Annotation of Amaya/amaya/AHTURLTools.c, revision 1.141
1.7 cvs 1: /*
2: *
1.133 vatton 3: * (c) COPYRIGHT MIT and INRIA, 1996-2002
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: ----------------------------------------------------------------------*/
1.124 vatton 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++)
1.123 vatton 48: string[i] = tolower (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 0x20: /* space */
1.140 kahan 103: case 0x26: /* & */
104: case 0x27: /* antislash */
1.75 cvs 105: new_chars = 3;
106: break;
107:
108: default:
1.122 kahan 109: if ((unsigned char )*ptr > 127)
110: new_chars = 3;
111: else
112: new_chars = 1;
1.75 cvs 113: break;
114: }
115:
116: /* see if we need extra room in the buffer */
117: if (new_chars > buffer_free_mem)
118: {
1.76 cvs 119: buffer_free_mem = 20;
1.106 cvs 120: status = TtaRealloc (buffer, sizeof (char)
1.75 cvs 121: * (buffer_len + buffer_free_mem + 1));
122: if (status)
1.114 cvs 123: buffer = (char *) status;
1.106 cvs 124: else
125: {
126: /* @@ maybe we should do some other behavior here, like
127: freeing the buffer and return a void thing */
128: buffer[buffer_len] = EOS;
129: break;
130: }
1.75 cvs 131: }
132: /* escape the char */
1.140 kahan 133: if (new_chars == 3)
134: {
1.106 cvs 135: buffer[buffer_len] = '%';
1.75 cvs 136: EscapeChar (&buffer[buffer_len+1], *ptr);
137: }
1.140 kahan 138: else
139: buffer[buffer_len] = *ptr;
140:
141: /* update the status */
142: buffer_len += new_chars;
143: buffer_free_mem -= new_chars;
144: /* examine the next char */
145: ptr++;
146: }
147: buffer[buffer_len] = EOS;
148: }
149: else
150: buffer = NULL;
151:
152: return (buffer);
153: }
154:
155: /*----------------------------------------------------------------------
156: EscapeXML
157: Takes a string and escapes all protected chars into entity
158: sequences.
159: Returns either NULL or a new buffer, which must be freed by the caller
160: ----------------------------------------------------------------------*/
161: char *EscapeXML (const char *string)
162: {
163: char *buffer;
164: int buffer_len;
165: int buffer_free_mem;
166: char *ptr;
167: char *entity = NULL;
168: int new_chars;
169: void *status;
170:
171: if (string && *string)
172: {
173: buffer_free_mem = strlen (string) + 20;
174: buffer = TtaGetMemory (buffer_free_mem + 1);
175: ptr = (char *) string;
176: buffer_len = 0;
177:
178: while (*ptr)
179: {
180: switch (*ptr)
181: {
182: case 0x26: /* & */
183: entity = "&";
1.141 ! kahan 184: new_chars = sizeof (entity) - 1;
! 185: break;
1.140 kahan 186:
1.141 ! kahan 187: case '<': /* < */
1.140 kahan 188: entity = "<";
1.141 ! kahan 189: new_chars = sizeof (entity) - 1;
1.140 kahan 190: break;
191:
1.141 ! kahan 192: case '>': /* > */
1.140 kahan 193: entity = ">";
1.141 ! kahan 194: new_chars = sizeof (entity) - 1;
! 195: break;
! 196:
! 197: case '"': /* "e; */
! 198: entity = ""e;";
! 199: new_chars = sizeof (entity) - 1;
1.140 kahan 200: break;
201:
202: default:
203: new_chars = 1;
204: break;
205: }
206:
207: /* see if we need extra room in the buffer */
208: if (new_chars > buffer_free_mem)
209: {
210: buffer_free_mem = 20;
211: status = TtaRealloc (buffer, sizeof (char)
212: * (buffer_len + buffer_free_mem + 1));
213: if (status)
214: buffer = (char *) status;
215: else
216: {
217: /* @@ maybe we should do some other behavior here, like
218: freeing the buffer and return a void thing */
219: buffer[buffer_len] = EOS;
220: break;
221: }
222: }
223: /* escape the char */
224: if (entity)
225: {
226: sprintf (&buffer[buffer_len], "%s", entity);
227: entity = NULL;
228: }
1.75 cvs 229: else
230: buffer[buffer_len] = *ptr;
231:
232: /* update the status */
233: buffer_len += new_chars;
234: buffer_free_mem -= new_chars;
235: /* examine the next char */
236: ptr++;
237: }
1.106 cvs 238: buffer[buffer_len] = EOS;
1.75 cvs 239: }
1.76 cvs 240: else
241: buffer = NULL;
242:
1.75 cvs 243: return (buffer);
1.122 kahan 244: }
245:
246: /*----------------------------------------------------------------------
247: URLToUTF8
248: If such convertion is needed, returns a converted string that the
249: user must free.
250: ----------------------------------------------------------------------*/
251: char *URLToUTF8 (char *url, Document doc)
252: {
253: unsigned char *tmp;
254:
255: if (!url || *url == EOS)
256: return NULL;
257:
258: /* does the URL contain chars > 127 ? */
259: tmp = url;
260: while (*tmp)
261: {
262: if (*tmp > 127)
263: break;
264: tmp++;
265: }
266:
267: if (*tmp == EOS)
268: return NULL; /* no such chars found */
269:
270: tmp = TtaConvertIsoToMbs (url, TtaGetDocumentCharset (doc));
271:
272: return tmp;
1.75 cvs 273: }
274:
275:
276: /*----------------------------------------------------------------------
1.11 cvs 277: ExplodeURL
1.8 cvs 278: ----------------------------------------------------------------------*/
1.106 cvs 279: void ExplodeURL (char *url, char **proto, char **host, char **dir,
280: char **file)
1.8 cvs 281: {
1.33 cvs 282: char *curr, *temp;
283: char used_sep;
1.32 cvs 284:
1.33 cvs 285: if (url && strchr (url, URL_SEP))
286: used_sep = URL_SEP;
287: else
288: used_sep = DIR_SEP;
1.8 cvs 289:
290: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
291: (dir == NULL) || (file == NULL))
292: return;
293:
294: /* initialize every pointer */
295: *proto = *host = *dir = *file = NULL;
296:
297: /* skip any leading space */
298: while ((*url == SPACE) || (*url == TAB))
299: url++;
1.9 cvs 300: curr = url;
301: if (*curr == 0)
1.8 cvs 302: goto finished;
303:
304: /* go to the end of the URL */
1.68 cvs 305: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
306: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 307: curr++;
1.8 cvs 308:
309: /* mark the end of the chain */
1.9 cvs 310: *curr = EOS;
311: curr--;
312: if (curr <= url)
1.8 cvs 313: goto finished;
314:
315: /* search the next DIR_SEP indicating the beginning of the file name */
316: do
1.11 cvs 317: curr--;
1.33 cvs 318: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 319:
1.9 cvs 320: if (curr < url)
1.8 cvs 321: goto finished;
1.9 cvs 322: *file = curr + 1;
1.8 cvs 323:
324: /* mark the end of the dir */
1.9 cvs 325: *curr = EOS;
326: curr--;
327: if (curr < url)
1.8 cvs 328: goto finished;
329:
1.29 cvs 330: /* search for the DIR_STR indicating the host name start */
1.33 cvs 331: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 332: curr--;
1.8 cvs 333:
334: /* if we found it, separate the host name from the directory */
1.102 kahan 335: if ((*curr == used_sep) && (*(curr + 1) == used_sep))
1.8 cvs 336: {
1.9 cvs 337: *host = temp = curr + 2;
1.33 cvs 338: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 339: temp++;
1.33 cvs 340: if (*temp == used_sep)
1.8 cvs 341: {
342: *temp = EOS;
343: *dir = temp + 1;
344: }
345: }
346: else
1.11 cvs 347: *dir = curr;
348:
1.9 cvs 349: if (curr <= url)
1.8 cvs 350: goto finished;
351:
352: /* mark the end of the proto */
1.9 cvs 353: *curr = EOS;
354: curr--;
355: if (curr < url)
1.8 cvs 356: goto finished;
357:
1.106 cvs 358: if (*curr == ':')
1.8 cvs 359: {
1.9 cvs 360: *curr = EOS;
361: curr--;
1.8 cvs 362: }
363: else
364: goto finished;
1.11 cvs 365:
1.9 cvs 366: if (curr < url)
1.8 cvs 367: goto finished;
1.9 cvs 368: while ((curr > url) && (isalpha (*curr)))
369: curr--;
370: *proto = curr;
1.8 cvs 371:
372: finished:;
373:
374: #ifdef AMAYA_DEBUG
375: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
376: if (*proto)
377: fprintf (stderr, "proto : %s, ", *proto);
378: if (*host)
379: fprintf (stderr, "host : %s, ", *host);
380: if (*dir)
381: fprintf (stderr, "dir : %s, ", *dir);
382: if (*file)
383: fprintf (stderr, "file : %s ", *file);
384: fprintf (stderr, "\n");
385: #endif
386:
387: }
1.3 cvs 388:
1.116 kahan 389: /*----------------------------------------------------------------------
390: PicTypeToMime
391: Converts a Thot PicType into the equivalent MIME type. If no convertion
392: is possible, it returns NULL.
393: ----------------------------------------------------------------------*/
394: char *PicTypeToMIME (PicType contentType)
395: {
396: char *mime_type;
397:
398: switch (contentType)
399: {
400: case xbm_type:
401: mime_type ="image/x-xbitmap";
402: break;
403: case eps_type:
404: mime_type ="application/postscript";
405: break;
406: case xpm_type:
407: mime_type ="image/x-xpicmap";
408: break;
409: case gif_type:
410: mime_type ="image/gif";
411: break;
412: case jpeg_type:
413: mime_type ="image/jpeg";
414: break;
415: case png_type:
416: mime_type ="image/png";
417: break;
418: case svg_type:
1.138 kahan 419: mime_type =AM_SVG_MIME_TYPE;
1.116 kahan 420: break;
421: case unknown_type:
422: default:
423: mime_type = NULL;
424: }
425:
426: return mime_type;
427: }
1.61 cvs 428:
429: /*----------------------------------------------------------------------
1.117 kahan 430: ImageElement
431: Returns the element (image parameter) and URL (url parameter) of an
432: image in a docImage document. The user must free the memory associated
1.120 kahan 433: with the url parameter if the function is succesful.
434: If the url parameter is NULL, we won't initialize it.
1.117 kahan 435: Returns TRUE if succesful, FALSE otherwise.
436: ----------------------------------------------------------------------*/
437: ThotBool ImageElement (Document doc, char **url, Element *image)
438: {
439: Element el, imgEl;
440: Attribute attr, srcAttr;
441: AttributeType attrType;
442: int length;
443: char *value;
444:
445: if (DocumentTypes[doc] != docImage)
446: return FALSE;
447:
448: /* find the value of the src attribute */
449: attrType.AttrSSchema = TtaGetSSchema ("HTML", doc);
450: attrType.AttrTypeNum = HTML_ATTR_SRC;
451: el = TtaGetRootElement (doc);
452: TtaSearchAttribute (attrType, SearchInTree, el, &imgEl, &srcAttr);
453:
454: if (!imgEl)
455: return FALSE;
456: *image = imgEl;
457:
1.120 kahan 458: if (url)
459: {
460: attr = TtaGetAttribute (imgEl, attrType);
461: length = TtaGetTextAttributeLength (srcAttr) + 1;
462: value = TtaGetMemory (length);
463: TtaGiveTextAttributeValue (srcAttr, value, &length);
464: *url = value;
465: }
1.117 kahan 466: return TRUE;
467: }
468:
469: /*----------------------------------------------------------------------
470: DocImageMimeType
471: Returns the MIME type of a docImage document.
472: ----------------------------------------------------------------------*/
473: char *DocImageMimeType (Document doc)
474: {
475: char *mime_type;
476: LoadedImageDesc *pImage;
477: PicType type;
478: Element image;
479:
480: if (DocumentTypes[doc] != docImage)
481: return NULL;
482:
483: mime_type = NULL;
484: if (!IsHTTPPath (DocumentURLs[doc]))
485: {
486: /* it is a local image */
1.120 kahan 487: if (ImageElement (doc, NULL, &image))
1.117 kahan 488: {
489: type = TtaGetPictureType (image);
490: mime_type = PicTypeToMIME (type);
491: }
492: }
493: else
494: {
495: /* find the value of the src attribute */
496: pImage = ImageURLs;
497: while (pImage != NULL)
498: {
499: if (pImage->document == doc)
500: {
501: if (pImage->content_type)
502: mime_type = pImage->content_type;
503: else if (pImage->elImage && pImage->elImage->currentElement)
504: {
505: type = TtaGetPictureType (pImage->elImage->currentElement);
506: mime_type = PicTypeToMIME (type);
507: }
508: break;
509: }
510: }
511: }
512: return (mime_type);
513: }
514:
1.4 cvs 515: /*----------------------------------------------------------------------
1.9 cvs 516: IsHTMLName
517: returns TRUE if path points to an HTML resource.
1.4 cvs 518: ----------------------------------------------------------------------*/
1.109 cvs 519: ThotBool IsHTMLName (const char *path)
1.106 cvs 520: {
1.136 cvs 521: char temppath[MAX_LENGTH];
522: char suffix[MAX_LENGTH];
523: char nsuffix[MAX_LENGTH];
524: int i;
1.5 cvs 525:
1.101 cvs 526: if (!path)
527: return (FALSE);
1.5 cvs 528:
1.106 cvs 529: strcpy (temppath, path);
1.124 vatton 530: TtaExtractSuffix (temppath, suffix);
1.101 cvs 531: i = 0;
1.106 cvs 532: while (suffix[i] != EOS)
1.101 cvs 533: {
534: /* Normalize the suffix */
535: i = 0;
1.106 cvs 536: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 537: {
1.123 vatton 538: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 539: i++;
540: }
1.106 cvs 541: nsuffix[i] = EOS;
542: if (!strcmp (nsuffix, "html") ||
543: !strcmp (nsuffix, "htm") ||
544: !strcmp (nsuffix, "shtml") ||
545: !strcmp (nsuffix, "jsp") ||
546: !strcmp (nsuffix, "xht") ||
547: !strcmp (nsuffix, "xhtm") ||
548: !strcmp (nsuffix, "xhtml"))
1.101 cvs 549: return (TRUE);
1.106 cvs 550: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 551: {
552: /* take into account compressed files */
1.124 vatton 553: TtaExtractSuffix (temppath, suffix);
1.101 cvs 554: /* Normalize the suffix */
555: i = 0;
1.106 cvs 556: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 557: {
1.123 vatton 558: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 559: i++;
560: }
1.106 cvs 561: nsuffix[i] = EOS;
562: if (!strcmp (nsuffix, "html") ||
563: !strcmp (nsuffix, "htm") ||
564: !strcmp (nsuffix, "shtml") ||
565: !strcmp (nsuffix, "jsp") ||
566: !strcmp (nsuffix, "xht") ||
567: !strcmp (nsuffix, "xhtm") ||
568: !strcmp (nsuffix, "xhtml"))
1.101 cvs 569: return (TRUE);
570: else
571: return (FALSE);
572: }
573: else
574: /* check if there is another suffix */
1.124 vatton 575: TtaExtractSuffix (temppath, suffix);
1.101 cvs 576: }
1.88 cvs 577: return (FALSE);
1.3 cvs 578: }
579:
1.4 cvs 580: /*----------------------------------------------------------------------
1.136 cvs 581: IsMathMLName
582: returns TRUE if path points to an MathML resource.
1.56 cvs 583: ----------------------------------------------------------------------*/
1.136 cvs 584: ThotBool IsMathMLName (const char *path)
1.56 cvs 585: {
1.136 cvs 586: char temppath[MAX_LENGTH];
587: char suffix[MAX_LENGTH];
1.56 cvs 588:
589: if (!path)
590: return (FALSE);
591:
1.106 cvs 592: strcpy (temppath, path);
1.124 vatton 593: TtaExtractSuffix (temppath, suffix);
1.56 cvs 594:
1.136 cvs 595: if (!strcasecmp (suffix, "mml"))
1.56 cvs 596: return (TRUE);
1.106 cvs 597: else if (!strcmp (suffix, "gz"))
1.56 cvs 598: {
599: /* take into account compressed files */
1.124 vatton 600: TtaExtractSuffix (temppath, suffix);
1.136 cvs 601: if (!strcasecmp (suffix, "mml"))
1.60 cvs 602: return (TRUE);
603: else
604: return (FALSE);
605: }
606: else
607: return (FALSE);
608: }
609:
610: /*----------------------------------------------------------------------
1.136 cvs 611: IsSVGName
612: returns TRUE if path points to an SVG resource.
1.133 vatton 613: ----------------------------------------------------------------------*/
1.136 cvs 614: ThotBool IsSVGName (const char *path)
1.133 vatton 615: {
1.136 cvs 616: char temppath[MAX_LENGTH];
617: char suffix[MAX_LENGTH];
1.133 vatton 618:
619: if (!path)
620: return (FALSE);
621:
622: strcpy (temppath, path);
623: TtaExtractSuffix (temppath, suffix);
624:
1.136 cvs 625: if (!strcasecmp (suffix, "svg"))
1.133 vatton 626: return (TRUE);
627: else if (!strcmp (suffix, "gz"))
628: {
629: /* take into account compressed files */
630: TtaExtractSuffix (temppath, suffix);
1.136 cvs 631: if (!strcasecmp (suffix, "svg"))
1.133 vatton 632: return (TRUE);
633: else
634: return (FALSE);
635: }
636: else
637: return (FALSE);
638: }
639:
640: /*----------------------------------------------------------------------
1.136 cvs 641: IsXMLName
642: returns TRUE if path points to an XML resource.
1.103 cvs 643: ----------------------------------------------------------------------*/
1.136 cvs 644: ThotBool IsXMLName (const char *path)
1.103 cvs 645: {
1.136 cvs 646: char temppath[MAX_LENGTH];
647: char suffix[MAX_LENGTH];
1.103 cvs 648:
649: if (!path)
650: return (FALSE);
651:
1.106 cvs 652: strcpy (temppath, path);
1.124 vatton 653: TtaExtractSuffix (temppath, suffix);
1.103 cvs 654:
1.136 cvs 655: if (!strcasecmp (suffix, "xml") ||
656: !strcasecmp (suffix, "xht") ||
657: !strcmp (suffix, "xhtm") ||
658: !strcmp (suffix, "xhtml"))
1.103 cvs 659: return (TRUE);
1.106 cvs 660: else if (!strcmp (suffix, "gz"))
1.103 cvs 661: {
662: /* take into account compressed files */
1.124 vatton 663: TtaExtractSuffix (temppath, suffix);
1.136 cvs 664: if (!strcasecmp (suffix, "xml") ||
665: !strcasecmp (suffix, "xht") ||
666: !strcmp (suffix, "xhtm") ||
667: !strcmp (suffix, "xhtml"))
1.103 cvs 668: return (TRUE);
669: else
670: return (FALSE);
671: }
672: else
673: return (FALSE);
674: }
675:
676: /*----------------------------------------------------------------------
1.136 cvs 677: IsUndisplayedName
678: returns TRUE if path points to an undisplayed resource.
1.103 cvs 679: ----------------------------------------------------------------------*/
1.136 cvs 680: ThotBool IsUndisplayedName (const char *path)
1.103 cvs 681: {
1.106 cvs 682: char temppath[MAX_LENGTH];
683: char suffix[MAX_LENGTH];
1.103 cvs 684:
685: if (!path)
686: return (FALSE);
687:
1.106 cvs 688: strcpy (temppath, path);
1.124 vatton 689: TtaExtractSuffix (temppath, suffix);
1.103 cvs 690:
1.136 cvs 691: if (!strcasecmp (suffix, "exe") ||
692: !strcasecmp (suffix, "zip") ||
693: !strcasecmp (suffix, "ppt") ||
694: !strcasecmp (suffix, "pdf") ||
695: !strcasecmp (suffix, "ps") ||
696: !strcasecmp (suffix, "eps") ||
697: !strcasecmp (suffix, "tar") ||
698: !strcasecmp (suffix, "tgz") ||
699: !strcasecmp (suffix, "ddl") ||
700: !strcasecmp (suffix, "o"))
1.103 cvs 701: return (TRUE);
1.106 cvs 702: else if (!strcmp (suffix, "gz"))
1.103 cvs 703: {
704: /* take into account compressed files */
1.124 vatton 705: TtaExtractSuffix (temppath, suffix);
1.136 cvs 706: if (!strcasecmp (suffix, "exe") ||
707: !strcasecmp (suffix, "zip") ||
708: !strcasecmp (suffix, "ppt") ||
709: !strcasecmp (suffix, "pdf") ||
710: !strcasecmp (suffix, "ps") ||
711: !strcasecmp (suffix, "eps") ||
712: !strcasecmp (suffix, "tar") ||
713: !strcasecmp (suffix, "ddl") ||
714: !strcasecmp (suffix, "o"))
1.103 cvs 715: return (TRUE);
716: else
717: return (FALSE);
718: }
719: else
720: return (FALSE);
721: }
722:
723: /*----------------------------------------------------------------------
1.60 cvs 724: IsCSSName
725: returns TRUE if path points to an XML resource.
726: ----------------------------------------------------------------------*/
1.111 cvs 727: ThotBool IsCSSName (const char *path)
1.60 cvs 728: {
1.106 cvs 729: char temppath[MAX_LENGTH];
730: char suffix[MAX_LENGTH];
1.60 cvs 731:
732: if (!path)
733: return (FALSE);
734:
1.106 cvs 735: strcpy (temppath, path);
1.124 vatton 736: TtaExtractSuffix (temppath, suffix);
1.60 cvs 737:
1.106 cvs 738: if (!strcasecmp (suffix, "css"))
1.60 cvs 739: return (TRUE);
1.106 cvs 740: else if (!strcmp (suffix, "gz"))
1.60 cvs 741: {
742: /* take into account compressed files */
1.124 vatton 743: TtaExtractSuffix (temppath, suffix);
1.106 cvs 744: if (!strcasecmp (suffix, "css"))
1.56 cvs 745: return (TRUE);
746: else
747: return (FALSE);
748: }
749: else
750: return (FALSE);
751: }
752:
753: /*----------------------------------------------------------------------
1.9 cvs 754: IsImageName
755: returns TRUE if path points to an image resource.
1.4 cvs 756: ----------------------------------------------------------------------*/
1.111 cvs 757: ThotBool IsImageName (const char *path)
1.106 cvs 758: {
759: char temppath[MAX_LENGTH];
760: char suffix[MAX_LENGTH];
761: char nsuffix[MAX_LENGTH];
1.5 cvs 762: int i;
763:
764: if (!path)
1.13 cvs 765: return (FALSE);
1.5 cvs 766:
1.106 cvs 767: strcpy (temppath, path);
1.124 vatton 768: TtaExtractSuffix (temppath, suffix);
1.5 cvs 769:
770: /* Normalize the suffix */
771: i = 0;
1.106 cvs 772: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 773: {
1.123 vatton 774: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 775: i++;
776: }
1.106 cvs 777: nsuffix[i] = EOS;
778: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
779: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
780: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 781: return (TRUE);
782: return (FALSE);
1.3 cvs 783: }
784:
1.4 cvs 785: /*----------------------------------------------------------------------
1.58 cvs 786: IsImageType
787: returns TRUE if type points to an image resource.
788: ----------------------------------------------------------------------*/
1.111 cvs 789: ThotBool IsImageType (const char *type)
1.58 cvs 790: {
1.106 cvs 791: char temptype[MAX_LENGTH];
1.58 cvs 792: int i;
793:
794: if (!type)
795: return (FALSE);
796:
1.106 cvs 797: strcpy (temptype, type);
1.58 cvs 798: /* Normalize the type */
799: i = 0;
1.106 cvs 800: while (temptype[i] != EOS)
1.58 cvs 801: {
802: temptype[i] = tolower (temptype[i]);
803: i++;
804: }
1.111 cvs 805: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
806: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
807: !strcmp (temptype, "png"))
1.58 cvs 808: return (TRUE);
809: return (FALSE);
810: }
811:
812: /*----------------------------------------------------------------------
1.9 cvs 813: IsTextName
1.4 cvs 814: ----------------------------------------------------------------------*/
1.111 cvs 815: ThotBool IsTextName (const char *path)
1.106 cvs 816: {
817: char temppath[MAX_LENGTH];
818: char suffix[MAX_LENGTH];
819: char nsuffix[MAX_LENGTH];
1.5 cvs 820: int i;
821:
822: if (!path)
1.13 cvs 823: return (FALSE);
1.5 cvs 824:
1.106 cvs 825: strcpy (temppath, path);
1.124 vatton 826: TtaExtractSuffix (temppath, suffix);
1.5 cvs 827:
828: /* Normalize the suffix */
829: i = 0;
1.106 cvs 830: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 831: {
1.25 cvs 832: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 833: i++;
834: }
1.106 cvs 835: nsuffix[i] = EOS;
1.5 cvs 836:
1.111 cvs 837: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 838: return (TRUE);
1.106 cvs 839: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 840: {
1.39 cvs 841: /* take into account compressed files */
1.124 vatton 842: TtaExtractSuffix (temppath, suffix);
1.13 cvs 843: /* Normalize the suffix */
844: i = 0;
1.106 cvs 845: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 846: {
1.25 cvs 847: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 848: i++;
849: }
1.106 cvs 850: nsuffix[i] = EOS;
1.111 cvs 851: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 852: return (TRUE);
853: else
854: return (FALSE);
855: }
856: else
857: return (FALSE);
1.3 cvs 858: }
859:
1.4 cvs 860: /*----------------------------------------------------------------------
1.9 cvs 861: IsHTTPPath
862: returns TRUE if path is in fact an http URL.
1.4 cvs 863: ----------------------------------------------------------------------*/
1.112 cvs 864: ThotBool IsHTTPPath (const char *path)
1.3 cvs 865: {
1.5 cvs 866: if (!path)
867: return FALSE;
1.3 cvs 868:
1.106 cvs 869: if ((!strncmp (path, "http:", 5) != 0)
870: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
871: || !strncmp (path, "internal:", 9))
1.58 cvs 872: return TRUE;
873: return FALSE;
1.3 cvs 874: }
875:
1.4 cvs 876: /*----------------------------------------------------------------------
1.9 cvs 877: IsWithParameters
878: returns TRUE if url has a concatenated query string.
1.4 cvs 879: ----------------------------------------------------------------------*/
1.133 vatton 880: ThotBool IsWithParameters (const char *url)
1.3 cvs 881: {
1.5 cvs 882: int i;
1.3 cvs 883:
1.9 cvs 884: if ((!url) || (url[0] == EOS))
1.5 cvs 885: return FALSE;
1.3 cvs 886:
1.9 cvs 887: i = strlen (url) - 1;
888: while (i > 0 && url[i--] != '?')
1.5 cvs 889: if (i < 0)
890: return FALSE;
1.3 cvs 891:
1.5 cvs 892: /* There is a parameter */
893: return TRUE;
1.3 cvs 894: }
895:
1.4 cvs 896: /*----------------------------------------------------------------------
1.9 cvs 897: IsW3Path
898: returns TRUE if path is in fact a URL.
1.4 cvs 899: ----------------------------------------------------------------------*/
1.133 vatton 900: ThotBool IsW3Path (const char *path)
1.106 cvs 901: {
902: if (strncmp (path, "http:", 5) &&
903: strncmp (path, "ftp:", 4) &&
904: strncmp (path, "telnet:", 7) &&
905: strncmp (path, "wais:", 5) &&
906: strncmp (path, "news:", 5) &&
907: strncmp (path, "gopher:", 7) &&
908: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 909: strncmp (path, "archie:", 7) &&
910: strncmp (path, "https:", 6))
1.72 cvs 911: return FALSE;
912: return TRUE;
1.3 cvs 913: }
914:
1.4 cvs 915: /*----------------------------------------------------------------------
1.90 cvs 916: IsFilePath
917: returns TRUE if path is in fact a URL.
918: ----------------------------------------------------------------------*/
1.133 vatton 919: ThotBool IsFilePath (const char *path)
1.90 cvs 920: {
1.106 cvs 921: if (strncmp (path, "file:", 5))
1.90 cvs 922: return FALSE;
923: return TRUE;
924: }
925:
926: /*----------------------------------------------------------------------
1.9 cvs 927: IsValidProtocol
928: returns true if the url protocol is supported by Amaya.
1.4 cvs 929: ----------------------------------------------------------------------*/
1.133 vatton 930: ThotBool IsValidProtocol (const char *url)
1.106 cvs 931: {
932: if (!strncmp (url, "http:", 5)
933: || !strncmp (url, "internal:", 9)
934: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 935: /* experimental */
1.24 cvs 936: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 937: return (TRUE);
1.5 cvs 938: else
1.8 cvs 939: return (FALSE);
1.3 cvs 940: }
941:
1.31 cvs 942:
943: /*----------------------------------------------------------------------
944: GetBaseURL
945: normalizes orgName according to a base associated with doc, and
946: following the standard URL format rules.
947: The function returns the base used to solve relative URL and SRC:
948: - the base of the document,
949: - or the document path (without document name).
950: ----------------------------------------------------------------------*/
1.106 cvs 951: char *GetBaseURL (Document doc)
1.31 cvs 952: {
953: Element el;
954: ElementType elType;
955: AttributeType attrType;
956: Attribute attr;
1.106 cvs 957: char *ptr, *basename;
1.31 cvs 958: int length;
959:
1.113 cvs 960: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 961: return NULL;
1.106 cvs 962: basename = TtaGetMemory (MAX_LENGTH);
963: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
964: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 965: length = MAX_LENGTH -1;
1.113 cvs 966: /* is it a HTML document ? */
1.31 cvs 967: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.113 cvs 968: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
969: /* it's a HTML document */
1.65 cvs 970: {
1.113 cvs 971: /* get the document element */
972: el = TtaGetMainRoot (doc);
973: /* search the BASE element */
974: elType.ElTypeNum = HTML_EL_HEAD;
975: el = TtaSearchTypedElement (elType, SearchForward, el);
976: if (el)
977: /* there is a HEAD element */
978: {
979: /* look for a BASE element within the HEAD */
980: elType.ElTypeNum = HTML_EL_BASE;
981: el = TtaSearchTypedElement (elType, SearchInTree, el);
982: }
983: if (el)
1.31 cvs 984: {
1.113 cvs 985: /* The document has a BASE element. Get the HREF attribute of the
986: BASE element */
987: attrType.AttrSSchema = elType.ElSSchema;
988: attrType.AttrTypeNum = HTML_ATTR_HREF_;
989: attr = TtaGetAttribute (el, attrType);
990: if (attr)
1.31 cvs 991: {
1.113 cvs 992: /* Use the base path of the document */
993: TtaGiveTextAttributeValue (attr, basename, &length);
994: /* base and orgName have to be separated by a DIR_SEP */
995: length--;
996: if (basename[0] != EOS && basename[length] != URL_SEP &&
997: basename[length] != DIR_SEP)
998: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 999: {
1.113 cvs 1000: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1001: AMAYA_PARSE_HOST |
1002: AMAYA_PARSE_PUNCTUATION);
1003: if (ptr && !strcmp (ptr, basename))
1004: {
1005: /* it has this form, complete it by adding a URL_STR */
1006: if (strchr (basename, DIR_SEP))
1007: strcat (basename, DIR_STR);
1008: else
1009: strcat (basename, URL_STR);
1010: length++;
1011: }
1012: if (ptr)
1013: TtaFreeMemory (ptr);
1.31 cvs 1014: }
1015: }
1016: }
1.113 cvs 1017: }
1018:
1.31 cvs 1019: /* Remove anything after the last DIR_SEP char. If no such char is found,
1020: * then search for the first ":" char, hoping that what's before that is a
1021: * protocol. If found, end the string there. If neither char is found,
1022: * then discard the whole base element.
1023: */
1.106 cvs 1024: length = strlen (basename) - 1;
1.31 cvs 1025: /* search for the last DIR_SEP char */
1.106 cvs 1026: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 1027: length--;
1028: if (length >= 0)
1029: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1030: basename[length + 1] = EOS;
1.31 cvs 1031: else
1032: /* search for the first PATH_STR char */
1033: {
1.106 cvs 1034: for (length = 0; basename[length] != ':' &&
1035: basename[length] != EOS; length ++);
1036: if (basename[length] == ':')
1.31 cvs 1037: /* found, so end the string there */
1.106 cvs 1038: basename[length + 1] = EOS;
1.31 cvs 1039: else
1040: /* not found, discard the base */
1.106 cvs 1041: basename[0] = EOS;
1.31 cvs 1042: }
1043: return (basename);
1044: }
1045:
1046:
1.4 cvs 1047: /*----------------------------------------------------------------------
1.40 cvs 1048: GetLocalPath
1049: Allocate and return the local document path associated to the url
1050: ----------------------------------------------------------------------*/
1.106 cvs 1051: char *GetLocalPath (Document doc, char *url)
1052: {
1053: char *ptr;
1054: char *n;
1055: char *documentname;
1056: char url_sep;
1.83 cvs 1057: int len;
1.67 cvs 1058: ThotBool noFile;
1.40 cvs 1059:
1060: if (url != NULL)
1061: {
1062: /* check whether the file name exists */
1.106 cvs 1063: len = strlen (url) - 1;
1.71 cvs 1064: if (IsW3Path (url))
1.106 cvs 1065: url_sep = '/';
1.41 cvs 1066: else
1.106 cvs 1067: url_sep = DIR_SEP;
1.41 cvs 1068: noFile = (url[len] == url_sep);
1.40 cvs 1069: if (noFile)
1.106 cvs 1070: url[len] = EOS;
1071: ptr = TtaGetMemory (MAX_LENGTH);
1072: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 1073: TtaExtractName (url, ptr, documentname);
1.106 cvs 1074: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1075: if (!TtaCheckDirectory (ptr))
1076: /* directory did not exist */
1.72 cvs 1077: TtaMakeDirectory (ptr);
1.47 cvs 1078:
1079: /* don't include the query string within document name */
1.106 cvs 1080: n = strrchr (documentname, '?');
1.47 cvs 1081: if (n != NULL)
1.106 cvs 1082: *n = EOS;
1.46 cvs 1083: /* don't include ':' within document name */
1.106 cvs 1084: n = strchr (documentname, ':');
1.46 cvs 1085: if (n != NULL)
1.106 cvs 1086: *n = EOS;
1.69 cvs 1087: /* if after all this operations document name
1088: is empty, let's use noname.html instead */
1.106 cvs 1089: if (documentname[0] == EOS)
1090: strcat (ptr, "noname.html");
1.69 cvs 1091: else
1.106 cvs 1092: strcat (ptr, documentname);
1.40 cvs 1093: TtaFreeMemory (documentname);
1094: /* restore the url */
1095: if (noFile)
1.41 cvs 1096: url[len] = url_sep;
1.40 cvs 1097: return (ptr);
1098: }
1099: else
1100: return (NULL);
1101: }
1102:
1.73 cvs 1103: /*----------------------------------------------------------------------
1.79 cvs 1104: ExtractTarget extract the target name from document nane.
1105: ----------------------------------------------------------------------*/
1.106 cvs 1106: void ExtractTarget (char *aName, char *target)
1.79 cvs 1107: {
1.106 cvs 1108: int lg, i;
1109: char *ptr;
1110: char *oldptr;
1.79 cvs 1111:
1112: if (!target || !aName)
1113: /* bad target */
1114: return;
1115:
1.106 cvs 1116: target[0] = EOS;
1117: lg = strlen (aName);
1.79 cvs 1118: if (lg)
1119: {
1120: /* the name is not empty */
1121: oldptr = ptr = &aName[0];
1122: do
1123: {
1.106 cvs 1124: ptr = strrchr (oldptr, '#');
1.79 cvs 1125: if (ptr)
1126: oldptr = &ptr[1];
1127: }
1128: while (ptr);
1129:
1130: i = (int) (oldptr) - (int) (aName); /* name length */
1131: if (i > 1)
1132: {
1.106 cvs 1133: aName[i - 1] = EOS;
1.79 cvs 1134: if (i != lg)
1.106 cvs 1135: strcpy (target, oldptr);
1.79 cvs 1136: }
1137: }
1138: }
1139:
1140: /*----------------------------------------------------------------------
1.90 cvs 1141: RemoveNewLines (text)
1142: Removes any '\n' chars that are found in text.
1143: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1144: ----------------------------------------------------------------------*/
1.106 cvs 1145: ThotBool RemoveNewLines (char *text)
1146: {
1147: ThotBool change = FALSE;
1148: char *src;
1149: char *dest;
1.90 cvs 1150:
1151: src = text;
1152: dest = text;
1.115 kahan 1153:
1154: /* remove any preceding whitespace */
1155: while (*src && *src == ' ')
1156: {
1157: src++;
1158: change = 1;
1159: }
1160:
1.90 cvs 1161: while (*src)
1162: {
1163: switch (*src)
1164: {
1.106 cvs 1165: case '\n':
1.90 cvs 1166: /* don't copy the newline */
1167: change = 1;
1168: break;
1169: default:
1170: *dest = *src;
1171: dest++;
1172: break;
1173: }
1174: src++;
1175: }
1176: /* copy the last EOS char */
1177: *dest = *src;
1178:
1179: return (change);
1180: }
1181:
1182: /*----------------------------------------------------------------------
1183: CleanCopyFileURL
1184: Copies a file url from a src string to destination string.
1.97 cvs 1185: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1186: we want to do).
1.90 cvs 1187: ----------------------------------------------------------------------*/
1.106 cvs 1188: static void CleanCopyFileURL (char *dest, char *src,
1189: ConvertionType convertion)
1.90 cvs 1190: {
1191: while (*src)
1.89 cvs 1192: {
1.90 cvs 1193: switch (*src)
1.89 cvs 1194: {
1195: #ifdef _WINDOWS
1.106 cvs 1196: case URL_SEP:
1.96 cvs 1197: /* make DIR_SEP transformation */
1.97 cvs 1198: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1199: *dest = DIR_SEP;
1.96 cvs 1200: else
1201: *dest = *src;
1.90 cvs 1202: dest++;
1.96 cvs 1203: src++;
1.90 cvs 1204: break;
1.89 cvs 1205: #endif /* _WINDOWS */
1.96 cvs 1206:
1.106 cvs 1207: case '%':
1.97 cvs 1208: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1209: {
1.97 cvs 1210: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1211: src++;
1.106 cvs 1212: if (*src != EOS)
1.97 cvs 1213: {
1214: *dest = UnEscapeChar (*src) * 16;
1215: src++;
1216: }
1.106 cvs 1217: if (*src != EOS)
1.97 cvs 1218: {
1219: *dest = *dest + UnEscapeChar (*src);
1220: src++;
1221: }
1222: dest++;
1.96 cvs 1223: }
1.97 cvs 1224: else
1.96 cvs 1225: {
1.97 cvs 1226: *dest = *src;
1227: dest++;
1.96 cvs 1228: src++;
1229: }
1230: break;
1231:
1.90 cvs 1232: default:
1233: *dest = *src;
1.89 cvs 1234: dest++;
1.96 cvs 1235: src++;
1.90 cvs 1236: break;
1.89 cvs 1237: }
1238: }
1.90 cvs 1239: /* copy the EOS char */
1240: *dest = *src;
1.73 cvs 1241: }
1.40 cvs 1242:
1243: /*----------------------------------------------------------------------
1.9 cvs 1244: NormalizeURL
1245: normalizes orgName according to a base associated with doc, and
1246: following the standard URL format rules.
1.113 cvs 1247: if doc is < 0, use as a base the URL of the document that contains
1248: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1249: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1250: other path.
1.9 cvs 1251: The function returns the new complete and normalized URL
1.12 cvs 1252: or file name path (newName) and the name of the document (docName).
1.9 cvs 1253: N.B. If the function can't find out what's the docName, it assigns
1254: the name "noname.html".
1.4 cvs 1255: ----------------------------------------------------------------------*/
1.106 cvs 1256: void NormalizeURL (char *orgName, Document doc, char *newName,
1257: char *docName, char *otherPath)
1258: {
1259: char *basename;
1260: char tempOrgName[MAX_LENGTH];
1261: char *ptr;
1262: char used_sep;
1.84 cvs 1263: int length;
1264: ThotBool check;
1.5 cvs 1265:
1.110 cvs 1266: #ifdef _WINDOWS
1.44 cvs 1267: int ndx;
1.110 cvs 1268: #endif /* _WINDOWS */
1.44 cvs 1269:
1.5 cvs 1270: if (!newName || !docName)
1271: return;
1.18 cvs 1272:
1.113 cvs 1273: if (doc < 0)
1274: basename = TtaStrdup (SavedDocumentURL);
1275: else if (doc > 0)
1.53 cvs 1276: basename = GetBaseURL (doc);
1277: else if (otherPath != NULL)
1.108 cvs 1278: basename = TtaStrdup (otherPath);
1.32 cvs 1279: else
1.53 cvs 1280: basename = NULL;
1.32 cvs 1281:
1.18 cvs 1282: /*
1.31 cvs 1283: * Clean orgName
1284: * Make sure we have a complete orgName, without any leading or trailing
1285: * white spaces, or trailinbg new lines
1286: */
1.5 cvs 1287: ptr = orgName;
1.18 cvs 1288: /* skip leading white space and new line characters */
1.106 cvs 1289: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1290: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1291: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1292: /*
1.31 cvs 1293: * Make orgName a complete URL
1294: * If the URL does not include a protocol, then try to calculate
1295: * one using the doc's base element (if it exists),
1296: */
1.106 cvs 1297: if (tempOrgName[0] == EOS)
1.53 cvs 1298: {
1.106 cvs 1299: newName[0] = EOS;
1300: docName[0] = EOS;
1.53 cvs 1301: TtaFreeMemory (basename);
1302: return;
1303: }
1.49 cvs 1304:
1305: /* clean trailing white space */
1.106 cvs 1306: length = strlen (tempOrgName) - 1;
1307: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1308: {
1.106 cvs 1309: tempOrgName[length] = EOS;
1.53 cvs 1310: length--;
1311: }
1.50 cvs 1312:
1.55 cvs 1313: /* remove extra dot (which dot???) */
1314: /* ugly, but faster than a strcmp */
1.106 cvs 1315: if (tempOrgName[length] == '.'
1316: && (length == 0 || tempOrgName[length-1] != '.'))
1317: tempOrgName[length] = EOS;
1.50 cvs 1318:
1.94 cvs 1319: if (IsW3Path (tempOrgName))
1.53 cvs 1320: {
1321: /* the name is complete, go to the Sixth Step */
1.106 cvs 1322: strcpy (newName, tempOrgName);
1.53 cvs 1323: SimplifyUrl (&newName);
1324: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1325: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1326: AMAYA_PARSE_HOST |
1327: AMAYA_PARSE_PUNCTUATION);
1328: if (ptr && !strcmp (ptr, newName))
1329: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1330: strcat (newName, URL_STR);
1.49 cvs 1331:
1.53 cvs 1332: if (ptr)
1.50 cvs 1333: TtaFreeMemory (ptr);
1.53 cvs 1334: }
1.113 cvs 1335: else if (basename == NULL)
1.53 cvs 1336: /* the name is complete, go to the Sixth Step */
1.106 cvs 1337: strcpy (newName, tempOrgName);
1.53 cvs 1338: else
1339: {
1.31 cvs 1340: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1341: #ifdef _WINDOWS
1.53 cvs 1342: if (!IsW3Path (basename))
1343: {
1.106 cvs 1344: length = strlen (tempOrgName);
1.53 cvs 1345: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1346: if (tempOrgName [ndx] == '/')
1347: tempOrgName [ndx] = '\\';
1.53 cvs 1348: }
1.110 cvs 1349: #endif /* _WINDOWS */
1.25 cvs 1350: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1351: if (ptr)
1352: {
1353: SimplifyUrl (&ptr);
1.106 cvs 1354: strcpy (newName, ptr);
1.53 cvs 1355: TtaFreeMemory (ptr);
1356: }
1357: else
1.106 cvs 1358: newName[0] = EOS;
1.53 cvs 1359: }
1.36 cvs 1360:
1361: TtaFreeMemory (basename);
1.18 cvs 1362: /*
1.31 cvs 1363: * Prepare the docname that will refer to this ressource in the
1364: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1365: * noname.html as a default ressource name
1.18 cvs 1366: */
1.106 cvs 1367: if (newName[0] != EOS)
1.53 cvs 1368: {
1.106 cvs 1369: length = strlen (newName) - 1;
1370: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1371: {
1372: used_sep = newName[length];
1373: check = TRUE;
1374: while (check)
1375: {
1.50 cvs 1376: length--;
1377: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1378: length--;
1.106 cvs 1379: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1380: {
1.106 cvs 1381: newName[length+1] = EOS;
1.53 cvs 1382: /* remove also previous directory */
1383: length--;
1384: while (length >= 0 && newName[length] != used_sep)
1385: length--;
1.106 cvs 1386: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1387: /* don't remove server name */
1.106 cvs 1388: newName[length+1] = EOS;
1.53 cvs 1389: }
1.106 cvs 1390: else if (!strncmp (&newName[length+1], ".", 1))
1391: newName[length+1] = EOS;
1.50 cvs 1392: else
1.53 cvs 1393: check = FALSE;
1394: }
1395: /* docname was not comprised inside the URL, so let's */
1396: /* assign the default ressource name */
1.106 cvs 1397: strcpy (docName, "noname.html");
1.53 cvs 1398: }
1399: else
1400: { /* docname is comprised inside the URL */
1.110 cvs 1401: while (length >= 0 && newName[length] != URL_SEP &&
1402: newName[length] != DIR_SEP)
1.53 cvs 1403: length--;
1404: if (length < 0)
1.106 cvs 1405: strcpy (docName, newName);
1.53 cvs 1406: else
1.106 cvs 1407: strcpy (docName, &newName[length+1]);
1.53 cvs 1408: }
1409: }
1410: else
1.106 cvs 1411: docName[0] = EOS;
1.18 cvs 1412: }
1.3 cvs 1413:
1.4 cvs 1414: /*----------------------------------------------------------------------
1.9 cvs 1415: IsSameHost
1.4 cvs 1416: ----------------------------------------------------------------------*/
1.106 cvs 1417: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1418: {
1.106 cvs 1419: char *basename_ptr1, *basename_ptr2;
1420: ThotBool result;
1.3 cvs 1421:
1.106 cvs 1422: basename_ptr1 = AmayaParseUrl (url1, "",
1423: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1424: basename_ptr2 = AmayaParseUrl (url2, "",
1425: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1426:
1.106 cvs 1427: if (strcmp (basename_ptr1, basename_ptr2))
1428: result = FALSE;
1429: else
1430: result = TRUE;
1431: TtaFreeMemory (basename_ptr1);
1432: TtaFreeMemory (basename_ptr2);
1433: return (result);
1.3 cvs 1434: }
1435:
1436:
1.4 cvs 1437: /*----------------------------------------------------------------------
1.22 cvs 1438: HasKnownFileSuffix
1439: returns TRUE if path points to a file ending with a suffix.
1440: ----------------------------------------------------------------------*/
1.106 cvs 1441: ThotBool HasKnownFileSuffix (const char *path)
1442: {
1443: char *root;
1444: char temppath[MAX_LENGTH];
1445: char suffix[MAX_LENGTH];
1.22 cvs 1446:
1.106 cvs 1447: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1448: return (FALSE);
1449:
1.106 cvs 1450: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1451:
1452: if (root)
1453: {
1.106 cvs 1454: strcpy (temppath, root);
1.25 cvs 1455: TtaFreeMemory (root);
1.22 cvs 1456: /* Get the suffix */
1.124 vatton 1457: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1458:
1.106 cvs 1459: if( suffix[0] == EOS)
1.22 cvs 1460: /* no suffix */
1461: return (FALSE);
1462:
1463: /* Normalize the suffix */
1464: ConvertToLowerCase (suffix);
1465:
1.106 cvs 1466: if (!strcmp (suffix, "gz"))
1.22 cvs 1467: /* skip the compressed suffix */
1468: {
1.124 vatton 1469: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1470: if(suffix[0] == EOS)
1.22 cvs 1471: /* no suffix */
1472: return (FALSE);
1473: /* Normalize the suffix */
1474: ConvertToLowerCase (suffix);
1475: }
1476:
1.106 cvs 1477: if (strcmp (suffix, "gif") &&
1478: strcmp (suffix, "xbm") &&
1479: strcmp (suffix, "xpm") &&
1480: strcmp (suffix, "jpg") &&
1481: strcmp (suffix, "pdf") &&
1482: strcmp (suffix, "png") &&
1483: strcmp (suffix, "tgz") &&
1484: strcmp (suffix, "xpg") &&
1485: strcmp (suffix, "xpd") &&
1486: strcmp (suffix, "ps") &&
1487: strcmp (suffix, "au") &&
1488: strcmp (suffix, "html") &&
1489: strcmp (suffix, "htm") &&
1490: strcmp (suffix, "shtml") &&
1491: strcmp (suffix, "xht") &&
1492: strcmp (suffix, "xhtm") &&
1493: strcmp (suffix, "xhtml") &&
1494: strcmp (suffix, "txt") &&
1495: strcmp (suffix, "css") &&
1496: strcmp (suffix, "eps"))
1.22 cvs 1497: return (FALSE);
1498: else
1499: return (TRUE);
1500: }
1501: else
1502: return (FALSE);
1503: }
1504:
1505:
1506: /*----------------------------------------------------------------------
1.24 cvs 1507: ChopURL
1508: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1509: If inputURL is bigger than that size, outputURL receives
1510: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1511: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1512: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1513: copied into outputURL.
1514: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1515: chars.
1516: ----------------------------------------------------------------------*/
1.106 cvs 1517: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1518: {
1519: int len;
1.9 cvs 1520:
1.106 cvs 1521: len = strlen (inputURL);
1.24 cvs 1522: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1523: strcpy (outputURL, inputURL);
1.24 cvs 1524: else
1525: /* make a truncated urlName on the status window */
1526: {
1.106 cvs 1527: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1528: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1529: strcat (outputURL, "...");
1530: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1531: }
1.25 cvs 1532: }
1533:
1534:
1535: /*----------------------------------------------------------------------
1536: scan
1.47 cvs 1537: Scan a filename for its constituents
1.25 cvs 1538: -----------------------------------
1539:
1540: On entry,
1541: name points to a document name which may be incomplete.
1542: On exit,
1543: absolute or relative may be nonzero (but not both).
1544: host, fragment and access may be nonzero if they were specified.
1545: Any which are nonzero point to zero terminated strings.
1546: ----------------------------------------------------------------------*/
1.106 cvs 1547: static void scan (char *name, HTURI *parts)
1.25 cvs 1548: {
1.106 cvs 1549: char * p;
1550: char * after_access = name;
1.32 cvs 1551:
1.43 cvs 1552: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1553: /* Look for fragment identifier */
1.106 cvs 1554: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1555: {
1.106 cvs 1556: *p++ = '\0';
1.28 cvs 1557: parts->fragment = p;
1.25 cvs 1558: }
1559:
1.28 cvs 1560: for (p=name; *p; p++)
1561: {
1.106 cvs 1562: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1563: break;
1.106 cvs 1564: if (*p == ':')
1.28 cvs 1565: {
1566: *p = 0;
1567: parts->access = after_access; /* Scheme has been specified */
1568:
1569: /* The combination of gcc, the "-O" flag and the HP platform is
1570: unhealthy. The following three lines is a quick & dirty fix, but is
1571: not recommended. Rather, turn off "-O". */
1572:
1573: /* after_access = p;*/
1574: /* while (*after_access == 0)*/
1575: /* after_access++;*/
1576: after_access = p+1;
1.106 cvs 1577: if (!strcasecmp("URL", parts->access))
1.28 cvs 1578: /* Ignore IETF's URL: pre-prefix */
1579: parts->access = NULL;
1580: else
1.25 cvs 1581: break;
1582: }
1583: }
1584:
1585: p = after_access;
1.43 cvs 1586: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1587: {
1.43 cvs 1588: if (p[1] == URL_SEP)
1.28 cvs 1589: {
1.25 cvs 1590: parts->host = p+2; /* host has been specified */
1.28 cvs 1591: *p = 0; /* Terminate access */
1592: /* look for end of host name if any */
1.106 cvs 1593: p = strchr (parts->host, URL_SEP);
1.28 cvs 1594: if (p)
1595: {
1.106 cvs 1596: *p = EOS; /* Terminate host */
1.25 cvs 1597: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1598: }
1599: }
1600: else
1601: /* Root found but no host */
1602: parts->absolute = p+1;
1603: }
1604: else
1605: {
1.25 cvs 1606: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1607: }
1.25 cvs 1608: }
1609:
1610:
1611: /*----------------------------------------------------------------------
1.28 cvs 1612: AmayaParseUrl: parse a Name relative to another name
1613:
1614: This returns those parts of a name which are given (and requested)
1615: substituting bits from the related name where necessary.
1.25 cvs 1616:
1.28 cvs 1617: On entry,
1.25 cvs 1618: aName A filename given
1619: relatedName A name relative to which aName is to be parsed. Give
1620: it an empty string if aName is absolute.
1621: wanted A mask for the bits which are wanted.
1622:
1.28 cvs 1623: On exit,
1.25 cvs 1624: returns A pointer to a malloc'd string which MUST BE FREED
1625: ----------------------------------------------------------------------*/
1.106 cvs 1626: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1627: {
1628: char *return_value;
1629: char result[MAX_LENGTH];
1630: char name[MAX_LENGTH];
1631: char rel[MAX_LENGTH];
1632: char *p, *access;
1.29 cvs 1633: HTURI given, related;
1634: int len;
1.106 cvs 1635: char used_sep;
1636: char *used_str;
1.32 cvs 1637:
1.106 cvs 1638: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1639: {
1.106 cvs 1640: used_str = DIR_STR;
1641: used_sep = DIR_SEP;
1.33 cvs 1642: }
1.32 cvs 1643: else
1.33 cvs 1644: {
1.106 cvs 1645: used_str = URL_STR;
1646: used_sep = URL_SEP;
1.33 cvs 1647: }
1.32 cvs 1648:
1.29 cvs 1649: /* Make working copies of input strings to cut up: */
1650: return_value = NULL;
1651: result[0] = 0; /* Clear string */
1.106 cvs 1652: strcpy (name, aName);
1.29 cvs 1653: if (relatedName != NULL)
1.106 cvs 1654: strcpy (rel, relatedName);
1.29 cvs 1655: else
1.106 cvs 1656: relatedName[0] = EOS;
1.29 cvs 1657:
1658: scan (name, &given);
1659: scan (rel, &related);
1660: access = given.access ? given.access : related.access;
1661: if (wanted & AMAYA_PARSE_ACCESS)
1662: if (access)
1663: {
1.106 cvs 1664: strcat (result, access);
1.29 cvs 1665: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1666: strcat (result, ":");
1.29 cvs 1667: }
1668:
1669: if (given.access && related.access)
1670: /* If different, inherit nothing. */
1.106 cvs 1671: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1672: {
1673: related.host = 0;
1674: related.absolute = 0;
1675: related.relative = 0;
1676: related.fragment = 0;
1677: }
1678:
1679: if (wanted & AMAYA_PARSE_HOST)
1680: if(given.host || related.host)
1681: {
1682: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1683: strcat (result, "//");
1684: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1685: }
1686:
1687: if (given.host && related.host)
1688: /* If different hosts, inherit no path. */
1.106 cvs 1689: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1690: {
1691: related.absolute = 0;
1692: related.relative = 0;
1693: related.fragment = 0;
1694: }
1695:
1696: if (wanted & AMAYA_PARSE_PATH)
1697: {
1698: if (given.absolute)
1699: {
1700: /* All is given */
1701: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1702: strcat (result, used_str);
1703: strcat (result, given.absolute);
1.25 cvs 1704: }
1.29 cvs 1705: else if (related.absolute)
1706: {
1707: /* Adopt path not name */
1.106 cvs 1708: strcat (result, used_str);
1709: strcat (result, related.absolute);
1.29 cvs 1710: if (given.relative)
1711: {
1712: /* Search part? */
1.106 cvs 1713: p = strchr (result, '?');
1.29 cvs 1714: if (!p)
1.106 cvs 1715: p=result+strlen(result)-1;
1.33 cvs 1716: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1717: /* Remove filename */
1718: p[1]=0;
1719: /* Add given one */
1.106 cvs 1720: strcat (result, given.relative);
1.25 cvs 1721: }
1722: }
1.29 cvs 1723: else if (given.relative)
1724: /* what we've got */
1.106 cvs 1725: strcat (result, given.relative);
1.29 cvs 1726: else if (related.relative)
1.106 cvs 1727: strcat (result, related.relative);
1.29 cvs 1728: else
1729: /* No inheritance */
1.106 cvs 1730: strcat (result, used_str);
1.25 cvs 1731: }
1.29 cvs 1732:
1733: if (wanted & AMAYA_PARSE_ANCHOR)
1734: if (given.fragment || related.fragment)
1735: {
1736: if (given.absolute && given.fragment)
1737: {
1738: /*Fixes for relURLs...*/
1739: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1740: strcat (result, "#");
1741: strcat (result, given.fragment);
1.29 cvs 1742: }
1743: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1744: strcat (result, "");
1.29 cvs 1745: else
1746: {
1.110 cvs 1747: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1748: strcat (result, "#");
1.110 cvs 1749: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1750: }
1751: }
1.106 cvs 1752: len = strlen (result);
1753: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1754: strcpy (return_value, result);
1.29 cvs 1755: return (return_value); /* exactly the right length */
1.25 cvs 1756: }
1757:
1758: /*----------------------------------------------------------------------
1759: HTCanon
1760: Canonicalizes the URL in the following manner starting from the host
1761: pointer:
1762:
1763: 1) The host name is converted to lowercase
1764: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1765:
1766: Return: OK The position of the current path part of the URL
1767: which might be the old one or a new one.
1768:
1769: ----------------------------------------------------------------------*/
1.106 cvs 1770: static char *HTCanon (char **filename, char *host)
1771: {
1772: char *newname = NULL;
1773: char used_sep;
1774: char *path;
1775: char *strptr;
1776: char *port;
1777: char *access = host-3;
1778:
1779: if (*filename && strchr (*filename, URL_SEP))
1780: used_sep = URL_SEP;
1.33 cvs 1781: else
1.106 cvs 1782: used_sep = DIR_SEP;
1.32 cvs 1783:
1.110 cvs 1784: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1785: access--;
1.110 cvs 1786: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1787: path = host + strlen (host);
1788: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1789: host = strptr;
1.110 cvs 1790: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1791: port = NULL;
1.25 cvs 1792:
1793: strptr = host; /* Convert to lower-case */
1.82 cvs 1794: while (strptr < path)
1.33 cvs 1795: {
1.123 vatton 1796: *strptr = tolower (*strptr);
1.82 cvs 1797: strptr++;
1.33 cvs 1798: }
1.25 cvs 1799:
1800: /* Does the URL contain a full domain name? This also works for a
1801: numerical host name. The domain name is already made lower-case
1802: and without a trailing dot. */
1803: {
1.106 cvs 1804: char *dot = port ? port : path;
1805: if (dot > *filename && *--dot == '.')
1.33 cvs 1806: {
1.106 cvs 1807: char *orig = dot;
1808: char *dest = dot + 1;
1.82 cvs 1809: while ((*orig++ = *dest++));
1810: if (port) port--;
1.33 cvs 1811: path--;
1.25 cvs 1812: }
1813: }
1814: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1815: if (port)
1816: {
1.82 cvs 1817: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1818: {
1819: if (!newname)
1820: {
1.106 cvs 1821: char *orig = port;
1822: char *dest = port + 1;
1.82 cvs 1823: while ((*orig++ = *dest++));
1.33 cvs 1824: }
1825: }
1.106 cvs 1826: else if ((!strncmp (access, "http", 4) &&
1827: (*(port + 1) == '8' &&
1828: *(port+2) == '0' &&
1.82 cvs 1829: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1830: (!strncmp (access, "gopher", 6) &&
1831: (*(port+1) == '7' &&
1832: *(port+2) == '0' &&
1.82 cvs 1833: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1834: (!strncmp (access, "ftp", 3) &&
1835: (*(port+1) == '2' &&
1836: *(port + 2) == '1' &&
1.82 cvs 1837: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1838: if (!newname)
1839: {
1.106 cvs 1840: char *orig = port;
1841: char *dest = port + 3;
1.33 cvs 1842: while((*orig++ = *dest++));
1843: /* Update path position, Henry Minsky */
1844: path -= 3;
1.25 cvs 1845: }
1.33 cvs 1846: }
1847: else if (newname)
1.106 cvs 1848: strncat (newname, port, (int) (path - port));
1.33 cvs 1849: }
1.25 cvs 1850:
1.33 cvs 1851: if (newname)
1852: {
1.106 cvs 1853: char *newpath = newname + strlen (newname);
1854: strcat (newname, path);
1.25 cvs 1855: path = newpath;
1.28 cvs 1856: /* Free old copy */
1857: TtaFreeMemory(*filename);
1.25 cvs 1858: *filename = newname;
1.33 cvs 1859: }
1.25 cvs 1860: return path;
1861: }
1862:
1863:
1864: /*----------------------------------------------------------------------
1.29 cvs 1865: SimplifyUrl: simplify a URI
1.32 cvs 1866: A URI is allowed to contain the sequence xxx/../ which may be
1867: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1868: Simplification helps us recognize duplicate URIs.
1.25 cvs 1869:
1.28 cvs 1870: Thus, /etc/junk/../fred becomes /etc/fred
1871: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1872:
1.28 cvs 1873: but we should NOT change
1874: http://fred.xxx.edu/../..
1.25 cvs 1875:
1876: or ../../albert.html
1877:
1.28 cvs 1878: In order to avoid empty URLs the following URLs become:
1.25 cvs 1879:
1880: /fred/.. becomes /fred/..
1881: /fred/././.. becomes /fred/..
1882: /fred/.././junk/.././ becomes /fred/..
1883:
1.28 cvs 1884: If more than one set of `://' is found (several proxies in cascade) then
1885: only the part after the last `://' is simplified.
1.25 cvs 1886:
1.28 cvs 1887: Returns: A string which might be the old one or a new one.
1.25 cvs 1888: ----------------------------------------------------------------------*/
1.106 cvs 1889: void SimplifyUrl (char **url)
1890: {
1891: char *path;
1892: char *access;
1893: char *newptr;
1894: char *p;
1895: char *orig, *dest, *end;
1.28 cvs 1896:
1.106 cvs 1897: char used_sep;
1.77 cvs 1898: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1899: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1900:
1.28 cvs 1901: if (!url || !*url)
1902: return;
1903:
1.106 cvs 1904: if (strchr (*url, URL_SEP))
1905: used_sep = URL_SEP;
1.32 cvs 1906: else
1.106 cvs 1907: used_sep = DIR_SEP;
1.32 cvs 1908:
1.77 cvs 1909: /* should we simplify double dot? */
1910: path = *url;
1.106 cvs 1911: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1912: ddot_simplify = FALSE;
1913: else
1914: ddot_simplify = TRUE;
1915:
1.28 cvs 1916: /* Find any scheme name */
1.106 cvs 1917: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1918: {
1919: /* Find host name */
1.28 cvs 1920: access = *url;
1.123 vatton 1921: while (access < path && (*access = tolower (*access)))
1.82 cvs 1922: access++;
1.28 cvs 1923: path += 3;
1.106 cvs 1924: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1925: /* For proxies */
1.106 cvs 1926: path = newptr + 3;
1.82 cvs 1927: /* We have a host name */
1.84 cvs 1928: path = HTCanon (url, path);
1.25 cvs 1929: }
1.106 cvs 1930: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1931: path += 2;
1932: else
1933: path = *url;
1.84 cvs 1934: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1935: /* Some URLs start //<foo> */
1936: path += 1;
1.94 cvs 1937: else if (IsFilePath (path))
1938: {
1939: /* doesn't need to do anything more */
1940: return;
1941: }
1.106 cvs 1942: else if (!strncmp (path, "news:", 5))
1.28 cvs 1943: {
1.106 cvs 1944: newptr = strchr (path+5, '@');
1.28 cvs 1945: if (!newptr)
1946: newptr = path + 5;
1947: while (*newptr)
1948: {
1949: /* Make group or host lower case */
1.123 vatton 1950: *newptr = tolower (*newptr);
1.28 cvs 1951: newptr++;
1.25 cvs 1952: }
1.28 cvs 1953: /* Doesn't need to do any more */
1954: return;
1.25 cvs 1955: }
1.130 cheyroul 1956:
1.126 cheyroul 1957:
1.28 cvs 1958: if ((p = path))
1959: {
1.106 cvs 1960: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1961: (end = strchr (path, '#'))))
1962: end = path + strlen (path);
1.28 cvs 1963:
1964: /* Parse string second time to simplify */
1965: p = path;
1966: while (p < end)
1967: {
1.110 cvs 1968: /* if we're pointing to a char, it's safe to reactivate the
1969: ../ convertion */
1.106 cvs 1970: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1971: ddot_simplify = TRUE;
1972:
1.33 cvs 1973: if (*p==used_sep)
1.28 cvs 1974: {
1.106 cvs 1975: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1976: {
1977: orig = p + 1;
1.84 cvs 1978: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1979: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1980: end = orig - 1;
1981: }
1.106 cvs 1982: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1983: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1984: {
1985: newptr = p;
1.52 cvs 1986: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1987: if (*newptr == used_sep)
1988: orig = newptr + 1;
1.28 cvs 1989: else
1.52 cvs 1990: orig = newptr;
1991:
1992: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1993: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1994: end = orig-1;
1995: /* Start again with prev slash */
1996: p = newptr;
1.28 cvs 1997: }
1.33 cvs 1998: else if (*(p+1) == used_sep)
1.28 cvs 1999: {
1.33 cvs 2000: while (*(p+1) == used_sep)
1.28 cvs 2001: {
2002: orig = p;
2003: dest = p + 1;
2004: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2005: end = orig-1;
2006: }
2007: }
2008: else
1.25 cvs 2009: p++;
1.28 cvs 2010: }
2011: else
2012: p++;
1.25 cvs 2013: }
2014: }
1.51 cvs 2015: /*
2016: ** Check for host/../.. kind of things
2017: */
1.106 cvs 2018: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 2019: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 2020: *(path+1) = EOS;
1.28 cvs 2021: return;
2022: }
2023:
2024:
2025: /*----------------------------------------------------------------------
1.96 cvs 2026: NormalizeFile normalizes local names.
1.28 cvs 2027: Return TRUE if target and src differ.
2028: ----------------------------------------------------------------------*/
1.106 cvs 2029: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2030: {
1.110 cvs 2031: #ifndef _WINDOWS
1.106 cvs 2032: char *s;
1.93 cvs 2033: int i;
1.110 cvs 2034: #endif /* !_WINDOWS */
1.82 cvs 2035: ThotBool change;
1.90 cvs 2036: int start_index; /* the first char that we'll copy */
1.28 cvs 2037:
1.54 cvs 2038: change = FALSE;
1.90 cvs 2039: start_index = 0;
2040:
1.106 cvs 2041: if (!src || src[0] == EOS)
1.96 cvs 2042: {
1.106 cvs 2043: target[0] = EOS;
1.96 cvs 2044: return FALSE;
2045: }
1.90 cvs 2046:
2047: /* @@ do I need file: or file:/ here? */
1.106 cvs 2048: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 2049: {
1.90 cvs 2050: /* remove the prefix file: */
2051: start_index += 5;
2052:
2053: /* remove the localhost prefix */
1.106 cvs 2054: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 2055: start_index += 11;
2056:
2057: /* remove the first two slashes in / / /path */
2058: while (src[start_index] &&
1.106 cvs 2059: src[start_index] == '/'
2060: && src[start_index + 1] == '/')
1.94 cvs 2061: start_index++;
2062:
2063: #ifdef _WINDOWS
2064: /* remove any extra slash before the drive name */
1.106 cvs 2065: if (src[start_index] == '/'
2066: &&src[start_index+2] == ':')
1.94 cvs 2067: start_index++;
2068: #endif /* _WINDOWS */
1.90 cvs 2069:
1.106 cvs 2070: if (src[start_index] == EOS)
1.90 cvs 2071: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 2072: strcpy (target, DIR_STR);
1.90 cvs 2073: else
1.97 cvs 2074: /* as we're inside a file: URL, we'll apply all the convertions
2075: we know */
2076: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 2077:
2078: change = TRUE;
2079: }
1.97 cvs 2080: else if (convertion != AM_CONV_NONE)
1.96 cvs 2081: {
2082: /* we are following a "local" relative link, we do all the
2083: convertions except for the HOME_DIR ~ one */
1.97 cvs 2084: CleanCopyFileURL (target, src, convertion);
1.28 cvs 2085: }
1.90 cvs 2086: #ifndef _WINDOWS
1.106 cvs 2087: else if (src[0] == '~')
1.53 cvs 2088: {
1.96 cvs 2089: /* it must be a URL typed in a text input field */
2090: /* do the HOME_DIR ~ substitution */
1.82 cvs 2091: s = TtaGetEnvString ("HOME");
1.106 cvs 2092: strcpy (target, s);
1.90 cvs 2093: #if 0
1.96 cvs 2094: /* JK: invalidated this part of the code as it's simpler
2095: to add the DIR_SEP whenever we have something to add
2096: to the path rather than adding it systematically */
1.106 cvs 2097: if (src[1] != DIR_SEP)
2098: strcat (target, DIR_STR);
1.90 cvs 2099: #endif
1.106 cvs 2100: i = strlen (target);
2101: strcpy (&target[i], &src[1]);
1.54 cvs 2102: change = TRUE;
1.53 cvs 2103: }
1.90 cvs 2104: #endif /* _WINDOWS */
1.28 cvs 2105: else
1.96 cvs 2106: /* leave it as it is */
1.106 cvs 2107: strcpy (target, src);
1.96 cvs 2108:
1.28 cvs 2109: /* remove /../ and /./ */
1.29 cvs 2110: SimplifyUrl (&target);
1.54 cvs 2111: if (!change)
1.106 cvs 2112: change = strcmp (src, target);
1.28 cvs 2113: return (change);
1.25 cvs 2114: }
2115:
1.28 cvs 2116:
1.25 cvs 2117: /*----------------------------------------------------------------------
1.31 cvs 2118: MakeRelativeURL: make relative name
1.25 cvs 2119:
1.28 cvs 2120: This function creates and returns a string which gives an expression of
2121: one address as related to another. Where there is no relation, an absolute
2122: address is retured.
1.25 cvs 2123:
1.28 cvs 2124: On entry,
1.25 cvs 2125: Both names must be absolute, fully qualified names of nodes
2126: (no fragment bits)
2127:
1.28 cvs 2128: On exit,
1.25 cvs 2129: The return result points to a newly allocated name which, if
2130: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2131: The caller is responsible for freeing the resulting name later.
2132: ----------------------------------------------------------------------*/
1.106 cvs 2133: char *MakeRelativeURL (char *aName, char *relatedName)
2134: {
2135: char *return_value;
2136: char result[MAX_LENGTH];
2137: char *p;
2138: char *q;
2139: char *after_access;
2140: char *last_slash = NULL;
2141: int slashes, levels, len;
1.110 cvs 2142: #ifdef _WINDOWS
1.44 cvs 2143: int ndx;
1.110 cvs 2144: #endif /* _WINDOWS */
1.44 cvs 2145:
1.29 cvs 2146: if (aName == NULL || relatedName == NULL)
2147: return (NULL);
2148:
2149: slashes = 0;
2150: after_access = NULL;
2151: p = aName;
2152: q = relatedName;
2153: for (; *p && (*p == *q); p++, q++)
1.27 cvs 2154: {
2155: /* Find extent of match */
1.106 cvs 2156: if (*p == ':')
1.29 cvs 2157: after_access = p + 1;
1.28 cvs 2158: if (*p == DIR_SEP)
1.27 cvs 2159: {
1.29 cvs 2160: /* memorize the last slash position and count them */
1.27 cvs 2161: last_slash = p;
2162: slashes++;
1.25 cvs 2163: }
2164: }
2165:
1.31 cvs 2166: /* q, p point to the first non-matching character or zero */
1.106 cvs 2167: if (*q == EOS)
1.31 cvs 2168: {
2169: /* New name is a subset of the related name */
2170: /* exactly the right length */
1.106 cvs 2171: len = strlen (p);
2172: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2173: strcpy (return_value, p);
1.31 cvs 2174: }
2175: else if ((slashes < 2 && after_access == NULL)
2176: || (slashes < 3 && after_access != NULL))
2177: {
2178: /* Two names whitout common path */
2179: /* exactly the right length */
1.106 cvs 2180: len = strlen (aName);
2181: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2182: strcpy (return_value, aName);
1.31 cvs 2183: }
2184: else
2185: {
2186: /* Some path in common */
1.106 cvs 2187: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2188: /* just the same server */
1.106 cvs 2189: strcpy (result, last_slash);
1.31 cvs 2190: else
2191: {
2192: levels= 0;
1.106 cvs 2193: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2194: if (*q == DIR_SEP)
2195: levels++;
2196:
1.106 cvs 2197: result[0] = EOS;
1.31 cvs 2198: for (;levels; levels--)
1.106 cvs 2199: strcat (result, "../");
2200: strcat (result, last_slash+1);
1.31 cvs 2201: }
1.52 cvs 2202:
2203: if (!*result)
1.106 cvs 2204: strcat (result, "./");
1.52 cvs 2205:
1.31 cvs 2206: /* exactly the right length */
1.106 cvs 2207: len = strlen (result);
2208: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2209: strcpy (return_value, result);
1.52 cvs 2210:
1.25 cvs 2211: }
1.110 cvs 2212: #ifdef _WINDOWS
1.106 cvs 2213: len = strlen (return_value);
1.44 cvs 2214: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2215: if (return_value[ndx] == '\\')
2216: return_value[ndx] = '/' ;
1.110 cvs 2217: #endif /* _WINDOWS */
1.29 cvs 2218: return (return_value);
1.24 cvs 2219: }
1.35 cvs 2220:
1.104 kahan 2221: /*----------------------------------------------------------------------
2222: AM_GetFileSize
2223: Returns TRUE and the filesize in the 2nd parameter.
2224: Otherwise, in case of a system error, returns FALSE, with a
2225: filesize of 0L.
2226: ---------------------------------------------------------------------*/
1.106 cvs 2227: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2228: {
1.106 cvs 2229: ThotFileHandle handle = ThotFile_BADHANDLE;
2230: ThotFileInfo info;
1.35 cvs 2231:
1.104 kahan 2232: *file_size = 0L;
2233: if (!TtaFileExist (filename))
2234: return FALSE;
2235:
2236: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2237: if (handle == ThotFile_BADHANDLE)
2238: /* ThotFile_BADHANDLE */
2239: return FALSE;
2240: if (TtaFileStat (handle, &info) == 0)
2241: /* bad stat */
2242: info.size = 0L;
2243: TtaFileClose (handle);
2244: *file_size = (unsigned long) info.size;
2245: return TRUE;
2246: }
1.139 kahan 2247:
2248: /*----------------------------------------------------------------------
2249: AM_UseXHTMLMimeType
2250: Returns TRUE if the user has configured Amaya to use this MIME type,
2251: FALSE otherwise.
2252: ---------------------------------------------------------------------*/
2253: ThotBool AM_UseXHTMLMimeType (void)
2254: {
2255: ThotBool xhtml_mimetype;
2256:
2257: /* does the user wants to use the new MIME type? */
2258: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2259:
2260: return (xhtml_mimetype);
2261: }
Webmaster