType.AttrSSchema = elType.ElSSchema;
721: attrType.AttrTypeNum = HTML_ATTR_HREF_;
722: attr = TtaGetAttribute (el, attrType);
723: if (attr)
724: {
725: /* Use the base path of the document */
726: TtaGiveTextAttributeValue (attr, basename, &length);
727: /* base and orgName have to be separated by a DIR_SEP */
728: length--;
1.110 cvs 729: if (basename[0] != EOS && basename[length] != URL_SEP &&
730: basename[length] != DIR_SEP)
1.31 cvs 731: /* verify if the base has the form "protocol://server:port" */
732: {
1.106 cvs 733: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1.33 cvs 734: AMAYA_PARSE_HOST |
735: AMAYA_PARSE_PUNCTUATION);
1.106 cvs 736: if (ptr && !strcmp (ptr, basename))
1.31 cvs 737: {
1.43 cvs 738: /* it has this form, we complete it by adding a URL_STR */
1.106 cvs 739: if (strchr (basename, DIR_SEP))
740: strcat (basename, DIR_STR);
1.43 cvs 741: else
1.106 cvs 742: strcat (basename, URL_STR);
1.31 cvs 743: length++;
744: }
745: if (ptr)
746: TtaFreeMemory (ptr);
747: }
748: }
1.33 cvs 749: }
750:
1.31 cvs 751: /* Remove anything after the last DIR_SEP char. If no such char is found,
752: * then search for the first ":" char, hoping that what's before that is a
753: * protocol. If found, end the string there. If neither char is found,
754: * then discard the whole base element.
755: */
1.106 cvs 756: length = strlen (basename) - 1;
1.31 cvs 757: /* search for the last DIR_SEP char */
1.106 cvs 758: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 759: length--;
760: if (length >= 0)
761: /* found the last DIR_SEP char, end the string there */
1.106 cvs 762: basename[length + 1] = EOS;
1.31 cvs 763: else
764: /* search for the first PATH_STR char */
765: {
1.106 cvs 766: for (length = 0; basename[length] != ':' &&
767: basename[length] != EOS; length ++);
768: if (basename[length] == ':')
1.31 cvs 769: /* found, so end the string there */
1.106 cvs 770: basename[length + 1] = EOS;
1.31 cvs 771: else
772: /* not found, discard the base */
1.106 cvs 773: basename[0] = EOS;
1.31 cvs 774: }
775: return (basename);
776: }
777:
778:
1.4 cvs 779: /*----------------------------------------------------------------------
1.40 cvs 780: GetLocalPath
781: Allocate and return the local document path associated to the url
782: ----------------------------------------------------------------------*/
1.106 cvs 783: char *GetLocalPath (Document doc, char *url)
784: {
785: char *ptr;
786: char *n;
787: char *documentname;
788: char url_sep;
1.83 cvs 789: int len;
1.67 cvs 790: ThotBool noFile;
1.40 cvs 791:
792: if (url != NULL)
793: {
794: /* check whether the file name exists */
1.106 cvs 795: len = strlen (url) - 1;
1.71 cvs 796: if (IsW3Path (url))
1.106 cvs 797: url_sep = '/';
1.41 cvs 798: else
1.106 cvs 799: url_sep = DIR_SEP;
1.41 cvs 800: noFile = (url[len] == url_sep);
1.40 cvs 801: if (noFile)
1.106 cvs 802: url[len] = EOS;
803: ptr = TtaGetMemory (MAX_LENGTH);
804: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 805: TtaExtractName (url, ptr, documentname);
1.106 cvs 806: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 807: if (!TtaCheckDirectory (ptr))
808: /* directory did not exist */
1.72 cvs 809: TtaMakeDirectory (ptr);
1.47 cvs 810:
811: /* don't include the query string within document name */
1.106 cvs 812: n = strrchr (documentname, '?');
1.47 cvs 813: if (n != NULL)
1.106 cvs 814: *n = EOS;
1.46 cvs 815: /* don't include ':' within document name */
1.106 cvs 816: n = strchr (documentname, ':');
1.46 cvs 817: if (n != NULL)
1.106 cvs 818: *n = EOS;
1.69 cvs 819: /* if after all this operations document name
820: is empty, let's use noname.html instead */
1.106 cvs 821: if (documentname[0] == EOS)
822: strcat (ptr, "noname.html");
1.69 cvs 823: else
1.106 cvs 824: strcat (ptr, documentname);
1.40 cvs 825: TtaFreeMemory (documentname);
826: /* restore the url */
827: if (noFile)
1.41 cvs 828: url[len] = url_sep;
1.40 cvs 829: return (ptr);
830: }
831: else
832: return (NULL);
833: }
834:
1.73 cvs 835: /*----------------------------------------------------------------------
1.79 cvs 836: ExtractTarget extract the target name from document nane.
837: ----------------------------------------------------------------------*/
1.106 cvs 838: void ExtractTarget (char *aName, char *target)
1.79 cvs 839: {
1.106 cvs 840: int lg, i;
841: char *ptr;
842: char *oldptr;
1.79 cvs 843:
844: if (!target || !aName)
845: /* bad target */
846: return;
847:
1.106 cvs 848: target[0] = EOS;
849: lg = strlen (aName);
1.79 cvs 850: if (lg)
851: {
852: /* the name is not empty */
853: oldptr = ptr = &aName[0];
854: do
855: {
1.106 cvs 856: ptr = strrchr (oldptr, '#');
1.79 cvs 857: if (ptr)
858: oldptr = &ptr[1];
859: }
860: while (ptr);
861:
862: i = (int) (oldptr) - (int) (aName); /* name length */
863: if (i > 1)
864: {
1.106 cvs 865: aName[i - 1] = EOS;
1.79 cvs 866: if (i != lg)
1.106 cvs 867: strcpy (target, oldptr);
1.79 cvs 868: }
869: }
870: }
871:
872: /*----------------------------------------------------------------------
1.90 cvs 873: RemoveNewLines (text)
874: Removes any '\n' chars that are found in text.
875: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 876: ----------------------------------------------------------------------*/
1.106 cvs 877: ThotBool RemoveNewLines (char *text)
878: {
879: ThotBool change = FALSE;
880: char *src;
881: char *dest;
1.90 cvs 882:
883: src = text;
884: dest = text;
885: while (*src)
886: {
887: switch (*src)
888: {
1.106 cvs 889: case '\n':
1.90 cvs 890: /* don't copy the newline */
891: change = 1;
892: break;
893: default:
894: *dest = *src;
895: dest++;
896: break;
897: }
898: src++;
899: }
900: /* copy the last EOS char */
901: *dest = *src;
902:
903: return (change);
904: }
905:
906: /*----------------------------------------------------------------------
907: CleanCopyFileURL
908: Copies a file url from a src string to destination string.
1.97 cvs 909: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
910: we want to do).
1.90 cvs 911: ----------------------------------------------------------------------*/
1.106 cvs 912: static void CleanCopyFileURL (char *dest, char *src,
913: ConvertionType convertion)
1.90 cvs 914: {
915: while (*src)
1.89 cvs 916: {
1.90 cvs 917: switch (*src)
1.89 cvs 918: {
919: #ifdef _WINDOWS
1.106 cvs 920: case URL_SEP:
1.96 cvs 921: /* make DIR_SEP transformation */
1.97 cvs 922: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 923: *dest = DIR_SEP;
1.96 cvs 924: else
925: *dest = *src;
1.90 cvs 926: dest++;
1.96 cvs 927: src++;
1.90 cvs 928: break;
1.89 cvs 929: #endif /* _WINDOWS */
1.96 cvs 930:
1.106 cvs 931: case '%':
1.97 cvs 932: if (convertion & AM_CONV_PERCENT)
1.96 cvs 933: {
1.97 cvs 934: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 935: src++;
1.106 cvs 936: if (*src != EOS)
1.97 cvs 937: {
938: *dest = UnEscapeChar (*src) * 16;
939: src++;
940: }
1.106 cvs 941: if (*src != EOS)
1.97 cvs 942: {
943: *dest = *dest + UnEscapeChar (*src);
944: src++;
945: }
946: dest++;
1.96 cvs 947: }
1.97 cvs 948: else
1.96 cvs 949: {
1.97 cvs 950: *dest = *src;
951: dest++;
1.96 cvs 952: src++;
953: }
954: break;
955:
1.90 cvs 956: default:
957: *dest = *src;
1.89 cvs 958: dest++;
1.96 cvs 959: src++;
1.90 cvs 960: break;
1.89 cvs 961: }
962: }
1.90 cvs 963: /* copy the EOS char */
964: *dest = *src;
1.73 cvs 965: }
1.40 cvs 966:
967: /*----------------------------------------------------------------------
1.9 cvs 968: NormalizeURL
969: normalizes orgName according to a base associated with doc, and
970: following the standard URL format rules.
1.53 cvs 971: if doc is 0 and otherPath not NULL, normalizes orgName according to this
972: other path.
1.9 cvs 973: The function returns the new complete and normalized URL
1.12 cvs 974: or file name path (newName) and the name of the document (docName).
1.9 cvs 975: N.B. If the function can't find out what's the docName, it assigns
976: the name "noname.html".
1.4 cvs 977: ----------------------------------------------------------------------*/
1.106 cvs 978: void NormalizeURL (char *orgName, Document doc, char *newName,
979: char *docName, char *otherPath)
980: {
981: char *basename;
982: char tempOrgName[MAX_LENGTH];
983: char *ptr;
984: char used_sep;
1.84 cvs 985: int length;
986: ThotBool check;
1.5 cvs 987:
1.110 cvs 988: #ifdef _WINDOWS
1.44 cvs 989: int ndx;
1.110 cvs 990: #endif /* _WINDOWS */
1.44 cvs 991:
1.5 cvs 992: if (!newName || !docName)
993: return;
1.18 cvs 994:
1.32 cvs 995: if (doc != 0)
1.53 cvs 996: basename = GetBaseURL (doc);
997: else if (otherPath != NULL)
1.108 cvs 998: basename = TtaStrdup (otherPath);
1.32 cvs 999: else
1.53 cvs 1000: basename = NULL;
1.32 cvs 1001:
1.18 cvs 1002: /*
1.31 cvs 1003: * Clean orgName
1004: * Make sure we have a complete orgName, without any leading or trailing
1005: * white spaces, or trailinbg new lines
1006: */
1.5 cvs 1007: ptr = orgName;
1.18 cvs 1008: /* skip leading white space and new line characters */
1.106 cvs 1009: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1010: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1011: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1012: /*
1.31 cvs 1013: * Make orgName a complete URL
1014: * If the URL does not include a protocol, then try to calculate
1015: * one using the doc's base element (if it exists),
1016: */
1.106 cvs 1017: if (tempOrgName[0] == EOS)
1.53 cvs 1018: {
1.106 cvs 1019: newName[0] = EOS;
1020: docName[0] = EOS;
1.53 cvs 1021: TtaFreeMemory (basename);
1022: return;
1023: }
1.49 cvs 1024:
1025: /* clean trailing white space */
1.106 cvs 1026: length = strlen (tempOrgName) - 1;
1027: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1028: {
1.106 cvs 1029: tempOrgName[length] = EOS;
1.53 cvs 1030: length--;
1031: }
1.50 cvs 1032:
1.55 cvs 1033: /* remove extra dot (which dot???) */
1034: /* ugly, but faster than a strcmp */
1.106 cvs 1035: if (tempOrgName[length] == '.'
1036: && (length == 0 || tempOrgName[length-1] != '.'))
1037: tempOrgName[length] = EOS;
1.50 cvs 1038:
1.94 cvs 1039: if (IsW3Path (tempOrgName))
1.53 cvs 1040: {
1041: /* the name is complete, go to the Sixth Step */
1.106 cvs 1042: strcpy (newName, tempOrgName);
1.53 cvs 1043: SimplifyUrl (&newName);
1044: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1045: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1046: AMAYA_PARSE_HOST |
1047: AMAYA_PARSE_PUNCTUATION);
1048: if (ptr && !strcmp (ptr, newName))
1049: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1050: strcat (newName, URL_STR);
1.49 cvs 1051:
1.53 cvs 1052: if (ptr)
1.50 cvs 1053: TtaFreeMemory (ptr);
1.53 cvs 1054: }
1055: else if ( basename == NULL)
1056: /* the name is complete, go to the Sixth Step */
1.106 cvs 1057: strcpy (newName, tempOrgName);
1.53 cvs 1058: else
1059: {
1.31 cvs 1060: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1061: #ifdef _WINDOWS
1.53 cvs 1062: if (!IsW3Path (basename))
1063: {
1.106 cvs 1064: length = strlen (tempOrgName);
1.53 cvs 1065: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1066: if (tempOrgName [ndx] == '/')
1067: tempOrgName [ndx] = '\\';
1.53 cvs 1068: }
1.110 cvs 1069: #endif /* _WINDOWS */
1.25 cvs 1070: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1071: if (ptr)
1072: {
1073: SimplifyUrl (&ptr);
1.106 cvs 1074: strcpy (newName, ptr);
1.53 cvs 1075: TtaFreeMemory (ptr);
1076: }
1077: else
1.106 cvs 1078: newName[0] = EOS;
1.53 cvs 1079: }
1.36 cvs 1080:
1081: TtaFreeMemory (basename);
1.18 cvs 1082: /*
1.31 cvs 1083: * Prepare the docname that will refer to this ressource in the
1084: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1085: * noname.html as a default ressource name
1.18 cvs 1086: */
1.106 cvs 1087: if (newName[0] != EOS)
1.53 cvs 1088: {
1.106 cvs 1089: length = strlen (newName) - 1;
1090: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1091: {
1092: used_sep = newName[length];
1093: check = TRUE;
1094: while (check)
1095: {
1.50 cvs 1096: length--;
1097: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1098: length--;
1.106 cvs 1099: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1100: {
1.106 cvs 1101: newName[length+1] = EOS;
1.53 cvs 1102: /* remove also previous directory */
1103: length--;
1104: while (length >= 0 && newName[length] != used_sep)
1105: length--;
1.106 cvs 1106: if (strncmp (&newName[length+1], "//", 2))
1.53 cvs 1107: /* don't remove server name */
1.106 cvs 1108: newName[length+1] = EOS;
1.53 cvs 1109: }
1.106 cvs 1110: else if (!strncmp (&newName[length+1], ".", 1))
1111: newName[length+1] = EOS;
1.50 cvs 1112: else
1.53 cvs 1113: check = FALSE;
1114: }
1.106 cvs 1115: strcpy (docName, "noname.html");
1.53 cvs 1116: /* docname was not comprised inside the URL, so let's */
1117: /* assign the default ressource name */
1.106 cvs 1118: strcpy (docName, "noname.html");
1.53 cvs 1119: }
1120: else
1121: { /* docname is comprised inside the URL */
1.110 cvs 1122: while (length >= 0 && newName[length] != URL_SEP &&
1123: newName[length] != DIR_SEP)
1.53 cvs 1124: length--;
1125: if (length < 0)
1.106 cvs 1126: strcpy (docName, newName);
1.53 cvs 1127: else
1.106 cvs 1128: strcpy (docName, &newName[length+1]);
1.53 cvs 1129: }
1130: }
1131: else
1.106 cvs 1132: docName[0] = EOS;
1.18 cvs 1133: }
1.3 cvs 1134:
1.4 cvs 1135: /*----------------------------------------------------------------------
1.9 cvs 1136: IsSameHost
1.4 cvs 1137: ----------------------------------------------------------------------*/
1.106 cvs 1138: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1139: {
1.106 cvs 1140: char *basename_ptr1, *basename_ptr2;
1141: ThotBool result;
1.3 cvs 1142:
1.106 cvs 1143: basename_ptr1 = AmayaParseUrl (url1, "",
1144: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1145: basename_ptr2 = AmayaParseUrl (url2, "",
1146: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1147:
1.106 cvs 1148: if (strcmp (basename_ptr1, basename_ptr2))
1149: result = FALSE;
1150: else
1151: result = TRUE;
1152: TtaFreeMemory (basename_ptr1);
1153: TtaFreeMemory (basename_ptr2);
1154: return (result);
1.3 cvs 1155: }
1156:
1157:
1.4 cvs 1158: /*----------------------------------------------------------------------
1.22 cvs 1159: HasKnownFileSuffix
1160: returns TRUE if path points to a file ending with a suffix.
1161: ----------------------------------------------------------------------*/
1.106 cvs 1162: ThotBool HasKnownFileSuffix (const char *path)
1163: {
1164: char *root;
1165: char temppath[MAX_LENGTH];
1166: char suffix[MAX_LENGTH];
1.22 cvs 1167:
1.106 cvs 1168: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1169: return (FALSE);
1170:
1.106 cvs 1171: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1172:
1173: if (root)
1174: {
1.106 cvs 1175: strcpy (temppath, root);
1.25 cvs 1176: TtaFreeMemory (root);
1.22 cvs 1177: /* Get the suffix */
1178: ExtractSuffix (temppath, suffix);
1179:
1.106 cvs 1180: if( suffix[0] == EOS)
1.22 cvs 1181: /* no suffix */
1182: return (FALSE);
1183:
1184: /* Normalize the suffix */
1185: ConvertToLowerCase (suffix);
1186:
1.106 cvs 1187: if (!strcmp (suffix, "gz"))
1.22 cvs 1188: /* skip the compressed suffix */
1189: {
1190: ExtractSuffix (temppath, suffix);
1.106 cvs 1191: if(suffix[0] == EOS)
1.22 cvs 1192: /* no suffix */
1193: return (FALSE);
1194: /* Normalize the suffix */
1195: ConvertToLowerCase (suffix);
1196: }
1197:
1.106 cvs 1198: if (strcmp (suffix, "gif") &&
1199: strcmp (suffix, "xbm") &&
1200: strcmp (suffix, "xpm") &&
1201: strcmp (suffix, "jpg") &&
1202: strcmp (suffix, "pdf") &&
1203: strcmp (suffix, "png") &&
1204: strcmp (suffix, "tgz") &&
1205: strcmp (suffix, "xpg") &&
1206: strcmp (suffix, "xpd") &&
1207: strcmp (suffix, "ps") &&
1208: strcmp (suffix, "au") &&
1209: strcmp (suffix, "html") &&
1210: strcmp (suffix, "htm") &&
1211: strcmp (suffix, "shtml") &&
1212: strcmp (suffix, "xht") &&
1213: strcmp (suffix, "xhtm") &&
1214: strcmp (suffix, "xhtml") &&
1215: strcmp (suffix, "txt") &&
1216: strcmp (suffix, "css") &&
1217: strcmp (suffix, "eps"))
1.22 cvs 1218: return (FALSE);
1219: else
1220: return (TRUE);
1221: }
1222: else
1223: return (FALSE);
1224: }
1225:
1226:
1227: /*----------------------------------------------------------------------
1.24 cvs 1228: ChopURL
1229: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1230: If inputURL is bigger than that size, outputURL receives
1231: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1232: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1233: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1234: copied into outputURL.
1235: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1236: chars.
1237: ----------------------------------------------------------------------*/
1.106 cvs 1238: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1239: {
1240: int len;
1.9 cvs 1241:
1.106 cvs 1242: len = strlen (inputURL);
1.24 cvs 1243: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1244: strcpy (outputURL, inputURL);
1.24 cvs 1245: else
1246: /* make a truncated urlName on the status window */
1247: {
1.106 cvs 1248: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1249: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1250: strcat (outputURL, "...");
1251: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1252: }
1.25 cvs 1253: }
1254:
1255:
1256: /*----------------------------------------------------------------------
1257: scan
1.47 cvs 1258: Scan a filename for its constituents
1.25 cvs 1259: -----------------------------------
1260:
1261: On entry,
1262: name points to a document name which may be incomplete.
1263: On exit,
1264: absolute or relative may be nonzero (but not both).
1265: host, fragment and access may be nonzero if they were specified.
1266: Any which are nonzero point to zero terminated strings.
1267: ----------------------------------------------------------------------*/
1.106 cvs 1268: static void scan (char *name, HTURI *parts)
1.25 cvs 1269: {
1.106 cvs 1270: char * p;
1271: char * after_access = name;
1.32 cvs 1272:
1.43 cvs 1273: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1274: /* Look for fragment identifier */
1.106 cvs 1275: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1276: {
1.106 cvs 1277: *p++ = '\0';
1.28 cvs 1278: parts->fragment = p;
1.25 cvs 1279: }
1280:
1.28 cvs 1281: for (p=name; *p; p++)
1282: {
1.106 cvs 1283: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1284: break;
1.106 cvs 1285: if (*p == ':')
1.28 cvs 1286: {
1287: *p = 0;
1288: parts->access = after_access; /* Scheme has been specified */
1289:
1290: /* The combination of gcc, the "-O" flag and the HP platform is
1291: unhealthy. The following three lines is a quick & dirty fix, but is
1292: not recommended. Rather, turn off "-O". */
1293:
1294: /* after_access = p;*/
1295: /* while (*after_access == 0)*/
1296: /* after_access++;*/
1297: after_access = p+1;
1.106 cvs 1298: if (!strcasecmp("URL", parts->access))
1.28 cvs 1299: /* Ignore IETF's URL: pre-prefix */
1300: parts->access = NULL;
1301: else
1.25 cvs 1302: break;
1303: }
1304: }
1305:
1306: p = after_access;
1.43 cvs 1307: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1308: {
1.43 cvs 1309: if (p[1] == URL_SEP)
1.28 cvs 1310: {
1.25 cvs 1311: parts->host = p+2; /* host has been specified */
1.28 cvs 1312: *p = 0; /* Terminate access */
1313: /* look for end of host name if any */
1.106 cvs 1314: p = strchr (parts->host, URL_SEP);
1.28 cvs 1315: if (p)
1316: {
1.106 cvs 1317: *p = EOS; /* Terminate host */
1.25 cvs 1318: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1319: }
1320: }
1321: else
1322: /* Root found but no host */
1323: parts->absolute = p+1;
1324: }
1325: else
1326: {
1.25 cvs 1327: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1328: }
1.25 cvs 1329: }
1330:
1331:
1332: /*----------------------------------------------------------------------
1.28 cvs 1333: AmayaParseUrl: parse a Name relative to another name
1334:
1335: This returns those parts of a name which are given (and requested)
1336: substituting bits from the related name where necessary.
1.25 cvs 1337:
1.28 cvs 1338: On entry,
1.25 cvs 1339: aName A filename given
1340: relatedName A name relative to which aName is to be parsed. Give
1341: it an empty string if aName is absolute.
1342: wanted A mask for the bits which are wanted.
1343:
1.28 cvs 1344: On exit,
1.25 cvs 1345: returns A pointer to a malloc'd string which MUST BE FREED
1346: ----------------------------------------------------------------------*/
1.106 cvs 1347: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1348: {
1349: char *return_value;
1350: char result[MAX_LENGTH];
1351: char name[MAX_LENGTH];
1352: char rel[MAX_LENGTH];
1353: char *p, *access;
1.29 cvs 1354: HTURI given, related;
1355: int len;
1.106 cvs 1356: char used_sep;
1357: char *used_str;
1.32 cvs 1358:
1.106 cvs 1359: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1360: {
1.106 cvs 1361: used_str = DIR_STR;
1362: used_sep = DIR_SEP;
1.33 cvs 1363: }
1.32 cvs 1364: else
1.33 cvs 1365: {
1.106 cvs 1366: used_str = URL_STR;
1367: used_sep = URL_SEP;
1.33 cvs 1368: }
1.32 cvs 1369:
1.29 cvs 1370: /* Make working copies of input strings to cut up: */
1371: return_value = NULL;
1372: result[0] = 0; /* Clear string */
1.106 cvs 1373: strcpy (name, aName);
1.29 cvs 1374: if (relatedName != NULL)
1.106 cvs 1375: strcpy (rel, relatedName);
1.29 cvs 1376: else
1.106 cvs 1377: relatedName[0] = EOS;
1.29 cvs 1378:
1379: scan (name, &given);
1380: scan (rel, &related);
1381: access = given.access ? given.access : related.access;
1382: if (wanted & AMAYA_PARSE_ACCESS)
1383: if (access)
1384: {
1.106 cvs 1385: strcat (result, access);
1.29 cvs 1386: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1387: strcat (result, ":");
1.29 cvs 1388: }
1389:
1390: if (given.access && related.access)
1391: /* If different, inherit nothing. */
1.106 cvs 1392: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1393: {
1394: related.host = 0;
1395: related.absolute = 0;
1396: related.relative = 0;
1397: related.fragment = 0;
1398: }
1399:
1400: if (wanted & AMAYA_PARSE_HOST)
1401: if(given.host || related.host)
1402: {
1403: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1404: strcat (result, "//");
1405: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1406: }
1407:
1408: if (given.host && related.host)
1409: /* If different hosts, inherit no path. */
1.106 cvs 1410: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1411: {
1412: related.absolute = 0;
1413: related.relative = 0;
1414: related.fragment = 0;
1415: }
1416:
1417: if (wanted & AMAYA_PARSE_PATH)
1418: {
1419: if (given.absolute)
1420: {
1421: /* All is given */
1422: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1423: strcat (result, used_str);
1424: strcat (result, given.absolute);
1.25 cvs 1425: }
1.29 cvs 1426: else if (related.absolute)
1427: {
1428: /* Adopt path not name */
1.106 cvs 1429: strcat (result, used_str);
1430: strcat (result, related.absolute);
1.29 cvs 1431: if (given.relative)
1432: {
1433: /* Search part? */
1.106 cvs 1434: p = strchr (result, '?');
1.29 cvs 1435: if (!p)
1.106 cvs 1436: p=result+strlen(result)-1;
1.33 cvs 1437: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1438: /* Remove filename */
1439: p[1]=0;
1440: /* Add given one */
1.106 cvs 1441: strcat (result, given.relative);
1.25 cvs 1442: }
1443: }
1.29 cvs 1444: else if (given.relative)
1445: /* what we've got */
1.106 cvs 1446: strcat (result, given.relative);
1.29 cvs 1447: else if (related.relative)
1.106 cvs 1448: strcat (result, related.relative);
1.29 cvs 1449: else
1450: /* No inheritance */
1.106 cvs 1451: strcat (result, used_str);
1.25 cvs 1452: }
1.29 cvs 1453:
1454: if (wanted & AMAYA_PARSE_ANCHOR)
1455: if (given.fragment || related.fragment)
1456: {
1457: if (given.absolute && given.fragment)
1458: {
1459: /*Fixes for relURLs...*/
1460: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1461: strcat (result, "#");
1462: strcat (result, given.fragment);
1.29 cvs 1463: }
1464: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1465: strcat (result, "");
1.29 cvs 1466: else
1467: {
1.110 cvs 1468: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1469: strcat (result, "#");
1.110 cvs 1470: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1471: }
1472: }
1.106 cvs 1473: len = strlen (result);
1474: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1475: strcpy (return_value, result);
1.29 cvs 1476: return (return_value); /* exactly the right length */
1.25 cvs 1477: }
1478:
1479: /*----------------------------------------------------------------------
1480: HTCanon
1481: Canonicalizes the URL in the following manner starting from the host
1482: pointer:
1483:
1484: 1) The host name is converted to lowercase
1485: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1486:
1487: Return: OK The position of the current path part of the URL
1488: which might be the old one or a new one.
1489:
1490: ----------------------------------------------------------------------*/
1.106 cvs 1491: static char *HTCanon (char **filename, char *host)
1492: {
1493: char *newname = NULL;
1494: char used_sep;
1495: char *path;
1496: char *strptr;
1497: char *port;
1498: char *access = host-3;
1499:
1500: if (*filename && strchr (*filename, URL_SEP))
1501: used_sep = URL_SEP;
1.33 cvs 1502: else
1.106 cvs 1503: used_sep = DIR_SEP;
1.32 cvs 1504:
1.110 cvs 1505: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1506: access--;
1.110 cvs 1507: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1508: path = host + strlen (host);
1509: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1510: host = strptr;
1.110 cvs 1511: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1512: port = NULL;
1.25 cvs 1513:
1514: strptr = host; /* Convert to lower-case */
1.82 cvs 1515: while (strptr < path)
1.33 cvs 1516: {
1.84 cvs 1517: *strptr = utolower (*strptr);
1.82 cvs 1518: strptr++;
1.33 cvs 1519: }
1.25 cvs 1520:
1521: /* Does the URL contain a full domain name? This also works for a
1522: numerical host name. The domain name is already made lower-case
1523: and without a trailing dot. */
1524: {
1.106 cvs 1525: char *dot = port ? port : path;
1526: if (dot > *filename && *--dot == '.')
1.33 cvs 1527: {
1.106 cvs 1528: char *orig = dot;
1529: char *dest = dot + 1;
1.82 cvs 1530: while ((*orig++ = *dest++));
1531: if (port) port--;
1.33 cvs 1532: path--;
1.25 cvs 1533: }
1534: }
1535: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1536: if (port)
1537: {
1.82 cvs 1538: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1539: {
1540: if (!newname)
1541: {
1.106 cvs 1542: char *orig = port;
1543: char *dest = port + 1;
1.82 cvs 1544: while ((*orig++ = *dest++));
1.33 cvs 1545: }
1546: }
1.106 cvs 1547: else if ((!strncmp (access, "http", 4) &&
1548: (*(port + 1) == '8' &&
1549: *(port+2) == '0' &&
1.82 cvs 1550: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1551: (!strncmp (access, "gopher", 6) &&
1552: (*(port+1) == '7' &&
1553: *(port+2) == '0' &&
1.82 cvs 1554: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1555: (!strncmp (access, "ftp", 3) &&
1556: (*(port+1) == '2' &&
1557: *(port + 2) == '1' &&
1.82 cvs 1558: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1559: if (!newname)
1560: {
1.106 cvs 1561: char *orig = port;
1562: char *dest = port + 3;
1.33 cvs 1563: while((*orig++ = *dest++));
1564: /* Update path position, Henry Minsky */
1565: path -= 3;
1.25 cvs 1566: }
1.33 cvs 1567: }
1568: else if (newname)
1.106 cvs 1569: strncat (newname, port, (int) (path - port));
1.33 cvs 1570: }
1.25 cvs 1571:
1.33 cvs 1572: if (newname)
1573: {
1.106 cvs 1574: char *newpath = newname + strlen (newname);
1575: strcat (newname, path);
1.25 cvs 1576: path = newpath;
1.28 cvs 1577: /* Free old copy */
1578: TtaFreeMemory(*filename);
1.25 cvs 1579: *filename = newname;
1.33 cvs 1580: }
1.25 cvs 1581: return path;
1582: }
1583:
1584:
1585: /*----------------------------------------------------------------------
1.29 cvs 1586: SimplifyUrl: simplify a URI
1.32 cvs 1587: A URI is allowed to contain the sequence xxx/../ which may be
1588: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1589: Simplification helps us recognize duplicate URIs.
1.25 cvs 1590:
1.28 cvs 1591: Thus, /etc/junk/../fred becomes /etc/fred
1592: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1593:
1.28 cvs 1594: but we should NOT change
1595: http://fred.xxx.edu/../..
1.25 cvs 1596:
1597: or ../../albert.html
1598:
1.28 cvs 1599: In order to avoid empty URLs the following URLs become:
1.25 cvs 1600:
1601: /fred/.. becomes /fred/..
1602: /fred/././.. becomes /fred/..
1603: /fred/.././junk/.././ becomes /fred/..
1604:
1.28 cvs 1605: If more than one set of `://' is found (several proxies in cascade) then
1606: only the part after the last `://' is simplified.
1.25 cvs 1607:
1.28 cvs 1608: Returns: A string which might be the old one or a new one.
1.25 cvs 1609: ----------------------------------------------------------------------*/
1.106 cvs 1610: void SimplifyUrl (char **url)
1611: {
1612: char *path;
1613: char *access;
1614: char *newptr;
1615: char *p;
1616: char *orig, *dest, *end;
1.28 cvs 1617:
1.106 cvs 1618: char used_sep;
1.77 cvs 1619: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1620: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1621:
1622:
1.28 cvs 1623: if (!url || !*url)
1624: return;
1625:
1.106 cvs 1626: if (strchr (*url, URL_SEP))
1627: used_sep = URL_SEP;
1.32 cvs 1628: else
1.106 cvs 1629: used_sep = DIR_SEP;
1.32 cvs 1630:
1.77 cvs 1631: /* should we simplify double dot? */
1632: path = *url;
1.106 cvs 1633: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1634: ddot_simplify = FALSE;
1635: else
1636: ddot_simplify = TRUE;
1637:
1.28 cvs 1638: /* Find any scheme name */
1.106 cvs 1639: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1640: {
1641: /* Find host name */
1.28 cvs 1642: access = *url;
1.84 cvs 1643: while (access < path && (*access = utolower (*access)))
1.82 cvs 1644: access++;
1.28 cvs 1645: path += 3;
1.106 cvs 1646: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1647: /* For proxies */
1.106 cvs 1648: path = newptr + 3;
1.82 cvs 1649: /* We have a host name */
1.84 cvs 1650: path = HTCanon (url, path);
1.25 cvs 1651: }
1.106 cvs 1652: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1653: path += 2;
1654: else
1655: path = *url;
1.25 cvs 1656:
1.84 cvs 1657: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1658: /* Some URLs start //<foo> */
1659: path += 1;
1.94 cvs 1660: else if (IsFilePath (path))
1661: {
1662: /* doesn't need to do anything more */
1663: return;
1664: }
1.106 cvs 1665: else if (!strncmp (path, "news:", 5))
1.28 cvs 1666: {
1.106 cvs 1667: newptr = strchr (path+5, '@');
1.28 cvs 1668: if (!newptr)
1669: newptr = path + 5;
1670: while (*newptr)
1671: {
1672: /* Make group or host lower case */
1.84 cvs 1673: *newptr = utolower (*newptr);
1.28 cvs 1674: newptr++;
1.25 cvs 1675: }
1.28 cvs 1676: /* Doesn't need to do any more */
1677: return;
1.25 cvs 1678: }
1.28 cvs 1679:
1680: if ((p = path))
1681: {
1.106 cvs 1682: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1683: (end = strchr (path, '#'))))
1684: end = path + strlen (path);
1.28 cvs 1685:
1686: /* Parse string second time to simplify */
1687: p = path;
1688: while (p < end)
1689: {
1.110 cvs 1690: /* if we're pointing to a char, it's safe to reactivate the
1691: ../ convertion */
1.106 cvs 1692: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1693: ddot_simplify = TRUE;
1694:
1.33 cvs 1695: if (*p==used_sep)
1.28 cvs 1696: {
1.106 cvs 1697: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1698: {
1699: orig = p + 1;
1.84 cvs 1700: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1701: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1702: end = orig - 1;
1703: }
1.106 cvs 1704: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1705: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1706: {
1707: newptr = p;
1.52 cvs 1708: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1709: if (*newptr == used_sep)
1710: orig = newptr + 1;
1.28 cvs 1711: else
1.52 cvs 1712: orig = newptr;
1713:
1714: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1715: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1716: end = orig-1;
1717: /* Start again with prev slash */
1718: p = newptr;
1.28 cvs 1719: }
1.33 cvs 1720: else if (*(p+1) == used_sep)
1.28 cvs 1721: {
1.33 cvs 1722: while (*(p+1) == used_sep)
1.28 cvs 1723: {
1724: orig = p;
1725: dest = p + 1;
1726: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1727: end = orig-1;
1728: }
1729: }
1730: else
1.25 cvs 1731: p++;
1.28 cvs 1732: }
1733: else
1734: p++;
1.25 cvs 1735: }
1736: }
1.51 cvs 1737:
1738: /*
1739: ** Check for host/../.. kind of things
1740: */
1.106 cvs 1741: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1742: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 1743: *(path+1) = EOS;
1.51 cvs 1744:
1.28 cvs 1745: return;
1746: }
1747:
1748:
1749: /*----------------------------------------------------------------------
1.96 cvs 1750: NormalizeFile normalizes local names.
1.28 cvs 1751: Return TRUE if target and src differ.
1752: ----------------------------------------------------------------------*/
1.106 cvs 1753: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 1754: {
1.110 cvs 1755: #ifndef _WINDOWS
1.106 cvs 1756: char *s;
1.93 cvs 1757: int i;
1.110 cvs 1758: #endif /* !_WINDOWS */
1.82 cvs 1759: ThotBool change;
1.90 cvs 1760: int start_index; /* the first char that we'll copy */
1.28 cvs 1761:
1.54 cvs 1762: change = FALSE;
1.90 cvs 1763: start_index = 0;
1764:
1.106 cvs 1765: if (!src || src[0] == EOS)
1.96 cvs 1766: {
1.106 cvs 1767: target[0] = EOS;
1.96 cvs 1768: return FALSE;
1769: }
1.90 cvs 1770:
1771: /* @@ do I need file: or file:/ here? */
1.106 cvs 1772: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1773: {
1.90 cvs 1774: /* remove the prefix file: */
1775: start_index += 5;
1776:
1777: /* remove the localhost prefix */
1.106 cvs 1778: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 1779: start_index += 11;
1780:
1781: /* remove the first two slashes in / / /path */
1782: while (src[start_index] &&
1.106 cvs 1783: src[start_index] == '/'
1784: && src[start_index + 1] == '/')
1.94 cvs 1785: start_index++;
1786:
1787: #ifdef _WINDOWS
1788: /* remove any extra slash before the drive name */
1.106 cvs 1789: if (src[start_index] == '/'
1790: &&src[start_index+2] == ':')
1.94 cvs 1791: start_index++;
1792: #endif /* _WINDOWS */
1.90 cvs 1793:
1.106 cvs 1794: if (src[start_index] == EOS)
1.90 cvs 1795: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 1796: strcpy (target, DIR_STR);
1.90 cvs 1797: else
1.97 cvs 1798: /* as we're inside a file: URL, we'll apply all the convertions
1799: we know */
1800: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1801:
1802: change = TRUE;
1803: }
1.97 cvs 1804: else if (convertion != AM_CONV_NONE)
1.96 cvs 1805: {
1806: /* we are following a "local" relative link, we do all the
1807: convertions except for the HOME_DIR ~ one */
1.97 cvs 1808: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1809: }
1.90 cvs 1810: #ifndef _WINDOWS
1.106 cvs 1811: else if (src[0] == '~')
1.53 cvs 1812: {
1.96 cvs 1813: /* it must be a URL typed in a text input field */
1814: /* do the HOME_DIR ~ substitution */
1.82 cvs 1815: s = TtaGetEnvString ("HOME");
1.106 cvs 1816: strcpy (target, s);
1.90 cvs 1817: #if 0
1.96 cvs 1818: /* JK: invalidated this part of the code as it's simpler
1819: to add the DIR_SEP whenever we have something to add
1820: to the path rather than adding it systematically */
1.106 cvs 1821: if (src[1] != DIR_SEP)
1822: strcat (target, DIR_STR);
1.90 cvs 1823: #endif
1.106 cvs 1824: i = strlen (target);
1825: strcpy (&target[i], &src[1]);
1.54 cvs 1826: change = TRUE;
1.53 cvs 1827: }
1.90 cvs 1828: #endif /* _WINDOWS */
1.28 cvs 1829: else
1.96 cvs 1830: /* leave it as it is */
1.106 cvs 1831: strcpy (target, src);
1.96 cvs 1832:
1.28 cvs 1833: /* remove /../ and /./ */
1.29 cvs 1834: SimplifyUrl (&target);
1.54 cvs 1835: if (!change)
1.106 cvs 1836: change = strcmp (src, target);
1.28 cvs 1837: return (change);
1.25 cvs 1838: }
1839:
1.28 cvs 1840:
1.25 cvs 1841: /*----------------------------------------------------------------------
1.31 cvs 1842: MakeRelativeURL: make relative name
1.25 cvs 1843:
1.28 cvs 1844: This function creates and returns a string which gives an expression of
1845: one address as related to another. Where there is no relation, an absolute
1846: address is retured.
1.25 cvs 1847:
1.28 cvs 1848: On entry,
1.25 cvs 1849: Both names must be absolute, fully qualified names of nodes
1850: (no fragment bits)
1851:
1.28 cvs 1852: On exit,
1.25 cvs 1853: The return result points to a newly allocated name which, if
1854: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1855: The caller is responsible for freeing the resulting name later.
1856: ----------------------------------------------------------------------*/
1.106 cvs 1857: char *MakeRelativeURL (char *aName, char *relatedName)
1858: {
1859: char *return_value;
1860: char result[MAX_LENGTH];
1861: char *p;
1862: char *q;
1863: char *after_access;
1864: char *last_slash = NULL;
1865: int slashes, levels, len;
1.110 cvs 1866: #ifdef _WINDOWS
1.44 cvs 1867: int ndx;
1.110 cvs 1868: #endif /* _WINDOWS */
1.44 cvs 1869:
1.29 cvs 1870: if (aName == NULL || relatedName == NULL)
1871: return (NULL);
1872:
1873: slashes = 0;
1874: after_access = NULL;
1875: p = aName;
1876: q = relatedName;
1877: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1878: {
1879: /* Find extent of match */
1.106 cvs 1880: if (*p == ':')
1.29 cvs 1881: after_access = p + 1;
1.28 cvs 1882: if (*p == DIR_SEP)
1.27 cvs 1883: {
1.29 cvs 1884: /* memorize the last slash position and count them */
1.27 cvs 1885: last_slash = p;
1886: slashes++;
1.25 cvs 1887: }
1888: }
1889:
1.31 cvs 1890: /* q, p point to the first non-matching character or zero */
1.106 cvs 1891: if (*q == EOS)
1.31 cvs 1892: {
1893: /* New name is a subset of the related name */
1894: /* exactly the right length */
1.106 cvs 1895: len = strlen (p);
1896: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1897: strcpy (return_value, p);
1.31 cvs 1898: }
1899: else if ((slashes < 2 && after_access == NULL)
1900: || (slashes < 3 && after_access != NULL))
1901: {
1902: /* Two names whitout common path */
1903: /* exactly the right length */
1.106 cvs 1904: len = strlen (aName);
1905: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1906: strcpy (return_value, aName);
1.31 cvs 1907: }
1908: else
1909: {
1910: /* Some path in common */
1.106 cvs 1911: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 1912: /* just the same server */
1.106 cvs 1913: strcpy (result, last_slash);
1.31 cvs 1914: else
1915: {
1916: levels= 0;
1.106 cvs 1917: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 1918: if (*q == DIR_SEP)
1919: levels++;
1920:
1.106 cvs 1921: result[0] = EOS;
1.31 cvs 1922: for (;levels; levels--)
1.106 cvs 1923: strcat (result, "../");
1924: strcat (result, last_slash+1);
1.31 cvs 1925: }
1.52 cvs 1926:
1927: if (!*result)
1.106 cvs 1928: strcat (result, "./");
1.52 cvs 1929:
1.31 cvs 1930: /* exactly the right length */
1.106 cvs 1931: len = strlen (result);
1932: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1933: strcpy (return_value, result);
1.52 cvs 1934:
1.25 cvs 1935: }
1.110 cvs 1936: #ifdef _WINDOWS
1.106 cvs 1937: len = strlen (return_value);
1.44 cvs 1938: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 1939: if (return_value[ndx] == '\\')
1940: return_value[ndx] = '/' ;
1.110 cvs 1941: #endif /* _WINDOWS */
1.29 cvs 1942: return (return_value);
1.24 cvs 1943: }
1.35 cvs 1944:
1.104 kahan 1945: /*----------------------------------------------------------------------
1946: AM_GetFileSize
1947: Returns TRUE and the filesize in the 2nd parameter.
1948: Otherwise, in case of a system error, returns FALSE, with a
1949: filesize of 0L.
1950: ---------------------------------------------------------------------*/
1.106 cvs 1951: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 1952: {
1.106 cvs 1953: ThotFileHandle handle = ThotFile_BADHANDLE;
1954: ThotFileInfo info;
1.35 cvs 1955:
1.104 kahan 1956: *file_size = 0L;
1957: if (!TtaFileExist (filename))
1958: return FALSE;
1959:
1960: handle = TtaFileOpen (filename, ThotFile_READWRITE);
1961: if (handle == ThotFile_BADHANDLE)
1962: /* ThotFile_BADHANDLE */
1963: return FALSE;
1964: if (TtaFileStat (handle, &info) == 0)
1965: /* bad stat */
1966: info.size = 0L;
1967: TtaFileClose (handle);
1968: *file_size = (unsigned long) info.size;
1969: return TRUE;
1970: }
Webmaster