Annotation of Amaya/amaya/AHTURLTools.c, revision 1.121
1.7 cvs 1: /*
2: *
1.110 cvs 3: * (c) COPYRIGHT MIT and INRIA, 1996-2001
1.7 cvs 4: * Please first read the full copyright statement in file COPYRIGHT.
5: *
6: */
1.9 cvs 7:
1.10 cvs 8: /*
9: * AHTURLTools.c: contains all the functions for testing, manipulating,
1.25 cvs 10: * and normalizing URLs. It also contains a local copy of the libWWW
11: * URL parsing functions.
1.10 cvs 12: *
13: * Authors: J. Kahan, I. Vatton
1.106 cvs 14: * R. Guetari: Windows.
1.10 cvs 15: *
16: */
1.7 cvs 17:
1.15 cvs 18: #define THOT_EXPORT extern
1.3 cvs 19: #include "amaya.h"
20:
1.8 cvs 21: #include "init_f.h"
22: #include "AHTURLTools_f.h"
1.100 kahan 23: #include "query_f.h"
1.8 cvs 24:
1.24 cvs 25: #define MAX_PRINT_URL_LENGTH 50
1.106 cvs 26: typedef struct _HTURI
27: {
28: char *access; /* Now known as "scheme" */
29: char *host;
30: char *absolute;
31: char *relative;
32: char *fragment;
1.29 cvs 33: } HTURI;
1.24 cvs 34:
1.28 cvs 35:
36: /*----------------------------------------------------------------------
37: ConvertToLowerCase
38: Converts a string to lowercase.
39: ----------------------------------------------------------------------*/
1.106 cvs 40: void ConvertToLowerCase (char *string)
1.28 cvs 41: {
42: int i;
1.93 cvs 43:
1.28 cvs 44: if (!string)
45: return;
46:
1.106 cvs 47: for (i = 0; string[i] != EOS; i++)
1.67 cvs 48: string[i] = utolower (string[i]);
1.28 cvs 49: }
1.22 cvs 50:
1.8 cvs 51: /*----------------------------------------------------------------------
1.75 cvs 52: EscapeChar
53: writes the equivalent escape code of a char in a string
54: ----------------------------------------------------------------------*/
1.109 cvs 55: void EscapeChar (char *string, char c)
1.75 cvs 56: {
1.109 cvs 57: unsigned int i;
58:
59: i = (unsigned char) c & 0xFF;
60: sprintf (string, "%02x", i);
1.75 cvs 61: }
62:
63: /*----------------------------------------------------------------------
1.96 cvs 64: UnEscapeChar
65: writes the equivalent hex code to a %xx coded char
66: ----------------------------------------------------------------------*/
1.109 cvs 67: static char UnEscapeChar (char c)
1.96 cvs 68: {
1.106 cvs 69: return c >= '0' && c <= '9' ? c - '0'
70: : c >= 'A' && c <= 'F' ? c - 'A' + 10
71: : c - 'a' + 10; /* accept small letters just in case */
1.96 cvs 72: }
73:
74: /*----------------------------------------------------------------------
1.75 cvs 75: EscapeURL
76: Takes a URL and escapes all protected chars into
77: %xx sequences. Also, removes any leading white spaces
78: Returns either NULL or a new buffer, which must be freed by the caller
79: ----------------------------------------------------------------------*/
1.106 cvs 80: char *EscapeURL (const char *url)
81: {
82: char *buffer;
83: int buffer_len;
84: int buffer_free_mem;
85: char *ptr;
86: int new_chars;
1.75 cvs 87: void *status;
88:
89: if (url && *url)
90: {
1.106 cvs 91: buffer_free_mem = strlen (url) + 20;
92: buffer = TtaGetMemory (buffer_free_mem + 1);
1.107 kahan 93: ptr = (char *) url;
1.75 cvs 94: buffer_len = 0;
95:
96: while (*ptr)
97: {
98: switch (*ptr)
99: {
100: /* put here below all the chars that need to
101: be escaped into %xx */
1.81 cvs 102: case 0x27: /* & */
103: case 0x20: /* space */
1.75 cvs 104: new_chars = 3;
105: break;
106:
107: default:
108: new_chars = 1;
109: break;
110: }
111:
112: /* see if we need extra room in the buffer */
113: if (new_chars > buffer_free_mem)
114: {
1.76 cvs 115: buffer_free_mem = 20;
1.106 cvs 116: status = TtaRealloc (buffer, sizeof (char)
1.75 cvs 117: * (buffer_len + buffer_free_mem + 1));
118: if (status)
1.114 cvs 119: buffer = (char *) status;
1.106 cvs 120: else
121: {
122: /* @@ maybe we should do some other behavior here, like
123: freeing the buffer and return a void thing */
124: buffer[buffer_len] = EOS;
125: break;
126: }
1.75 cvs 127: }
128: /* escape the char */
129: if (new_chars == 3)
130: {
1.106 cvs 131: buffer[buffer_len] = '%';
1.75 cvs 132: EscapeChar (&buffer[buffer_len+1], *ptr);
133: }
134: else
135: buffer[buffer_len] = *ptr;
136:
137: /* update the status */
138: buffer_len += new_chars;
139: buffer_free_mem -= new_chars;
140: /* examine the next char */
141: ptr++;
142: }
1.106 cvs 143: buffer[buffer_len] = EOS;
1.75 cvs 144: }
1.76 cvs 145: else
146: buffer = NULL;
147:
1.75 cvs 148: return (buffer);
149: }
150:
151:
152: /*----------------------------------------------------------------------
1.11 cvs 153: ExplodeURL
1.8 cvs 154: ----------------------------------------------------------------------*/
1.106 cvs 155: void ExplodeURL (char *url, char **proto, char **host, char **dir,
156: char **file)
1.8 cvs 157: {
1.33 cvs 158: char *curr, *temp;
159: char used_sep;
1.32 cvs 160:
1.33 cvs 161: if (url && strchr (url, URL_SEP))
162: used_sep = URL_SEP;
163: else
164: used_sep = DIR_SEP;
1.8 cvs 165:
166: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
167: (dir == NULL) || (file == NULL))
168: return;
169:
170: /* initialize every pointer */
171: *proto = *host = *dir = *file = NULL;
172:
173: /* skip any leading space */
174: while ((*url == SPACE) || (*url == TAB))
175: url++;
1.9 cvs 176: curr = url;
177: if (*curr == 0)
1.8 cvs 178: goto finished;
179:
180: /* go to the end of the URL */
1.68 cvs 181: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
182: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 183: curr++;
1.8 cvs 184:
185: /* mark the end of the chain */
1.9 cvs 186: *curr = EOS;
187: curr--;
188: if (curr <= url)
1.8 cvs 189: goto finished;
190:
191: /* search the next DIR_SEP indicating the beginning of the file name */
192: do
1.11 cvs 193: curr--;
1.33 cvs 194: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 195:
1.9 cvs 196: if (curr < url)
1.8 cvs 197: goto finished;
1.9 cvs 198: *file = curr + 1;
1.8 cvs 199:
200: /* mark the end of the dir */
1.9 cvs 201: *curr = EOS;
202: curr--;
203: if (curr < url)
1.8 cvs 204: goto finished;
205:
1.29 cvs 206: /* search for the DIR_STR indicating the host name start */
1.33 cvs 207: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 208: curr--;
1.8 cvs 209:
210: /* if we found it, separate the host name from the directory */
1.102 kahan 211: if ((*curr == used_sep) && (*(curr + 1) == used_sep))
1.8 cvs 212: {
1.9 cvs 213: *host = temp = curr + 2;
1.33 cvs 214: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 215: temp++;
1.33 cvs 216: if (*temp == used_sep)
1.8 cvs 217: {
218: *temp = EOS;
219: *dir = temp + 1;
220: }
221: }
222: else
1.11 cvs 223: *dir = curr;
224:
1.9 cvs 225: if (curr <= url)
1.8 cvs 226: goto finished;
227:
228: /* mark the end of the proto */
1.9 cvs 229: *curr = EOS;
230: curr--;
231: if (curr < url)
1.8 cvs 232: goto finished;
233:
1.106 cvs 234: if (*curr == ':')
1.8 cvs 235: {
1.9 cvs 236: *curr = EOS;
237: curr--;
1.8 cvs 238: }
239: else
240: goto finished;
1.11 cvs 241:
1.9 cvs 242: if (curr < url)
1.8 cvs 243: goto finished;
1.9 cvs 244: while ((curr > url) && (isalpha (*curr)))
245: curr--;
246: *proto = curr;
1.8 cvs 247:
248: finished:;
249:
250: #ifdef AMAYA_DEBUG
251: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
252: if (*proto)
253: fprintf (stderr, "proto : %s, ", *proto);
254: if (*host)
255: fprintf (stderr, "host : %s, ", *host);
256: if (*dir)
257: fprintf (stderr, "dir : %s, ", *dir);
258: if (*file)
259: fprintf (stderr, "file : %s ", *file);
260: fprintf (stderr, "\n");
261: #endif
262:
263: }
1.3 cvs 264:
1.116 kahan 265: /*----------------------------------------------------------------------
266: PicTypeToMime
267: Converts a Thot PicType into the equivalent MIME type. If no convertion
268: is possible, it returns NULL.
269: ----------------------------------------------------------------------*/
270: char *PicTypeToMIME (PicType contentType)
271: {
272: char *mime_type;
273:
274: switch (contentType)
275: {
276: case xbm_type:
277: mime_type ="image/x-xbitmap";
278: break;
279: case eps_type:
280: mime_type ="application/postscript";
281: break;
282: case xpm_type:
283: mime_type ="image/x-xpicmap";
284: break;
285: case gif_type:
286: mime_type ="image/gif";
287: break;
288: case jpeg_type:
289: mime_type ="image/jpeg";
290: break;
291: case png_type:
292: mime_type ="image/png";
293: break;
294: case svg_type:
1.118 kahan 295: mime_type ="image/svg+xml";
1.116 kahan 296: break;
297: case unknown_type:
298: default:
299: mime_type = NULL;
300: }
301:
302: return mime_type;
303: }
1.61 cvs 304:
305: /*----------------------------------------------------------------------
1.117 kahan 306: ImageElement
307: Returns the element (image parameter) and URL (url parameter) of an
308: image in a docImage document. The user must free the memory associated
1.120 kahan 309: with the url parameter if the function is succesful.
310: If the url parameter is NULL, we won't initialize it.
1.117 kahan 311: Returns TRUE if succesful, FALSE otherwise.
312: ----------------------------------------------------------------------*/
313: ThotBool ImageElement (Document doc, char **url, Element *image)
314: {
315: Element el, imgEl;
316: Attribute attr, srcAttr;
317: AttributeType attrType;
318: int length;
319: char *value;
320:
321: if (DocumentTypes[doc] != docImage)
322: return FALSE;
323:
324: /* find the value of the src attribute */
325: attrType.AttrSSchema = TtaGetSSchema ("HTML", doc);
326: attrType.AttrTypeNum = HTML_ATTR_SRC;
327: el = TtaGetRootElement (doc);
328: TtaSearchAttribute (attrType, SearchInTree, el, &imgEl, &srcAttr);
329:
330: if (!imgEl)
331: return FALSE;
332: *image = imgEl;
333:
1.120 kahan 334: if (url)
335: {
336: attr = TtaGetAttribute (imgEl, attrType);
337: length = TtaGetTextAttributeLength (srcAttr) + 1;
338: value = TtaGetMemory (length);
339: TtaGiveTextAttributeValue (srcAttr, value, &length);
340: *url = value;
341: }
1.117 kahan 342: return TRUE;
343: }
344:
345: /*----------------------------------------------------------------------
346: DocImageMimeType
347: Returns the MIME type of a docImage document.
348: ----------------------------------------------------------------------*/
349: char *DocImageMimeType (Document doc)
350: {
351: char *mime_type;
352: LoadedImageDesc *pImage;
353: PicType type;
354: Element image;
355:
356: if (DocumentTypes[doc] != docImage)
357: return NULL;
358:
359: mime_type = NULL;
360: if (!IsHTTPPath (DocumentURLs[doc]))
361: {
362: /* it is a local image */
1.120 kahan 363: if (ImageElement (doc, NULL, &image))
1.117 kahan 364: {
365: type = TtaGetPictureType (image);
366: mime_type = PicTypeToMIME (type);
367: }
368: }
369: else
370: {
371: /* find the value of the src attribute */
372: pImage = ImageURLs;
373: while (pImage != NULL)
374: {
375: if (pImage->document == doc)
376: {
377: if (pImage->content_type)
378: mime_type = pImage->content_type;
379: else if (pImage->elImage && pImage->elImage->currentElement)
380: {
381: type = TtaGetPictureType (pImage->elImage->currentElement);
382: mime_type = PicTypeToMIME (type);
383: }
384: break;
385: }
386: }
387: }
388: return (mime_type);
389: }
390:
391: /*----------------------------------------------------------------------
1.61 cvs 392: ExtractSuffix extract suffix from document nane.
393: ----------------------------------------------------------------------*/
1.106 cvs 394: void ExtractSuffix (char *aName, char *aSuffix)
1.61 cvs 395: {
1.106 cvs 396: int lg, i;
397: char *ptr, *oldptr;
1.61 cvs 398:
399: if (!aSuffix || !aName)
400: /* bad suffix */
401: return;
402:
1.106 cvs 403: aSuffix[0] = EOS;
404: lg = strlen (aName);
1.61 cvs 405: if (lg)
406: {
407: /* the name is not empty */
408: oldptr = ptr = &aName[0];
409: do
410: {
1.106 cvs 411: ptr = strrchr (oldptr, '.');
1.61 cvs 412: if (ptr)
413: oldptr = &ptr[1];
414: }
415: while (ptr);
416:
417: i = (int) (oldptr) - (int) (aName); /* name length */
418: if (i > 1)
419: {
1.106 cvs 420: aName[i - 1] = EOS;
1.61 cvs 421: if (i != lg)
1.106 cvs 422: strcpy (aSuffix, oldptr);
1.61 cvs 423: }
424: }
425: }
426:
1.4 cvs 427: /*----------------------------------------------------------------------
1.9 cvs 428: IsHTMLName
429: returns TRUE if path points to an HTML resource.
1.4 cvs 430: ----------------------------------------------------------------------*/
1.109 cvs 431: ThotBool IsHTMLName (const char *path)
1.106 cvs 432: {
433: char temppath[MAX_LENGTH];
434: char suffix[MAX_LENGTH];
435: char nsuffix[MAX_LENGTH];
1.101 cvs 436: int i;
1.5 cvs 437:
1.101 cvs 438: if (!path)
439: return (FALSE);
1.5 cvs 440:
1.106 cvs 441: strcpy (temppath, path);
1.101 cvs 442: ExtractSuffix (temppath, suffix);
443: i = 0;
1.106 cvs 444: while (suffix[i] != EOS)
1.101 cvs 445: {
446: /* Normalize the suffix */
447: i = 0;
1.106 cvs 448: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 449: {
450: nsuffix[i] = utolower (suffix[i]);
451: i++;
452: }
1.106 cvs 453: nsuffix[i] = EOS;
454: if (!strcmp (nsuffix, "html") ||
455: !strcmp (nsuffix, "htm") ||
456: !strcmp (nsuffix, "shtml") ||
457: !strcmp (nsuffix, "jsp") ||
458: !strcmp (nsuffix, "xht") ||
459: !strcmp (nsuffix, "xhtm") ||
460: !strcmp (nsuffix, "xhtml"))
1.101 cvs 461: return (TRUE);
1.106 cvs 462: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 463: {
464: /* take into account compressed files */
465: ExtractSuffix (temppath, suffix);
466: /* Normalize the suffix */
467: i = 0;
1.106 cvs 468: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 469: {
470: nsuffix[i] = utolower (suffix[i]);
471: i++;
472: }
1.106 cvs 473: nsuffix[i] = EOS;
474: if (!strcmp (nsuffix, "html") ||
475: !strcmp (nsuffix, "htm") ||
476: !strcmp (nsuffix, "shtml") ||
477: !strcmp (nsuffix, "jsp") ||
478: !strcmp (nsuffix, "xht") ||
479: !strcmp (nsuffix, "xhtm") ||
480: !strcmp (nsuffix, "xhtml"))
1.101 cvs 481: return (TRUE);
482: else
483: return (FALSE);
484: }
485: else
486: /* check if there is another suffix */
487: ExtractSuffix (temppath, suffix);
488: }
1.88 cvs 489: return (FALSE);
1.3 cvs 490: }
491:
1.4 cvs 492: /*----------------------------------------------------------------------
1.56 cvs 493: IsXMLName
494: returns TRUE if path points to an XML resource.
495: ----------------------------------------------------------------------*/
1.111 cvs 496: ThotBool IsXMLName (const char *path)
1.56 cvs 497: {
1.106 cvs 498: char temppath[MAX_LENGTH];
499: char suffix[MAX_LENGTH];
1.56 cvs 500:
501: if (!path)
502: return (FALSE);
503:
1.106 cvs 504: strcpy (temppath, path);
1.56 cvs 505: ExtractSuffix (temppath, suffix);
506:
1.106 cvs 507: if (!strcasecmp (suffix, "xml") ||
508: !strcasecmp (suffix, "xht") ||
509: !strcmp (suffix, "xhtm") ||
510: !strcmp (suffix, "xhtml"))
1.56 cvs 511: return (TRUE);
1.106 cvs 512: else if (!strcmp (suffix, "gz"))
1.56 cvs 513: {
514: /* take into account compressed files */
515: ExtractSuffix (temppath, suffix);
1.106 cvs 516: if (!strcasecmp (suffix, "xml") ||
517: !strcasecmp (suffix, "xht") ||
518: !strcmp (suffix, "xhtm") ||
519: !strcmp (suffix, "xhtml"))
1.60 cvs 520: return (TRUE);
521: else
522: return (FALSE);
523: }
524: else
525: return (FALSE);
526: }
527:
528: /*----------------------------------------------------------------------
1.103 cvs 529: IsMathMLName
530: returns TRUE if path points to an MathML resource.
531: ----------------------------------------------------------------------*/
1.111 cvs 532: ThotBool IsMathMLName (const char *path)
1.103 cvs 533: {
1.106 cvs 534: char temppath[MAX_LENGTH];
535: char suffix[MAX_LENGTH];
1.103 cvs 536:
537: if (!path)
538: return (FALSE);
539:
1.106 cvs 540: strcpy (temppath, path);
1.103 cvs 541: ExtractSuffix (temppath, suffix);
542:
1.106 cvs 543: if (!strcasecmp (suffix, "mml"))
1.103 cvs 544: return (TRUE);
1.106 cvs 545: else if (!strcmp (suffix, "gz"))
1.103 cvs 546: {
547: /* take into account compressed files */
548: ExtractSuffix (temppath, suffix);
1.106 cvs 549: if (!strcasecmp (suffix, "mml"))
1.103 cvs 550: return (TRUE);
551: else
552: return (FALSE);
553: }
554: else
555: return (FALSE);
556: }
557:
558: /*----------------------------------------------------------------------
559: IsSVGName
1.119 kahan 560: returns TRUE if path points to an SVG resource.
1.103 cvs 561: ----------------------------------------------------------------------*/
1.111 cvs 562: ThotBool IsSVGName (const char *path)
1.103 cvs 563: {
1.106 cvs 564: char temppath[MAX_LENGTH];
565: char suffix[MAX_LENGTH];
1.103 cvs 566:
567: if (!path)
568: return (FALSE);
569:
1.106 cvs 570: strcpy (temppath, path);
1.103 cvs 571: ExtractSuffix (temppath, suffix);
572:
1.106 cvs 573: if (!strcasecmp (suffix, "svg"))
1.103 cvs 574: return (TRUE);
1.106 cvs 575: else if (!strcmp (suffix, "gz"))
1.103 cvs 576: {
577: /* take into account compressed files */
578: ExtractSuffix (temppath, suffix);
1.106 cvs 579: if (!strcasecmp (suffix, "svg"))
1.103 cvs 580: return (TRUE);
581: else
582: return (FALSE);
583: }
584: else
585: return (FALSE);
586: }
587:
588: /*----------------------------------------------------------------------
1.60 cvs 589: IsCSSName
590: returns TRUE if path points to an XML resource.
591: ----------------------------------------------------------------------*/
1.111 cvs 592: ThotBool IsCSSName (const char *path)
1.60 cvs 593: {
1.106 cvs 594: char temppath[MAX_LENGTH];
595: char suffix[MAX_LENGTH];
1.60 cvs 596:
597: if (!path)
598: return (FALSE);
599:
1.106 cvs 600: strcpy (temppath, path);
1.60 cvs 601: ExtractSuffix (temppath, suffix);
602:
1.106 cvs 603: if (!strcasecmp (suffix, "css"))
1.60 cvs 604: return (TRUE);
1.106 cvs 605: else if (!strcmp (suffix, "gz"))
1.60 cvs 606: {
607: /* take into account compressed files */
608: ExtractSuffix (temppath, suffix);
1.106 cvs 609: if (!strcasecmp (suffix, "css"))
1.56 cvs 610: return (TRUE);
611: else
612: return (FALSE);
613: }
614: else
615: return (FALSE);
616: }
617:
618: /*----------------------------------------------------------------------
1.9 cvs 619: IsImageName
620: returns TRUE if path points to an image resource.
1.4 cvs 621: ----------------------------------------------------------------------*/
1.111 cvs 622: ThotBool IsImageName (const char *path)
1.106 cvs 623: {
624: char temppath[MAX_LENGTH];
625: char suffix[MAX_LENGTH];
626: char nsuffix[MAX_LENGTH];
1.5 cvs 627: int i;
628:
629: if (!path)
1.13 cvs 630: return (FALSE);
1.5 cvs 631:
1.106 cvs 632: strcpy (temppath, path);
1.5 cvs 633: ExtractSuffix (temppath, suffix);
634:
635: /* Normalize the suffix */
636: i = 0;
1.106 cvs 637: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 638: {
1.67 cvs 639: nsuffix[i] = utolower (suffix[i]);
1.13 cvs 640: i++;
641: }
1.106 cvs 642: nsuffix[i] = EOS;
643: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
644: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
645: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.39 cvs 646: return (TRUE);
647: return (FALSE);
1.3 cvs 648: }
649:
1.4 cvs 650: /*----------------------------------------------------------------------
1.58 cvs 651: IsImageType
652: returns TRUE if type points to an image resource.
653: ----------------------------------------------------------------------*/
1.111 cvs 654: ThotBool IsImageType (const char *type)
1.58 cvs 655: {
1.106 cvs 656: char temptype[MAX_LENGTH];
1.58 cvs 657: int i;
658:
659: if (!type)
660: return (FALSE);
661:
1.106 cvs 662: strcpy (temptype, type);
1.58 cvs 663: /* Normalize the type */
664: i = 0;
1.106 cvs 665: while (temptype[i] != EOS)
1.58 cvs 666: {
667: temptype[i] = tolower (temptype[i]);
668: i++;
669: }
1.111 cvs 670: if (!strcmp (temptype, "gif") || !strcmp (temptype, "x-xbitmap") ||
671: !strcmp (temptype, "x-xpixmap") || !strcmp (temptype, "jpeg") ||
672: !strcmp (temptype, "png"))
1.58 cvs 673: return (TRUE);
674: return (FALSE);
675: }
676:
677: /*----------------------------------------------------------------------
1.9 cvs 678: IsTextName
1.4 cvs 679: ----------------------------------------------------------------------*/
1.111 cvs 680: ThotBool IsTextName (const char *path)
1.106 cvs 681: {
682: char temppath[MAX_LENGTH];
683: char suffix[MAX_LENGTH];
684: char nsuffix[MAX_LENGTH];
1.5 cvs 685: int i;
686:
687: if (!path)
1.13 cvs 688: return (FALSE);
1.5 cvs 689:
1.106 cvs 690: strcpy (temppath, path);
1.5 cvs 691: ExtractSuffix (temppath, suffix);
692:
693: /* Normalize the suffix */
694: i = 0;
1.106 cvs 695: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 696: {
1.25 cvs 697: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 698: i++;
699: }
1.106 cvs 700: nsuffix[i] = EOS;
1.5 cvs 701:
1.111 cvs 702: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 703: return (TRUE);
1.106 cvs 704: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 705: {
1.39 cvs 706: /* take into account compressed files */
1.13 cvs 707: ExtractSuffix (temppath, suffix);
708: /* Normalize the suffix */
709: i = 0;
1.106 cvs 710: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 711: {
1.25 cvs 712: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 713: i++;
714: }
1.106 cvs 715: nsuffix[i] = EOS;
1.111 cvs 716: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.13 cvs 717: return (TRUE);
718: else
719: return (FALSE);
720: }
721: else
722: return (FALSE);
1.3 cvs 723: }
724:
1.4 cvs 725: /*----------------------------------------------------------------------
1.9 cvs 726: IsHTTPPath
727: returns TRUE if path is in fact an http URL.
1.4 cvs 728: ----------------------------------------------------------------------*/
1.112 cvs 729: ThotBool IsHTTPPath (const char *path)
1.3 cvs 730: {
1.5 cvs 731: if (!path)
732: return FALSE;
1.3 cvs 733:
1.106 cvs 734: if ((!strncmp (path, "http:", 5) != 0)
735: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
736: || !strncmp (path, "internal:", 9))
1.58 cvs 737: return TRUE;
738: return FALSE;
1.3 cvs 739: }
740:
1.4 cvs 741: /*----------------------------------------------------------------------
1.9 cvs 742: IsWithParameters
743: returns TRUE if url has a concatenated query string.
1.4 cvs 744: ----------------------------------------------------------------------*/
1.66 cvs 745: ThotBool IsWithParameters (const char *url)
1.3 cvs 746: {
1.5 cvs 747: int i;
1.3 cvs 748:
1.9 cvs 749: if ((!url) || (url[0] == EOS))
1.5 cvs 750: return FALSE;
1.3 cvs 751:
1.9 cvs 752: i = strlen (url) - 1;
753: while (i > 0 && url[i--] != '?')
1.5 cvs 754: if (i < 0)
755: return FALSE;
1.3 cvs 756:
1.5 cvs 757: /* There is a parameter */
758: return TRUE;
1.3 cvs 759: }
760:
1.4 cvs 761: /*----------------------------------------------------------------------
1.9 cvs 762: IsW3Path
763: returns TRUE if path is in fact a URL.
1.4 cvs 764: ----------------------------------------------------------------------*/
1.106 cvs 765: ThotBool IsW3Path (const char *path)
766: {
767: if (strncmp (path, "http:", 5) &&
768: strncmp (path, "ftp:", 4) &&
769: strncmp (path, "telnet:", 7) &&
770: strncmp (path, "wais:", 5) &&
771: strncmp (path, "news:", 5) &&
772: strncmp (path, "gopher:", 7) &&
773: strncmp (path, "mailto:", 7) &&
774: strncmp (path, "archie:", 7))
1.72 cvs 775: return FALSE;
776: return TRUE;
1.3 cvs 777: }
778:
1.4 cvs 779: /*----------------------------------------------------------------------
1.90 cvs 780: IsFilePath
781: returns TRUE if path is in fact a URL.
782: ----------------------------------------------------------------------*/
1.106 cvs 783: ThotBool IsFilePath (const char *path)
1.90 cvs 784: {
1.106 cvs 785: if (strncmp (path, "file:", 5))
1.90 cvs 786: return FALSE;
787: return TRUE;
788: }
789:
790: /*----------------------------------------------------------------------
1.9 cvs 791: IsValidProtocol
792: returns true if the url protocol is supported by Amaya.
1.4 cvs 793: ----------------------------------------------------------------------*/
1.106 cvs 794: ThotBool IsValidProtocol (const char *url)
795: {
796: if (!strncmp (url, "http:", 5)
797: || !strncmp (url, "internal:", 9)
798: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 799: /* experimental */
1.24 cvs 800: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 801: return (TRUE);
1.5 cvs 802: else
1.8 cvs 803: return (FALSE);
1.3 cvs 804: }
805:
1.31 cvs 806:
807: /*----------------------------------------------------------------------
808: GetBaseURL
809: normalizes orgName according to a base associated with doc, and
810: following the standard URL format rules.
811: The function returns the base used to solve relative URL and SRC:
812: - the base of the document,
813: - or the document path (without document name).
814: ----------------------------------------------------------------------*/
1.106 cvs 815: char *GetBaseURL (Document doc)
1.31 cvs 816: {
817: Element el;
818: ElementType elType;
819: AttributeType attrType;
820: Attribute attr;
1.106 cvs 821: char *ptr, *basename;
1.31 cvs 822: int length;
823:
1.113 cvs 824: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 825: return NULL;
1.106 cvs 826: basename = TtaGetMemory (MAX_LENGTH);
827: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
828: basename[MAX_LENGTH-1] = EOS;
1.31 cvs 829: length = MAX_LENGTH -1;
1.113 cvs 830: /* is it a HTML document ? */
1.31 cvs 831: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1.113 cvs 832: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
833: /* it's a HTML document */
1.65 cvs 834: {
1.113 cvs 835: /* get the document element */
836: el = TtaGetMainRoot (doc);
837: /* search the BASE element */
838: elType.ElTypeNum = HTML_EL_HEAD;
839: el = TtaSearchTypedElement (elType, SearchForward, el);
840: if (el)
841: /* there is a HEAD element */
842: {
843: /* look for a BASE element within the HEAD */
844: elType.ElTypeNum = HTML_EL_BASE;
845: el = TtaSearchTypedElement (elType, SearchInTree, el);
846: }
847: if (el)
1.31 cvs 848: {
1.113 cvs 849: /* The document has a BASE element. Get the HREF attribute of the
850: BASE element */
851: attrType.AttrSSchema = elType.ElSSchema;
852: attrType.AttrTypeNum = HTML_ATTR_HREF_;
853: attr = TtaGetAttribute (el, attrType);
854: if (attr)
1.31 cvs 855: {
1.113 cvs 856: /* Use the base path of the document */
857: TtaGiveTextAttributeValue (attr, basename, &length);
858: /* base and orgName have to be separated by a DIR_SEP */
859: length--;
860: if (basename[0] != EOS && basename[length] != URL_SEP &&
861: basename[length] != DIR_SEP)
862: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 863: {
1.113 cvs 864: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
865: AMAYA_PARSE_HOST |
866: AMAYA_PARSE_PUNCTUATION);
867: if (ptr && !strcmp (ptr, basename))
868: {
869: /* it has this form, complete it by adding a URL_STR */
870: if (strchr (basename, DIR_SEP))
871: strcat (basename, DIR_STR);
872: else
873: strcat (basename, URL_STR);
874: length++;
875: }
876: if (ptr)
877: TtaFreeMemory (ptr);
1.31 cvs 878: }
879: }
880: }
1.113 cvs 881: }
882:
1.31 cvs 883: /* Remove anything after the last DIR_SEP char. If no such char is found,
884: * then search for the first ":" char, hoping that what's before that is a
885: * protocol. If found, end the string there. If neither char is found,
886: * then discard the whole base element.
887: */
1.106 cvs 888: length = strlen (basename) - 1;
1.31 cvs 889: /* search for the last DIR_SEP char */
1.106 cvs 890: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 891: length--;
892: if (length >= 0)
893: /* found the last DIR_SEP char, end the string there */
1.106 cvs 894: basename[length + 1] = EOS;
1.31 cvs 895: else
896: /* search for the first PATH_STR char */
897: {
1.106 cvs 898: for (length = 0; basename[length] != ':' &&
899: basename[length] != EOS; length ++);
900: if (basename[length] == ':')
1.31 cvs 901: /* found, so end the string there */
1.106 cvs 902: basename[length + 1] = EOS;
1.31 cvs 903: else
904: /* not found, discard the base */
1.106 cvs 905: basename[0] = EOS;
1.31 cvs 906: }
907: return (basename);
908: }
909:
910:
1.4 cvs 911: /*----------------------------------------------------------------------
1.40 cvs 912: GetLocalPath
913: Allocate and return the local document path associated to the url
914: ----------------------------------------------------------------------*/
1.106 cvs 915: char *GetLocalPath (Document doc, char *url)
916: {
917: char *ptr;
918: char *n;
919: char *documentname;
920: char url_sep;
1.83 cvs 921: int len;
1.67 cvs 922: ThotBool noFile;
1.40 cvs 923:
924: if (url != NULL)
925: {
926: /* check whether the file name exists */
1.106 cvs 927: len = strlen (url) - 1;
1.71 cvs 928: if (IsW3Path (url))
1.106 cvs 929: url_sep = '/';
1.41 cvs 930: else
1.106 cvs 931: url_sep = DIR_SEP;
1.41 cvs 932: noFile = (url[len] == url_sep);
1.40 cvs 933: if (noFile)
1.106 cvs 934: url[len] = EOS;
935: ptr = TtaGetMemory (MAX_LENGTH);
936: documentname = TtaGetMemory (MAX_LENGTH);
1.78 cvs 937: TtaExtractName (url, ptr, documentname);
1.106 cvs 938: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 939: if (!TtaCheckDirectory (ptr))
940: /* directory did not exist */
1.72 cvs 941: TtaMakeDirectory (ptr);
1.47 cvs 942:
943: /* don't include the query string within document name */
1.106 cvs 944: n = strrchr (documentname, '?');
1.47 cvs 945: if (n != NULL)
1.106 cvs 946: *n = EOS;
1.46 cvs 947: /* don't include ':' within document name */
1.106 cvs 948: n = strchr (documentname, ':');
1.46 cvs 949: if (n != NULL)
1.106 cvs 950: *n = EOS;
1.69 cvs 951: /* if after all this operations document name
952: is empty, let's use noname.html instead */
1.106 cvs 953: if (documentname[0] == EOS)
954: strcat (ptr, "noname.html");
1.69 cvs 955: else
1.106 cvs 956: strcat (ptr, documentname);
1.40 cvs 957: TtaFreeMemory (documentname);
958: /* restore the url */
959: if (noFile)
1.41 cvs 960: url[len] = url_sep;
1.40 cvs 961: return (ptr);
962: }
963: else
964: return (NULL);
965: }
966:
1.73 cvs 967: /*----------------------------------------------------------------------
1.79 cvs 968: ExtractTarget extract the target name from document nane.
969: ----------------------------------------------------------------------*/
1.106 cvs 970: void ExtractTarget (char *aName, char *target)
1.79 cvs 971: {
1.106 cvs 972: int lg, i;
973: char *ptr;
974: char *oldptr;
1.79 cvs 975:
976: if (!target || !aName)
977: /* bad target */
978: return;
979:
1.106 cvs 980: target[0] = EOS;
981: lg = strlen (aName);
1.79 cvs 982: if (lg)
983: {
984: /* the name is not empty */
985: oldptr = ptr = &aName[0];
986: do
987: {
1.106 cvs 988: ptr = strrchr (oldptr, '#');
1.79 cvs 989: if (ptr)
990: oldptr = &ptr[1];
991: }
992: while (ptr);
993:
994: i = (int) (oldptr) - (int) (aName); /* name length */
995: if (i > 1)
996: {
1.106 cvs 997: aName[i - 1] = EOS;
1.79 cvs 998: if (i != lg)
1.106 cvs 999: strcpy (target, oldptr);
1.79 cvs 1000: }
1001: }
1002: }
1003:
1004: /*----------------------------------------------------------------------
1.90 cvs 1005: RemoveNewLines (text)
1006: Removes any '\n' chars that are found in text.
1007: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1008: ----------------------------------------------------------------------*/
1.106 cvs 1009: ThotBool RemoveNewLines (char *text)
1010: {
1011: ThotBool change = FALSE;
1012: char *src;
1013: char *dest;
1.90 cvs 1014:
1015: src = text;
1016: dest = text;
1.115 kahan 1017:
1018: /* remove any preceding whitespace */
1019: while (*src && *src == ' ')
1020: {
1021: src++;
1022: change = 1;
1023: }
1024:
1.90 cvs 1025: while (*src)
1026: {
1027: switch (*src)
1028: {
1.106 cvs 1029: case '\n':
1.90 cvs 1030: /* don't copy the newline */
1031: change = 1;
1032: break;
1033: default:
1034: *dest = *src;
1035: dest++;
1036: break;
1037: }
1038: src++;
1039: }
1040: /* copy the last EOS char */
1041: *dest = *src;
1042:
1043: return (change);
1044: }
1045:
1046: /*----------------------------------------------------------------------
1047: CleanCopyFileURL
1048: Copies a file url from a src string to destination string.
1.97 cvs 1049: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1050: we want to do).
1.90 cvs 1051: ----------------------------------------------------------------------*/
1.106 cvs 1052: static void CleanCopyFileURL (char *dest, char *src,
1053: ConvertionType convertion)
1.90 cvs 1054: {
1055: while (*src)
1.89 cvs 1056: {
1.90 cvs 1057: switch (*src)
1.89 cvs 1058: {
1059: #ifdef _WINDOWS
1.106 cvs 1060: case URL_SEP:
1.96 cvs 1061: /* make DIR_SEP transformation */
1.97 cvs 1062: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1063: *dest = DIR_SEP;
1.96 cvs 1064: else
1065: *dest = *src;
1.90 cvs 1066: dest++;
1.96 cvs 1067: src++;
1.90 cvs 1068: break;
1.89 cvs 1069: #endif /* _WINDOWS */
1.96 cvs 1070:
1.106 cvs 1071: case '%':
1.97 cvs 1072: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1073: {
1.97 cvs 1074: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1075: src++;
1.106 cvs 1076: if (*src != EOS)
1.97 cvs 1077: {
1078: *dest = UnEscapeChar (*src) * 16;
1079: src++;
1080: }
1.106 cvs 1081: if (*src != EOS)
1.97 cvs 1082: {
1083: *dest = *dest + UnEscapeChar (*src);
1084: src++;
1085: }
1086: dest++;
1.96 cvs 1087: }
1.97 cvs 1088: else
1.96 cvs 1089: {
1.97 cvs 1090: *dest = *src;
1091: dest++;
1.96 cvs 1092: src++;
1093: }
1094: break;
1095:
1.90 cvs 1096: default:
1097: *dest = *src;
1.89 cvs 1098: dest++;
1.96 cvs 1099: src++;
1.90 cvs 1100: break;
1.89 cvs 1101: }
1102: }
1.90 cvs 1103: /* copy the EOS char */
1104: *dest = *src;
1.73 cvs 1105: }
1.40 cvs 1106:
1107: /*----------------------------------------------------------------------
1.9 cvs 1108: NormalizeURL
1109: normalizes orgName according to a base associated with doc, and
1110: following the standard URL format rules.
1.113 cvs 1111: if doc is < 0, use as a base the URL of the document that contains
1112: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1113: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1114: other path.
1.9 cvs 1115: The function returns the new complete and normalized URL
1.12 cvs 1116: or file name path (newName) and the name of the document (docName).
1.9 cvs 1117: N.B. If the function can't find out what's the docName, it assigns
1118: the name "noname.html".
1.4 cvs 1119: ----------------------------------------------------------------------*/
1.106 cvs 1120: void NormalizeURL (char *orgName, Document doc, char *newName,
1121: char *docName, char *otherPath)
1122: {
1123: char *basename;
1124: char tempOrgName[MAX_LENGTH];
1125: char *ptr;
1126: char used_sep;
1.84 cvs 1127: int length;
1128: ThotBool check;
1.5 cvs 1129:
1.110 cvs 1130: #ifdef _WINDOWS
1.44 cvs 1131: int ndx;
1.110 cvs 1132: #endif /* _WINDOWS */
1.44 cvs 1133:
1.5 cvs 1134: if (!newName || !docName)
1135: return;
1.18 cvs 1136:
1.113 cvs 1137: if (doc < 0)
1138: basename = TtaStrdup (SavedDocumentURL);
1139: else if (doc > 0)
1.53 cvs 1140: basename = GetBaseURL (doc);
1141: else if (otherPath != NULL)
1.108 cvs 1142: basename = TtaStrdup (otherPath);
1.32 cvs 1143: else
1.53 cvs 1144: basename = NULL;
1.32 cvs 1145:
1.18 cvs 1146: /*
1.31 cvs 1147: * Clean orgName
1148: * Make sure we have a complete orgName, without any leading or trailing
1149: * white spaces, or trailinbg new lines
1150: */
1.5 cvs 1151: ptr = orgName;
1.18 cvs 1152: /* skip leading white space and new line characters */
1.106 cvs 1153: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1154: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1155: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1156: /*
1.31 cvs 1157: * Make orgName a complete URL
1158: * If the URL does not include a protocol, then try to calculate
1159: * one using the doc's base element (if it exists),
1160: */
1.106 cvs 1161: if (tempOrgName[0] == EOS)
1.53 cvs 1162: {
1.106 cvs 1163: newName[0] = EOS;
1164: docName[0] = EOS;
1.53 cvs 1165: TtaFreeMemory (basename);
1166: return;
1167: }
1.49 cvs 1168:
1169: /* clean trailing white space */
1.106 cvs 1170: length = strlen (tempOrgName) - 1;
1171: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1172: {
1.106 cvs 1173: tempOrgName[length] = EOS;
1.53 cvs 1174: length--;
1175: }
1.50 cvs 1176:
1.55 cvs 1177: /* remove extra dot (which dot???) */
1178: /* ugly, but faster than a strcmp */
1.106 cvs 1179: if (tempOrgName[length] == '.'
1180: && (length == 0 || tempOrgName[length-1] != '.'))
1181: tempOrgName[length] = EOS;
1.50 cvs 1182:
1.94 cvs 1183: if (IsW3Path (tempOrgName))
1.53 cvs 1184: {
1185: /* the name is complete, go to the Sixth Step */
1.106 cvs 1186: strcpy (newName, tempOrgName);
1.53 cvs 1187: SimplifyUrl (&newName);
1188: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1189: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1190: AMAYA_PARSE_HOST |
1191: AMAYA_PARSE_PUNCTUATION);
1192: if (ptr && !strcmp (ptr, newName))
1193: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1194: strcat (newName, URL_STR);
1.49 cvs 1195:
1.53 cvs 1196: if (ptr)
1.50 cvs 1197: TtaFreeMemory (ptr);
1.53 cvs 1198: }
1.113 cvs 1199: else if (basename == NULL)
1.53 cvs 1200: /* the name is complete, go to the Sixth Step */
1.106 cvs 1201: strcpy (newName, tempOrgName);
1.53 cvs 1202: else
1203: {
1.31 cvs 1204: /* Calculate the absolute URL, using the base or document URL */
1.110 cvs 1205: #ifdef _WINDOWS
1.53 cvs 1206: if (!IsW3Path (basename))
1207: {
1.106 cvs 1208: length = strlen (tempOrgName);
1.53 cvs 1209: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1210: if (tempOrgName [ndx] == '/')
1211: tempOrgName [ndx] = '\\';
1.53 cvs 1212: }
1.110 cvs 1213: #endif /* _WINDOWS */
1.25 cvs 1214: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1215: if (ptr)
1216: {
1217: SimplifyUrl (&ptr);
1.106 cvs 1218: strcpy (newName, ptr);
1.53 cvs 1219: TtaFreeMemory (ptr);
1220: }
1221: else
1.106 cvs 1222: newName[0] = EOS;
1.53 cvs 1223: }
1.36 cvs 1224:
1225: TtaFreeMemory (basename);
1.18 cvs 1226: /*
1.31 cvs 1227: * Prepare the docname that will refer to this ressource in the
1228: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1229: * noname.html as a default ressource name
1.18 cvs 1230: */
1.106 cvs 1231: if (newName[0] != EOS)
1.53 cvs 1232: {
1.106 cvs 1233: length = strlen (newName) - 1;
1234: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1235: {
1236: used_sep = newName[length];
1237: check = TRUE;
1238: while (check)
1239: {
1.50 cvs 1240: length--;
1241: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1242: length--;
1.106 cvs 1243: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1244: {
1.106 cvs 1245: newName[length+1] = EOS;
1.53 cvs 1246: /* remove also previous directory */
1247: length--;
1248: while (length >= 0 && newName[length] != used_sep)
1249: length--;
1.106 cvs 1250: if (strncmp (&newName[length+1], "//", 2))
1.53 cvs 1251: /* don't remove server name */
1.106 cvs 1252: newName[length+1] = EOS;
1.53 cvs 1253: }
1.106 cvs 1254: else if (!strncmp (&newName[length+1], ".", 1))
1255: newName[length+1] = EOS;
1.50 cvs 1256: else
1.53 cvs 1257: check = FALSE;
1258: }
1259: /* docname was not comprised inside the URL, so let's */
1260: /* assign the default ressource name */
1.106 cvs 1261: strcpy (docName, "noname.html");
1.53 cvs 1262: }
1263: else
1264: { /* docname is comprised inside the URL */
1.110 cvs 1265: while (length >= 0 && newName[length] != URL_SEP &&
1266: newName[length] != DIR_SEP)
1.53 cvs 1267: length--;
1268: if (length < 0)
1.106 cvs 1269: strcpy (docName, newName);
1.53 cvs 1270: else
1.106 cvs 1271: strcpy (docName, &newName[length+1]);
1.53 cvs 1272: }
1273: }
1274: else
1.106 cvs 1275: docName[0] = EOS;
1.18 cvs 1276: }
1.3 cvs 1277:
1.4 cvs 1278: /*----------------------------------------------------------------------
1.9 cvs 1279: IsSameHost
1.4 cvs 1280: ----------------------------------------------------------------------*/
1.106 cvs 1281: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1282: {
1.106 cvs 1283: char *basename_ptr1, *basename_ptr2;
1284: ThotBool result;
1.3 cvs 1285:
1.106 cvs 1286: basename_ptr1 = AmayaParseUrl (url1, "",
1287: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1288: basename_ptr2 = AmayaParseUrl (url2, "",
1289: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1290:
1.106 cvs 1291: if (strcmp (basename_ptr1, basename_ptr2))
1292: result = FALSE;
1293: else
1294: result = TRUE;
1295: TtaFreeMemory (basename_ptr1);
1296: TtaFreeMemory (basename_ptr2);
1297: return (result);
1.3 cvs 1298: }
1299:
1300:
1.4 cvs 1301: /*----------------------------------------------------------------------
1.22 cvs 1302: HasKnownFileSuffix
1303: returns TRUE if path points to a file ending with a suffix.
1304: ----------------------------------------------------------------------*/
1.106 cvs 1305: ThotBool HasKnownFileSuffix (const char *path)
1306: {
1307: char *root;
1308: char temppath[MAX_LENGTH];
1309: char suffix[MAX_LENGTH];
1.22 cvs 1310:
1.106 cvs 1311: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1312: return (FALSE);
1313:
1.106 cvs 1314: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1315:
1316: if (root)
1317: {
1.106 cvs 1318: strcpy (temppath, root);
1.25 cvs 1319: TtaFreeMemory (root);
1.22 cvs 1320: /* Get the suffix */
1321: ExtractSuffix (temppath, suffix);
1322:
1.106 cvs 1323: if( suffix[0] == EOS)
1.22 cvs 1324: /* no suffix */
1325: return (FALSE);
1326:
1327: /* Normalize the suffix */
1328: ConvertToLowerCase (suffix);
1329:
1.106 cvs 1330: if (!strcmp (suffix, "gz"))
1.22 cvs 1331: /* skip the compressed suffix */
1332: {
1333: ExtractSuffix (temppath, suffix);
1.106 cvs 1334: if(suffix[0] == EOS)
1.22 cvs 1335: /* no suffix */
1336: return (FALSE);
1337: /* Normalize the suffix */
1338: ConvertToLowerCase (suffix);
1339: }
1340:
1.106 cvs 1341: if (strcmp (suffix, "gif") &&
1342: strcmp (suffix, "xbm") &&
1343: strcmp (suffix, "xpm") &&
1344: strcmp (suffix, "jpg") &&
1345: strcmp (suffix, "pdf") &&
1346: strcmp (suffix, "png") &&
1347: strcmp (suffix, "tgz") &&
1348: strcmp (suffix, "xpg") &&
1349: strcmp (suffix, "xpd") &&
1350: strcmp (suffix, "ps") &&
1351: strcmp (suffix, "au") &&
1352: strcmp (suffix, "html") &&
1353: strcmp (suffix, "htm") &&
1354: strcmp (suffix, "shtml") &&
1355: strcmp (suffix, "xht") &&
1356: strcmp (suffix, "xhtm") &&
1357: strcmp (suffix, "xhtml") &&
1358: strcmp (suffix, "txt") &&
1359: strcmp (suffix, "css") &&
1360: strcmp (suffix, "eps"))
1.22 cvs 1361: return (FALSE);
1362: else
1363: return (TRUE);
1364: }
1365: else
1366: return (FALSE);
1367: }
1368:
1369:
1370: /*----------------------------------------------------------------------
1.24 cvs 1371: ChopURL
1372: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1373: If inputURL is bigger than that size, outputURL receives
1374: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1375: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1376: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1377: copied into outputURL.
1378: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1379: chars.
1380: ----------------------------------------------------------------------*/
1.106 cvs 1381: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1382: {
1383: int len;
1.9 cvs 1384:
1.106 cvs 1385: len = strlen (inputURL);
1.24 cvs 1386: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1387: strcpy (outputURL, inputURL);
1.24 cvs 1388: else
1389: /* make a truncated urlName on the status window */
1390: {
1.106 cvs 1391: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1392: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1393: strcat (outputURL, "...");
1394: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1395: }
1.25 cvs 1396: }
1397:
1398:
1399: /*----------------------------------------------------------------------
1400: scan
1.47 cvs 1401: Scan a filename for its constituents
1.25 cvs 1402: -----------------------------------
1403:
1404: On entry,
1405: name points to a document name which may be incomplete.
1406: On exit,
1407: absolute or relative may be nonzero (but not both).
1408: host, fragment and access may be nonzero if they were specified.
1409: Any which are nonzero point to zero terminated strings.
1410: ----------------------------------------------------------------------*/
1.106 cvs 1411: static void scan (char *name, HTURI *parts)
1.25 cvs 1412: {
1.106 cvs 1413: char * p;
1414: char * after_access = name;
1.32 cvs 1415:
1.43 cvs 1416: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1417: /* Look for fragment identifier */
1.106 cvs 1418: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1419: {
1.106 cvs 1420: *p++ = '\0';
1.28 cvs 1421: parts->fragment = p;
1.25 cvs 1422: }
1423:
1.28 cvs 1424: for (p=name; *p; p++)
1425: {
1.106 cvs 1426: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1427: break;
1.106 cvs 1428: if (*p == ':')
1.28 cvs 1429: {
1430: *p = 0;
1431: parts->access = after_access; /* Scheme has been specified */
1432:
1433: /* The combination of gcc, the "-O" flag and the HP platform is
1434: unhealthy. The following three lines is a quick & dirty fix, but is
1435: not recommended. Rather, turn off "-O". */
1436:
1437: /* after_access = p;*/
1438: /* while (*after_access == 0)*/
1439: /* after_access++;*/
1440: after_access = p+1;
1.106 cvs 1441: if (!strcasecmp("URL", parts->access))
1.28 cvs 1442: /* Ignore IETF's URL: pre-prefix */
1443: parts->access = NULL;
1444: else
1.25 cvs 1445: break;
1446: }
1447: }
1448:
1449: p = after_access;
1.43 cvs 1450: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1451: {
1.43 cvs 1452: if (p[1] == URL_SEP)
1.28 cvs 1453: {
1.25 cvs 1454: parts->host = p+2; /* host has been specified */
1.28 cvs 1455: *p = 0; /* Terminate access */
1456: /* look for end of host name if any */
1.106 cvs 1457: p = strchr (parts->host, URL_SEP);
1.28 cvs 1458: if (p)
1459: {
1.106 cvs 1460: *p = EOS; /* Terminate host */
1.25 cvs 1461: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1462: }
1463: }
1464: else
1465: /* Root found but no host */
1466: parts->absolute = p+1;
1467: }
1468: else
1469: {
1.25 cvs 1470: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1471: }
1.25 cvs 1472: }
1473:
1474:
1475: /*----------------------------------------------------------------------
1.28 cvs 1476: AmayaParseUrl: parse a Name relative to another name
1477:
1478: This returns those parts of a name which are given (and requested)
1479: substituting bits from the related name where necessary.
1.25 cvs 1480:
1.28 cvs 1481: On entry,
1.25 cvs 1482: aName A filename given
1483: relatedName A name relative to which aName is to be parsed. Give
1484: it an empty string if aName is absolute.
1485: wanted A mask for the bits which are wanted.
1486:
1.28 cvs 1487: On exit,
1.25 cvs 1488: returns A pointer to a malloc'd string which MUST BE FREED
1489: ----------------------------------------------------------------------*/
1.106 cvs 1490: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1491: {
1492: char *return_value;
1493: char result[MAX_LENGTH];
1494: char name[MAX_LENGTH];
1495: char rel[MAX_LENGTH];
1496: char *p, *access;
1.29 cvs 1497: HTURI given, related;
1498: int len;
1.106 cvs 1499: char used_sep;
1500: char *used_str;
1.32 cvs 1501:
1.106 cvs 1502: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1503: {
1.106 cvs 1504: used_str = DIR_STR;
1505: used_sep = DIR_SEP;
1.33 cvs 1506: }
1.32 cvs 1507: else
1.33 cvs 1508: {
1.106 cvs 1509: used_str = URL_STR;
1510: used_sep = URL_SEP;
1.33 cvs 1511: }
1.32 cvs 1512:
1.29 cvs 1513: /* Make working copies of input strings to cut up: */
1514: return_value = NULL;
1515: result[0] = 0; /* Clear string */
1.106 cvs 1516: strcpy (name, aName);
1.29 cvs 1517: if (relatedName != NULL)
1.106 cvs 1518: strcpy (rel, relatedName);
1.29 cvs 1519: else
1.106 cvs 1520: relatedName[0] = EOS;
1.29 cvs 1521:
1522: scan (name, &given);
1523: scan (rel, &related);
1524: access = given.access ? given.access : related.access;
1525: if (wanted & AMAYA_PARSE_ACCESS)
1526: if (access)
1527: {
1.106 cvs 1528: strcat (result, access);
1.29 cvs 1529: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1530: strcat (result, ":");
1.29 cvs 1531: }
1532:
1533: if (given.access && related.access)
1534: /* If different, inherit nothing. */
1.106 cvs 1535: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1536: {
1537: related.host = 0;
1538: related.absolute = 0;
1539: related.relative = 0;
1540: related.fragment = 0;
1541: }
1542:
1543: if (wanted & AMAYA_PARSE_HOST)
1544: if(given.host || related.host)
1545: {
1546: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1547: strcat (result, "//");
1548: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1549: }
1550:
1551: if (given.host && related.host)
1552: /* If different hosts, inherit no path. */
1.106 cvs 1553: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1554: {
1555: related.absolute = 0;
1556: related.relative = 0;
1557: related.fragment = 0;
1558: }
1559:
1560: if (wanted & AMAYA_PARSE_PATH)
1561: {
1562: if (given.absolute)
1563: {
1564: /* All is given */
1565: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1566: strcat (result, used_str);
1567: strcat (result, given.absolute);
1.25 cvs 1568: }
1.29 cvs 1569: else if (related.absolute)
1570: {
1571: /* Adopt path not name */
1.106 cvs 1572: strcat (result, used_str);
1573: strcat (result, related.absolute);
1.29 cvs 1574: if (given.relative)
1575: {
1576: /* Search part? */
1.106 cvs 1577: p = strchr (result, '?');
1.29 cvs 1578: if (!p)
1.106 cvs 1579: p=result+strlen(result)-1;
1.33 cvs 1580: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1581: /* Remove filename */
1582: p[1]=0;
1583: /* Add given one */
1.106 cvs 1584: strcat (result, given.relative);
1.25 cvs 1585: }
1586: }
1.29 cvs 1587: else if (given.relative)
1588: /* what we've got */
1.106 cvs 1589: strcat (result, given.relative);
1.29 cvs 1590: else if (related.relative)
1.106 cvs 1591: strcat (result, related.relative);
1.29 cvs 1592: else
1593: /* No inheritance */
1.106 cvs 1594: strcat (result, used_str);
1.25 cvs 1595: }
1.29 cvs 1596:
1597: if (wanted & AMAYA_PARSE_ANCHOR)
1598: if (given.fragment || related.fragment)
1599: {
1600: if (given.absolute && given.fragment)
1601: {
1602: /*Fixes for relURLs...*/
1603: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1604: strcat (result, "#");
1605: strcat (result, given.fragment);
1.29 cvs 1606: }
1607: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1608: strcat (result, "");
1.29 cvs 1609: else
1610: {
1.110 cvs 1611: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1612: strcat (result, "#");
1.110 cvs 1613: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1614: }
1615: }
1.106 cvs 1616: len = strlen (result);
1617: if ((return_value = TtaGetMemory (len + 1)) != NULL)
1618: strcpy (return_value, result);
1.29 cvs 1619: return (return_value); /* exactly the right length */
1.25 cvs 1620: }
1621:
1622: /*----------------------------------------------------------------------
1623: HTCanon
1624: Canonicalizes the URL in the following manner starting from the host
1625: pointer:
1626:
1627: 1) The host name is converted to lowercase
1628: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1629:
1630: Return: OK The position of the current path part of the URL
1631: which might be the old one or a new one.
1632:
1633: ----------------------------------------------------------------------*/
1.106 cvs 1634: static char *HTCanon (char **filename, char *host)
1635: {
1636: char *newname = NULL;
1637: char used_sep;
1638: char *path;
1639: char *strptr;
1640: char *port;
1641: char *access = host-3;
1642:
1643: if (*filename && strchr (*filename, URL_SEP))
1644: used_sep = URL_SEP;
1.33 cvs 1645: else
1.106 cvs 1646: used_sep = DIR_SEP;
1.32 cvs 1647:
1.110 cvs 1648: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1649: access--;
1.110 cvs 1650: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1651: path = host + strlen (host);
1652: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1653: host = strptr;
1.110 cvs 1654: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1655: port = NULL;
1.25 cvs 1656:
1657: strptr = host; /* Convert to lower-case */
1.82 cvs 1658: while (strptr < path)
1.33 cvs 1659: {
1.84 cvs 1660: *strptr = utolower (*strptr);
1.82 cvs 1661: strptr++;
1.33 cvs 1662: }
1.25 cvs 1663:
1664: /* Does the URL contain a full domain name? This also works for a
1665: numerical host name. The domain name is already made lower-case
1666: and without a trailing dot. */
1667: {
1.106 cvs 1668: char *dot = port ? port : path;
1669: if (dot > *filename && *--dot == '.')
1.33 cvs 1670: {
1.106 cvs 1671: char *orig = dot;
1672: char *dest = dot + 1;
1.82 cvs 1673: while ((*orig++ = *dest++));
1674: if (port) port--;
1.33 cvs 1675: path--;
1.25 cvs 1676: }
1677: }
1678: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1679: if (port)
1680: {
1.82 cvs 1681: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1682: {
1683: if (!newname)
1684: {
1.106 cvs 1685: char *orig = port;
1686: char *dest = port + 1;
1.82 cvs 1687: while ((*orig++ = *dest++));
1.33 cvs 1688: }
1689: }
1.106 cvs 1690: else if ((!strncmp (access, "http", 4) &&
1691: (*(port + 1) == '8' &&
1692: *(port+2) == '0' &&
1.82 cvs 1693: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 1694: (!strncmp (access, "gopher", 6) &&
1695: (*(port+1) == '7' &&
1696: *(port+2) == '0' &&
1.82 cvs 1697: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 1698: (!strncmp (access, "ftp", 3) &&
1699: (*(port+1) == '2' &&
1700: *(port + 2) == '1' &&
1.82 cvs 1701: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 1702: if (!newname)
1703: {
1.106 cvs 1704: char *orig = port;
1705: char *dest = port + 3;
1.33 cvs 1706: while((*orig++ = *dest++));
1707: /* Update path position, Henry Minsky */
1708: path -= 3;
1.25 cvs 1709: }
1.33 cvs 1710: }
1711: else if (newname)
1.106 cvs 1712: strncat (newname, port, (int) (path - port));
1.33 cvs 1713: }
1.25 cvs 1714:
1.33 cvs 1715: if (newname)
1716: {
1.106 cvs 1717: char *newpath = newname + strlen (newname);
1718: strcat (newname, path);
1.25 cvs 1719: path = newpath;
1.28 cvs 1720: /* Free old copy */
1721: TtaFreeMemory(*filename);
1.25 cvs 1722: *filename = newname;
1.33 cvs 1723: }
1.25 cvs 1724: return path;
1725: }
1726:
1727:
1728: /*----------------------------------------------------------------------
1.29 cvs 1729: SimplifyUrl: simplify a URI
1.32 cvs 1730: A URI is allowed to contain the sequence xxx/../ which may be
1731: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 1732: Simplification helps us recognize duplicate URIs.
1.25 cvs 1733:
1.28 cvs 1734: Thus, /etc/junk/../fred becomes /etc/fred
1735: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 1736:
1.28 cvs 1737: but we should NOT change
1738: http://fred.xxx.edu/../..
1.25 cvs 1739:
1740: or ../../albert.html
1741:
1.28 cvs 1742: In order to avoid empty URLs the following URLs become:
1.25 cvs 1743:
1744: /fred/.. becomes /fred/..
1745: /fred/././.. becomes /fred/..
1746: /fred/.././junk/.././ becomes /fred/..
1747:
1.28 cvs 1748: If more than one set of `://' is found (several proxies in cascade) then
1749: only the part after the last `://' is simplified.
1.25 cvs 1750:
1.28 cvs 1751: Returns: A string which might be the old one or a new one.
1.25 cvs 1752: ----------------------------------------------------------------------*/
1.106 cvs 1753: void SimplifyUrl (char **url)
1754: {
1755: char *path;
1756: char *access;
1757: char *newptr;
1758: char *p;
1759: char *orig, *dest, *end;
1.28 cvs 1760:
1.106 cvs 1761: char used_sep;
1.77 cvs 1762: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
1763: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 1764:
1765:
1.28 cvs 1766: if (!url || !*url)
1767: return;
1768:
1.106 cvs 1769: if (strchr (*url, URL_SEP))
1770: used_sep = URL_SEP;
1.32 cvs 1771: else
1.106 cvs 1772: used_sep = DIR_SEP;
1.32 cvs 1773:
1.77 cvs 1774: /* should we simplify double dot? */
1775: path = *url;
1.106 cvs 1776: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 1777: ddot_simplify = FALSE;
1778: else
1779: ddot_simplify = TRUE;
1780:
1.28 cvs 1781: /* Find any scheme name */
1.106 cvs 1782: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 1783: {
1784: /* Find host name */
1.28 cvs 1785: access = *url;
1.84 cvs 1786: while (access < path && (*access = utolower (*access)))
1.82 cvs 1787: access++;
1.28 cvs 1788: path += 3;
1.106 cvs 1789: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 1790: /* For proxies */
1.106 cvs 1791: path = newptr + 3;
1.82 cvs 1792: /* We have a host name */
1.84 cvs 1793: path = HTCanon (url, path);
1.25 cvs 1794: }
1.106 cvs 1795: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 1796: path += 2;
1797: else
1798: path = *url;
1.25 cvs 1799:
1.84 cvs 1800: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 1801: /* Some URLs start //<foo> */
1802: path += 1;
1.94 cvs 1803: else if (IsFilePath (path))
1804: {
1805: /* doesn't need to do anything more */
1806: return;
1807: }
1.106 cvs 1808: else if (!strncmp (path, "news:", 5))
1.28 cvs 1809: {
1.106 cvs 1810: newptr = strchr (path+5, '@');
1.28 cvs 1811: if (!newptr)
1812: newptr = path + 5;
1813: while (*newptr)
1814: {
1815: /* Make group or host lower case */
1.84 cvs 1816: *newptr = utolower (*newptr);
1.28 cvs 1817: newptr++;
1.25 cvs 1818: }
1.28 cvs 1819: /* Doesn't need to do any more */
1820: return;
1.25 cvs 1821: }
1.28 cvs 1822:
1823: if ((p = path))
1824: {
1.106 cvs 1825: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
1826: (end = strchr (path, '#'))))
1827: end = path + strlen (path);
1.28 cvs 1828:
1829: /* Parse string second time to simplify */
1830: p = path;
1831: while (p < end)
1832: {
1.110 cvs 1833: /* if we're pointing to a char, it's safe to reactivate the
1834: ../ convertion */
1.106 cvs 1835: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 1836: ddot_simplify = TRUE;
1837:
1.33 cvs 1838: if (*p==used_sep)
1.28 cvs 1839: {
1.106 cvs 1840: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 1841: {
1842: orig = p + 1;
1.84 cvs 1843: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 1844: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 1845: end = orig - 1;
1846: }
1.106 cvs 1847: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 1848: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 1849: {
1850: newptr = p;
1.52 cvs 1851: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
1852: if (*newptr == used_sep)
1853: orig = newptr + 1;
1.28 cvs 1854: else
1.52 cvs 1855: orig = newptr;
1856:
1857: dest = (*(p+3) != used_sep) ? p+3 : p+4;
1858: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
1859: end = orig-1;
1860: /* Start again with prev slash */
1861: p = newptr;
1.28 cvs 1862: }
1.33 cvs 1863: else if (*(p+1) == used_sep)
1.28 cvs 1864: {
1.33 cvs 1865: while (*(p+1) == used_sep)
1.28 cvs 1866: {
1867: orig = p;
1868: dest = p + 1;
1869: while ((*orig++ = *dest++)); /* Remove multiple /'s */
1870: end = orig-1;
1871: }
1872: }
1873: else
1.25 cvs 1874: p++;
1.28 cvs 1875: }
1876: else
1877: p++;
1.25 cvs 1878: }
1879: }
1.51 cvs 1880:
1881: /*
1882: ** Check for host/../.. kind of things
1883: */
1.106 cvs 1884: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 1885: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 1886: *(path+1) = EOS;
1.51 cvs 1887:
1.28 cvs 1888: return;
1889: }
1890:
1891:
1892: /*----------------------------------------------------------------------
1.96 cvs 1893: NormalizeFile normalizes local names.
1.28 cvs 1894: Return TRUE if target and src differ.
1895: ----------------------------------------------------------------------*/
1.106 cvs 1896: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 1897: {
1.110 cvs 1898: #ifndef _WINDOWS
1.106 cvs 1899: char *s;
1.93 cvs 1900: int i;
1.110 cvs 1901: #endif /* !_WINDOWS */
1.82 cvs 1902: ThotBool change;
1.90 cvs 1903: int start_index; /* the first char that we'll copy */
1.28 cvs 1904:
1.54 cvs 1905: change = FALSE;
1.90 cvs 1906: start_index = 0;
1907:
1.106 cvs 1908: if (!src || src[0] == EOS)
1.96 cvs 1909: {
1.106 cvs 1910: target[0] = EOS;
1.96 cvs 1911: return FALSE;
1912: }
1.90 cvs 1913:
1914: /* @@ do I need file: or file:/ here? */
1.106 cvs 1915: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 1916: {
1.90 cvs 1917: /* remove the prefix file: */
1918: start_index += 5;
1919:
1920: /* remove the localhost prefix */
1.106 cvs 1921: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 1922: start_index += 11;
1923:
1924: /* remove the first two slashes in / / /path */
1925: while (src[start_index] &&
1.106 cvs 1926: src[start_index] == '/'
1927: && src[start_index + 1] == '/')
1.94 cvs 1928: start_index++;
1929:
1930: #ifdef _WINDOWS
1931: /* remove any extra slash before the drive name */
1.106 cvs 1932: if (src[start_index] == '/'
1933: &&src[start_index+2] == ':')
1.94 cvs 1934: start_index++;
1935: #endif /* _WINDOWS */
1.90 cvs 1936:
1.106 cvs 1937: if (src[start_index] == EOS)
1.90 cvs 1938: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 1939: strcpy (target, DIR_STR);
1.90 cvs 1940: else
1.97 cvs 1941: /* as we're inside a file: URL, we'll apply all the convertions
1942: we know */
1943: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 1944:
1945: change = TRUE;
1946: }
1.97 cvs 1947: else if (convertion != AM_CONV_NONE)
1.96 cvs 1948: {
1949: /* we are following a "local" relative link, we do all the
1950: convertions except for the HOME_DIR ~ one */
1.97 cvs 1951: CleanCopyFileURL (target, src, convertion);
1.28 cvs 1952: }
1.90 cvs 1953: #ifndef _WINDOWS
1.106 cvs 1954: else if (src[0] == '~')
1.53 cvs 1955: {
1.96 cvs 1956: /* it must be a URL typed in a text input field */
1957: /* do the HOME_DIR ~ substitution */
1.82 cvs 1958: s = TtaGetEnvString ("HOME");
1.106 cvs 1959: strcpy (target, s);
1.90 cvs 1960: #if 0
1.96 cvs 1961: /* JK: invalidated this part of the code as it's simpler
1962: to add the DIR_SEP whenever we have something to add
1963: to the path rather than adding it systematically */
1.106 cvs 1964: if (src[1] != DIR_SEP)
1965: strcat (target, DIR_STR);
1.90 cvs 1966: #endif
1.106 cvs 1967: i = strlen (target);
1968: strcpy (&target[i], &src[1]);
1.54 cvs 1969: change = TRUE;
1.53 cvs 1970: }
1.90 cvs 1971: #endif /* _WINDOWS */
1.28 cvs 1972: else
1.96 cvs 1973: /* leave it as it is */
1.106 cvs 1974: strcpy (target, src);
1.96 cvs 1975:
1.28 cvs 1976: /* remove /../ and /./ */
1.29 cvs 1977: SimplifyUrl (&target);
1.54 cvs 1978: if (!change)
1.106 cvs 1979: change = strcmp (src, target);
1.28 cvs 1980: return (change);
1.25 cvs 1981: }
1982:
1.28 cvs 1983:
1.25 cvs 1984: /*----------------------------------------------------------------------
1.31 cvs 1985: MakeRelativeURL: make relative name
1.25 cvs 1986:
1.28 cvs 1987: This function creates and returns a string which gives an expression of
1988: one address as related to another. Where there is no relation, an absolute
1989: address is retured.
1.25 cvs 1990:
1.28 cvs 1991: On entry,
1.25 cvs 1992: Both names must be absolute, fully qualified names of nodes
1993: (no fragment bits)
1994:
1.28 cvs 1995: On exit,
1.25 cvs 1996: The return result points to a newly allocated name which, if
1997: parsed by AmayaParseUrl relative to relatedName, will yield aName.
1998: The caller is responsible for freeing the resulting name later.
1999: ----------------------------------------------------------------------*/
1.106 cvs 2000: char *MakeRelativeURL (char *aName, char *relatedName)
2001: {
2002: char *return_value;
2003: char result[MAX_LENGTH];
2004: char *p;
2005: char *q;
2006: char *after_access;
2007: char *last_slash = NULL;
2008: int slashes, levels, len;
1.110 cvs 2009: #ifdef _WINDOWS
1.44 cvs 2010: int ndx;
1.110 cvs 2011: #endif /* _WINDOWS */
1.44 cvs 2012:
1.29 cvs 2013: if (aName == NULL || relatedName == NULL)
2014: return (NULL);
2015:
2016: slashes = 0;
2017: after_access = NULL;
2018: p = aName;
2019: q = relatedName;
2020: for (; *p && (*p == *q); p++, q++)
1.27 cvs 2021: {
2022: /* Find extent of match */
1.106 cvs 2023: if (*p == ':')
1.29 cvs 2024: after_access = p + 1;
1.28 cvs 2025: if (*p == DIR_SEP)
1.27 cvs 2026: {
1.29 cvs 2027: /* memorize the last slash position and count them */
1.27 cvs 2028: last_slash = p;
2029: slashes++;
1.25 cvs 2030: }
2031: }
2032:
1.31 cvs 2033: /* q, p point to the first non-matching character or zero */
1.106 cvs 2034: if (*q == EOS)
1.31 cvs 2035: {
2036: /* New name is a subset of the related name */
2037: /* exactly the right length */
1.106 cvs 2038: len = strlen (p);
2039: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2040: strcpy (return_value, p);
1.31 cvs 2041: }
2042: else if ((slashes < 2 && after_access == NULL)
2043: || (slashes < 3 && after_access != NULL))
2044: {
2045: /* Two names whitout common path */
2046: /* exactly the right length */
1.106 cvs 2047: len = strlen (aName);
2048: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2049: strcpy (return_value, aName);
1.31 cvs 2050: }
2051: else
2052: {
2053: /* Some path in common */
1.106 cvs 2054: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2055: /* just the same server */
1.106 cvs 2056: strcpy (result, last_slash);
1.31 cvs 2057: else
2058: {
2059: levels= 0;
1.106 cvs 2060: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2061: if (*q == DIR_SEP)
2062: levels++;
2063:
1.106 cvs 2064: result[0] = EOS;
1.31 cvs 2065: for (;levels; levels--)
1.106 cvs 2066: strcat (result, "../");
2067: strcat (result, last_slash+1);
1.31 cvs 2068: }
1.52 cvs 2069:
2070: if (!*result)
1.106 cvs 2071: strcat (result, "./");
1.52 cvs 2072:
1.31 cvs 2073: /* exactly the right length */
1.106 cvs 2074: len = strlen (result);
2075: if ((return_value = TtaGetMemory (len + 1)) != NULL)
2076: strcpy (return_value, result);
1.52 cvs 2077:
1.25 cvs 2078: }
1.110 cvs 2079: #ifdef _WINDOWS
1.106 cvs 2080: len = strlen (return_value);
1.44 cvs 2081: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2082: if (return_value[ndx] == '\\')
2083: return_value[ndx] = '/' ;
1.110 cvs 2084: #endif /* _WINDOWS */
1.29 cvs 2085: return (return_value);
1.24 cvs 2086: }
1.35 cvs 2087:
1.104 kahan 2088: /*----------------------------------------------------------------------
2089: AM_GetFileSize
2090: Returns TRUE and the filesize in the 2nd parameter.
2091: Otherwise, in case of a system error, returns FALSE, with a
2092: filesize of 0L.
2093: ---------------------------------------------------------------------*/
1.106 cvs 2094: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2095: {
1.106 cvs 2096: ThotFileHandle handle = ThotFile_BADHANDLE;
2097: ThotFileInfo info;
1.35 cvs 2098:
1.104 kahan 2099: *file_size = 0L;
2100: if (!TtaFileExist (filename))
2101: return FALSE;
2102:
2103: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2104: if (handle == ThotFile_BADHANDLE)
2105: /* ThotFile_BADHANDLE */
2106: return FALSE;
2107: if (TtaFileStat (handle, &info) == 0)
2108: /* bad stat */
2109: info.size = 0L;
2110: TtaFileClose (handle);
2111: *file_size = (unsigned long) info.size;
2112: return TRUE;
2113: }
Webmaster