Annotation of Amaya/amaya/AHTURLTools.c, revision 1.116
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)
1.114 cvs 119: buffer = (char *) 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:
1.116 ! kahan 265: /*----------------------------------------------------------------------
! 266: PicTypeToMime
! 267: Converts a Thot PicType into the equivalent MIME type. If no convertion
! 268: is possible, it returns NULL.
! 269: ----------------------------------------------------------------------*/
! 270: char *PicTypeToMIME (PicType contentType)
! 271: {
! 272: char *mime_type;
! 273:
! 274: switch (contentType)
! 275: {
! 276: case xbm_type:
! 277: mime_type ="image/x-xbitmap";
! 278: break;
! 279: case eps_type:
! 280: mime_type ="application/postscript";
! 281: break;
! 282: case xpm_type:
! 283: mime_type ="image/x-xpicmap";
! 284: break;
! 285: case gif_type:
! 286: mime_type ="image/gif";
! 287: break;
! 288: case jpeg_type:
! 289: mime_type ="image/jpeg";
! 290: break;
! 291: case png_type:
! 292: mime_type ="image/png";
! 293: break;
! 294: case svg_type:
! 295: mime_type ="image/svg";
! 296: break;
! 297: case unknown_type:
! 298: default:
! 299: mime_type = NULL;
! 300: }
! 301:
! 302: return mime_type;
! 303: }
1.61 cvs 304:
305: /*----------------------------------------------------------------------
306: ExtractSuffix extract suffix from document nane.
307: ----------------------------------------------------------------------*/
1.106 cvs 308: void ExtractSuffix (char *aName, char *aSuffix)
1.61 cvs 309: {
1.106 cvs 310: int lg, i;
311: char *ptr, *oldptr;
1.61 cvs 312:
313: if (!aSuffix || !aName)
314: /* bad suffix */
315: return;
316:
1.106 cvs 317: aSuffix[0] = EOS;
318: lg = strlen (aName);
1.61 cvs 319: if (lg)
320: {
321: /* the name is not empty */
322: oldptr = ptr = &aName[0];
323: do
324: {
1.106 cvs 325: ptr = strrchr (oldptr, '.');
1.61 cvs 326: if (ptr)
327: oldptr = &ptr[1];
328: }
329: while (ptr);
330:
331: i = (int) (oldptr) - (int) (aName); /* name length */
332: if (i > 1)
333: {
1.106 cvs 334: aName[i - 1] = EOS;
1.61 cvs 335: if (i != lg)
1.106 cvs 336: strcpy (aSuffix, oldptr);
1.61 cvs 337: }
338: }
339: }
340:
1.4 cvs 341: /*----------------------------------------------------------------------
1.9 cvs 342: IsHTMLName
343: returns TRUE if path points to an HTML resource.
1.4 cvs 344: ----------------------------------------------------------------------*/
1.109 cvs 345: ThotBool IsHTMLName (const char *path)
1.106 cvs 346: {
347: char temppath[MAX_LENGTH];
348: char suffix[MAX_LENGTH];
349: char nsuffix[MAX_LENGTH];
1.101 cvs 350: int i;
1.5 cvs 351:
1.101 cvs 352: if (!path)
353: return (FALSE);
1.5 cvs 354:
1.106 cvs 355: strcpy (temppath, path);
1.101 cvs 356: ExtractSuffix (temppath, suffix);
357: i = 0;
1.106 cvs 358: while (suffix[i] != EOS)
1.101 cvs 359: {
360: /* Normalize the suffix */
361: i = 0;
1.106 cvs 362: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 363: {
364: nsuffix[i] = utolower (suffix[i]);
365: i++;
366: }
1.106 cvs 367: nsuffix[i] = EOS;
368: if (!strcmp (nsuffix, "html") ||
369: !strcmp (nsuffix, "htm") ||
370: !strcmp (nsuffix, "shtml") ||
371: !strcmp (nsuffix, "jsp") ||
372: !strcmp (nsuffix, "xht") ||
373: !strcmp (nsuffix, "xhtm") ||
374: !strcmp (nsuffix, "xhtml"))
1.101 cvs 375: return (TRUE);
1.106 cvs 376: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 377: {
378: /* take into account compressed files */
379: ExtractSuffix (temppath, suffix);
380: /* Normalize the suffix */
381: i = 0;
1.106 cvs 382: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 383: {
384: nsuffix[i] = utolower (suffix[i]);
385: i++;
386: }
1.106 cvs 387: nsuffix[i] = EOS;
388: if (!strcmp (nsuffix, "html") ||
389: !strcmp (nsuffix, "htm") ||
390: !strcmp (nsuffix, "shtml") ||
391: !strcmp (nsuffix, "jsp") ||
392: !strcmp (nsuffix, "xht") ||
393: !strcmp (nsuffix, "xhtm") ||
394: !strcmp (nsuffix, "xhtml"))
1.101 cvs 395: return (TRUE);
396: else
397: return (FALSE);
398: }
399: else
400: /* check if there is another suffix */
401: ExtractSuffix (temppath, suffix);
402: }
1.88 cvs 403: return (FALSE);
1.3 cvs 404: }
405:
1.4 cvs 406: /*----------------------------------------------------------------------
1.56 cvs 407: IsXMLName
408: returns TRUE if path points to an XML resource.
409: ----------------------------------------------------------------------*/
1.111 cvs 410: ThotBool IsXMLName (const char *path)
1.56 cvs 411: {
1.106 cvs 412: char temppath[MAX_LENGTH];
413: char suffix[MAX_LENGTH];
1.56 cvs 414:
415: if (!path)
416: return (FALSE);
417:
1.106 cvs 418: strcpy (temppath, path);
1.56 cvs 419: ExtractSuffix (temppath, suffix);
420:
1.106 cvs 421: if (!strcasecmp (suffix, "xml") ||
422: !strcasecmp (suffix, "xht") ||
423: !strcmp (suffix, "xhtm") ||
424: !strcmp (suffix, "xhtml"))
1.56 cvs 425: return (TRUE);
1.106 cvs 426: else if (!strcmp (suffix, "gz"))
1.56 cvs 427: {
428: /* take into account compressed files */
429: ExtractSuffix (temppath, suffix);
1.106 cvs 430: if (!strcasecmp (suffix, "xml") ||
431: !strcasecmp (suffix, "xht") ||
432: !strcmp (suffix, "xhtm") ||
433: !strcmp (suffix, "xhtml"))
1.60 cvs 434: return (TRUE);
435: else
436: return (FALSE);
437: }
438: else
439: return (FALSE);
440: }
441:
442: /*----------------------------------------------------------------------
1.103 cvs 443: IsMathMLName
444: returns TRUE if path points to an MathML resource.
445: ----------------------------------------------------------------------*/
1.111 cvs 446: ThotBool IsMathMLName (const char *path)
1.103 cvs 447: {
1.106 cvs 448: char temppath[MAX_LENGTH];
449: char suffix[MAX_LENGTH];
1.103 cvs 450:
451: if (!path)
452: return (FALSE);
453:
1.106 cvs 454: strcpy (temppath, path);
1.103 cvs 455: ExtractSuffix (temppath, suffix);
456:
1.106 cvs 457: if (!strcasecmp (suffix, "mml"))
1.103 cvs 458: return (TRUE);
1.106 cvs 459: else if (!strcmp (suffix, "gz"))
1.103 cvs 460: {
461: /* take into account compressed files */
462: ExtractSuffix (temppath, suffix);
1.106 cvs 463: if (!strcasecmp (suffix, "mml"))
1.103 cvs 464: return (TRUE);
465: else
466: return (FALSE);
467: }
468: else
469: return (FALSE);
470: }
471:
472: /*----------------------------------------------------------------------
473: IsSVGName
474: returns TRUE if path points to an MathML resource.
475: ----------------------------------------------------------------------*/
1.111 cvs 476: ThotBool IsSVGName (const char *path)
1.103 cvs 477: {
1.106 cvs 478: char temppath[MAX_LENGTH];
479: char suffix[MAX_LENGTH];
1.103 cvs 480:
481: if (!path)
482: return (FALSE);
483:
1.106 cvs 484: strcpy (temppath, path);
1.103 cvs 485: ExtractSuffix (temppath, suffix);
486:
1.106 cvs 487: if (!strcasecmp (suffix, "svg"))
1.103 cvs 488: return (TRUE);
1.106 cvs 489: else if (!strcmp (suffix, "gz"))
1.103 cvs 490: {
491: /* take into account compressed files */
492: ExtractSuffix (temppath, suffix);
1.106 cvs 493: if (!strcasecmp (suffix, "svg"))
1.103 cvs 494: return (TRUE);
495: else
496: return (FALSE);
497: }
498: else
499: return (FALSE);
500: }
501:
502: /*----------------------------------------------------------------------
1.60 cvs 503: IsCSSName
504: returns TRUE if path points to an XML resource.
505: ----------------------------------------------------------------------*/
1.111 cvs 506: ThotBool IsCSSName (const char *path)
1.60 cvs 507: {
1.106 cvs 508: char temppath[MAX_LENGTH];
509: char suffix[MAX_LENGTH];
1.60 cvs 510:
511: if (!path)
512: return (FALSE);
513:
1.106 cvs 514: strcpy (temppath, path);
1.60 cvs 515: ExtractSuffix (temppath, suffix);
516:
1.106 cvs 517: if (!strcasecmp (suffix, "css"))
1.60 cvs 518: return (TRUE);
1.106 cvs 519: else if (!strcmp (suffix, "gz"))
1.60 cvs 520: {
521: /* take into account compressed files */
522: ExtractSuffix (temppath, suffix);
1.106 cvs 523: if (!strcasecmp (suffix, "css"))
1.56 cvs 524: return (TRUE);
525: else
526: return (FALSE);
527: }
528: else
529: return (FALSE);
530: }
531:
532: /*----------------------------------------------------------------------
1.9 cvs 533: IsImageName
534: returns TRUE if path points to an image resource.
1.4 cvs 535: ----------------------------------------------------------------------*/
1.111 cvs 536: ThotBool IsImageName (const char *path)
1.106 cvs 537: {
538: char temppath[MAX_LENGTH];
539: char suffix[MAX_LENGTH];
540: char nsuffix[MAX_LENGTH];
1.5 cvs 541: int i;
542:
543: if (!path)
1.13 cvs 544: return (FALSE);
1.5 cvs 545:
1.106 cvs 546: strcpy (temppath, path);
1.5 cvs 547: ExtractSuffix (temppath, suffix);
548:
549: /* Normalize the suffix */
550: i = 0;
1.106 cvs 551: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 552: {
1.67 cvs 553: nsuffix[i] = utolower (suffix[i]);
1.13 cvs 554: i++;
555: }
1.106 cvs 556: nsuffix[i] = EOS;
557: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
558: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
559: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 560: return (TRUE);
561: return (FALSE);
1.3 cvs 562: }
563:
1.4 cvs 564: /*----------------------------------------------------------------------
1.58 cvs 565: IsImageType
566: returns TRUE if type points to an image resource.
567: ----------------------------------------------------------------------*/
1.111 cvs 568: ThotBool IsImageType (const char *type)
1.58 cvs 569: {
1.106 cvs 570: char temptype[MAX_LENGTH];
1.58 cvs 571: int i;
572:
573: if (!type)
574: return (FALSE);
575:
1.106 cvs 576: strcpy (temptype, type);
1.58 cvs 577: /* Normalize the type */
578: i = 0;
1.106 cvs 579: while (temptype[i] != EOS)
1.58 cvs 580: {
581: temptype[i] = tolower (temptype[i]);
582: i++;
583: }
1.111 cvs 584: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
585: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
586: !strcmp (temptype, "png"))
1.58 cvs 587: return (TRUE);
588: return (FALSE);
589: }
590:
591: /*----------------------------------------------------------------------
1.9 cvs 592: IsTextName
1.4 cvs 593: ----------------------------------------------------------------------*/
1.111 cvs 594: ThotBool IsTextName (const char *path)
1.106 cvs 595: {
596: char temppath[MAX_LENGTH];
597: char suffix[MAX_LENGTH];
598: char nsuffix[MAX_LENGTH];
1.5 cvs 599: int i;
600:
601: if (!path)
1.13 cvs 602: return (FALSE);
1.5 cvs 603:
1.106 cvs 604: strcpy (temppath, path);
1.5 cvs 605: ExtractSuffix (temppath, suffix);
606:
607: /* Normalize the suffix */
608: i = 0;
1.106 cvs 609: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 610: {
1.25 cvs 611: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 612: i++;
613: }
1.106 cvs 614: nsuffix[i] = EOS;
1.5 cvs 615:
1.111 cvs 616: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 617: return (TRUE);
1.106 cvs 618: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 619: {
1.39 cvs 620: /* take into account compressed files */
1.13 cvs 621: ExtractSuffix (temppath, suffix);
622: /* Normalize the suffix */
623: i = 0;
1.106 cvs 624: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 625: {
1.25 cvs 626: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 627: i++;
628: }
1.106 cvs 629: nsuffix[i] = EOS;
1.111 cvs 630: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 631: return (TRUE);
632: else
633: return (FALSE);
634: }
635: else
636: return (FALSE);
1.3 cvs 637: }
638:
1.4 cvs 639: /*----------------------------------------------------------------------
1.9 cvs 640: IsHTTPPath
641: returns TRUE if path is in fact an http URL.
1.4 cvs 642: ----------------------------------------------------------------------*/
1.112 cvs 643: ThotBool IsHTTPPath (const char *path)
1.3 cvs 644: {
1.5 cvs 645: if (!path)
646: return FALSE;
1.3 cvs 647:
1.106 cvs 648: if ((!strncmp (path, "http:", 5) != 0)
649: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
650: || !strncmp (path, "internal:", 9))
1.58 cvs 651: return TRUE;
652: return FALSE;
1.3 cvs 653: }
654:
1.4 cvs 655: /*----------------------------------------------------------------------
1.9 cvs 656: IsWithParameters
657: returns TRUE if url has a concatenated query string.
1.4 cvs 658: ----------------------------------------------------------------------*/
1.66 cvs 659: ThotBool IsWithParameters (const char *url)
1.3 cvs 660: {
1.5 cvs 661: int i;
1.3 cvs 662:
1.9 cvs 663: if ((!url) || (url[0] == EOS))
1.5 cvs 664: return FALSE;
1.3 cvs 665:
1.9 cvs 666: i = strlen (url) - 1;
667: while (i > 0 && url[i--] != '?')
1.5 cvs 668: if (i < 0)
669: return FALSE;
1.3 cvs 670:
1.5 cvs 671: /* There is a parameter */
672: return TRUE;
1.3 cvs 673: }
674:
1.4 cvs 675: /*----------------------------------------------------------------------
1.9 cvs 676: IsW3Path
677: returns TRUE if path is in fact a URL.
1.4 cvs 678: ----------------------------------------------------------------------*/
1.106 cvs 679: ThotBool IsW3Path (const char *path)
680: {
681: if (strncmp (path, "http:", 5) &&
682: strncmp (path, "ftp:", 4) &&
683: strncmp (path, "telnet:", 7) &&
684: strncmp (path, "wais:", 5) &&
685: strncmp (path, "news:", 5) &&
686: strncmp (path, "gopher:", 7) &&
687: strncmp (path, "mailto:", 7) &&
688: strncmp (path, "archie:", 7))
1.72 cvs 689: return FALSE;
690: return TRUE;
1.3 cvs 691: }
692:
1.4 cvs 693: /*----------------------------------------------------------------------
1.90 cvs 694: IsFilePath
695: returns TRUE if path is in fact a URL.
696: ----------------------------------------------------------------------*/
1.106 cvs 697: ThotBool IsFilePath (const char *path)
1.90 cvs 698: {
1.106 cvs 699: if (strncmp (path, "file:", 5))
1.90 cvs 700: return FALSE;
701: return TRUE;
702: }
703:
704: /*----------------------------------------------------------------------
1.9 cvs 705: IsValidProtocol
706: returns true if the url protocol is supported by Amaya.
1.4 cvs 707: ----------------------------------------------------------------------*/
1.106 cvs 708: ThotBool IsValidProtocol (const char *url)
709: {
710: if (!strncmp (url, "http:", 5)
711: || !strncmp (url, "internal:", 9)
712: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 713: /* experimental */
1.24 cvs 714: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 715: return (TRUE);
1.5 cvs 716: else
1.8 cvs 717: return (FALSE);
1.3 cvs 718: }
719:
1.31 cvs 720:
721: /*----------------------------------------------------------------------
722: GetBaseURL
723: normalizes orgName according to a base associated with doc, and
724: following the standard URL format rules.
725: The function returns the base used to solve relative URL and SRC:
726: - the base of the document,
727: - or the document path (without document name).
728: ----------------------------------------------------------------------*/
1.106 cvs 729: char *GetBaseURL (Document doc)
1.31 cvs 730: {
731: Element el;
732: ElementType elType;
733: AttributeType attrType;
734: Attribute attr;
1.106 cvs 735: char *ptr, *basename;
1.31 cvs 736: int length;
737:
1.113 cvs 738: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 739: return NULL;
1.106 cvs 740: basename = TtaGetMemory (MAX_LENGTH);
741: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
742: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 743: length = MAX_LENGTH -1;
1.113 cvs 744: /* is it a HTML document ? */
1.31 cvs 745: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.113 cvs 746: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
747: /* it's a HTML document */
1.65 cvs 748: {
1.113 cvs 749: /* get the document element */
750: el = TtaGetMainRoot (doc);
751: /* search the BASE element */
752: elType.ElTypeNum = HTML_EL_HEAD;
753: el = TtaSearchTypedElement (elType, SearchForward, el);
754: if (el)
755: /* there is a HEAD element */
756: {
757: /* look for a BASE element within the HEAD */
758: elType.ElTypeNum = HTML_EL_BASE;
759: el = TtaSearchTypedElement (elType, SearchInTree, el);
760: }
761: if (el)
1.31 cvs 762: {
1.113 cvs 763: /* The document has a BASE element. Get the HREF attribute of the
764: BASE element */
765: attrType.AttrSSchema = elType.ElSSchema;
766: attrType.AttrTypeNum = HTML_ATTR_HREF_;
767: attr = TtaGetAttribute (el, attrType);
768: if (attr)
1.31 cvs 769: {
1.113 cvs 770: /* Use the base path of the document */
771: TtaGiveTextAttributeValue (attr, basename, &length);
772: /* base and orgName have to be separated by a DIR_SEP */
773: length--;
774: if (basename[0] != EOS && basename[length] != URL_SEP &&
775: basename[length] != DIR_SEP)
776: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 777: {
1.113 cvs 778: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
779: AMAYA_PARSE_HOST |
780: AMAYA_PARSE_PUNCTUATION);
781: if (ptr && !strcmp (ptr, basename))
782: {
783: /* it has this form, complete it by adding a URL_STR */
784: if (strchr (basename, DIR_SEP))
785: strcat (basename, DIR_STR);
786: else
787: strcat (basename, URL_STR);
788: length++;
789: }
790: if (ptr)
791: TtaFreeMemory (ptr);
1.31 cvs 792: }
793: }
794: }
1.113 cvs 795: }
796:
1.31 cvs 797: /* Remove anything after the last DIR_SEP char. If no such char is found,
798: * then search for the first ":" char, hoping that what's before that is a
799: * protocol. If found, end the string there. If neither char is found,
800: * then discard the whole base element.
801: */
1.106 cvs 802: length = strlen (basename) - 1;
1.31 cvs 803: /* search for the last DIR_SEP char */
1.106 cvs 804: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 805: length--;
806: if (length >= 0)
807: /* found the last DIR_SEP char, end the string there */
1.106 cvs 808: basename[length + 1] = EOS;
1.31 cvs 809: else
810: /* search for the first PATH_STR char */
811: {
1.106 cvs 812: for (length = 0; basename[length] != ':' &&
813: basename[length] != EOS; length ++);
814: if (basename[length] == ':')
1.31 cvs 815: /* found, so end the string there */
1.106 cvs 816: basename[length + 1] = EOS;
1.31 cvs 817: else
818: /* not found, discard the base */
1.106 cvs 819: basename[0] = EOS;
1.31 cvs 820: }
821: return (basename);
822: }
823:
824:
1.4 cvs 825: /*----------------------------------------------------------------------
1.40 cvs 826: GetLocalPath
827: Allocate and return the local document path associated to the url
828: ----------------------------------------------------------------------*/
1.106 cvs 829: char *GetLocalPath (Document doc, char *url)
830: {
831: char *ptr;
832: char *n;
833: char *documentname;
834: char url_sep;
1.83 cvs 835: int len;
1.67 cvs 836: ThotBool noFile;
1.40 cvs 837:
838: if (url != NULL)
839: {
840: /* check whether the file name exists */
1.106 cvs 841: len = strlen (url) - 1;
1.71 cvs 842: if (IsW3Path (url))
1.106 cvs 843: url_sep = '/';
1.41 cvs 844: else
1.106 cvs 845: url_sep = DIR_SEP;
1.41 cvs 846: noFile = (url[len] == url_sep);
1.40 cvs 847: if (noFile)
1.106 cvs 848: url[len] = EOS;
849: ptr = TtaGetMemory (MAX_LENGTH);
850: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 851: TtaExtractName (url, ptr, documentname);
1.106 cvs 852: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 853: if (!TtaCheckDirectory (ptr))
854: /* directory did not exist */
1.72 cvs 855: TtaMakeDirectory (ptr);
1.47 cvs 856:
857: /* don't include the query string within document name */
1.106 cvs 858: n = strrchr (documentname, '?');
1.47 cvs 859: if (n != NULL)
1.106 cvs 860: *n = EOS;
1.46 cvs 861: /* don't include ':' within document name */
1.106 cvs 862: n = strchr (documentname, ':');
1.46 cvs 863: if (n != NULL)
1.106 cvs 864: *n = EOS;
1.69 cvs 865: /* if after all this operations document name
866: is empty, let's use noname.html instead */
1.106 cvs 867: if (documentname[0] == EOS)
868: strcat (ptr, "noname.html");
1.69 cvs 869: else
1.106 cvs 870: strcat (ptr, documentname);
1.40 cvs 871: TtaFreeMemory (documentname);
872: /* restore the url */
873: if (noFile)
1.41 cvs 874: url[len] = url_sep;
1.40 cvs 875: return (ptr);
876: }
877: else
878: return (NULL);
879: }
880:
1.73 cvs 881: /*----------------------------------------------------------------------
1.79 cvs 882: ExtractTarget extract the target name from document nane.
883: ----------------------------------------------------------------------*/
1.106 cvs 884: void ExtractTarget (char *aName, char *target)
1.79 cvs 885: {
1.106 cvs 886: int lg, i;
887: char *ptr;
888: char *oldptr;
1.79 cvs 889:
890: if (!target || !aName)
891: /* bad target */
892: return;
893:
1.106 cvs 894: target[0] = EOS;
895: lg = strlen (aName);
1.79 cvs 896: if (lg)
897: {
898: /* the name is not empty */
899: oldptr = ptr = &aName[0];
900: do
901: {
1.106 cvs 902: ptr = strrchr (oldptr, '#');
1.79 cvs 903: if (ptr)
904: oldptr = &ptr[1];
905: }
906: while (ptr);
907:
908: i = (int) (oldptr) - (int) (aName); /* name length */
909: if (i > 1)
910: {
1.106 cvs 911: aName[i - 1] = EOS;
1.79 cvs 912: if (i != lg)
1.106 cvs 913: strcpy (target, oldptr);
1.79 cvs 914: }
915: }
916: }
917:
918: /*----------------------------------------------------------------------
1.90 cvs 919: RemoveNewLines (text)
920: Removes any '\n' chars that are found in text.
921: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 922: ----------------------------------------------------------------------*/
1.106 cvs 923: ThotBool RemoveNewLines (char *text)
924: {
925: ThotBool change = FALSE;
926: char *src;
927: char *dest;
1.90 cvs 928:
929: src = text;
930: dest = text;
1.115 kahan 931:
932: /* remove any preceding whitespace */
933: while (*src && *src == ' ')
934: {
935: src++;
936: change = 1;
937: }
938:
1.90 cvs 939: while (*src)
940: {
941: switch (*src)
942: {
1.106 cvs 943: case '\n':
1.90 cvs 944: /* don't copy the newline */
945: change = 1;
946: break;
947: default:
948: *dest = *src;
949: dest++;
950: break;
951: }
952: src++;
953: }
954: /* copy the last EOS char */
955: *dest = *src;
956:
957: return (change);
958: }
959:
960: /*----------------------------------------------------------------------
961: CleanCopyFileURL
962: Copies a file url from a src string to destination string.
1.97 cvs 963: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
964: we want to do).
1.90 cvs 965: ----------------------------------------------------------------------*/
1.106 cvs 966: static void CleanCopyFileURL (char *dest, char *src,
967: ConvertionType convertion)
1.90 cvs 968: {
969: while (*src)
1.89 cvs 970: {
1.90 cvs 971: switch (*src)
1.89 cvs 972: {
973: #ifdef _WINDOWS
1.106 cvs 974: case URL_SEP:
1.96 cvs 975: /* make DIR_SEP transformation */
1.97 cvs 976: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 977: *dest = DIR_SEP;
1.96 cvs 978: else
979: *dest = *src;
1.90 cvs 980: dest++;
1.96 cvs 981: src++;
1.90 cvs 982: break;
1.89 cvs 983: #endif /* _WINDOWS */
1.96 cvs 984:
1.106 cvs 985: case '%':
1.97 cvs 986: if (convertion & AM_CONV_PERCENT)
1.96 cvs 987: {
1.97 cvs 988: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 989: src++;
1.106 cvs 990: if (*src != EOS)
1.97 cvs 991: {
992: *dest = UnEscapeChar (*src) * 16;
993: src++;
994: }
1.106 cvs 995: if (*src != EOS)
1.97 cvs 996: {
997: *dest = *dest + UnEscapeChar (*src);
998: src++;
999: }
1000: dest++;
1.96 cvs 1001: }
1.97 cvs 1002: else
1.96 cvs 1003: {
1.97 cvs 1004: *dest = *src;
1005: dest++;
1.96 cvs 1006: src++;
1007: }
1008: break;
1009:
1.90 cvs 1010: default:
1011: *dest = *src;
1.89 cvs 1012: dest++;
1.96 cvs 1013: src++;
1.90 cvs 1014: break;
1.89 cvs 1015: }
1016: }
1.90 cvs 1017: /* copy the EOS char */
1018: *dest = *src;
1.73 cvs 1019: }
1.40 cvs 1020:
1021: /*----------------------------------------------------------------------
1.9 cvs 1022: NormalizeURL
1023: normalizes orgName according to a base associated with doc, and
1024: following the standard URL format rules.
1.113 cvs 1025: if doc is < 0, use as a base the URL of the document that contains
1026: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1027: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1028: other path.
1.9 cvs 1029: The function returns the new complete and normalized URL
1.12 cvs 1030: or file name path (newName) and the name of the document (docName).
1.9 cvs 1031: N.B. If the function can't find out what's the docName, it assigns
1032: the name "noname.html".
1.4 cvs 1033: ----------------------------------------------------------------------*/
1.106 cvs 1034: void NormalizeURL (char *orgName, Document doc, char *newName,
1035: char *docName, char *otherPath)
1036: {
1037: char *basename;
1038: char tempOrgName[MAX_LENGTH];
1039: char *ptr;
1040: char used_sep;
1.84 cvs 1041: int length;
1042: ThotBool check;
1.5 cvs 1043:
1.110 cvs 1044: #ifdef _WINDOWS
1.44 cvs 1045: int ndx;
1.110 cvs 1046: #endif /* _WINDOWS */
1.44 cvs 1047:
1.5 cvs 1048: if (!newName || !docName)
1049: return;
1.18 cvs 1050:
1.113 cvs 1051: if (doc < 0)
1052: basename = TtaStrdup (SavedDocumentURL);
1053: else if (doc > 0)
1.53 cvs 1054: basename = GetBaseURL (doc);
1055: else if (otherPath != NULL)
1.108 cvs 1056: basename = TtaStrdup (otherPath);
1.32 cvs 1057: else
1.53 cvs 1058: basename = NULL;
1.32 cvs 1059:
1.18 cvs 1060: /*
1.31 cvs 1061: * Clean orgName
1062: * Make sure we have a complete orgName, without any leading or trailing
1063: * white spaces, or trailinbg new lines
1064: */
1.5 cvs 1065: ptr = orgName;
1.18 cvs 1066: /* skip leading white space and new line characters */
1.106 cvs 1067: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1068: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1069: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1070: /*
1.31 cvs 1071: * Make orgName a complete URL
1072: * If the URL does not include a protocol, then try to calculate
1073: * one using the doc's base element (if it exists),
1074: */
1.106 cvs 1075: if (tempOrgName[0] == EOS)
1.53 cvs 1076: {
1.106 cvs 1077: newName[0] = EOS;
1078: docName[0] = EOS;
1.53 cvs 1079: TtaFreeMemory (basename);
1080: return;
1081: }
1.49 cvs 1082:
1083: /* clean trailing white space */
1.106 cvs 1084: length = strlen (tempOrgName) - 1;
1085: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1086: {
1.106 cvs 1087: tempOrgName[length] = EOS;
1.53 cvs 1088: length--;
1089: }
1.50 cvs 1090:
1.55 cvs 1091: /* remove extra dot (which dot???) */
1092: /* ugly, but faster than a strcmp */
1.106 cvs 1093: if (tempOrgName[length] == '.'
1094: && (length == 0 || tempOrgName[length-1] != '.'))
1095: tempOrgName[length] = EOS;
1.50 cvs 1096:
1.94 cvs 1097: if (IsW3Path (tempOrgName))
1.53 cvs 1098: {
1099: /* the name is complete, go to the Sixth Step */
1.106 cvs 1100: strcpy (newName, tempOrgName);
1.53 cvs 1101: SimplifyUrl (&newName);
1102: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1103: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1104: AMAYA_PARSE_HOST |
1105: AMAYA_PARSE_PUNCTUATION);
1106: if (ptr && !strcmp (ptr, newName))
1107: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1108: strcat (newName, URL_STR);
1.49 cvs 1109:
1.53 cvs 1110: if (ptr)
1.50 cvs 1111: TtaFreeMemory (ptr);
1.53 cvs 1112: }
1.113 cvs 1113: else if (basename == NULL)
1.53 cvs 1114: /* the name is complete, go to the Sixth Step */
1.106 cvs 1115: strcpy (newName, tempOrgName);
1.53 cvs 1116: else
1117: {
1.31 cvs 1118: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1119: #ifdef _WINDOWS
1.53 cvs 1120: if (!IsW3Path (basename))
1121: {
1.106 cvs 1122: length = strlen (tempOrgName);
1.53 cvs 1123: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1124: if (tempOrgName [ndx] == '/')
1125: tempOrgName [ndx] = '\\';
1.53 cvs 1126: }
1.110 cvs 1127: #endif /* _WINDOWS */
1.25 cvs 1128: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1129: if (ptr)
1130: {
1131: SimplifyUrl (&ptr);
1.106 cvs 1132: strcpy (newName, ptr);
1.53 cvs 1133: TtaFreeMemory (ptr);
1134: }
1135: else
1.106 cvs 1136: newName[0] = EOS;
1.53 cvs 1137: }
1.36 cvs 1138:
1139: TtaFreeMemory (basename);
1.18 cvs 1140: /*
1.31 cvs 1141: * Prepare the docname that will refer to this ressource in the
1142: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1143: * noname.html as a default ressource name
1.18 cvs 1144: */
1.106 cvs 1145: if (newName[0] != EOS)
1.53 cvs 1146: {
1.106 cvs 1147: length = strlen (newName) - 1;
1148: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1149: {
1150: used_sep = newName[length];
1151: check = TRUE;
1152: while (check)
1153: {
1.50 cvs 1154: length--;
1155: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1156: length--;
1.106 cvs 1157: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1158: {
1.106 cvs 1159: newName[length+1] = EOS;
1.53 cvs 1160: /* remove also previous directory */
1161: length--;
1162: while (length >= 0 && newName[length] != used_sep)
1163: length--;
1.106 cvs 1164: if (strncmp (&newName[length+1], "//", 2))
1.53 cvs 1165: /* don't remove server name */
1.106 cvs 1166: newName[length+1] = EOS;
1.53 cvs 1167: }
1.106 cvs 1168: else if (!strncmp (&newName[length+1], ".", 1))
1169: newName[length+1] = EOS;
1.50 cvs 1170: else
1.53 cvs 1171: check = FALSE;
1172: }
1173: /* docname was not comprised inside the URL, so let's */
1174: /* assign the default ressource name */
1.106 cvs 1175: strcpy (docName, "noname.html");
1.53 cvs 1176: }
1177: else
1178: { /* docname is comprised inside the URL */
1.110 cvs 1179: while (length >= 0 && newName[length] != URL_SEP &&
1180: newName[length] != DIR_SEP)
1.53 cvs 1181: length--;
1182: if (length < 0)
1.106 cvs 1183: strcpy (docName, newName);
1.53 cvs 1184: else
1.106 cvs 1185: strcpy (docName, &newName[length+1]);
1.53 cvs 1186: }
1187: }
1188: else
1.106 cvs 1189: docName[0] = EOS;
1.18 cvs 1190: }
1.3 cvs 1191:
1.4 cvs 1192: /*----------------------------------------------------------------------
1.9 cvs 1193: IsSameHost
1.4 cvs 1194: ----------------------------------------------------------------------*/
1.106 cvs 1195: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1196: {
1.106 cvs 1197: char *basename_ptr1, *basename_ptr2;
1198: ThotBool result;
1.3 cvs 1199:
1.106 cvs 1200: basename_ptr1 = AmayaParseUrl (url1, "",
1201: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1202: basename_ptr2 = AmayaParseUrl (url2, "",
1203: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1204:
1.106 cvs 1205: if (strcmp (basename_ptr1, basename_ptr2))
1206: result = FALSE;
1207: else
1208: result = TRUE;
1209: TtaFreeMemory (basename_ptr1);
1210: TtaFreeMemory (basename_ptr2);
1211: return (result);
1.3 cvs 1212: }
1213:
1214:
1.4 cvs 1215: /*----------------------------------------------------------------------
1.22 cvs 1216: HasKnownFileSuffix
1217: returns TRUE if path points to a file ending with a suffix.
1218: ----------------------------------------------------------------------*/
1.106 cvs 1219: ThotBool HasKnownFileSuffix (const char *path)
1220: {
1221: char *root;
1222: char temppath[MAX_LENGTH];
1223: char suffix[MAX_LENGTH];
1.22 cvs 1224:
1.106 cvs 1225: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1226: return (FALSE);
1227:
1.106 cvs 1228: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1229:
1230: if (root)
1231: {
1.106 cvs 1232: strcpy (temppath, root);
1.25 cvs 1233: TtaFreeMemory (root);
1.22 cvs 1234: /* Get the suffix */
1235: ExtractSuffix (temppath, suffix);
1236:
1.106 cvs 1237: if( suffix[0] == EOS)
1.22 cvs 1238: /* no suffix */
1239: return (FALSE);
1240:
1241: /* Normalize the suffix */
1242: ConvertToLowerCase (suffix);
1243:
1.106 cvs 1244: if (!strcmp (suffix, "gz"))
1.22 cvs 1245: /* skip the compressed suffix */
1246: {
1247: ExtractSuffix (temppath, suffix);
1.106 cvs 1248: if(suffix[0] == EOS)
1.22 cvs 1249: /* no suffix */
1250: return (FALSE);
1251: /* Normalize the suffix */
1252: ConvertToLowerCase (suffix);
1253: }
1254:
1.106 cvs 1255: if (strcmp (suffix, "gif") &&
1256: strcmp (suffix, "xbm") &&
1257: strcmp (suffix, "xpm") &&
1258: strcmp (suffix, "jpg") &&
1259: strcmp (suffix, "pdf") &&
1260: strcmp (suffix, "png") &&
1261: strcmp (suffix, "tgz") &&
1262: strcmp (suffix, "xpg") &&
1263: strcmp (suffix, "xpd") &&
1264: strcmp (suffix, "ps") &&
1265: strcmp (suffix, "au") &&
1266: strcmp (suffix, "html") &&
1267: strcmp (suffix, "htm") &&
1268: strcmp (suffix, "shtml") &&
1269: strcmp (suffix, "xht") &&
1270: strcmp (suffix, "xhtm") &&
1271: strcmp (suffix, "xhtml") &&
1272: strcmp (suffix, "txt") &&
1273: strcmp (suffix, "css") &&
1274: strcmp (suffix, "eps"))
1.22 cvs 1275: return (FALSE);
1276: else
1277: return (TRUE);
1278: }
1279: else
1280: return (FALSE);
1281: }
1282:
1283:
1284: /*----------------------------------------------------------------------
1.24 cvs 1285: ChopURL
1286: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1287: If inputURL is bigger than that size, outputURL receives
1288: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1289: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1290: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1291: copied into outputURL.
1292: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1293: chars.
1294: ----------------------------------------------------------------------*/
1.106 cvs 1295: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1296: {
1297: int len;
1.9 cvs 1298:
1.106 cvs 1299: len = strlen (inputURL);
1.24 cvs 1300: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1301: strcpy (outputURL, inputURL);
1.24 cvs 1302: else
1303: /* make a truncated urlName on the status window */
1304: {
1.106 cvs 1305: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1306: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1307: strcat (outputURL, "...");
1308: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1309: }
1.25 cvs 1310: }
1311:
1312:
1313: /*----------------------------------------------------------------------
1314: scan
1.47 cvs 1315: Scan a filename for its constituents
1.25 cvs 1316: -----------------------------------
1317:
1318: On entry,
1319: name points to a document name which may be incomplete.
1320: On exit,
1321: absolute or relative may be nonzero (but not both).
1322: host, fragment and access may be nonzero if they were specified.
1323: Any which are nonzero point to zero terminated strings.
1324: ----------------------------------------------------------------------*/
1.106 cvs 1325: static void scan (char *name, HTURI *parts)
1.25 cvs 1326: {
1.106 cvs 1327: char * p;
1328: char * after_access = name;
1.32 cvs 1329:
1.43 cvs 1330: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1331: /* Look for fragment identifier */
1.106 cvs 1332: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1333: {
1.106 cvs 1334: *p++ = '\0';
1.28 cvs 1335: parts->fragment = p;
1.25 cvs 1336: }
1337:
1.28 cvs 1338: for (p=name; *p; p++)
1339: {
1.106 cvs 1340: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1341: break;
1.106 cvs 1342: if (*p == ':')
1.28 cvs 1343: {
1344: *p = 0;
1345: parts->access = after_access; /* Scheme has been specified */
1346:
1347: /* The combination of gcc, the "-O" flag and the HP platform is
1348: unhealthy. The following three lines is a quick & dirty fix, but is
1349: not recommended. Rather, turn off "-O". */
1350:
1351: /* after_access = p;*/
1352: /* while (*after_access == 0)*/
1353: /* after_access++;*/
1354: after_access = p+1;
1.106 cvs 1355: if (!strcasecmp("URL", parts->access))
1.28 cvs 1356: /* Ignore IETF's URL: pre-prefix */
1357: parts->access = NULL;
1358: else
1.25 cvs 1359: break;
1360: }
1361: }
1362:
1363: p = after_access;
1.43 cvs 1364: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1365: {
1.43 cvs 1366: if (p[1] == URL_SEP)
1.28 cvs 1367: {
1.25 cvs 1368: parts->host = p+2; /* host has been specified */
1.28 cvs 1369: *p = 0; /* Terminate access */
1370: /* look for end of host name if any */
1.106 cvs 1371: p = strchr (parts->host, URL_SEP);
1.28 cvs 1372: if (p)
1373: {
1.106 cvs 1374: *p = EOS; /* Terminate host */
1.25 cvs 1375: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1376: }
1377: }
1378: else
1379: /* Root found but no host */
1380: parts->absolute = p+1;
1381: }
1382: else
1383: {
1.25 cvs 1384: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1385: }
1.25 cvs 1386: }
1387:
1388:
1389: /*----------------------------------------------------------------------
1.28 cvs 1390: AmayaParseUrl: parse a Name relative to another name
1391:
1392: This returns those parts of a name which are given (and requested)
1393: substituting bits from the related name where necessary.
1.25 cvs 1394:
1.28 cvs 1395: On entry,
1.25 cvs 1396: aName A filename given
1397: relatedName A name relative to which aName is to be parsed. Give
1398: it an empty string if aName is absolute.
1399: wanted A mask for the bits which are wanted.
1400:
1.28 cvs 1401: On exit,
1.25 cvs 1402: returns A pointer to a malloc'd string which MUST BE FREED
1403: ----------------------------------------------------------------------*/
1.106 cvs 1404: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1405: {
1406: char *return_value;
1407: char result[MAX_LENGTH];
1408: char name[MAX_LENGTH];
1409: char rel[MAX_LENGTH];
1410: char *p, *access;
1.29 cvs 1411: HTURI given, related;
1412: int len;
1.106 cvs 1413: char used_sep;
1414: char *used_str;
1.32 cvs 1415:
1.106 cvs 1416: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1417: {
1.106 cvs 1418: used_str = DIR_STR;
1419: used_sep = DIR_SEP;
1.33 cvs 1420: }
1.32 cvs 1421: else
1.33 cvs 1422: {
1.106 cvs 1423: used_str = URL_STR;
1424: used_sep = URL_SEP;
1.33 cvs 1425: }
1.32 cvs 1426:
1.29 cvs 1427: /* Make working copies of input strings to cut up: */
1428: return_value = NULL;
1429: result[0] = 0; /* Clear string */
1.106 cvs 1430: strcpy (name, aName);
1.29 cvs 1431: if (relatedName != NULL)
1.106 cvs 1432: strcpy (rel, relatedName);
1.29 cvs 1433: else
1.106 cvs 1434: relatedName[0] = EOS;
1.29 cvs 1435:
1436: scan (name, &given);
1437: scan (rel, &related);
1438: access = given.access ? given.access : related.access;
1439: if (wanted & AMAYA_PARSE_ACCESS)
1440: if (access)
1441: {
1.106 cvs 1442: strcat (result, access);
1.29 cvs 1443: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1444: strcat (result, ":");
1.29 cvs 1445: }
1446:
1447: if (given.access && related.access)
1448: /* If different, inherit nothing. */
1.106 cvs 1449: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1450: {
1451: related.host = 0;
1452: related.absolute = 0;
1453: related.relative = 0;
1454: related.fragment = 0;
1455: }
1456:
1457: if (wanted & AMAYA_PARSE_HOST)
1458: if(given.host || related.host)
1459: {
1460: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1461: strcat (result, "//");
1462: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1463: }
1464:
1465: if (given.host && related.host)
1466: /* If different hosts, inherit no path. */
1.106 cvs 1467: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1468: {
1469: related.absolute = 0;
1470: related.relative = 0;
1471: related.fragment = 0;
1472: }
1473:
1474: if (wanted & AMAYA_PARSE_PATH)
1475: {
1476: if (given.absolute)
1477: {
1478: /* All is given */
1479: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1480: strcat (result, used_str);
1481: strcat (result, given.absolute);
1.25 cvs 1482: }
1.29 cvs 1483: else if (related.absolute)
1484: {
1485: /* Adopt path not name */
1.106 cvs 1486: strcat (result, used_str);
1487: strcat (result, related.absolute);
1.29 cvs 1488: if (given.relative)
1489: {
1490: /* Search part? */
1.106 cvs 1491: p = strchr (result, '?');
1.29 cvs 1492: if (!p)
1.106 cvs 1493: p=result+strlen(result)-1;
1.33 cvs 1494: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1495: /* Remove filename */
1496: p[1]=0;
1497: /* Add given one */
1.106 cvs 1498: strcat (result, given.relative);
1.25 cvs 1499: }
1500: }
1.29 cvs 1501: else if (given.relative)
1502: /* what we've got */
1.106 cvs 1503: strcat (result, given.relative);
1.29 cvs 1504: else if (related.relative)
1.106 cvs 1505: strcat (result, related.relative);
1.29 cvs 1506: else
1507: /* No inheritance */
1.106 cvs 1508: strcat (result, used_str);
1.25 cvs 1509: }
1.29 cvs 1510:
1511: if (wanted & AMAYA_PARSE_ANCHOR)
1512: if (given.fragment || related.fragment)
1513: {
1514: if (given.absolute && given.fragment)
1515: {
1516: /*Fixes for relURLs...*/
1517: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1518: strcat (result, "#");
1519: strcat (result, given.fragment);
1.29 cvs 1520: }
1521: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1522: strcat (result, "");
1.29 cvs 1523: else
1524: {
1.110 cvs 1525: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1526: strcat (result, "#");
1.110 cvs 1527: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1528: }
1529: }
1.106 cvs 1530: len = strlen (result);
1531: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1532: strcpy (return_value, result);
1.29 cvs 1533: return (return_value); /* exactly the right length */
1.25 cvs 1534: }
1535:
1536: /*----------------------------------------------------------------------
1537: HTCanon
1538: Canonicalizes the URL in the following manner starting from the host
1539: pointer:
1540:
1541: 1) The host name is converted to lowercase
1542: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1543:
1544: Return: OK The position of the current path part of the URL
1545: which might be the old one or a new one.
1546:
1547: ----------------------------------------------------------------------*/
1.106 cvs 1548: static char *HTCanon (char **filename, char *host)
1549: {
1550: char *newname = NULL;
1551: char used_sep;
1552: char *path;
1553: char *strptr;
1554: char *port;
1555: char *access = host-3;
1556:
1557: if (*filename && strchr (*filename, URL_SEP))
1558: used_sep = URL_SEP;
1.33 cvs 1559: else
1.106 cvs 1560: used_sep = DIR_SEP;
1.32 cvs 1561:
1.110 cvs 1562: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1563: access--;
1.110 cvs 1564: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1565: path = host + strlen (host);
1566: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1567: host = strptr;
1.110 cvs 1568: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1569: port = NULL;
1.25 cvs 1570:
1571: strptr = host; /* Convert to lower-case */
1.82 cvs 1572: while (strptr < path)
1.33 cvs 1573: {
1.84 cvs 1574: *strptr = utolower (*strptr);
1.82 cvs 1575: strptr++;
1.33 cvs 1576: }
1.25 cvs 1577:
1578: /* Does the URL contain a full domain name? This also works for a
1579: numerical host name. The domain name is already made lower-case
1580: and without a trailing dot. */
1581: {
1.106 cvs 1582: char *dot = port ? port : path;
1583: if (dot > *filename && *--dot == '.')
1.33 cvs 1584: {
1.106 cvs 1585: char *orig = dot;
1586: char *dest = dot + 1;
1.82 cvs 1587: while ((*orig++ = *dest++));
1588: if (port) port--;
1.33 cvs 1589: path--;
1.25 cvs 1590: }
1591: }
1592: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1593: if (port)
1594: {
1.82 cvs 1595: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1596: {
1597: if (!newname)
1598: {
1.106 cvs 1599: char *orig = port;
1600: char *dest = port + 1;
1.82 cvs 1601: while ((*orig++ = *dest++));
1.33 cvs 1602: }
1603: }
1.106 cvs 1604: else if ((!strncmp (access, "http", 4) &&
1605: (*(port + 1) == '8' &&
1606: *(port+2) == '0' &&
1.82 cvs 1607: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1608: (!strncmp (access, "gopher", 6) &&
1609: (*(port+1) == '7' &&
1610: *(port+2) == '0' &&
1.82 cvs 1611: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1612: (!strncmp (access, "ftp", 3) &&
1613: (*(port+1) == '2' &&
1614: *(port + 2) == '1' &&
1.82 cvs 1615: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1616: if (!newname)
1617: {
1.106 cvs 1618: char *orig = port;
1619: char *dest = port + 3;
1.33 cvs 1620: while((*orig++ = *dest++));
1621: /* Update path position, Henry Minsky */
1622: path -= 3;
1.25 cvs 1623: }
1.33 cvs 1624: }
1625: else if (newname)
1.106 cvs 1626: strncat (newname, port, (int) (path - port));
1.33 cvs 1627: }
1.25 cvs 1628:
1.33 cvs 1629: if (newname)
1630: {
1.106 cvs 1631: char *newpath = newname + strlen (newname);
1632: strcat (newname, path);
1.25 cvs 1633: path = newpath;
1.28 cvs 1634: /* Free old copy */
1635: TtaFreeMemory(*filename);
1.25 cvs 1636: *filename = newname;
1.33 cvs 1637: }
1.25 cvs 1638: return path;
1639: }
1640:
1641:
1642: /*----------------------------------------------------------------------
1.29 cvs 1643: SimplifyUrl: simplify a URI
1.32 cvs 1644: A URI is allowed to contain the sequence xxx/../ which may be
1645: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1646: Simplification helps us recognize duplicate URIs.
1.25 cvs 1647:
1.28 cvs 1648: Thus, /etc/junk/../fred becomes /etc/fred
1649: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1650:
1.28 cvs 1651: but we should NOT change
1652: http://fred.xxx.edu/../..
1.25 cvs 1653:
1654: or ../../albert.html
1655:
1.28 cvs 1656: In order to avoid empty URLs the following URLs become:
1.25 cvs 1657:
1658: /fred/.. becomes /fred/..
1659: /fred/././.. becomes /fred/..
1660: /fred/.././junk/.././ becomes /fred/..
1661:
1.28 cvs 1662: If more than one set of `://' is found (several proxies in cascade) then
1663: only the part after the last `://' is simplified.
1.25 cvs 1664:
1.28 cvs 1665: Returns: A string which might be the old one or a new one.
1.25 cvs 1666: ----------------------------------------------------------------------*/
1.106 cvs 1667: void SimplifyUrl (char **url)
1668: {
1669: char *path;
1670: char *access;
1671: char *newptr;
1672: char *p;
1673: char *orig, *dest, *end;
1.28 cvs 1674:
1.106 cvs 1675: char used_sep;
1.77 cvs 1676: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1677: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1678:
1679:
1.28 cvs 1680: if (!url || !*url)
1681: return;
1682:
1.106 cvs 1683: if (strchr (*url, URL_SEP))
1684: used_sep = URL_SEP;
1.32 cvs 1685: else
1.106 cvs 1686: used_sep = DIR_SEP;
1.32 cvs 1687:
1.77 cvs 1688: /* should we simplify double dot? */
1689: path = *url;
1.106 cvs 1690: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1691: ddot_simplify = FALSE;
1692: else
1693: ddot_simplify = TRUE;
1694:
1.28 cvs 1695: /* Find any scheme name */
1.106 cvs 1696: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1697: {
1698: /* Find host name */
1.28 cvs 1699: access = *url;
1.84 cvs 1700: while (access < path && (*access = utolower (*access)))
1.82 cvs 1701: access++;
1.28 cvs 1702: path += 3;
1.106 cvs 1703: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1704: /* For proxies */
1.106 cvs 1705: path = newptr + 3;
1.82 cvs 1706: /* We have a host name */
1.84 cvs 1707: path = HTCanon (url, path);
1.25 cvs 1708: }
1.106 cvs 1709: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1710: path += 2;
1711: else
1712: path = *url;
1.25 cvs 1713:
1.84 cvs 1714: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1715: /* Some URLs start //<foo> */
1716: path += 1;
1.94 cvs 1717: else if (IsFilePath (path))
1718: {
1719: /* doesn't need to do anything more */
1720: return;
1721: }
1.106 cvs 1722: else if (!strncmp (path, "news:", 5))
1.28 cvs 1723: {
1.106 cvs 1724: newptr = strchr (path+5, '@');
1.28 cvs 1725: if (!newptr)
1726: newptr = path + 5;
1727: while (*newptr)
1728: {
1729: /* Make group or host lower case */
1.84 cvs 1730: *newptr = utolower (*newptr);
1.28 cvs 1731: newptr++;
1.25 cvs 1732: }
1.28 cvs 1733: /* Doesn't need to do any more */
1734: return;
1.25 cvs 1735: }
1.28 cvs 1736:
1737: if ((p = path))
1738: {
1.106 cvs 1739: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1740: (end = strchr (path, '#'))))
1741: end = path + strlen (path);
1.28 cvs 1742:
1743: /* Parse string second time to simplify */
1744: p = path;
1745: while (p < end)
1746: {
1.110 cvs 1747: /* if we're pointing to a char, it's safe to reactivate the
1748: ../ convertion */
1.106 cvs 1749: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1750: ddot_simplify = TRUE;
1751:
1.33 cvs 1752: if (*p==used_sep)
1.28 cvs 1753: {
1.106 cvs 1754: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1755: {
1756: orig = p + 1;
1.84 cvs 1757: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1758: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1759: end = orig - 1;
1760: }
1.106 cvs 1761: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1762: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1763: {
1764: newptr = p;
1.52 cvs 1765: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1766: if (*newptr == used_sep)
1767: orig = newptr + 1;
1.28 cvs 1768: else
1.52 cvs 1769: orig = newptr;
1770:
1771: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1772: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1773: end = orig-1;
1774: /* Start again with prev slash */
1775: p = newptr;
1.28 cvs 1776: }
1.33 cvs 1777: else if (*(p+1) == used_sep)
1.28 cvs 1778: {
1.33 cvs 1779: while (*(p+1) == used_sep)
1.28 cvs 1780: {
1781: orig = p;
1782: dest = p + 1;
1783: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1784: end = orig-1;
1785: }
1786: }
1787: else
1.25 cvs 1788: p++;
1.28 cvs 1789: }
1790: else
1791: p++;
1.25 cvs 1792: }
1793: }
1.51 cvs 1794:
1795: /*
1796: ** Check for host/../.. kind of things
1797: */
1.106 cvs 1798: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1799: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 1800: *(path+1) = EOS;
1.51 cvs 1801:
1.28 cvs 1802: return;
1803: }
1804:
1805:
1806: /*----------------------------------------------------------------------
1.96 cvs 1807: NormalizeFile normalizes local names.
1.28 cvs 1808: Return TRUE if target and src differ.
1809: ----------------------------------------------------------------------*/
1.106 cvs 1810: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 1811: {
1.110 cvs 1812: #ifndef _WINDOWS
1.106 cvs 1813: char *s;
1.93 cvs 1814: int i;
1.110 cvs 1815: #endif /* !_WINDOWS */
1.82 cvs 1816: ThotBool change;
1.90 cvs 1817: int start_index; /* the first char that we'll copy */
1.28 cvs 1818:
1.54 cvs 1819: change = FALSE;
1.90 cvs 1820: start_index = 0;
1821:
1.106 cvs 1822: if (!src || src[0] == EOS)
1.96 cvs 1823: {
1.106 cvs 1824: target[0] = EOS;
1.96 cvs 1825: return FALSE;
1826: }
1.90 cvs 1827:
1828: /* @@ do I need file: or file:/ here? */
1.106 cvs 1829: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1830: {
1.90 cvs 1831: /* remove the prefix file: */
1832: start_index += 5;
1833:
1834: /* remove the localhost prefix */
1.106 cvs 1835: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 1836: start_index += 11;
1837:
1838: /* remove the first two slashes in / / /path */
1839: while (src[start_index] &&
1.106 cvs 1840: src[start_index] == '/'
1841: && src[start_index + 1] == '/')
1.94 cvs 1842: start_index++;
1843:
1844: #ifdef _WINDOWS
1845: /* remove any extra slash before the drive name */
1.106 cvs 1846: if (src[start_index] == '/'
1847: &&src[start_index+2] == ':')
1.94 cvs 1848: start_index++;
1849: #endif /* _WINDOWS */
1.90 cvs 1850:
1.106 cvs 1851: if (src[start_index] == EOS)
1.90 cvs 1852: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 1853: strcpy (target, DIR_STR);
1.90 cvs 1854: else
1.97 cvs 1855: /* as we're inside a file: URL, we'll apply all the convertions
1856: we know */
1857: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1858:
1859: change = TRUE;
1860: }
1.97 cvs 1861: else if (convertion != AM_CONV_NONE)
1.96 cvs 1862: {
1863: /* we are following a "local" relative link, we do all the
1864: convertions except for the HOME_DIR ~ one */
1.97 cvs 1865: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1866: }
1.90 cvs 1867: #ifndef _WINDOWS
1.106 cvs 1868: else if (src[0] == '~')
1.53 cvs 1869: {
1.96 cvs 1870: /* it must be a URL typed in a text input field */
1871: /* do the HOME_DIR ~ substitution */
1.82 cvs 1872: s = TtaGetEnvString ("HOME");
1.106 cvs 1873: strcpy (target, s);
1.90 cvs 1874: #if 0
1.96 cvs 1875: /* JK: invalidated this part of the code as it's simpler
1876: to add the DIR_SEP whenever we have something to add
1877: to the path rather than adding it systematically */
1.106 cvs 1878: if (src[1] != DIR_SEP)
1879: strcat (target, DIR_STR);
1.90 cvs 1880: #endif
1.106 cvs 1881: i = strlen (target);
1882: strcpy (&target[i], &src[1]);
1.54 cvs 1883: change = TRUE;
1.53 cvs 1884: }
1.90 cvs 1885: #endif /* _WINDOWS */
1.28 cvs 1886: else
1.96 cvs 1887: /* leave it as it is */
1.106 cvs 1888: strcpy (target, src);
1.96 cvs 1889:
1.28 cvs 1890: /* remove /../ and /./ */
1.29 cvs 1891: SimplifyUrl (&target);
1.54 cvs 1892: if (!change)
1.106 cvs 1893: change = strcmp (src, target);
1.28 cvs 1894: return (change);
1.25 cvs 1895: }
1896:
1.28 cvs 1897:
1.25 cvs 1898: /*----------------------------------------------------------------------
1.31 cvs 1899: MakeRelativeURL: make relative name
1.25 cvs 1900:
1.28 cvs 1901: This function creates and returns a string which gives an expression of
1902: one address as related to another. Where there is no relation, an absolute
1903: address is retured.
1.25 cvs 1904:
1.28 cvs 1905: On entry,
1.25 cvs 1906: Both names must be absolute, fully qualified names of nodes
1907: (no fragment bits)
1908:
1.28 cvs 1909: On exit,
1.25 cvs 1910: The return result points to a newly allocated name which, if
1911: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1912: The caller is responsible for freeing the resulting name later.
1913: ----------------------------------------------------------------------*/
1.106 cvs 1914: char *MakeRelativeURL (char *aName, char *relatedName)
1915: {
1916: char *return_value;
1917: char result[MAX_LENGTH];
1918: char *p;
1919: char *q;
1920: char *after_access;
1921: char *last_slash = NULL;
1922: int slashes, levels, len;
1.110 cvs 1923: #ifdef _WINDOWS
1.44 cvs 1924: int ndx;
1.110 cvs 1925: #endif /* _WINDOWS */
1.44 cvs 1926:
1.29 cvs 1927: if (aName == NULL || relatedName == NULL)
1928: return (NULL);
1929:
1930: slashes = 0;
1931: after_access = NULL;
1932: p = aName;
1933: q = relatedName;
1934: for (; *p && (*p == *q); p++, q++)
1.27 cvs 1935: {
1936: /* Find extent of match */
1.106 cvs 1937: if (*p == ':')
1.29 cvs 1938: after_access = p + 1;
1.28 cvs 1939: if (*p == DIR_SEP)
1.27 cvs 1940: {
1.29 cvs 1941: /* memorize the last slash position and count them */
1.27 cvs 1942: last_slash = p;
1943: slashes++;
1.25 cvs 1944: }
1945: }
1946:
1.31 cvs 1947: /* q, p point to the first non-matching character or zero */
1.106 cvs 1948: if (*q == EOS)
1.31 cvs 1949: {
1950: /* New name is a subset of the related name */
1951: /* exactly the right length */
1.106 cvs 1952: len = strlen (p);
1953: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1954: strcpy (return_value, p);
1.31 cvs 1955: }
1956: else if ((slashes < 2 && after_access == NULL)
1957: || (slashes < 3 && after_access != NULL))
1958: {
1959: /* Two names whitout common path */
1960: /* exactly the right length */
1.106 cvs 1961: len = strlen (aName);
1962: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1963: strcpy (return_value, aName);
1.31 cvs 1964: }
1965: else
1966: {
1967: /* Some path in common */
1.106 cvs 1968: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 1969: /* just the same server */
1.106 cvs 1970: strcpy (result, last_slash);
1.31 cvs 1971: else
1972: {
1973: levels= 0;
1.106 cvs 1974: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 1975: if (*q == DIR_SEP)
1976: levels++;
1977:
1.106 cvs 1978: result[0] = EOS;
1.31 cvs 1979: for (;levels; levels--)
1.106 cvs 1980: strcat (result, "../");
1981: strcat (result, last_slash+1);
1.31 cvs 1982: }
1.52 cvs 1983:
1984: if (!*result)
1.106 cvs 1985: strcat (result, "./");
1.52 cvs 1986:
1.31 cvs 1987: /* exactly the right length */
1.106 cvs 1988: len = strlen (result);
1989: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1990: strcpy (return_value, result);
1.52 cvs 1991:
1.25 cvs 1992: }
1.110 cvs 1993: #ifdef _WINDOWS
1.106 cvs 1994: len = strlen (return_value);
1.44 cvs 1995: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 1996: if (return_value[ndx] == '\\')
1997: return_value[ndx] = '/' ;
1.110 cvs 1998: #endif /* _WINDOWS */
1.29 cvs 1999: return (return_value);
1.24 cvs 2000: }
1.35 cvs 2001:
1.104 kahan 2002: /*----------------------------------------------------------------------
2003: AM_GetFileSize
2004: Returns TRUE and the filesize in the 2nd parameter.
2005: Otherwise, in case of a system error, returns FALSE, with a
2006: filesize of 0L.
2007: ---------------------------------------------------------------------*/
1.106 cvs 2008: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2009: {
1.106 cvs 2010: ThotFileHandle handle = ThotFile_BADHANDLE;
2011: ThotFileInfo info;
1.35 cvs 2012:
1.104 kahan 2013: *file_size = 0L;
2014: if (!TtaFileExist (filename))
2015: return FALSE;
2016:
2017: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2018: if (handle == ThotFile_BADHANDLE)
2019: /* ThotFile_BADHANDLE */
2020: return FALSE;
2021: if (TtaFileStat (handle, &info) == 0)
2022: /* bad stat */
2023: info.size = 0L;
2024: TtaFileClose (handle);
2025: *file_size = (unsigned long) info.size;
2026: return TRUE;
2027: }
Webmaster
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to AHTURLTools.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Public] / Amaya / amaya