Annotation of Amaya/amaya/AHTURLTools.c, revision 1.152
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:
1.152 ! kahan 35: #ifdef _WINDOWS
! 36: #define TMPDIR "TMP"
! 37: #else
! 38: #define TMPDIR "TMPDIR"
! 39: #endif /* _WINDOWS */
1.28 cvs 40:
41: /*----------------------------------------------------------------------
42: ConvertToLowerCase
43: Converts a string to lowercase.
44: ----------------------------------------------------------------------*/
1.124 vatton 45: void ConvertToLowerCase (char *string)
1.28 cvs 46: {
47: int i;
1.93 cvs 48:
1.28 cvs 49: if (!string)
50: return;
51:
1.106 cvs 52: for (i = 0; string[i] != EOS; i++)
1.123 vatton 53: string[i] = tolower (string[i]);
1.28 cvs 54: }
1.22 cvs 55:
1.8 cvs 56: /*----------------------------------------------------------------------
1.75 cvs 57: EscapeChar
58: writes the equivalent escape code of a char in a string
59: ----------------------------------------------------------------------*/
1.109 cvs 60: void EscapeChar (char *string, char c)
1.75 cvs 61: {
1.109 cvs 62: unsigned int i;
63:
64: i = (unsigned char) c & 0xFF;
65: sprintf (string, "%02x", i);
1.75 cvs 66: }
67:
68: /*----------------------------------------------------------------------
1.96 cvs 69: UnEscapeChar
70: writes the equivalent hex code to a %xx coded char
71: ----------------------------------------------------------------------*/
1.109 cvs 72: static char UnEscapeChar (char c)
1.96 cvs 73: {
1.106 cvs 74: return c >= '0' && c <= '9' ? c - '0'
75: : c >= 'A' && c <= 'F' ? c - 'A' + 10
76: : c - 'a' + 10; /* accept small letters just in case */
1.96 cvs 77: }
78:
79: /*----------------------------------------------------------------------
1.75 cvs 80: EscapeURL
81: Takes a URL and escapes all protected chars into
82: %xx sequences. Also, removes any leading white spaces
83: Returns either NULL or a new buffer, which must be freed by the caller
84: ----------------------------------------------------------------------*/
1.106 cvs 85: char *EscapeURL (const char *url)
86: {
87: char *buffer;
88: int buffer_len;
89: int buffer_free_mem;
90: char *ptr;
91: int new_chars;
1.75 cvs 92: void *status;
93:
94: if (url && *url)
95: {
1.106 cvs 96: buffer_free_mem = strlen (url) + 20;
97: buffer = TtaGetMemory (buffer_free_mem + 1);
1.107 kahan 98: ptr = (char *) url;
1.75 cvs 99: buffer_len = 0;
100:
101: while (*ptr)
102: {
103: switch (*ptr)
104: {
105: /* put here below all the chars that need to
106: be escaped into %xx */
1.81 cvs 107: case 0x20: /* space */
1.143 vatton 108: /*case 0x26:*/ /* & */
1.140 kahan 109: case 0x27: /* antislash */
1.75 cvs 110: new_chars = 3;
111: break;
112:
113: default:
1.122 kahan 114: if ((unsigned char )*ptr > 127)
115: new_chars = 3;
116: else
117: new_chars = 1;
1.75 cvs 118: break;
119: }
120:
121: /* see if we need extra room in the buffer */
122: if (new_chars > buffer_free_mem)
123: {
1.76 cvs 124: buffer_free_mem = 20;
1.106 cvs 125: status = TtaRealloc (buffer, sizeof (char)
1.75 cvs 126: * (buffer_len + buffer_free_mem + 1));
127: if (status)
1.114 cvs 128: buffer = (char *) status;
1.106 cvs 129: else
130: {
131: /* @@ maybe we should do some other behavior here, like
132: freeing the buffer and return a void thing */
133: buffer[buffer_len] = EOS;
134: break;
135: }
1.75 cvs 136: }
137: /* escape the char */
1.140 kahan 138: if (new_chars == 3)
139: {
1.106 cvs 140: buffer[buffer_len] = '%';
1.75 cvs 141: EscapeChar (&buffer[buffer_len+1], *ptr);
142: }
1.140 kahan 143: else
144: buffer[buffer_len] = *ptr;
145:
146: /* update the status */
147: buffer_len += new_chars;
148: buffer_free_mem -= new_chars;
149: /* examine the next char */
150: ptr++;
151: }
152: buffer[buffer_len] = EOS;
153: }
154: else
155: buffer = NULL;
156:
157: return (buffer);
158: }
159:
160: /*----------------------------------------------------------------------
161: EscapeXML
162: Takes a string and escapes all protected chars into entity
163: sequences.
164: Returns either NULL or a new buffer, which must be freed by the caller
165: ----------------------------------------------------------------------*/
166: char *EscapeXML (const char *string)
167: {
168: char *buffer;
169: int buffer_len;
170: int buffer_free_mem;
171: char *ptr;
172: char *entity = NULL;
173: int new_chars;
174: void *status;
175:
176: if (string && *string)
177: {
178: buffer_free_mem = strlen (string) + 20;
179: buffer = TtaGetMemory (buffer_free_mem + 1);
180: ptr = (char *) string;
181: buffer_len = 0;
182:
183: while (*ptr)
184: {
185: switch (*ptr)
186: {
187: case 0x26: /* & */
188: entity = "&";
1.141 kahan 189: new_chars = sizeof (entity) - 1;
190: break;
1.140 kahan 191:
1.141 kahan 192: case '<': /* < */
1.140 kahan 193: entity = "<";
1.141 kahan 194: new_chars = sizeof (entity) - 1;
1.140 kahan 195: break;
196:
1.141 kahan 197: case '>': /* > */
1.140 kahan 198: entity = ">";
1.141 kahan 199: new_chars = sizeof (entity) - 1;
200: break;
201:
202: case '"': /* "e; */
203: entity = ""e;";
204: new_chars = sizeof (entity) - 1;
1.140 kahan 205: break;
206:
207: default:
208: new_chars = 1;
209: break;
210: }
211:
212: /* see if we need extra room in the buffer */
213: if (new_chars > buffer_free_mem)
214: {
215: buffer_free_mem = 20;
216: status = TtaRealloc (buffer, sizeof (char)
217: * (buffer_len + buffer_free_mem + 1));
218: if (status)
219: buffer = (char *) status;
220: else
221: {
222: /* @@ maybe we should do some other behavior here, like
223: freeing the buffer and return a void thing */
224: buffer[buffer_len] = EOS;
225: break;
226: }
227: }
228: /* escape the char */
229: if (entity)
230: {
231: sprintf (&buffer[buffer_len], "%s", entity);
232: entity = NULL;
233: }
1.75 cvs 234: else
235: buffer[buffer_len] = *ptr;
236:
237: /* update the status */
238: buffer_len += new_chars;
239: buffer_free_mem -= new_chars;
240: /* examine the next char */
241: ptr++;
242: }
1.106 cvs 243: buffer[buffer_len] = EOS;
1.75 cvs 244: }
1.76 cvs 245: else
246: buffer = NULL;
247:
1.75 cvs 248: return (buffer);
1.122 kahan 249: }
250:
1.75 cvs 251:
252: /*----------------------------------------------------------------------
1.11 cvs 253: ExplodeURL
1.8 cvs 254: ----------------------------------------------------------------------*/
1.106 cvs 255: void ExplodeURL (char *url, char **proto, char **host, char **dir,
256: char **file)
1.8 cvs 257: {
1.33 cvs 258: char *curr, *temp;
259: char used_sep;
1.32 cvs 260:
1.33 cvs 261: if (url && strchr (url, URL_SEP))
262: used_sep = URL_SEP;
263: else
264: used_sep = DIR_SEP;
1.8 cvs 265:
266: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
267: (dir == NULL) || (file == NULL))
268: return;
269:
270: /* initialize every pointer */
271: *proto = *host = *dir = *file = NULL;
272:
273: /* skip any leading space */
274: while ((*url == SPACE) || (*url == TAB))
275: url++;
1.9 cvs 276: curr = url;
277: if (*curr == 0)
1.8 cvs 278: goto finished;
279:
280: /* go to the end of the URL */
1.68 cvs 281: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
282: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 283: curr++;
1.8 cvs 284:
285: /* mark the end of the chain */
1.9 cvs 286: *curr = EOS;
287: curr--;
288: if (curr <= url)
1.8 cvs 289: goto finished;
290:
291: /* search the next DIR_SEP indicating the beginning of the file name */
292: do
1.11 cvs 293: curr--;
1.33 cvs 294: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 295:
1.9 cvs 296: if (curr < url)
1.8 cvs 297: goto finished;
1.9 cvs 298: *file = curr + 1;
1.8 cvs 299:
300: /* mark the end of the dir */
1.9 cvs 301: *curr = EOS;
302: curr--;
303: if (curr < url)
1.8 cvs 304: goto finished;
305:
1.29 cvs 306: /* search for the DIR_STR indicating the host name start */
1.33 cvs 307: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 308: curr--;
1.8 cvs 309:
310: /* if we found it, separate the host name from the directory */
1.102 kahan 311: if ((*curr == used_sep) && (*(curr + 1) == used_sep))
1.8 cvs 312: {
1.9 cvs 313: *host = temp = curr + 2;
1.33 cvs 314: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 315: temp++;
1.33 cvs 316: if (*temp == used_sep)
1.8 cvs 317: {
318: *temp = EOS;
319: *dir = temp + 1;
320: }
321: }
322: else
1.11 cvs 323: *dir = curr;
324:
1.9 cvs 325: if (curr <= url)
1.8 cvs 326: goto finished;
327:
328: /* mark the end of the proto */
1.9 cvs 329: *curr = EOS;
330: curr--;
331: if (curr < url)
1.8 cvs 332: goto finished;
333:
1.106 cvs 334: if (*curr == ':')
1.8 cvs 335: {
1.9 cvs 336: *curr = EOS;
337: curr--;
1.8 cvs 338: }
339: else
340: goto finished;
1.11 cvs 341:
1.9 cvs 342: if (curr < url)
1.8 cvs 343: goto finished;
1.9 cvs 344: while ((curr > url) && (isalpha (*curr)))
345: curr--;
346: *proto = curr;
1.8 cvs 347:
348: finished:;
349:
350: #ifdef AMAYA_DEBUG
351: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
352: if (*proto)
353: fprintf (stderr, "proto : %s, ", *proto);
354: if (*host)
355: fprintf (stderr, "host : %s, ", *host);
356: if (*dir)
357: fprintf (stderr, "dir : %s, ", *dir);
358: if (*file)
359: fprintf (stderr, "file : %s ", *file);
360: fprintf (stderr, "\n");
361: #endif
362:
363: }
1.3 cvs 364:
1.116 kahan 365: /*----------------------------------------------------------------------
366: PicTypeToMime
367: Converts a Thot PicType into the equivalent MIME type. If no convertion
368: is possible, it returns NULL.
369: ----------------------------------------------------------------------*/
370: char *PicTypeToMIME (PicType contentType)
371: {
372: char *mime_type;
373:
374: switch (contentType)
375: {
376: case xbm_type:
377: mime_type ="image/x-xbitmap";
378: break;
379: case eps_type:
380: mime_type ="application/postscript";
381: break;
382: case xpm_type:
383: mime_type ="image/x-xpicmap";
384: break;
385: case gif_type:
386: mime_type ="image/gif";
387: break;
388: case jpeg_type:
389: mime_type ="image/jpeg";
390: break;
391: case png_type:
392: mime_type ="image/png";
393: break;
394: case svg_type:
1.138 kahan 395: mime_type =AM_SVG_MIME_TYPE;
1.116 kahan 396: break;
397: case unknown_type:
398: default:
399: mime_type = NULL;
400: }
401:
402: return mime_type;
403: }
1.61 cvs 404:
405: /*----------------------------------------------------------------------
1.117 kahan 406: ImageElement
407: Returns the element (image parameter) and URL (url parameter) of an
408: image in a docImage document. The user must free the memory associated
1.120 kahan 409: with the url parameter if the function is succesful.
410: If the url parameter is NULL, we won't initialize it.
1.117 kahan 411: Returns TRUE if succesful, FALSE otherwise.
412: ----------------------------------------------------------------------*/
413: ThotBool ImageElement (Document doc, char **url, Element *image)
414: {
415: Element el, imgEl;
416: Attribute attr, srcAttr;
417: AttributeType attrType;
418: int length;
419: char *value;
420:
421: if (DocumentTypes[doc] != docImage)
422: return FALSE;
423:
424: /* find the value of the src attribute */
425: attrType.AttrSSchema = TtaGetSSchema ("HTML", doc);
426: attrType.AttrTypeNum = HTML_ATTR_SRC;
427: el = TtaGetRootElement (doc);
428: TtaSearchAttribute (attrType, SearchInTree, el, &imgEl, &srcAttr);
429:
430: if (!imgEl)
431: return FALSE;
432: *image = imgEl;
433:
1.120 kahan 434: if (url)
435: {
436: attr = TtaGetAttribute (imgEl, attrType);
437: length = TtaGetTextAttributeLength (srcAttr) + 1;
438: value = TtaGetMemory (length);
439: TtaGiveTextAttributeValue (srcAttr, value, &length);
440: *url = value;
441: }
1.117 kahan 442: return TRUE;
443: }
444:
445: /*----------------------------------------------------------------------
446: DocImageMimeType
447: Returns the MIME type of a docImage document.
448: ----------------------------------------------------------------------*/
449: char *DocImageMimeType (Document doc)
450: {
451: char *mime_type;
452: LoadedImageDesc *pImage;
453: PicType type;
454: Element image;
455:
456: if (DocumentTypes[doc] != docImage)
457: return NULL;
458:
459: mime_type = NULL;
460: if (!IsHTTPPath (DocumentURLs[doc]))
461: {
462: /* it is a local image */
1.120 kahan 463: if (ImageElement (doc, NULL, &image))
1.117 kahan 464: {
465: type = TtaGetPictureType (image);
466: mime_type = PicTypeToMIME (type);
467: }
468: }
469: else
470: {
471: /* find the value of the src attribute */
472: pImage = ImageURLs;
473: while (pImage != NULL)
474: {
475: if (pImage->document == doc)
476: {
477: if (pImage->content_type)
478: mime_type = pImage->content_type;
479: else if (pImage->elImage && pImage->elImage->currentElement)
480: {
481: type = TtaGetPictureType (pImage->elImage->currentElement);
482: mime_type = PicTypeToMIME (type);
483: }
484: break;
485: }
486: }
487: }
488: return (mime_type);
489: }
490:
1.4 cvs 491: /*----------------------------------------------------------------------
1.9 cvs 492: IsHTMLName
493: returns TRUE if path points to an HTML resource.
1.4 cvs 494: ----------------------------------------------------------------------*/
1.109 cvs 495: ThotBool IsHTMLName (const char *path)
1.106 cvs 496: {
1.136 cvs 497: char temppath[MAX_LENGTH];
498: char suffix[MAX_LENGTH];
499: char nsuffix[MAX_LENGTH];
500: int i;
1.5 cvs 501:
1.101 cvs 502: if (!path)
503: return (FALSE);
1.5 cvs 504:
1.106 cvs 505: strcpy (temppath, path);
1.124 vatton 506: TtaExtractSuffix (temppath, suffix);
1.101 cvs 507: i = 0;
1.106 cvs 508: while (suffix[i] != EOS)
1.101 cvs 509: {
510: /* Normalize the suffix */
511: i = 0;
1.106 cvs 512: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 513: {
1.123 vatton 514: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 515: i++;
516: }
1.106 cvs 517: nsuffix[i] = EOS;
518: if (!strcmp (nsuffix, "html") ||
519: !strcmp (nsuffix, "htm") ||
520: !strcmp (nsuffix, "shtml") ||
521: !strcmp (nsuffix, "jsp") ||
522: !strcmp (nsuffix, "xht") ||
523: !strcmp (nsuffix, "xhtm") ||
1.144 cvs 524: !strcmp (nsuffix, "lhtml") ||
1.106 cvs 525: !strcmp (nsuffix, "xhtml"))
1.101 cvs 526: return (TRUE);
1.106 cvs 527: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 528: {
529: /* take into account compressed files */
1.124 vatton 530: TtaExtractSuffix (temppath, suffix);
1.101 cvs 531: /* Normalize the suffix */
532: i = 0;
1.106 cvs 533: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 534: {
1.123 vatton 535: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 536: i++;
537: }
1.106 cvs 538: nsuffix[i] = EOS;
539: if (!strcmp (nsuffix, "html") ||
540: !strcmp (nsuffix, "htm") ||
541: !strcmp (nsuffix, "shtml") ||
542: !strcmp (nsuffix, "jsp") ||
543: !strcmp (nsuffix, "xht") ||
544: !strcmp (nsuffix, "xhtm") ||
1.144 cvs 545: !strcmp (nsuffix, "lhtml") ||
1.106 cvs 546: !strcmp (nsuffix, "xhtml"))
1.101 cvs 547: return (TRUE);
548: else
549: return (FALSE);
550: }
551: else
552: /* check if there is another suffix */
1.124 vatton 553: TtaExtractSuffix (temppath, suffix);
1.101 cvs 554: }
1.88 cvs 555: return (FALSE);
1.3 cvs 556: }
557:
1.4 cvs 558: /*----------------------------------------------------------------------
1.136 cvs 559: IsMathMLName
560: returns TRUE if path points to an MathML resource.
1.56 cvs 561: ----------------------------------------------------------------------*/
1.136 cvs 562: ThotBool IsMathMLName (const char *path)
1.56 cvs 563: {
1.136 cvs 564: char temppath[MAX_LENGTH];
565: char suffix[MAX_LENGTH];
1.56 cvs 566:
567: if (!path)
568: return (FALSE);
569:
1.106 cvs 570: strcpy (temppath, path);
1.124 vatton 571: TtaExtractSuffix (temppath, suffix);
1.56 cvs 572:
1.136 cvs 573: if (!strcasecmp (suffix, "mml"))
1.56 cvs 574: return (TRUE);
1.106 cvs 575: else if (!strcmp (suffix, "gz"))
1.56 cvs 576: {
577: /* take into account compressed files */
1.124 vatton 578: TtaExtractSuffix (temppath, suffix);
1.136 cvs 579: if (!strcasecmp (suffix, "mml"))
1.60 cvs 580: return (TRUE);
581: else
582: return (FALSE);
583: }
584: else
585: return (FALSE);
586: }
587:
588: /*----------------------------------------------------------------------
1.136 cvs 589: IsSVGName
590: returns TRUE if path points to an SVG resource.
1.133 vatton 591: ----------------------------------------------------------------------*/
1.136 cvs 592: ThotBool IsSVGName (const char *path)
1.133 vatton 593: {
1.136 cvs 594: char temppath[MAX_LENGTH];
595: char suffix[MAX_LENGTH];
1.133 vatton 596:
597: if (!path)
598: return (FALSE);
599:
600: strcpy (temppath, path);
601: TtaExtractSuffix (temppath, suffix);
602:
1.136 cvs 603: if (!strcasecmp (suffix, "svg"))
1.133 vatton 604: return (TRUE);
605: else if (!strcmp (suffix, "gz"))
606: {
607: /* take into account compressed files */
608: TtaExtractSuffix (temppath, suffix);
1.136 cvs 609: if (!strcasecmp (suffix, "svg"))
1.133 vatton 610: return (TRUE);
611: else
612: return (FALSE);
613: }
614: else
615: return (FALSE);
616: }
617:
618: /*----------------------------------------------------------------------
1.136 cvs 619: IsXMLName
620: returns TRUE if path points to an XML resource.
1.103 cvs 621: ----------------------------------------------------------------------*/
1.136 cvs 622: ThotBool IsXMLName (const char *path)
1.103 cvs 623: {
1.136 cvs 624: char temppath[MAX_LENGTH];
625: char suffix[MAX_LENGTH];
1.103 cvs 626:
627: if (!path)
628: return (FALSE);
629:
1.106 cvs 630: strcpy (temppath, path);
1.124 vatton 631: TtaExtractSuffix (temppath, suffix);
1.103 cvs 632:
1.136 cvs 633: if (!strcasecmp (suffix, "xml") ||
634: !strcasecmp (suffix, "xht") ||
635: !strcmp (suffix, "xhtm") ||
1.145 kahan 636: !strcmp (suffix, "xhtml") ||
637: !strcmp (suffix, "smi"))
1.103 cvs 638: return (TRUE);
1.106 cvs 639: else if (!strcmp (suffix, "gz"))
1.103 cvs 640: {
641: /* take into account compressed files */
1.124 vatton 642: TtaExtractSuffix (temppath, suffix);
1.136 cvs 643: if (!strcasecmp (suffix, "xml") ||
644: !strcasecmp (suffix, "xht") ||
645: !strcmp (suffix, "xhtm") ||
1.145 kahan 646: !strcmp (suffix, "xhtml") ||
647: !strcmp (suffix, "smi"))
1.103 cvs 648: return (TRUE);
649: else
650: return (FALSE);
651: }
652: else
653: return (FALSE);
654: }
655:
656: /*----------------------------------------------------------------------
1.136 cvs 657: IsUndisplayedName
658: returns TRUE if path points to an undisplayed resource.
1.103 cvs 659: ----------------------------------------------------------------------*/
1.136 cvs 660: ThotBool IsUndisplayedName (const char *path)
1.103 cvs 661: {
1.106 cvs 662: char temppath[MAX_LENGTH];
663: char suffix[MAX_LENGTH];
1.103 cvs 664:
665: if (!path)
666: return (FALSE);
667:
1.106 cvs 668: strcpy (temppath, path);
1.124 vatton 669: TtaExtractSuffix (temppath, suffix);
1.103 cvs 670:
1.136 cvs 671: if (!strcasecmp (suffix, "exe") ||
672: !strcasecmp (suffix, "zip") ||
673: !strcasecmp (suffix, "ppt") ||
674: !strcasecmp (suffix, "pdf") ||
675: !strcasecmp (suffix, "ps") ||
676: !strcasecmp (suffix, "eps") ||
677: !strcasecmp (suffix, "tar") ||
678: !strcasecmp (suffix, "tgz") ||
679: !strcasecmp (suffix, "ddl") ||
680: !strcasecmp (suffix, "o"))
1.103 cvs 681: return (TRUE);
1.106 cvs 682: else if (!strcmp (suffix, "gz"))
1.103 cvs 683: {
684: /* take into account compressed files */
1.124 vatton 685: TtaExtractSuffix (temppath, suffix);
1.136 cvs 686: if (!strcasecmp (suffix, "exe") ||
687: !strcasecmp (suffix, "zip") ||
688: !strcasecmp (suffix, "ppt") ||
689: !strcasecmp (suffix, "pdf") ||
690: !strcasecmp (suffix, "ps") ||
691: !strcasecmp (suffix, "eps") ||
692: !strcasecmp (suffix, "tar") ||
693: !strcasecmp (suffix, "ddl") ||
694: !strcasecmp (suffix, "o"))
1.103 cvs 695: return (TRUE);
696: else
697: return (FALSE);
698: }
699: else
700: return (FALSE);
701: }
702:
703: /*----------------------------------------------------------------------
1.60 cvs 704: IsCSSName
705: returns TRUE if path points to an XML resource.
706: ----------------------------------------------------------------------*/
1.111 cvs 707: ThotBool IsCSSName (const char *path)
1.60 cvs 708: {
1.106 cvs 709: char temppath[MAX_LENGTH];
710: char suffix[MAX_LENGTH];
1.60 cvs 711:
712: if (!path)
713: return (FALSE);
714:
1.106 cvs 715: strcpy (temppath, path);
1.124 vatton 716: TtaExtractSuffix (temppath, suffix);
1.60 cvs 717:
1.106 cvs 718: if (!strcasecmp (suffix, "css"))
1.60 cvs 719: return (TRUE);
1.106 cvs 720: else if (!strcmp (suffix, "gz"))
1.60 cvs 721: {
722: /* take into account compressed files */
1.124 vatton 723: TtaExtractSuffix (temppath, suffix);
1.106 cvs 724: if (!strcasecmp (suffix, "css"))
1.56 cvs 725: return (TRUE);
726: else
727: return (FALSE);
728: }
729: else
730: return (FALSE);
731: }
732:
733: /*----------------------------------------------------------------------
1.9 cvs 734: IsImageName
735: returns TRUE if path points to an image resource.
1.4 cvs 736: ----------------------------------------------------------------------*/
1.111 cvs 737: ThotBool IsImageName (const char *path)
1.106 cvs 738: {
739: char temppath[MAX_LENGTH];
740: char suffix[MAX_LENGTH];
741: char nsuffix[MAX_LENGTH];
1.5 cvs 742: int i;
743:
744: if (!path)
1.13 cvs 745: return (FALSE);
1.5 cvs 746:
1.106 cvs 747: strcpy (temppath, path);
1.124 vatton 748: TtaExtractSuffix (temppath, suffix);
1.5 cvs 749:
750: /* Normalize the suffix */
751: i = 0;
1.106 cvs 752: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 753: {
1.123 vatton 754: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 755: i++;
756: }
1.106 cvs 757: nsuffix[i] = EOS;
758: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
759: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
760: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 761: return (TRUE);
762: return (FALSE);
1.3 cvs 763: }
764:
1.4 cvs 765: /*----------------------------------------------------------------------
1.58 cvs 766: IsImageType
767: returns TRUE if type points to an image resource.
768: ----------------------------------------------------------------------*/
1.111 cvs 769: ThotBool IsImageType (const char *type)
1.58 cvs 770: {
1.106 cvs 771: char temptype[MAX_LENGTH];
1.58 cvs 772: int i;
773:
774: if (!type)
775: return (FALSE);
776:
1.106 cvs 777: strcpy (temptype, type);
1.58 cvs 778: /* Normalize the type */
779: i = 0;
1.106 cvs 780: while (temptype[i] != EOS)
1.58 cvs 781: {
782: temptype[i] = tolower (temptype[i]);
783: i++;
784: }
1.111 cvs 785: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
786: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
787: !strcmp (temptype, "png"))
1.58 cvs 788: return (TRUE);
789: return (FALSE);
790: }
791:
792: /*----------------------------------------------------------------------
1.9 cvs 793: IsTextName
1.4 cvs 794: ----------------------------------------------------------------------*/
1.111 cvs 795: ThotBool IsTextName (const char *path)
1.106 cvs 796: {
797: char temppath[MAX_LENGTH];
798: char suffix[MAX_LENGTH];
799: char nsuffix[MAX_LENGTH];
1.5 cvs 800: int i;
801:
802: if (!path)
1.13 cvs 803: return (FALSE);
1.5 cvs 804:
1.106 cvs 805: strcpy (temppath, path);
1.124 vatton 806: TtaExtractSuffix (temppath, suffix);
1.5 cvs 807:
808: /* Normalize the suffix */
809: i = 0;
1.106 cvs 810: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 811: {
1.25 cvs 812: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 813: i++;
814: }
1.106 cvs 815: nsuffix[i] = EOS;
1.5 cvs 816:
1.111 cvs 817: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 818: return (TRUE);
1.106 cvs 819: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 820: {
1.39 cvs 821: /* take into account compressed files */
1.124 vatton 822: TtaExtractSuffix (temppath, suffix);
1.13 cvs 823: /* Normalize the suffix */
824: i = 0;
1.106 cvs 825: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 826: {
1.25 cvs 827: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 828: i++;
829: }
1.106 cvs 830: nsuffix[i] = EOS;
1.111 cvs 831: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 832: return (TRUE);
833: else
834: return (FALSE);
835: }
836: else
837: return (FALSE);
1.3 cvs 838: }
839:
1.4 cvs 840: /*----------------------------------------------------------------------
1.9 cvs 841: IsHTTPPath
842: returns TRUE if path is in fact an http URL.
1.4 cvs 843: ----------------------------------------------------------------------*/
1.112 cvs 844: ThotBool IsHTTPPath (const char *path)
1.3 cvs 845: {
1.5 cvs 846: if (!path)
847: return FALSE;
1.3 cvs 848:
1.106 cvs 849: if ((!strncmp (path, "http:", 5) != 0)
850: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
851: || !strncmp (path, "internal:", 9))
1.58 cvs 852: return TRUE;
853: return FALSE;
1.3 cvs 854: }
855:
1.4 cvs 856: /*----------------------------------------------------------------------
1.9 cvs 857: IsWithParameters
858: returns TRUE if url has a concatenated query string.
1.4 cvs 859: ----------------------------------------------------------------------*/
1.133 vatton 860: ThotBool IsWithParameters (const char *url)
1.3 cvs 861: {
1.5 cvs 862: int i;
1.3 cvs 863:
1.9 cvs 864: if ((!url) || (url[0] == EOS))
1.5 cvs 865: return FALSE;
1.3 cvs 866:
1.9 cvs 867: i = strlen (url) - 1;
868: while (i > 0 && url[i--] != '?')
1.5 cvs 869: if (i < 0)
870: return FALSE;
1.3 cvs 871:
1.5 cvs 872: /* There is a parameter */
873: return TRUE;
1.3 cvs 874: }
875:
1.4 cvs 876: /*----------------------------------------------------------------------
1.9 cvs 877: IsW3Path
878: returns TRUE if path is in fact a URL.
1.4 cvs 879: ----------------------------------------------------------------------*/
1.133 vatton 880: ThotBool IsW3Path (const char *path)
1.106 cvs 881: {
882: if (strncmp (path, "http:", 5) &&
883: strncmp (path, "ftp:", 4) &&
884: strncmp (path, "telnet:", 7) &&
885: strncmp (path, "wais:", 5) &&
886: strncmp (path, "news:", 5) &&
887: strncmp (path, "gopher:", 7) &&
888: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 889: strncmp (path, "archie:", 7) &&
890: strncmp (path, "https:", 6))
1.72 cvs 891: return FALSE;
892: return TRUE;
1.3 cvs 893: }
894:
1.4 cvs 895: /*----------------------------------------------------------------------
1.90 cvs 896: IsFilePath
897: returns TRUE if path is in fact a URL.
898: ----------------------------------------------------------------------*/
1.133 vatton 899: ThotBool IsFilePath (const char *path)
1.90 cvs 900: {
1.106 cvs 901: if (strncmp (path, "file:", 5))
1.90 cvs 902: return FALSE;
903: return TRUE;
904: }
905:
906: /*----------------------------------------------------------------------
1.9 cvs 907: IsValidProtocol
908: returns true if the url protocol is supported by Amaya.
1.4 cvs 909: ----------------------------------------------------------------------*/
1.133 vatton 910: ThotBool IsValidProtocol (const char *url)
1.106 cvs 911: {
912: if (!strncmp (url, "http:", 5)
913: || !strncmp (url, "internal:", 9)
914: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 915: /* experimental */
1.24 cvs 916: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 917: return (TRUE);
1.5 cvs 918: else
1.8 cvs 919: return (FALSE);
1.3 cvs 920: }
921:
1.31 cvs 922:
923: /*----------------------------------------------------------------------
924: GetBaseURL
925: normalizes orgName according to a base associated with doc, and
926: following the standard URL format rules.
927: The function returns the base used to solve relative URL and SRC:
928: - the base of the document,
929: - or the document path (without document name).
930: ----------------------------------------------------------------------*/
1.106 cvs 931: char *GetBaseURL (Document doc)
1.31 cvs 932: {
933: Element el;
934: ElementType elType;
935: AttributeType attrType;
936: Attribute attr;
1.106 cvs 937: char *ptr, *basename;
1.31 cvs 938: int length;
1.151 kahan 939: ThotBool hasDocBase;
1.31 cvs 940:
1.113 cvs 941: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 942: return NULL;
1.148 kahan 943: /* the other functions expect basename to have no more than MAX_LENGTH chars */
1.106 cvs 944: basename = TtaGetMemory (MAX_LENGTH);
1.148 kahan 945: basename[0] = EOS;
1.31 cvs 946: length = MAX_LENGTH -1;
1.151 kahan 947: hasDocBase = FALSE;
948:
949: /* If the document has a base URL, it has a priority over the headers. */
950: /* @@ We need to do this too when we support XML:base */
951:
952: /* is it a HTML document ? */
953: elType.ElSSchema = TtaGetDocumentSSchema (doc);
954: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
955: /* it's a HTML document */
956: {
957: /* get the document element */
958: el = TtaGetMainRoot (doc);
959: /* search the BASE element */
960: elType.ElTypeNum = HTML_EL_HEAD;
961: el = TtaSearchTypedElement (elType, SearchForward, el);
962: if (el)
963: /* there is a HEAD element */
964: {
965: /* look for a BASE element within the HEAD */
966: elType.ElTypeNum = HTML_EL_BASE;
967: el = TtaSearchTypedElement (elType, SearchInTree, el);
968: }
969: if (el)
970: {
971: /* The document has a BASE element. Get the HREF attribute of the
972: BASE element */
973: hasDocBase = TRUE;
974: attrType.AttrSSchema = elType.ElSSchema;
975: attrType.AttrTypeNum = HTML_ATTR_HREF_;
976: attr = TtaGetAttribute (el, attrType);
977: if (attr)
978: {
979: /* Use the base path of the document */
980: TtaGiveTextAttributeValue (attr, basename, &length);
981: }
982: }
983: }
984:
985: /* there was no BASE. Do we have a location header? */
986: if (!hasDocBase && DocumentMeta[doc] && DocumentMeta[doc]->full_content_location
1.148 kahan 987: && DocumentMeta[doc]->full_content_location[0] != EOS)
1.65 cvs 988: {
1.148 kahan 989: strncpy (basename, DocumentMeta[doc]->full_content_location, MAX_LENGTH-1);
990: basename[MAX_LENGTH-1] = EOS;
991: length = strlen (basename);
992: }
993:
994: if (basename[0] != EOS)
995: {
996: /* base and orgName have to be separated by a DIR_SEP */
997: length--;
998: if (basename[0] != EOS && basename[length] != URL_SEP &&
999: basename[length] != DIR_SEP)
1000: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 1001: {
1.148 kahan 1002: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1003: AMAYA_PARSE_HOST |
1004: AMAYA_PARSE_PUNCTUATION);
1005: if (ptr && !strcmp (ptr, basename))
1.31 cvs 1006: {
1.148 kahan 1007: /* it has this form, complete it by adding a URL_STR */
1008: if (strchr (basename, DIR_SEP))
1009: strcat (basename, DIR_STR);
1010: else
1011: strcat (basename, URL_STR);
1012: length++;
1.31 cvs 1013: }
1.149 kahan 1014: else if (!ptr || ptr[0] == EOS)
1015: {
1016: /* no host was detected, we may have a relative URL. We test
1017: if it begins with a URL_SEP, DIR_SEP or period. If yes, it's
1018: relative. */
1019: if (! (basename[0] == '.' || basename[0] == URL_SEP
1020: || basename[0] == DIR_SEP))
1021: basename[0] = EOS;
1022: }
1.148 kahan 1023: if (ptr)
1024: TtaFreeMemory (ptr);
1.31 cvs 1025: }
1.113 cvs 1026: }
1027:
1.148 kahan 1028: /* there was no base element and no location header, we use the DocumentURL */
1029: if (basename[0] == EOS)
1030: {
1031: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1032: basename[MAX_LENGTH-1] = EOS;
1033: }
1034:
1.31 cvs 1035: /* Remove anything after the last DIR_SEP char. If no such char is found,
1036: * then search for the first ":" char, hoping that what's before that is a
1037: * protocol. If found, end the string there. If neither char is found,
1038: * then discard the whole base element.
1039: */
1.106 cvs 1040: length = strlen (basename) - 1;
1.31 cvs 1041: /* search for the last DIR_SEP char */
1.106 cvs 1042: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 1043: length--;
1044: if (length >= 0)
1045: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1046: basename[length + 1] = EOS;
1.31 cvs 1047: else
1048: /* search for the first PATH_STR char */
1049: {
1.106 cvs 1050: for (length = 0; basename[length] != ':' &&
1051: basename[length] != EOS; length ++);
1052: if (basename[length] == ':')
1.31 cvs 1053: /* found, so end the string there */
1.106 cvs 1054: basename[length + 1] = EOS;
1.31 cvs 1055: else
1056: /* not found, discard the base */
1.106 cvs 1057: basename[0] = EOS;
1.31 cvs 1058: }
1059: return (basename);
1060: }
1061:
1062:
1.4 cvs 1063: /*----------------------------------------------------------------------
1.40 cvs 1064: GetLocalPath
1065: Allocate and return the local document path associated to the url
1066: ----------------------------------------------------------------------*/
1.150 vatton 1067: char *GetLocalPath (Document doc, char *url)
1.106 cvs 1068: {
1069: char *ptr;
1070: char *n;
1071: char *documentname;
1072: char url_sep;
1.83 cvs 1073: int len;
1.67 cvs 1074: ThotBool noFile;
1.40 cvs 1075:
1076: if (url != NULL)
1077: {
1078: /* check whether the file name exists */
1.106 cvs 1079: len = strlen (url) - 1;
1.71 cvs 1080: if (IsW3Path (url))
1.106 cvs 1081: url_sep = '/';
1.41 cvs 1082: else
1.106 cvs 1083: url_sep = DIR_SEP;
1.41 cvs 1084: noFile = (url[len] == url_sep);
1.40 cvs 1085: if (noFile)
1.106 cvs 1086: url[len] = EOS;
1087: ptr = TtaGetMemory (MAX_LENGTH);
1088: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 1089: TtaExtractName (url, ptr, documentname);
1.106 cvs 1090: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1091: if (!TtaCheckDirectory (ptr))
1092: /* directory did not exist */
1.72 cvs 1093: TtaMakeDirectory (ptr);
1.47 cvs 1094:
1095: /* don't include the query string within document name */
1.106 cvs 1096: n = strrchr (documentname, '?');
1.47 cvs 1097: if (n != NULL)
1.106 cvs 1098: *n = EOS;
1.46 cvs 1099: /* don't include ':' within document name */
1.106 cvs 1100: n = strchr (documentname, ':');
1.46 cvs 1101: if (n != NULL)
1.106 cvs 1102: *n = EOS;
1.69 cvs 1103: /* if after all this operations document name
1104: is empty, let's use noname.html instead */
1.106 cvs 1105: if (documentname[0] == EOS)
1106: strcat (ptr, "noname.html");
1.69 cvs 1107: else
1.106 cvs 1108: strcat (ptr, documentname);
1.40 cvs 1109: TtaFreeMemory (documentname);
1110: /* restore the url */
1111: if (noFile)
1.41 cvs 1112: url[len] = url_sep;
1.40 cvs 1113: return (ptr);
1114: }
1115: else
1116: return (NULL);
1117: }
1118:
1.73 cvs 1119: /*----------------------------------------------------------------------
1.79 cvs 1120: ExtractTarget extract the target name from document nane.
1121: ----------------------------------------------------------------------*/
1.150 vatton 1122: void ExtractTarget (char *aName, char *target)
1.79 cvs 1123: {
1.106 cvs 1124: int lg, i;
1125: char *ptr;
1126: char *oldptr;
1.79 cvs 1127:
1128: if (!target || !aName)
1129: /* bad target */
1130: return;
1131:
1.106 cvs 1132: target[0] = EOS;
1133: lg = strlen (aName);
1.79 cvs 1134: if (lg)
1135: {
1136: /* the name is not empty */
1137: oldptr = ptr = &aName[0];
1138: do
1139: {
1.106 cvs 1140: ptr = strrchr (oldptr, '#');
1.79 cvs 1141: if (ptr)
1142: oldptr = &ptr[1];
1143: }
1144: while (ptr);
1145:
1146: i = (int) (oldptr) - (int) (aName); /* name length */
1147: if (i > 1)
1148: {
1.106 cvs 1149: aName[i - 1] = EOS;
1.79 cvs 1150: if (i != lg)
1.106 cvs 1151: strcpy (target, oldptr);
1.79 cvs 1152: }
1153: }
1154: }
1155:
1156: /*----------------------------------------------------------------------
1.90 cvs 1157: RemoveNewLines (text)
1158: Removes any '\n' chars that are found in text.
1159: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1160: ----------------------------------------------------------------------*/
1.106 cvs 1161: ThotBool RemoveNewLines (char *text)
1162: {
1163: ThotBool change = FALSE;
1164: char *src;
1165: char *dest;
1.90 cvs 1166:
1167: src = text;
1168: dest = text;
1.115 kahan 1169:
1170: /* remove any preceding whitespace */
1171: while (*src && *src == ' ')
1172: {
1173: src++;
1174: change = 1;
1175: }
1176:
1.90 cvs 1177: while (*src)
1178: {
1179: switch (*src)
1180: {
1.106 cvs 1181: case '\n':
1.90 cvs 1182: /* don't copy the newline */
1183: change = 1;
1184: break;
1185: default:
1186: *dest = *src;
1187: dest++;
1188: break;
1189: }
1190: src++;
1191: }
1192: /* copy the last EOS char */
1193: *dest = *src;
1194:
1195: return (change);
1196: }
1197:
1198: /*----------------------------------------------------------------------
1199: CleanCopyFileURL
1200: Copies a file url from a src string to destination string.
1.97 cvs 1201: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1202: we want to do).
1.90 cvs 1203: ----------------------------------------------------------------------*/
1.106 cvs 1204: static void CleanCopyFileURL (char *dest, char *src,
1205: ConvertionType convertion)
1.90 cvs 1206: {
1207: while (*src)
1.89 cvs 1208: {
1.90 cvs 1209: switch (*src)
1.89 cvs 1210: {
1211: #ifdef _WINDOWS
1.106 cvs 1212: case URL_SEP:
1.96 cvs 1213: /* make DIR_SEP transformation */
1.97 cvs 1214: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1215: *dest = DIR_SEP;
1.96 cvs 1216: else
1217: *dest = *src;
1.90 cvs 1218: dest++;
1.96 cvs 1219: src++;
1.90 cvs 1220: break;
1.89 cvs 1221: #endif /* _WINDOWS */
1.96 cvs 1222:
1.106 cvs 1223: case '%':
1.97 cvs 1224: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1225: {
1.97 cvs 1226: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1227: src++;
1.106 cvs 1228: if (*src != EOS)
1.97 cvs 1229: {
1230: *dest = UnEscapeChar (*src) * 16;
1231: src++;
1232: }
1.106 cvs 1233: if (*src != EOS)
1.97 cvs 1234: {
1235: *dest = *dest + UnEscapeChar (*src);
1236: src++;
1237: }
1238: dest++;
1.96 cvs 1239: }
1.97 cvs 1240: else
1.96 cvs 1241: {
1.97 cvs 1242: *dest = *src;
1243: dest++;
1.96 cvs 1244: src++;
1245: }
1246: break;
1247:
1.90 cvs 1248: default:
1249: *dest = *src;
1.89 cvs 1250: dest++;
1.96 cvs 1251: src++;
1.90 cvs 1252: break;
1.89 cvs 1253: }
1254: }
1.90 cvs 1255: /* copy the EOS char */
1256: *dest = *src;
1.73 cvs 1257: }
1.40 cvs 1258:
1259: /*----------------------------------------------------------------------
1.9 cvs 1260: NormalizeURL
1261: normalizes orgName according to a base associated with doc, and
1262: following the standard URL format rules.
1.113 cvs 1263: if doc is < 0, use as a base the URL of the document that contains
1264: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1265: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1266: other path.
1.9 cvs 1267: The function returns the new complete and normalized URL
1.12 cvs 1268: or file name path (newName) and the name of the document (docName).
1.9 cvs 1269: N.B. If the function can't find out what's the docName, it assigns
1270: the name "noname.html".
1.4 cvs 1271: ----------------------------------------------------------------------*/
1.106 cvs 1272: void NormalizeURL (char *orgName, Document doc, char *newName,
1273: char *docName, char *otherPath)
1274: {
1275: char *basename;
1276: char tempOrgName[MAX_LENGTH];
1277: char *ptr;
1278: char used_sep;
1.84 cvs 1279: int length;
1280: ThotBool check;
1.5 cvs 1281:
1.110 cvs 1282: #ifdef _WINDOWS
1.44 cvs 1283: int ndx;
1.110 cvs 1284: #endif /* _WINDOWS */
1.44 cvs 1285:
1.5 cvs 1286: if (!newName || !docName)
1287: return;
1.18 cvs 1288:
1.113 cvs 1289: if (doc < 0)
1290: basename = TtaStrdup (SavedDocumentURL);
1291: else if (doc > 0)
1.53 cvs 1292: basename = GetBaseURL (doc);
1293: else if (otherPath != NULL)
1.108 cvs 1294: basename = TtaStrdup (otherPath);
1.32 cvs 1295: else
1.53 cvs 1296: basename = NULL;
1.32 cvs 1297:
1.18 cvs 1298: /*
1.31 cvs 1299: * Clean orgName
1300: * Make sure we have a complete orgName, without any leading or trailing
1301: * white spaces, or trailinbg new lines
1302: */
1.5 cvs 1303: ptr = orgName;
1.18 cvs 1304: /* skip leading white space and new line characters */
1.106 cvs 1305: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1306: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1307: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1308: /*
1.31 cvs 1309: * Make orgName a complete URL
1310: * If the URL does not include a protocol, then try to calculate
1311: * one using the doc's base element (if it exists),
1312: */
1.106 cvs 1313: if (tempOrgName[0] == EOS)
1.53 cvs 1314: {
1.106 cvs 1315: newName[0] = EOS;
1316: docName[0] = EOS;
1.53 cvs 1317: TtaFreeMemory (basename);
1318: return;
1319: }
1.49 cvs 1320:
1321: /* clean trailing white space */
1.106 cvs 1322: length = strlen (tempOrgName) - 1;
1323: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1324: {
1.106 cvs 1325: tempOrgName[length] = EOS;
1.53 cvs 1326: length--;
1327: }
1.50 cvs 1328:
1.55 cvs 1329: /* remove extra dot (which dot???) */
1330: /* ugly, but faster than a strcmp */
1.106 cvs 1331: if (tempOrgName[length] == '.'
1332: && (length == 0 || tempOrgName[length-1] != '.'))
1333: tempOrgName[length] = EOS;
1.50 cvs 1334:
1.94 cvs 1335: if (IsW3Path (tempOrgName))
1.53 cvs 1336: {
1337: /* the name is complete, go to the Sixth Step */
1.106 cvs 1338: strcpy (newName, tempOrgName);
1.53 cvs 1339: SimplifyUrl (&newName);
1340: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1341: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1342: AMAYA_PARSE_HOST |
1343: AMAYA_PARSE_PUNCTUATION);
1344: if (ptr && !strcmp (ptr, newName))
1345: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1346: strcat (newName, URL_STR);
1.49 cvs 1347:
1.53 cvs 1348: if (ptr)
1.50 cvs 1349: TtaFreeMemory (ptr);
1.53 cvs 1350: }
1.113 cvs 1351: else if (basename == NULL)
1.53 cvs 1352: /* the name is complete, go to the Sixth Step */
1.106 cvs 1353: strcpy (newName, tempOrgName);
1.53 cvs 1354: else
1355: {
1.31 cvs 1356: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1357: #ifdef _WINDOWS
1.53 cvs 1358: if (!IsW3Path (basename))
1359: {
1.106 cvs 1360: length = strlen (tempOrgName);
1.53 cvs 1361: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1362: if (tempOrgName [ndx] == '/')
1363: tempOrgName [ndx] = '\\';
1.53 cvs 1364: }
1.110 cvs 1365: #endif /* _WINDOWS */
1.25 cvs 1366: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1367: if (ptr)
1368: {
1369: SimplifyUrl (&ptr);
1.106 cvs 1370: strcpy (newName, ptr);
1.53 cvs 1371: TtaFreeMemory (ptr);
1372: }
1373: else
1.106 cvs 1374: newName[0] = EOS;
1.53 cvs 1375: }
1.36 cvs 1376:
1377: TtaFreeMemory (basename);
1.18 cvs 1378: /*
1.31 cvs 1379: * Prepare the docname that will refer to this ressource in the
1380: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1381: * noname.html as a default ressource name
1.18 cvs 1382: */
1.106 cvs 1383: if (newName[0] != EOS)
1.53 cvs 1384: {
1.106 cvs 1385: length = strlen (newName) - 1;
1386: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1387: {
1388: used_sep = newName[length];
1389: check = TRUE;
1390: while (check)
1391: {
1.50 cvs 1392: length--;
1393: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1394: length--;
1.106 cvs 1395: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1396: {
1.106 cvs 1397: newName[length+1] = EOS;
1.53 cvs 1398: /* remove also previous directory */
1399: length--;
1400: while (length >= 0 && newName[length] != used_sep)
1401: length--;
1.106 cvs 1402: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1403: /* don't remove server name */
1.106 cvs 1404: newName[length+1] = EOS;
1.53 cvs 1405: }
1.106 cvs 1406: else if (!strncmp (&newName[length+1], ".", 1))
1407: newName[length+1] = EOS;
1.50 cvs 1408: else
1.53 cvs 1409: check = FALSE;
1410: }
1411: /* docname was not comprised inside the URL, so let's */
1412: /* assign the default ressource name */
1.106 cvs 1413: strcpy (docName, "noname.html");
1.53 cvs 1414: }
1415: else
1416: { /* docname is comprised inside the URL */
1.110 cvs 1417: while (length >= 0 && newName[length] != URL_SEP &&
1418: newName[length] != DIR_SEP)
1.53 cvs 1419: length--;
1420: if (length < 0)
1.106 cvs 1421: strcpy (docName, newName);
1.53 cvs 1422: else
1.106 cvs 1423: strcpy (docName, &newName[length+1]);
1.53 cvs 1424: }
1425: }
1426: else
1.106 cvs 1427: docName[0] = EOS;
1.18 cvs 1428: }
1.3 cvs 1429:
1.4 cvs 1430: /*----------------------------------------------------------------------
1.9 cvs 1431: IsSameHost
1.4 cvs 1432: ----------------------------------------------------------------------*/
1.106 cvs 1433: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1434: {
1.106 cvs 1435: char *basename_ptr1, *basename_ptr2;
1436: ThotBool result;
1.3 cvs 1437:
1.106 cvs 1438: basename_ptr1 = AmayaParseUrl (url1, "",
1439: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1440: basename_ptr2 = AmayaParseUrl (url2, "",
1441: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1442:
1.106 cvs 1443: if (strcmp (basename_ptr1, basename_ptr2))
1444: result = FALSE;
1445: else
1446: result = TRUE;
1447: TtaFreeMemory (basename_ptr1);
1448: TtaFreeMemory (basename_ptr2);
1449: return (result);
1.3 cvs 1450: }
1451:
1452:
1.4 cvs 1453: /*----------------------------------------------------------------------
1.22 cvs 1454: HasKnownFileSuffix
1455: returns TRUE if path points to a file ending with a suffix.
1456: ----------------------------------------------------------------------*/
1.106 cvs 1457: ThotBool HasKnownFileSuffix (const char *path)
1458: {
1459: char *root;
1460: char temppath[MAX_LENGTH];
1461: char suffix[MAX_LENGTH];
1.22 cvs 1462:
1.106 cvs 1463: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1464: return (FALSE);
1465:
1.106 cvs 1466: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1467:
1468: if (root)
1469: {
1.106 cvs 1470: strcpy (temppath, root);
1.25 cvs 1471: TtaFreeMemory (root);
1.22 cvs 1472: /* Get the suffix */
1.124 vatton 1473: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1474:
1.106 cvs 1475: if( suffix[0] == EOS)
1.22 cvs 1476: /* no suffix */
1477: return (FALSE);
1478:
1479: /* Normalize the suffix */
1480: ConvertToLowerCase (suffix);
1481:
1.106 cvs 1482: if (!strcmp (suffix, "gz"))
1.22 cvs 1483: /* skip the compressed suffix */
1484: {
1.124 vatton 1485: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1486: if(suffix[0] == EOS)
1.22 cvs 1487: /* no suffix */
1488: return (FALSE);
1489: /* Normalize the suffix */
1490: ConvertToLowerCase (suffix);
1491: }
1492:
1.106 cvs 1493: if (strcmp (suffix, "gif") &&
1494: strcmp (suffix, "xbm") &&
1495: strcmp (suffix, "xpm") &&
1496: strcmp (suffix, "jpg") &&
1497: strcmp (suffix, "pdf") &&
1498: strcmp (suffix, "png") &&
1499: strcmp (suffix, "tgz") &&
1500: strcmp (suffix, "xpg") &&
1501: strcmp (suffix, "xpd") &&
1502: strcmp (suffix, "ps") &&
1503: strcmp (suffix, "au") &&
1504: strcmp (suffix, "html") &&
1505: strcmp (suffix, "htm") &&
1506: strcmp (suffix, "shtml") &&
1507: strcmp (suffix, "xht") &&
1508: strcmp (suffix, "xhtm") &&
1509: strcmp (suffix, "xhtml") &&
1510: strcmp (suffix, "txt") &&
1511: strcmp (suffix, "css") &&
1512: strcmp (suffix, "eps"))
1.22 cvs 1513: return (FALSE);
1514: else
1515: return (TRUE);
1516: }
1517: else
1518: return (FALSE);
1519: }
1520:
1521:
1522: /*----------------------------------------------------------------------
1.24 cvs 1523: ChopURL
1524: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1525: If inputURL is bigger than that size, outputURL receives
1526: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1527: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1528: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1529: copied into outputURL.
1530: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1531: chars.
1532: ----------------------------------------------------------------------*/
1.106 cvs 1533: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1534: {
1535: int len;
1.9 cvs 1536:
1.106 cvs 1537: len = strlen (inputURL);
1.24 cvs 1538: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1539: strcpy (outputURL, inputURL);
1.24 cvs 1540: else
1541: /* make a truncated urlName on the status window */
1542: {
1.106 cvs 1543: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1544: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1545: strcat (outputURL, "...");
1546: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1547: }
1.25 cvs 1548: }
1549:
1550:
1551: /*----------------------------------------------------------------------
1552: scan
1.47 cvs 1553: Scan a filename for its constituents
1.25 cvs 1554: -----------------------------------
1555:
1556: On entry,
1557: name points to a document name which may be incomplete.
1558: On exit,
1559: absolute or relative may be nonzero (but not both).
1560: host, fragment and access may be nonzero if they were specified.
1561: Any which are nonzero point to zero terminated strings.
1562: ----------------------------------------------------------------------*/
1.106 cvs 1563: static void scan (char *name, HTURI *parts)
1.25 cvs 1564: {
1.106 cvs 1565: char * p;
1566: char * after_access = name;
1.32 cvs 1567:
1.43 cvs 1568: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1569: /* Look for fragment identifier */
1.106 cvs 1570: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1571: {
1.106 cvs 1572: *p++ = '\0';
1.28 cvs 1573: parts->fragment = p;
1.25 cvs 1574: }
1575:
1.28 cvs 1576: for (p=name; *p; p++)
1577: {
1.106 cvs 1578: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1579: break;
1.106 cvs 1580: if (*p == ':')
1.28 cvs 1581: {
1582: *p = 0;
1583: parts->access = after_access; /* Scheme has been specified */
1584:
1585: /* The combination of gcc, the "-O" flag and the HP platform is
1586: unhealthy. The following three lines is a quick & dirty fix, but is
1587: not recommended. Rather, turn off "-O". */
1588:
1589: /* after_access = p;*/
1590: /* while (*after_access == 0)*/
1591: /* after_access++;*/
1592: after_access = p+1;
1.106 cvs 1593: if (!strcasecmp("URL", parts->access))
1.28 cvs 1594: /* Ignore IETF's URL: pre-prefix */
1595: parts->access = NULL;
1596: else
1.25 cvs 1597: break;
1598: }
1599: }
1600:
1601: p = after_access;
1.43 cvs 1602: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1603: {
1.43 cvs 1604: if (p[1] == URL_SEP)
1.28 cvs 1605: {
1.25 cvs 1606: parts->host = p+2; /* host has been specified */
1.28 cvs 1607: *p = 0; /* Terminate access */
1608: /* look for end of host name if any */
1.106 cvs 1609: p = strchr (parts->host, URL_SEP);
1.28 cvs 1610: if (p)
1611: {
1.106 cvs 1612: *p = EOS; /* Terminate host */
1.25 cvs 1613: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1614: }
1615: }
1616: else
1617: /* Root found but no host */
1618: parts->absolute = p+1;
1619: }
1620: else
1621: {
1.25 cvs 1622: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1623: }
1.25 cvs 1624: }
1625:
1626:
1627: /*----------------------------------------------------------------------
1.28 cvs 1628: AmayaParseUrl: parse a Name relative to another name
1629:
1630: This returns those parts of a name which are given (and requested)
1631: substituting bits from the related name where necessary.
1.25 cvs 1632:
1.28 cvs 1633: On entry,
1.25 cvs 1634: aName A filename given
1635: relatedName A name relative to which aName is to be parsed. Give
1636: it an empty string if aName is absolute.
1637: wanted A mask for the bits which are wanted.
1638:
1.28 cvs 1639: On exit,
1.25 cvs 1640: returns A pointer to a malloc'd string which MUST BE FREED
1641: ----------------------------------------------------------------------*/
1.106 cvs 1642: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1643: {
1644: char *return_value;
1645: char result[MAX_LENGTH];
1646: char name[MAX_LENGTH];
1647: char rel[MAX_LENGTH];
1648: char *p, *access;
1.29 cvs 1649: HTURI given, related;
1650: int len;
1.106 cvs 1651: char used_sep;
1652: char *used_str;
1.32 cvs 1653:
1.106 cvs 1654: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1655: {
1.106 cvs 1656: used_str = DIR_STR;
1657: used_sep = DIR_SEP;
1.33 cvs 1658: }
1.32 cvs 1659: else
1.33 cvs 1660: {
1.106 cvs 1661: used_str = URL_STR;
1662: used_sep = URL_SEP;
1.33 cvs 1663: }
1.32 cvs 1664:
1.29 cvs 1665: /* Make working copies of input strings to cut up: */
1666: return_value = NULL;
1667: result[0] = 0; /* Clear string */
1.106 cvs 1668: strcpy (name, aName);
1.29 cvs 1669: if (relatedName != NULL)
1.106 cvs 1670: strcpy (rel, relatedName);
1.29 cvs 1671: else
1.106 cvs 1672: relatedName[0] = EOS;
1.29 cvs 1673:
1674: scan (name, &given);
1675: scan (rel, &related);
1676: access = given.access ? given.access : related.access;
1677: if (wanted & AMAYA_PARSE_ACCESS)
1678: if (access)
1679: {
1.106 cvs 1680: strcat (result, access);
1.29 cvs 1681: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1682: strcat (result, ":");
1.29 cvs 1683: }
1684:
1685: if (given.access && related.access)
1686: /* If different, inherit nothing. */
1.106 cvs 1687: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1688: {
1689: related.host = 0;
1690: related.absolute = 0;
1691: related.relative = 0;
1692: related.fragment = 0;
1693: }
1694:
1695: if (wanted & AMAYA_PARSE_HOST)
1696: if(given.host || related.host)
1697: {
1698: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1699: strcat (result, "//");
1700: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1701: }
1702:
1703: if (given.host && related.host)
1704: /* If different hosts, inherit no path. */
1.106 cvs 1705: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1706: {
1707: related.absolute = 0;
1708: related.relative = 0;
1709: related.fragment = 0;
1710: }
1711:
1712: if (wanted & AMAYA_PARSE_PATH)
1713: {
1714: if (given.absolute)
1715: {
1716: /* All is given */
1717: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1718: strcat (result, used_str);
1719: strcat (result, given.absolute);
1.25 cvs 1720: }
1.29 cvs 1721: else if (related.absolute)
1722: {
1723: /* Adopt path not name */
1.106 cvs 1724: strcat (result, used_str);
1725: strcat (result, related.absolute);
1.29 cvs 1726: if (given.relative)
1727: {
1728: /* Search part? */
1.106 cvs 1729: p = strchr (result, '?');
1.29 cvs 1730: if (!p)
1.106 cvs 1731: p=result+strlen(result)-1;
1.33 cvs 1732: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1733: /* Remove filename */
1734: p[1]=0;
1735: /* Add given one */
1.106 cvs 1736: strcat (result, given.relative);
1.25 cvs 1737: }
1738: }
1.29 cvs 1739: else if (given.relative)
1740: /* what we've got */
1.106 cvs 1741: strcat (result, given.relative);
1.29 cvs 1742: else if (related.relative)
1.106 cvs 1743: strcat (result, related.relative);
1.29 cvs 1744: else
1745: /* No inheritance */
1.106 cvs 1746: strcat (result, used_str);
1.25 cvs 1747: }
1.29 cvs 1748:
1749: if (wanted & AMAYA_PARSE_ANCHOR)
1750: if (given.fragment || related.fragment)
1751: {
1752: if (given.absolute && given.fragment)
1753: {
1754: /*Fixes for relURLs...*/
1755: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1756: strcat (result, "#");
1757: strcat (result, given.fragment);
1.29 cvs 1758: }
1759: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1760: strcat (result, "");
1.29 cvs 1761: else
1762: {
1.110 cvs 1763: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1764: strcat (result, "#");
1.110 cvs 1765: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1766: }
1767: }
1.106 cvs 1768: len = strlen (result);
1769: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1770: strcpy (return_value, result);
1.29 cvs 1771: return (return_value); /* exactly the right length */
1.25 cvs 1772: }
1773:
1774: /*----------------------------------------------------------------------
1775: HTCanon
1776: Canonicalizes the URL in the following manner starting from the host
1777: pointer:
1778:
1779: 1) The host name is converted to lowercase
1780: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1781:
1782: Return: OK The position of the current path part of the URL
1783: which might be the old one or a new one.
1784:
1785: ----------------------------------------------------------------------*/
1.106 cvs 1786: static char *HTCanon (char **filename, char *host)
1787: {
1788: char *newname = NULL;
1789: char used_sep;
1790: char *path;
1791: char *strptr;
1792: char *port;
1793: char *access = host-3;
1794:
1795: if (*filename && strchr (*filename, URL_SEP))
1796: used_sep = URL_SEP;
1.33 cvs 1797: else
1.106 cvs 1798: used_sep = DIR_SEP;
1.32 cvs 1799:
1.110 cvs 1800: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1801: access--;
1.110 cvs 1802: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1803: path = host + strlen (host);
1804: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1805: host = strptr;
1.110 cvs 1806: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1807: port = NULL;
1.25 cvs 1808:
1809: strptr = host; /* Convert to lower-case */
1.82 cvs 1810: while (strptr < path)
1.33 cvs 1811: {
1.123 vatton 1812: *strptr = tolower (*strptr);
1.82 cvs 1813: strptr++;
1.33 cvs 1814: }
1.25 cvs 1815:
1816: /* Does the URL contain a full domain name? This also works for a
1817: numerical host name. The domain name is already made lower-case
1818: and without a trailing dot. */
1819: {
1.106 cvs 1820: char *dot = port ? port : path;
1821: if (dot > *filename && *--dot == '.')
1.33 cvs 1822: {
1.106 cvs 1823: char *orig = dot;
1824: char *dest = dot + 1;
1.82 cvs 1825: while ((*orig++ = *dest++));
1826: if (port) port--;
1.33 cvs 1827: path--;
1.25 cvs 1828: }
1829: }
1830: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1831: if (port)
1832: {
1.82 cvs 1833: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1834: {
1835: if (!newname)
1836: {
1.106 cvs 1837: char *orig = port;
1838: char *dest = port + 1;
1.82 cvs 1839: while ((*orig++ = *dest++));
1.33 cvs 1840: }
1841: }
1.106 cvs 1842: else if ((!strncmp (access, "http", 4) &&
1843: (*(port + 1) == '8' &&
1844: *(port+2) == '0' &&
1.82 cvs 1845: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1846: (!strncmp (access, "gopher", 6) &&
1847: (*(port+1) == '7' &&
1848: *(port+2) == '0' &&
1.82 cvs 1849: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1850: (!strncmp (access, "ftp", 3) &&
1851: (*(port+1) == '2' &&
1852: *(port + 2) == '1' &&
1.82 cvs 1853: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1854: if (!newname)
1855: {
1.106 cvs 1856: char *orig = port;
1857: char *dest = port + 3;
1.33 cvs 1858: while((*orig++ = *dest++));
1859: /* Update path position, Henry Minsky */
1860: path -= 3;
1.25 cvs 1861: }
1.33 cvs 1862: }
1863: else if (newname)
1.106 cvs 1864: strncat (newname, port, (int) (path - port));
1.33 cvs 1865: }
1.25 cvs 1866:
1.33 cvs 1867: if (newname)
1868: {
1.106 cvs 1869: char *newpath = newname + strlen (newname);
1870: strcat (newname, path);
1.25 cvs 1871: path = newpath;
1.28 cvs 1872: /* Free old copy */
1873: TtaFreeMemory(*filename);
1.25 cvs 1874: *filename = newname;
1.33 cvs 1875: }
1.25 cvs 1876: return path;
1877: }
1878:
1879:
1880: /*----------------------------------------------------------------------
1.29 cvs 1881: SimplifyUrl: simplify a URI
1.32 cvs 1882: A URI is allowed to contain the sequence xxx/../ which may be
1883: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1884: Simplification helps us recognize duplicate URIs.
1.25 cvs 1885:
1.28 cvs 1886: Thus, /etc/junk/../fred becomes /etc/fred
1887: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1888:
1.28 cvs 1889: but we should NOT change
1890: http://fred.xxx.edu/../..
1.25 cvs 1891:
1892: or ../../albert.html
1893:
1.28 cvs 1894: In order to avoid empty URLs the following URLs become:
1.25 cvs 1895:
1896: /fred/.. becomes /fred/..
1897: /fred/././.. becomes /fred/..
1898: /fred/.././junk/.././ becomes /fred/..
1899:
1.28 cvs 1900: If more than one set of `://' is found (several proxies in cascade) then
1901: only the part after the last `://' is simplified.
1.25 cvs 1902:
1.28 cvs 1903: Returns: A string which might be the old one or a new one.
1.25 cvs 1904: ----------------------------------------------------------------------*/
1.106 cvs 1905: void SimplifyUrl (char **url)
1906: {
1907: char *path;
1908: char *access;
1909: char *newptr;
1910: char *p;
1911: char *orig, *dest, *end;
1.28 cvs 1912:
1.106 cvs 1913: char used_sep;
1.77 cvs 1914: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1915: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1916:
1.28 cvs 1917: if (!url || !*url)
1918: return;
1919:
1.106 cvs 1920: if (strchr (*url, URL_SEP))
1921: used_sep = URL_SEP;
1.32 cvs 1922: else
1.106 cvs 1923: used_sep = DIR_SEP;
1.32 cvs 1924:
1.77 cvs 1925: /* should we simplify double dot? */
1926: path = *url;
1.106 cvs 1927: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1928: ddot_simplify = FALSE;
1929: else
1930: ddot_simplify = TRUE;
1931:
1.28 cvs 1932: /* Find any scheme name */
1.106 cvs 1933: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1934: {
1935: /* Find host name */
1.28 cvs 1936: access = *url;
1.123 vatton 1937: while (access < path && (*access = tolower (*access)))
1.82 cvs 1938: access++;
1.28 cvs 1939: path += 3;
1.106 cvs 1940: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1941: /* For proxies */
1.106 cvs 1942: path = newptr + 3;
1.82 cvs 1943: /* We have a host name */
1.84 cvs 1944: path = HTCanon (url, path);
1.25 cvs 1945: }
1.106 cvs 1946: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1947: path += 2;
1948: else
1949: path = *url;
1.84 cvs 1950: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1951: /* Some URLs start //<foo> */
1952: path += 1;
1.94 cvs 1953: else if (IsFilePath (path))
1954: {
1955: /* doesn't need to do anything more */
1956: return;
1957: }
1.106 cvs 1958: else if (!strncmp (path, "news:", 5))
1.28 cvs 1959: {
1.106 cvs 1960: newptr = strchr (path+5, '@');
1.28 cvs 1961: if (!newptr)
1962: newptr = path + 5;
1963: while (*newptr)
1964: {
1965: /* Make group or host lower case */
1.123 vatton 1966: *newptr = tolower (*newptr);
1.28 cvs 1967: newptr++;
1.25 cvs 1968: }
1.28 cvs 1969: /* Doesn't need to do any more */
1970: return;
1.25 cvs 1971: }
1.130 cheyroul 1972:
1.126 cheyroul 1973:
1.28 cvs 1974: if ((p = path))
1975: {
1.106 cvs 1976: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1977: (end = strchr (path, '#'))))
1978: end = path + strlen (path);
1.28 cvs 1979:
1980: /* Parse string second time to simplify */
1981: p = path;
1982: while (p < end)
1983: {
1.110 cvs 1984: /* if we're pointing to a char, it's safe to reactivate the
1985: ../ convertion */
1.106 cvs 1986: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1987: ddot_simplify = TRUE;
1988:
1.33 cvs 1989: if (*p==used_sep)
1.28 cvs 1990: {
1.106 cvs 1991: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1992: {
1993: orig = p + 1;
1.84 cvs 1994: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1995: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1996: end = orig - 1;
1997: }
1.106 cvs 1998: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1999: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 2000: {
2001: newptr = p;
1.52 cvs 2002: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
2003: if (*newptr == used_sep)
2004: orig = newptr + 1;
1.28 cvs 2005: else
1.52 cvs 2006: orig = newptr;
2007:
2008: dest = (*(p+3) != used_sep) ? p+3 : p+4;
2009: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
2010: end = orig-1;
2011: /* Start again with prev slash */
2012: p = newptr;
1.28 cvs 2013: }
1.33 cvs 2014: else if (*(p+1) == used_sep)
1.28 cvs 2015: {
1.33 cvs 2016: while (*(p+1) == used_sep)
1.28 cvs 2017: {
2018: orig = p;
2019: dest = p + 1;
2020: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2021: end = orig-1;
2022: }
2023: }
2024: else
1.25 cvs 2025: p++;
1.28 cvs 2026: }
2027: else
2028: p++;
1.25 cvs 2029: }
2030: }
1.51 cvs 2031: /*
2032: ** Check for host/../.. kind of things
2033: */
1.106 cvs 2034: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 2035: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 2036: *(path+1) = EOS;
1.28 cvs 2037: return;
2038: }
2039:
2040:
2041: /*----------------------------------------------------------------------
1.96 cvs 2042: NormalizeFile normalizes local names.
1.28 cvs 2043: Return TRUE if target and src differ.
2044: ----------------------------------------------------------------------*/
1.106 cvs 2045: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2046: {
1.110 cvs 2047: #ifndef _WINDOWS
1.106 cvs 2048: char *s;
1.93 cvs 2049: int i;
1.110 cvs 2050: #endif /* !_WINDOWS */
1.82 cvs 2051: ThotBool change;
1.90 cvs 2052: int start_index; /* the first char that we'll copy */
1.28 cvs 2053:
1.54 cvs 2054: change = FALSE;
1.90 cvs 2055: start_index = 0;
2056:
1.106 cvs 2057: if (!src || src[0] == EOS)
1.96 cvs 2058: {
1.106 cvs 2059: target[0] = EOS;
1.96 cvs 2060: return FALSE;
2061: }
1.90 cvs 2062:
2063: /* @@ do I need file: or file:/ here? */
1.106 cvs 2064: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 2065: {
1.90 cvs 2066: /* remove the prefix file: */
2067: start_index += 5;
2068:
2069: /* remove the localhost prefix */
1.106 cvs 2070: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 2071: start_index += 11;
2072:
2073: /* remove the first two slashes in / / /path */
2074: while (src[start_index] &&
1.106 cvs 2075: src[start_index] == '/'
2076: && src[start_index + 1] == '/')
1.94 cvs 2077: start_index++;
2078:
2079: #ifdef _WINDOWS
2080: /* remove any extra slash before the drive name */
1.106 cvs 2081: if (src[start_index] == '/'
2082: &&src[start_index+2] == ':')
1.94 cvs 2083: start_index++;
2084: #endif /* _WINDOWS */
1.90 cvs 2085:
1.106 cvs 2086: if (src[start_index] == EOS)
1.90 cvs 2087: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 2088: strcpy (target, DIR_STR);
1.90 cvs 2089: else
1.97 cvs 2090: /* as we're inside a file: URL, we'll apply all the convertions
2091: we know */
2092: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 2093:
2094: change = TRUE;
2095: }
1.97 cvs 2096: else if (convertion != AM_CONV_NONE)
1.96 cvs 2097: {
2098: /* we are following a "local" relative link, we do all the
2099: convertions except for the HOME_DIR ~ one */
1.97 cvs 2100: CleanCopyFileURL (target, src, convertion);
1.28 cvs 2101: }
1.90 cvs 2102: #ifndef _WINDOWS
1.106 cvs 2103: else if (src[0] == '~')
1.53 cvs 2104: {
1.96 cvs 2105: /* it must be a URL typed in a text input field */
2106: /* do the HOME_DIR ~ substitution */
1.82 cvs 2107: s = TtaGetEnvString ("HOME");
1.106 cvs 2108: strcpy (target, s);
1.90 cvs 2109: #if 0
1.96 cvs 2110: /* JK: invalidated this part of the code as it's simpler
2111: to add the DIR_SEP whenever we have something to add
2112: to the path rather than adding it systematically */
1.106 cvs 2113: if (src[1] != DIR_SEP)
2114: strcat (target, DIR_STR);
1.90 cvs 2115: #endif
1.106 cvs 2116: i = strlen (target);
2117: strcpy (&target[i], &src[1]);
1.54 cvs 2118: change = TRUE;
1.53 cvs 2119: }
1.90 cvs 2120: #endif /* _WINDOWS */
1.28 cvs 2121: else
1.96 cvs 2122: /* leave it as it is */
1.106 cvs 2123: strcpy (target, src);
1.96 cvs 2124:
1.28 cvs 2125: /* remove /../ and /./ */
1.29 cvs 2126: SimplifyUrl (&target);
1.54 cvs 2127: if (!change)
1.106 cvs 2128: change = strcmp (src, target);
1.28 cvs 2129: return (change);
1.25 cvs 2130: }
2131:
1.28 cvs 2132:
1.25 cvs 2133: /*----------------------------------------------------------------------
1.31 cvs 2134: MakeRelativeURL: make relative name
1.25 cvs 2135:
1.28 cvs 2136: This function creates and returns a string which gives an expression of
2137: one address as related to another. Where there is no relation, an absolute
2138: address is retured.
1.25 cvs 2139:
1.28 cvs 2140: On entry,
1.25 cvs 2141: Both names must be absolute, fully qualified names of nodes
2142: (no fragment bits)
2143:
1.28 cvs 2144: On exit,
1.25 cvs 2145: The return result points to a newly allocated name which, if
2146: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2147: The caller is responsible for freeing the resulting name later.
2148: ----------------------------------------------------------------------*/
1.106 cvs 2149: char *MakeRelativeURL (char *aName, char *relatedName)
2150: {
2151: char *return_value;
2152: char result[MAX_LENGTH];
2153: char *p;
2154: char *q;
2155: char *after_access;
2156: char *last_slash = NULL;
2157: int slashes, levels, len;
1.110 cvs 2158: #ifdef _WINDOWS
1.44 cvs 2159: int ndx;
1.110 cvs 2160: #endif /* _WINDOWS */
1.44 cvs 2161:
1.29 cvs 2162: if (aName == NULL || relatedName == NULL)
2163: return (NULL);
2164:
2165: slashes = 0;
2166: after_access = NULL;
2167: p = aName;
2168: q = relatedName;
1.147 vatton 2169: len = 0;
2170: for (; *p && !strncasecmp (p, q, 1); p++, q++, len++)
1.27 cvs 2171: {
2172: /* Find extent of match */
1.106 cvs 2173: if (*p == ':')
1.146 cvs 2174: {
2175: after_access = p + 1;
1.147 vatton 2176: if (len == 1)
2177: /* it's a local Windows path like c:... */
2178: slashes++;
1.146 cvs 2179: }
1.28 cvs 2180: if (*p == DIR_SEP)
1.27 cvs 2181: {
1.29 cvs 2182: /* memorize the last slash position and count them */
1.27 cvs 2183: last_slash = p;
2184: slashes++;
1.25 cvs 2185: }
2186: }
2187:
1.31 cvs 2188: /* q, p point to the first non-matching character or zero */
1.106 cvs 2189: if (*q == EOS)
1.31 cvs 2190: {
2191: /* New name is a subset of the related name */
2192: /* exactly the right length */
1.106 cvs 2193: len = strlen (p);
2194: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2195: strcpy (return_value, p);
1.31 cvs 2196: }
2197: else if ((slashes < 2 && after_access == NULL)
2198: || (slashes < 3 && after_access != NULL))
2199: {
2200: /* Two names whitout common path */
2201: /* exactly the right length */
1.106 cvs 2202: len = strlen (aName);
2203: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2204: strcpy (return_value, aName);
1.31 cvs 2205: }
2206: else
2207: {
2208: /* Some path in common */
1.106 cvs 2209: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2210: /* just the same server */
1.106 cvs 2211: strcpy (result, last_slash);
1.31 cvs 2212: else
2213: {
2214: levels= 0;
1.106 cvs 2215: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2216: if (*q == DIR_SEP)
2217: levels++;
2218:
1.106 cvs 2219: result[0] = EOS;
1.31 cvs 2220: for (;levels; levels--)
1.106 cvs 2221: strcat (result, "../");
2222: strcat (result, last_slash+1);
1.31 cvs 2223: }
1.52 cvs 2224:
2225: if (!*result)
1.106 cvs 2226: strcat (result, "./");
1.52 cvs 2227:
1.31 cvs 2228: /* exactly the right length */
1.106 cvs 2229: len = strlen (result);
2230: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2231: strcpy (return_value, result);
1.52 cvs 2232:
1.25 cvs 2233: }
1.110 cvs 2234: #ifdef _WINDOWS
1.106 cvs 2235: len = strlen (return_value);
1.44 cvs 2236: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2237: if (return_value[ndx] == '\\')
2238: return_value[ndx] = '/' ;
1.110 cvs 2239: #endif /* _WINDOWS */
1.29 cvs 2240: return (return_value);
1.24 cvs 2241: }
1.35 cvs 2242:
1.104 kahan 2243: /*----------------------------------------------------------------------
2244: AM_GetFileSize
2245: Returns TRUE and the filesize in the 2nd parameter.
2246: Otherwise, in case of a system error, returns FALSE, with a
2247: filesize of 0L.
2248: ---------------------------------------------------------------------*/
1.106 cvs 2249: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2250: {
1.106 cvs 2251: ThotFileHandle handle = ThotFile_BADHANDLE;
2252: ThotFileInfo info;
1.35 cvs 2253:
1.104 kahan 2254: *file_size = 0L;
2255: if (!TtaFileExist (filename))
2256: return FALSE;
2257:
2258: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2259: if (handle == ThotFile_BADHANDLE)
2260: /* ThotFile_BADHANDLE */
2261: return FALSE;
2262: if (TtaFileStat (handle, &info) == 0)
2263: /* bad stat */
2264: info.size = 0L;
2265: TtaFileClose (handle);
2266: *file_size = (unsigned long) info.size;
2267: return TRUE;
2268: }
1.139 kahan 2269:
2270: /*----------------------------------------------------------------------
2271: AM_UseXHTMLMimeType
2272: Returns TRUE if the user has configured Amaya to use this MIME type,
2273: FALSE otherwise.
2274: ---------------------------------------------------------------------*/
2275: ThotBool AM_UseXHTMLMimeType (void)
2276: {
2277: ThotBool xhtml_mimetype;
2278:
2279: /* does the user wants to use the new MIME type? */
2280: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2281:
2282: return (xhtml_mimetype);
1.152 ! kahan 2283: }
! 2284:
! 2285: /*-----------------------------------------------------------------------
! 2286: GetTempName
! 2287: Front end to the Unix tempnam function, which is independent of the
! 2288: value of the TMPDIR env value
! 2289: Returns a dynamically allocated string with a tempname. The user
! 2290: must free this memory.
! 2291: -----------------------------------------------------------------------*/
! 2292: char *GetTempName (const char *dir, const char *prefix)
! 2293: {
! 2294: char *tmpdir;
! 2295: char *tmp;
! 2296: char *name = NULL;
! 2297:
! 2298: /* save the value of TMPDIR */
! 2299: tmp = getenv (TMPDIR);
! 2300:
! 2301: if (tmp)
! 2302: {
! 2303: tmpdir = TtaStrdup (tmp);
! 2304: }
! 2305: else
! 2306: tmpdir = NULL;
! 2307:
! 2308: /* remove TMPDIR from the environment */
! 2309: if (tmpdir)
! 2310: {
! 2311: tmp = TtaGetMemory (strlen (tmpdir) + 2);
! 2312: sprintf (tmp, "%s=", TMPDIR);
! 2313: #ifdef _WINDOWS
! 2314: _putenv (tmp);
! 2315: #else
! 2316: putenv (tmp);
! 2317: #endif /* _WINDOWS */
! 2318: }
! 2319:
! 2320: /* create the tempname */
! 2321: #ifdef _WINDOWS
! 2322: /* Under Windows, _tempnam returns the same name until the file is created */
! 2323: {
! 2324: char *altprefix;
! 2325: name = tmpnam (NULL); /* get a possibly unique string */
! 2326: altprefix = TtaGetMemory(strlen (prefix) + strlen(name) + 1);
! 2327: sprintf (altprefix, "%s%s", prefix, name + strlen(_P_tmpdir));
! 2328: name = _tempnam (dir, altprefix); /* get a name that isn't yet in use */
! 2329: TtaFreeMemory (altprefix);
! 2330: }
! 2331: #else
! 2332: name = tempnam (dir, prefix);
! 2333: #endif /* _WINDOWS */
! 2334:
! 2335: /* restore the value of TMPDIR */
! 2336: if (tmpdir)
! 2337: {
! 2338: #ifdef _WINDOWS
! 2339: _putenv (tmpdir);
! 2340: #else
! 2341: putenv (tmpdir);
! 2342: #endif /* _WINDOWS */
! 2343: /* Shouldn't be free (see man for putenv ()) */
! 2344: /* TtaFreeMemory (tmpdir); */
! 2345: }
! 2346: return (name);
1.139 kahan 2347: }
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