Annotation of Amaya/amaya/AHTURLTools.c, revision 1.109
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: ----------------------------------------------------------------------*/
1.109 ! 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: ----------------------------------------------------------------------*/
1.106 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: ----------------------------------------------------------------------*/
1.106 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: ----------------------------------------------------------------------*/
1.106 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: ----------------------------------------------------------------------*/
1.106 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: ----------------------------------------------------------------------*/
1.106 cvs 497: ThotBool IsImageName (const char *path)
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: ----------------------------------------------------------------------*/
1.106 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: }
1.106 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: ----------------------------------------------------------------------*/
1.106 cvs 555: ThotBool IsTextName (const char *path)
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:
1.106 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;
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])
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;
706: /* get the root element */
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--;
1.106 cvs 729: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 730: /* verify if the base has the form "protocol://server:port" */
731: {
1.106 cvs 732: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1.33 cvs 733: AMAYA_PARSE_HOST |
734: AMAYA_PARSE_PUNCTUATION);
1.106 cvs 735: if (ptr && !strcmp (ptr, basename))
1.31 cvs 736: {
1.43 cvs 737: /* it has this form, we complete it by adding a URL_STR */
1.106 cvs 738: if (strchr (basename, DIR_SEP))
739: strcat (basename, DIR_STR);
1.43 cvs 740: else
1.106 cvs 741: strcat (basename, URL_STR);
1.31 cvs 742: length++;
743: }
744: if (ptr)
745: TtaFreeMemory (ptr);
746: }
747: }
1.33 cvs 748: }
749:
1.31 cvs 750: /* Remove anything after the last DIR_SEP char. If no such char is found,
751: * then search for the first ":" char, hoping that what's before that is a
752: * protocol. If found, end the string there. If neither char is found,
753: * then discard the whole base element.
754: */
1.106 cvs 755: length = strlen (basename) - 1;
1.31 cvs 756: /* search for the last DIR_SEP char */
1.106 cvs 757: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 758: length--;
759: if (length >= 0)
760: /* found the last DIR_SEP char, end the string there */
1.106 cvs 761: basename[length + 1] = EOS;
1.31 cvs 762: else
763: /* search for the first PATH_STR char */
764: {
1.106 cvs 765: for (length = 0; basename[length] != ':' &&
766: basename[length] != EOS; length ++);
767: if (basename[length] == ':')
1.31 cvs 768: /* found, so end the string there */
1.106 cvs 769: basename[length + 1] = EOS;
1.31 cvs 770: else
771: /* not found, discard the base */
1.106 cvs 772: basename[0] = EOS;
1.31 cvs 773: }
774: return (basename);
775: }
776:
777:
1.4 cvs 778: /*----------------------------------------------------------------------
1.40 cvs 779: GetLocalPath
780: Allocate and return the local document path associated to the url
781: ----------------------------------------------------------------------*/
1.106 cvs 782: char *GetLocalPath (Document doc, char *url)
783: {
784: char *ptr;
785: char *n;
786: char *documentname;
787: char url_sep;
1.83 cvs 788: int len;
1.67 cvs 789: ThotBool noFile;
1.40 cvs 790:
791: if (url != NULL)
792: {
793: /* check whether the file name exists */
1.106 cvs 794: len = strlen (url) - 1;
1.71 cvs 795: if (IsW3Path (url))
1.106 cvs 796: url_sep = '/';
1.41 cvs 797: else
1.106 cvs 798: url_sep = DIR_SEP;
1.41 cvs 799: noFile = (url[len] == url_sep);
1.40 cvs 800: if (noFile)
1.106 cvs 801: url[len] = EOS;
802: ptr = TtaGetMemory (MAX_LENGTH);
803: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 804: TtaExtractName (url, ptr, documentname);
1.106 cvs 805: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 806: if (!TtaCheckDirectory (ptr))
807: /* directory did not exist */
1.72 cvs 808: TtaMakeDirectory (ptr);
1.47 cvs 809:
810: /* don't include the query string within document name */
1.106 cvs 811: n = strrchr (documentname, '?');
1.47 cvs 812: if (n != NULL)
1.106 cvs 813: *n = EOS;
1.46 cvs 814: /* don't include ':' within document name */
1.106 cvs 815: n = strchr (documentname, ':');
1.46 cvs 816: if (n != NULL)
1.106 cvs 817: *n = EOS;
1.69 cvs 818: /* if after all this operations document name
819: is empty, let's use noname.html instead */
1.106 cvs 820: if (documentname[0] == EOS)
821: strcat (ptr, "noname.html");
1.69 cvs 822: else
1.106 cvs 823: strcat (ptr, documentname);
1.40 cvs 824: TtaFreeMemory (documentname);
825: /* restore the url */
826: if (noFile)
1.41 cvs 827: url[len] = url_sep;
1.40 cvs 828: return (ptr);
829: }
830: else
831: return (NULL);
832: }
833:
1.73 cvs 834: /*----------------------------------------------------------------------
1.79 cvs 835: ExtractTarget extract the target name from document nane.
836: ----------------------------------------------------------------------*/
1.106 cvs 837: void ExtractTarget (char *aName, char *target)
1.79 cvs 838: {
1.106 cvs 839: int lg, i;
840: char *ptr;
841: char *oldptr;
1.79 cvs 842:
843: if (!target || !aName)
844: /* bad target */
845: return;
846:
1.106 cvs 847: target[0] = EOS;
848: lg = strlen (aName);
1.79 cvs 849: if (lg)
850: {
851: /* the name is not empty */
852: oldptr = ptr = &aName[0];
853: do
854: {
1.106 cvs 855: ptr = strrchr (oldptr, '#');
1.79 cvs 856: if (ptr)
857: oldptr = &ptr[1];
858: }
859: while (ptr);
860:
861: i = (int) (oldptr) - (int) (aName); /* name length */
862: if (i > 1)
863: {
1.106 cvs 864: aName[i - 1] = EOS;
1.79 cvs 865: if (i != lg)
1.106 cvs 866: strcpy (target, oldptr);
1.79 cvs 867: }
868: }
869: }
870:
871: /*----------------------------------------------------------------------
1.90 cvs 872: RemoveNewLines (text)
873: Removes any '\n' chars that are found in text.
874: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 875: ----------------------------------------------------------------------*/
1.106 cvs 876: ThotBool RemoveNewLines (char *text)
877: {
878: ThotBool change = FALSE;
879: char *src;
880: char *dest;
1.90 cvs 881:
882: src = text;
883: dest = text;
884: while (*src)
885: {
886: switch (*src)
887: {
1.106 cvs 888: case '\n':
1.90 cvs 889: /* don't copy the newline */
890: change = 1;
891: break;
892: default:
893: *dest = *src;
894: dest++;
895: break;
896: }
897: src++;
898: }
899: /* copy the last EOS char */
900: *dest = *src;
901:
902: return (change);
903: }
904:
905: /*----------------------------------------------------------------------
906: CleanCopyFileURL
907: Copies a file url from a src string to destination string.
1.97 cvs 908: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
909: we want to do).
1.90 cvs 910: ----------------------------------------------------------------------*/
1.106 cvs 911: static void CleanCopyFileURL (char *dest, char *src,
912: ConvertionType convertion)
1.90 cvs 913: {
914: while (*src)
1.89 cvs 915: {
1.90 cvs 916: switch (*src)
1.89 cvs 917: {
918: #ifdef _WINDOWS
1.106 cvs 919: case URL_SEP:
1.96 cvs 920: /* make DIR_SEP transformation */
1.97 cvs 921: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 922: *dest = DIR_SEP;
1.96 cvs 923: else
924: *dest = *src;
1.90 cvs 925: dest++;
1.96 cvs 926: src++;
1.90 cvs 927: break;
1.89 cvs 928: #endif /* _WINDOWS */
1.96 cvs 929:
1.106 cvs 930: case '%':
1.97 cvs 931: if (convertion & AM_CONV_PERCENT)
1.96 cvs 932: {
1.97 cvs 933: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 934: src++;
1.106 cvs 935: if (*src != EOS)
1.97 cvs 936: {
937: *dest = UnEscapeChar (*src) * 16;
938: src++;
939: }
1.106 cvs 940: if (*src != EOS)
1.97 cvs 941: {
942: *dest = *dest + UnEscapeChar (*src);
943: src++;
944: }
945: dest++;
1.96 cvs 946: }
1.97 cvs 947: else
1.96 cvs 948: {
1.97 cvs 949: *dest = *src;
950: dest++;
1.96 cvs 951: src++;
952: }
953: break;
954:
1.90 cvs 955: default:
956: *dest = *src;
1.89 cvs 957: dest++;
1.96 cvs 958: src++;
1.90 cvs 959: break;
1.89 cvs 960: }
961: }
1.90 cvs 962: /* copy the EOS char */
963: *dest = *src;
1.73 cvs 964: }
1.40 cvs 965:
966: /*----------------------------------------------------------------------
1.9 cvs 967: NormalizeURL
968: normalizes orgName according to a base associated with doc, and
969: following the standard URL format rules.
1.53 cvs 970: if doc is 0 and otherPath not NULL, normalizes orgName according to this
971: other path.
1.9 cvs 972: The function returns the new complete and normalized URL
1.12 cvs 973: or file name path (newName) and the name of the document (docName).
1.9 cvs 974: N.B. If the function can't find out what's the docName, it assigns
975: the name "noname.html".
1.4 cvs 976: ----------------------------------------------------------------------*/
1.106 cvs 977: void NormalizeURL (char *orgName, Document doc, char *newName,
978: char *docName, char *otherPath)
979: {
980: char *basename;
981: char tempOrgName[MAX_LENGTH];
982: char *ptr;
983: char used_sep;
1.84 cvs 984: int length;
985: ThotBool check;
1.5 cvs 986:
1.44 cvs 987: # ifdef _WINDOWS
988: int ndx;
989: # endif /* _WINDOWS */
990:
1.5 cvs 991: if (!newName || !docName)
992: return;
1.18 cvs 993:
1.32 cvs 994: if (doc != 0)
1.53 cvs 995: basename = GetBaseURL (doc);
996: else if (otherPath != NULL)
1.108 cvs 997: basename = TtaStrdup (otherPath);
1.32 cvs 998: else
1.53 cvs 999: basename = NULL;
1.32 cvs 1000:
1.18 cvs 1001: /*
1.31 cvs 1002: * Clean orgName
1003: * Make sure we have a complete orgName, without any leading or trailing
1004: * white spaces, or trailinbg new lines
1005: */
1.5 cvs 1006: ptr = orgName;
1.18 cvs 1007: /* skip leading white space and new line characters */
1.106 cvs 1008: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1009: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1010: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1011: /*
1.31 cvs 1012: * Make orgName a complete URL
1013: * If the URL does not include a protocol, then try to calculate
1014: * one using the doc's base element (if it exists),
1015: */
1.106 cvs 1016: if (tempOrgName[0] == EOS)
1.53 cvs 1017: {
1.106 cvs 1018: newName[0] = EOS;
1019: docName[0] = EOS;
1.53 cvs 1020: TtaFreeMemory (basename);
1021: return;
1022: }
1.49 cvs 1023:
1024: /* clean trailing white space */
1.106 cvs 1025: length = strlen (tempOrgName) - 1;
1026: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1027: {
1.106 cvs 1028: tempOrgName[length] = EOS;
1.53 cvs 1029: length--;
1030: }
1.50 cvs 1031:
1.55 cvs 1032: /* remove extra dot (which dot???) */
1033: /* ugly, but faster than a strcmp */
1.106 cvs 1034: if (tempOrgName[length] == '.'
1035: && (length == 0 || tempOrgName[length-1] != '.'))
1036: tempOrgName[length] = EOS;
1.50 cvs 1037:
1.94 cvs 1038: if (IsW3Path (tempOrgName))
1.53 cvs 1039: {
1040: /* the name is complete, go to the Sixth Step */
1.106 cvs 1041: strcpy (newName, tempOrgName);
1.53 cvs 1042: SimplifyUrl (&newName);
1043: /* verify if the URL has the form "protocol://server:port" */
1.106 cvs 1044: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1045: if (ptr && !strcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */
1046: strcat (newName, URL_STR);
1.49 cvs 1047:
1.53 cvs 1048: if (ptr)
1.50 cvs 1049: TtaFreeMemory (ptr);
1.53 cvs 1050: }
1051: else if ( basename == NULL)
1052: /* the name is complete, go to the Sixth Step */
1.106 cvs 1053: strcpy (newName, tempOrgName);
1.53 cvs 1054: else
1055: {
1.31 cvs 1056: /* Calculate the absolute URL, using the base or document URL */
1.44 cvs 1057: # ifdef _WINDOWS
1.53 cvs 1058: if (!IsW3Path (basename))
1059: {
1.106 cvs 1060: length = strlen (tempOrgName);
1.53 cvs 1061: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1062: if (tempOrgName [ndx] == '/')
1063: tempOrgName [ndx] = '\\';
1.53 cvs 1064: }
1.44 cvs 1065: # endif /* _WINDOWS */
1.25 cvs 1066: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1067: if (ptr)
1068: {
1069: SimplifyUrl (&ptr);
1.106 cvs 1070: strcpy (newName, ptr);
1.53 cvs 1071: TtaFreeMemory (ptr);
1072: }
1073: else
1.106 cvs 1074: newName[0] = EOS;
1.53 cvs 1075: }
1.36 cvs 1076:
1077: TtaFreeMemory (basename);
1.18 cvs 1078: /*
1.31 cvs 1079: * Prepare the docname that will refer to this ressource in the
1080: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1081: * noname.html as a default ressource name
1.18 cvs 1082: */
1.106 cvs 1083: if (newName[0] != EOS)
1.53 cvs 1084: {
1.106 cvs 1085: length = strlen (newName) - 1;
1086: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1087: {
1088: used_sep = newName[length];
1089: check = TRUE;
1090: while (check)
1091: {
1.50 cvs 1092: length--;
1093: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1094: length--;
1.106 cvs 1095: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1096: {
1.106 cvs 1097: newName[length+1] = EOS;
1.53 cvs 1098: /* remove also previous directory */
1099: length--;
1100: while (length >= 0 && newName[length] != used_sep)
1101: length--;
1.106 cvs 1102: if (strncmp (&newName[length+1], "//", 2))
1.53 cvs 1103: /* don't remove server name */
1.106 cvs 1104: newName[length+1] = EOS;
1.53 cvs 1105: }
1.106 cvs 1106: else if (!strncmp (&newName[length+1], ".", 1))
1107: newName[length+1] = EOS;
1.50 cvs 1108: else
1.53 cvs 1109: check = FALSE;
1110: }
1.106 cvs 1111: strcpy (docName, "noname.html");
1.53 cvs 1112: /* docname was not comprised inside the URL, so let's */
1113: /* assign the default ressource name */
1.106 cvs 1114: strcpy (docName, "noname.html");
1.53 cvs 1115: }
1116: else
1117: { /* docname is comprised inside the URL */
1.106 cvs 1118: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP)
1.53 cvs 1119: length--;
1120: if (length < 0)
1.106 cvs 1121: strcpy (docName, newName);
1.53 cvs 1122: else
1.106 cvs 1123: strcpy (docName, &newName[length+1]);
1.53 cvs 1124: }
1125: }
1126: else
1.106 cvs 1127: docName[0] = EOS;
1.18 cvs 1128: }
1.3 cvs 1129:
1.4 cvs 1130: /*----------------------------------------------------------------------
1.9 cvs 1131: IsSameHost
1.4 cvs 1132: ----------------------------------------------------------------------*/
1.106 cvs 1133: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1134: {
1.106 cvs 1135: char *basename_ptr1, *basename_ptr2;
1136: ThotBool result;
1.3 cvs 1137:
1.106 cvs 1138: basename_ptr1 = AmayaParseUrl (url1, "",
1139: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1140: basename_ptr2 = AmayaParseUrl (url2, "",
1141: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1142:
1.106 cvs 1143: if (strcmp (basename_ptr1, basename_ptr2))
1144: result = FALSE;
1145: else
1146: result = TRUE;
1147: TtaFreeMemory (basename_ptr1);
1148: TtaFreeMemory (basename_ptr2);
1149: return (result);
1.3 cvs 1150: }
1151:
1152:
1.4 cvs 1153: /*----------------------------------------------------------------------
1.22 cvs 1154: HasKnownFileSuffix
1155: returns TRUE if path points to a file ending with a suffix.
1156: ----------------------------------------------------------------------*/
1.106 cvs 1157: ThotBool HasKnownFileSuffix (const char *path)
1158: {
1159: char *root;
1160: char temppath[MAX_LENGTH];
1161: char suffix[MAX_LENGTH];
1.22 cvs 1162:
1.106 cvs 1163: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1164: return (FALSE);
1165:
1.106 cvs 1166: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1167:
1168: if (root)
1169: {
1.106 cvs 1170: strcpy (temppath, root);
1.25 cvs 1171: TtaFreeMemory (root);
1.22 cvs 1172: /* Get the suffix */
1173: ExtractSuffix (temppath, suffix);
1174:
1.106 cvs 1175: if( suffix[0] == EOS)
1.22 cvs 1176: /* no suffix */
1177: return (FALSE);
1178:
1179: /* Normalize the suffix */
1180: ConvertToLowerCase (suffix);
1181:
1.106 cvs 1182: if (!strcmp (suffix, "gz"))
1.22 cvs 1183: /* skip the compressed suffix */
1184: {
1185: ExtractSuffix (temppath, suffix);
1.106 cvs 1186: if(suffix[0] == EOS)
1.22 cvs 1187: /* no suffix */
1188: return (FALSE);
1189: /* Normalize the suffix */
1190: ConvertToLowerCase (suffix);
1191: }
1192:
1.106 cvs 1193: if (strcmp (suffix, "gif") &&
1194: strcmp (suffix, "xbm") &&
1195: strcmp (suffix, "xpm") &&
1196: strcmp (suffix, "jpg") &&
1197: strcmp (suffix, "pdf") &&
1198: strcmp (suffix, "png") &&
1199: strcmp (suffix, "tgz") &&
1200: strcmp (suffix, "xpg") &&
1201: strcmp (suffix, "xpd") &&
1202: strcmp (suffix, "ps") &&
1203: strcmp (suffix, "au") &&
1204: strcmp (suffix, "html") &&
1205: strcmp (suffix, "htm") &&
1206: strcmp (suffix, "shtml") &&
1207: strcmp (suffix, "xht") &&
1208: strcmp (suffix, "xhtm") &&
1209: strcmp (suffix, "xhtml") &&
1210: strcmp (suffix, "txt") &&
1211: strcmp (suffix, "css") &&
1212: strcmp (suffix, "eps"))
1.22 cvs 1213: return (FALSE);
1214: else
1215: return (TRUE);
1216: }
1217: else
1218: return (FALSE);
1219: }
1220:
1221:
1222: /*----------------------------------------------------------------------
1.24 cvs 1223: ChopURL
1224: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1225: If inputURL is bigger than that size, outputURL receives
1226: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1227: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1228: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1229: copied into outputURL.
1230: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1231: chars.
1232: ----------------------------------------------------------------------*/
1.106 cvs 1233: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1234: {
1235: int len;
1.9 cvs 1236:
1.106 cvs 1237: len = strlen (inputURL);
1.24 cvs 1238: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1239: strcpy (outputURL, inputURL);
1.24 cvs 1240: else
1241: /* make a truncated urlName on the status window */
1242: {
1.106 cvs 1243: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1244: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1245: strcat (outputURL, "...");
1246: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1247: }
1.25 cvs 1248: }
1249:
1250:
1251: /*----------------------------------------------------------------------
1252: scan
1.47 cvs 1253: Scan a filename for its constituents
1.25 cvs 1254: -----------------------------------
1255:
1256: On entry,
1257: name points to a document name which may be incomplete.
1258: On exit,
1259: absolute or relative may be nonzero (but not both).
1260: host, fragment and access may be nonzero if they were specified.
1261: Any which are nonzero point to zero terminated strings.
1262: ----------------------------------------------------------------------*/
1.106 cvs 1263: static void scan (char *name, HTURI *parts)
1.25 cvs 1264: {
1.106 cvs 1265: char * p;
1266: char * after_access = name;
1.32 cvs 1267:
1.43 cvs 1268: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1269: /* Look for fragment identifier */
1.106 cvs 1270: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1271: {
1.106 cvs 1272: *p++ = '\0';
1.28 cvs 1273: parts->fragment = p;
1.25 cvs 1274: }
1275:
1.28 cvs 1276: for (p=name; *p; p++)
1277: {
1.106 cvs 1278: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1279: break;
1.106 cvs 1280: if (*p == ':')
1.28 cvs 1281: {
1282: *p = 0;
1283: parts->access = after_access; /* Scheme has been specified */
1284:
1285: /* The combination of gcc, the "-O" flag and the HP platform is
1286: unhealthy. The following three lines is a quick & dirty fix, but is
1287: not recommended. Rather, turn off "-O". */
1288:
1289: /* after_access = p;*/
1290: /* while (*after_access == 0)*/
1291: /* after_access++;*/
1292: after_access = p+1;
1.106 cvs 1293: if (!strcasecmp("URL", parts->access))
1.28 cvs 1294: /* Ignore IETF's URL: pre-prefix */
1295: parts->access = NULL;
1296: else
1.25 cvs 1297: break;
1298: }
1299: }
1300:
1301: p = after_access;
1.43 cvs 1302: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1303: {
1.43 cvs 1304: if (p[1] == URL_SEP)
1.28 cvs 1305: {
1.25 cvs 1306: parts->host = p+2; /* host has been specified */
1.28 cvs 1307: *p = 0; /* Terminate access */
1308: /* look for end of host name if any */
1.106 cvs 1309: p = strchr (parts->host, URL_SEP);
1.28 cvs 1310: if (p)
1311: {
1.106 cvs 1312: *p = EOS; /* Terminate host */
1.25 cvs 1313: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1314: }
1315: }
1316: else
1317: /* Root found but no host */
1318: parts->absolute = p+1;
1319: }
1320: else
1321: {
1.25 cvs 1322: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1323: }
1.25 cvs 1324: }
1325:
1326:
1327: /*----------------------------------------------------------------------
1.28 cvs 1328: AmayaParseUrl: parse a Name relative to another name
1329:
1330: This returns those parts of a name which are given (and requested)
1331: substituting bits from the related name where necessary.
1.25 cvs 1332:
1.28 cvs 1333: On entry,
1.25 cvs 1334: aName A filename given
1335: relatedName A name relative to which aName is to be parsed. Give
1336: it an empty string if aName is absolute.
1337: wanted A mask for the bits which are wanted.
1338:
1.28 cvs 1339: On exit,
1.25 cvs 1340: returns A pointer to a malloc'd string which MUST BE FREED
1341: ----------------------------------------------------------------------*/
1.106 cvs 1342: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1343: {
1344: char *return_value;
1345: char result[MAX_LENGTH];
1346: char name[MAX_LENGTH];
1347: char rel[MAX_LENGTH];
1348: char *p, *access;
1.29 cvs 1349: HTURI given, related;
1350: int len;
1.106 cvs 1351: char used_sep;
1352: char *used_str;
1.32 cvs 1353:
1.106 cvs 1354: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1355: {
1.106 cvs 1356: used_str = DIR_STR;
1357: used_sep = DIR_SEP;
1.33 cvs 1358: }
1.32 cvs 1359: else
1.33 cvs 1360: {
1.106 cvs 1361: used_str = URL_STR;
1362: used_sep = URL_SEP;
1.33 cvs 1363: }
1.32 cvs 1364:
1.29 cvs 1365: /* Make working copies of input strings to cut up: */
1366: return_value = NULL;
1367: result[0] = 0; /* Clear string */
1.106 cvs 1368: strcpy (name, aName);
1.29 cvs 1369: if (relatedName != NULL)
1.106 cvs 1370: strcpy (rel, relatedName);
1.29 cvs 1371: else
1.106 cvs 1372: relatedName[0] = EOS;
1.29 cvs 1373:
1374: scan (name, &given);
1375: scan (rel, &related);
1376: access = given.access ? given.access : related.access;
1377: if (wanted & AMAYA_PARSE_ACCESS)
1378: if (access)
1379: {
1.106 cvs 1380: strcat (result, access);
1.29 cvs 1381: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1382: strcat (result, ":");
1.29 cvs 1383: }
1384:
1385: if (given.access && related.access)
1386: /* If different, inherit nothing. */
1.106 cvs 1387: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1388: {
1389: related.host = 0;
1390: related.absolute = 0;
1391: related.relative = 0;
1392: related.fragment = 0;
1393: }
1394:
1395: if (wanted & AMAYA_PARSE_HOST)
1396: if(given.host || related.host)
1397: {
1398: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1399: strcat (result, "//");
1400: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1401: }
1402:
1403: if (given.host && related.host)
1404: /* If different hosts, inherit no path. */
1.106 cvs 1405: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1406: {
1407: related.absolute = 0;
1408: related.relative = 0;
1409: related.fragment = 0;
1410: }
1411:
1412: if (wanted & AMAYA_PARSE_PATH)
1413: {
1414: if (given.absolute)
1415: {
1416: /* All is given */
1417: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1418: strcat (result, used_str);
1419: strcat (result, given.absolute);
1.25 cvs 1420: }
1.29 cvs 1421: else if (related.absolute)
1422: {
1423: /* Adopt path not name */
1.106 cvs 1424: strcat (result, used_str);
1425: strcat (result, related.absolute);
1.29 cvs 1426: if (given.relative)
1427: {
1428: /* Search part? */
1.106 cvs 1429: p = strchr (result, '?');
1.29 cvs 1430: if (!p)
1.106 cvs 1431: p=result+strlen(result)-1;
1.33 cvs 1432: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1433: /* Remove filename */
1434: p[1]=0;
1435: /* Add given one */
1.106 cvs 1436: strcat (result, given.relative);
1.25 cvs 1437: }
1438: }
1.29 cvs 1439: else if (given.relative)
1440: /* what we've got */
1.106 cvs 1441: strcat (result, given.relative);
1.29 cvs 1442: else if (related.relative)
1.106 cvs 1443: strcat (result, related.relative);
1.29 cvs 1444: else
1445: /* No inheritance */
1.106 cvs 1446: strcat (result, used_str);
1.25 cvs 1447: }
1.29 cvs 1448:
1449: if (wanted & AMAYA_PARSE_ANCHOR)
1450: if (given.fragment || related.fragment)
1451: {
1452: if (given.absolute && given.fragment)
1453: {
1454: /*Fixes for relURLs...*/
1455: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1456: strcat (result, "#");
1457: strcat (result, given.fragment);
1.29 cvs 1458: }
1459: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1460: strcat (result, "");
1.29 cvs 1461: else
1462: {
1463: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1464: strcat (result, "#");
1465: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1466: }
1467: }
1.106 cvs 1468: len = strlen (result);
1469: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1470: strcpy (return_value, result);
1.29 cvs 1471: return (return_value); /* exactly the right length */
1.25 cvs 1472: }
1473:
1474: /*----------------------------------------------------------------------
1475: HTCanon
1476: Canonicalizes the URL in the following manner starting from the host
1477: pointer:
1478:
1479: 1) The host name is converted to lowercase
1480: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1481:
1482: Return: OK The position of the current path part of the URL
1483: which might be the old one or a new one.
1484:
1485: ----------------------------------------------------------------------*/
1.106 cvs 1486: static char *HTCanon (char **filename, char *host)
1487: {
1488: char *newname = NULL;
1489: char used_sep;
1490: char *path;
1491: char *strptr;
1492: char *port;
1493: char *access = host-3;
1494:
1495: if (*filename && strchr (*filename, URL_SEP))
1496: used_sep = URL_SEP;
1.33 cvs 1497: else
1.106 cvs 1498: used_sep = DIR_SEP;
1.32 cvs 1499:
1.82 cvs 1500: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1501: access--;
1.106 cvs 1502: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1503: path = host + strlen (host);
1504: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1505: host = strptr;
1.106 cvs 1506: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1507: port = NULL;
1.25 cvs 1508:
1509: strptr = host; /* Convert to lower-case */
1.82 cvs 1510: while (strptr < path)
1.33 cvs 1511: {
1.84 cvs 1512: *strptr = utolower (*strptr);
1.82 cvs 1513: strptr++;
1.33 cvs 1514: }
1.25 cvs 1515:
1516: /* Does the URL contain a full domain name? This also works for a
1517: numerical host name. The domain name is already made lower-case
1518: and without a trailing dot. */
1519: {
1.106 cvs 1520: char *dot = port ? port : path;
1521: if (dot > *filename && *--dot == '.')
1.33 cvs 1522: {
1.106 cvs 1523: char *orig = dot;
1524: char *dest = dot + 1;
1.82 cvs 1525: while ((*orig++ = *dest++));
1526: if (port) port--;
1.33 cvs 1527: path--;
1.25 cvs 1528: }
1529: }
1530: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1531: if (port)
1532: {
1.82 cvs 1533: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1534: {
1535: if (!newname)
1536: {
1.106 cvs 1537: char *orig = port;
1538: char *dest = port + 1;
1.82 cvs 1539: while ((*orig++ = *dest++));
1.33 cvs 1540: }
1541: }
1.106 cvs 1542: else if ((!strncmp (access, "http", 4) &&
1543: (*(port + 1) == '8' &&
1544: *(port+2) == '0' &&
1.82 cvs 1545: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1546: (!strncmp (access, "gopher", 6) &&
1547: (*(port+1) == '7' &&
1548: *(port+2) == '0' &&
1.82 cvs 1549: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1550: (!strncmp (access, "ftp", 3) &&
1551: (*(port+1) == '2' &&
1552: *(port + 2) == '1' &&
1.82 cvs 1553: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1554: if (!newname)
1555: {
1.106 cvs 1556: char *orig = port;
1557: char *dest = port + 3;
1.33 cvs 1558: while((*orig++ = *dest++));
1559: /* Update path position, Henry Minsky */
1560: path -= 3;
1.25 cvs 1561: }
1.33 cvs 1562: }
1563: else if (newname)
1.106 cvs 1564: strncat (newname, port, (int) (path - port));
1.33 cvs 1565: }
1.25 cvs 1566:
1.33 cvs 1567: if (newname)
1568: {
1.106 cvs 1569: char *newpath = newname + strlen (newname);
1570: strcat (newname, path);
1.25 cvs 1571: path = newpath;
1.28 cvs 1572: /* Free old copy */
1573: TtaFreeMemory(*filename);
1.25 cvs 1574: *filename = newname;
1.33 cvs 1575: }
1.25 cvs 1576: return path;
1577: }
1578:
1579:
1580: /*----------------------------------------------------------------------
1.29 cvs 1581: SimplifyUrl: simplify a URI
1.32 cvs 1582: A URI is allowed to contain the sequence xxx/../ which may be
1583: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1584: Simplification helps us recognize duplicate URIs.
1.25 cvs 1585:
1.28 cvs 1586: Thus, /etc/junk/../fred becomes /etc/fred
1587: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1588:
1.28 cvs 1589: but we should NOT change
1590: http://fred.xxx.edu/../..
1.25 cvs 1591:
1592: or ../../albert.html
1593:
1.28 cvs 1594: In order to avoid empty URLs the following URLs become:
1.25 cvs 1595:
1596: /fred/.. becomes /fred/..
1597: /fred/././.. becomes /fred/..
1598: /fred/.././junk/.././ becomes /fred/..
1599:
1.28 cvs 1600: If more than one set of `://' is found (several proxies in cascade) then
1601: only the part after the last `://' is simplified.
1.25 cvs 1602:
1.28 cvs 1603: Returns: A string which might be the old one or a new one.
1.25 cvs 1604: ----------------------------------------------------------------------*/
1.106 cvs 1605: void SimplifyUrl (char **url)
1606: {
1607: char *path;
1608: char *access;
1609: char *newptr;
1610: char *p;
1611: char *orig, *dest, *end;
1.28 cvs 1612:
1.106 cvs 1613: char used_sep;
1.77 cvs 1614: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1615: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1616:
1617:
1.28 cvs 1618: if (!url || !*url)
1619: return;
1620:
1.106 cvs 1621: if (strchr (*url, URL_SEP))
1622: used_sep = URL_SEP;
1.32 cvs 1623: else
1.106 cvs 1624: used_sep = DIR_SEP;
1.32 cvs 1625:
1.77 cvs 1626: /* should we simplify double dot? */
1627: path = *url;
1.106 cvs 1628: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1629: ddot_simplify = FALSE;
1630: else
1631: ddot_simplify = TRUE;
1632:
1.28 cvs 1633: /* Find any scheme name */
1.106 cvs 1634: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1635: {
1636: /* Find host name */
1.28 cvs 1637: access = *url;
1.84 cvs 1638: while (access < path && (*access = utolower (*access)))
1.82 cvs 1639: access++;
1.28 cvs 1640: path += 3;
1.106 cvs 1641: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1642: /* For proxies */
1.106 cvs 1643: path = newptr + 3;
1.82 cvs 1644: /* We have a host name */
1.84 cvs 1645: path = HTCanon (url, path);
1.25 cvs 1646: }
1.106 cvs 1647: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1648: path += 2;
1649: else
1650: path = *url;
1.25 cvs 1651:
1.84 cvs 1652: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1653: /* Some URLs start //<foo> */
1654: path += 1;
1.94 cvs 1655: else if (IsFilePath (path))
1656: {
1657: /* doesn't need to do anything more */
1658: return;
1659: }
1.106 cvs 1660: else if (!strncmp (path, "news:", 5))
1.28 cvs 1661: {
1.106 cvs 1662: newptr = strchr (path+5, '@');
1.28 cvs 1663: if (!newptr)
1664: newptr = path + 5;
1665: while (*newptr)
1666: {
1667: /* Make group or host lower case */
1.84 cvs 1668: *newptr = utolower (*newptr);
1.28 cvs 1669: newptr++;
1.25 cvs 1670: }
1.28 cvs 1671: /* Doesn't need to do any more */
1672: return;
1.25 cvs 1673: }
1.28 cvs 1674:
1675: if ((p = path))
1676: {
1.106 cvs 1677: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1678: (end = strchr (path, '#'))))
1679: end = path + strlen (path);
1.28 cvs 1680:
1681: /* Parse string second time to simplify */
1682: p = path;
1683: while (p < end)
1684: {
1.77 cvs 1685: /* if we're pointing to a char, it's safe to reactivate the ../ convertion */
1.106 cvs 1686: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1687: ddot_simplify = TRUE;
1688:
1.33 cvs 1689: if (*p==used_sep)
1.28 cvs 1690: {
1.106 cvs 1691: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1692: {
1693: orig = p + 1;
1.84 cvs 1694: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1695: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1696: end = orig - 1;
1697: }
1.106 cvs 1698: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1699: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1700: {
1701: newptr = p;
1.52 cvs 1702: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1703: if (*newptr == used_sep)
1704: orig = newptr + 1;
1.28 cvs 1705: else
1.52 cvs 1706: orig = newptr;
1707:
1708: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1709: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1710: end = orig-1;
1711: /* Start again with prev slash */
1712: p = newptr;
1.28 cvs 1713: }
1.33 cvs 1714: else if (*(p+1) == used_sep)
1.28 cvs 1715: {
1.33 cvs 1716: while (*(p+1) == used_sep)
1.28 cvs 1717: {
1718: orig = p;
1719: dest = p + 1;
1720: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1721: end = orig-1;
1722: }
1723: }
1724: else
1.25 cvs 1725: p++;
1.28 cvs 1726: }
1727: else
1728: p++;
1.25 cvs 1729: }
1730: }
1.51 cvs 1731:
1732: /*
1733: ** Check for host/../.. kind of things
1734: */
1.106 cvs 1735: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1736: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 1737: *(path+1) = EOS;
1.51 cvs 1738:
1.28 cvs 1739: return;
1740: }
1741:
1742:
1743: /*----------------------------------------------------------------------
1.96 cvs 1744: NormalizeFile normalizes local names.
1.28 cvs 1745: Return TRUE if target and src differ.
1746: ----------------------------------------------------------------------*/
1.106 cvs 1747: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 1748: {
1.93 cvs 1749: # ifndef _WINDOWS
1.106 cvs 1750: char *s;
1.93 cvs 1751: int i;
1752: # endif /* !_WINDOWS */
1.82 cvs 1753: ThotBool change;
1.90 cvs 1754: int start_index; /* the first char that we'll copy */
1.28 cvs 1755:
1.54 cvs 1756: change = FALSE;
1.90 cvs 1757: start_index = 0;
1758:
1.106 cvs 1759: if (!src || src[0] == EOS)
1.96 cvs 1760: {
1.106 cvs 1761: target[0] = EOS;
1.96 cvs 1762: return FALSE;
1763: }
1.90 cvs 1764:
1765: /* @@ do I need file: or file:/ here? */
1.106 cvs 1766: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1767: {
1.90 cvs 1768: /* remove the prefix file: */
1769: start_index += 5;
1770:
1771: /* remove the localhost prefix */
1.106 cvs 1772: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 1773: start_index += 11;
1774:
1775: /* remove the first two slashes in / / /path */
1776: while (src[start_index] &&
1.106 cvs 1777: src[start_index] == '/'
1778: && src[start_index + 1] == '/')
1.94 cvs 1779: start_index++;
1780:
1781: #ifdef _WINDOWS
1782: /* remove any extra slash before the drive name */
1.106 cvs 1783: if (src[start_index] == '/'
1784: &&src[start_index+2] == ':')
1.94 cvs 1785: start_index++;
1786: #endif /* _WINDOWS */
1.90 cvs 1787:
1.106 cvs 1788: if (src[start_index] == EOS)
1.90 cvs 1789: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 1790: strcpy (target, DIR_STR);
1.90 cvs 1791: else
1.97 cvs 1792: /* as we're inside a file: URL, we'll apply all the convertions
1793: we know */
1794: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1795:
1796: change = TRUE;
1797: }
1.97 cvs 1798: else if (convertion != AM_CONV_NONE)
1.96 cvs 1799: {
1800: /* we are following a "local" relative link, we do all the
1801: convertions except for the HOME_DIR ~ one */
1.97 cvs 1802: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1803: }
1.90 cvs 1804: #ifndef _WINDOWS
1.106 cvs 1805: else if (src[0] == '~')
1.53 cvs 1806: {
1.96 cvs 1807: /* it must be a URL typed in a text input field */
1808: /* do the HOME_DIR ~ substitution */
1.82 cvs 1809: s = TtaGetEnvString ("HOME");
1.106 cvs 1810: strcpy (target, s);
1.90 cvs 1811: #if 0
1.96 cvs 1812: /* JK: invalidated this part of the code as it's simpler
1813: to add the DIR_SEP whenever we have something to add
1814: to the path rather than adding it systematically */
1.106 cvs 1815: if (src[1] != DIR_SEP)
1816: strcat (target, DIR_STR);
1.90 cvs 1817: #endif
1.106 cvs 1818: i = strlen (target);
1819: strcpy (&target[i], &src[1]);
1.54 cvs 1820: change = TRUE;
1.53 cvs 1821: }
1.90 cvs 1822: #endif /* _WINDOWS */
1.28 cvs 1823: else
1.96 cvs 1824: /* leave it as it is */
1.106 cvs 1825: strcpy (target, src);
1.96 cvs 1826:
1.28 cvs 1827: /* remove /../ and /./ */
1.29 cvs 1828: SimplifyUrl (&target);
1.54 cvs 1829: if (!change)
1.106 cvs 1830: change = strcmp (src, target);
1.28 cvs 1831: return (change);
1.25 cvs 1832: }
1833:
1.28 cvs 1834:
1.25 cvs 1835: /*----------------------------------------------------------------------
1.31 cvs 1836: MakeRelativeURL: make relative name
1.25 cvs 1837:
1.28 cvs 1838: This function creates and returns a string which gives an expression of
1839: one address as related to another. Where there is no relation, an absolute
1840: address is retured.
1.25 cvs 1841:
1.28 cvs 1842: On entry,
1.25 cvs 1843: Both names must be absolute, fully qualified names of nodes
1844: (no fragment bits)
1845:
1.28 cvs 1846: On exit,
1.25 cvs 1847: The return result points to a newly allocated name which, if
1848: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1849: The caller is responsible for freeing the resulting name later.
1850: ----------------------------------------------------------------------*/
1.106 cvs 1851: char *MakeRelativeURL (char *aName, char *relatedName)
1852: {
1853: char *return_value;
1854: char result[MAX_LENGTH];
1855: char *p;
1856: char *q;
1857: char *after_access;
1858: char *last_slash = NULL;
1859: int slashes, levels, len;
1.29 cvs 1860:
1.44 cvs 1861: # ifdef _WINDOWS
1862: int ndx;
1863: # endif /* _WINDOWS */
1864:
1.29 cvs 1865: if (aName == NULL || relatedName == NULL)
1866: return (NULL);
1867:
1868: slashes = 0;
1869: after_access = NULL;
1870: p = aName;
1871: q = relatedName;
1872: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1873: {
1874: /* Find extent of match */
1.106 cvs 1875: if (*p == ':')
1.29 cvs 1876: after_access = p + 1;
1.28 cvs 1877: if (*p == DIR_SEP)
1.27 cvs 1878: {
1.29 cvs 1879: /* memorize the last slash position and count them */
1.27 cvs 1880: last_slash = p;
1881: slashes++;
1.25 cvs 1882: }
1883: }
1884:
1.31 cvs 1885: /* q, p point to the first non-matching character or zero */
1.106 cvs 1886: if (*q == EOS)
1.31 cvs 1887: {
1888: /* New name is a subset of the related name */
1889: /* exactly the right length */
1.106 cvs 1890: len = strlen (p);
1891: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1892: strcpy (return_value, p);
1.31 cvs 1893: }
1894: else if ((slashes < 2 && after_access == NULL)
1895: || (slashes < 3 && after_access != NULL))
1896: {
1897: /* Two names whitout common path */
1898: /* exactly the right length */
1.106 cvs 1899: len = strlen (aName);
1900: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1901: strcpy (return_value, aName);
1.31 cvs 1902: }
1903: else
1904: {
1905: /* Some path in common */
1.106 cvs 1906: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 1907: /* just the same server */
1.106 cvs 1908: strcpy (result, last_slash);
1.31 cvs 1909: else
1910: {
1911: levels= 0;
1.106 cvs 1912: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 1913: if (*q == DIR_SEP)
1914: levels++;
1915:
1.106 cvs 1916: result[0] = EOS;
1.31 cvs 1917: for (;levels; levels--)
1.106 cvs 1918: strcat (result, "../");
1919: strcat (result, last_slash+1);
1.31 cvs 1920: }
1.52 cvs 1921:
1922: if (!*result)
1.106 cvs 1923: strcat (result, "./");
1.52 cvs 1924:
1.31 cvs 1925: /* exactly the right length */
1.106 cvs 1926: len = strlen (result);
1927: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1928: strcpy (return_value, result);
1.52 cvs 1929:
1.25 cvs 1930: }
1.44 cvs 1931: # ifdef _WINDOWS
1.106 cvs 1932: len = strlen (return_value);
1.44 cvs 1933: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 1934: if (return_value[ndx] == '\\')
1935: return_value[ndx] = '/' ;
1.44 cvs 1936: # endif /* _WINDOWS */
1.29 cvs 1937: return (return_value);
1.24 cvs 1938: }
1.35 cvs 1939:
1.104 kahan 1940: /*----------------------------------------------------------------------
1941: AM_GetFileSize
1942: Returns TRUE and the filesize in the 2nd parameter.
1943: Otherwise, in case of a system error, returns FALSE, with a
1944: filesize of 0L.
1945: ---------------------------------------------------------------------*/
1.106 cvs 1946: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 1947: {
1.106 cvs 1948: ThotFileHandle handle = ThotFile_BADHANDLE;
1949: ThotFileInfo info;
1.35 cvs 1950:
1.104 kahan 1951: *file_size = 0L;
1952: if (!TtaFileExist (filename))
1953: return FALSE;
1954:
1955: handle = TtaFileOpen (filename, ThotFile_READWRITE);
1956: if (handle == ThotFile_BADHANDLE)
1957: /* ThotFile_BADHANDLE */
1958: return FALSE;
1959: if (TtaFileStat (handle, &info) == 0)
1960: /* bad stat */
1961: info.size = 0L;
1962: TtaFileClose (handle);
1963: *file_size = (unsigned long) info.size;
1964: return TRUE;
1965: }
Webmaster