Annotation of Amaya/amaya/AHTURLTools.c, revision 1.189
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") ||
1.186 quint 727: !strcasecmp (suffix, "ddl") ||
728: !strcasecmp (suffix, "mpg") ||
729: !strcasecmp (suffix, "mpeg") ||
730: !strcasecmp (suffix, "wmv") ||
731: !strcasecmp (suffix, "wma") ||
1.136 cvs 732: !strcasecmp (suffix, "o"))
1.183 quint 733: ret = TRUE;
1.106 cvs 734: else if (!strcmp (suffix, "gz"))
1.103 cvs 735: {
736: /* take into account compressed files */
1.124 vatton 737: TtaExtractSuffix (temppath, suffix);
1.136 cvs 738: if (!strcasecmp (suffix, "exe") ||
739: !strcasecmp (suffix, "zip") ||
740: !strcasecmp (suffix, "ppt") ||
741: !strcasecmp (suffix, "pdf") ||
742: !strcasecmp (suffix, "ps") ||
743: !strcasecmp (suffix, "eps") ||
744: !strcasecmp (suffix, "tar") ||
745: !strcasecmp (suffix, "ddl") ||
1.186 quint 746: !strcasecmp (suffix, "mpg") ||
747: !strcasecmp (suffix, "mpeg") ||
748: !strcasecmp (suffix, "wmv") ||
749: !strcasecmp (suffix, "wma") ||
1.136 cvs 750: !strcasecmp (suffix, "o"))
1.183 quint 751: ret = TRUE;
1.103 cvs 752: else
1.183 quint 753: ret = FALSE;
1.103 cvs 754: }
755: else
1.183 quint 756: ret = FALSE;
757:
758: TtaFreeMemory (temppath);
759: TtaFreeMemory (suffix);
760: return (ret);
1.103 cvs 761: }
762:
763: /*----------------------------------------------------------------------
1.60 cvs 764: IsCSSName
765: returns TRUE if path points to an XML resource.
766: ----------------------------------------------------------------------*/
1.111 cvs 767: ThotBool IsCSSName (const char *path)
1.60 cvs 768: {
1.183 quint 769: char *temppath;
770: char *suffix;
771: ThotBool ret;
1.60 cvs 772:
773: if (!path)
1.183 quint 774: return (FALSE);
1.60 cvs 775:
1.183 quint 776: temppath = TtaStrdup ((char *)path);
777: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 778: TtaExtractSuffix (temppath, suffix);
1.60 cvs 779:
1.106 cvs 780: if (!strcasecmp (suffix, "css"))
1.183 quint 781: ret = TRUE;
1.106 cvs 782: else if (!strcmp (suffix, "gz"))
1.60 cvs 783: {
784: /* take into account compressed files */
1.124 vatton 785: TtaExtractSuffix (temppath, suffix);
1.106 cvs 786: if (!strcasecmp (suffix, "css"))
1.183 quint 787: ret = TRUE;
1.56 cvs 788: else
1.183 quint 789: ret = FALSE;
1.56 cvs 790: }
791: else
1.183 quint 792: ret = FALSE;
793:
794: TtaFreeMemory (temppath);
795: TtaFreeMemory (suffix);
796: return (ret);
1.56 cvs 797: }
798:
799: /*----------------------------------------------------------------------
1.177 kahan 800: MultipleBookmarks
801: returns the value of the multiple bookmarks environment variable.
802: (TRUE if enabled).
803: ----------------------------------------------------------------------*/
804: ThotBool MultipleBookmarks (void)
805: {
806: ThotBool multiple_bm;
807:
1.180 kahan 808: TtaGetEnvBoolean ("DISABLE_MULTIPLE_BM", &multiple_bm);
1.177 kahan 809:
1.180 kahan 810: return (multiple_bm == FALSE);
1.177 kahan 811: }
812:
813: /*----------------------------------------------------------------------
1.172 kahan 814: IsRDFName
815: returns TRUE if path points to an RDF resource.
816: ----------------------------------------------------------------------*/
817: ThotBool IsRDFName (const char *path)
818: {
1.183 quint 819: char *temppath;
820: char *suffix;
821: ThotBool ret;
1.172 kahan 822:
1.177 kahan 823: /* temporarily disabling this function */
824: if (!MultipleBookmarks ())
1.183 quint 825: return (FALSE);
1.177 kahan 826:
1.172 kahan 827: if (!path)
1.183 quint 828: return (FALSE);
1.172 kahan 829:
1.183 quint 830: temppath = TtaStrdup ((char *)path);
831: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.172 kahan 832: TtaExtractSuffix (temppath, suffix);
833:
834: if (!strcasecmp (suffix, "rdf"))
1.183 quint 835: ret = TRUE;
1.172 kahan 836: else if (!strcmp (suffix, "gz"))
837: {
838: /* take into account compressed files */
839: TtaExtractSuffix (temppath, suffix);
840: if (!strcasecmp (suffix, "rdf"))
1.183 quint 841: ret = TRUE;
1.172 kahan 842: else
1.183 quint 843: ret = FALSE;
1.172 kahan 844: }
845: else
1.183 quint 846: ret = FALSE;
847:
848: TtaFreeMemory (temppath);
849: TtaFreeMemory (suffix);
850: return (ret);
1.172 kahan 851: }
852:
853: /*----------------------------------------------------------------------
1.9 cvs 854: IsImageName
855: returns TRUE if path points to an image resource.
1.4 cvs 856: ----------------------------------------------------------------------*/
1.111 cvs 857: ThotBool IsImageName (const char *path)
1.106 cvs 858: {
1.183 quint 859: char *temppath;
860: char *suffix;
1.106 cvs 861: char nsuffix[MAX_LENGTH];
1.5 cvs 862: int i;
1.183 quint 863: ThotBool ret;
1.5 cvs 864:
865: if (!path)
1.13 cvs 866: return (FALSE);
1.5 cvs 867:
1.183 quint 868: temppath = TtaStrdup ((char *)path);
869: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 870: TtaExtractSuffix (temppath, suffix);
1.5 cvs 871:
872: /* Normalize the suffix */
873: i = 0;
1.106 cvs 874: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 875: {
1.123 vatton 876: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 877: i++;
878: }
1.106 cvs 879: nsuffix[i] = EOS;
880: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) ||
881: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) ||
882: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au")))
1.183 quint 883: ret = TRUE;
884: else
885: ret = FALSE;
886:
887: TtaFreeMemory (temppath);
888: TtaFreeMemory (suffix);
889: return (ret);
1.3 cvs 890: }
891:
1.4 cvs 892: /*----------------------------------------------------------------------
1.58 cvs 893: IsImageType
894: returns TRUE if type points to an image resource.
895: ----------------------------------------------------------------------*/
1.111 cvs 896: ThotBool IsImageType (const char *type)
1.58 cvs 897: {
1.183 quint 898: char *temptype;
1.58 cvs 899: int i;
1.183 quint 900: ThotBool ret;
1.58 cvs 901:
902: if (!type)
903: return (FALSE);
904:
1.183 quint 905: temptype = TtaStrdup ((char *)type);
1.58 cvs 906: /* Normalize the type */
907: i = 0;
1.106 cvs 908: while (temptype[i] != EOS)
1.58 cvs 909: {
910: temptype[i] = tolower (temptype[i]);
911: i++;
912: }
1.166 vatton 913: if (!strncmp (temptype, "image/", sizeof ("image/") - 1))
914: i = sizeof ("image/") - 1;
915: else
916: i = 0;
917: if (!strcmp (&temptype[i], "gif") || !strcmp (&temptype[i], "x-xbitmap") ||
918: !strcmp (&temptype[i], "x-xpixmap") || !strcmp (&temptype[i], "jpeg") ||
919: !strcmp (&temptype[i], "png"))
1.183 quint 920: ret = TRUE;
921: else
922: ret = FALSE;
923: TtaFreeMemory (temptype);
924: return (ret);
1.58 cvs 925: }
926:
927: /*----------------------------------------------------------------------
1.9 cvs 928: IsTextName
1.4 cvs 929: ----------------------------------------------------------------------*/
1.111 cvs 930: ThotBool IsTextName (const char *path)
1.106 cvs 931: {
1.183 quint 932: char *temppath;
933: char *suffix;
1.106 cvs 934: char nsuffix[MAX_LENGTH];
1.5 cvs 935: int i;
1.183 quint 936: ThotBool ret;
1.5 cvs 937:
938: if (!path)
1.13 cvs 939: return (FALSE);
1.5 cvs 940:
1.183 quint 941: temppath = TtaStrdup ((char *)path);
942: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 943: TtaExtractSuffix (temppath, suffix);
1.5 cvs 944:
945: /* Normalize the suffix */
946: i = 0;
1.106 cvs 947: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 948: {
1.25 cvs 949: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 950: i++;
951: }
1.106 cvs 952: nsuffix[i] = EOS;
1.5 cvs 953:
1.111 cvs 954: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.183 quint 955: ret = TRUE;
1.106 cvs 956: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 957: {
1.39 cvs 958: /* take into account compressed files */
1.124 vatton 959: TtaExtractSuffix (temppath, suffix);
1.13 cvs 960: /* Normalize the suffix */
961: i = 0;
1.106 cvs 962: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 963: {
1.25 cvs 964: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 965: i++;
966: }
1.106 cvs 967: nsuffix[i] = EOS;
1.111 cvs 968: if (!strcmp (nsuffix, "txt") || !strcmp (nsuffix, "dtd"))
1.183 quint 969: ret = TRUE;
1.13 cvs 970: else
1.183 quint 971: ret = FALSE;
1.13 cvs 972: }
973: else
1.183 quint 974: ret = FALSE;
975:
976: TtaFreeMemory (temppath);
977: TtaFreeMemory (suffix);
978: return (ret);
1.3 cvs 979: }
980:
1.4 cvs 981: /*----------------------------------------------------------------------
1.9 cvs 982: IsHTTPPath
983: returns TRUE if path is in fact an http URL.
1.4 cvs 984: ----------------------------------------------------------------------*/
1.112 cvs 985: ThotBool IsHTTPPath (const char *path)
1.3 cvs 986: {
1.5 cvs 987: if (!path)
988: return FALSE;
1.3 cvs 989:
1.106 cvs 990: if ((!strncmp (path, "http:", 5) != 0)
991: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
992: || !strncmp (path, "internal:", 9))
1.58 cvs 993: return TRUE;
994: return FALSE;
1.3 cvs 995: }
996:
1.4 cvs 997: /*----------------------------------------------------------------------
1.9 cvs 998: IsWithParameters
999: returns TRUE if url has a concatenated query string.
1.4 cvs 1000: ----------------------------------------------------------------------*/
1.133 vatton 1001: ThotBool IsWithParameters (const char *url)
1.3 cvs 1002: {
1.5 cvs 1003: int i;
1.3 cvs 1004:
1.9 cvs 1005: if ((!url) || (url[0] == EOS))
1.5 cvs 1006: return FALSE;
1.3 cvs 1007:
1.9 cvs 1008: i = strlen (url) - 1;
1009: while (i > 0 && url[i--] != '?')
1.5 cvs 1010: if (i < 0)
1011: return FALSE;
1.3 cvs 1012:
1.5 cvs 1013: /* There is a parameter */
1014: return TRUE;
1.3 cvs 1015: }
1016:
1.4 cvs 1017: /*----------------------------------------------------------------------
1.9 cvs 1018: IsW3Path
1019: returns TRUE if path is in fact a URL.
1.4 cvs 1020: ----------------------------------------------------------------------*/
1.133 vatton 1021: ThotBool IsW3Path (const char *path)
1.106 cvs 1022: {
1.170 quint 1023: if (path == NULL)
1024: return FALSE;
1.106 cvs 1025: if (strncmp (path, "http:", 5) &&
1026: strncmp (path, "ftp:", 4) &&
1027: strncmp (path, "telnet:", 7) &&
1028: strncmp (path, "wais:", 5) &&
1029: strncmp (path, "news:", 5) &&
1030: strncmp (path, "gopher:", 7) &&
1031: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 1032: strncmp (path, "archie:", 7) &&
1033: strncmp (path, "https:", 6))
1.72 cvs 1034: return FALSE;
1035: return TRUE;
1.3 cvs 1036: }
1037:
1.4 cvs 1038: /*----------------------------------------------------------------------
1.90 cvs 1039: IsFilePath
1040: returns TRUE if path is in fact a URL.
1041: ----------------------------------------------------------------------*/
1.133 vatton 1042: ThotBool IsFilePath (const char *path)
1.90 cvs 1043: {
1.106 cvs 1044: if (strncmp (path, "file:", 5))
1.90 cvs 1045: return FALSE;
1046: return TRUE;
1047: }
1048:
1049: /*----------------------------------------------------------------------
1.9 cvs 1050: IsValidProtocol
1051: returns true if the url protocol is supported by Amaya.
1.4 cvs 1052: ----------------------------------------------------------------------*/
1.133 vatton 1053: ThotBool IsValidProtocol (const char *url)
1.106 cvs 1054: {
1055: if (!strncmp (url, "http:", 5)
1056: || !strncmp (url, "internal:", 9)
1057: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 1058: /* experimental */
1.24 cvs 1059: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 1060: return (TRUE);
1.5 cvs 1061: else
1.8 cvs 1062: return (FALSE);
1.3 cvs 1063: }
1064:
1.31 cvs 1065:
1066: /*----------------------------------------------------------------------
1067: GetBaseURL
1068: normalizes orgName according to a base associated with doc, and
1069: following the standard URL format rules.
1070: The function returns the base used to solve relative URL and SRC:
1071: - the base of the document,
1072: - or the document path (without document name).
1073: ----------------------------------------------------------------------*/
1.106 cvs 1074: char *GetBaseURL (Document doc)
1.31 cvs 1075: {
1076: Element el;
1077: ElementType elType;
1078: AttributeType attrType;
1079: Attribute attr;
1.176 vatton 1080: char *ptr, *basename, *utf8path;
1.31 cvs 1081: int length;
1.151 kahan 1082: ThotBool hasDocBase;
1.31 cvs 1083:
1.113 cvs 1084: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 1085: return NULL;
1.148 kahan 1086: /* the other functions expect basename to have no more than MAX_LENGTH chars */
1.171 gully 1087: basename = (char *)TtaGetMemory (MAX_LENGTH);
1.148 kahan 1088: basename[0] = EOS;
1.31 cvs 1089: length = MAX_LENGTH -1;
1.151 kahan 1090: hasDocBase = FALSE;
1091:
1092: /* If the document has a base URL, it has a priority over the headers. */
1093: /* @@ We need to do this too when we support XML:base */
1094:
1095: /* is it a HTML document ? */
1096: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1097: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
1098: /* it's a HTML document */
1099: {
1100: /* get the document element */
1101: el = TtaGetMainRoot (doc);
1102: /* search the BASE element */
1103: elType.ElTypeNum = HTML_EL_HEAD;
1104: el = TtaSearchTypedElement (elType, SearchForward, el);
1105: if (el)
1106: /* there is a HEAD element */
1107: {
1108: /* look for a BASE element within the HEAD */
1109: elType.ElTypeNum = HTML_EL_BASE;
1110: el = TtaSearchTypedElement (elType, SearchInTree, el);
1111: }
1112: if (el)
1113: {
1114: /* The document has a BASE element. Get the HREF attribute of the
1115: BASE element */
1116: hasDocBase = TRUE;
1117: attrType.AttrSSchema = elType.ElSSchema;
1118: attrType.AttrTypeNum = HTML_ATTR_HREF_;
1119: attr = TtaGetAttribute (el, attrType);
1120: if (attr)
1121: {
1122: /* Use the base path of the document */
1123: TtaGiveTextAttributeValue (attr, basename, &length);
1.176 vatton 1124: utf8path = basename;
1125: basename = (char *)TtaConvertMbsToByte ((unsigned char *)utf8path,
1126: TtaGetDefaultCharset ());
1127: TtaFreeMemory (utf8path);
1.151 kahan 1128: }
1129: }
1130: }
1131:
1132: /* there was no BASE. Do we have a location header? */
1133: if (!hasDocBase && DocumentMeta[doc] && DocumentMeta[doc]->full_content_location
1.148 kahan 1134: && DocumentMeta[doc]->full_content_location[0] != EOS)
1.65 cvs 1135: {
1.148 kahan 1136: strncpy (basename, DocumentMeta[doc]->full_content_location, MAX_LENGTH-1);
1137: basename[MAX_LENGTH-1] = EOS;
1138: length = strlen (basename);
1139: }
1140:
1141: if (basename[0] != EOS)
1142: {
1143: /* base and orgName have to be separated by a DIR_SEP */
1144: length--;
1145: if (basename[0] != EOS && basename[length] != URL_SEP &&
1146: basename[length] != DIR_SEP)
1147: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 1148: {
1.148 kahan 1149: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1150: AMAYA_PARSE_HOST |
1151: AMAYA_PARSE_PUNCTUATION);
1152: if (ptr && !strcmp (ptr, basename))
1.31 cvs 1153: {
1.148 kahan 1154: /* it has this form, complete it by adding a URL_STR */
1155: if (strchr (basename, DIR_SEP))
1156: strcat (basename, DIR_STR);
1157: else
1158: strcat (basename, URL_STR);
1159: length++;
1.31 cvs 1160: }
1.149 kahan 1161: else if (!ptr || ptr[0] == EOS)
1162: {
1163: /* no host was detected, we may have a relative URL. We test
1164: if it begins with a URL_SEP, DIR_SEP or period. If yes, it's
1165: relative. */
1166: if (! (basename[0] == '.' || basename[0] == URL_SEP
1167: || basename[0] == DIR_SEP))
1168: basename[0] = EOS;
1169: }
1.148 kahan 1170: if (ptr)
1171: TtaFreeMemory (ptr);
1.31 cvs 1172: }
1.113 cvs 1173: }
1174:
1.148 kahan 1175: /* there was no base element and no location header, we use the DocumentURL */
1176: if (basename[0] == EOS)
1177: {
1178: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1179: basename[MAX_LENGTH-1] = EOS;
1180: }
1181:
1.31 cvs 1182: /* Remove anything after the last DIR_SEP char. If no such char is found,
1183: * then search for the first ":" char, hoping that what's before that is a
1184: * protocol. If found, end the string there. If neither char is found,
1185: * then discard the whole base element.
1186: */
1.106 cvs 1187: length = strlen (basename) - 1;
1.31 cvs 1188: /* search for the last DIR_SEP char */
1.106 cvs 1189: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 1190: length--;
1191: if (length >= 0)
1192: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1193: basename[length + 1] = EOS;
1.31 cvs 1194: else
1195: /* search for the first PATH_STR char */
1196: {
1.106 cvs 1197: for (length = 0; basename[length] != ':' &&
1198: basename[length] != EOS; length ++);
1199: if (basename[length] == ':')
1.31 cvs 1200: /* found, so end the string there */
1.106 cvs 1201: basename[length + 1] = EOS;
1.31 cvs 1202: else
1203: /* not found, discard the base */
1.106 cvs 1204: basename[0] = EOS;
1.31 cvs 1205: }
1206: return (basename);
1207: }
1208:
1209:
1.4 cvs 1210: /*----------------------------------------------------------------------
1.40 cvs 1211: GetLocalPath
1212: Allocate and return the local document path associated to the url
1213: ----------------------------------------------------------------------*/
1.150 vatton 1214: char *GetLocalPath (Document doc, char *url)
1.106 cvs 1215: {
1216: char *ptr;
1217: char *n;
1218: char *documentname;
1219: char url_sep;
1.83 cvs 1220: int len;
1.67 cvs 1221: ThotBool noFile;
1.40 cvs 1222:
1.153 vatton 1223: if (url)
1.40 cvs 1224: {
1225: /* check whether the file name exists */
1.106 cvs 1226: len = strlen (url) - 1;
1.71 cvs 1227: if (IsW3Path (url))
1.157 kahan 1228: url_sep = URL_SEP;
1.41 cvs 1229: else
1.106 cvs 1230: url_sep = DIR_SEP;
1.41 cvs 1231: noFile = (url[len] == url_sep);
1.40 cvs 1232: if (noFile)
1.106 cvs 1233: url[len] = EOS;
1.171 gully 1234: ptr = (char *)TtaGetMemory (MAX_LENGTH);
1235: documentname = (char *)TtaGetMemory (MAX_LENGTH);
1.78 cvs 1236: TtaExtractName (url, ptr, documentname);
1.106 cvs 1237: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1238: if (!TtaCheckDirectory (ptr))
1239: /* directory did not exist */
1.72 cvs 1240: TtaMakeDirectory (ptr);
1.47 cvs 1241:
1.153 vatton 1242: if (doc == 0)
1243: {
1244: n = strrchr (documentname, '.');
1245: if (n)
1246: *n = EOS;
1247: if (documentname[0] == EOS)
1248: strcpy (documentname, "noname");
1249: n = GetTempName (ptr, documentname);
1250: TtaFreeMemory (ptr);
1251: ptr = n;
1252: }
1.69 cvs 1253: else
1.153 vatton 1254: {
1255: /* don't include the query string within document name */
1256: n = strrchr (documentname, '?');
1257: if (n)
1258: *n = EOS;
1259: /* don't include ':' within document name */
1260: n = strchr (documentname, ':');
1261: if (n)
1262: *n = EOS;
1263: /* if after all this operations document name
1264: is empty, let's use noname.html instead */
1265: if (documentname[0] == EOS)
1266: strcat (ptr, "noname.html");
1267: else
1268: strcat (ptr, documentname);
1269: }
1.40 cvs 1270: TtaFreeMemory (documentname);
1.157 kahan 1271: /* substitute invalid chars in file names by a _ */
1.188 gully 1272: n = strrchr(ptr,DIR_SEP);
1.157 kahan 1273: while (*n)
1274: {
1275: if (*n == '*' || *n == ',')
1276: *n = '_';
1.187 cvs 1277: if ((unsigned char)*n >= 0x80) /* avoid non-ASCII */
1278: *n = 'A' + ((unsigned char)*n % 26);
1.157 kahan 1279: n++;
1280: }
1.40 cvs 1281: /* restore the url */
1282: if (noFile)
1.41 cvs 1283: url[len] = url_sep;
1.40 cvs 1284: return (ptr);
1285: }
1286: else
1287: return (NULL);
1288: }
1289:
1.73 cvs 1290: /*----------------------------------------------------------------------
1.79 cvs 1291: ExtractTarget extract the target name from document nane.
1292: ----------------------------------------------------------------------*/
1.150 vatton 1293: void ExtractTarget (char *aName, char *target)
1.79 cvs 1294: {
1.106 cvs 1295: int lg, i;
1296: char *ptr;
1297: char *oldptr;
1.79 cvs 1298:
1299: if (!target || !aName)
1300: /* bad target */
1301: return;
1302:
1.106 cvs 1303: target[0] = EOS;
1304: lg = strlen (aName);
1.79 cvs 1305: if (lg)
1306: {
1307: /* the name is not empty */
1308: oldptr = ptr = &aName[0];
1309: do
1310: {
1.106 cvs 1311: ptr = strrchr (oldptr, '#');
1.79 cvs 1312: if (ptr)
1313: oldptr = &ptr[1];
1314: }
1315: while (ptr);
1316:
1317: i = (int) (oldptr) - (int) (aName); /* name length */
1318: if (i > 1)
1319: {
1.106 cvs 1320: aName[i - 1] = EOS;
1.79 cvs 1321: if (i != lg)
1.106 cvs 1322: strcpy (target, oldptr);
1.79 cvs 1323: }
1324: }
1325: }
1326:
1327: /*----------------------------------------------------------------------
1.90 cvs 1328: RemoveNewLines (text)
1329: Removes any '\n' chars that are found in text.
1330: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1331: ----------------------------------------------------------------------*/
1.106 cvs 1332: ThotBool RemoveNewLines (char *text)
1333: {
1334: ThotBool change = FALSE;
1335: char *src;
1336: char *dest;
1.90 cvs 1337:
1338: src = text;
1339: dest = text;
1.115 kahan 1340:
1341: /* remove any preceding whitespace */
1342: while (*src && *src == ' ')
1343: {
1344: src++;
1345: change = 1;
1346: }
1347:
1.90 cvs 1348: while (*src)
1349: {
1350: switch (*src)
1351: {
1.106 cvs 1352: case '\n':
1.90 cvs 1353: /* don't copy the newline */
1354: change = 1;
1355: break;
1356: default:
1357: *dest = *src;
1358: dest++;
1359: break;
1360: }
1361: src++;
1362: }
1363: /* copy the last EOS char */
1364: *dest = *src;
1365:
1366: return (change);
1367: }
1368:
1369: /*----------------------------------------------------------------------
1370: CleanCopyFileURL
1371: Copies a file url from a src string to destination string.
1.97 cvs 1372: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1373: we want to do).
1.90 cvs 1374: ----------------------------------------------------------------------*/
1.106 cvs 1375: static void CleanCopyFileURL (char *dest, char *src,
1376: ConvertionType convertion)
1.90 cvs 1377: {
1378: while (*src)
1.89 cvs 1379: {
1.90 cvs 1380: switch (*src)
1.89 cvs 1381: {
1.184 gully 1382: #ifdef _WINDOWS
1.106 cvs 1383: case URL_SEP:
1.96 cvs 1384: /* make DIR_SEP transformation */
1.97 cvs 1385: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1386: *dest = DIR_SEP;
1.96 cvs 1387: else
1388: *dest = *src;
1.90 cvs 1389: dest++;
1.96 cvs 1390: src++;
1.90 cvs 1391: break;
1.184 gully 1392: #endif /* _WINDOWS */
1.96 cvs 1393:
1.106 cvs 1394: case '%':
1.97 cvs 1395: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1396: {
1.97 cvs 1397: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1398: src++;
1.106 cvs 1399: if (*src != EOS)
1.97 cvs 1400: {
1401: *dest = UnEscapeChar (*src) * 16;
1402: src++;
1403: }
1.106 cvs 1404: if (*src != EOS)
1.97 cvs 1405: {
1406: *dest = *dest + UnEscapeChar (*src);
1407: src++;
1408: }
1409: dest++;
1.96 cvs 1410: }
1.97 cvs 1411: else
1.96 cvs 1412: {
1.97 cvs 1413: *dest = *src;
1414: dest++;
1.96 cvs 1415: src++;
1416: }
1417: break;
1418:
1.90 cvs 1419: default:
1420: *dest = *src;
1.89 cvs 1421: dest++;
1.96 cvs 1422: src++;
1.90 cvs 1423: break;
1.89 cvs 1424: }
1425: }
1.90 cvs 1426: /* copy the EOS char */
1427: *dest = *src;
1.73 cvs 1428: }
1.40 cvs 1429:
1430: /*----------------------------------------------------------------------
1.9 cvs 1431: NormalizeURL
1432: normalizes orgName according to a base associated with doc, and
1433: following the standard URL format rules.
1.113 cvs 1434: if doc is < 0, use as a base the URL of the document that contains
1435: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1436: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1437: other path.
1.9 cvs 1438: The function returns the new complete and normalized URL
1.12 cvs 1439: or file name path (newName) and the name of the document (docName).
1.9 cvs 1440: N.B. If the function can't find out what's the docName, it assigns
1441: the name "noname.html".
1.4 cvs 1442: ----------------------------------------------------------------------*/
1.106 cvs 1443: void NormalizeURL (char *orgName, Document doc, char *newName,
1444: char *docName, char *otherPath)
1445: {
1446: char *basename;
1447: char tempOrgName[MAX_LENGTH];
1448: char *ptr;
1449: char used_sep;
1.84 cvs 1450: int length;
1451: ThotBool check;
1.5 cvs 1452:
1.184 gully 1453: #ifdef _WINDOWS
1.44 cvs 1454: int ndx;
1.184 gully 1455: #endif /* _WINDOWS */
1.44 cvs 1456:
1.5 cvs 1457: if (!newName || !docName)
1458: return;
1.18 cvs 1459:
1.113 cvs 1460: if (doc < 0)
1461: basename = TtaStrdup (SavedDocumentURL);
1462: else if (doc > 0)
1.53 cvs 1463: basename = GetBaseURL (doc);
1464: else if (otherPath != NULL)
1.108 cvs 1465: basename = TtaStrdup (otherPath);
1.32 cvs 1466: else
1.53 cvs 1467: basename = NULL;
1.32 cvs 1468:
1.18 cvs 1469: /*
1.31 cvs 1470: * Clean orgName
1471: * Make sure we have a complete orgName, without any leading or trailing
1472: * white spaces, or trailinbg new lines
1473: */
1.5 cvs 1474: ptr = orgName;
1.18 cvs 1475: /* skip leading white space and new line characters */
1.106 cvs 1476: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1477: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1478: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1479: /*
1.31 cvs 1480: * Make orgName a complete URL
1481: * If the URL does not include a protocol, then try to calculate
1482: * one using the doc's base element (if it exists),
1483: */
1.106 cvs 1484: if (tempOrgName[0] == EOS)
1.53 cvs 1485: {
1.106 cvs 1486: newName[0] = EOS;
1487: docName[0] = EOS;
1.53 cvs 1488: TtaFreeMemory (basename);
1489: return;
1490: }
1.49 cvs 1491:
1492: /* clean trailing white space */
1.106 cvs 1493: length = strlen (tempOrgName) - 1;
1494: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1495: {
1.106 cvs 1496: tempOrgName[length] = EOS;
1.53 cvs 1497: length--;
1498: }
1.50 cvs 1499:
1.55 cvs 1500: /* remove extra dot (which dot???) */
1501: /* ugly, but faster than a strcmp */
1.106 cvs 1502: if (tempOrgName[length] == '.'
1503: && (length == 0 || tempOrgName[length-1] != '.'))
1504: tempOrgName[length] = EOS;
1.50 cvs 1505:
1.94 cvs 1506: if (IsW3Path (tempOrgName))
1.53 cvs 1507: {
1508: /* the name is complete, go to the Sixth Step */
1.106 cvs 1509: strcpy (newName, tempOrgName);
1.53 cvs 1510: SimplifyUrl (&newName);
1511: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1512: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1513: AMAYA_PARSE_HOST |
1514: AMAYA_PARSE_PUNCTUATION);
1515: if (ptr && !strcmp (ptr, newName))
1516: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1517: strcat (newName, URL_STR);
1.49 cvs 1518:
1.53 cvs 1519: if (ptr)
1.50 cvs 1520: TtaFreeMemory (ptr);
1.53 cvs 1521: }
1.113 cvs 1522: else if (basename == NULL)
1.53 cvs 1523: /* the name is complete, go to the Sixth Step */
1.106 cvs 1524: strcpy (newName, tempOrgName);
1.53 cvs 1525: else
1526: {
1.31 cvs 1527: /* Calculate the absolute URL, using the base or document URL */
1.184 gully 1528: #ifdef _WINDOWS
1.53 cvs 1529: if (!IsW3Path (basename))
1530: {
1.106 cvs 1531: length = strlen (tempOrgName);
1.53 cvs 1532: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1533: if (tempOrgName [ndx] == '/')
1534: tempOrgName [ndx] = '\\';
1.53 cvs 1535: }
1.184 gully 1536: #endif /* _WINDOWS */
1.25 cvs 1537: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1538: if (ptr)
1539: {
1540: SimplifyUrl (&ptr);
1.106 cvs 1541: strcpy (newName, ptr);
1.53 cvs 1542: TtaFreeMemory (ptr);
1543: }
1544: else
1.106 cvs 1545: newName[0] = EOS;
1.53 cvs 1546: }
1.36 cvs 1547:
1548: TtaFreeMemory (basename);
1.18 cvs 1549: /*
1.31 cvs 1550: * Prepare the docname that will refer to this ressource in the
1551: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1552: * noname.html as a default ressource name
1.18 cvs 1553: */
1.106 cvs 1554: if (newName[0] != EOS)
1.53 cvs 1555: {
1.106 cvs 1556: length = strlen (newName) - 1;
1557: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1558: {
1559: used_sep = newName[length];
1560: check = TRUE;
1561: while (check)
1562: {
1.50 cvs 1563: length--;
1564: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1565: length--;
1.106 cvs 1566: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1567: {
1.106 cvs 1568: newName[length+1] = EOS;
1.53 cvs 1569: /* remove also previous directory */
1570: length--;
1571: while (length >= 0 && newName[length] != used_sep)
1572: length--;
1.106 cvs 1573: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1574: /* don't remove server name */
1.106 cvs 1575: newName[length+1] = EOS;
1.53 cvs 1576: }
1.106 cvs 1577: else if (!strncmp (&newName[length+1], ".", 1))
1578: newName[length+1] = EOS;
1.50 cvs 1579: else
1.53 cvs 1580: check = FALSE;
1581: }
1582: /* docname was not comprised inside the URL, so let's */
1583: /* assign the default ressource name */
1.106 cvs 1584: strcpy (docName, "noname.html");
1.53 cvs 1585: }
1586: else
1587: { /* docname is comprised inside the URL */
1.110 cvs 1588: while (length >= 0 && newName[length] != URL_SEP &&
1589: newName[length] != DIR_SEP)
1.53 cvs 1590: length--;
1591: if (length < 0)
1.106 cvs 1592: strcpy (docName, newName);
1.53 cvs 1593: else
1.106 cvs 1594: strcpy (docName, &newName[length+1]);
1.53 cvs 1595: }
1596: }
1597: else
1.106 cvs 1598: docName[0] = EOS;
1.18 cvs 1599: }
1.3 cvs 1600:
1.4 cvs 1601: /*----------------------------------------------------------------------
1.9 cvs 1602: IsSameHost
1.4 cvs 1603: ----------------------------------------------------------------------*/
1.106 cvs 1604: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1605: {
1.106 cvs 1606: char *basename_ptr1, *basename_ptr2;
1607: ThotBool result;
1.3 cvs 1608:
1.106 cvs 1609: basename_ptr1 = AmayaParseUrl (url1, "",
1610: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1611: basename_ptr2 = AmayaParseUrl (url2, "",
1612: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1613:
1.106 cvs 1614: if (strcmp (basename_ptr1, basename_ptr2))
1615: result = FALSE;
1616: else
1617: result = TRUE;
1618: TtaFreeMemory (basename_ptr1);
1619: TtaFreeMemory (basename_ptr2);
1620: return (result);
1.3 cvs 1621: }
1622:
1623:
1.4 cvs 1624: /*----------------------------------------------------------------------
1.22 cvs 1625: HasKnownFileSuffix
1626: returns TRUE if path points to a file ending with a suffix.
1627: ----------------------------------------------------------------------*/
1.153 vatton 1628: ThotBool HasKnownFileSuffix (const char *path)
1.106 cvs 1629: {
1630: char *root;
1631: char temppath[MAX_LENGTH];
1632: char suffix[MAX_LENGTH];
1.22 cvs 1633:
1.106 cvs 1634: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1635: return (FALSE);
1636:
1.106 cvs 1637: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1638:
1639: if (root)
1640: {
1.106 cvs 1641: strcpy (temppath, root);
1.25 cvs 1642: TtaFreeMemory (root);
1.22 cvs 1643: /* Get the suffix */
1.124 vatton 1644: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1645:
1.106 cvs 1646: if( suffix[0] == EOS)
1.22 cvs 1647: /* no suffix */
1648: return (FALSE);
1649:
1650: /* Normalize the suffix */
1651: ConvertToLowerCase (suffix);
1652:
1.106 cvs 1653: if (!strcmp (suffix, "gz"))
1.22 cvs 1654: /* skip the compressed suffix */
1655: {
1.124 vatton 1656: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1657: if(suffix[0] == EOS)
1.22 cvs 1658: /* no suffix */
1659: return (FALSE);
1660: /* Normalize the suffix */
1661: ConvertToLowerCase (suffix);
1662: }
1663:
1.106 cvs 1664: if (strcmp (suffix, "gif") &&
1665: strcmp (suffix, "xbm") &&
1666: strcmp (suffix, "xpm") &&
1667: strcmp (suffix, "jpg") &&
1668: strcmp (suffix, "pdf") &&
1669: strcmp (suffix, "png") &&
1670: strcmp (suffix, "tgz") &&
1671: strcmp (suffix, "xpg") &&
1672: strcmp (suffix, "xpd") &&
1673: strcmp (suffix, "ps") &&
1674: strcmp (suffix, "au") &&
1675: strcmp (suffix, "html") &&
1676: strcmp (suffix, "htm") &&
1677: strcmp (suffix, "shtml") &&
1678: strcmp (suffix, "xht") &&
1679: strcmp (suffix, "xhtm") &&
1680: strcmp (suffix, "xhtml") &&
1681: strcmp (suffix, "txt") &&
1682: strcmp (suffix, "css") &&
1683: strcmp (suffix, "eps"))
1.22 cvs 1684: return (FALSE);
1685: else
1686: return (TRUE);
1687: }
1688: else
1689: return (FALSE);
1690: }
1691:
1692:
1693: /*----------------------------------------------------------------------
1.24 cvs 1694: ChopURL
1695: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1696: If inputURL is bigger than that size, outputURL receives
1697: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1698: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1699: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1700: copied into outputURL.
1701: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1702: chars.
1703: ----------------------------------------------------------------------*/
1.106 cvs 1704: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1705: {
1706: int len;
1.9 cvs 1707:
1.106 cvs 1708: len = strlen (inputURL);
1.24 cvs 1709: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1710: strcpy (outputURL, inputURL);
1.24 cvs 1711: else
1712: /* make a truncated urlName on the status window */
1713: {
1.106 cvs 1714: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1715: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1716: strcat (outputURL, "...");
1717: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1718: }
1.25 cvs 1719: }
1720:
1721:
1722: /*----------------------------------------------------------------------
1723: scan
1.47 cvs 1724: Scan a filename for its constituents
1.25 cvs 1725: -----------------------------------
1726:
1727: On entry,
1728: name points to a document name which may be incomplete.
1729: On exit,
1730: absolute or relative may be nonzero (but not both).
1731: host, fragment and access may be nonzero if they were specified.
1732: Any which are nonzero point to zero terminated strings.
1733: ----------------------------------------------------------------------*/
1.106 cvs 1734: static void scan (char *name, HTURI *parts)
1.25 cvs 1735: {
1.106 cvs 1736: char * p;
1737: char * after_access = name;
1.32 cvs 1738:
1.43 cvs 1739: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1740: /* Look for fragment identifier */
1.106 cvs 1741: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1742: {
1.106 cvs 1743: *p++ = '\0';
1.28 cvs 1744: parts->fragment = p;
1.25 cvs 1745: }
1746:
1.28 cvs 1747: for (p=name; *p; p++)
1748: {
1.106 cvs 1749: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1750: break;
1.106 cvs 1751: if (*p == ':')
1.28 cvs 1752: {
1753: *p = 0;
1754: parts->access = after_access; /* Scheme has been specified */
1755:
1756: /* The combination of gcc, the "-O" flag and the HP platform is
1757: unhealthy. The following three lines is a quick & dirty fix, but is
1758: not recommended. Rather, turn off "-O". */
1759:
1760: /* after_access = p;*/
1761: /* while (*after_access == 0)*/
1762: /* after_access++;*/
1763: after_access = p+1;
1.106 cvs 1764: if (!strcasecmp("URL", parts->access))
1.28 cvs 1765: /* Ignore IETF's URL: pre-prefix */
1766: parts->access = NULL;
1767: else
1.25 cvs 1768: break;
1769: }
1770: }
1771:
1772: p = after_access;
1.43 cvs 1773: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1774: {
1.43 cvs 1775: if (p[1] == URL_SEP)
1.28 cvs 1776: {
1.25 cvs 1777: parts->host = p+2; /* host has been specified */
1.28 cvs 1778: *p = 0; /* Terminate access */
1779: /* look for end of host name if any */
1.106 cvs 1780: p = strchr (parts->host, URL_SEP);
1.28 cvs 1781: if (p)
1782: {
1.106 cvs 1783: *p = EOS; /* Terminate host */
1.25 cvs 1784: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1785: }
1786: }
1787: else
1788: /* Root found but no host */
1789: parts->absolute = p+1;
1790: }
1791: else
1792: {
1.25 cvs 1793: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1794: }
1.25 cvs 1795: }
1796:
1797:
1798: /*----------------------------------------------------------------------
1.28 cvs 1799: AmayaParseUrl: parse a Name relative to another name
1800:
1801: This returns those parts of a name which are given (and requested)
1802: substituting bits from the related name where necessary.
1.25 cvs 1803:
1.28 cvs 1804: On entry,
1.25 cvs 1805: aName A filename given
1806: relatedName A name relative to which aName is to be parsed. Give
1807: it an empty string if aName is absolute.
1808: wanted A mask for the bits which are wanted.
1809:
1.28 cvs 1810: On exit,
1.25 cvs 1811: returns A pointer to a malloc'd string which MUST BE FREED
1812: ----------------------------------------------------------------------*/
1.106 cvs 1813: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1814: {
1815: char *return_value;
1816: char result[MAX_LENGTH];
1817: char name[MAX_LENGTH];
1818: char rel[MAX_LENGTH];
1819: char *p, *access;
1.29 cvs 1820: HTURI given, related;
1821: int len;
1.106 cvs 1822: char used_sep;
1823: char *used_str;
1.32 cvs 1824:
1.106 cvs 1825: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1826: {
1.106 cvs 1827: used_str = DIR_STR;
1828: used_sep = DIR_SEP;
1.33 cvs 1829: }
1.32 cvs 1830: else
1.33 cvs 1831: {
1.106 cvs 1832: used_str = URL_STR;
1833: used_sep = URL_SEP;
1.33 cvs 1834: }
1.32 cvs 1835:
1.29 cvs 1836: /* Make working copies of input strings to cut up: */
1837: return_value = NULL;
1838: result[0] = 0; /* Clear string */
1.169 quint 1839: rel[0] = EOS;
1840: strncpy (name, aName, MAX_LENGTH - 1);
1841: name[MAX_LENGTH - 1] = EOS;
1842: if (relatedName != NULL)
1843: {
1844: strncpy (rel, relatedName, MAX_LENGTH - 1);
1845: rel[MAX_LENGTH - 1] = EOS;
1846: }
1.29 cvs 1847: else
1.106 cvs 1848: relatedName[0] = EOS;
1.29 cvs 1849:
1850: scan (name, &given);
1851: scan (rel, &related);
1852: access = given.access ? given.access : related.access;
1853: if (wanted & AMAYA_PARSE_ACCESS)
1854: if (access)
1855: {
1.106 cvs 1856: strcat (result, access);
1.29 cvs 1857: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1858: strcat (result, ":");
1.29 cvs 1859: }
1860:
1861: if (given.access && related.access)
1862: /* If different, inherit nothing. */
1.106 cvs 1863: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1864: {
1865: related.host = 0;
1866: related.absolute = 0;
1867: related.relative = 0;
1868: related.fragment = 0;
1869: }
1870:
1871: if (wanted & AMAYA_PARSE_HOST)
1872: if(given.host || related.host)
1873: {
1874: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1875: strcat (result, "//");
1876: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1877: }
1878:
1879: if (given.host && related.host)
1880: /* If different hosts, inherit no path. */
1.106 cvs 1881: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1882: {
1883: related.absolute = 0;
1884: related.relative = 0;
1885: related.fragment = 0;
1886: }
1887:
1888: if (wanted & AMAYA_PARSE_PATH)
1889: {
1890: if (given.absolute)
1891: {
1892: /* All is given */
1893: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1894: strcat (result, used_str);
1895: strcat (result, given.absolute);
1.25 cvs 1896: }
1.29 cvs 1897: else if (related.absolute)
1898: {
1899: /* Adopt path not name */
1.106 cvs 1900: strcat (result, used_str);
1901: strcat (result, related.absolute);
1.29 cvs 1902: if (given.relative)
1903: {
1904: /* Search part? */
1.106 cvs 1905: p = strchr (result, '?');
1.29 cvs 1906: if (!p)
1.106 cvs 1907: p=result+strlen(result)-1;
1.33 cvs 1908: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1909: /* Remove filename */
1910: p[1]=0;
1911: /* Add given one */
1.106 cvs 1912: strcat (result, given.relative);
1.25 cvs 1913: }
1914: }
1.29 cvs 1915: else if (given.relative)
1916: /* what we've got */
1.106 cvs 1917: strcat (result, given.relative);
1.29 cvs 1918: else if (related.relative)
1.106 cvs 1919: strcat (result, related.relative);
1.29 cvs 1920: else
1921: /* No inheritance */
1.106 cvs 1922: strcat (result, used_str);
1.25 cvs 1923: }
1.29 cvs 1924:
1925: if (wanted & AMAYA_PARSE_ANCHOR)
1926: if (given.fragment || related.fragment)
1927: {
1928: if (given.absolute && given.fragment)
1929: {
1930: /*Fixes for relURLs...*/
1931: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1932: strcat (result, "#");
1933: strcat (result, given.fragment);
1.29 cvs 1934: }
1935: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1936: strcat (result, "");
1.29 cvs 1937: else
1938: {
1.110 cvs 1939: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1940: strcat (result, "#");
1.110 cvs 1941: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1942: }
1943: }
1.106 cvs 1944: len = strlen (result);
1.171 gully 1945: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 1946: strcpy (return_value, result);
1.29 cvs 1947: return (return_value); /* exactly the right length */
1.25 cvs 1948: }
1949:
1950: /*----------------------------------------------------------------------
1951: HTCanon
1952: Canonicalizes the URL in the following manner starting from the host
1953: pointer:
1954:
1955: 1) The host name is converted to lowercase
1956: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1957:
1958: Return: OK The position of the current path part of the URL
1959: which might be the old one or a new one.
1960:
1961: ----------------------------------------------------------------------*/
1.106 cvs 1962: static char *HTCanon (char **filename, char *host)
1963: {
1964: char *newname = NULL;
1965: char used_sep;
1966: char *path;
1967: char *strptr;
1968: char *port;
1969: char *access = host-3;
1970:
1971: if (*filename && strchr (*filename, URL_SEP))
1972: used_sep = URL_SEP;
1.33 cvs 1973: else
1.106 cvs 1974: used_sep = DIR_SEP;
1.32 cvs 1975:
1.110 cvs 1976: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1977: access--;
1.110 cvs 1978: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1979: path = host + strlen (host);
1980: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1981: host = strptr;
1.110 cvs 1982: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1983: port = NULL;
1.25 cvs 1984:
1985: strptr = host; /* Convert to lower-case */
1.82 cvs 1986: while (strptr < path)
1.33 cvs 1987: {
1.123 vatton 1988: *strptr = tolower (*strptr);
1.82 cvs 1989: strptr++;
1.33 cvs 1990: }
1.25 cvs 1991:
1992: /* Does the URL contain a full domain name? This also works for a
1993: numerical host name. The domain name is already made lower-case
1994: and without a trailing dot. */
1995: {
1.106 cvs 1996: char *dot = port ? port : path;
1997: if (dot > *filename && *--dot == '.')
1.33 cvs 1998: {
1.106 cvs 1999: char *orig = dot;
2000: char *dest = dot + 1;
1.82 cvs 2001: while ((*orig++ = *dest++));
2002: if (port) port--;
1.33 cvs 2003: path--;
1.25 cvs 2004: }
2005: }
2006: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 2007: if (port)
2008: {
1.82 cvs 2009: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 2010: {
2011: if (!newname)
2012: {
1.106 cvs 2013: char *orig = port;
2014: char *dest = port + 1;
1.82 cvs 2015: while ((*orig++ = *dest++));
1.33 cvs 2016: }
2017: }
1.106 cvs 2018: else if ((!strncmp (access, "http", 4) &&
2019: (*(port + 1) == '8' &&
2020: *(port+2) == '0' &&
1.82 cvs 2021: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 2022: (!strncmp (access, "gopher", 6) &&
2023: (*(port+1) == '7' &&
2024: *(port+2) == '0' &&
1.82 cvs 2025: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 2026: (!strncmp (access, "ftp", 3) &&
2027: (*(port+1) == '2' &&
2028: *(port + 2) == '1' &&
1.82 cvs 2029: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 2030: if (!newname)
2031: {
1.106 cvs 2032: char *orig = port;
2033: char *dest = port + 3;
1.33 cvs 2034: while((*orig++ = *dest++));
2035: /* Update path position, Henry Minsky */
2036: path -= 3;
1.25 cvs 2037: }
1.33 cvs 2038: }
2039: else if (newname)
1.106 cvs 2040: strncat (newname, port, (int) (path - port));
1.33 cvs 2041: }
1.25 cvs 2042:
1.33 cvs 2043: if (newname)
2044: {
1.106 cvs 2045: char *newpath = newname + strlen (newname);
2046: strcat (newname, path);
1.25 cvs 2047: path = newpath;
1.28 cvs 2048: /* Free old copy */
2049: TtaFreeMemory(*filename);
1.25 cvs 2050: *filename = newname;
1.33 cvs 2051: }
1.25 cvs 2052: return path;
2053: }
2054:
2055:
2056: /*----------------------------------------------------------------------
1.29 cvs 2057: SimplifyUrl: simplify a URI
1.32 cvs 2058: A URI is allowed to contain the sequence xxx/../ which may be
2059: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 2060: Simplification helps us recognize duplicate URIs.
1.25 cvs 2061:
1.28 cvs 2062: Thus, /etc/junk/../fred becomes /etc/fred
2063: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 2064:
1.28 cvs 2065: but we should NOT change
2066: http://fred.xxx.edu/../..
1.25 cvs 2067:
2068: or ../../albert.html
2069:
1.28 cvs 2070: In order to avoid empty URLs the following URLs become:
1.25 cvs 2071:
2072: /fred/.. becomes /fred/..
2073: /fred/././.. becomes /fred/..
2074: /fred/.././junk/.././ becomes /fred/..
2075:
1.28 cvs 2076: If more than one set of `://' is found (several proxies in cascade) then
2077: only the part after the last `://' is simplified.
1.25 cvs 2078:
1.28 cvs 2079: Returns: A string which might be the old one or a new one.
1.25 cvs 2080: ----------------------------------------------------------------------*/
1.106 cvs 2081: void SimplifyUrl (char **url)
2082: {
2083: char *path;
2084: char *access;
2085: char *newptr;
2086: char *p;
2087: char *orig, *dest, *end;
1.28 cvs 2088:
1.106 cvs 2089: char used_sep;
1.77 cvs 2090: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
2091: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 2092:
1.28 cvs 2093: if (!url || !*url)
2094: return;
2095:
1.106 cvs 2096: if (strchr (*url, URL_SEP))
2097: used_sep = URL_SEP;
1.32 cvs 2098: else
1.106 cvs 2099: used_sep = DIR_SEP;
1.32 cvs 2100:
1.77 cvs 2101: /* should we simplify double dot? */
2102: path = *url;
1.106 cvs 2103: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 2104: ddot_simplify = FALSE;
2105: else
2106: ddot_simplify = TRUE;
2107:
1.28 cvs 2108: /* Find any scheme name */
1.106 cvs 2109: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 2110: {
2111: /* Find host name */
1.28 cvs 2112: access = *url;
1.123 vatton 2113: while (access < path && (*access = tolower (*access)))
1.82 cvs 2114: access++;
1.28 cvs 2115: path += 3;
1.106 cvs 2116: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 2117: /* For proxies */
1.106 cvs 2118: path = newptr + 3;
1.82 cvs 2119: /* We have a host name */
1.84 cvs 2120: path = HTCanon (url, path);
1.25 cvs 2121: }
1.106 cvs 2122: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 2123: path += 2;
2124: else
2125: path = *url;
1.84 cvs 2126: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 2127: /* Some URLs start //<foo> */
2128: path += 1;
1.94 cvs 2129: else if (IsFilePath (path))
2130: {
2131: /* doesn't need to do anything more */
2132: return;
2133: }
1.106 cvs 2134: else if (!strncmp (path, "news:", 5))
1.28 cvs 2135: {
1.106 cvs 2136: newptr = strchr (path+5, '@');
1.28 cvs 2137: if (!newptr)
2138: newptr = path + 5;
2139: while (*newptr)
2140: {
2141: /* Make group or host lower case */
1.123 vatton 2142: *newptr = tolower (*newptr);
1.28 cvs 2143: newptr++;
1.25 cvs 2144: }
1.28 cvs 2145: /* Doesn't need to do any more */
2146: return;
1.25 cvs 2147: }
1.126 cheyroul 2148:
1.189 ! vatton 2149: /* remove initial multiple /'s */
! 2150: p = path - 1;
! 2151: if (path != *url && *p == used_sep && *path == used_sep)
! 2152: {
! 2153: while (*path == used_sep)
! 2154: {
! 2155: orig = path;
! 2156: dest = path + 1;
! 2157: while ((*orig++ = *dest++)); /* Remove multiple /'s */
! 2158: end = orig-1;
! 2159: }
! 2160: }
! 2161:
! 2162: if (path)
1.28 cvs 2163: {
1.106 cvs 2164: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
2165: (end = strchr (path, '#'))))
2166: end = path + strlen (path);
1.28 cvs 2167:
2168: /* Parse string second time to simplify */
2169: p = path;
2170: while (p < end)
2171: {
1.110 cvs 2172: /* if we're pointing to a char, it's safe to reactivate the
2173: ../ convertion */
1.106 cvs 2174: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 2175: ddot_simplify = TRUE;
2176:
1.33 cvs 2177: if (*p==used_sep)
1.28 cvs 2178: {
1.106 cvs 2179: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 2180: {
2181: orig = p + 1;
1.84 cvs 2182: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 2183: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 2184: end = orig - 1;
2185: }
1.106 cvs 2186: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 2187: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 2188: {
2189: newptr = p;
1.52 cvs 2190: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
2191: if (*newptr == used_sep)
2192: orig = newptr + 1;
1.28 cvs 2193: else
1.52 cvs 2194: orig = newptr;
2195:
2196: dest = (*(p+3) != used_sep) ? p+3 : p+4;
2197: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
2198: end = orig-1;
2199: /* Start again with prev slash */
2200: p = newptr;
1.28 cvs 2201: }
1.33 cvs 2202: else if (*(p+1) == used_sep)
1.28 cvs 2203: {
1.33 cvs 2204: while (*(p+1) == used_sep)
1.28 cvs 2205: {
2206: orig = p;
2207: dest = p + 1;
2208: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2209: end = orig-1;
2210: }
2211: }
2212: else
1.25 cvs 2213: p++;
1.28 cvs 2214: }
2215: else
2216: p++;
1.25 cvs 2217: }
2218: }
1.51 cvs 2219: /*
2220: ** Check for host/../.. kind of things
2221: */
1.106 cvs 2222: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 2223: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 2224: *(path+1) = EOS;
1.28 cvs 2225: return;
2226: }
2227:
2228:
2229: /*----------------------------------------------------------------------
1.96 cvs 2230: NormalizeFile normalizes local names.
1.28 cvs 2231: Return TRUE if target and src differ.
2232: ----------------------------------------------------------------------*/
1.106 cvs 2233: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2234: {
1.184 gully 2235: #ifndef _WINDOWS
1.106 cvs 2236: char *s;
1.93 cvs 2237: int i;
1.184 gully 2238: #endif /* _WINDOWS */
1.82 cvs 2239: ThotBool change;
1.90 cvs 2240: int start_index; /* the first char that we'll copy */
1.28 cvs 2241:
1.54 cvs 2242: change = FALSE;
1.90 cvs 2243: start_index = 0;
2244:
1.106 cvs 2245: if (!src || src[0] == EOS)
1.96 cvs 2246: {
1.106 cvs 2247: target[0] = EOS;
1.96 cvs 2248: return FALSE;
2249: }
1.90 cvs 2250:
2251: /* @@ do I need file: or file:/ here? */
1.106 cvs 2252: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 2253: {
1.90 cvs 2254: /* remove the prefix file: */
2255: start_index += 5;
2256:
2257: /* remove the localhost prefix */
1.106 cvs 2258: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 2259: start_index += 11;
2260:
2261: /* remove the first two slashes in / / /path */
2262: while (src[start_index] &&
1.106 cvs 2263: src[start_index] == '/'
2264: && src[start_index + 1] == '/')
1.94 cvs 2265: start_index++;
2266:
1.184 gully 2267: #ifdef _WINDOWS
1.94 cvs 2268: /* remove any extra slash before the drive name */
1.106 cvs 2269: if (src[start_index] == '/'
2270: &&src[start_index+2] == ':')
1.94 cvs 2271: start_index++;
1.184 gully 2272: #endif /* _WINDOWS */
1.90 cvs 2273:
1.106 cvs 2274: if (src[start_index] == EOS)
1.90 cvs 2275: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 2276: strcpy (target, DIR_STR);
1.90 cvs 2277: else
1.97 cvs 2278: /* as we're inside a file: URL, we'll apply all the convertions
2279: we know */
2280: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 2281:
2282: change = TRUE;
2283: }
1.97 cvs 2284: else if (convertion != AM_CONV_NONE)
1.96 cvs 2285: {
2286: /* we are following a "local" relative link, we do all the
2287: convertions except for the HOME_DIR ~ one */
1.97 cvs 2288: CleanCopyFileURL (target, src, convertion);
1.28 cvs 2289: }
1.184 gully 2290: #ifndef _WINDOWS
1.106 cvs 2291: else if (src[0] == '~')
1.53 cvs 2292: {
1.96 cvs 2293: /* it must be a URL typed in a text input field */
2294: /* do the HOME_DIR ~ substitution */
1.82 cvs 2295: s = TtaGetEnvString ("HOME");
1.106 cvs 2296: strcpy (target, s);
1.90 cvs 2297: #if 0
1.96 cvs 2298: /* JK: invalidated this part of the code as it's simpler
2299: to add the DIR_SEP whenever we have something to add
2300: to the path rather than adding it systematically */
1.106 cvs 2301: if (src[1] != DIR_SEP)
2302: strcat (target, DIR_STR);
1.90 cvs 2303: #endif
1.106 cvs 2304: i = strlen (target);
2305: strcpy (&target[i], &src[1]);
1.54 cvs 2306: change = TRUE;
1.53 cvs 2307: }
1.184 gully 2308: #endif /* _WINDOWS */
1.28 cvs 2309: else
1.96 cvs 2310: /* leave it as it is */
1.106 cvs 2311: strcpy (target, src);
1.96 cvs 2312:
1.28 cvs 2313: /* remove /../ and /./ */
1.29 cvs 2314: SimplifyUrl (&target);
1.54 cvs 2315: if (!change)
1.106 cvs 2316: change = strcmp (src, target);
1.28 cvs 2317: return (change);
1.25 cvs 2318: }
2319:
1.28 cvs 2320:
1.25 cvs 2321: /*----------------------------------------------------------------------
1.31 cvs 2322: MakeRelativeURL: make relative name
1.25 cvs 2323:
1.28 cvs 2324: This function creates and returns a string which gives an expression of
2325: one address as related to another. Where there is no relation, an absolute
2326: address is retured.
1.25 cvs 2327:
1.28 cvs 2328: On entry,
1.25 cvs 2329: Both names must be absolute, fully qualified names of nodes
2330: (no fragment bits)
2331:
1.28 cvs 2332: On exit,
1.25 cvs 2333: The return result points to a newly allocated name which, if
2334: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2335: The caller is responsible for freeing the resulting name later.
2336: ----------------------------------------------------------------------*/
1.106 cvs 2337: char *MakeRelativeURL (char *aName, char *relatedName)
2338: {
2339: char *return_value;
2340: char result[MAX_LENGTH];
2341: char *p;
2342: char *q;
2343: char *after_access;
2344: char *last_slash = NULL;
2345: int slashes, levels, len;
1.184 gully 2346: #ifdef _WINDOWS
1.44 cvs 2347: int ndx;
1.184 gully 2348: #endif /* _WINDOWS */
1.44 cvs 2349:
1.29 cvs 2350: if (aName == NULL || relatedName == NULL)
2351: return (NULL);
2352:
2353: slashes = 0;
2354: after_access = NULL;
2355: p = aName;
2356: q = relatedName;
1.147 vatton 2357: len = 0;
2358: for (; *p && !strncasecmp (p, q, 1); p++, q++, len++)
1.27 cvs 2359: {
2360: /* Find extent of match */
1.106 cvs 2361: if (*p == ':')
1.146 cvs 2362: {
1.168 cvs 2363: after_access = p + 1;
1.184 gully 2364: #ifdef _WINDOWS
1.168 cvs 2365: if (len == 1)
2366: {
2367: /* it's a local Windows path like c:... */
2368: slashes+=2;
2369: }
1.184 gully 2370: #endif /* _WINDOWS */
1.168 cvs 2371: }
2372: if (*p == DIR_SEP)
2373: {
2374: /* memorize the last slash position and count them */
2375: last_slash = p;
1.147 vatton 2376: slashes++;
1.146 cvs 2377: }
1.25 cvs 2378: }
2379:
1.31 cvs 2380: /* q, p point to the first non-matching character or zero */
1.106 cvs 2381: if (*q == EOS)
1.31 cvs 2382: {
2383: /* New name is a subset of the related name */
2384: /* exactly the right length */
1.106 cvs 2385: len = strlen (p);
1.171 gully 2386: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 2387: strcpy (return_value, p);
1.31 cvs 2388: }
2389: else if ((slashes < 2 && after_access == NULL)
2390: || (slashes < 3 && after_access != NULL))
1.168 cvs 2391: {
1.31 cvs 2392: /* Two names whitout common path */
2393: /* exactly the right length */
1.106 cvs 2394: len = strlen (aName);
1.171 gully 2395: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 2396: strcpy (return_value, aName);
1.31 cvs 2397: }
2398: else
2399: {
2400: /* Some path in common */
1.106 cvs 2401: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2402: /* just the same server */
1.106 cvs 2403: strcpy (result, last_slash);
1.31 cvs 2404: else
2405: {
2406: levels= 0;
1.106 cvs 2407: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2408: if (*q == DIR_SEP)
2409: levels++;
2410:
1.106 cvs 2411: result[0] = EOS;
1.31 cvs 2412: for (;levels; levels--)
1.106 cvs 2413: strcat (result, "../");
2414: strcat (result, last_slash+1);
1.31 cvs 2415: }
1.52 cvs 2416:
2417: if (!*result)
1.106 cvs 2418: strcat (result, "./");
1.52 cvs 2419:
1.31 cvs 2420: /* exactly the right length */
1.106 cvs 2421: len = strlen (result);
1.171 gully 2422: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 2423: strcpy (return_value, result);
1.52 cvs 2424:
1.25 cvs 2425: }
1.184 gully 2426: #ifdef _WINDOWS
1.106 cvs 2427: len = strlen (return_value);
1.168 cvs 2428: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2429: if (return_value[ndx] == '\\')
2430: return_value[ndx] = '/' ;
1.184 gully 2431: #endif /* _WINDOWS */
1.29 cvs 2432: return (return_value);
1.24 cvs 2433: }
1.35 cvs 2434:
1.104 kahan 2435: /*----------------------------------------------------------------------
2436: AM_GetFileSize
2437: Returns TRUE and the filesize in the 2nd parameter.
2438: Otherwise, in case of a system error, returns FALSE, with a
2439: filesize of 0L.
2440: ---------------------------------------------------------------------*/
1.106 cvs 2441: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2442: {
2443: if (!TtaFileExist (filename))
2444: return FALSE;
1.185 vatton 2445: *file_size = TtaGetFileSize (filename);
2446: return TRUE;
1.104 kahan 2447: }
1.139 kahan 2448:
2449: /*----------------------------------------------------------------------
2450: AM_UseXHTMLMimeType
2451: Returns TRUE if the user has configured Amaya to use this MIME type,
2452: FALSE otherwise.
2453: ---------------------------------------------------------------------*/
2454: ThotBool AM_UseXHTMLMimeType (void)
2455: {
2456: ThotBool xhtml_mimetype;
2457:
2458: /* does the user wants to use the new MIME type? */
2459: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2460:
2461: return (xhtml_mimetype);
1.152 kahan 2462: }
2463:
1.154 kahan 2464:
2465: /********************************************
2466: The following routines were adapted from the GNU libc functions
2467: for generating a tmpnam.
2468: *********************************************/
2469:
2470: /* These are the characters used in temporary filenames. */
2471: static const char letters[] =
2472: "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
2473:
2474: /* Generate a temporary file name based on TMPL. TMPL must match the
2475: rules for mk[s]temp (i.e. end in "XXXXXX"). The name constructed
2476: does not exist at the time of the call to __gen_tempname. TMPL is
2477: overwritten with the result.
2478:
2479: We use a clever algorithm to get hard-to-predict names. */
2480: void
2481: AM_gen_tempname (char *tmpl)
2482: {
2483: int len;
2484: char *XXXXXX;
2485: static uint64_t value;
2486: uint64_t random_time_bits;
2487: unsigned int count;
2488: int save_errno = errno;
2489: struct stat st;
2490:
2491: /* A lower bound on the number of temporary files to attempt to
2492: generate. The maximum total number of temporary file names that
2493: can exist for a given template is 62**6. It should never be
2494: necessary to try all these combinations. Instead if a reasonable
2495: number of names is tried (we define reasonable as 62**3) fail to
2496: give the system administrator the chance to remove the problems. */
2497: unsigned int attempts_min = 62 * 62 * 62;
2498:
2499: /* The number of times to attempt to generate a temporary file. To
2500: conform to POSIX, this must be no smaller than TMP_MAX. */
2501: unsigned int attempts = attempts_min < TMP_MAX ? TMP_MAX : attempts_min;
2502:
2503: len = strlen (tmpl);
2504: if (len < 6 || strcmp (&tmpl[len - 6], "XXXXXX"))
2505: {
2506: /* @@ JK ? */
2507: errno = EINVAL;
2508: return;
2509: }
2510:
2511: /* This is where the Xs start. */
2512: XXXXXX = &tmpl[len - 6];
2513:
2514: /* Get some more or less random data. */
2515: #ifdef RANDOM_BITS
2516: RANDOM_BITS (random_time_bits);
2517: #else
2518: # if HAVE_GETTIMEOFDAY || _LIBC
2519: {
2520: struct timeval tv;
2521: gettimeofday (&tv, NULL);
2522: random_time_bits = ((uint64_t) tv.tv_usec << 16) ^ tv.tv_sec;
2523: }
2524: # else
2525: random_time_bits = time (NULL);
2526: # endif
2527: #endif
2528: value += random_time_bits ^ getpid ();
2529:
2530: for (count = 0; count < attempts; value += 7777, ++count)
2531: {
2532: uint64_t v = value;
2533:
2534: /* Fill in the random bits. */
2535: XXXXXX[0] = letters[v % 62];
2536: v /= 62;
2537: XXXXXX[1] = letters[v % 62];
2538: v /= 62;
2539: XXXXXX[2] = letters[v % 62];
2540: v /= 62;
2541: XXXXXX[3] = letters[v % 62];
2542: v /= 62;
2543: XXXXXX[4] = letters[v % 62];
2544: v /= 62;
2545: XXXXXX[5] = letters[v % 62];
2546:
2547: /* This case is backward from the other three. AM_gen_tempname
2548: succeeds if __xstat fails because the name does not exist.
2549: Note the continue to bypass the common logic at the bottom
2550: of the loop. */
2551: if (stat (tmpl, &st) < 0)
2552: break;
2553:
2554: continue;
2555: }
2556:
2557: if (count == attempts || errno != ENOENT)
2558: tmpl[0] = EOS;
2559: else
2560: errno = save_errno;
2561:
2562: return;
2563: }
2564:
2565: #define JOSE 1
2566:
1.152 kahan 2567: /*-----------------------------------------------------------------------
2568: GetTempName
2569: Front end to the Unix tempnam function, which is independent of the
2570: value of the TMPDIR env value
2571: Returns a dynamically allocated string with a tempname. The user
2572: must free this memory.
2573: -----------------------------------------------------------------------*/
2574: char *GetTempName (const char *dir, const char *prefix)
2575: {
1.154 kahan 2576: #ifdef JOSE
2577:
1.162 kahan 2578: static char tmpbufmem[PATH_MAX + 1];
1.154 kahan 2579: int len;
2580: int i;
2581:
1.155 cvs 2582: if (!dir || *dir == EOS || !TtaDirExists (dir))
1.154 kahan 2583: return NULL;
2584:
1.162 kahan 2585: /* make sure that the name is no bigger than PATH_MAX + the 6 tempname chars we
2586: will add */
1.154 kahan 2587:
1.156 cvs 2588: len = strlen (dir);
1.162 kahan 2589: if (len + 6 > PATH_MAX)
1.154 kahan 2590: return NULL;
2591:
2592: /* copy the dir name, and add a DIR_SEP if it's missing */
2593: if (dir[strlen (dir) - 1] == DIR_SEP)
2594: strcpy (tmpbufmem, dir);
2595: else
1.156 cvs 2596: {
1.154 kahan 2597: sprintf (tmpbufmem, "%s%c", dir, DIR_SEP);
1.156 cvs 2598: len++;
2599: }
1.154 kahan 2600:
1.161 kahan 2601: /* copy the prefix (no more than L_tmpnam chars, to respect POSIX). Save
1.156 cvs 2602: space for the 6 X and EOS chars that will become the random bits */
2603: if (prefix)
2604: {
2605: i = 0;
1.167 cheyroul 2606: while (prefix[i] != EOS && i < L_tmpnam - 8)
1.156 cvs 2607: tmpbufmem[len++] = prefix[i++];
2608: tmpbufmem[len] = EOS;
2609: }
2610:
2611: /* Add the 6 X chars */
2612: len = strlen (tmpbufmem);
2613: i = 0;
2614: while (i < 6)
2615: {
2616: tmpbufmem[len++] = 'X';
2617: i++;
2618: }
2619: tmpbufmem[len] = EOS;
1.154 kahan 2620:
2621: AM_gen_tempname (tmpbufmem);
2622:
2623: if (tmpbufmem[0] == EOS)
2624: return NULL;
2625: else
2626: return (TtaStrdup (tmpbufmem));
2627:
2628: #else
1.152 kahan 2629: char *tmpdir;
1.181 vatton 2630: char *tmp = NULL;
1.152 kahan 2631: char *name = NULL;
2632:
2633: /* save the value of TMPDIR */
1.181 vatton 2634: tmpdir = getenv ("TMPDIR");
1.152 kahan 2635: if (tmpdir)
2636: {
1.181 vatton 2637: /* remove TMPDIR from the environment */
2638: tmp = TtaGetMemory (strlen (tmpdir) + 20);
2639: sprintf (tmp, "TMPDIR=");
1.184 gully 2640: #ifdef _WINDOWS
1.152 kahan 2641: _putenv (tmp);
2642: #else
2643: putenv (tmp);
1.184 gully 2644: #endif /* _WINDOWS */
1.181 vatton 2645: /* prepare the string to restore the value of TMPDIR */
2646: strrcat (tmp, tmpdir);
1.152 kahan 2647: }
2648:
2649: /* create the tempname */
1.184 gully 2650: #ifdef _WINDOWS
1.152 kahan 2651: /* Under Windows, _tempnam returns the same name until the file is created */
2652: {
2653: char *altprefix;
2654: name = tmpnam (NULL); /* get a possibly unique string */
1.181 vatton 2655: altprefix = TtaGetMemory (strlen (prefix) + strlen(name) + 1);
1.152 kahan 2656: sprintf (altprefix, "%s%s", prefix, name + strlen(_P_tmpdir));
2657: name = _tempnam (dir, altprefix); /* get a name that isn't yet in use */
2658: TtaFreeMemory (altprefix);
2659: }
2660: #else
2661: name = tempnam (dir, prefix);
1.184 gully 2662: #endif /* _WINDOWS */
1.152 kahan 2663:
2664: if (tmpdir)
2665: {
1.181 vatton 2666: /* restore the value of TMPDIR */
1.184 gully 2667: #ifdef _WINDOWS
1.152 kahan 2668: _putenv (tmpdir);
2669: #else
2670: putenv (tmpdir);
1.184 gully 2671: #endif /* _WINDOWS */
1.181 vatton 2672: TtaFreeMemory (tmpdir);
1.152 kahan 2673: }
2674: return (name);
1.154 kahan 2675: #endif
1.139 kahan 2676: }
Webmaster
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to AHTURLTools.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Public] / Amaya / amaya