Annotation of Amaya/amaya/AHTURLTools.c, revision 1.151
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 */
1.143 vatton 103: /*case 0x26:*/ /* & */
1.140 kahan 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") ||
1.144 cvs 519: !strcmp (nsuffix, "lhtml") ||
1.106 cvs 520: !strcmp (nsuffix, "xhtml"))
1.101 cvs 521: return (TRUE);
1.106 cvs 522: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 523: {
524: /* take into account compressed files */
1.124 vatton 525: TtaExtractSuffix (temppath, suffix);
1.101 cvs 526: /* Normalize the suffix */
527: i = 0;
1.106 cvs 528: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 529: {
1.123 vatton 530: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 531: i++;
532: }
1.106 cvs 533: nsuffix[i] = EOS;
534: if (!strcmp (nsuffix, "html") ||
535: !strcmp (nsuffix, "htm") ||
536: !strcmp (nsuffix, "shtml") ||
537: !strcmp (nsuffix, "jsp") ||
538: !strcmp (nsuffix, "xht") ||
539: !strcmp (nsuffix, "xhtm") ||
1.144 cvs 540: !strcmp (nsuffix, "lhtml") ||
1.106 cvs 541: !strcmp (nsuffix, "xhtml"))
1.101 cvs 542: return (TRUE);
543: else
544: return (FALSE);
545: }
546: else
547: /* check if there is another suffix */
1.124 vatton 548: TtaExtractSuffix (temppath, suffix);
1.101 cvs 549: }
1.88 cvs 550: return (FALSE);
1.3 cvs 551: }
552:
1.4 cvs 553: /*----------------------------------------------------------------------
1.136 cvs 554: IsMathMLName
555: returns TRUE if path points to an MathML resource.
1.56 cvs 556: ----------------------------------------------------------------------*/
1.136 cvs 557: ThotBool IsMathMLName (const char *path)
1.56 cvs 558: {
1.136 cvs 559: char temppath[MAX_LENGTH];
560: char suffix[MAX_LENGTH];
1.56 cvs 561:
562: if (!path)
563: return (FALSE);
564:
1.106 cvs 565: strcpy (temppath, path);
1.124 vatton 566: TtaExtractSuffix (temppath, suffix);
1.56 cvs 567:
1.136 cvs 568: if (!strcasecmp (suffix, "mml"))
1.56 cvs 569: return (TRUE);
1.106 cvs 570: else if (!strcmp (suffix, "gz"))
1.56 cvs 571: {
572: /* take into account compressed files */
1.124 vatton 573: TtaExtractSuffix (temppath, suffix);
1.136 cvs 574: if (!strcasecmp (suffix, "mml"))
1.60 cvs 575: return (TRUE);
576: else
577: return (FALSE);
578: }
579: else
580: return (FALSE);
581: }
582:
583: /*----------------------------------------------------------------------
1.136 cvs 584: IsSVGName
585: returns TRUE if path points to an SVG resource.
1.133 vatton 586: ----------------------------------------------------------------------*/
1.136 cvs 587: ThotBool IsSVGName (const char *path)
1.133 vatton 588: {
1.136 cvs 589: char temppath[MAX_LENGTH];
590: char suffix[MAX_LENGTH];
1.133 vatton 591:
592: if (!path)
593: return (FALSE);
594:
595: strcpy (temppath, path);
596: TtaExtractSuffix (temppath, suffix);
597:
1.136 cvs 598: if (!strcasecmp (suffix, "svg"))
1.133 vatton 599: return (TRUE);
600: else if (!strcmp (suffix, "gz"))
601: {
602: /* take into account compressed files */
603: TtaExtractSuffix (temppath, suffix);
1.136 cvs 604: if (!strcasecmp (suffix, "svg"))
1.133 vatton 605: return (TRUE);
606: else
607: return (FALSE);
608: }
609: else
610: return (FALSE);
611: }
612:
613: /*----------------------------------------------------------------------
1.136 cvs 614: IsXMLName
615: returns TRUE if path points to an XML resource.
1.103 cvs 616: ----------------------------------------------------------------------*/
1.136 cvs 617: ThotBool IsXMLName (const char *path)
1.103 cvs 618: {
1.136 cvs 619: char temppath[MAX_LENGTH];
620: char suffix[MAX_LENGTH];
1.103 cvs 621:
622: if (!path)
623: return (FALSE);
624:
1.106 cvs 625: strcpy (temppath, path);
1.124 vatton 626: TtaExtractSuffix (temppath, suffix);
1.103 cvs 627:
1.136 cvs 628: if (!strcasecmp (suffix, "xml") ||
629: !strcasecmp (suffix, "xht") ||
630: !strcmp (suffix, "xhtm") ||
1.145 kahan 631: !strcmp (suffix, "xhtml") ||
632: !strcmp (suffix, "smi"))
1.103 cvs 633: return (TRUE);
1.106 cvs 634: else if (!strcmp (suffix, "gz"))
1.103 cvs 635: {
636: /* take into account compressed files */
1.124 vatton 637: TtaExtractSuffix (temppath, suffix);
1.136 cvs 638: if (!strcasecmp (suffix, "xml") ||
639: !strcasecmp (suffix, "xht") ||
640: !strcmp (suffix, "xhtm") ||
1.145 kahan 641: !strcmp (suffix, "xhtml") ||
642: !strcmp (suffix, "smi"))
1.103 cvs 643: return (TRUE);
644: else
645: return (FALSE);
646: }
647: else
648: return (FALSE);
649: }
650:
651: /*----------------------------------------------------------------------
1.136 cvs 652: IsUndisplayedName
653: returns TRUE if path points to an undisplayed resource.
1.103 cvs 654: ----------------------------------------------------------------------*/
1.136 cvs 655: ThotBool IsUndisplayedName (const char *path)
1.103 cvs 656: {
1.106 cvs 657: char temppath[MAX_LENGTH];
658: char suffix[MAX_LENGTH];
1.103 cvs 659:
660: if (!path)
661: return (FALSE);
662:
1.106 cvs 663: strcpy (temppath, path);
1.124 vatton 664: TtaExtractSuffix (temppath, suffix);
1.103 cvs 665:
1.136 cvs 666: if (!strcasecmp (suffix, "exe") ||
667: !strcasecmp (suffix, "zip") ||
668: !strcasecmp (suffix, "ppt") ||
669: !strcasecmp (suffix, "pdf") ||
670: !strcasecmp (suffix, "ps") ||
671: !strcasecmp (suffix, "eps") ||
672: !strcasecmp (suffix, "tar") ||
673: !strcasecmp (suffix, "tgz") ||
674: !strcasecmp (suffix, "ddl") ||
675: !strcasecmp (suffix, "o"))
1.103 cvs 676: return (TRUE);
1.106 cvs 677: else if (!strcmp (suffix, "gz"))
1.103 cvs 678: {
679: /* take into account compressed files */
1.124 vatton 680: TtaExtractSuffix (temppath, suffix);
1.136 cvs 681: if (!strcasecmp (suffix, "exe") ||
682: !strcasecmp (suffix, "zip") ||
683: !strcasecmp (suffix, "ppt") ||
684: !strcasecmp (suffix, "pdf") ||
685: !strcasecmp (suffix, "ps") ||
686: !strcasecmp (suffix, "eps") ||
687: !strcasecmp (suffix, "tar") ||
688: !strcasecmp (suffix, "ddl") ||
689: !strcasecmp (suffix, "o"))
1.103 cvs 690: return (TRUE);
691: else
692: return (FALSE);
693: }
694: else
695: return (FALSE);
696: }
697:
698: /*----------------------------------------------------------------------
1.60 cvs 699: IsCSSName
700: returns TRUE if path points to an XML resource.
701: ----------------------------------------------------------------------*/
1.111 cvs 702: ThotBool IsCSSName (const char *path)
1.60 cvs 703: {
1.106 cvs 704: char temppath[MAX_LENGTH];
705: char suffix[MAX_LENGTH];
1.60 cvs 706:
707: if (!path)
708: return (FALSE);
709:
1.106 cvs 710: strcpy (temppath, path);
1.124 vatton 711: TtaExtractSuffix (temppath, suffix);
1.60 cvs 712:
1.106 cvs 713: if (!strcasecmp (suffix, "css"))
1.60 cvs 714: return (TRUE);
1.106 cvs 715: else if (!strcmp (suffix, "gz"))
1.60 cvs 716: {
717: /* take into account compressed files */
1.124 vatton 718: TtaExtractSuffix (temppath, suffix);
1.106 cvs 719: if (!strcasecmp (suffix, "css"))
1.56 cvs 720: return (TRUE);
721: else
722: return (FALSE);
723: }
724: else
725: return (FALSE);
726: }
727:
728: /*----------------------------------------------------------------------
1.9 cvs 729: IsImageName
730: returns TRUE if path points to an image resource.
1.4 cvs 731: ----------------------------------------------------------------------*/
1.111 cvs 732: ThotBool IsImageName (const char *path)
1.106 cvs 733: {
734: char temppath[MAX_LENGTH];
735: char suffix[MAX_LENGTH];
736: char nsuffix[MAX_LENGTH];
1.5 cvs 737: int i;
738:
739: if (!path)
1.13 cvs 740: return (FALSE);
1.5 cvs 741:
1.106 cvs 742: strcpy (temppath, path);
1.124 vatton 743: TtaExtractSuffix (temppath, suffix);
1.5 cvs 744:
745: /* Normalize the suffix */
746: i = 0;
1.106 cvs 747: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 748: {
1.123 vatton 749: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 750: i++;
751: }
1.106 cvs 752: nsuffix[i] = EOS;
753: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
754: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
755: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 756: return (TRUE);
757: return (FALSE);
1.3 cvs 758: }
759:
1.4 cvs 760: /*----------------------------------------------------------------------
1.58 cvs 761: IsImageType
762: returns TRUE if type points to an image resource.
763: ----------------------------------------------------------------------*/
1.111 cvs 764: ThotBool IsImageType (const char *type)
1.58 cvs 765: {
1.106 cvs 766: char temptype[MAX_LENGTH];
1.58 cvs 767: int i;
768:
769: if (!type)
770: return (FALSE);
771:
1.106 cvs 772: strcpy (temptype, type);
1.58 cvs 773: /* Normalize the type */
774: i = 0;
1.106 cvs 775: while (temptype[i] != EOS)
1.58 cvs 776: {
777: temptype[i] = tolower (temptype[i]);
778: i++;
779: }
1.111 cvs 780: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
781: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
782: !strcmp (temptype, "png"))
1.58 cvs 783: return (TRUE);
784: return (FALSE);
785: }
786:
787: /*----------------------------------------------------------------------
1.9 cvs 788: IsTextName
1.4 cvs 789: ----------------------------------------------------------------------*/
1.111 cvs 790: ThotBool IsTextName (const char *path)
1.106 cvs 791: {
792: char temppath[MAX_LENGTH];
793: char suffix[MAX_LENGTH];
794: char nsuffix[MAX_LENGTH];
1.5 cvs 795: int i;
796:
797: if (!path)
1.13 cvs 798: return (FALSE);
1.5 cvs 799:
1.106 cvs 800: strcpy (temppath, path);
1.124 vatton 801: TtaExtractSuffix (temppath, suffix);
1.5 cvs 802:
803: /* Normalize the suffix */
804: i = 0;
1.106 cvs 805: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 806: {
1.25 cvs 807: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 808: i++;
809: }
1.106 cvs 810: nsuffix[i] = EOS;
1.5 cvs 811:
1.111 cvs 812: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 813: return (TRUE);
1.106 cvs 814: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 815: {
1.39 cvs 816: /* take into account compressed files */
1.124 vatton 817: TtaExtractSuffix (temppath, suffix);
1.13 cvs 818: /* Normalize the suffix */
819: i = 0;
1.106 cvs 820: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 821: {
1.25 cvs 822: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 823: i++;
824: }
1.106 cvs 825: nsuffix[i] = EOS;
1.111 cvs 826: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 827: return (TRUE);
828: else
829: return (FALSE);
830: }
831: else
832: return (FALSE);
1.3 cvs 833: }
834:
1.4 cvs 835: /*----------------------------------------------------------------------
1.9 cvs 836: IsHTTPPath
837: returns TRUE if path is in fact an http URL.
1.4 cvs 838: ----------------------------------------------------------------------*/
1.112 cvs 839: ThotBool IsHTTPPath (const char *path)
1.3 cvs 840: {
1.5 cvs 841: if (!path)
842: return FALSE;
1.3 cvs 843:
1.106 cvs 844: if ((!strncmp (path, "http:", 5) != 0)
845: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
846: || !strncmp (path, "internal:", 9))
1.58 cvs 847: return TRUE;
848: return FALSE;
1.3 cvs 849: }
850:
1.4 cvs 851: /*----------------------------------------------------------------------
1.9 cvs 852: IsWithParameters
853: returns TRUE if url has a concatenated query string.
1.4 cvs 854: ----------------------------------------------------------------------*/
1.133 vatton 855: ThotBool IsWithParameters (const char *url)
1.3 cvs 856: {
1.5 cvs 857: int i;
1.3 cvs 858:
1.9 cvs 859: if ((!url) || (url[0] == EOS))
1.5 cvs 860: return FALSE;
1.3 cvs 861:
1.9 cvs 862: i = strlen (url) - 1;
863: while (i > 0 && url[i--] != '?')
1.5 cvs 864: if (i < 0)
865: return FALSE;
1.3 cvs 866:
1.5 cvs 867: /* There is a parameter */
868: return TRUE;
1.3 cvs 869: }
870:
1.4 cvs 871: /*----------------------------------------------------------------------
1.9 cvs 872: IsW3Path
873: returns TRUE if path is in fact a URL.
1.4 cvs 874: ----------------------------------------------------------------------*/
1.133 vatton 875: ThotBool IsW3Path (const char *path)
1.106 cvs 876: {
877: if (strncmp (path, "http:", 5) &&
878: strncmp (path, "ftp:", 4) &&
879: strncmp (path, "telnet:", 7) &&
880: strncmp (path, "wais:", 5) &&
881: strncmp (path, "news:", 5) &&
882: strncmp (path, "gopher:", 7) &&
883: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 884: strncmp (path, "archie:", 7) &&
885: strncmp (path, "https:", 6))
1.72 cvs 886: return FALSE;
887: return TRUE;
1.3 cvs 888: }
889:
1.4 cvs 890: /*----------------------------------------------------------------------
1.90 cvs 891: IsFilePath
892: returns TRUE if path is in fact a URL.
893: ----------------------------------------------------------------------*/
1.133 vatton 894: ThotBool IsFilePath (const char *path)
1.90 cvs 895: {
1.106 cvs 896: if (strncmp (path, "file:", 5))
1.90 cvs 897: return FALSE;
898: return TRUE;
899: }
900:
901: /*----------------------------------------------------------------------
1.9 cvs 902: IsValidProtocol
903: returns true if the url protocol is supported by Amaya.
1.4 cvs 904: ----------------------------------------------------------------------*/
1.133 vatton 905: ThotBool IsValidProtocol (const char *url)
1.106 cvs 906: {
907: if (!strncmp (url, "http:", 5)
908: || !strncmp (url, "internal:", 9)
909: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 910: /* experimental */
1.24 cvs 911: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 912: return (TRUE);
1.5 cvs 913: else
1.8 cvs 914: return (FALSE);
1.3 cvs 915: }
916:
1.31 cvs 917:
918: /*----------------------------------------------------------------------
919: GetBaseURL
920: normalizes orgName according to a base associated with doc, and
921: following the standard URL format rules.
922: The function returns the base used to solve relative URL and SRC:
923: - the base of the document,
924: - or the document path (without document name).
925: ----------------------------------------------------------------------*/
1.106 cvs 926: char *GetBaseURL (Document doc)
1.31 cvs 927: {
928: Element el;
929: ElementType elType;
930: AttributeType attrType;
931: Attribute attr;
1.106 cvs 932: char *ptr, *basename;
1.31 cvs 933: int length;
1.151 ! kahan 934: ThotBool hasDocBase;
1.31 cvs 935:
1.113 cvs 936: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 937: return NULL;
1.148 kahan 938: /* the other functions expect basename to have no more than MAX_LENGTH chars */
1.106 cvs 939: basename = TtaGetMemory (MAX_LENGTH);
1.148 kahan 940: basename[0] = EOS;
1.31 cvs 941: length = MAX_LENGTH -1;
1.151 ! kahan 942: hasDocBase = FALSE;
! 943:
! 944: /* If the document has a base URL, it has a priority over the headers. */
! 945: /* @@ We need to do this too when we support XML:base */
! 946:
! 947: /* is it a HTML document ? */
! 948: elType.ElSSchema = TtaGetDocumentSSchema (doc);
! 949: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
! 950: /* it's a HTML document */
! 951: {
! 952: /* get the document element */
! 953: el = TtaGetMainRoot (doc);
! 954: /* search the BASE element */
! 955: elType.ElTypeNum = HTML_EL_HEAD;
! 956: el = TtaSearchTypedElement (elType, SearchForward, el);
! 957: if (el)
! 958: /* there is a HEAD element */
! 959: {
! 960: /* look for a BASE element within the HEAD */
! 961: elType.ElTypeNum = HTML_EL_BASE;
! 962: el = TtaSearchTypedElement (elType, SearchInTree, el);
! 963: }
! 964: if (el)
! 965: {
! 966: /* The document has a BASE element. Get the HREF attribute of the
! 967: BASE element */
! 968: hasDocBase = TRUE;
! 969: attrType.AttrSSchema = elType.ElSSchema;
! 970: attrType.AttrTypeNum = HTML_ATTR_HREF_;
! 971: attr = TtaGetAttribute (el, attrType);
! 972: if (attr)
! 973: {
! 974: /* Use the base path of the document */
! 975: TtaGiveTextAttributeValue (attr, basename, &length);
! 976: }
! 977: }
! 978: }
! 979:
! 980: /* there was no BASE. Do we have a location header? */
! 981: if (!hasDocBase && DocumentMeta[doc] && DocumentMeta[doc]->full_content_location
1.148 kahan 982: && DocumentMeta[doc]->full_content_location[0] != EOS)
1.65 cvs 983: {
1.148 kahan 984: strncpy (basename, DocumentMeta[doc]->full_content_location, MAX_LENGTH-1);
985: basename[MAX_LENGTH-1] = EOS;
986: length = strlen (basename);
987: }
988:
989: if (basename[0] != EOS)
990: {
991: /* base and orgName have to be separated by a DIR_SEP */
992: length--;
993: if (basename[0] != EOS && basename[length] != URL_SEP &&
994: basename[length] != DIR_SEP)
995: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 996: {
1.148 kahan 997: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
998: AMAYA_PARSE_HOST |
999: AMAYA_PARSE_PUNCTUATION);
1000: if (ptr && !strcmp (ptr, basename))
1.31 cvs 1001: {
1.148 kahan 1002: /* it has this form, complete it by adding a URL_STR */
1003: if (strchr (basename, DIR_SEP))
1004: strcat (basename, DIR_STR);
1005: else
1006: strcat (basename, URL_STR);
1007: length++;
1.31 cvs 1008: }
1.149 kahan 1009: else if (!ptr || ptr[0] == EOS)
1010: {
1011: /* no host was detected, we may have a relative URL. We test
1012: if it begins with a URL_SEP, DIR_SEP or period. If yes, it's
1013: relative. */
1014: if (! (basename[0] == '.' || basename[0] == URL_SEP
1015: || basename[0] == DIR_SEP))
1016: basename[0] = EOS;
1017: }
1.148 kahan 1018: if (ptr)
1019: TtaFreeMemory (ptr);
1.31 cvs 1020: }
1.113 cvs 1021: }
1022:
1.148 kahan 1023: /* there was no base element and no location header, we use the DocumentURL */
1024: if (basename[0] == EOS)
1025: {
1026: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1027: basename[MAX_LENGTH-1] = EOS;
1028: }
1029:
1.31 cvs 1030: /* Remove anything after the last DIR_SEP char. If no such char is found,
1031: * then search for the first ":" char, hoping that what's before that is a
1032: * protocol. If found, end the string there. If neither char is found,
1033: * then discard the whole base element.
1034: */
1.106 cvs 1035: length = strlen (basename) - 1;
1.31 cvs 1036: /* search for the last DIR_SEP char */
1.106 cvs 1037: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 1038: length--;
1039: if (length >= 0)
1040: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1041: basename[length + 1] = EOS;
1.31 cvs 1042: else
1043: /* search for the first PATH_STR char */
1044: {
1.106 cvs 1045: for (length = 0; basename[length] != ':' &&
1046: basename[length] != EOS; length ++);
1047: if (basename[length] == ':')
1.31 cvs 1048: /* found, so end the string there */
1.106 cvs 1049: basename[length + 1] = EOS;
1.31 cvs 1050: else
1051: /* not found, discard the base */
1.106 cvs 1052: basename[0] = EOS;
1.31 cvs 1053: }
1054: return (basename);
1055: }
1056:
1057:
1.4 cvs 1058: /*----------------------------------------------------------------------
1.40 cvs 1059: GetLocalPath
1060: Allocate and return the local document path associated to the url
1061: ----------------------------------------------------------------------*/
1.150 vatton 1062: char *GetLocalPath (Document doc, char *url)
1.106 cvs 1063: {
1064: char *ptr;
1065: char *n;
1066: char *documentname;
1067: char url_sep;
1.83 cvs 1068: int len;
1.67 cvs 1069: ThotBool noFile;
1.40 cvs 1070:
1071: if (url != NULL)
1072: {
1073: /* check whether the file name exists */
1.106 cvs 1074: len = strlen (url) - 1;
1.71 cvs 1075: if (IsW3Path (url))
1.106 cvs 1076: url_sep = '/';
1.41 cvs 1077: else
1.106 cvs 1078: url_sep = DIR_SEP;
1.41 cvs 1079: noFile = (url[len] == url_sep);
1.40 cvs 1080: if (noFile)
1.106 cvs 1081: url[len] = EOS;
1082: ptr = TtaGetMemory (MAX_LENGTH);
1083: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 1084: TtaExtractName (url, ptr, documentname);
1.106 cvs 1085: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1086: if (!TtaCheckDirectory (ptr))
1087: /* directory did not exist */
1.72 cvs 1088: TtaMakeDirectory (ptr);
1.47 cvs 1089:
1090: /* don't include the query string within document name */
1.106 cvs 1091: n = strrchr (documentname, '?');
1.47 cvs 1092: if (n != NULL)
1.106 cvs 1093: *n = EOS;
1.46 cvs 1094: /* don't include ':' within document name */
1.106 cvs 1095: n = strchr (documentname, ':');
1.46 cvs 1096: if (n != NULL)
1.106 cvs 1097: *n = EOS;
1.69 cvs 1098: /* if after all this operations document name
1099: is empty, let's use noname.html instead */
1.106 cvs 1100: if (documentname[0] == EOS)
1101: strcat (ptr, "noname.html");
1.69 cvs 1102: else
1.106 cvs 1103: strcat (ptr, documentname);
1.40 cvs 1104: TtaFreeMemory (documentname);
1105: /* restore the url */
1106: if (noFile)
1.41 cvs 1107: url[len] = url_sep;
1.40 cvs 1108: return (ptr);
1109: }
1110: else
1111: return (NULL);
1112: }
1113:
1.73 cvs 1114: /*----------------------------------------------------------------------
1.79 cvs 1115: ExtractTarget extract the target name from document nane.
1116: ----------------------------------------------------------------------*/
1.150 vatton 1117: void ExtractTarget (char *aName, char *target)
1.79 cvs 1118: {
1.106 cvs 1119: int lg, i;
1120: char *ptr;
1121: char *oldptr;
1.79 cvs 1122:
1123: if (!target || !aName)
1124: /* bad target */
1125: return;
1126:
1.106 cvs 1127: target[0] = EOS;
1128: lg = strlen (aName);
1.79 cvs 1129: if (lg)
1130: {
1131: /* the name is not empty */
1132: oldptr = ptr = &aName[0];
1133: do
1134: {
1.106 cvs 1135: ptr = strrchr (oldptr, '#');
1.79 cvs 1136: if (ptr)
1137: oldptr = &ptr[1];
1138: }
1139: while (ptr);
1140:
1141: i = (int) (oldptr) - (int) (aName); /* name length */
1142: if (i > 1)
1143: {
1.106 cvs 1144: aName[i - 1] = EOS;
1.79 cvs 1145: if (i != lg)
1.106 cvs 1146: strcpy (target, oldptr);
1.79 cvs 1147: }
1148: }
1149: }
1150:
1151: /*----------------------------------------------------------------------
1.90 cvs 1152: RemoveNewLines (text)
1153: Removes any '\n' chars that are found in text.
1154: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1155: ----------------------------------------------------------------------*/
1.106 cvs 1156: ThotBool RemoveNewLines (char *text)
1157: {
1158: ThotBool change = FALSE;
1159: char *src;
1160: char *dest;
1.90 cvs 1161:
1162: src = text;
1163: dest = text;
1.115 kahan 1164:
1165: /* remove any preceding whitespace */
1166: while (*src && *src == ' ')
1167: {
1168: src++;
1169: change = 1;
1170: }
1171:
1.90 cvs 1172: while (*src)
1173: {
1174: switch (*src)
1175: {
1.106 cvs 1176: case '\n':
1.90 cvs 1177: /* don't copy the newline */
1178: change = 1;
1179: break;
1180: default:
1181: *dest = *src;
1182: dest++;
1183: break;
1184: }
1185: src++;
1186: }
1187: /* copy the last EOS char */
1188: *dest = *src;
1189:
1190: return (change);
1191: }
1192:
1193: /*----------------------------------------------------------------------
1194: CleanCopyFileURL
1195: Copies a file url from a src string to destination string.
1.97 cvs 1196: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1197: we want to do).
1.90 cvs 1198: ----------------------------------------------------------------------*/
1.106 cvs 1199: static void CleanCopyFileURL (char *dest, char *src,
1200: ConvertionType convertion)
1.90 cvs 1201: {
1202: while (*src)
1.89 cvs 1203: {
1.90 cvs 1204: switch (*src)
1.89 cvs 1205: {
1206: #ifdef _WINDOWS
1.106 cvs 1207: case URL_SEP:
1.96 cvs 1208: /* make DIR_SEP transformation */
1.97 cvs 1209: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1210: *dest = DIR_SEP;
1.96 cvs 1211: else
1212: *dest = *src;
1.90 cvs 1213: dest++;
1.96 cvs 1214: src++;
1.90 cvs 1215: break;
1.89 cvs 1216: #endif /* _WINDOWS */
1.96 cvs 1217:
1.106 cvs 1218: case '%':
1.97 cvs 1219: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1220: {
1.97 cvs 1221: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1222: src++;
1.106 cvs 1223: if (*src != EOS)
1.97 cvs 1224: {
1225: *dest = UnEscapeChar (*src) * 16;
1226: src++;
1227: }
1.106 cvs 1228: if (*src != EOS)
1.97 cvs 1229: {
1230: *dest = *dest + UnEscapeChar (*src);
1231: src++;
1232: }
1233: dest++;
1.96 cvs 1234: }
1.97 cvs 1235: else
1.96 cvs 1236: {
1.97 cvs 1237: *dest = *src;
1238: dest++;
1.96 cvs 1239: src++;
1240: }
1241: break;
1242:
1.90 cvs 1243: default:
1244: *dest = *src;
1.89 cvs 1245: dest++;
1.96 cvs 1246: src++;
1.90 cvs 1247: break;
1.89 cvs 1248: }
1249: }
1.90 cvs 1250: /* copy the EOS char */
1251: *dest = *src;
1.73 cvs 1252: }
1.40 cvs 1253:
1254: /*----------------------------------------------------------------------
1.9 cvs 1255: NormalizeURL
1256: normalizes orgName according to a base associated with doc, and
1257: following the standard URL format rules.
1.113 cvs 1258: if doc is < 0, use as a base the URL of the document that contains
1259: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1260: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1261: other path.
1.9 cvs 1262: The function returns the new complete and normalized URL
1.12 cvs 1263: or file name path (newName) and the name of the document (docName).
1.9 cvs 1264: N.B. If the function can't find out what's the docName, it assigns
1265: the name "noname.html".
1.4 cvs 1266: ----------------------------------------------------------------------*/
1.106 cvs 1267: void NormalizeURL (char *orgName, Document doc, char *newName,
1268: char *docName, char *otherPath)
1269: {
1270: char *basename;
1271: char tempOrgName[MAX_LENGTH];
1272: char *ptr;
1273: char used_sep;
1.84 cvs 1274: int length;
1275: ThotBool check;
1.5 cvs 1276:
1.110 cvs 1277: #ifdef _WINDOWS
1.44 cvs 1278: int ndx;
1.110 cvs 1279: #endif /* _WINDOWS */
1.44 cvs 1280:
1.5 cvs 1281: if (!newName || !docName)
1282: return;
1.18 cvs 1283:
1.113 cvs 1284: if (doc < 0)
1285: basename = TtaStrdup (SavedDocumentURL);
1286: else if (doc > 0)
1.53 cvs 1287: basename = GetBaseURL (doc);
1288: else if (otherPath != NULL)
1.108 cvs 1289: basename = TtaStrdup (otherPath);
1.32 cvs 1290: else
1.53 cvs 1291: basename = NULL;
1.32 cvs 1292:
1.18 cvs 1293: /*
1.31 cvs 1294: * Clean orgName
1295: * Make sure we have a complete orgName, without any leading or trailing
1296: * white spaces, or trailinbg new lines
1297: */
1.5 cvs 1298: ptr = orgName;
1.18 cvs 1299: /* skip leading white space and new line characters */
1.106 cvs 1300: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1301: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1302: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1303: /*
1.31 cvs 1304: * Make orgName a complete URL
1305: * If the URL does not include a protocol, then try to calculate
1306: * one using the doc's base element (if it exists),
1307: */
1.106 cvs 1308: if (tempOrgName[0] == EOS)
1.53 cvs 1309: {
1.106 cvs 1310: newName[0] = EOS;
1311: docName[0] = EOS;
1.53 cvs 1312: TtaFreeMemory (basename);
1313: return;
1314: }
1.49 cvs 1315:
1316: /* clean trailing white space */
1.106 cvs 1317: length = strlen (tempOrgName) - 1;
1318: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1319: {
1.106 cvs 1320: tempOrgName[length] = EOS;
1.53 cvs 1321: length--;
1322: }
1.50 cvs 1323:
1.55 cvs 1324: /* remove extra dot (which dot???) */
1325: /* ugly, but faster than a strcmp */
1.106 cvs 1326: if (tempOrgName[length] == '.'
1327: && (length == 0 || tempOrgName[length-1] != '.'))
1328: tempOrgName[length] = EOS;
1.50 cvs 1329:
1.94 cvs 1330: if (IsW3Path (tempOrgName))
1.53 cvs 1331: {
1332: /* the name is complete, go to the Sixth Step */
1.106 cvs 1333: strcpy (newName, tempOrgName);
1.53 cvs 1334: SimplifyUrl (&newName);
1335: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1336: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1337: AMAYA_PARSE_HOST |
1338: AMAYA_PARSE_PUNCTUATION);
1339: if (ptr && !strcmp (ptr, newName))
1340: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1341: strcat (newName, URL_STR);
1.49 cvs 1342:
1.53 cvs 1343: if (ptr)
1.50 cvs 1344: TtaFreeMemory (ptr);
1.53 cvs 1345: }
1.113 cvs 1346: else if (basename == NULL)
1.53 cvs 1347: /* the name is complete, go to the Sixth Step */
1.106 cvs 1348: strcpy (newName, tempOrgName);
1.53 cvs 1349: else
1350: {
1.31 cvs 1351: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1352: #ifdef _WINDOWS
1.53 cvs 1353: if (!IsW3Path (basename))
1354: {
1.106 cvs 1355: length = strlen (tempOrgName);
1.53 cvs 1356: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1357: if (tempOrgName [ndx] == '/')
1358: tempOrgName [ndx] = '\\';
1.53 cvs 1359: }
1.110 cvs 1360: #endif /* _WINDOWS */
1.25 cvs 1361: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1362: if (ptr)
1363: {
1364: SimplifyUrl (&ptr);
1.106 cvs 1365: strcpy (newName, ptr);
1.53 cvs 1366: TtaFreeMemory (ptr);
1367: }
1368: else
1.106 cvs 1369: newName[0] = EOS;
1.53 cvs 1370: }
1.36 cvs 1371:
1372: TtaFreeMemory (basename);
1.18 cvs 1373: /*
1.31 cvs 1374: * Prepare the docname that will refer to this ressource in the
1375: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1376: * noname.html as a default ressource name
1.18 cvs 1377: */
1.106 cvs 1378: if (newName[0] != EOS)
1.53 cvs 1379: {
1.106 cvs 1380: length = strlen (newName) - 1;
1381: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1382: {
1383: used_sep = newName[length];
1384: check = TRUE;
1385: while (check)
1386: {
1.50 cvs 1387: length--;
1388: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1389: length--;
1.106 cvs 1390: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1391: {
1.106 cvs 1392: newName[length+1] = EOS;
1.53 cvs 1393: /* remove also previous directory */
1394: length--;
1395: while (length >= 0 && newName[length] != used_sep)
1396: length--;
1.106 cvs 1397: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1398: /* don't remove server name */
1.106 cvs 1399: newName[length+1] = EOS;
1.53 cvs 1400: }
1.106 cvs 1401: else if (!strncmp (&newName[length+1], ".", 1))
1402: newName[length+1] = EOS;
1.50 cvs 1403: else
1.53 cvs 1404: check = FALSE;
1405: }
1406: /* docname was not comprised inside the URL, so let's */
1407: /* assign the default ressource name */
1.106 cvs 1408: strcpy (docName, "noname.html");
1.53 cvs 1409: }
1410: else
1411: { /* docname is comprised inside the URL */
1.110 cvs 1412: while (length >= 0 && newName[length] != URL_SEP &&
1413: newName[length] != DIR_SEP)
1.53 cvs 1414: length--;
1415: if (length < 0)
1.106 cvs 1416: strcpy (docName, newName);
1.53 cvs 1417: else
1.106 cvs 1418: strcpy (docName, &newName[length+1]);
1.53 cvs 1419: }
1420: }
1421: else
1.106 cvs 1422: docName[0] = EOS;
1.18 cvs 1423: }
1.3 cvs 1424:
1.4 cvs 1425: /*----------------------------------------------------------------------
1.9 cvs 1426: IsSameHost
1.4 cvs 1427: ----------------------------------------------------------------------*/
1.106 cvs 1428: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1429: {
1.106 cvs 1430: char *basename_ptr1, *basename_ptr2;
1431: ThotBool result;
1.3 cvs 1432:
1.106 cvs 1433: basename_ptr1 = AmayaParseUrl (url1, "",
1434: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1435: basename_ptr2 = AmayaParseUrl (url2, "",
1436: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1437:
1.106 cvs 1438: if (strcmp (basename_ptr1, basename_ptr2))
1439: result = FALSE;
1440: else
1441: result = TRUE;
1442: TtaFreeMemory (basename_ptr1);
1443: TtaFreeMemory (basename_ptr2);
1444: return (result);
1.3 cvs 1445: }
1446:
1447:
1.4 cvs 1448: /*----------------------------------------------------------------------
1.22 cvs 1449: HasKnownFileSuffix
1450: returns TRUE if path points to a file ending with a suffix.
1451: ----------------------------------------------------------------------*/
1.106 cvs 1452: ThotBool HasKnownFileSuffix (const char *path)
1453: {
1454: char *root;
1455: char temppath[MAX_LENGTH];
1456: char suffix[MAX_LENGTH];
1.22 cvs 1457:
1.106 cvs 1458: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1459: return (FALSE);
1460:
1.106 cvs 1461: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1462:
1463: if (root)
1464: {
1.106 cvs 1465: strcpy (temppath, root);
1.25 cvs 1466: TtaFreeMemory (root);
1.22 cvs 1467: /* Get the suffix */
1.124 vatton 1468: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1469:
1.106 cvs 1470: if( suffix[0] == EOS)
1.22 cvs 1471: /* no suffix */
1472: return (FALSE);
1473:
1474: /* Normalize the suffix */
1475: ConvertToLowerCase (suffix);
1476:
1.106 cvs 1477: if (!strcmp (suffix, "gz"))
1.22 cvs 1478: /* skip the compressed suffix */
1479: {
1.124 vatton 1480: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1481: if(suffix[0] == EOS)
1.22 cvs 1482: /* no suffix */
1483: return (FALSE);
1484: /* Normalize the suffix */
1485: ConvertToLowerCase (suffix);
1486: }
1487:
1.106 cvs 1488: if (strcmp (suffix, "gif") &&
1489: strcmp (suffix, "xbm") &&
1490: strcmp (suffix, "xpm") &&
1491: strcmp (suffix, "jpg") &&
1492: strcmp (suffix, "pdf") &&
1493: strcmp (suffix, "png") &&
1494: strcmp (suffix, "tgz") &&
1495: strcmp (suffix, "xpg") &&
1496: strcmp (suffix, "xpd") &&
1497: strcmp (suffix, "ps") &&
1498: strcmp (suffix, "au") &&
1499: strcmp (suffix, "html") &&
1500: strcmp (suffix, "htm") &&
1501: strcmp (suffix, "shtml") &&
1502: strcmp (suffix, "xht") &&
1503: strcmp (suffix, "xhtm") &&
1504: strcmp (suffix, "xhtml") &&
1505: strcmp (suffix, "txt") &&
1506: strcmp (suffix, "css") &&
1507: strcmp (suffix, "eps"))
1.22 cvs 1508: return (FALSE);
1509: else
1510: return (TRUE);
1511: }
1512: else
1513: return (FALSE);
1514: }
1515:
1516:
1517: /*----------------------------------------------------------------------
1.24 cvs 1518: ChopURL
1519: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1520: If inputURL is bigger than that size, outputURL receives
1521: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1522: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1523: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1524: copied into outputURL.
1525: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1526: chars.
1527: ----------------------------------------------------------------------*/
1.106 cvs 1528: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1529: {
1530: int len;
1.9 cvs 1531:
1.106 cvs 1532: len = strlen (inputURL);
1.24 cvs 1533: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1534: strcpy (outputURL, inputURL);
1.24 cvs 1535: else
1536: /* make a truncated urlName on the status window */
1537: {
1.106 cvs 1538: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1539: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1540: strcat (outputURL, "...");
1541: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1542: }
1.25 cvs 1543: }
1544:
1545:
1546: /*----------------------------------------------------------------------
1547: scan
1.47 cvs 1548: Scan a filename for its constituents
1.25 cvs 1549: -----------------------------------
1550:
1551: On entry,
1552: name points to a document name which may be incomplete.
1553: On exit,
1554: absolute or relative may be nonzero (but not both).
1555: host, fragment and access may be nonzero if they were specified.
1556: Any which are nonzero point to zero terminated strings.
1557: ----------------------------------------------------------------------*/
1.106 cvs 1558: static void scan (char *name, HTURI *parts)
1.25 cvs 1559: {
1.106 cvs 1560: char * p;
1561: char * after_access = name;
1.32 cvs 1562:
1.43 cvs 1563: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1564: /* Look for fragment identifier */
1.106 cvs 1565: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1566: {
1.106 cvs 1567: *p++ = '\0';
1.28 cvs 1568: parts->fragment = p;
1.25 cvs 1569: }
1570:
1.28 cvs 1571: for (p=name; *p; p++)
1572: {
1.106 cvs 1573: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1574: break;
1.106 cvs 1575: if (*p == ':')
1.28 cvs 1576: {
1577: *p = 0;
1578: parts->access = after_access; /* Scheme has been specified */
1579:
1580: /* The combination of gcc, the "-O" flag and the HP platform is
1581: unhealthy. The following three lines is a quick & dirty fix, but is
1582: not recommended. Rather, turn off "-O". */
1583:
1584: /* after_access = p;*/
1585: /* while (*after_access == 0)*/
1586: /* after_access++;*/
1587: after_access = p+1;
1.106 cvs 1588: if (!strcasecmp("URL", parts->access))
1.28 cvs 1589: /* Ignore IETF's URL: pre-prefix */
1590: parts->access = NULL;
1591: else
1.25 cvs 1592: break;
1593: }
1594: }
1595:
1596: p = after_access;
1.43 cvs 1597: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1598: {
1.43 cvs 1599: if (p[1] == URL_SEP)
1.28 cvs 1600: {
1.25 cvs 1601: parts->host = p+2; /* host has been specified */
1.28 cvs 1602: *p = 0; /* Terminate access */
1603: /* look for end of host name if any */
1.106 cvs 1604: p = strchr (parts->host, URL_SEP);
1.28 cvs 1605: if (p)
1606: {
1.106 cvs 1607: *p = EOS; /* Terminate host */
1.25 cvs 1608: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1609: }
1610: }
1611: else
1612: /* Root found but no host */
1613: parts->absolute = p+1;
1614: }
1615: else
1616: {
1.25 cvs 1617: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1618: }
1.25 cvs 1619: }
1620:
1621:
1622: /*----------------------------------------------------------------------
1.28 cvs 1623: AmayaParseUrl: parse a Name relative to another name
1624:
1625: This returns those parts of a name which are given (and requested)
1626: substituting bits from the related name where necessary.
1.25 cvs 1627:
1.28 cvs 1628: On entry,
1.25 cvs 1629: aName A filename given
1630: relatedName A name relative to which aName is to be parsed. Give
1631: it an empty string if aName is absolute.
1632: wanted A mask for the bits which are wanted.
1633:
1.28 cvs 1634: On exit,
1.25 cvs 1635: returns A pointer to a malloc'd string which MUST BE FREED
1636: ----------------------------------------------------------------------*/
1.106 cvs 1637: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1638: {
1639: char *return_value;
1640: char result[MAX_LENGTH];
1641: char name[MAX_LENGTH];
1642: char rel[MAX_LENGTH];
1643: char *p, *access;
1.29 cvs 1644: HTURI given, related;
1645: int len;
1.106 cvs 1646: char used_sep;
1647: char *used_str;
1.32 cvs 1648:
1.106 cvs 1649: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1650: {
1.106 cvs 1651: used_str = DIR_STR;
1652: used_sep = DIR_SEP;
1.33 cvs 1653: }
1.32 cvs 1654: else
1.33 cvs 1655: {
1.106 cvs 1656: used_str = URL_STR;
1657: used_sep = URL_SEP;
1.33 cvs 1658: }
1.32 cvs 1659:
1.29 cvs 1660: /* Make working copies of input strings to cut up: */
1661: return_value = NULL;
1662: result[0] = 0; /* Clear string */
1.106 cvs 1663: strcpy (name, aName);
1.29 cvs 1664: if (relatedName != NULL)
1.106 cvs 1665: strcpy (rel, relatedName);
1.29 cvs 1666: else
1.106 cvs 1667: relatedName[0] = EOS;
1.29 cvs 1668:
1669: scan (name, &given);
1670: scan (rel, &related);
1671: access = given.access ? given.access : related.access;
1672: if (wanted & AMAYA_PARSE_ACCESS)
1673: if (access)
1674: {
1.106 cvs 1675: strcat (result, access);
1.29 cvs 1676: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1677: strcat (result, ":");
1.29 cvs 1678: }
1679:
1680: if (given.access && related.access)
1681: /* If different, inherit nothing. */
1.106 cvs 1682: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1683: {
1684: related.host = 0;
1685: related.absolute = 0;
1686: related.relative = 0;
1687: related.fragment = 0;
1688: }
1689:
1690: if (wanted & AMAYA_PARSE_HOST)
1691: if(given.host || related.host)
1692: {
1693: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1694: strcat (result, "//");
1695: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1696: }
1697:
1698: if (given.host && related.host)
1699: /* If different hosts, inherit no path. */
1.106 cvs 1700: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1701: {
1702: related.absolute = 0;
1703: related.relative = 0;
1704: related.fragment = 0;
1705: }
1706:
1707: if (wanted & AMAYA_PARSE_PATH)
1708: {
1709: if (given.absolute)
1710: {
1711: /* All is given */
1712: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1713: strcat (result, used_str);
1714: strcat (result, given.absolute);
1.25 cvs 1715: }
1.29 cvs 1716: else if (related.absolute)
1717: {
1718: /* Adopt path not name */
1.106 cvs 1719: strcat (result, used_str);
1720: strcat (result, related.absolute);
1.29 cvs 1721: if (given.relative)
1722: {
1723: /* Search part? */
1.106 cvs 1724: p = strchr (result, '?');
1.29 cvs 1725: if (!p)
1.106 cvs 1726: p=result+strlen(result)-1;
1.33 cvs 1727: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1728: /* Remove filename */
1729: p[1]=0;
1730: /* Add given one */
1.106 cvs 1731: strcat (result, given.relative);
1.25 cvs 1732: }
1733: }
1.29 cvs 1734: else if (given.relative)
1735: /* what we've got */
1.106 cvs 1736: strcat (result, given.relative);
1.29 cvs 1737: else if (related.relative)
1.106 cvs 1738: strcat (result, related.relative);
1.29 cvs 1739: else
1740: /* No inheritance */
1.106 cvs 1741: strcat (result, used_str);
1.25 cvs 1742: }
1.29 cvs 1743:
1744: if (wanted & AMAYA_PARSE_ANCHOR)
1745: if (given.fragment || related.fragment)
1746: {
1747: if (given.absolute && given.fragment)
1748: {
1749: /*Fixes for relURLs...*/
1750: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1751: strcat (result, "#");
1752: strcat (result, given.fragment);
1.29 cvs 1753: }
1754: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1755: strcat (result, "");
1.29 cvs 1756: else
1757: {
1.110 cvs 1758: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1759: strcat (result, "#");
1.110 cvs 1760: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1761: }
1762: }
1.106 cvs 1763: len = strlen (result);
1764: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1765: strcpy (return_value, result);
1.29 cvs 1766: return (return_value); /* exactly the right length */
1.25 cvs 1767: }
1768:
1769: /*----------------------------------------------------------------------
1770: HTCanon
1771: Canonicalizes the URL in the following manner starting from the host
1772: pointer:
1773:
1774: 1) The host name is converted to lowercase
1775: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1776:
1777: Return: OK The position of the current path part of the URL
1778: which might be the old one or a new one.
1779:
1780: ----------------------------------------------------------------------*/
1.106 cvs 1781: static char *HTCanon (char **filename, char *host)
1782: {
1783: char *newname = NULL;
1784: char used_sep;
1785: char *path;
1786: char *strptr;
1787: char *port;
1788: char *access = host-3;
1789:
1790: if (*filename && strchr (*filename, URL_SEP))
1791: used_sep = URL_SEP;
1.33 cvs 1792: else
1.106 cvs 1793: used_sep = DIR_SEP;
1.32 cvs 1794:
1.110 cvs 1795: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1796: access--;
1.110 cvs 1797: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1798: path = host + strlen (host);
1799: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1800: host = strptr;
1.110 cvs 1801: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1802: port = NULL;
1.25 cvs 1803:
1804: strptr = host; /* Convert to lower-case */
1.82 cvs 1805: while (strptr < path)
1.33 cvs 1806: {
1.123 vatton 1807: *strptr = tolower (*strptr);
1.82 cvs 1808: strptr++;
1.33 cvs 1809: }
1.25 cvs 1810:
1811: /* Does the URL contain a full domain name? This also works for a
1812: numerical host name. The domain name is already made lower-case
1813: and without a trailing dot. */
1814: {
1.106 cvs 1815: char *dot = port ? port : path;
1816: if (dot > *filename && *--dot == '.')
1.33 cvs 1817: {
1.106 cvs 1818: char *orig = dot;
1819: char *dest = dot + 1;
1.82 cvs 1820: while ((*orig++ = *dest++));
1821: if (port) port--;
1.33 cvs 1822: path--;
1.25 cvs 1823: }
1824: }
1825: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1826: if (port)
1827: {
1.82 cvs 1828: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1829: {
1830: if (!newname)
1831: {
1.106 cvs 1832: char *orig = port;
1833: char *dest = port + 1;
1.82 cvs 1834: while ((*orig++ = *dest++));
1.33 cvs 1835: }
1836: }
1.106 cvs 1837: else if ((!strncmp (access, "http", 4) &&
1838: (*(port + 1) == '8' &&
1839: *(port+2) == '0' &&
1.82 cvs 1840: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1841: (!strncmp (access, "gopher", 6) &&
1842: (*(port+1) == '7' &&
1843: *(port+2) == '0' &&
1.82 cvs 1844: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1845: (!strncmp (access, "ftp", 3) &&
1846: (*(port+1) == '2' &&
1847: *(port + 2) == '1' &&
1.82 cvs 1848: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1849: if (!newname)
1850: {
1.106 cvs 1851: char *orig = port;
1852: char *dest = port + 3;
1.33 cvs 1853: while((*orig++ = *dest++));
1854: /* Update path position, Henry Minsky */
1855: path -= 3;
1.25 cvs 1856: }
1.33 cvs 1857: }
1858: else if (newname)
1.106 cvs 1859: strncat (newname, port, (int) (path - port));
1.33 cvs 1860: }
1.25 cvs 1861:
1.33 cvs 1862: if (newname)
1863: {
1.106 cvs 1864: char *newpath = newname + strlen (newname);
1865: strcat (newname, path);
1.25 cvs 1866: path = newpath;
1.28 cvs 1867: /* Free old copy */
1868: TtaFreeMemory(*filename);
1.25 cvs 1869: *filename = newname;
1.33 cvs 1870: }
1.25 cvs 1871: return path;
1872: }
1873:
1874:
1875: /*----------------------------------------------------------------------
1.29 cvs 1876: SimplifyUrl: simplify a URI
1.32 cvs 1877: A URI is allowed to contain the sequence xxx/../ which may be
1878: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1879: Simplification helps us recognize duplicate URIs.
1.25 cvs 1880:
1.28 cvs 1881: Thus, /etc/junk/../fred becomes /etc/fred
1882: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1883:
1.28 cvs 1884: but we should NOT change
1885: http://fred.xxx.edu/../..
1.25 cvs 1886:
1887: or ../../albert.html
1888:
1.28 cvs 1889: In order to avoid empty URLs the following URLs become:
1.25 cvs 1890:
1891: /fred/.. becomes /fred/..
1892: /fred/././.. becomes /fred/..
1893: /fred/.././junk/.././ becomes /fred/..
1894:
1.28 cvs 1895: If more than one set of `://' is found (several proxies in cascade) then
1896: only the part after the last `://' is simplified.
1.25 cvs 1897:
1.28 cvs 1898: Returns: A string which might be the old one or a new one.
1.25 cvs 1899: ----------------------------------------------------------------------*/
1.106 cvs 1900: void SimplifyUrl (char **url)
1901: {
1902: char *path;
1903: char *access;
1904: char *newptr;
1905: char *p;
1906: char *orig, *dest, *end;
1.28 cvs 1907:
1.106 cvs 1908: char used_sep;
1.77 cvs 1909: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1910: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1911:
1.28 cvs 1912: if (!url || !*url)
1913: return;
1914:
1.106 cvs 1915: if (strchr (*url, URL_SEP))
1916: used_sep = URL_SEP;
1.32 cvs 1917: else
1.106 cvs 1918: used_sep = DIR_SEP;
1.32 cvs 1919:
1.77 cvs 1920: /* should we simplify double dot? */
1921: path = *url;
1.106 cvs 1922: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1923: ddot_simplify = FALSE;
1924: else
1925: ddot_simplify = TRUE;
1926:
1.28 cvs 1927: /* Find any scheme name */
1.106 cvs 1928: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1929: {
1930: /* Find host name */
1.28 cvs 1931: access = *url;
1.123 vatton 1932: while (access < path && (*access = tolower (*access)))
1.82 cvs 1933: access++;
1.28 cvs 1934: path += 3;
1.106 cvs 1935: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1936: /* For proxies */
1.106 cvs 1937: path = newptr + 3;
1.82 cvs 1938: /* We have a host name */
1.84 cvs 1939: path = HTCanon (url, path);
1.25 cvs 1940: }
1.106 cvs 1941: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1942: path += 2;
1943: else
1944: path = *url;
1.84 cvs 1945: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1946: /* Some URLs start //<foo> */
1947: path += 1;
1.94 cvs 1948: else if (IsFilePath (path))
1949: {
1950: /* doesn't need to do anything more */
1951: return;
1952: }
1.106 cvs 1953: else if (!strncmp (path, "news:", 5))
1.28 cvs 1954: {
1.106 cvs 1955: newptr = strchr (path+5, '@');
1.28 cvs 1956: if (!newptr)
1957: newptr = path + 5;
1958: while (*newptr)
1959: {
1960: /* Make group or host lower case */
1.123 vatton 1961: *newptr = tolower (*newptr);
1.28 cvs 1962: newptr++;
1.25 cvs 1963: }
1.28 cvs 1964: /* Doesn't need to do any more */
1965: return;
1.25 cvs 1966: }
1.130 cheyroul 1967:
1.126 cheyroul 1968:
1.28 cvs 1969: if ((p = path))
1970: {
1.106 cvs 1971: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1972: (end = strchr (path, '#'))))
1973: end = path + strlen (path);
1.28 cvs 1974:
1975: /* Parse string second time to simplify */
1976: p = path;
1977: while (p < end)
1978: {
1.110 cvs 1979: /* if we're pointing to a char, it's safe to reactivate the
1980: ../ convertion */
1.106 cvs 1981: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1982: ddot_simplify = TRUE;
1983:
1.33 cvs 1984: if (*p==used_sep)
1.28 cvs 1985: {
1.106 cvs 1986: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1987: {
1988: orig = p + 1;
1.84 cvs 1989: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1990: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1991: end = orig - 1;
1992: }
1.106 cvs 1993: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1994: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1995: {
1996: newptr = p;
1.52 cvs 1997: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1998: if (*newptr == used_sep)
1999: orig = newptr + 1;
1.28 cvs 2000: else
1.52 cvs 2001: orig = newptr;
2002:
2003: dest = (*(p+3) != used_sep) ? p+3 : p+4;
2004: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
2005: end = orig-1;
2006: /* Start again with prev slash */
2007: p = newptr;
1.28 cvs 2008: }
1.33 cvs 2009: else if (*(p+1) == used_sep)
1.28 cvs 2010: {
1.33 cvs 2011: while (*(p+1) == used_sep)
1.28 cvs 2012: {
2013: orig = p;
2014: dest = p + 1;
2015: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2016: end = orig-1;
2017: }
2018: }
2019: else
1.25 cvs 2020: p++;
1.28 cvs 2021: }
2022: else
2023: p++;
1.25 cvs 2024: }
2025: }
1.51 cvs 2026: /*
2027: ** Check for host/../.. kind of things
2028: */
1.106 cvs 2029: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 2030: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 2031: *(path+1) = EOS;
1.28 cvs 2032: return;
2033: }
2034:
2035:
2036: /*----------------------------------------------------------------------
1.96 cvs 2037: NormalizeFile normalizes local names.
1.28 cvs 2038: Return TRUE if target and src differ.
2039: ----------------------------------------------------------------------*/
1.106 cvs 2040: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2041: {
1.110 cvs 2042: #ifndef _WINDOWS
1.106 cvs 2043: char *s;
1.93 cvs 2044: int i;
1.110 cvs 2045: #endif /* !_WINDOWS */
1.82 cvs 2046: ThotBool change;
1.90 cvs 2047: int start_index; /* the first char that we'll copy */
1.28 cvs 2048:
1.54 cvs 2049: change = FALSE;
1.90 cvs 2050: start_index = 0;
2051:
1.106 cvs 2052: if (!src || src[0] == EOS)
1.96 cvs 2053: {
1.106 cvs 2054: target[0] = EOS;
1.96 cvs 2055: return FALSE;
2056: }
1.90 cvs 2057:
2058: /* @@ do I need file: or file:/ here? */
1.106 cvs 2059: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 2060: {
1.90 cvs 2061: /* remove the prefix file: */
2062: start_index += 5;
2063:
2064: /* remove the localhost prefix */
1.106 cvs 2065: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 2066: start_index += 11;
2067:
2068: /* remove the first two slashes in / / /path */
2069: while (src[start_index] &&
1.106 cvs 2070: src[start_index] == '/'
2071: && src[start_index + 1] == '/')
1.94 cvs 2072: start_index++;
2073:
2074: #ifdef _WINDOWS
2075: /* remove any extra slash before the drive name */
1.106 cvs 2076: if (src[start_index] == '/'
2077: &&src[start_index+2] == ':')
1.94 cvs 2078: start_index++;
2079: #endif /* _WINDOWS */
1.90 cvs 2080:
1.106 cvs 2081: if (src[start_index] == EOS)
1.90 cvs 2082: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 2083: strcpy (target, DIR_STR);
1.90 cvs 2084: else
1.97 cvs 2085: /* as we're inside a file: URL, we'll apply all the convertions
2086: we know */
2087: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 2088:
2089: change = TRUE;
2090: }
1.97 cvs 2091: else if (convertion != AM_CONV_NONE)
1.96 cvs 2092: {
2093: /* we are following a "local" relative link, we do all the
2094: convertions except for the HOME_DIR ~ one */
1.97 cvs 2095: CleanCopyFileURL (target, src, convertion);
1.28 cvs 2096: }
1.90 cvs 2097: #ifndef _WINDOWS
1.106 cvs 2098: else if (src[0] == '~')
1.53 cvs 2099: {
1.96 cvs 2100: /* it must be a URL typed in a text input field */
2101: /* do the HOME_DIR ~ substitution */
1.82 cvs 2102: s = TtaGetEnvString ("HOME");
1.106 cvs 2103: strcpy (target, s);
1.90 cvs 2104: #if 0
1.96 cvs 2105: /* JK: invalidated this part of the code as it's simpler
2106: to add the DIR_SEP whenever we have something to add
2107: to the path rather than adding it systematically */
1.106 cvs 2108: if (src[1] != DIR_SEP)
2109: strcat (target, DIR_STR);
1.90 cvs 2110: #endif
1.106 cvs 2111: i = strlen (target);
2112: strcpy (&target[i], &src[1]);
1.54 cvs 2113: change = TRUE;
1.53 cvs 2114: }
1.90 cvs 2115: #endif /* _WINDOWS */
1.28 cvs 2116: else
1.96 cvs 2117: /* leave it as it is */
1.106 cvs 2118: strcpy (target, src);
1.96 cvs 2119:
1.28 cvs 2120: /* remove /../ and /./ */
1.29 cvs 2121: SimplifyUrl (&target);
1.54 cvs 2122: if (!change)
1.106 cvs 2123: change = strcmp (src, target);
1.28 cvs 2124: return (change);
1.25 cvs 2125: }
2126:
1.28 cvs 2127:
1.25 cvs 2128: /*----------------------------------------------------------------------
1.31 cvs 2129: MakeRelativeURL: make relative name
1.25 cvs 2130:
1.28 cvs 2131: This function creates and returns a string which gives an expression of
2132: one address as related to another. Where there is no relation, an absolute
2133: address is retured.
1.25 cvs 2134:
1.28 cvs 2135: On entry,
1.25 cvs 2136: Both names must be absolute, fully qualified names of nodes
2137: (no fragment bits)
2138:
1.28 cvs 2139: On exit,
1.25 cvs 2140: The return result points to a newly allocated name which, if
2141: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2142: The caller is responsible for freeing the resulting name later.
2143: ----------------------------------------------------------------------*/
1.106 cvs 2144: char *MakeRelativeURL (char *aName, char *relatedName)
2145: {
2146: char *return_value;
2147: char result[MAX_LENGTH];
2148: char *p;
2149: char *q;
2150: char *after_access;
2151: char *last_slash = NULL;
2152: int slashes, levels, len;
1.110 cvs 2153: #ifdef _WINDOWS
1.44 cvs 2154: int ndx;
1.110 cvs 2155: #endif /* _WINDOWS */
1.44 cvs 2156:
1.29 cvs 2157: if (aName == NULL || relatedName == NULL)
2158: return (NULL);
2159:
2160: slashes = 0;
2161: after_access = NULL;
2162: p = aName;
2163: q = relatedName;
1.147 vatton 2164: len = 0;
2165: for (; *p && !strncasecmp (p, q, 1); p++, q++, len++)
1.27 cvs 2166: {
2167: /* Find extent of match */
1.106 cvs 2168: if (*p == ':')
1.146 cvs 2169: {
2170: after_access = p + 1;
1.147 vatton 2171: if (len == 1)
2172: /* it's a local Windows path like c:... */
2173: slashes++;
1.146 cvs 2174: }
1.28 cvs 2175: if (*p == DIR_SEP)
1.27 cvs 2176: {
1.29 cvs 2177: /* memorize the last slash position and count them */
1.27 cvs 2178: last_slash = p;
2179: slashes++;
1.25 cvs 2180: }
2181: }
2182:
1.31 cvs 2183: /* q, p point to the first non-matching character or zero */
1.106 cvs 2184: if (*q == EOS)
1.31 cvs 2185: {
2186: /* New name is a subset of the related name */
2187: /* exactly the right length */
1.106 cvs 2188: len = strlen (p);
2189: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2190: strcpy (return_value, p);
1.31 cvs 2191: }
2192: else if ((slashes < 2 && after_access == NULL)
2193: || (slashes < 3 && after_access != NULL))
2194: {
2195: /* Two names whitout common path */
2196: /* exactly the right length */
1.106 cvs 2197: len = strlen (aName);
2198: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2199: strcpy (return_value, aName);
1.31 cvs 2200: }
2201: else
2202: {
2203: /* Some path in common */
1.106 cvs 2204: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2205: /* just the same server */
1.106 cvs 2206: strcpy (result, last_slash);
1.31 cvs 2207: else
2208: {
2209: levels= 0;
1.106 cvs 2210: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2211: if (*q == DIR_SEP)
2212: levels++;
2213:
1.106 cvs 2214: result[0] = EOS;
1.31 cvs 2215: for (;levels; levels--)
1.106 cvs 2216: strcat (result, "../");
2217: strcat (result, last_slash+1);
1.31 cvs 2218: }
1.52 cvs 2219:
2220: if (!*result)
1.106 cvs 2221: strcat (result, "./");
1.52 cvs 2222:
1.31 cvs 2223: /* exactly the right length */
1.106 cvs 2224: len = strlen (result);
2225: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2226: strcpy (return_value, result);
1.52 cvs 2227:
1.25 cvs 2228: }
1.110 cvs 2229: #ifdef _WINDOWS
1.106 cvs 2230: len = strlen (return_value);
1.44 cvs 2231: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2232: if (return_value[ndx] == '\\')
2233: return_value[ndx] = '/' ;
1.110 cvs 2234: #endif /* _WINDOWS */
1.29 cvs 2235: return (return_value);
1.24 cvs 2236: }
1.35 cvs 2237:
1.104 kahan 2238: /*----------------------------------------------------------------------
2239: AM_GetFileSize
2240: Returns TRUE and the filesize in the 2nd parameter.
2241: Otherwise, in case of a system error, returns FALSE, with a
2242: filesize of 0L.
2243: ---------------------------------------------------------------------*/
1.106 cvs 2244: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2245: {
1.106 cvs 2246: ThotFileHandle handle = ThotFile_BADHANDLE;
2247: ThotFileInfo info;
1.35 cvs 2248:
1.104 kahan 2249: *file_size = 0L;
2250: if (!TtaFileExist (filename))
2251: return FALSE;
2252:
2253: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2254: if (handle == ThotFile_BADHANDLE)
2255: /* ThotFile_BADHANDLE */
2256: return FALSE;
2257: if (TtaFileStat (handle, &info) == 0)
2258: /* bad stat */
2259: info.size = 0L;
2260: TtaFileClose (handle);
2261: *file_size = (unsigned long) info.size;
2262: return TRUE;
2263: }
1.139 kahan 2264:
2265: /*----------------------------------------------------------------------
2266: AM_UseXHTMLMimeType
2267: Returns TRUE if the user has configured Amaya to use this MIME type,
2268: FALSE otherwise.
2269: ---------------------------------------------------------------------*/
2270: ThotBool AM_UseXHTMLMimeType (void)
2271: {
2272: ThotBool xhtml_mimetype;
2273:
2274: /* does the user wants to use the new MIME type? */
2275: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2276:
2277: return (xhtml_mimetype);
2278: }
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