Annotation of Amaya/amaya/AHTURLTools.c, revision 1.184
1.7 cvs 1: /*
2: *
1.176 vatton 3: * (c) COPYRIGHT INRIA and W3C, 1996-2004
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: */
1.15 cvs 17: #define THOT_EXPORT extern
1.3 cvs 18: #include "amaya.h"
19:
1.8 cvs 20: #include "init_f.h"
21: #include "AHTURLTools_f.h"
1.100 kahan 22: #include "query_f.h"
1.8 cvs 23:
1.24 cvs 24: #define MAX_PRINT_URL_LENGTH 50
1.106 cvs 25: typedef struct _HTURI
26: {
27: char *access; /* Now known as "scheme" */
28: char *host;
29: char *absolute;
30: char *relative;
31: char *fragment;
1.29 cvs 32: } HTURI;
1.24 cvs 33:
1.175 cvs 34: #ifdef _WINDOWS
1.164 kahan 35: #ifndef PATH_MAX
1.163 cvs 36: #define PATH_MAX MAX_PATH
1.164 kahan 37: #endif
1.155 cvs 38: #define stat _stat
39: #define uint64_t unsigned __int64
40: #define getpid _getpid
1.175 cvs 41: #else /* _WINDOWS */
1.161 kahan 42: #if HAVE_STDINT_H
43: #include <stdint.h>
44: #endif /* HAVE_STDINT_H */
1.175 cvs 45: #endif /* _WINDOWS */
1.28 cvs 46:
47: /*----------------------------------------------------------------------
48: ConvertToLowerCase
49: Converts a string to lowercase.
50: ----------------------------------------------------------------------*/
1.124 vatton 51: void ConvertToLowerCase (char *string)
1.28 cvs 52: {
53: int i;
1.93 cvs 54:
1.28 cvs 55: if (!string)
56: return;
57:
1.106 cvs 58: for (i = 0; string[i] != EOS; i++)
1.123 vatton 59: string[i] = tolower (string[i]);
1.28 cvs 60: }
1.22 cvs 61:
1.8 cvs 62: /*----------------------------------------------------------------------
1.75 cvs 63: EscapeChar
64: writes the equivalent escape code of a char in a string
65: ----------------------------------------------------------------------*/
1.109 cvs 66: void EscapeChar (char *string, char c)
1.75 cvs 67: {
1.109 cvs 68: unsigned int i;
69:
70: i = (unsigned char) c & 0xFF;
71: sprintf (string, "%02x", i);
1.75 cvs 72: }
73:
74: /*----------------------------------------------------------------------
1.96 cvs 75: UnEscapeChar
76: writes the equivalent hex code to a %xx coded char
77: ----------------------------------------------------------------------*/
1.109 cvs 78: static char UnEscapeChar (char c)
1.96 cvs 79: {
1.106 cvs 80: return c >= '0' && c <= '9' ? c - '0'
81: : c >= 'A' && c <= 'F' ? c - 'A' + 10
82: : c - 'a' + 10; /* accept small letters just in case */
1.96 cvs 83: }
84:
85: /*----------------------------------------------------------------------
1.75 cvs 86: EscapeURL
87: Takes a URL and escapes all protected chars into
88: %xx sequences. Also, removes any leading white spaces
89: Returns either NULL or a new buffer, which must be freed by the caller
90: ----------------------------------------------------------------------*/
1.106 cvs 91: char *EscapeURL (const char *url)
92: {
93: char *buffer;
94: int buffer_len;
95: int buffer_free_mem;
96: char *ptr;
97: int new_chars;
1.75 cvs 98: void *status;
99:
100: if (url && *url)
101: {
1.106 cvs 102: buffer_free_mem = strlen (url) + 20;
1.171 gully 103: buffer = (char *)TtaGetMemory (buffer_free_mem + 1);
1.107 kahan 104: ptr = (char *) url;
1.75 cvs 105: buffer_len = 0;
106:
107: while (*ptr)
108: {
109: switch (*ptr)
110: {
111: /* put here below all the chars that need to
112: be escaped into %xx */
1.81 cvs 113: case 0x20: /* space */
1.143 vatton 114: /*case 0x26:*/ /* & */
1.140 kahan 115: case 0x27: /* antislash */
1.75 cvs 116: new_chars = 3;
117: break;
118:
119: default:
1.122 kahan 120: if ((unsigned char )*ptr > 127)
121: new_chars = 3;
122: else
123: new_chars = 1;
1.75 cvs 124: break;
125: }
126:
127: /* see if we need extra room in the buffer */
128: if (new_chars > buffer_free_mem)
129: {
1.76 cvs 130: buffer_free_mem = 20;
1.106 cvs 131: status = TtaRealloc (buffer, sizeof (char)
1.75 cvs 132: * (buffer_len + buffer_free_mem + 1));
133: if (status)
1.114 cvs 134: buffer = (char *) status;
1.106 cvs 135: else
136: {
137: /* @@ maybe we should do some other behavior here, like
138: freeing the buffer and return a void thing */
139: buffer[buffer_len] = EOS;
140: break;
141: }
1.75 cvs 142: }
143: /* escape the char */
1.140 kahan 144: if (new_chars == 3)
145: {
1.106 cvs 146: buffer[buffer_len] = '%';
1.75 cvs 147: EscapeChar (&buffer[buffer_len+1], *ptr);
148: }
1.140 kahan 149: else
150: buffer[buffer_len] = *ptr;
151:
152: /* update the status */
153: buffer_len += new_chars;
154: buffer_free_mem -= new_chars;
155: /* examine the next char */
156: ptr++;
157: }
158: buffer[buffer_len] = EOS;
159: }
160: else
161: buffer = NULL;
162:
163: return (buffer);
164: }
165:
166: /*----------------------------------------------------------------------
167: EscapeXML
168: Takes a string and escapes all protected chars into entity
169: sequences.
170: Returns either NULL or a new buffer, which must be freed by the caller
171: ----------------------------------------------------------------------*/
172: char *EscapeXML (const char *string)
173: {
174: char *buffer;
175: int buffer_len;
176: int buffer_free_mem;
177: char *ptr;
178: char *entity = NULL;
179: int new_chars;
180: void *status;
181:
182: if (string && *string)
183: {
184: buffer_free_mem = strlen (string) + 20;
1.171 gully 185: buffer = (char *)TtaGetMemory (buffer_free_mem + 1);
1.140 kahan 186: ptr = (char *) string;
187: buffer_len = 0;
188:
189: while (*ptr)
190: {
191: switch (*ptr)
192: {
193: case 0x26: /* & */
194: entity = "&";
1.141 kahan 195: new_chars = sizeof (entity) - 1;
196: break;
1.140 kahan 197:
1.141 kahan 198: case '<': /* < */
1.140 kahan 199: entity = "<";
1.141 kahan 200: new_chars = sizeof (entity) - 1;
1.140 kahan 201: break;
202:
1.141 kahan 203: case '>': /* > */
1.140 kahan 204: entity = ">";
1.141 kahan 205: new_chars = sizeof (entity) - 1;
206: break;
207:
208: case '"': /* "e; */
209: entity = ""e;";
210: new_chars = sizeof (entity) - 1;
1.140 kahan 211: break;
212:
213: default:
214: new_chars = 1;
215: break;
216: }
217:
218: /* see if we need extra room in the buffer */
219: if (new_chars > buffer_free_mem)
220: {
221: buffer_free_mem = 20;
222: status = TtaRealloc (buffer, sizeof (char)
223: * (buffer_len + buffer_free_mem + 1));
224: if (status)
225: buffer = (char *) status;
226: else
227: {
228: /* @@ maybe we should do some other behavior here, like
229: freeing the buffer and return a void thing */
230: buffer[buffer_len] = EOS;
231: break;
232: }
233: }
234: /* escape the char */
235: if (entity)
236: {
237: sprintf (&buffer[buffer_len], "%s", entity);
238: entity = NULL;
239: }
1.75 cvs 240: else
241: buffer[buffer_len] = *ptr;
242:
243: /* update the status */
244: buffer_len += new_chars;
245: buffer_free_mem -= new_chars;
246: /* examine the next char */
247: ptr++;
248: }
1.106 cvs 249: buffer[buffer_len] = EOS;
1.75 cvs 250: }
1.76 cvs 251: else
252: buffer = NULL;
253:
1.75 cvs 254: return (buffer);
1.122 kahan 255: }
256:
1.75 cvs 257:
258: /*----------------------------------------------------------------------
1.11 cvs 259: ExplodeURL
1.8 cvs 260: ----------------------------------------------------------------------*/
1.106 cvs 261: void ExplodeURL (char *url, char **proto, char **host, char **dir,
262: char **file)
1.8 cvs 263: {
1.33 cvs 264: char *curr, *temp;
265: char used_sep;
1.32 cvs 266:
1.33 cvs 267: if (url && strchr (url, URL_SEP))
268: used_sep = URL_SEP;
269: else
270: used_sep = DIR_SEP;
1.8 cvs 271:
272: if ((url == NULL) || (proto == NULL) || (host == NULL) ||
273: (dir == NULL) || (file == NULL))
274: return;
275:
276: /* initialize every pointer */
277: *proto = *host = *dir = *file = NULL;
278:
279: /* skip any leading space */
280: while ((*url == SPACE) || (*url == TAB))
281: url++;
1.9 cvs 282: curr = url;
283: if (*curr == 0)
1.8 cvs 284: goto finished;
285:
286: /* go to the end of the URL */
1.68 cvs 287: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) &&
288: (*curr != __CR__) && (*curr != EOL))
1.9 cvs 289: curr++;
1.8 cvs 290:
291: /* mark the end of the chain */
1.9 cvs 292: *curr = EOS;
293: curr--;
294: if (curr <= url)
1.8 cvs 295: goto finished;
296:
297: /* search the next DIR_SEP indicating the beginning of the file name */
298: do
1.11 cvs 299: curr--;
1.33 cvs 300: while ((curr >= url) && (*curr != used_sep));
1.11 cvs 301:
1.9 cvs 302: if (curr < url)
1.8 cvs 303: goto finished;
1.9 cvs 304: *file = curr + 1;
1.8 cvs 305:
306: /* mark the end of the dir */
1.9 cvs 307: *curr = EOS;
308: curr--;
309: if (curr < url)
1.8 cvs 310: goto finished;
311:
1.29 cvs 312: /* search for the DIR_STR indicating the host name start */
1.33 cvs 313: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep)))
1.9 cvs 314: curr--;
1.8 cvs 315:
316: /* if we found it, separate the host name from the directory */
1.102 kahan 317: if ((*curr == used_sep) && (*(curr + 1) == used_sep))
1.8 cvs 318: {
1.9 cvs 319: *host = temp = curr + 2;
1.33 cvs 320: while ((*temp != 0) && (*temp != used_sep))
1.8 cvs 321: temp++;
1.33 cvs 322: if (*temp == used_sep)
1.8 cvs 323: {
324: *temp = EOS;
325: *dir = temp + 1;
326: }
327: }
328: else
1.11 cvs 329: *dir = curr;
330:
1.9 cvs 331: if (curr <= url)
1.8 cvs 332: goto finished;
333:
334: /* mark the end of the proto */
1.9 cvs 335: *curr = EOS;
336: curr--;
337: if (curr < url)
1.8 cvs 338: goto finished;
339:
1.106 cvs 340: if (*curr == ':')
1.8 cvs 341: {
1.9 cvs 342: *curr = EOS;
343: curr--;
1.8 cvs 344: }
345: else
346: goto finished;
1.11 cvs 347:
1.9 cvs 348: if (curr < url)
1.8 cvs 349: goto finished;
1.9 cvs 350: while ((curr > url) && (isalpha (*curr)))
351: curr--;
352: *proto = curr;
1.8 cvs 353:
354: finished:;
355:
356: #ifdef AMAYA_DEBUG
357: fprintf (stderr, "ExplodeURL(%s)\n\t", url);
358: if (*proto)
359: fprintf (stderr, "proto : %s, ", *proto);
360: if (*host)
361: fprintf (stderr, "host : %s, ", *host);
362: if (*dir)
363: fprintf (stderr, "dir : %s, ", *dir);
364: if (*file)
365: fprintf (stderr, "file : %s ", *file);
366: fprintf (stderr, "\n");
367: #endif
368:
369: }
1.3 cvs 370:
1.116 kahan 371: /*----------------------------------------------------------------------
372: PicTypeToMime
373: Converts a Thot PicType into the equivalent MIME type. If no convertion
374: is possible, it returns NULL.
375: ----------------------------------------------------------------------*/
376: char *PicTypeToMIME (PicType contentType)
377: {
378: char *mime_type;
379:
380: switch (contentType)
381: {
1.179 vatton 382: case xbm_type:
383: mime_type ="image/x-xbitmap";
384: break;
1.116 kahan 385: case eps_type:
386: mime_type ="application/postscript";
387: break;
388: case xpm_type:
389: mime_type ="image/x-xpicmap";
390: break;
391: case gif_type:
392: mime_type ="image/gif";
393: break;
394: case jpeg_type:
395: mime_type ="image/jpeg";
396: break;
397: case png_type:
398: mime_type ="image/png";
399: break;
400: case svg_type:
1.165 cvs 401: mime_type = AM_SVG_MIME_TYPE;
402: break;
403: case html_type:
404: mime_type = AM_XHTML_MIME_TYPE;
405: break;
406: case mathml_type:
407: mime_type = AM_MATHML_MIME_TYPE;
1.116 kahan 408: break;
409: case unknown_type:
410: default:
411: mime_type = NULL;
412: }
413:
414: return mime_type;
415: }
1.61 cvs 416:
417: /*----------------------------------------------------------------------
1.117 kahan 418: ImageElement
419: Returns the element (image parameter) and URL (url parameter) of an
420: image in a docImage document. The user must free the memory associated
1.120 kahan 421: with the url parameter if the function is succesful.
422: If the url parameter is NULL, we won't initialize it.
1.117 kahan 423: Returns TRUE if succesful, FALSE otherwise.
424: ----------------------------------------------------------------------*/
425: ThotBool ImageElement (Document doc, char **url, Element *image)
426: {
427: Element el, imgEl;
428: Attribute attr, srcAttr;
429: AttributeType attrType;
430: int length;
1.176 vatton 431: char *utf8value;
1.117 kahan 432:
433: if (DocumentTypes[doc] != docImage)
434: return FALSE;
435:
436: /* find the value of the src attribute */
437: attrType.AttrSSchema = TtaGetSSchema ("HTML", doc);
438: attrType.AttrTypeNum = HTML_ATTR_SRC;
439: el = TtaGetRootElement (doc);
440: TtaSearchAttribute (attrType, SearchInTree, el, &imgEl, &srcAttr);
441:
442: if (!imgEl)
443: return FALSE;
444: *image = imgEl;
445:
1.120 kahan 446: if (url)
447: {
448: attr = TtaGetAttribute (imgEl, attrType);
449: length = TtaGetTextAttributeLength (srcAttr) + 1;
1.176 vatton 450: utf8value = (char *)TtaGetMemory (length);
451: TtaGiveTextAttributeValue (srcAttr, utf8value, &length);
452: *url = (char *)TtaConvertMbsToByte ((unsigned char *)utf8value,
453: TtaGetDefaultCharset ());
454: TtaFreeMemory (utf8value);
1.120 kahan 455: }
1.117 kahan 456: return TRUE;
457: }
458:
459: /*----------------------------------------------------------------------
460: DocImageMimeType
461: Returns the MIME type of a docImage document.
462: ----------------------------------------------------------------------*/
463: char *DocImageMimeType (Document doc)
464: {
465: char *mime_type;
466: LoadedImageDesc *pImage;
467: PicType type;
468: Element image;
469:
470: if (DocumentTypes[doc] != docImage)
471: return NULL;
472:
473: mime_type = NULL;
474: if (!IsHTTPPath (DocumentURLs[doc]))
475: {
476: /* it is a local image */
1.120 kahan 477: if (ImageElement (doc, NULL, &image))
1.117 kahan 478: {
479: type = TtaGetPictureType (image);
480: mime_type = PicTypeToMIME (type);
481: }
482: }
483: else
484: {
485: /* find the value of the src attribute */
486: pImage = ImageURLs;
487: while (pImage != NULL)
488: {
489: if (pImage->document == doc)
490: {
491: if (pImage->content_type)
492: mime_type = pImage->content_type;
493: else if (pImage->elImage && pImage->elImage->currentElement)
494: {
495: type = TtaGetPictureType (pImage->elImage->currentElement);
496: mime_type = PicTypeToMIME (type);
497: }
498: break;
499: }
500: }
501: }
502: return (mime_type);
503: }
504:
1.4 cvs 505: /*----------------------------------------------------------------------
1.9 cvs 506: IsHTMLName
507: returns TRUE if path points to an HTML resource.
1.4 cvs 508: ----------------------------------------------------------------------*/
1.109 cvs 509: ThotBool IsHTMLName (const char *path)
1.106 cvs 510: {
1.183 quint 511: char *temppath;
512: char *suffix;
1.136 cvs 513: char nsuffix[MAX_LENGTH];
514: int i;
1.5 cvs 515:
1.101 cvs 516: if (!path)
517: return (FALSE);
1.5 cvs 518:
1.183 quint 519: temppath = TtaStrdup ((char *)path);
520: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 521: TtaExtractSuffix (temppath, suffix);
1.101 cvs 522: i = 0;
1.106 cvs 523: while (suffix[i] != EOS)
1.101 cvs 524: {
525: /* Normalize the suffix */
526: i = 0;
1.106 cvs 527: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 528: {
1.123 vatton 529: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 530: i++;
531: }
1.106 cvs 532: nsuffix[i] = EOS;
533: if (!strcmp (nsuffix, "html") ||
534: !strcmp (nsuffix, "htm") ||
535: !strcmp (nsuffix, "shtml") ||
536: !strcmp (nsuffix, "jsp") ||
1.159 vatton 537: !strcmp (nsuffix, "tpl") ||
1.106 cvs 538: !strcmp (nsuffix, "xht") ||
539: !strcmp (nsuffix, "xhtm") ||
1.144 cvs 540: !strcmp (nsuffix, "lhtml") ||
1.106 cvs 541: !strcmp (nsuffix, "xhtml"))
1.183 quint 542: {
543: TtaFreeMemory (temppath);
544: TtaFreeMemory (suffix);
545: return (TRUE);
546: }
1.106 cvs 547: else if (!strcmp (nsuffix, "gz"))
1.101 cvs 548: {
549: /* take into account compressed files */
1.124 vatton 550: TtaExtractSuffix (temppath, suffix);
1.101 cvs 551: /* Normalize the suffix */
552: i = 0;
1.106 cvs 553: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.101 cvs 554: {
1.123 vatton 555: nsuffix[i] = tolower (suffix[i]);
1.101 cvs 556: i++;
557: }
1.106 cvs 558: nsuffix[i] = EOS;
1.183 quint 559: TtaFreeMemory (temppath);
560: TtaFreeMemory (suffix);
1.106 cvs 561: if (!strcmp (nsuffix, "html") ||
562: !strcmp (nsuffix, "htm") ||
563: !strcmp (nsuffix, "shtml") ||
564: !strcmp (nsuffix, "jsp") ||
1.159 vatton 565: !strcmp (nsuffix, "tpl") ||
1.106 cvs 566: !strcmp (nsuffix, "xht") ||
567: !strcmp (nsuffix, "xhtm") ||
1.144 cvs 568: !strcmp (nsuffix, "lhtml") ||
1.106 cvs 569: !strcmp (nsuffix, "xhtml"))
1.101 cvs 570: return (TRUE);
571: else
572: return (FALSE);
573: }
574: else
575: /* check if there is another suffix */
1.124 vatton 576: TtaExtractSuffix (temppath, suffix);
1.101 cvs 577: }
1.183 quint 578: TtaFreeMemory (temppath);
579: TtaFreeMemory (suffix);
580: return (FALSE);
1.3 cvs 581: }
582:
1.4 cvs 583: /*----------------------------------------------------------------------
1.136 cvs 584: IsMathMLName
585: returns TRUE if path points to an MathML resource.
1.56 cvs 586: ----------------------------------------------------------------------*/
1.136 cvs 587: ThotBool IsMathMLName (const char *path)
1.56 cvs 588: {
1.183 quint 589: char *temppath;
590: char *suffix;
591: ThotBool ret;
1.56 cvs 592:
593: if (!path)
1.183 quint 594: return (FALSE);
1.56 cvs 595:
1.183 quint 596: temppath = TtaStrdup ((char *)path);
597: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 598: TtaExtractSuffix (temppath, suffix);
1.56 cvs 599:
1.136 cvs 600: if (!strcasecmp (suffix, "mml"))
1.183 quint 601: ret = TRUE;
1.106 cvs 602: else if (!strcmp (suffix, "gz"))
1.56 cvs 603: {
604: /* take into account compressed files */
1.124 vatton 605: TtaExtractSuffix (temppath, suffix);
1.136 cvs 606: if (!strcasecmp (suffix, "mml"))
1.183 quint 607: ret = TRUE;
1.60 cvs 608: else
1.183 quint 609: ret = FALSE;
1.60 cvs 610: }
611: else
1.183 quint 612: ret = FALSE;
613:
614: TtaFreeMemory (temppath);
615: TtaFreeMemory (suffix);
616: return (ret);
1.60 cvs 617: }
618:
619: /*----------------------------------------------------------------------
1.136 cvs 620: IsSVGName
621: returns TRUE if path points to an SVG resource.
1.133 vatton 622: ----------------------------------------------------------------------*/
1.136 cvs 623: ThotBool IsSVGName (const char *path)
1.133 vatton 624: {
1.183 quint 625: char *temppath;
626: char *suffix;
627: ThotBool ret;
1.133 vatton 628:
629: if (!path)
630: return (FALSE);
631:
1.183 quint 632: temppath = TtaStrdup ((char *)path);
633: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.133 vatton 634: TtaExtractSuffix (temppath, suffix);
635:
1.158 vatton 636: if (!strcasecmp (suffix, "svg") || !strcasecmp (suffix, "svgz"))
1.183 quint 637: ret = TRUE;
1.133 vatton 638: else if (!strcmp (suffix, "gz"))
639: {
640: /* take into account compressed files */
641: TtaExtractSuffix (temppath, suffix);
1.136 cvs 642: if (!strcasecmp (suffix, "svg"))
1.183 quint 643: ret = TRUE;
1.133 vatton 644: else
1.183 quint 645: ret = FALSE;
1.133 vatton 646: }
647: else
1.183 quint 648: ret = FALSE;
649:
650: TtaFreeMemory (temppath);
651: TtaFreeMemory (suffix);
652: return (ret);
1.133 vatton 653: }
654:
655: /*----------------------------------------------------------------------
1.136 cvs 656: IsXMLName
657: returns TRUE if path points to an XML resource.
1.103 cvs 658: ----------------------------------------------------------------------*/
1.136 cvs 659: ThotBool IsXMLName (const char *path)
1.103 cvs 660: {
1.183 quint 661: char *temppath;
662: char *suffix;
663: ThotBool ret;
1.103 cvs 664:
665: if (!path)
666: return (FALSE);
667:
1.183 quint 668: temppath = TtaStrdup ((char *)path);
669: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 670: TtaExtractSuffix (temppath, suffix);
1.103 cvs 671:
1.136 cvs 672: if (!strcasecmp (suffix, "xml") ||
673: !strcasecmp (suffix, "xht") ||
674: !strcmp (suffix, "xhtm") ||
1.145 kahan 675: !strcmp (suffix, "xhtml") ||
1.182 vatton 676: !strcmp (suffix, "smi") ||
677: !strcmp (suffix, "zsl"))
1.183 quint 678: ret = TRUE;
1.106 cvs 679: else if (!strcmp (suffix, "gz"))
1.103 cvs 680: {
681: /* take into account compressed files */
1.124 vatton 682: TtaExtractSuffix (temppath, suffix);
1.136 cvs 683: if (!strcasecmp (suffix, "xml") ||
684: !strcasecmp (suffix, "xht") ||
685: !strcmp (suffix, "xhtm") ||
1.145 kahan 686: !strcmp (suffix, "xhtml") ||
1.182 vatton 687: !strcmp (suffix, "smi") |
688: !strcmp (suffix, "xsl"))
1.183 quint 689: ret = TRUE;
1.103 cvs 690: else
1.183 quint 691: ret = FALSE;
1.103 cvs 692: }
693: else
1.183 quint 694: ret = FALSE;
695:
696: TtaFreeMemory (temppath);
697: TtaFreeMemory (suffix);
698: return (ret);
1.103 cvs 699: }
700:
701: /*----------------------------------------------------------------------
1.136 cvs 702: IsUndisplayedName
703: returns TRUE if path points to an undisplayed resource.
1.103 cvs 704: ----------------------------------------------------------------------*/
1.136 cvs 705: ThotBool IsUndisplayedName (const char *path)
1.103 cvs 706: {
1.183 quint 707: char *temppath;
708: char *suffix;
709: ThotBool ret;
1.103 cvs 710:
711: if (!path)
712: return (FALSE);
713:
1.183 quint 714: temppath = TtaStrdup ((char *)path);
715: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 716: TtaExtractSuffix (temppath, suffix);
1.103 cvs 717:
1.136 cvs 718: if (!strcasecmp (suffix, "exe") ||
719: !strcasecmp (suffix, "zip") ||
720: !strcasecmp (suffix, "ppt") ||
721: !strcasecmp (suffix, "pdf") ||
722: !strcasecmp (suffix, "ps") ||
723: !strcasecmp (suffix, "eps") ||
724: !strcasecmp (suffix, "tar") ||
725: !strcasecmp (suffix, "tgz") ||
726: !strcasecmp (suffix, "ddl") ||
727: !strcasecmp (suffix, "o"))
1.183 quint 728: ret = TRUE;
1.106 cvs 729: else if (!strcmp (suffix, "gz"))
1.103 cvs 730: {
731: /* take into account compressed files */
1.124 vatton 732: TtaExtractSuffix (temppath, suffix);
1.136 cvs 733: if (!strcasecmp (suffix, "exe") ||
734: !strcasecmp (suffix, "zip") ||
735: !strcasecmp (suffix, "ppt") ||
736: !strcasecmp (suffix, "pdf") ||
737: !strcasecmp (suffix, "ps") ||
738: !strcasecmp (suffix, "eps") ||
739: !strcasecmp (suffix, "tar") ||
740: !strcasecmp (suffix, "ddl") ||
741: !strcasecmp (suffix, "o"))
1.183 quint 742: ret = TRUE;
1.103 cvs 743: else
1.183 quint 744: ret = FALSE;
1.103 cvs 745: }
746: else
1.183 quint 747: ret = FALSE;
748:
749: TtaFreeMemory (temppath);
750: TtaFreeMemory (suffix);
751: return (ret);
1.103 cvs 752: }
753:
754: /*----------------------------------------------------------------------
1.60 cvs 755: IsCSSName
756: returns TRUE if path points to an XML resource.
757: ----------------------------------------------------------------------*/
1.111 cvs 758: ThotBool IsCSSName (const char *path)
1.60 cvs 759: {
1.183 quint 760: char *temppath;
761: char *suffix;
762: ThotBool ret;
1.60 cvs 763:
764: if (!path)
1.183 quint 765: return (FALSE);
1.60 cvs 766:
1.183 quint 767: temppath = TtaStrdup ((char *)path);
768: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 769: TtaExtractSuffix (temppath, suffix);
1.60 cvs 770:
1.106 cvs 771: if (!strcasecmp (suffix, "css"))
1.183 quint 772: ret = TRUE;
1.106 cvs 773: else if (!strcmp (suffix, "gz"))
1.60 cvs 774: {
775: /* take into account compressed files */
1.124 vatton 776: TtaExtractSuffix (temppath, suffix);
1.106 cvs 777: if (!strcasecmp (suffix, "css"))
1.183 quint 778: ret = TRUE;
1.56 cvs 779: else
1.183 quint 780: ret = FALSE;
1.56 cvs 781: }
782: else
1.183 quint 783: ret = FALSE;
784:
785: TtaFreeMemory (temppath);
786: TtaFreeMemory (suffix);
787: return (ret);
1.56 cvs 788: }
789:
790: /*----------------------------------------------------------------------
1.177 kahan 791: MultipleBookmarks
792: returns the value of the multiple bookmarks environment variable.
793: (TRUE if enabled).
794: ----------------------------------------------------------------------*/
795: ThotBool MultipleBookmarks (void)
796: {
797: ThotBool multiple_bm;
798:
1.180 kahan 799: TtaGetEnvBoolean ("DISABLE_MULTIPLE_BM", &multiple_bm);
1.177 kahan 800:
1.180 kahan 801: return (multiple_bm == FALSE);
1.177 kahan 802: }
803:
804: /*----------------------------------------------------------------------
1.172 kahan 805: IsRDFName
806: returns TRUE if path points to an RDF resource.
807: ----------------------------------------------------------------------*/
808: ThotBool IsRDFName (const char *path)
809: {
1.183 quint 810: char *temppath;
811: char *suffix;
812: ThotBool ret;
1.172 kahan 813:
1.177 kahan 814: /* temporarily disabling this function */
815: if (!MultipleBookmarks ())
1.183 quint 816: return (FALSE);
1.177 kahan 817:
1.172 kahan 818: if (!path)
1.183 quint 819: return (FALSE);
1.172 kahan 820:
1.183 quint 821: temppath = TtaStrdup ((char *)path);
822: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.172 kahan 823: TtaExtractSuffix (temppath, suffix);
824:
825: if (!strcasecmp (suffix, "rdf"))
1.183 quint 826: ret = TRUE;
1.172 kahan 827: else if (!strcmp (suffix, "gz"))
828: {
829: /* take into account compressed files */
830: TtaExtractSuffix (temppath, suffix);
831: if (!strcasecmp (suffix, "rdf"))
1.183 quint 832: ret = TRUE;
1.172 kahan 833: else
1.183 quint 834: ret = FALSE;
1.172 kahan 835: }
836: else
1.183 quint 837: ret = FALSE;
838:
839: TtaFreeMemory (temppath);
840: TtaFreeMemory (suffix);
841: return (ret);
1.172 kahan 842: }
843:
844: /*----------------------------------------------------------------------
1.9 cvs 845: IsImageName
846: returns TRUE if path points to an image resource.
1.4 cvs 847: ----------------------------------------------------------------------*/
1.111 cvs 848: ThotBool IsImageName (const char *path)
1.106 cvs 849: {
1.183 quint 850: char *temppath;
851: char *suffix;
1.106 cvs 852: char nsuffix[MAX_LENGTH];
1.5 cvs 853: int i;
1.183 quint 854: ThotBool ret;
1.5 cvs 855:
856: if (!path)
1.13 cvs 857: return (FALSE);
1.5 cvs 858:
1.183 quint 859: temppath = TtaStrdup ((char *)path);
860: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 861: TtaExtractSuffix (temppath, suffix);
1.5 cvs 862:
863: /* Normalize the suffix */
864: i = 0;
1.106 cvs 865: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 866: {
1.123 vatton 867: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 868: i++;
869: }
1.106 cvs 870: nsuffix[i] = EOS;
871: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
872: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
873: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.183 quint 874: ret = TRUE;
875: else
876: ret = FALSE;
877:
878: TtaFreeMemory (temppath);
879: TtaFreeMemory (suffix);
880: return (ret);
1.3 cvs 881: }
882:
1.4 cvs 883: /*----------------------------------------------------------------------
1.58 cvs 884: IsImageType
885: returns TRUE if type points to an image resource.
886: ----------------------------------------------------------------------*/
1.111 cvs 887: ThotBool IsImageType (const char *type)
1.58 cvs 888: {
1.183 quint 889: char *temptype;
1.58 cvs 890: int i;
1.183 quint 891: ThotBool ret;
1.58 cvs 892:
893: if (!type)
894: return (FALSE);
895:
1.183 quint 896: temptype = TtaStrdup ((char *)type);
1.58 cvs 897: /* Normalize the type */
898: i = 0;
1.106 cvs 899: while (temptype[i] != EOS)
1.58 cvs 900: {
901: temptype[i] = tolower (temptype[i]);
902: i++;
903: }
1.166 vatton 904: if (!strncmp (temptype, "image/", sizeof ("image/") - 1))
905: i = sizeof ("image/") - 1;
906: else
907: i = 0;
908: if (!strcmp (&temptype[i], "gif") || !strcmp (&temptype[i], "x-xbitmap") ||
909: !strcmp (&temptype[i], "x-xpixmap") || !strcmp (&temptype[i], "jpeg") ||
910: !strcmp (&temptype[i], "png"))
1.183 quint 911: ret = TRUE;
912: else
913: ret = FALSE;
914: TtaFreeMemory (temptype);
915: return (ret);
1.58 cvs 916: }
917:
918: /*----------------------------------------------------------------------
1.9 cvs 919: IsTextName
1.4 cvs 920: ----------------------------------------------------------------------*/
1.111 cvs 921: ThotBool IsTextName (const char *path)
1.106 cvs 922: {
1.183 quint 923: char *temppath;
924: char *suffix;
1.106 cvs 925: char nsuffix[MAX_LENGTH];
1.5 cvs 926: int i;
1.183 quint 927: ThotBool ret;
1.5 cvs 928:
929: if (!path)
1.13 cvs 930: return (FALSE);
1.5 cvs 931:
1.183 quint 932: temppath = TtaStrdup ((char *)path);
933: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 934: TtaExtractSuffix (temppath, suffix);
1.5 cvs 935:
936: /* Normalize the suffix */
937: i = 0;
1.106 cvs 938: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 939: {
1.25 cvs 940: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 941: i++;
942: }
1.106 cvs 943: nsuffix[i] = EOS;
1.5 cvs 944:
1.111 cvs 945: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.183 quint 946: ret = TRUE;
1.106 cvs 947: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 948: {
1.39 cvs 949: /* take into account compressed files */
1.124 vatton 950: TtaExtractSuffix (temppath, suffix);
1.13 cvs 951: /* Normalize the suffix */
952: i = 0;
1.106 cvs 953: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 954: {
1.25 cvs 955: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 956: i++;
957: }
1.106 cvs 958: nsuffix[i] = EOS;
1.111 cvs 959: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.183 quint 960: ret = TRUE;
1.13 cvs 961: else
1.183 quint 962: ret = FALSE;
1.13 cvs 963: }
964: else
1.183 quint 965: ret = FALSE;
966:
967: TtaFreeMemory (temppath);
968: TtaFreeMemory (suffix);
969: return (ret);
1.3 cvs 970: }
971:
1.4 cvs 972: /*----------------------------------------------------------------------
1.9 cvs 973: IsHTTPPath
974: returns TRUE if path is in fact an http URL.
1.4 cvs 975: ----------------------------------------------------------------------*/
1.112 cvs 976: ThotBool IsHTTPPath (const char *path)
1.3 cvs 977: {
1.5 cvs 978: if (!path)
979: return FALSE;
1.3 cvs 980:
1.106 cvs 981: if ((!strncmp (path, "http:", 5) != 0)
982: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
983: || !strncmp (path, "internal:", 9))
1.58 cvs 984: return TRUE;
985: return FALSE;
1.3 cvs 986: }
987:
1.4 cvs 988: /*----------------------------------------------------------------------
1.9 cvs 989: IsWithParameters
990: returns TRUE if url has a concatenated query string.
1.4 cvs 991: ----------------------------------------------------------------------*/
1.133 vatton 992: ThotBool IsWithParameters (const char *url)
1.3 cvs 993: {
1.5 cvs 994: int i;
1.3 cvs 995:
1.9 cvs 996: if ((!url) || (url[0] == EOS))
1.5 cvs 997: return FALSE;
1.3 cvs 998:
1.9 cvs 999: i = strlen (url) - 1;
1000: while (i > 0 && url[i--] != '?')
1.5 cvs 1001: if (i < 0)
1002: return FALSE;
1.3 cvs 1003:
1.5 cvs 1004: /* There is a parameter */
1005: return TRUE;
1.3 cvs 1006: }
1007:
1.4 cvs 1008: /*----------------------------------------------------------------------
1.9 cvs 1009: IsW3Path
1010: returns TRUE if path is in fact a URL.
1.4 cvs 1011: ----------------------------------------------------------------------*/
1.133 vatton 1012: ThotBool IsW3Path (const char *path)
1.106 cvs 1013: {
1.170 quint 1014: if (path == NULL)
1015: return FALSE;
1.106 cvs 1016: if (strncmp (path, "http:", 5) &&
1017: strncmp (path, "ftp:", 4) &&
1018: strncmp (path, "telnet:", 7) &&
1019: strncmp (path, "wais:", 5) &&
1020: strncmp (path, "news:", 5) &&
1021: strncmp (path, "gopher:", 7) &&
1022: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 1023: strncmp (path, "archie:", 7) &&
1024: strncmp (path, "https:", 6))
1.72 cvs 1025: return FALSE;
1026: return TRUE;
1.3 cvs 1027: }
1028:
1.4 cvs 1029: /*----------------------------------------------------------------------
1.90 cvs 1030: IsFilePath
1031: returns TRUE if path is in fact a URL.
1032: ----------------------------------------------------------------------*/
1.133 vatton 1033: ThotBool IsFilePath (const char *path)
1.90 cvs 1034: {
1.106 cvs 1035: if (strncmp (path, "file:", 5))
1.90 cvs 1036: return FALSE;
1037: return TRUE;
1038: }
1039:
1040: /*----------------------------------------------------------------------
1.9 cvs 1041: IsValidProtocol
1042: returns true if the url protocol is supported by Amaya.
1.4 cvs 1043: ----------------------------------------------------------------------*/
1.133 vatton 1044: ThotBool IsValidProtocol (const char *url)
1.106 cvs 1045: {
1046: if (!strncmp (url, "http:", 5)
1047: || !strncmp (url, "internal:", 9)
1048: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 1049: /* experimental */
1.24 cvs 1050: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 1051: return (TRUE);
1.5 cvs 1052: else
1.8 cvs 1053: return (FALSE);
1.3 cvs 1054: }
1055:
1.31 cvs 1056:
1057: /*----------------------------------------------------------------------
1058: GetBaseURL
1059: normalizes orgName according to a base associated with doc, and
1060: following the standard URL format rules.
1061: The function returns the base used to solve relative URL and SRC:
1062: - the base of the document,
1063: - or the document path (without document name).
1064: ----------------------------------------------------------------------*/
1.106 cvs 1065: char *GetBaseURL (Document doc)
1.31 cvs 1066: {
1067: Element el;
1068: ElementType elType;
1069: AttributeType attrType;
1070: Attribute attr;
1.176 vatton 1071: char *ptr, *basename, *utf8path;
1.31 cvs 1072: int length;
1.151 kahan 1073: ThotBool hasDocBase;
1.31 cvs 1074:
1.113 cvs 1075: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 1076: return NULL;
1.148 kahan 1077: /* the other functions expect basename to have no more than MAX_LENGTH chars */
1.171 gully 1078: basename = (char *)TtaGetMemory (MAX_LENGTH);
1.148 kahan 1079: basename[0] = EOS;
1.31 cvs 1080: length = MAX_LENGTH -1;
1.151 kahan 1081: hasDocBase = FALSE;
1082:
1083: /* If the document has a base URL, it has a priority over the headers. */
1084: /* @@ We need to do this too when we support XML:base */
1085:
1086: /* is it a HTML document ? */
1087: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1088: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
1089: /* it's a HTML document */
1090: {
1091: /* get the document element */
1092: el = TtaGetMainRoot (doc);
1093: /* search the BASE element */
1094: elType.ElTypeNum = HTML_EL_HEAD;
1095: el = TtaSearchTypedElement (elType, SearchForward, el);
1096: if (el)
1097: /* there is a HEAD element */
1098: {
1099: /* look for a BASE element within the HEAD */
1100: elType.ElTypeNum = HTML_EL_BASE;
1101: el = TtaSearchTypedElement (elType, SearchInTree, el);
1102: }
1103: if (el)
1104: {
1105: /* The document has a BASE element. Get the HREF attribute of the
1106: BASE element */
1107: hasDocBase = TRUE;
1108: attrType.AttrSSchema = elType.ElSSchema;
1109: attrType.AttrTypeNum = HTML_ATTR_HREF_;
1110: attr = TtaGetAttribute (el, attrType);
1111: if (attr)
1112: {
1113: /* Use the base path of the document */
1114: TtaGiveTextAttributeValue (attr, basename, &length);
1.176 vatton 1115: utf8path = basename;
1116: basename = (char *)TtaConvertMbsToByte ((unsigned char *)utf8path,
1117: TtaGetDefaultCharset ());
1118: TtaFreeMemory (utf8path);
1.151 kahan 1119: }
1120: }
1121: }
1122:
1123: /* there was no BASE. Do we have a location header? */
1124: if (!hasDocBase && DocumentMeta[doc] && DocumentMeta[doc]->full_content_location
1.148 kahan 1125: && DocumentMeta[doc]->full_content_location[0] != EOS)
1.65 cvs 1126: {
1.148 kahan 1127: strncpy (basename, DocumentMeta[doc]->full_content_location, MAX_LENGTH-1);
1128: basename[MAX_LENGTH-1] = EOS;
1129: length = strlen (basename);
1130: }
1131:
1132: if (basename[0] != EOS)
1133: {
1134: /* base and orgName have to be separated by a DIR_SEP */
1135: length--;
1136: if (basename[0] != EOS && basename[length] != URL_SEP &&
1137: basename[length] != DIR_SEP)
1138: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 1139: {
1.148 kahan 1140: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1141: AMAYA_PARSE_HOST |
1142: AMAYA_PARSE_PUNCTUATION);
1143: if (ptr && !strcmp (ptr, basename))
1.31 cvs 1144: {
1.148 kahan 1145: /* it has this form, complete it by adding a URL_STR */
1146: if (strchr (basename, DIR_SEP))
1147: strcat (basename, DIR_STR);
1148: else
1149: strcat (basename, URL_STR);
1150: length++;
1.31 cvs 1151: }
1.149 kahan 1152: else if (!ptr || ptr[0] == EOS)
1153: {
1154: /* no host was detected, we may have a relative URL. We test
1155: if it begins with a URL_SEP, DIR_SEP or period. If yes, it's
1156: relative. */
1157: if (! (basename[0] == '.' || basename[0] == URL_SEP
1158: || basename[0] == DIR_SEP))
1159: basename[0] = EOS;
1160: }
1.148 kahan 1161: if (ptr)
1162: TtaFreeMemory (ptr);
1.31 cvs 1163: }
1.113 cvs 1164: }
1165:
1.148 kahan 1166: /* there was no base element and no location header, we use the DocumentURL */
1167: if (basename[0] == EOS)
1168: {
1169: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1170: basename[MAX_LENGTH-1] = EOS;
1171: }
1172:
1.31 cvs 1173: /* Remove anything after the last DIR_SEP char. If no such char is found,
1174: * then search for the first ":" char, hoping that what's before that is a
1175: * protocol. If found, end the string there. If neither char is found,
1176: * then discard the whole base element.
1177: */
1.106 cvs 1178: length = strlen (basename) - 1;
1.31 cvs 1179: /* search for the last DIR_SEP char */
1.106 cvs 1180: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 1181: length--;
1182: if (length >= 0)
1183: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1184: basename[length + 1] = EOS;
1.31 cvs 1185: else
1186: /* search for the first PATH_STR char */
1187: {
1.106 cvs 1188: for (length = 0; basename[length] != ':' &&
1189: basename[length] != EOS; length ++);
1190: if (basename[length] == ':')
1.31 cvs 1191: /* found, so end the string there */
1.106 cvs 1192: basename[length + 1] = EOS;
1.31 cvs 1193: else
1194: /* not found, discard the base */
1.106 cvs 1195: basename[0] = EOS;
1.31 cvs 1196: }
1197: return (basename);
1198: }
1199:
1200:
1.4 cvs 1201: /*----------------------------------------------------------------------
1.40 cvs 1202: GetLocalPath
1203: Allocate and return the local document path associated to the url
1204: ----------------------------------------------------------------------*/
1.150 vatton 1205: char *GetLocalPath (Document doc, char *url)
1.106 cvs 1206: {
1207: char *ptr;
1208: char *n;
1209: char *documentname;
1210: char url_sep;
1.83 cvs 1211: int len;
1.67 cvs 1212: ThotBool noFile;
1.40 cvs 1213:
1.153 vatton 1214: if (url)
1.40 cvs 1215: {
1216: /* check whether the file name exists */
1.106 cvs 1217: len = strlen (url) - 1;
1.71 cvs 1218: if (IsW3Path (url))
1.157 kahan 1219: url_sep = URL_SEP;
1.41 cvs 1220: else
1.106 cvs 1221: url_sep = DIR_SEP;
1.41 cvs 1222: noFile = (url[len] == url_sep);
1.40 cvs 1223: if (noFile)
1.106 cvs 1224: url[len] = EOS;
1.171 gully 1225: ptr = (char *)TtaGetMemory (MAX_LENGTH);
1226: documentname = (char *)TtaGetMemory (MAX_LENGTH);
1.78 cvs 1227: TtaExtractName (url, ptr, documentname);
1.106 cvs 1228: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1229: if (!TtaCheckDirectory (ptr))
1230: /* directory did not exist */
1.72 cvs 1231: TtaMakeDirectory (ptr);
1.47 cvs 1232:
1.153 vatton 1233: if (doc == 0)
1234: {
1235: n = strrchr (documentname, '.');
1236: if (n)
1237: *n = EOS;
1238: if (documentname[0] == EOS)
1239: strcpy (documentname, "noname");
1240: n = GetTempName (ptr, documentname);
1241: TtaFreeMemory (ptr);
1242: ptr = n;
1243: }
1.69 cvs 1244: else
1.153 vatton 1245: {
1246: /* don't include the query string within document name */
1247: n = strrchr (documentname, '?');
1248: if (n)
1249: *n = EOS;
1250: /* don't include ':' within document name */
1251: n = strchr (documentname, ':');
1252: if (n)
1253: *n = EOS;
1254: /* if after all this operations document name
1255: is empty, let's use noname.html instead */
1256: if (documentname[0] == EOS)
1257: strcat (ptr, "noname.html");
1258: else
1259: strcat (ptr, documentname);
1260: }
1.40 cvs 1261: TtaFreeMemory (documentname);
1.157 kahan 1262: /* substitute invalid chars in file names by a _ */
1263: n = ptr;
1264: while (*n)
1265: {
1266: if (*n == '*' || *n == ',')
1267: *n = '_';
1268: n++;
1269: }
1.40 cvs 1270: /* restore the url */
1271: if (noFile)
1.41 cvs 1272: url[len] = url_sep;
1.40 cvs 1273: return (ptr);
1274: }
1275: else
1276: return (NULL);
1277: }
1278:
1.73 cvs 1279: /*----------------------------------------------------------------------
1.79 cvs 1280: ExtractTarget extract the target name from document nane.
1281: ----------------------------------------------------------------------*/
1.150 vatton 1282: void ExtractTarget (char *aName, char *target)
1.79 cvs 1283: {
1.106 cvs 1284: int lg, i;
1285: char *ptr;
1286: char *oldptr;
1.79 cvs 1287:
1288: if (!target || !aName)
1289: /* bad target */
1290: return;
1291:
1.106 cvs 1292: target[0] = EOS;
1293: lg = strlen (aName);
1.79 cvs 1294: if (lg)
1295: {
1296: /* the name is not empty */
1297: oldptr = ptr = &aName[0];
1298: do
1299: {
1.106 cvs 1300: ptr = strrchr (oldptr, '#');
1.79 cvs 1301: if (ptr)
1302: oldptr = &ptr[1];
1303: }
1304: while (ptr);
1305:
1306: i = (int) (oldptr) - (int) (aName); /* name length */
1307: if (i > 1)
1308: {
1.106 cvs 1309: aName[i - 1] = EOS;
1.79 cvs 1310: if (i != lg)
1.106 cvs 1311: strcpy (target, oldptr);
1.79 cvs 1312: }
1313: }
1314: }
1315:
1316: /*----------------------------------------------------------------------
1.90 cvs 1317: RemoveNewLines (text)
1318: Removes any '\n' chars that are found in text.
1319: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1320: ----------------------------------------------------------------------*/
1.106 cvs 1321: ThotBool RemoveNewLines (char *text)
1322: {
1323: ThotBool change = FALSE;
1324: char *src;
1325: char *dest;
1.90 cvs 1326:
1327: src = text;
1328: dest = text;
1.115 kahan 1329:
1330: /* remove any preceding whitespace */
1331: while (*src && *src == ' ')
1332: {
1333: src++;
1334: change = 1;
1335: }
1336:
1.90 cvs 1337: while (*src)
1338: {
1339: switch (*src)
1340: {
1.106 cvs 1341: case '\n':
1.90 cvs 1342: /* don't copy the newline */
1343: change = 1;
1344: break;
1345: default:
1346: *dest = *src;
1347: dest++;
1348: break;
1349: }
1350: src++;
1351: }
1352: /* copy the last EOS char */
1353: *dest = *src;
1354:
1355: return (change);
1356: }
1357:
1358: /*----------------------------------------------------------------------
1359: CleanCopyFileURL
1360: Copies a file url from a src string to destination string.
1.97 cvs 1361: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1362: we want to do).
1.90 cvs 1363: ----------------------------------------------------------------------*/
1.106 cvs 1364: static void CleanCopyFileURL (char *dest, char *src,
1365: ConvertionType convertion)
1.90 cvs 1366: {
1367: while (*src)
1.89 cvs 1368: {
1.90 cvs 1369: switch (*src)
1.89 cvs 1370: {
1.184 ! gully 1371: #ifdef _WINDOWS
1.106 cvs 1372: case URL_SEP:
1.96 cvs 1373: /* make DIR_SEP transformation */
1.97 cvs 1374: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1375: *dest = DIR_SEP;
1.96 cvs 1376: else
1377: *dest = *src;
1.90 cvs 1378: dest++;
1.96 cvs 1379: src++;
1.90 cvs 1380: break;
1.184 ! gully 1381: #endif /* _WINDOWS */
1.96 cvs 1382:
1.106 cvs 1383: case '%':
1.97 cvs 1384: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1385: {
1.97 cvs 1386: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1387: src++;
1.106 cvs 1388: if (*src != EOS)
1.97 cvs 1389: {
1390: *dest = UnEscapeChar (*src) * 16;
1391: src++;
1392: }
1.106 cvs 1393: if (*src != EOS)
1.97 cvs 1394: {
1395: *dest = *dest + UnEscapeChar (*src);
1396: src++;
1397: }
1398: dest++;
1.96 cvs 1399: }
1.97 cvs 1400: else
1.96 cvs 1401: {
1.97 cvs 1402: *dest = *src;
1403: dest++;
1.96 cvs 1404: src++;
1405: }
1406: break;
1407:
1.90 cvs 1408: default:
1409: *dest = *src;
1.89 cvs 1410: dest++;
1.96 cvs 1411: src++;
1.90 cvs 1412: break;
1.89 cvs 1413: }
1414: }
1.90 cvs 1415: /* copy the EOS char */
1416: *dest = *src;
1.73 cvs 1417: }
1.40 cvs 1418:
1419: /*----------------------------------------------------------------------
1.9 cvs 1420: NormalizeURL
1421: normalizes orgName according to a base associated with doc, and
1422: following the standard URL format rules.
1.113 cvs 1423: if doc is < 0, use as a base the URL of the document that contains
1424: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1425: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1426: other path.
1.9 cvs 1427: The function returns the new complete and normalized URL
1.12 cvs 1428: or file name path (newName) and the name of the document (docName).
1.9 cvs 1429: N.B. If the function can't find out what's the docName, it assigns
1430: the name "noname.html".
1.4 cvs 1431: ----------------------------------------------------------------------*/
1.106 cvs 1432: void NormalizeURL (char *orgName, Document doc, char *newName,
1433: char *docName, char *otherPath)
1434: {
1435: char *basename;
1436: char tempOrgName[MAX_LENGTH];
1437: char *ptr;
1438: char used_sep;
1.84 cvs 1439: int length;
1440: ThotBool check;
1.5 cvs 1441:
1.184 ! gully 1442: #ifdef _WINDOWS
1.44 cvs 1443: int ndx;
1.184 ! gully 1444: #endif /* _WINDOWS */
1.44 cvs 1445:
1.5 cvs 1446: if (!newName || !docName)
1447: return;
1.18 cvs 1448:
1.113 cvs 1449: if (doc < 0)
1450: basename = TtaStrdup (SavedDocumentURL);
1451: else if (doc > 0)
1.53 cvs 1452: basename = GetBaseURL (doc);
1453: else if (otherPath != NULL)
1.108 cvs 1454: basename = TtaStrdup (otherPath);
1.32 cvs 1455: else
1.53 cvs 1456: basename = NULL;
1.32 cvs 1457:
1.18 cvs 1458: /*
1.31 cvs 1459: * Clean orgName
1460: * Make sure we have a complete orgName, without any leading or trailing
1461: * white spaces, or trailinbg new lines
1462: */
1.5 cvs 1463: ptr = orgName;
1.18 cvs 1464: /* skip leading white space and new line characters */
1.106 cvs 1465: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1466: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1467: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1468: /*
1.31 cvs 1469: * Make orgName a complete URL
1470: * If the URL does not include a protocol, then try to calculate
1471: * one using the doc's base element (if it exists),
1472: */
1.106 cvs 1473: if (tempOrgName[0] == EOS)
1.53 cvs 1474: {
1.106 cvs 1475: newName[0] = EOS;
1476: docName[0] = EOS;
1.53 cvs 1477: TtaFreeMemory (basename);
1478: return;
1479: }
1.49 cvs 1480:
1481: /* clean trailing white space */
1.106 cvs 1482: length = strlen (tempOrgName) - 1;
1483: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1484: {
1.106 cvs 1485: tempOrgName[length] = EOS;
1.53 cvs 1486: length--;
1487: }
1.50 cvs 1488:
1.55 cvs 1489: /* remove extra dot (which dot???) */
1490: /* ugly, but faster than a strcmp */
1.106 cvs 1491: if (tempOrgName[length] == '.'
1492: && (length == 0 || tempOrgName[length-1] != '.'))
1493: tempOrgName[length] = EOS;
1.50 cvs 1494:
1.94 cvs 1495: if (IsW3Path (tempOrgName))
1.53 cvs 1496: {
1497: /* the name is complete, go to the Sixth Step */
1.106 cvs 1498: strcpy (newName, tempOrgName);
1.53 cvs 1499: SimplifyUrl (&newName);
1500: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1501: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1502: AMAYA_PARSE_HOST |
1503: AMAYA_PARSE_PUNCTUATION);
1504: if (ptr && !strcmp (ptr, newName))
1505: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1506: strcat (newName, URL_STR);
1.49 cvs 1507:
1.53 cvs 1508: if (ptr)
1.50 cvs 1509: TtaFreeMemory (ptr);
1.53 cvs 1510: }
1.113 cvs 1511: else if (basename == NULL)
1.53 cvs 1512: /* the name is complete, go to the Sixth Step */
1.106 cvs 1513: strcpy (newName, tempOrgName);
1.53 cvs 1514: else
1515: {
1.31 cvs 1516: /* Calculate the absolute URL, using the base or document URL */
1.184 ! gully 1517: #ifdef _WINDOWS
1.53 cvs 1518: if (!IsW3Path (basename))
1519: {
1.106 cvs 1520: length = strlen (tempOrgName);
1.53 cvs 1521: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1522: if (tempOrgName [ndx] == '/')
1523: tempOrgName [ndx] = '\\';
1.53 cvs 1524: }
1.184 ! gully 1525: #endif /* _WINDOWS */
1.25 cvs 1526: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1527: if (ptr)
1528: {
1529: SimplifyUrl (&ptr);
1.106 cvs 1530: strcpy (newName, ptr);
1.53 cvs 1531: TtaFreeMemory (ptr);
1532: }
1533: else
1.106 cvs 1534: newName[0] = EOS;
1.53 cvs 1535: }
1.36 cvs 1536:
1537: TtaFreeMemory (basename);
1.18 cvs 1538: /*
1.31 cvs 1539: * Prepare the docname that will refer to this ressource in the
1540: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1541: * noname.html as a default ressource name
1.18 cvs 1542: */
1.106 cvs 1543: if (newName[0] != EOS)
1.53 cvs 1544: {
1.106 cvs 1545: length = strlen (newName) - 1;
1546: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1547: {
1548: used_sep = newName[length];
1549: check = TRUE;
1550: while (check)
1551: {
1.50 cvs 1552: length--;
1553: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1554: length--;
1.106 cvs 1555: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1556: {
1.106 cvs 1557: newName[length+1] = EOS;
1.53 cvs 1558: /* remove also previous directory */
1559: length--;
1560: while (length >= 0 && newName[length] != used_sep)
1561: length--;
1.106 cvs 1562: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1563: /* don't remove server name */
1.106 cvs 1564: newName[length+1] = EOS;
1.53 cvs 1565: }
1.106 cvs 1566: else if (!strncmp (&newName[length+1], ".", 1))
1567: newName[length+1] = EOS;
1.50 cvs 1568: else
1.53 cvs 1569: check = FALSE;
1570: }
1571: /* docname was not comprised inside the URL, so let's */
1572: /* assign the default ressource name */
1.106 cvs 1573: strcpy (docName, "noname.html");
1.53 cvs 1574: }
1575: else
1576: { /* docname is comprised inside the URL */
1.110 cvs 1577: while (length >= 0 && newName[length] != URL_SEP &&
1578: newName[length] != DIR_SEP)
1.53 cvs 1579: length--;
1580: if (length < 0)
1.106 cvs 1581: strcpy (docName, newName);
1.53 cvs 1582: else
1.106 cvs 1583: strcpy (docName, &newName[length+1]);
1.53 cvs 1584: }
1585: }
1586: else
1.106 cvs 1587: docName[0] = EOS;
1.18 cvs 1588: }
1.3 cvs 1589:
1.4 cvs 1590: /*----------------------------------------------------------------------
1.9 cvs 1591: IsSameHost
1.4 cvs 1592: ----------------------------------------------------------------------*/
1.106 cvs 1593: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1594: {
1.106 cvs 1595: char *basename_ptr1, *basename_ptr2;
1596: ThotBool result;
1.3 cvs 1597:
1.106 cvs 1598: basename_ptr1 = AmayaParseUrl (url1, "",
1599: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1600: basename_ptr2 = AmayaParseUrl (url2, "",
1601: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1602:
1.106 cvs 1603: if (strcmp (basename_ptr1, basename_ptr2))
1604: result = FALSE;
1605: else
1606: result = TRUE;
1607: TtaFreeMemory (basename_ptr1);
1608: TtaFreeMemory (basename_ptr2);
1609: return (result);
1.3 cvs 1610: }
1611:
1612:
1.4 cvs 1613: /*----------------------------------------------------------------------
1.22 cvs 1614: HasKnownFileSuffix
1615: returns TRUE if path points to a file ending with a suffix.
1616: ----------------------------------------------------------------------*/
1.153 vatton 1617: ThotBool HasKnownFileSuffix (const char *path)
1.106 cvs 1618: {
1619: char *root;
1620: char temppath[MAX_LENGTH];
1621: char suffix[MAX_LENGTH];
1.22 cvs 1622:
1.106 cvs 1623: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1624: return (FALSE);
1625:
1.106 cvs 1626: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1627:
1628: if (root)
1629: {
1.106 cvs 1630: strcpy (temppath, root);
1.25 cvs 1631: TtaFreeMemory (root);
1.22 cvs 1632: /* Get the suffix */
1.124 vatton 1633: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1634:
1.106 cvs 1635: if( suffix[0] == EOS)
1.22 cvs 1636: /* no suffix */
1637: return (FALSE);
1638:
1639: /* Normalize the suffix */
1640: ConvertToLowerCase (suffix);
1641:
1.106 cvs 1642: if (!strcmp (suffix, "gz"))
1.22 cvs 1643: /* skip the compressed suffix */
1644: {
1.124 vatton 1645: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1646: if(suffix[0] == EOS)
1.22 cvs 1647: /* no suffix */
1648: return (FALSE);
1649: /* Normalize the suffix */
1650: ConvertToLowerCase (suffix);
1651: }
1652:
1.106 cvs 1653: if (strcmp (suffix, "gif") &&
1654: strcmp (suffix, "xbm") &&
1655: strcmp (suffix, "xpm") &&
1656: strcmp (suffix, "jpg") &&
1657: strcmp (suffix, "pdf") &&
1658: strcmp (suffix, "png") &&
1659: strcmp (suffix, "tgz") &&
1660: strcmp (suffix, "xpg") &&
1661: strcmp (suffix, "xpd") &&
1662: strcmp (suffix, "ps") &&
1663: strcmp (suffix, "au") &&
1664: strcmp (suffix, "html") &&
1665: strcmp (suffix, "htm") &&
1666: strcmp (suffix, "shtml") &&
1667: strcmp (suffix, "xht") &&
1668: strcmp (suffix, "xhtm") &&
1669: strcmp (suffix, "xhtml") &&
1670: strcmp (suffix, "txt") &&
1671: strcmp (suffix, "css") &&
1672: strcmp (suffix, "eps"))
1.22 cvs 1673: return (FALSE);
1674: else
1675: return (TRUE);
1676: }
1677: else
1678: return (FALSE);
1679: }
1680:
1681:
1682: /*----------------------------------------------------------------------
1.24 cvs 1683: ChopURL
1684: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1685: If inputURL is bigger than that size, outputURL receives
1686: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1687: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1688: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1689: copied into outputURL.
1690: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1691: chars.
1692: ----------------------------------------------------------------------*/
1.106 cvs 1693: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1694: {
1695: int len;
1.9 cvs 1696:
1.106 cvs 1697: len = strlen (inputURL);
1.24 cvs 1698: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1699: strcpy (outputURL, inputURL);
1.24 cvs 1700: else
1701: /* make a truncated urlName on the status window */
1702: {
1.106 cvs 1703: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1704: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1705: strcat (outputURL, "...");
1706: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1707: }
1.25 cvs 1708: }
1709:
1710:
1711: /*----------------------------------------------------------------------
1712: scan
1.47 cvs 1713: Scan a filename for its constituents
1.25 cvs 1714: -----------------------------------
1715:
1716: On entry,
1717: name points to a document name which may be incomplete.
1718: On exit,
1719: absolute or relative may be nonzero (but not both).
1720: host, fragment and access may be nonzero if they were specified.
1721: Any which are nonzero point to zero terminated strings.
1722: ----------------------------------------------------------------------*/
1.106 cvs 1723: static void scan (char *name, HTURI *parts)
1.25 cvs 1724: {
1.106 cvs 1725: char * p;
1726: char * after_access = name;
1.32 cvs 1727:
1.43 cvs 1728: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1729: /* Look for fragment identifier */
1.106 cvs 1730: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1731: {
1.106 cvs 1732: *p++ = '\0';
1.28 cvs 1733: parts->fragment = p;
1.25 cvs 1734: }
1735:
1.28 cvs 1736: for (p=name; *p; p++)
1737: {
1.106 cvs 1738: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1739: break;
1.106 cvs 1740: if (*p == ':')
1.28 cvs 1741: {
1742: *p = 0;
1743: parts->access = after_access; /* Scheme has been specified */
1744:
1745: /* The combination of gcc, the "-O" flag and the HP platform is
1746: unhealthy. The following three lines is a quick & dirty fix, but is
1747: not recommended. Rather, turn off "-O". */
1748:
1749: /* after_access = p;*/
1750: /* while (*after_access == 0)*/
1751: /* after_access++;*/
1752: after_access = p+1;
1.106 cvs 1753: if (!strcasecmp("URL", parts->access))
1.28 cvs 1754: /* Ignore IETF's URL: pre-prefix */
1755: parts->access = NULL;
1756: else
1.25 cvs 1757: break;
1758: }
1759: }
1760:
1761: p = after_access;
1.43 cvs 1762: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1763: {
1.43 cvs 1764: if (p[1] == URL_SEP)
1.28 cvs 1765: {
1.25 cvs 1766: parts->host = p+2; /* host has been specified */
1.28 cvs 1767: *p = 0; /* Terminate access */
1768: /* look for end of host name if any */
1.106 cvs 1769: p = strchr (parts->host, URL_SEP);
1.28 cvs 1770: if (p)
1771: {
1.106 cvs 1772: *p = EOS; /* Terminate host */
1.25 cvs 1773: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1774: }
1775: }
1776: else
1777: /* Root found but no host */
1778: parts->absolute = p+1;
1779: }
1780: else
1781: {
1.25 cvs 1782: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1783: }
1.25 cvs 1784: }
1785:
1786:
1787: /*----------------------------------------------------------------------
1.28 cvs 1788: AmayaParseUrl: parse a Name relative to another name
1789:
1790: This returns those parts of a name which are given (and requested)
1791: substituting bits from the related name where necessary.
1.25 cvs 1792:
1.28 cvs 1793: On entry,
1.25 cvs 1794: aName A filename given
1795: relatedName A name relative to which aName is to be parsed. Give
1796: it an empty string if aName is absolute.
1797: wanted A mask for the bits which are wanted.
1798:
1.28 cvs 1799: On exit,
1.25 cvs 1800: returns A pointer to a malloc'd string which MUST BE FREED
1801: ----------------------------------------------------------------------*/
1.106 cvs 1802: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1803: {
1804: char *return_value;
1805: char result[MAX_LENGTH];
1806: char name[MAX_LENGTH];
1807: char rel[MAX_LENGTH];
1808: char *p, *access;
1.29 cvs 1809: HTURI given, related;
1810: int len;
1.106 cvs 1811: char used_sep;
1812: char *used_str;
1.32 cvs 1813:
1.106 cvs 1814: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1815: {
1.106 cvs 1816: used_str = DIR_STR;
1817: used_sep = DIR_SEP;
1.33 cvs 1818: }
1.32 cvs 1819: else
1.33 cvs 1820: {
1.106 cvs 1821: used_str = URL_STR;
1822: used_sep = URL_SEP;
1.33 cvs 1823: }
1.32 cvs 1824:
1.29 cvs 1825: /* Make working copies of input strings to cut up: */
1826: return_value = NULL;
1827: result[0] = 0; /* Clear string */
1.169 quint 1828: rel[0] = EOS;
1829: strncpy (name, aName, MAX_LENGTH - 1);
1830: name[MAX_LENGTH - 1] = EOS;
1831: if (relatedName != NULL)
1832: {
1833: strncpy (rel, relatedName, MAX_LENGTH - 1);
1834: rel[MAX_LENGTH - 1] = EOS;
1835: }
1.29 cvs 1836: else
1.106 cvs 1837: relatedName[0] = EOS;
1.29 cvs 1838:
1839: scan (name, &given);
1840: scan (rel, &related);
1841: access = given.access ? given.access : related.access;
1842: if (wanted & AMAYA_PARSE_ACCESS)
1843: if (access)
1844: {
1.106 cvs 1845: strcat (result, access);
1.29 cvs 1846: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1847: strcat (result, ":");
1.29 cvs 1848: }
1849:
1850: if (given.access && related.access)
1851: /* If different, inherit nothing. */
1.106 cvs 1852: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1853: {
1854: related.host = 0;
1855: related.absolute = 0;
1856: related.relative = 0;
1857: related.fragment = 0;
1858: }
1859:
1860: if (wanted & AMAYA_PARSE_HOST)
1861: if(given.host || related.host)
1862: {
1863: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1864: strcat (result, "//");
1865: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1866: }
1867:
1868: if (given.host && related.host)
1869: /* If different hosts, inherit no path. */
1.106 cvs 1870: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1871: {
1872: related.absolute = 0;
1873: related.relative = 0;
1874: related.fragment = 0;
1875: }
1876:
1877: if (wanted & AMAYA_PARSE_PATH)
1878: {
1879: if (given.absolute)
1880: {
1881: /* All is given */
1882: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1883: strcat (result, used_str);
1884: strcat (result, given.absolute);
1.25 cvs 1885: }
1.29 cvs 1886: else if (related.absolute)
1887: {
1888: /* Adopt path not name */
1.106 cvs 1889: strcat (result, used_str);
1890: strcat (result, related.absolute);
1.29 cvs 1891: if (given.relative)
1892: {
1893: /* Search part? */
1.106 cvs 1894: p = strchr (result, '?');
1.29 cvs 1895: if (!p)
1.106 cvs 1896: p=result+strlen(result)-1;
1.33 cvs 1897: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1898: /* Remove filename */
1899: p[1]=0;
1900: /* Add given one */
1.106 cvs 1901: strcat (result, given.relative);
1.25 cvs 1902: }
1903: }
1.29 cvs 1904: else if (given.relative)
1905: /* what we've got */
1.106 cvs 1906: strcat (result, given.relative);
1.29 cvs 1907: else if (related.relative)
1.106 cvs 1908: strcat (result, related.relative);
1.29 cvs 1909: else
1910: /* No inheritance */
1.106 cvs 1911: strcat (result, used_str);
1.25 cvs 1912: }
1.29 cvs 1913:
1914: if (wanted & AMAYA_PARSE_ANCHOR)
1915: if (given.fragment || related.fragment)
1916: {
1917: if (given.absolute && given.fragment)
1918: {
1919: /*Fixes for relURLs...*/
1920: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1921: strcat (result, "#");
1922: strcat (result, given.fragment);
1.29 cvs 1923: }
1924: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1925: strcat (result, "");
1.29 cvs 1926: else
1927: {
1.110 cvs 1928: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1929: strcat (result, "#");
1.110 cvs 1930: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1931: }
1932: }
1.106 cvs 1933: len = strlen (result);
1.171 gully 1934: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 1935: strcpy (return_value, result);
1.29 cvs 1936: return (return_value); /* exactly the right length */
1.25 cvs 1937: }
1938:
1939: /*----------------------------------------------------------------------
1940: HTCanon
1941: Canonicalizes the URL in the following manner starting from the host
1942: pointer:
1943:
1944: 1) The host name is converted to lowercase
1945: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1946:
1947: Return: OK The position of the current path part of the URL
1948: which might be the old one or a new one.
1949:
1950: ----------------------------------------------------------------------*/
1.106 cvs 1951: static char *HTCanon (char **filename, char *host)
1952: {
1953: char *newname = NULL;
1954: char used_sep;
1955: char *path;
1956: char *strptr;
1957: char *port;
1958: char *access = host-3;
1959:
1960: if (*filename && strchr (*filename, URL_SEP))
1961: used_sep = URL_SEP;
1.33 cvs 1962: else
1.106 cvs 1963: used_sep = DIR_SEP;
1.32 cvs 1964:
1.110 cvs 1965: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1966: access--;
1.110 cvs 1967: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1968: path = host + strlen (host);
1969: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1970: host = strptr;
1.110 cvs 1971: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1972: port = NULL;
1.25 cvs 1973:
1974: strptr = host; /* Convert to lower-case */
1.82 cvs 1975: while (strptr < path)
1.33 cvs 1976: {
1.123 vatton 1977: *strptr = tolower (*strptr);
1.82 cvs 1978: strptr++;
1.33 cvs 1979: }
1.25 cvs 1980:
1981: /* Does the URL contain a full domain name? This also works for a
1982: numerical host name. The domain name is already made lower-case
1983: and without a trailing dot. */
1984: {
1.106 cvs 1985: char *dot = port ? port : path;
1986: if (dot > *filename && *--dot == '.')
1.33 cvs 1987: {
1.106 cvs 1988: char *orig = dot;
1989: char *dest = dot + 1;
1.82 cvs 1990: while ((*orig++ = *dest++));
1991: if (port) port--;
1.33 cvs 1992: path--;
1.25 cvs 1993: }
1994: }
1995: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 1996: if (port)
1997: {
1.82 cvs 1998: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 1999: {
2000: if (!newname)
2001: {
1.106 cvs 2002: char *orig = port;
2003: char *dest = port + 1;
1.82 cvs 2004: while ((*orig++ = *dest++));
1.33 cvs 2005: }
2006: }
1.106 cvs 2007: else if ((!strncmp (access, "http", 4) &&
2008: (*(port + 1) == '8' &&
2009: *(port+2) == '0' &&
1.82 cvs 2010: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 2011: (!strncmp (access, "gopher", 6) &&
2012: (*(port+1) == '7' &&
2013: *(port+2) == '0' &&
1.82 cvs 2014: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 2015: (!strncmp (access, "ftp", 3) &&
2016: (*(port+1) == '2' &&
2017: *(port + 2) == '1' &&
1.82 cvs 2018: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 2019: if (!newname)
2020: {
1.106 cvs 2021: char *orig = port;
2022: char *dest = port + 3;
1.33 cvs 2023: while((*orig++ = *dest++));
2024: /* Update path position, Henry Minsky */
2025: path -= 3;
1.25 cvs 2026: }
1.33 cvs 2027: }
2028: else if (newname)
1.106 cvs 2029: strncat (newname, port, (int) (path - port));
1.33 cvs 2030: }
1.25 cvs 2031:
1.33 cvs 2032: if (newname)
2033: {
1.106 cvs 2034: char *newpath = newname + strlen (newname);
2035: strcat (newname, path);
1.25 cvs 2036: path = newpath;
1.28 cvs 2037: /* Free old copy */
2038: TtaFreeMemory(*filename);
1.25 cvs 2039: *filename = newname;
1.33 cvs 2040: }
1.25 cvs 2041: return path;
2042: }
2043:
2044:
2045: /*----------------------------------------------------------------------
1.29 cvs 2046: SimplifyUrl: simplify a URI
1.32 cvs 2047: A URI is allowed to contain the sequence xxx/../ which may be
2048: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 2049: Simplification helps us recognize duplicate URIs.
1.25 cvs 2050:
1.28 cvs 2051: Thus, /etc/junk/../fred becomes /etc/fred
2052: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 2053:
1.28 cvs 2054: but we should NOT change
2055: http://fred.xxx.edu/../..
1.25 cvs 2056:
2057: or ../../albert.html
2058:
1.28 cvs 2059: In order to avoid empty URLs the following URLs become:
1.25 cvs 2060:
2061: /fred/.. becomes /fred/..
2062: /fred/././.. becomes /fred/..
2063: /fred/.././junk/.././ becomes /fred/..
2064:
1.28 cvs 2065: If more than one set of `://' is found (several proxies in cascade) then
2066: only the part after the last `://' is simplified.
1.25 cvs 2067:
1.28 cvs 2068: Returns: A string which might be the old one or a new one.
1.25 cvs 2069: ----------------------------------------------------------------------*/
1.106 cvs 2070: void SimplifyUrl (char **url)
2071: {
2072: char *path;
2073: char *access;
2074: char *newptr;
2075: char *p;
2076: char *orig, *dest, *end;
1.28 cvs 2077:
1.106 cvs 2078: char used_sep;
1.77 cvs 2079: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
2080: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 2081:
1.28 cvs 2082: if (!url || !*url)
2083: return;
2084:
1.106 cvs 2085: if (strchr (*url, URL_SEP))
2086: used_sep = URL_SEP;
1.32 cvs 2087: else
1.106 cvs 2088: used_sep = DIR_SEP;
1.32 cvs 2089:
1.77 cvs 2090: /* should we simplify double dot? */
2091: path = *url;
1.106 cvs 2092: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 2093: ddot_simplify = FALSE;
2094: else
2095: ddot_simplify = TRUE;
2096:
1.28 cvs 2097: /* Find any scheme name */
1.106 cvs 2098: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 2099: {
2100: /* Find host name */
1.28 cvs 2101: access = *url;
1.123 vatton 2102: while (access < path && (*access = tolower (*access)))
1.82 cvs 2103: access++;
1.28 cvs 2104: path += 3;
1.106 cvs 2105: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 2106: /* For proxies */
1.106 cvs 2107: path = newptr + 3;
1.82 cvs 2108: /* We have a host name */
1.84 cvs 2109: path = HTCanon (url, path);
1.25 cvs 2110: }
1.106 cvs 2111: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 2112: path += 2;
2113: else
2114: path = *url;
1.84 cvs 2115: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 2116: /* Some URLs start //<foo> */
2117: path += 1;
1.94 cvs 2118: else if (IsFilePath (path))
2119: {
2120: /* doesn't need to do anything more */
2121: return;
2122: }
1.106 cvs 2123: else if (!strncmp (path, "news:", 5))
1.28 cvs 2124: {
1.106 cvs 2125: newptr = strchr (path+5, '@');
1.28 cvs 2126: if (!newptr)
2127: newptr = path + 5;
2128: while (*newptr)
2129: {
2130: /* Make group or host lower case */
1.123 vatton 2131: *newptr = tolower (*newptr);
1.28 cvs 2132: newptr++;
1.25 cvs 2133: }
1.28 cvs 2134: /* Doesn't need to do any more */
2135: return;
1.25 cvs 2136: }
1.130 cheyroul 2137:
1.126 cheyroul 2138:
1.28 cvs 2139: if ((p = path))
2140: {
1.106 cvs 2141: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
2142: (end = strchr (path, '#'))))
2143: end = path + strlen (path);
1.28 cvs 2144:
2145: /* Parse string second time to simplify */
2146: p = path;
2147: while (p < end)
2148: {
1.110 cvs 2149: /* if we're pointing to a char, it's safe to reactivate the
2150: ../ convertion */
1.106 cvs 2151: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 2152: ddot_simplify = TRUE;
2153:
1.33 cvs 2154: if (*p==used_sep)
1.28 cvs 2155: {
1.106 cvs 2156: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 2157: {
2158: orig = p + 1;
1.84 cvs 2159: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 2160: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 2161: end = orig - 1;
2162: }
1.106 cvs 2163: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 2164: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 2165: {
2166: newptr = p;
1.52 cvs 2167: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
2168: if (*newptr == used_sep)
2169: orig = newptr + 1;
1.28 cvs 2170: else
1.52 cvs 2171: orig = newptr;
2172:
2173: dest = (*(p+3) != used_sep) ? p+3 : p+4;
2174: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
2175: end = orig-1;
2176: /* Start again with prev slash */
2177: p = newptr;
1.28 cvs 2178: }
1.33 cvs 2179: else if (*(p+1) == used_sep)
1.28 cvs 2180: {
1.33 cvs 2181: while (*(p+1) == used_sep)
1.28 cvs 2182: {
2183: orig = p;
2184: dest = p + 1;
2185: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2186: end = orig-1;
2187: }
2188: }
2189: else
1.25 cvs 2190: p++;
1.28 cvs 2191: }
2192: else
2193: p++;
1.25 cvs 2194: }
2195: }
1.51 cvs 2196: /*
2197: ** Check for host/../.. kind of things
2198: */
1.106 cvs 2199: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 2200: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 2201: *(path+1) = EOS;
1.28 cvs 2202: return;
2203: }
2204:
2205:
2206: /*----------------------------------------------------------------------
1.96 cvs 2207: NormalizeFile normalizes local names.
1.28 cvs 2208: Return TRUE if target and src differ.
2209: ----------------------------------------------------------------------*/
1.106 cvs 2210: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2211: {
1.184 ! gully 2212: #ifndef _WINDOWS
1.106 cvs 2213: char *s;
1.93 cvs 2214: int i;
1.184 ! gully 2215: #endif /* _WINDOWS */
1.82 cvs 2216: ThotBool change;
1.90 cvs 2217: int start_index; /* the first char that we'll copy */
1.28 cvs 2218:
1.54 cvs 2219: change = FALSE;
1.90 cvs 2220: start_index = 0;
2221:
1.106 cvs 2222: if (!src || src[0] == EOS)
1.96 cvs 2223: {
1.106 cvs 2224: target[0] = EOS;
1.96 cvs 2225: return FALSE;
2226: }
1.90 cvs 2227:
2228: /* @@ do I need file: or file:/ here? */
1.106 cvs 2229: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 2230: {
1.90 cvs 2231: /* remove the prefix file: */
2232: start_index += 5;
2233:
2234: /* remove the localhost prefix */
1.106 cvs 2235: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 2236: start_index += 11;
2237:
2238: /* remove the first two slashes in / / /path */
2239: while (src[start_index] &&
1.106 cvs 2240: src[start_index] == '/'
2241: && src[start_index + 1] == '/')
1.94 cvs 2242: start_index++;
2243:
1.184 ! gully 2244: #ifdef _WINDOWS
1.94 cvs 2245: /* remove any extra slash before the drive name */
1.106 cvs 2246: if (src[start_index] == '/'
2247: &&src[start_index+2] == ':')
1.94 cvs 2248: start_index++;
1.184 ! gully 2249: #endif /* _WINDOWS */
1.90 cvs 2250:
1.106 cvs 2251: if (src[start_index] == EOS)
1.90 cvs 2252: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 2253: strcpy (target, DIR_STR);
1.90 cvs 2254: else
1.97 cvs 2255: /* as we're inside a file: URL, we'll apply all the convertions
2256: we know */
2257: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 2258:
2259: change = TRUE;
2260: }
1.97 cvs 2261: else if (convertion != AM_CONV_NONE)
1.96 cvs 2262: {
2263: /* we are following a "local" relative link, we do all the
2264: convertions except for the HOME_DIR ~ one */
1.97 cvs 2265: CleanCopyFileURL (target, src, convertion);
1.28 cvs 2266: }
1.184 ! gully 2267: #ifndef _WINDOWS
1.106 cvs 2268: else if (src[0] == '~')
1.53 cvs 2269: {
1.96 cvs 2270: /* it must be a URL typed in a text input field */
2271: /* do the HOME_DIR ~ substitution */
1.82 cvs 2272: s = TtaGetEnvString ("HOME");
1.106 cvs 2273: strcpy (target, s);
1.90 cvs 2274: #if 0
1.96 cvs 2275: /* JK: invalidated this part of the code as it's simpler
2276: to add the DIR_SEP whenever we have something to add
2277: to the path rather than adding it systematically */
1.106 cvs 2278: if (src[1] != DIR_SEP)
2279: strcat (target, DIR_STR);
1.90 cvs 2280: #endif
1.106 cvs 2281: i = strlen (target);
2282: strcpy (&target[i], &src[1]);
1.54 cvs 2283: change = TRUE;
1.53 cvs 2284: }
1.184 ! gully 2285: #endif /* _WINDOWS */
1.28 cvs 2286: else
1.96 cvs 2287: /* leave it as it is */
1.106 cvs 2288: strcpy (target, src);
1.96 cvs 2289:
1.28 cvs 2290: /* remove /../ and /./ */
1.29 cvs 2291: SimplifyUrl (&target);
1.54 cvs 2292: if (!change)
1.106 cvs 2293: change = strcmp (src, target);
1.28 cvs 2294: return (change);
1.25 cvs 2295: }
2296:
1.28 cvs 2297:
1.25 cvs 2298: /*----------------------------------------------------------------------
1.31 cvs 2299: MakeRelativeURL: make relative name
1.25 cvs 2300:
1.28 cvs 2301: This function creates and returns a string which gives an expression of
2302: one address as related to another. Where there is no relation, an absolute
2303: address is retured.
1.25 cvs 2304:
1.28 cvs 2305: On entry,
1.25 cvs 2306: Both names must be absolute, fully qualified names of nodes
2307: (no fragment bits)
2308:
1.28 cvs 2309: On exit,
1.25 cvs 2310: The return result points to a newly allocated name which, if
2311: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2312: The caller is responsible for freeing the resulting name later.
2313: ----------------------------------------------------------------------*/
1.106 cvs 2314: char *MakeRelativeURL (char *aName, char *relatedName)
2315: {
2316: char *return_value;
2317: char result[MAX_LENGTH];
2318: char *p;
2319: char *q;
2320: char *after_access;
2321: char *last_slash = NULL;
2322: int slashes, levels, len;
1.184 ! gully 2323: #ifdef _WINDOWS
1.44 cvs 2324: int ndx;
1.184 ! gully 2325: #endif /* _WINDOWS */
1.44 cvs 2326:
1.29 cvs 2327: if (aName == NULL || relatedName == NULL)
2328: return (NULL);
2329:
2330: slashes = 0;
2331: after_access = NULL;
2332: p = aName;
2333: q = relatedName;
1.147 vatton 2334: len = 0;
2335: for (; *p && !strncasecmp (p, q, 1); p++, q++, len++)
1.27 cvs 2336: {
2337: /* Find extent of match */
1.106 cvs 2338: if (*p == ':')
1.146 cvs 2339: {
1.168 cvs 2340: after_access = p + 1;
1.184 ! gully 2341: #ifdef _WINDOWS
1.168 cvs 2342: if (len == 1)
2343: {
2344: /* it's a local Windows path like c:... */
2345: slashes+=2;
2346: }
1.184 ! gully 2347: #endif /* _WINDOWS */
1.168 cvs 2348: }
2349: if (*p == DIR_SEP)
2350: {
2351: /* memorize the last slash position and count them */
2352: last_slash = p;
1.147 vatton 2353: slashes++;
1.146 cvs 2354: }
1.25 cvs 2355: }
2356:
1.31 cvs 2357: /* q, p point to the first non-matching character or zero */
1.106 cvs 2358: if (*q == EOS)
1.31 cvs 2359: {
2360: /* New name is a subset of the related name */
2361: /* exactly the right length */
1.106 cvs 2362: len = strlen (p);
1.171 gully 2363: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 2364: strcpy (return_value, p);
1.31 cvs 2365: }
2366: else if ((slashes < 2 && after_access == NULL)
2367: || (slashes < 3 && after_access != NULL))
1.168 cvs 2368: {
1.31 cvs 2369: /* Two names whitout common path */
2370: /* exactly the right length */
1.106 cvs 2371: len = strlen (aName);
1.171 gully 2372: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 2373: strcpy (return_value, aName);
1.31 cvs 2374: }
2375: else
2376: {
2377: /* Some path in common */
1.106 cvs 2378: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2379: /* just the same server */
1.106 cvs 2380: strcpy (result, last_slash);
1.31 cvs 2381: else
2382: {
2383: levels= 0;
1.106 cvs 2384: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2385: if (*q == DIR_SEP)
2386: levels++;
2387:
1.106 cvs 2388: result[0] = EOS;
1.31 cvs 2389: for (;levels; levels--)
1.106 cvs 2390: strcat (result, "../");
2391: strcat (result, last_slash+1);
1.31 cvs 2392: }
1.52 cvs 2393:
2394: if (!*result)
1.106 cvs 2395: strcat (result, "./");
1.52 cvs 2396:
1.31 cvs 2397: /* exactly the right length */
1.106 cvs 2398: len = strlen (result);
1.171 gully 2399: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 2400: strcpy (return_value, result);
1.52 cvs 2401:
1.25 cvs 2402: }
1.184 ! gully 2403: #ifdef _WINDOWS
1.106 cvs 2404: len = strlen (return_value);
1.168 cvs 2405: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2406: if (return_value[ndx] == '\\')
2407: return_value[ndx] = '/' ;
1.184 ! gully 2408: #endif /* _WINDOWS */
1.29 cvs 2409: return (return_value);
1.24 cvs 2410: }
1.35 cvs 2411:
1.104 kahan 2412: /*----------------------------------------------------------------------
2413: AM_GetFileSize
2414: Returns TRUE and the filesize in the 2nd parameter.
2415: Otherwise, in case of a system error, returns FALSE, with a
2416: filesize of 0L.
2417: ---------------------------------------------------------------------*/
1.106 cvs 2418: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2419: {
1.106 cvs 2420: ThotFileHandle handle = ThotFile_BADHANDLE;
2421: ThotFileInfo info;
1.35 cvs 2422:
1.104 kahan 2423: *file_size = 0L;
2424: if (!TtaFileExist (filename))
2425: return FALSE;
2426:
2427: handle = TtaFileOpen (filename, ThotFile_READWRITE);
2428: if (handle == ThotFile_BADHANDLE)
2429: /* ThotFile_BADHANDLE */
2430: return FALSE;
2431: if (TtaFileStat (handle, &info) == 0)
2432: /* bad stat */
2433: info.size = 0L;
2434: TtaFileClose (handle);
2435: *file_size = (unsigned long) info.size;
2436: return TRUE;
2437: }
1.139 kahan 2438:
2439: /*----------------------------------------------------------------------
2440: AM_UseXHTMLMimeType
2441: Returns TRUE if the user has configured Amaya to use this MIME type,
2442: FALSE otherwise.
2443: ---------------------------------------------------------------------*/
2444: ThotBool AM_UseXHTMLMimeType (void)
2445: {
2446: ThotBool xhtml_mimetype;
2447:
2448: /* does the user wants to use the new MIME type? */
2449: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2450:
2451: return (xhtml_mimetype);
1.152 kahan 2452: }
2453:
1.154 kahan 2454:
2455: /********************************************
2456: The following routines were adapted from the GNU libc functions
2457: for generating a tmpnam.
2458: *********************************************/
2459:
2460: /* These are the characters used in temporary filenames. */
2461: static const char letters[] =
2462: "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
2463:
2464: /* Generate a temporary file name based on TMPL. TMPL must match the
2465: rules for mk[s]temp (i.e. end in "XXXXXX"). The name constructed
2466: does not exist at the time of the call to __gen_tempname. TMPL is
2467: overwritten with the result.
2468:
2469: We use a clever algorithm to get hard-to-predict names. */
2470: void
2471: AM_gen_tempname (char *tmpl)
2472: {
2473: int len;
2474: char *XXXXXX;
2475: static uint64_t value;
2476: uint64_t random_time_bits;
2477: unsigned int count;
2478: int save_errno = errno;
2479: struct stat st;
2480:
2481: /* A lower bound on the number of temporary files to attempt to
2482: generate. The maximum total number of temporary file names that
2483: can exist for a given template is 62**6. It should never be
2484: necessary to try all these combinations. Instead if a reasonable
2485: number of names is tried (we define reasonable as 62**3) fail to
2486: give the system administrator the chance to remove the problems. */
2487: unsigned int attempts_min = 62 * 62 * 62;
2488:
2489: /* The number of times to attempt to generate a temporary file. To
2490: conform to POSIX, this must be no smaller than TMP_MAX. */
2491: unsigned int attempts = attempts_min < TMP_MAX ? TMP_MAX : attempts_min;
2492:
2493: len = strlen (tmpl);
2494: if (len < 6 || strcmp (&tmpl[len - 6], "XXXXXX"))
2495: {
2496: /* @@ JK ? */
2497: errno = EINVAL;
2498: return;
2499: }
2500:
2501: /* This is where the Xs start. */
2502: XXXXXX = &tmpl[len - 6];
2503:
2504: /* Get some more or less random data. */
2505: #ifdef RANDOM_BITS
2506: RANDOM_BITS (random_time_bits);
2507: #else
2508: # if HAVE_GETTIMEOFDAY || _LIBC
2509: {
2510: struct timeval tv;
2511: gettimeofday (&tv, NULL);
2512: random_time_bits = ((uint64_t) tv.tv_usec << 16) ^ tv.tv_sec;
2513: }
2514: # else
2515: random_time_bits = time (NULL);
2516: # endif
2517: #endif
2518: value += random_time_bits ^ getpid ();
2519:
2520: for (count = 0; count < attempts; value += 7777, ++count)
2521: {
2522: uint64_t v = value;
2523:
2524: /* Fill in the random bits. */
2525: XXXXXX[0] = letters[v % 62];
2526: v /= 62;
2527: XXXXXX[1] = letters[v % 62];
2528: v /= 62;
2529: XXXXXX[2] = letters[v % 62];
2530: v /= 62;
2531: XXXXXX[3] = letters[v % 62];
2532: v /= 62;
2533: XXXXXX[4] = letters[v % 62];
2534: v /= 62;
2535: XXXXXX[5] = letters[v % 62];
2536:
2537: /* This case is backward from the other three. AM_gen_tempname
2538: succeeds if __xstat fails because the name does not exist.
2539: Note the continue to bypass the common logic at the bottom
2540: of the loop. */
2541: if (stat (tmpl, &st) < 0)
2542: break;
2543:
2544: continue;
2545: }
2546:
2547: if (count == attempts || errno != ENOENT)
2548: tmpl[0] = EOS;
2549: else
2550: errno = save_errno;
2551:
2552: return;
2553: }
2554:
2555: #define JOSE 1
2556:
1.152 kahan 2557: /*-----------------------------------------------------------------------
2558: GetTempName
2559: Front end to the Unix tempnam function, which is independent of the
2560: value of the TMPDIR env value
2561: Returns a dynamically allocated string with a tempname. The user
2562: must free this memory.
2563: -----------------------------------------------------------------------*/
2564: char *GetTempName (const char *dir, const char *prefix)
2565: {
1.154 kahan 2566: #ifdef JOSE
2567:
1.162 kahan 2568: static char tmpbufmem[PATH_MAX + 1];
1.154 kahan 2569: int len;
2570: int i;
2571:
1.155 cvs 2572: if (!dir || *dir == EOS || !TtaDirExists (dir))
1.154 kahan 2573: return NULL;
2574:
1.162 kahan 2575: /* make sure that the name is no bigger than PATH_MAX + the 6 tempname chars we
2576: will add */
1.154 kahan 2577:
1.156 cvs 2578: len = strlen (dir);
1.162 kahan 2579: if (len + 6 > PATH_MAX)
1.154 kahan 2580: return NULL;
2581:
2582: /* copy the dir name, and add a DIR_SEP if it's missing */
2583: if (dir[strlen (dir) - 1] == DIR_SEP)
2584: strcpy (tmpbufmem, dir);
2585: else
1.156 cvs 2586: {
1.154 kahan 2587: sprintf (tmpbufmem, "%s%c", dir, DIR_SEP);
1.156 cvs 2588: len++;
2589: }
1.154 kahan 2590:
1.161 kahan 2591: /* copy the prefix (no more than L_tmpnam chars, to respect POSIX). Save
1.156 cvs 2592: space for the 6 X and EOS chars that will become the random bits */
2593: if (prefix)
2594: {
2595: i = 0;
1.167 cheyroul 2596: while (prefix[i] != EOS && i < L_tmpnam - 8)
1.156 cvs 2597: tmpbufmem[len++] = prefix[i++];
2598: tmpbufmem[len] = EOS;
2599: }
2600:
2601: /* Add the 6 X chars */
2602: len = strlen (tmpbufmem);
2603: i = 0;
2604: while (i < 6)
2605: {
2606: tmpbufmem[len++] = 'X';
2607: i++;
2608: }
2609: tmpbufmem[len] = EOS;
1.154 kahan 2610:
2611: AM_gen_tempname (tmpbufmem);
2612:
2613: if (tmpbufmem[0] == EOS)
2614: return NULL;
2615: else
2616: return (TtaStrdup (tmpbufmem));
2617:
2618: #else
1.152 kahan 2619: char *tmpdir;
1.181 vatton 2620: char *tmp = NULL;
1.152 kahan 2621: char *name = NULL;
2622:
2623: /* save the value of TMPDIR */
1.181 vatton 2624: tmpdir = getenv ("TMPDIR");
1.152 kahan 2625: if (tmpdir)
2626: {
1.181 vatton 2627: /* remove TMPDIR from the environment */
2628: tmp = TtaGetMemory (strlen (tmpdir) + 20);
2629: sprintf (tmp, "TMPDIR=");
1.184 ! gully 2630: #ifdef _WINDOWS
1.152 kahan 2631: _putenv (tmp);
2632: #else
2633: putenv (tmp);
1.184 ! gully 2634: #endif /* _WINDOWS */
1.181 vatton 2635: /* prepare the string to restore the value of TMPDIR */
2636: strrcat (tmp, tmpdir);
1.152 kahan 2637: }
2638:
2639: /* create the tempname */
1.184 ! gully 2640: #ifdef _WINDOWS
1.152 kahan 2641: /* Under Windows, _tempnam returns the same name until the file is created */
2642: {
2643: char *altprefix;
2644: name = tmpnam (NULL); /* get a possibly unique string */
1.181 vatton 2645: altprefix = TtaGetMemory (strlen (prefix) + strlen(name) + 1);
1.152 kahan 2646: sprintf (altprefix, "%s%s", prefix, name + strlen(_P_tmpdir));
2647: name = _tempnam (dir, altprefix); /* get a name that isn't yet in use */
2648: TtaFreeMemory (altprefix);
2649: }
2650: #else
2651: name = tempnam (dir, prefix);
1.184 ! gully 2652: #endif /* _WINDOWS */
1.152 kahan 2653:
2654: if (tmpdir)
2655: {
1.181 vatton 2656: /* restore the value of TMPDIR */
1.184 ! gully 2657: #ifdef _WINDOWS
1.152 kahan 2658: _putenv (tmpdir);
2659: #else
2660: putenv (tmpdir);
1.184 ! gully 2661: #endif /* _WINDOWS */
1.181 vatton 2662: TtaFreeMemory (tmpdir);
1.152 kahan 2663: }
2664: return (name);
1.154 kahan 2665: #endif
1.139 kahan 2666: }
Webmaster