Annotation of Amaya/amaya/AHTURLTools.c, revision 1.142
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:
1.75 cvs 246:
247: /*----------------------------------------------------------------------
1.11 cvs 248: ExplodeURL
1.8 cvs 249: ----------------------------------------------------------------------*/
1.106 cvs 250: void ExplodeURL (char *url, char **proto, char **host, char **dir,
251: char **file)
1.8 cvs 252: {
1.33 cvs 253: char *curr, *temp;
254: char used_sep;
1.32 cvs 255:
1.33 cvs 256: if (url && strchr (url, URL_SEP))
257: used_sep = URL_SEP;
258: else
259: used_sep = DIR_SEP;
1.8 cvs 260:
261: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
262: (dir == NULL) || (file == NULL))
263: return;
264:
265: /* initialize every pointer */
266: *proto = *host = *dir = *file = NULL;
267:
268: /* skip any leading space */
269: while ((*url == SPACE) || (*url == TAB))
270: url++;
1.9 cvs 271: curr = url;
272: if (*curr == 0)
1.8 cvs 273: goto finished;
274:
275: /* go to the end of the URL */
1.68 cvs 276: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
277: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 278: curr++;
1.8 cvs 279:
280: /* mark the end of the chain */
1.9 cvs 281: *curr = EOS;
282: curr--;
283: if (curr <= url)
1.8 cvs 284: goto finished;
285:
286: /* search the next DIR_SEP indicating the beginning of the file name */
287: do
1.11 cvs 288: curr--;
1.33 cvs 289: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 290:
1.9 cvs 291: if (curr < url)
1.8 cvs 292: goto finished;
1.9 cvs 293: *file = curr + 1;
1.8 cvs 294:
295: /* mark the end of the dir */
1.9 cvs 296: *curr = EOS;
297: curr--;
298: if (curr < url)
1.8 cvs 299: goto finished;
300:
1.29 cvs 301: /* search for the DIR_STR indicating the host name start */
1.33 cvs 302: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 303: curr--;
1.8 cvs 304:
305: /* if we found it, separate the host name from the directory */
1.102 kahan 306: if ((*curr == used_sep) && (*(curr + 1) == used_sep))
1.8 cvs 307: {
1.9 cvs 308: *host = temp = curr + 2;
1.33 cvs 309: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 310: temp++;
1.33 cvs 311: if (*temp == used_sep)
1.8 cvs 312: {
313: *temp = EOS;
314: *dir = temp + 1;
315: }
316: }
317: else
1.11 cvs 318: *dir = curr;
319:
1.9 cvs 320: if (curr <= url)
1.8 cvs 321: goto finished;
322:
323: /* mark the end of the proto */
1.9 cvs 324: *curr = EOS;
325: curr--;
326: if (curr < url)
1.8 cvs 327: goto finished;
328:
1.106 cvs 329: if (*curr == ':')
1.8 cvs 330: {
1.9 cvs 331: *curr = EOS;
332: curr--;
1.8 cvs 333: }
334: else
335: goto finished;
1.11 cvs 336:
1.9 cvs 337: if (curr < url)
1.8 cvs 338: goto finished;
1.9 cvs 339: while ((curr > url) && (isalpha (*curr)))
340: curr--;
341: *proto = curr;
1.8 cvs 342:
343: finished:;
344:
345: #ifdef AMAYA_DEBUG
346: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
347: if (*proto)
348: fprintf (stderr, "proto : %s, ", *proto);
349: if (*host)
350: fprintf (stderr, "host : %s, ", *host);
351: if (*dir)
352: fprintf (stderr, "dir : %s, ", *dir);
353: if (*file)
354: fprintf (stderr, "file : %s ", *file);
355: fprintf (stderr, "\n");
356: #endif
357:
358: }
1.3 cvs 359:
1.116 kahan 360: /*----------------------------------------------------------------------
361: PicTypeToMime
362: Converts a Thot PicType into the equivalent MIME type. If no convertion
363: is possible, it returns NULL.
364: ----------------------------------------------------------------------*/
365: char *PicTypeToMIME (PicType contentType)
366: {
367: char *mime_type;
368:
369: switch (contentType)
370: {
371: case xbm_type:
372: mime_type ="image/x-xbitmap";
373: break;
374: case eps_type:
375: mime_type ="application/postscript";
376: break;
377: case xpm_type:
378: mime_type ="image/x-xpicmap";
379: break;
380: case gif_type:
381: mime_type ="image/gif";
382: break;
383: case jpeg_type:
384: mime_type ="image/jpeg";
385: break;
386: case png_type:
387: mime_type ="image/png";
388: break;
389: case svg_type:
1.138 kahan 390: mime_type =AM_SVG_MIME_TYPE;
1.116 kahan 391: break;
392: case unknown_type:
393: default:
394: mime_type = NULL;
395: }
396:
397: return mime_type;
398: }
1.61 cvs 399:
400: /*----------------------------------------------------------------------
1.117 kahan 401: ImageElement
402: Returns the element (image parameter) and URL (url parameter) of an
403: image in a docImage document. The user must free the memory associated
1.120 kahan 404: with the url parameter if the function is succesful.
405: If the url parameter is NULL, we won't initialize it.
1.117 kahan 406: Returns TRUE if succesful, FALSE otherwise.
407: ----------------------------------------------------------------------*/
408: ThotBool ImageElement (Document doc, char **url, Element *image)
409: {
410: Element el, imgEl;
411: Attribute attr, srcAttr;
412: AttributeType attrType;
413: int length;
414: char *value;
415:
416: if (DocumentTypes[doc] != docImage)
417: return FALSE;
418:
419: /* find the value of the src attribute */
420: attrType.AttrSSchema = TtaGetSSchema ("HTML", doc);
421: attrType.AttrTypeNum = HTML_ATTR_SRC;
422: el = TtaGetRootElement (doc);
423: TtaSearchAttribute (attrType, SearchInTree, el, &imgEl, &srcAttr);
424:
425: if (!imgEl)
426: return FALSE;
427: *image = imgEl;
428:
1.120 kahan 429: if (url)
430: {
431: attr = TtaGetAttribute (imgEl, attrType);
432: length = TtaGetTextAttributeLength (srcAttr) + 1;
433: value = TtaGetMemory (length);
434: TtaGiveTextAttributeValue (srcAttr, value, &length);
435: *url = value;
436: }
1.117 kahan 437: return TRUE;
438: }
439:
440: /*----------------------------------------------------------------------
441: DocImageMimeType
442: Returns the MIME type of a docImage document.
443: ----------------------------------------------------------------------*/
444: char *DocImageMimeType (Document doc)
445: {
446: char *mime_type;
447: LoadedImageDesc *pImage;
448: PicType type;
449: Element image;
450:
451: if (DocumentTypes[doc] != docImage)
452: return NULL;
453:
454: mime_type = NULL;
455: if (!IsHTTPPath (DocumentURLs[doc]))
456: {
457: /* it is a local image */
1.120 kahan 458: if (ImageElement (doc, NULL, &image))
1.117 kahan 459: {
460: type = TtaGetPictureType (image);
461: mime_type = PicTypeToMIME (type);
462: }
463: }
464: else
465: {
466: /* find the value of the src attribute */
467: pImage = ImageURLs;
468: while (pImage != NULL)
469: {
470: if (pImage->document == doc)
471: {
472: if (pImage->content_type)
473: mime_type = pImage->content_type;
474: else if (pImage->elImage && pImage->elImage->currentElement)
475: {
476: type = TtaGetPictureType (pImage->elImage->currentElement);
477: mime_type = PicTypeToMIME (type);
478: }
479: break;
480: }
481: }
482: }
483: return (mime_type);
484: }
485:
1.4 cvs 486: /*----------------------------------------------------------------------
1.9 cvs 487: IsHTMLName
488: returns TRUE if path points to an HTML resource.
1.4 cvs 489: ----------------------------------------------------------------------*/
1.109 cvs 490: ThotBool IsHTMLName (const char *path)
1.106 cvs 491: {
1.136 cvs 492: char temppath[MAX_LENGTH];
493: char suffix[MAX_LENGTH];
494: char nsuffix[MAX_LENGTH];
495: int i;
1.5 cvs 496:
1.101 cvs 497: if (!path)
498: return (FALSE);
1.5 cvs 499:
1.106 cvs 500: strcpy (temppath, path);
1.124 vatton 501: TtaExtractSuffix (temppath, suffix);
1.101 cvs 502: i = 0;
1.106 cvs 503: while (suffix[i] != EOS)
1.101 cvs 504: {
505: /* Normalize the suffix */
506: i = 0;
1.106 cvs 507: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 508: {
1.123 vatton 509: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 510: i++;
511: }
1.106 cvs 512: nsuffix[i] = EOS;
513: if (!strcmp (nsuffix, "html") ||
514: !strcmp (nsuffix, "htm") ||
515: !strcmp (nsuffix, "shtml") ||
516: !strcmp (nsuffix, "jsp") ||
517: !strcmp (nsuffix, "xht") ||
518: !strcmp (nsuffix, "xhtm") ||
519: !strcmp (nsuffix, "xhtml"))
1.101 cvs 520: return (TRUE);
1.106 cvs 521: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 522: {
523: /* take into account compressed files */
1.124 vatton 524: TtaExtractSuffix (temppath, suffix);
1.101 cvs 525: /* Normalize the suffix */
526: i = 0;
1.106 cvs 527: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 528: {
1.123 vatton 529: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 530: i++;
531: }
1.106 cvs 532: nsuffix[i] = EOS;
533: if (!strcmp (nsuffix, "html") ||
534: !strcmp (nsuffix, "htm") ||
535: !strcmp (nsuffix, "shtml") ||
536: !strcmp (nsuffix, "jsp") ||
537: !strcmp (nsuffix, "xht") ||
538: !strcmp (nsuffix, "xhtm") ||
539: !strcmp (nsuffix, "xhtml"))
1.101 cvs 540: return (TRUE);
541: else
542: return (FALSE);
543: }
544: else
545: /* check if there is another suffix */
1.124 vatton 546: TtaExtractSuffix (temppath, suffix);
1.101 cvs 547: }
1.88 cvs 548: return (FALSE);
1.3 cvs 549: }
550:
1.4 cvs 551: /*----------------------------------------------------------------------
1.136 cvs 552: IsMathMLName
553: returns TRUE if path points to an MathML resource.
1.56 cvs 554: ----------------------------------------------------------------------*/
1.136 cvs 555: ThotBool IsMathMLName (const char *path)
1.56 cvs 556: {
1.136 cvs 557: char temppath[MAX_LENGTH];
558: char suffix[MAX_LENGTH];
1.56 cvs 559:
560: if (!path)
561: return (FALSE);
562:
1.106 cvs 563: strcpy (temppath, path);
1.124 vatton 564: TtaExtractSuffix (temppath, suffix);
1.56 cvs 565:
1.136 cvs 566: if (!strcasecmp (suffix, "mml"))
1.56 cvs 567: return (TRUE);
1.106 cvs 568: else if (!strcmp (suffix, "gz"))
1.56 cvs 569: {
570: /* take into account compressed files */
1.124 vatton 571: TtaExtractSuffix (temppath, suffix);
1.136 cvs 572: if (!strcasecmp (suffix, "mml"))
1.60 cvs 573: return (TRUE);
574: else
575: return (FALSE);
576: }
577: else
578: return (FALSE);
579: }
580:
581: /*----------------------------------------------------------------------
1.136 cvs 582: IsSVGName
583: returns TRUE if path points to an SVG resource.
1.133 vatton 584: ----------------------------------------------------------------------*/
1.136 cvs 585: ThotBool IsSVGName (const char *path)
1.133 vatton 586: {
1.136 cvs 587: char temppath[MAX_LENGTH];
588: char suffix[MAX_LENGTH];
1.133 vatton 589:
590: if (!path)
591: return (FALSE);
592:
593: strcpy (temppath, path);
594: TtaExtractSuffix (temppath, suffix);
595:
1.136 cvs 596: if (!strcasecmp (suffix, "svg"))
1.133 vatton 597: return (TRUE);
598: else if (!strcmp (suffix, "gz"))
599: {
600: /* take into account compressed files */
601: TtaExtractSuffix (temppath, suffix);
1.136 cvs 602: if (!strcasecmp (suffix, "svg"))
1.133 vatton 603: return (TRUE);
604: else
605: return (FALSE);
606: }
607: else
608: return (FALSE);
609: }
610:
611: /*----------------------------------------------------------------------
1.136 cvs 612: IsXMLName
613: returns TRUE if path points to an XML resource.
1.103 cvs 614: ----------------------------------------------------------------------*/
1.136 cvs 615: ThotBool IsXMLName (const char *path)
1.103 cvs 616: {
1.136 cvs 617: char temppath[MAX_LENGTH];
618: char suffix[MAX_LENGTH];
1.103 cvs 619:
620: if (!path)
621: return (FALSE);
622:
1.106 cvs 623: strcpy (temppath, path);
1.124 vatton 624: TtaExtractSuffix (temppath, suffix);
1.103 cvs 625:
1.136 cvs 626: if (!strcasecmp (suffix, "xml") ||
627: !strcasecmp (suffix, "xht") ||
628: !strcmp (suffix, "xhtm") ||
629: !strcmp (suffix, "xhtml"))
1.103 cvs 630: return (TRUE);
1.106 cvs 631: else if (!strcmp (suffix, "gz"))
1.103 cvs 632: {
633: /* take into account compressed files */
1.124 vatton 634: TtaExtractSuffix (temppath, suffix);
1.136 cvs 635: if (!strcasecmp (suffix, "xml") ||
636: !strcasecmp (suffix, "xht") ||
637: !strcmp (suffix, "xhtm") ||
638: !strcmp (suffix, "xhtml"))
1.103 cvs 639: return (TRUE);
640: else
641: return (FALSE);
642: }
643: else
644: return (FALSE);
645: }
646:
647: /*----------------------------------------------------------------------
1.136 cvs 648: IsUndisplayedName
649: returns TRUE if path points to an undisplayed resource.
1.103 cvs 650: ----------------------------------------------------------------------*/
1.136 cvs 651: ThotBool IsUndisplayedName (const char *path)
1.103 cvs 652: {
1.106 cvs 653: char temppath[MAX_LENGTH];
654: char suffix[MAX_LENGTH];
1.103 cvs 655:
656: if (!path)
657: return (FALSE);
658:
1.106 cvs 659: strcpy (temppath, path);
1.124 vatton 660: TtaExtractSuffix (temppath, suffix);
1.103 cvs 661:
1.136 cvs 662: if (!strcasecmp (suffix, "exe") ||
663: !strcasecmp (suffix, "zip") ||
664: !strcasecmp (suffix, "ppt") ||
665: !strcasecmp (suffix, "pdf") ||
666: !strcasecmp (suffix, "ps") ||
667: !strcasecmp (suffix, "eps") ||
668: !strcasecmp (suffix, "tar") ||
669: !strcasecmp (suffix, "tgz") ||
670: !strcasecmp (suffix, "ddl") ||
671: !strcasecmp (suffix, "o"))
1.103 cvs 672: return (TRUE);
1.106 cvs 673: else if (!strcmp (suffix, "gz"))
1.103 cvs 674: {
675: /* take into account compressed files */
1.124 vatton 676: TtaExtractSuffix (temppath, suffix);
1.136 cvs 677: if (!strcasecmp (suffix, "exe") ||
678: !strcasecmp (suffix, "zip") ||
679: !strcasecmp (suffix, "ppt") ||
680: !strcasecmp (suffix, "pdf") ||
681: !strcasecmp (suffix, "ps") ||
682: !strcasecmp (suffix, "eps") ||
683: !strcasecmp (suffix, "tar") ||
684: !strcasecmp (suffix, "ddl") ||
685: !strcasecmp (suffix, "o"))
1.103 cvs 686: return (TRUE);
687: else
688: return (FALSE);
689: }
690: else
691: return (FALSE);
692: }
693:
694: /*----------------------------------------------------------------------
1.60 cvs 695: IsCSSName
696: returns TRUE if path points to an XML resource.
697: ----------------------------------------------------------------------*/
1.111 cvs 698: ThotBool IsCSSName (const char *path)
1.60 cvs 699: {
1.106 cvs 700: char temppath[MAX_LENGTH];
701: char suffix[MAX_LENGTH];
1.60 cvs 702:
703: if (!path)
704: return (FALSE);
705:
1.106 cvs 706: strcpy (temppath, path);
1.124 vatton 707: TtaExtractSuffix (temppath, suffix);
1.60 cvs 708:
1.106 cvs 709: if (!strcasecmp (suffix, "css"))
1.60 cvs 710: return (TRUE);
1.106 cvs 711: else if (!strcmp (suffix, "gz"))
1.60 cvs 712: {
713: /* take into account compressed files */
1.124 vatton 714: TtaExtractSuffix (temppath, suffix);
1.106 cvs 715: if (!strcasecmp (suffix, "css"))
1.56 cvs 716: return (TRUE);
717: else
718: return (FALSE);
719: }
720: else
721: return (FALSE);
722: }
723:
724: /*----------------------------------------------------------------------
1.9 cvs 725: IsImageName
726: returns TRUE if path points to an image resource.
1.4 cvs 727: ----------------------------------------------------------------------*/
1.111 cvs 728: ThotBool IsImageName (const char *path)
1.106 cvs 729: {
730: char temppath[MAX_LENGTH];
731: char suffix[MAX_LENGTH];
732: char nsuffix[MAX_LENGTH];
1.5 cvs 733: int i;
734:
735: if (!path)
1.13 cvs 736: return (FALSE);
1.5 cvs 737:
1.106 cvs 738: strcpy (temppath, path);
1.124 vatton 739: TtaExtractSuffix (temppath, suffix);
1.5 cvs 740:
741: /* Normalize the suffix */
742: i = 0;
1.106 cvs 743: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 744: {
1.123 vatton 745: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 746: i++;
747: }
1.106 cvs 748: nsuffix[i] = EOS;
749: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
750: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
751: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 752: return (TRUE);
753: return (FALSE);
1.3 cvs 754: }
755:
1.4 cvs 756: /*----------------------------------------------------------------------
1.58 cvs 757: IsImageType
758: returns TRUE if type points to an image resource.
759: ----------------------------------------------------------------------*/
1.111 cvs 760: ThotBool IsImageType (const char *type)
1.58 cvs 761: {
1.106 cvs 762: char temptype[MAX_LENGTH];
1.58 cvs 763: int i;
764:
765: if (!type)
766: return (FALSE);
767:
1.106 cvs 768: strcpy (temptype, type);
1.58 cvs 769: /* Normalize the type */
770: i = 0;
1.106 cvs 771: while (temptype[i] != EOS)
1.58 cvs 772: {
773: temptype[i] = tolower (temptype[i]);
774: i++;
775: }
1.111 cvs 776: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
777: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
778: !strcmp (temptype, "png"))
1.58 cvs 779: return (TRUE);
780: return (FALSE);
781: }
782:
783: /*----------------------------------------------------------------------
1.9 cvs 784: IsTextName
1.4 cvs 785: ----------------------------------------------------------------------*/
1.111 cvs 786: ThotBool IsTextName (const char *path)
1.106 cvs 787: {
788: char temppath[MAX_LENGTH];
789: char suffix[MAX_LENGTH];
790: char nsuffix[MAX_LENGTH];
1.5 cvs 791: int i;
792:
793: if (!path)
1.13 cvs 794: return (FALSE);
1.5 cvs 795:
1.106 cvs 796: strcpy (temppath, path);
1.124 vatton 797: TtaExtractSuffix (temppath, suffix);
1.5 cvs 798:
799: /* Normalize the suffix */
800: i = 0;
1.106 cvs 801: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 802: {
1.25 cvs 803: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 804: i++;
805: }
1.106 cvs 806: nsuffix[i] = EOS;
1.5 cvs 807:
1.111 cvs 808: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 809: return (TRUE);
1.106 cvs 810: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 811: {
1.39 cvs 812: /* take into account compressed files */
1.124 vatton 813: TtaExtractSuffix (temppath, suffix);
1.13 cvs 814: /* Normalize the suffix */
815: i = 0;
1.106 cvs 816: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 817: {
1.25 cvs 818: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 819: i++;
820: }
1.106 cvs 821: nsuffix[i] = EOS;
1.111 cvs 822: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 823: return (TRUE);
824: else
825: return (FALSE);
826: }
827: else
828: return (FALSE);
1.3 cvs 829: }
830:
1.4 cvs 831: /*----------------------------------------------------------------------
1.9 cvs 832: IsHTTPPath
833: returns TRUE if path is in fact an http URL.
1.4 cvs 834: ----------------------------------------------------------------------*/
1.112 cvs 835: ThotBool IsHTTPPath (const char *path)
1.3 cvs 836: {
1.5 cvs 837: if (!path)
838: return FALSE;
1.3 cvs 839:
1.106 cvs 840: if ((!strncmp (path, "http:", 5) != 0)
841: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
842: || !strncmp (path, "internal:", 9))
1.58 cvs 843: return TRUE;
844: return FALSE;
1.3 cvs 845: }
846:
1.4 cvs 847: /*----------------------------------------------------------------------
1.9 cvs 848: IsWithParameters
849: returns TRUE if url has a concatenated query string.
1.4 cvs 850: ----------------------------------------------------------------------*/
1.133 vatton 851: ThotBool IsWithParameters (const char *url)
1.3 cvs 852: {
1.5 cvs 853: int i;
1.3 cvs 854:
1.9 cvs 855: if ((!url) || (url[0] == EOS))
1.5 cvs 856: return FALSE;
1.3 cvs 857:
1.9 cvs 858: i = strlen (url) - 1;
859: while (i > 0 && url[i--] != '?')
1.5 cvs 860: if (i < 0)
861: return FALSE;
1.3 cvs 862:
1.5 cvs 863: /* There is a parameter */
864: return TRUE;
1.3 cvs 865: }
866:
1.4 cvs 867: /*----------------------------------------------------------------------
1.9 cvs 868: IsW3Path
869: returns TRUE if path is in fact a URL.
1.4 cvs 870: ----------------------------------------------------------------------*/
1.133 vatton 871: ThotBool IsW3Path (const char *path)
1.106 cvs 872: {
873: if (strncmp (path, "http:", 5) &&
874: strncmp (path, "ftp:", 4) &&
875: strncmp (path, "telnet:", 7) &&
876: strncmp (path, "wais:", 5) &&
877: strncmp (path, "news:", 5) &&
878: strncmp (path, "gopher:", 7) &&
879: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 880: strncmp (path, "archie:", 7) &&
881: strncmp (path, "https:", 6))
1.72 cvs 882: return FALSE;
883: return TRUE;
1.3 cvs 884: }
885:
1.4 cvs 886: /*----------------------------------------------------------------------
1.90 cvs 887: IsFilePath
888: returns TRUE if path is in fact a URL.
889: ----------------------------------------------------------------------*/
1.133 vatton 890: ThotBool IsFilePath (const char *path)
1.90 cvs 891: {
1.106 cvs 892: if (strncmp (path, "file:", 5))
1.90 cvs 893: return FALSE;
894: return TRUE;
895: }
896:
897: /*----------------------------------------------------------------------
1.9 cvs 898: IsValidProtocol
899: returns true if the url protocol is supported by Amaya.
1.4 cvs 900: ----------------------------------------------------------------------*/
1.133 vatton 901: ThotBool IsValidProtocol (const char *url)
1.106 cvs 902: {
903: if (!strncmp (url, "http:", 5)
904: || !strncmp (url, "internal:", 9)
905: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 906: /* experimental */
1.24 cvs 907: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 908: return (TRUE);
1.5 cvs 909: else
1.8 cvs 910: return (FALSE);
1.3 cvs 911: }
912:
1.31 cvs 913:
914: /*----------------------------------------------------------------------
915: GetBaseURL
916: normalizes orgName according to a base associated with doc, and
917: following the standard URL format rules.
918: The function returns the base used to solve relative URL and SRC:
919: - the base of the document,
920: - or the document path (without document name).
921: ----------------------------------------------------------------------*/
1.106 cvs 922: char *GetBaseURL (Document doc)
1.31 cvs 923: {
924: Element el;
925: ElementType elType;
926: AttributeType attrType;
927: Attribute attr;
1.106 cvs 928: char *ptr, *basename;
1.31 cvs 929: int length;
930:
1.113 cvs 931: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 932: return NULL;
1.106 cvs 933: basename = TtaGetMemory (MAX_LENGTH);
934: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
935: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 936: length = MAX_LENGTH -1;
1.113 cvs 937: /* is it a HTML document ? */
1.31 cvs 938: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.113 cvs 939: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
940: /* it's a HTML document */
1.65 cvs 941: {
1.113 cvs 942: /* get the document element */
943: el = TtaGetMainRoot (doc);
944: /* search the BASE element */
945: elType.ElTypeNum = HTML_EL_HEAD;
946: el = TtaSearchTypedElement (elType, SearchForward, el);
947: if (el)
948: /* there is a HEAD element */
949: {
950: /* look for a BASE element within the HEAD */
951: elType.ElTypeNum = HTML_EL_BASE;
952: el = TtaSearchTypedElement (elType, SearchInTree, el);
953: }
954: if (el)
1.31 cvs 955: {
1.113 cvs 956: /* The document has a BASE element. Get the HREF attribute of the
957: BASE element */
958: attrType.AttrSSchema = elType.ElSSchema;
959: attrType.AttrTypeNum = HTML_ATTR_HREF_;
960: attr = TtaGetAttribute (el, attrType);
961: if (attr)
1.31 cvs 962: {
1.113 cvs 963: /* Use the base path of the document */
964: TtaGiveTextAttributeValue (attr, basename, &length);
965: /* base and orgName have to be separated by a DIR_SEP */
966: length--;
967: if (basename[0] != EOS && basename[length] != URL_SEP &&
968: basename[length] != DIR_SEP)
969: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 970: {
1.113 cvs 971: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
972: AMAYA_PARSE_HOST |
973: AMAYA_PARSE_PUNCTUATION);
974: if (ptr && !strcmp (ptr, basename))
975: {
976: /* it has this form, complete it by adding a URL_STR */
977: if (strchr (basename, DIR_SEP))
978: strcat (basename, DIR_STR);
979: else
980: strcat (basename, URL_STR);
981: length++;
982: }
983: if (ptr)
984: TtaFreeMemory (ptr);
1.31 cvs 985: }
986: }
987: }
1.113 cvs 988: }
989:
1.31 cvs 990: /* Remove anything after the last DIR_SEP char. If no such char is found,
991: * then search for the first ":" char, hoping that what's before that is a
992: * protocol. If found, end the string there. If neither char is found,
993: * then discard the whole base element.
994: */
1.106 cvs 995: length = strlen (basename) - 1;
1.31 cvs 996: /* search for the last DIR_SEP char */
1.106 cvs 997: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 998: length--;
999: if (length >= 0)
1000: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1001: basename[length + 1] = EOS;
1.31 cvs 1002: else
1003: /* search for the first PATH_STR char */
1004: {
1.106 cvs 1005: for (length = 0; basename[length] != ':' &&
1006: basename[length] != EOS; length ++);
1007: if (basename[length] == ':')
1.31 cvs 1008: /* found, so end the string there */
1.106 cvs 1009: basename[length + 1] = EOS;
1.31 cvs 1010: else
1011: /* not found, discard the base */
1.106 cvs 1012: basename[0] = EOS;
1.31 cvs 1013: }
1014: return (basename);
1015: }
1016:
1017:
1.4 cvs 1018: /*----------------------------------------------------------------------
1.40 cvs 1019: GetLocalPath
1020: Allocate and return the local document path associated to the url
1021: ----------------------------------------------------------------------*/
1.106 cvs 1022: char *GetLocalPath (Document doc, char *url)
1023: {
1024: char *ptr;
1025: char *n;
1026: char *documentname;
1027: char url_sep;
1.83 cvs 1028: int len;
1.67 cvs 1029: ThotBool noFile;
1.40 cvs 1030:
1031: if (url != NULL)
1032: {
1033: /* check whether the file name exists */
1.106 cvs 1034: len = strlen (url) - 1;
1.71 cvs 1035: if (IsW3Path (url))
1.106 cvs 1036: url_sep = '/';
1.41 cvs 1037: else
1.106 cvs 1038: url_sep = DIR_SEP;
1.41 cvs 1039: noFile = (url[len] == url_sep);
1.40 cvs 1040: if (noFile)
1.106 cvs 1041: url[len] = EOS;
1042: ptr = TtaGetMemory (MAX_LENGTH);
1043: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 1044: TtaExtractName (url, ptr, documentname);
1.106 cvs 1045: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1046: if (!TtaCheckDirectory (ptr))
1047: /* directory did not exist */
1.72 cvs 1048: TtaMakeDirectory (ptr);
1.47 cvs 1049:
1050: /* don't include the query string within document name */
1.106 cvs 1051: n = strrchr (documentname, '?');
1.47 cvs 1052: if (n != NULL)
1.106 cvs 1053: *n = EOS;
1.46 cvs 1054: /* don't include ':' within document name */
1.106 cvs 1055: n = strchr (documentname, ':');
1.46 cvs 1056: if (n != NULL)
1.106 cvs 1057: *n = EOS;
1.69 cvs 1058: /* if after all this operations document name
1059: is empty, let's use noname.html instead */
1.106 cvs 1060: if (documentname[0] == EOS)
1061: strcat (ptr, "noname.html");
1.69 cvs 1062: else
1.106 cvs 1063: strcat (ptr, documentname);
1.40 cvs 1064: TtaFreeMemory (documentname);
1065: /* restore the url */
1066: if (noFile)
1.41 cvs 1067: url[len] = url_sep;
1.40 cvs 1068: return (ptr);
1069: }
1070: else
1071: return (NULL);
1072: }
1073:
1.73 cvs 1074: /*----------------------------------------------------------------------
1.79 cvs 1075: ExtractTarget extract the target name from document nane.
1076: ----------------------------------------------------------------------*/
1.106 cvs 1077: void ExtractTarget (char *aName, char *target)
1.79 cvs 1078: {
1.106 cvs 1079: int lg, i;
1080: char *ptr;
1081: char *oldptr;
1.79 cvs 1082:
1083: if (!target || !aName)
1084: /* bad target */
1085: return;
1086:
1.106 cvs 1087: target[0] = EOS;
1088: lg = strlen (aName);
1.79 cvs 1089: if (lg)
1090: {
1091: /* the name is not empty */
1092: oldptr = ptr = &aName[0];
1093: do
1094: {
1.106 cvs 1095: ptr = strrchr (oldptr, '#');
1.79 cvs 1096: if (ptr)
1097: oldptr = &ptr[1];
1098: }
1099: while (ptr);
1100:
1101: i = (int) (oldptr) - (int) (aName); /* name length */
1102: if (i > 1)
1103: {
1.106 cvs 1104: aName[i - 1] = EOS;
1.79 cvs 1105: if (i != lg)
1.106 cvs 1106: strcpy (target, oldptr);
1.79 cvs 1107: }
1108: }
1109: }
1110:
1111: /*----------------------------------------------------------------------
1.90 cvs 1112: RemoveNewLines (text)
1113: Removes any '\n' chars that are found in text.
1114: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1115: ----------------------------------------------------------------------*/
1.106 cvs 1116: ThotBool RemoveNewLines (char *text)
1117: {
1118: ThotBool change = FALSE;
1119: char *src;
1120: char *dest;
1.90 cvs 1121:
1122: src = text;
1123: dest = text;
1.115 kahan 1124:
1125: /* remove any preceding whitespace */
1126: while (*src && *src == ' ')
1127: {
1128: src++;
1129: change = 1;
1130: }
1131:
1.90 cvs 1132: while (*src)
1133: {
1134: switch (*src)
1135: {
1.106 cvs 1136: case '\n':
1.90 cvs 1137: /* don't copy the newline */
1138: change = 1;
1139: break;
1140: default:
1141: *dest = *src;
1142: dest++;
1143: break;
1144: }
1145: src++;
1146: }
1147: /* copy the last EOS char */
1148: *dest = *src;
1149:
1150: return (change);
1151: }
1152:
1153: /*----------------------------------------------------------------------
1154: CleanCopyFileURL
1155: Copies a file url from a src string to destination string.
1.97 cvs 1156: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1157: we want to do).
1.90 cvs 1158: ----------------------------------------------------------------------*/
1.106 cvs 1159: static void CleanCopyFileURL (char *dest, char *src,
1160: ConvertionType convertion)
1.90 cvs 1161: {
1162: while (*src)
1.89 cvs 1163: {
1.90 cvs 1164: switch (*src)
1.89 cvs 1165: {
1166: #ifdef _WINDOWS
1.106 cvs 1167: case URL_SEP:
1.96 cvs 1168: /* make DIR_SEP transformation */
1.97 cvs 1169: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1170: *dest = DIR_SEP;
1.96 cvs 1171: else
1172: *dest = *src;
1.90 cvs 1173: dest++;
1.96 cvs 1174: src++;
1.90 cvs 1175: break;
1.89 cvs 1176: #endif /* _WINDOWS */
1.96 cvs 1177:
1.106 cvs 1178: case '%':
1.97 cvs 1179: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1180: {
1.97 cvs 1181: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1182: src++;
1.106 cvs 1183: if (*src != EOS)
1.97 cvs 1184: {
1185: *dest = UnEscapeChar (*src) * 16;
1186: src++;
1187: }
1.106 cvs 1188: if (*src != EOS)
1.97 cvs 1189: {
1190: *dest = *dest + UnEscapeChar (*src);
1191: src++;
1192: }
1193: dest++;
1.96 cvs 1194: }
1.97 cvs 1195: else
1.96 cvs 1196: {
1.97 cvs 1197: *dest = *src;
1198: dest++;
1.96 cvs 1199: src++;
1200: }
1201: break;
1202:
1.90 cvs 1203: default:
1204: *dest = *src;
1.89 cvs 1205: dest++;
1.96 cvs 1206: src++;
1.90 cvs 1207: break;
1.89 cvs 1208: }
1209: }
1.90 cvs 1210: /* copy the EOS char */
1211: *dest = *src;
1.73 cvs 1212: }
1.40 cvs 1213:
1214: /*----------------------------------------------------------------------
1.9 cvs 1215: NormalizeURL
1216: normalizes orgName according to a base associated with doc, and
1217: following the standard URL format rules.
1.113 cvs 1218: if doc is < 0, use as a base the URL of the document that contains
1219: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1220: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1221: other path.
1.9 cvs 1222: The function returns the new complete and normalized URL
1.12 cvs 1223: or file name path (newName) and the name of the document (docName).
1.9 cvs 1224: N.B. If the function can't find out what's the docName, it assigns
1225: the name "noname.html".
1.4 cvs 1226: ----------------------------------------------------------------------*/
1.106 cvs 1227: void NormalizeURL (char *orgName, Document doc, char *newName,
1228: char *docName, char *otherPath)
1229: {
1230: char *basename;
1231: char tempOrgName[MAX_LENGTH];
1232: char *ptr;
1233: char used_sep;
1.84 cvs 1234: int length;
1235: ThotBool check;
1.5 cvs 1236:
1.110 cvs 1237: #ifdef _WINDOWS
1.44 cvs 1238: int ndx;
1.110 cvs 1239: #endif /* _WINDOWS */
1.44 cvs 1240:
1.5 cvs 1241: if (!newName || !docName)
1242: return;
1.18 cvs 1243:
1.113 cvs 1244: if (doc < 0)
1245: basename = TtaStrdup (SavedDocumentURL);
1246: else if (doc > 0)
1.53 cvs 1247: basename = GetBaseURL (doc);
1248: else if (otherPath != NULL)
1.108 cvs 1249: basename = TtaStrdup (otherPath);
1.32 cvs 1250: else
1.53 cvs 1251: basename = NULL;
1.32 cvs 1252:
1.18 cvs 1253: /*
1.31 cvs 1254: * Clean orgName
1255: * Make sure we have a complete orgName, without any leading or trailing
1256: * white spaces, or trailinbg new lines
1257: */
1.5 cvs 1258: ptr = orgName;
1.18 cvs 1259: /* skip leading white space and new line characters */
1.106 cvs 1260: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1261: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1262: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1263: /*
1.31 cvs 1264: * Make orgName a complete URL
1265: * If the URL does not include a protocol, then try to calculate
1266: * one using the doc's base element (if it exists),
1267: */
1.106 cvs 1268: if (tempOrgName[0] == EOS)
1.53 cvs 1269: {
1.106 cvs 1270: newName[0] = EOS;
1271: docName[0] = EOS;
1.53 cvs 1272: TtaFreeMemory (basename);
1273: return;
1274: }
1.49 cvs 1275:
1276: /* clean trailing white space */
1.106 cvs 1277: length = strlen (tempOrgName) - 1;
1278: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1279: {
1.106 cvs 1280: tempOrgName[length] = EOS;
1.53 cvs 1281: length--;
1282: }
1.50 cvs 1283:
1.55 cvs 1284: /* remove extra dot (which dot???) */
1285: /* ugly, but faster than a strcmp */
1.106 cvs 1286: if (tempOrgName[length] == '.'
1287: && (length == 0 || tempOrgName[length-1] != '.'))
1288: tempOrgName[length] = EOS;
1.50 cvs 1289:
1.94 cvs 1290: if (IsW3Path (tempOrgName))
1.53 cvs 1291: {
1292: /* the name is complete, go to the Sixth Step */
1.106 cvs 1293: strcpy (newName, tempOrgName);
1.53 cvs 1294: SimplifyUrl (&newName);
1295: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1296: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1297: AMAYA_PARSE_HOST |
1298: AMAYA_PARSE_PUNCTUATION);
1299: if (ptr && !strcmp (ptr, newName))
1300: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1301: strcat (newName, URL_STR);
1.49 cvs 1302:
1.53 cvs 1303: if (ptr)
1.50 cvs 1304: TtaFreeMemory (ptr);
1.53 cvs 1305: }
1.113 cvs 1306: else if (basename == NULL)
1.53 cvs 1307: /* the name is complete, go to the Sixth Step */
1.106 cvs 1308: strcpy (newName, tempOrgName);
1.53 cvs 1309: else
1310: {
1.31 cvs 1311: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1312: #ifdef _WINDOWS
1.53 cvs 1313: if (!IsW3Path (basename))
1314: {
1.106 cvs 1315: length = strlen (tempOrgName);
1.53 cvs 1316: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1317: if (tempOrgName [ndx] == '/')
1318: tempOrgName [ndx] = '\\';
1.53 cvs 1319: }
1.110 cvs 1320: #endif /* _WINDOWS */
1.25 cvs 1321: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1322: if (ptr)
1323: {
1324: SimplifyUrl (&ptr);
1.106 cvs 1325: strcpy (newName, ptr);
1.53 cvs 1326: TtaFreeMemory (ptr);
1327: }
1328: else
1.106 cvs 1329: newName[0] = EOS;
1.53 cvs 1330: }
1.36 cvs 1331:
1332: TtaFreeMemory (basename);
1.18 cvs 1333: /*
1.31 cvs 1334: * Prepare the docname that will refer to this ressource in the
1335: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1336: * noname.html as a default ressource name
1.18 cvs 1337: */
1.106 cvs 1338: if (newName[0] != EOS)
1.53 cvs 1339: {
1.106 cvs 1340: length = strlen (newName) - 1;
1341: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1342: {
1343: used_sep = newName[length];
1344: check = TRUE;
1345: while (check)
1346: {
1.50 cvs 1347: length--;
1348: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1349: length--;
1.106 cvs 1350: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1351: {
1.106 cvs 1352: newName[length+1] = EOS;
1.53 cvs 1353: /* remove also previous directory */
1354: length--;
1355: while (length >= 0 && newName[length] != used_sep)
1356: length--;
1.106 cvs 1357: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1358: /* don't remove server name */
1.106 cvs 1359: newName[length+1] = EOS;
1.53 cvs 1360: }
1.106 cvs 1361: else if (!strncmp (&newName[length+1], ".", 1))
1362: newName[length+1] = EOS;
1.50 cvs 1363: else
1.53 cvs 1364: check = FALSE;
1365: }
1366: /* docname was not comprised inside the URL, so let's */
1367: /* assign the default ressource name */
1.106 cvs 1368: strcpy (docName, "noname.html");
1.53 cvs 1369: }
1370: else
1371: { /* docname is comprised inside the URL */
1.110 cvs 1372: while (length >= 0 && newName[length] != URL_SEP &&
1373: newName[length] != DIR_SEP)
1.53 cvs 1374: length--;
1375: if (length < 0)
1.106 cvs 1376: strcpy (docName, newName);
1.53 cvs 1377: else
1.106 cvs 1378: strcpy (docName, &newName[length+1]);
1.53 cvs 1379: }
1380: }
1381: else
1.106 cvs 1382: docName[0] = EOS;
1.18 cvs 1383: }
1.3 cvs 1384:
1.4 cvs 1385: /*----------------------------------------------------------------------
1.9 cvs 1386: IsSameHost
1.4 cvs 1387: ----------------------------------------------------------------------*/
1.106 cvs 1388: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1389: {
1.106 cvs 1390: char *basename_ptr1, *basename_ptr2;
1391: ThotBool result;
1.3 cvs 1392:
1.106 cvs 1393: basename_ptr1 = AmayaParseUrl (url1, "",
1394: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1395: basename_ptr2 = AmayaParseUrl (url2, "",
1396: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1397:
1.106 cvs 1398: if (strcmp (basename_ptr1, basename_ptr2))
1399: result = FALSE;
1400: else
1401: result = TRUE;
1402: TtaFreeMemory (basename_ptr1);
1403: TtaFreeMemory (basename_ptr2);
1404: return (result);
1.3 cvs 1405: }
1406:
1407:
1.4 cvs 1408: /*----------------------------------------------------------------------
1.22 cvs 1409: HasKnownFileSuffix
1410: returns TRUE if path points to a file ending with a suffix.
1411: ----------------------------------------------------------------------*/
1.106 cvs 1412: ThotBool HasKnownFileSuffix (const char *path)
1413: {
1414: char *root;
1415: char temppath[MAX_LENGTH];
1416: char suffix[MAX_LENGTH];
1.22 cvs 1417:
1.106 cvs 1418: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1419: return (FALSE);
1420:
1.106 cvs 1421: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1422:
1423: if (root)
1424: {
1.106 cvs 1425: strcpy (temppath, root);
1.25 cvs 1426: TtaFreeMemory (root);
1.22 cvs 1427: /* Get the suffix */
1.124 vatton 1428: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1429:
1.106 cvs 1430: if( suffix[0] == EOS)
1.22 cvs 1431: /* no suffix */
1432: return (FALSE);
1433:
1434: /* Normalize the suffix */
1435: ConvertToLowerCase (suffix);
1436:
1.106 cvs 1437: if (!strcmp (suffix, "gz"))
1.22 cvs 1438: /* skip the compressed suffix */
1439: {
1.124 vatton 1440: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1441: if(suffix[0] == EOS)
1.22 cvs 1442: /* no suffix */
1443: return (FALSE);
1444: /* Normalize the suffix */
1445: ConvertToLowerCase (suffix);
1446: }
1447:
1.106 cvs 1448: if (strcmp (suffix, "gif") &&
1449: strcmp (suffix, "xbm") &&
1450: strcmp (suffix, "xpm") &&
1451: strcmp (suffix, "jpg") &&
1452: strcmp (suffix, "pdf") &&
1453: strcmp (suffix, "png") &&
1454: strcmp (suffix, "tgz") &&
1455: strcmp (suffix, "xpg") &&
1456: strcmp (suffix, "xpd") &&
1457: strcmp (suffix, "ps") &&
1458: strcmp (suffix, "au") &&
1459: strcmp (suffix, "html") &&
1460: strcmp (suffix, "htm") &&
1461: strcmp (suffix, "shtml") &&
1462: strcmp (suffix, "xht") &&
1463: strcmp (suffix, "xhtm") &&
1464: strcmp (suffix, "xhtml") &&
1465: strcmp (suffix, "txt") &&
1466: strcmp (suffix, "css") &&
1467: strcmp (suffix, "eps"))
1.22 cvs 1468: return (FALSE);
1469: else
1470: return (TRUE);
1471: }
1472: else
1473: return (FALSE);
1474: }
1475:
1476:
1477: /*----------------------------------------------------------------------
1.24 cvs 1478: ChopURL
1479: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1480: If inputURL is bigger than that size, outputURL receives
1481: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1482: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1483: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1484: copied into outputURL.
1485: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1486: chars.
1487: ----------------------------------------------------------------------*/
1.106 cvs 1488: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1489: {
1490: int len;
1.9 cvs 1491:
1.106 cvs 1492: len = strlen (inputURL);
1.24 cvs 1493: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1494: strcpy (outputURL, inputURL);
1.24 cvs 1495: else
1496: /* make a truncated urlName on the status window */
1497: {
1.106 cvs 1498: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1499: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1500: strcat (outputURL, "...");
1501: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1502: }
1.25 cvs 1503: }
1504:
1505:
1506: /*----------------------------------------------------------------------
1507: scan
1.47 cvs 1508: Scan a filename for its constituents
1.25 cvs 1509: -----------------------------------
1510:
1511: On entry,
1512: name points to a document name which may be incomplete.
1513: On exit,
1514: absolute or relative may be nonzero (but not both).
1515: host, fragment and access may be nonzero if they were specified.
1516: Any which are nonzero point to zero terminated strings.
1517: ----------------------------------------------------------------------*/
1.106 cvs 1518: static void scan (char *name, HTURI *parts)
1.25 cvs 1519: {
1.106 cvs 1520: char * p;
1521: char * after_access = name;
1.32 cvs 1522:
1.43 cvs 1523: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1524: /* Look for fragment identifier */
1.106 cvs 1525: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1526: {
1.106 cvs 1527: *p++ = '\0';
1.28 cvs 1528: parts->fragment = p;
1.25 cvs 1529: }
1530:
1.28 cvs 1531: for (p=name; *p; p++)
1532: {
1.106 cvs 1533: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1534: break;
1.106 cvs 1535: if (*p == ':')
1.28 cvs 1536: {
1537: *p = 0;
1538: parts->access = after_access; /* Scheme has been specified */
1539:
1540: /* The combination of gcc, the "-O" flag and the HP platform is
1541: unhealthy. The following three lines is a quick & dirty fix, but is
1542: not recommended. Rather, turn off "-O". */
1543:
1544: /* after_access = p;*/
1545: /* while (*after_access == 0)*/
1546: /* after_access++;*/
1547: after_access = p+1;
1.106 cvs 1548: if (!strcasecmp("URL", parts->access))
1.28 cvs 1549: /* Ignore IETF's URL: pre-prefix */
1550: parts->access = NULL;
1551: else
1.25 cvs 1552: break;
1553: }
1554: }
1555:
1556: p = after_access;
1.43 cvs 1557: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1558: {
1.43 cvs 1559: if (p[1] == URL_SEP)
1.28 cvs 1560: {
1.25 cvs 1561: parts->host = p+2; /* host has been specified */
1.28 cvs 1562: *p = 0; /* Terminate access */
1563: /* look for end of host name if any */
1.106 cvs 1564: p = strchr (parts->host, URL_SEP);
1.28 cvs 1565: if (p)
1566: {
1.106 cvs 1567: *p = EOS; /* Terminate host */
1.25 cvs 1568: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1569: }
1570: }
1571: else
1572: /* Root found but no host */
1573: parts->absolute = p+1;
1574: }
1575: else
1576: {
1.25 cvs 1577: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1578: }
1.25 cvs 1579: }
1580:
1581:
1582: /*----------------------------------------------------------------------
1.28 cvs 1583: AmayaParseUrl: parse a Name relative to another name
1584:
1585: This returns those parts of a name which are given (and requested)
1586: substituting bits from the related name where necessary.
1.25 cvs 1587:
1.28 cvs 1588: On entry,
1.25 cvs 1589: aName A filename given
1590: relatedName A name relative to which aName is to be parsed. Give
1591: it an empty string if aName is absolute.
1592: wanted A mask for the bits which are wanted.
1593:
1.28 cvs 1594: On exit,
1.25 cvs 1595: returns A pointer to a malloc'd string which MUST BE FREED
1596: ----------------------------------------------------------------------*/
1.106 cvs 1597: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1598: {
1599: char *return_value;
1600: char result[MAX_LENGTH];
1601: char name[MAX_LENGTH];
1602: char rel[MAX_LENGTH];
1603: char *p, *access;
1.29 cvs 1604: HTURI given, related;
1605: int len;
1.106 cvs 1606: char used_sep;
1607: char *used_str;
1.32 cvs 1608:
1.106 cvs 1609: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1610: {
1.106 cvs 1611: used_str = DIR_STR;
1612: used_sep = DIR_SEP;
1.33 cvs 1613: }
1.32 cvs 1614: else
1.33 cvs 1615: {
1.106 cvs 1616: used_str = URL_STR;
1617: used_sep = URL_SEP;
1.33 cvs 1618: }
1.32 cvs 1619:
1.29 cvs 1620: /* Make working copies of input strings to cut up: */
1621: return_value = NULL;
1622: result[0] = 0; /* Clear string */
1.106 cvs 1623: strcpy (name, aName);
1.29 cvs 1624: if (relatedName != NULL)
1.106 cvs 1625: strcpy (rel, relatedName);
1.29 cvs 1626: else
1.106 cvs 1627: relatedName[0] = EOS;
1.29 cvs 1628:
1629: scan (name, &given);
1630: scan (rel, &related);
1631: access = given.access ? given.access : related.access;
1632: if (wanted & AMAYA_PARSE_ACCESS)
1633: if (access)
1634: {
1.106 cvs 1635: strcat (result, access);
1.29 cvs 1636: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1637: strcat (result, ":");
1.29 cvs 1638: }
1639:
1640: if (given.access && related.access)
1641: /* If different, inherit nothing. */
1.106 cvs 1642: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1643: {
1644: related.host = 0;
1645: related.absolute = 0;
1646: related.relative = 0;
1647: related.fragment = 0;
1648: }
1649:
1650: if (wanted & AMAYA_PARSE_HOST)
1651: if(given.host || related.host)
1652: {
1653: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1654: strcat (result, "//");
1655: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1656: }
1657:
1658: if (given.host && related.host)
1659: /* If different hosts, inherit no path. */
1.106 cvs 1660: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1661: {
1662: related.absolute = 0;
1663: related.relative = 0;
1664: related.fragment = 0;
1665: }
1666:
1667: if (wanted & AMAYA_PARSE_PATH)
1668: {
1669: if (given.absolute)
1670: {
1671: /* All is given */
1672: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1673: strcat (result, used_str);
1674: strcat (result, given.absolute);
1.25 cvs 1675: }
1.29 cvs 1676: else if (related.absolute)
1677: {
1678: /* Adopt path not name */
1.106 cvs 1679: strcat (result, used_str);
1680: strcat (result, related.absolute);
1.29 cvs 1681: if (given.relative)
1682: {
1683: /* Search part? */
1.106 cvs 1684: p = strchr (result, '?');
1.29 cvs 1685: if (!p)
1.106 cvs 1686: p=result+strlen(result)-1;
1.33 cvs 1687: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1688: /* Remove filename */
1689: p[1]=0;
1690: /* Add given one */
1.106 cvs 1691: strcat (result, given.relative);
1.25 cvs 1692: }
1693: }
1.29 cvs 1694: else if (given.relative)
1695: /* what we've got */
1.106 cvs 1696: strcat (result, given.relative);
1.29 cvs 1697: else if (related.relative)
1.106 cvs 1698: strcat (result, related.relative);
1.29 cvs 1699: else
1700: /* No inheritance */
1.106 cvs 1701: strcat (result, used_str);
1.25 cvs 1702: }
1.29 cvs 1703:
1704: if (wanted & AMAYA_PARSE_ANCHOR)
1705: if (given.fragment || related.fragment)
1706: {
1707: if (given.absolute && given.fragment)
1708: {
1709: /*Fixes for relURLs...*/
1710: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1711: strcat (result, "#");
1712: strcat (result, given.fragment);
1.29 cvs 1713: }
1714: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1715: strcat (result, "");
1.29 cvs 1716: else
1717: {
1.110 cvs 1718: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1719: strcat (result, "#");
1.110 cvs 1720: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1721: }
1722: }
1.106 cvs 1723: len = strlen (result);
1724: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1725: strcpy (return_value, result);
1.29 cvs 1726: return (return_value); /* exactly the right length */
1.25 cvs 1727: }
1728:
1729: /*----------------------------------------------------------------------
1730: HTCanon
1731: Canonicalizes the URL in the following manner starting from the host
1732: pointer:
1733:
1734: 1) The host name is converted to lowercase
1735: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1736:
1737: Return: OK The position of the current path part of the URL
1738: which might be the old one or a new one.
1739:
1740: ----------------------------------------------------------------------*/
1.106 cvs 1741: static char *HTCanon (char **filename, char *host)
1742: {
1743: char *newname = NULL;
1744: char used_sep;
1745: char *path;
1746: char *strptr;
1747: char *port;
1748: char *access = host-3;
1749:
1750: if (*filename && strchr (*filename, URL_SEP))
1751: used_sep = URL_SEP;
1.33 cvs 1752: else
1.106 cvs 1753: used_sep = DIR_SEP;
1.32 cvs 1754:
1.110 cvs 1755: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1756: access--;
1.110 cvs 1757: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1758: path = host + strlen (host);
1759: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1760: host = strptr;
1.110 cvs 1761: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1762: port = NULL;
1.25 cvs 1763:
1764: strptr = host; /* Convert to lower-case */
1.82 cvs 1765: while (strptr < path)
1.33 cvs 1766: {
1.123 vatton 1767: *strptr = tolower (*strptr);
1.82 cvs 1768: strptr++;
1.33 cvs 1769: }
1.25 cvs 1770:
1771: /* Does the URL contain a full domain name? This also works for a
1772: numerical host name. The domain name is already made lower-case
1773: and without a trailing dot. */
1774: {
1.106 cvs 1775: char *dot = port ? port : path;
1776: if (dot > *filename && *--dot == '.')
1.33 cvs 1777: {
1.106 cvs 1778: char *orig = dot;
1779: char *dest = dot + 1;
1.82 cvs 1780: while ((*orig++ = *dest++));
1781: if (port) port--;
1.33 cvs 1782: path--;
1.25 cvs 1783: }
1784: }
1785: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1786: if (port)
1787: {
1.82 cvs 1788: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1789: {
1790: if (!newname)
1791: {
1.106 cvs 1792: char *orig = port;
1793: char *dest = port + 1;
1.82 cvs 1794: while ((*orig++ = *dest++));
1.33 cvs 1795: }
1796: }
1.106 cvs 1797: else if ((!strncmp (access, "http", 4) &&
1798: (*(port + 1) == '8' &&
1799: *(port+2) == '0' &&
1.82 cvs 1800: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1801: (!strncmp (access, "gopher", 6) &&
1802: (*(port+1) == '7' &&
1803: *(port+2) == '0' &&
1.82 cvs 1804: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1805: (!strncmp (access, "ftp", 3) &&
1806: (*(port+1) == '2' &&
1807: *(port + 2) == '1' &&
1.82 cvs 1808: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1809: if (!newname)
1810: {
1.106 cvs 1811: char *orig = port;
1812: char *dest = port + 3;
1.33 cvs 1813: while((*orig++ = *dest++));
1814: /* Update path position, Henry Minsky */
1815: path -= 3;
1.25 cvs 1816: }
1.33 cvs 1817: }
1818: else if (newname)
1.106 cvs 1819: strncat (newname, port, (int) (path - port));
1.33 cvs 1820: }
1.25 cvs 1821:
1.33 cvs 1822: if (newname)
1823: {
1.106 cvs 1824: char *newpath = newname + strlen (newname);
1825: strcat (newname, path);
1.25 cvs 1826: path = newpath;
1.28 cvs 1827: /* Free old copy */
1828: TtaFreeMemory(*filename);
1.25 cvs 1829: *filename = newname;
1.33 cvs 1830: }
1.25 cvs 1831: return path;
1832: }
1833:
1834:
1835: /*----------------------------------------------------------------------
1.29 cvs 1836: SimplifyUrl: simplify a URI
1.32 cvs 1837: A URI is allowed to contain the sequence xxx/../ which may be
1838: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1839: Simplification helps us recognize duplicate URIs.
1.25 cvs 1840:
1.28 cvs 1841: Thus, /etc/junk/../fred becomes /etc/fred
1842: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1843:
1.28 cvs 1844: but we should NOT change
1845: http://fred.xxx.edu/../..
1.25 cvs 1846:
1847: or ../../albert.html
1848:
1.28 cvs 1849: In order to avoid empty URLs the following URLs become:
1.25 cvs 1850:
1851: /fred/.. becomes /fred/..
1852: /fred/././.. becomes /fred/..
1853: /fred/.././junk/.././ becomes /fred/..
1854:
1.28 cvs 1855: If more than one set of `://' is found (several proxies in cascade) then
1856: only the part after the last `://' is simplified.
1.25 cvs 1857:
1.28 cvs 1858: Returns: A string which might be the old one or a new one.
1.25 cvs 1859: ----------------------------------------------------------------------*/
1.106 cvs 1860: void SimplifyUrl (char **url)
1861: {
1862: char *path;
1863: char *access;
1864: char *newptr;
1865: char *p;
1866: char *orig, *dest, *end;
1.28 cvs 1867:
1.106 cvs 1868: char used_sep;
1.77 cvs 1869: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1870: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1871:
1.28 cvs 1872: if (!url || !*url)
1873: return;
1874:
1.106 cvs 1875: if (strchr (*url, URL_SEP))
1876: used_sep = URL_SEP;
1.32 cvs 1877: else
1.106 cvs 1878: used_sep = DIR_SEP;
1.32 cvs 1879:
1.77 cvs 1880: /* should we simplify double dot? */
1881: path = *url;
1.106 cvs 1882: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1883: ddot_simplify = FALSE;
1884: else
1885: ddot_simplify = TRUE;
1886:
1.28 cvs 1887: /* Find any scheme name */
1.106 cvs 1888: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1889: {
1890: /* Find host name */
1.28 cvs 1891: access = *url;
1.123 vatton 1892: while (access < path && (*access = tolower (*access)))
1.82 cvs 1893: access++;
1.28 cvs 1894: path += 3;
1.106 cvs 1895: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1896: /* For proxies */
1.106 cvs 1897: path = newptr + 3;
1.82 cvs 1898: /* We have a host name */
1.84 cvs 1899: path = HTCanon (url, path);
1.25 cvs 1900: }
1.106 cvs 1901: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1902: path += 2;
1903: else
1904: path = *url;
1.84 cvs 1905: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1906: /* Some URLs start //<foo> */
1907: path += 1;
1.94 cvs 1908: else if (IsFilePath (path))
1909: {
1910: /* doesn't need to do anything more */
1911: return;
1912: }
1.106 cvs 1913: else if (!strncmp (path, "news:", 5))
1.28 cvs 1914: {
1.106 cvs 1915: newptr = strchr (path+5, '@');
1.28 cvs 1916: if (!newptr)
1917: newptr = path + 5;
1918: while (*newptr)
1919: {
1920: /* Make group or host lower case */
1.123 vatton 1921: *newptr = tolower (*newptr);
1.28 cvs 1922: newptr++;
1.25 cvs 1923: }
1.28 cvs 1924: /* Doesn't need to do any more */
1925: return;
1.25 cvs 1926: }
1.130 cheyroul 1927:
1.126 cheyroul 1928:
1.28 cvs 1929: if ((p = path))
1930: {
1.106 cvs 1931: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1932: (end = strchr (path, '#'))))
1933: end = path + strlen (path);
1.28 cvs 1934:
1935: /* Parse string second time to simplify */
1936: p = path;
1937: while (p < end)
1938: {
1.110 cvs 1939: /* if we're pointing to a char, it's safe to reactivate the
1940: ../ convertion */
1.106 cvs 1941: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1942: ddot_simplify = TRUE;
1943:
1.33 cvs 1944: if (*p==used_sep)
1.28 cvs 1945: {
1.106 cvs 1946: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1947: {
1948: orig = p + 1;
1.84 cvs 1949: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1950: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1951: end = orig - 1;
1952: }
1.106 cvs 1953: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1954: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1955: {
1956: newptr = p;
1.52 cvs 1957: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1958: if (*newptr == used_sep)
1959: orig = newptr + 1;
1.28 cvs 1960: else
1.52 cvs 1961: orig = newptr;
1962:
1963: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1964: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1965: end = orig-1;
1966: /* Start again with prev slash */
1967: p = newptr;
1.28 cvs 1968: }
1.33 cvs 1969: else if (*(p+1) == used_sep)
1.28 cvs 1970: {
1.33 cvs 1971: while (*(p+1) == used_sep)
1.28 cvs 1972: {
1973: orig = p;
1974: dest = p + 1;
1975: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1976: end = orig-1;
1977: }
1978: }
1979: else
1.25 cvs 1980: p++;
1.28 cvs 1981: }
1982: else
1983: p++;
1.25 cvs 1984: }
1985: }
1.51 cvs 1986: /*
1987: ** Check for host/../.. kind of things
1988: */
1.106 cvs 1989: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1990: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 1991: *(path+1) = EOS;
1.28 cvs 1992: return;
1993: }
1994:
1995:
1996: /*----------------------------------------------------------------------
1.96 cvs 1997: NormalizeFile normalizes local names.
1.28 cvs 1998: Return TRUE if target and src differ.
1999: ----------------------------------------------------------------------*/
1.106 cvs 2000: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2001: {
1.110 cvs 2002: #ifndef _WINDOWS
1.106 cvs 2003: char *s;
1.93 cvs 2004: int i;
1.110 cvs 2005: #endif /* !_WINDOWS */
1.82 cvs 2006: ThotBool change;
1.90 cvs 2007: int start_index; /* the first char that we'll copy */
1.28 cvs 2008:
1.54 cvs 2009: change = FALSE;
1.90 cvs 2010: start_index = 0;
2011:
1.106 cvs 2012: if (!src || src[0] == EOS)
1.96 cvs 2013: {
1.106 cvs 2014: target[0] = EOS;
1.96 cvs 2015: return FALSE;
2016: }
1.90 cvs 2017:
2018: /* @@ do I need file: or file:/ here? */
1.106 cvs 2019: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 2020: {
1.90 cvs 2021: /* remove the prefix file: */
2022: start_index += 5;
2023:
2024: /* remove the localhost prefix */
1.106 cvs 2025: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 2026: start_index += 11;
2027:
2028: /* remove the first two slashes in / / /path */
2029: while (src[start_index] &&
1.106 cvs 2030: src[start_index] == '/'
2031: && src[start_index + 1] == '/')
1.94 cvs 2032: start_index++;
2033:
2034: #ifdef _WINDOWS
2035: /* remove any extra slash before the drive name */
1.106 cvs 2036: if (src[start_index] == '/'
2037: &&src[start_index+2] == ':')
1.94 cvs 2038: start_index++;
2039: #endif /* _WINDOWS */
1.90 cvs 2040:
1.106 cvs 2041: if (src[start_index] == EOS)
1.90 cvs 2042: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 2043: strcpy (target, DIR_STR);
1.90 cvs 2044: else
1.97 cvs 2045: /* as we're inside a file: URL, we'll apply all the convertions
2046: we know */
2047: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 2048:
2049: change = TRUE;
2050: }
1.97 cvs 2051: else if (convertion != AM_CONV_NONE)
1.96 cvs 2052: {
2053: /* we are following a "local" relative link, we do all the
2054: convertions except for the HOME_DIR ~ one */
1.97 cvs 2055: CleanCopyFileURL (target, src, convertion);
1.28 cvs 2056: }
1.90 cvs 2057: #ifndef _WINDOWS
1.106 cvs 2058: else if (src[0] == '~')
1.53 cvs 2059: {
1.96 cvs 2060: /* it must be a URL typed in a text input field */
2061: /* do the HOME_DIR ~ substitution */
1.82 cvs 2062: s = TtaGetEnvString ("HOME");
1.106 cvs 2063: strcpy (target, s);
1.90 cvs 2064: #if 0
1.96 cvs 2065: /* JK: invalidated this part of the code as it's simpler
2066: to add the DIR_SEP whenever we have something to add
2067: to the path rather than adding it systematically */
1.106 cvs 2068: if (src[1] != DIR_SEP)
2069: strcat (target, DIR_STR);
1.90 cvs 2070: #endif
1.106 cvs 2071: i = strlen (target);
2072: strcpy (&target[i], &src[1]);
1.54 cvs 2073: change = TRUE;
1.53 cvs 2074: }
1.90 cvs 2075: #endif /* _WINDOWS */
1.28 cvs 2076: else
1.96 cvs 2077: /* leave it as it is */
1.106 cvs 2078: strcpy (target, src);
1.96 cvs 2079:
1.28 cvs 2080: /* remove /../ and /./ */
1.29 cvs 2081: SimplifyUrl (&target);
1.54 cvs 2082: if (!change)
1.106 cvs 2083: change = strcmp (src, target);
1.28 cvs 2084: return (change);
1.25 cvs 2085: }
2086:
1.28 cvs 2087:
1.25 cvs 2088: /*----------------------------------------------------------------------
1.31 cvs 2089: MakeRelativeURL: make relative name
1.25 cvs 2090:
1.28 cvs 2091: This function creates and returns a string which gives an expression of
2092: one address as related to another. Where there is no relation, an absolute
2093: address is retured.
1.25 cvs 2094:
1.28 cvs 2095: On entry,
1.25 cvs 2096: Both names must be absolute, fully qualified names of nodes
2097: (no fragment bits)
2098:
1.28 cvs 2099: On exit,
1.25 cvs 2100: The return result points to a newly allocated name which, if
2101: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2102: The caller is responsible for freeing the resulting name later.
2103: ----------------------------------------------------------------------*/
1.106 cvs 2104: char *MakeRelativeURL (char *aName, char *relatedName)
2105: {
2106: char *return_value;
2107: char result[MAX_LENGTH];
2108: char *p;
2109: char *q;
2110: char *after_access;
2111: char *last_slash = NULL;
2112: int slashes, levels, len;
1.110 cvs 2113: #ifdef _WINDOWS
1.44 cvs 2114: int ndx;
1.110 cvs 2115: #endif /* _WINDOWS */
1.44 cvs 2116:
1.29 cvs 2117: if (aName == NULL || relatedName == NULL)
2118: return (NULL);
2119:
2120: slashes = 0;
2121: after_access = NULL;
2122: p = aName;
2123: q = relatedName;
2124: for (; *p && (*p == *q); p++, q++)
1.27 cvs 2125: {
2126: /* Find extent of match */
1.106 cvs 2127: if (*p == ':')
1.29 cvs 2128: after_access = p + 1;
1.28 cvs 2129: if (*p == DIR_SEP)
1.27 cvs 2130: {
1.29 cvs 2131: /* memorize the last slash position and count them */
1.27 cvs 2132: last_slash = p;
2133: slashes++;
1.25 cvs 2134: }
2135: }
2136:
1.31 cvs 2137: /* q, p point to the first non-matching character or zero */
1.106 cvs 2138: if (*q == EOS)
1.31 cvs 2139: {
2140: /* New name is a subset of the related name */
2141: /* exactly the right length */
1.106 cvs 2142: len = strlen (p);
2143: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2144: strcpy (return_value, p);
1.31 cvs 2145: }
2146: else if ((slashes < 2 && after_access == NULL)
2147: || (slashes < 3 && after_access != NULL))
2148: {
2149: /* Two names whitout common path */
2150: /* exactly the right length */
1.106 cvs 2151: len = strlen (aName);
2152: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2153: strcpy (return_value, aName);
1.31 cvs 2154: }
2155: else
2156: {
2157: /* Some path in common */
1.106 cvs 2158: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2159: /* just the same server */
1.106 cvs 2160: strcpy (result, last_slash);
1.31 cvs 2161: else
2162: {
2163: levels= 0;
1.106 cvs 2164: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2165: if (*q == DIR_SEP)
2166: levels++;
2167:
1.106 cvs 2168: result[0] = EOS;
1.31 cvs 2169: for (;levels; levels--)
1.106 cvs 2170: strcat (result, "../");
2171: strcat (result, last_slash+1);
1.31 cvs 2172: }
1.52 cvs 2173:
2174: if (!*result)
1.106 cvs 2175: strcat (result, "./");
1.52 cvs 2176:
1.31 cvs 2177: /* exactly the right length */
1.106 cvs 2178: len = strlen (result);
2179: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2180: strcpy (return_value, result);
1.52 cvs 2181:
1.25 cvs 2182: }
1.110 cvs 2183: #ifdef _WINDOWS
1.106 cvs 2184: len = strlen (return_value);
1.44 cvs 2185: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2186: if (return_value[ndx] == '\\')
2187: return_value[ndx] = '/' ;
1.110 cvs 2188: #endif /* _WINDOWS */
1.29 cvs 2189: return (return_value);
1.24 cvs 2190: }
1.35 cvs 2191:
1.104 kahan 2192: /*----------------------------------------------------------------------
2193: AM_GetFileSize
2194: Returns TRUE and the filesize in the 2nd parameter.
2195: Otherwise, in case of a system error, returns FALSE, with a
2196: filesize of 0L.
2197: ---------------------------------------------------------------------*/
1.106 cvs 2198: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2199: {
1.106 cvs 2200: ThotFileHandle handle = ThotFile_BADHANDLE;
2201: ThotFileInfo info;
1.35 cvs 2202:
1.104 kahan 2203: *file_size = 0L;
2204: if (!TtaFileExist (filename))
2205: return FALSE;
2206:
2207: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2208: if (handle == ThotFile_BADHANDLE)
2209: /* ThotFile_BADHANDLE */
2210: return FALSE;
2211: if (TtaFileStat (handle, &info) == 0)
2212: /* bad stat */
2213: info.size = 0L;
2214: TtaFileClose (handle);
2215: *file_size = (unsigned long) info.size;
2216: return TRUE;
2217: }
1.139 kahan 2218:
2219: /*----------------------------------------------------------------------
2220: AM_UseXHTMLMimeType
2221: Returns TRUE if the user has configured Amaya to use this MIME type,
2222: FALSE otherwise.
2223: ---------------------------------------------------------------------*/
2224: ThotBool AM_UseXHTMLMimeType (void)
2225: {
2226: ThotBool xhtml_mimetype;
2227:
2228: /* does the user wants to use the new MIME type? */
2229: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2230:
2231: return (xhtml_mimetype);
2232: }
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