Annotation of Amaya/amaya/AHTURLTools.c, revision 1.194
1.7 cvs 1: /*
2: *
1.190 vatton 3: * (c) COPYRIGHT INRIA and W3C, 1996-2005
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;
1.192 vatton 917: if (!strcmp (&temptype[i], "gif") ||
918: !strcmp (&temptype[i], "x-xbitmap") ||
919: !strcmp (&temptype[i], "x-xpixmap") ||
920: !strcmp (&temptype[i], "jpeg") ||
1.166 vatton 921: !strcmp (&temptype[i], "png"))
1.183 quint 922: ret = TRUE;
923: else
924: ret = FALSE;
925: TtaFreeMemory (temptype);
926: return (ret);
1.58 cvs 927: }
928:
929: /*----------------------------------------------------------------------
1.9 cvs 930: IsTextName
1.4 cvs 931: ----------------------------------------------------------------------*/
1.111 cvs 932: ThotBool IsTextName (const char *path)
1.106 cvs 933: {
1.183 quint 934: char *temppath;
935: char *suffix;
1.106 cvs 936: char nsuffix[MAX_LENGTH];
1.5 cvs 937: int i;
1.183 quint 938: ThotBool ret;
1.5 cvs 939:
940: if (!path)
1.13 cvs 941: return (FALSE);
1.5 cvs 942:
1.183 quint 943: temppath = TtaStrdup ((char *)path);
944: suffix = (char *)TtaGetMemory (strlen (path) + 1);
1.124 vatton 945: TtaExtractSuffix (temppath, suffix);
1.5 cvs 946:
947: /* Normalize the suffix */
948: i = 0;
1.106 cvs 949: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.5 cvs 950: {
1.25 cvs 951: nsuffix[i] = tolower (suffix[i]);
1.5 cvs 952: i++;
953: }
1.106 cvs 954: nsuffix[i] = EOS;
1.5 cvs 955:
1.192 vatton 956: if (!strcmp (nsuffix, "txt") ||
957: !strcmp (nsuffix, "dtd"))
1.183 quint 958: ret = TRUE;
1.106 cvs 959: else if (!strcmp (nsuffix, "gz"))
1.13 cvs 960: {
1.39 cvs 961: /* take into account compressed files */
1.124 vatton 962: TtaExtractSuffix (temppath, suffix);
1.13 cvs 963: /* Normalize the suffix */
964: i = 0;
1.106 cvs 965: while (suffix[i] != EOS && i < MAX_LENGTH -1)
1.13 cvs 966: {
1.25 cvs 967: nsuffix[i] = tolower (suffix[i]);
1.13 cvs 968: i++;
969: }
1.106 cvs 970: nsuffix[i] = EOS;
1.192 vatton 971: if (!strcmp (nsuffix, "txt") ||
972: !strcmp (nsuffix, "dtd"))
1.183 quint 973: ret = TRUE;
1.13 cvs 974: else
1.183 quint 975: ret = FALSE;
1.13 cvs 976: }
977: else
1.183 quint 978: ret = FALSE;
979:
980: TtaFreeMemory (temppath);
981: TtaFreeMemory (suffix);
982: return (ret);
1.3 cvs 983: }
984:
1.4 cvs 985: /*----------------------------------------------------------------------
1.9 cvs 986: IsHTTPPath
987: returns TRUE if path is in fact an http URL.
1.4 cvs 988: ----------------------------------------------------------------------*/
1.112 cvs 989: ThotBool IsHTTPPath (const char *path)
1.3 cvs 990: {
1.5 cvs 991: if (!path)
992: return FALSE;
1.3 cvs 993:
1.106 cvs 994: if ((!strncmp (path, "http:", 5) != 0)
995: || (AHTFTPURL_flag () && !strncmp (path, "ftp:", 4))
996: || !strncmp (path, "internal:", 9))
1.58 cvs 997: return TRUE;
998: return FALSE;
1.3 cvs 999: }
1000:
1.4 cvs 1001: /*----------------------------------------------------------------------
1.9 cvs 1002: IsWithParameters
1003: returns TRUE if url has a concatenated query string.
1.4 cvs 1004: ----------------------------------------------------------------------*/
1.133 vatton 1005: ThotBool IsWithParameters (const char *url)
1.3 cvs 1006: {
1.5 cvs 1007: int i;
1.3 cvs 1008:
1.9 cvs 1009: if ((!url) || (url[0] == EOS))
1.5 cvs 1010: return FALSE;
1.3 cvs 1011:
1.9 cvs 1012: i = strlen (url) - 1;
1013: while (i > 0 && url[i--] != '?')
1.5 cvs 1014: if (i < 0)
1015: return FALSE;
1.3 cvs 1016:
1.5 cvs 1017: /* There is a parameter */
1018: return TRUE;
1.3 cvs 1019: }
1020:
1.4 cvs 1021: /*----------------------------------------------------------------------
1.9 cvs 1022: IsW3Path
1023: returns TRUE if path is in fact a URL.
1.4 cvs 1024: ----------------------------------------------------------------------*/
1.133 vatton 1025: ThotBool IsW3Path (const char *path)
1.106 cvs 1026: {
1.170 quint 1027: if (path == NULL)
1028: return FALSE;
1.106 cvs 1029: if (strncmp (path, "http:", 5) &&
1030: strncmp (path, "ftp:", 4) &&
1031: strncmp (path, "telnet:", 7) &&
1032: strncmp (path, "wais:", 5) &&
1033: strncmp (path, "news:", 5) &&
1034: strncmp (path, "gopher:", 7) &&
1035: strncmp (path, "mailto:", 7) &&
1.132 cheyroul 1036: strncmp (path, "archie:", 7) &&
1037: strncmp (path, "https:", 6))
1.72 cvs 1038: return FALSE;
1039: return TRUE;
1.3 cvs 1040: }
1041:
1.4 cvs 1042: /*----------------------------------------------------------------------
1.90 cvs 1043: IsFilePath
1044: returns TRUE if path is in fact a URL.
1045: ----------------------------------------------------------------------*/
1.133 vatton 1046: ThotBool IsFilePath (const char *path)
1.90 cvs 1047: {
1.106 cvs 1048: if (strncmp (path, "file:", 5))
1.90 cvs 1049: return FALSE;
1050: return TRUE;
1051: }
1052:
1053: /*----------------------------------------------------------------------
1.9 cvs 1054: IsValidProtocol
1055: returns true if the url protocol is supported by Amaya.
1.4 cvs 1056: ----------------------------------------------------------------------*/
1.133 vatton 1057: ThotBool IsValidProtocol (const char *url)
1.106 cvs 1058: {
1059: if (!strncmp (url, "http:", 5)
1060: || !strncmp (url, "internal:", 9)
1061: || (AHTFTPURL_flag () && !strncmp (url, "ftp:", 4)))
1.22 cvs 1062: /* experimental */
1.24 cvs 1063: /*** || !strncmp (path, "news:", 5)***/
1.8 cvs 1064: return (TRUE);
1.5 cvs 1065: else
1.8 cvs 1066: return (FALSE);
1.3 cvs 1067: }
1068:
1.31 cvs 1069:
1070: /*----------------------------------------------------------------------
1071: GetBaseURL
1072: normalizes orgName according to a base associated with doc, and
1073: following the standard URL format rules.
1074: The function returns the base used to solve relative URL and SRC:
1075: - the base of the document,
1076: - or the document path (without document name).
1077: ----------------------------------------------------------------------*/
1.106 cvs 1078: char *GetBaseURL (Document doc)
1.31 cvs 1079: {
1080: Element el;
1081: ElementType elType;
1082: AttributeType attrType;
1083: Attribute attr;
1.176 vatton 1084: char *ptr, *basename, *utf8path;
1.31 cvs 1085: int length;
1.151 kahan 1086: ThotBool hasDocBase;
1.31 cvs 1087:
1.113 cvs 1088: if (doc == 0 || !DocumentURLs[doc])
1.110 cvs 1089: return NULL;
1.148 kahan 1090: /* the other functions expect basename to have no more than MAX_LENGTH chars */
1.171 gully 1091: basename = (char *)TtaGetMemory (MAX_LENGTH);
1.148 kahan 1092: basename[0] = EOS;
1.31 cvs 1093: length = MAX_LENGTH -1;
1.151 kahan 1094: hasDocBase = FALSE;
1095:
1096: /* If the document has a base URL, it has a priority over the headers. */
1097: /* @@ We need to do this too when we support XML:base */
1098:
1099: /* is it a HTML document ? */
1100: elType.ElSSchema = TtaGetDocumentSSchema (doc);
1101: if (!strcmp (TtaGetSSchemaName (elType.ElSSchema), "HTML"))
1102: /* it's a HTML document */
1103: {
1104: /* get the document element */
1105: el = TtaGetMainRoot (doc);
1106: /* search the BASE element */
1107: elType.ElTypeNum = HTML_EL_HEAD;
1108: el = TtaSearchTypedElement (elType, SearchForward, el);
1109: if (el)
1110: /* there is a HEAD element */
1111: {
1112: /* look for a BASE element within the HEAD */
1113: elType.ElTypeNum = HTML_EL_BASE;
1114: el = TtaSearchTypedElement (elType, SearchInTree, el);
1115: }
1116: if (el)
1117: {
1118: /* The document has a BASE element. Get the HREF attribute of the
1119: BASE element */
1120: hasDocBase = TRUE;
1121: attrType.AttrSSchema = elType.ElSSchema;
1122: attrType.AttrTypeNum = HTML_ATTR_HREF_;
1123: attr = TtaGetAttribute (el, attrType);
1124: if (attr)
1125: {
1126: /* Use the base path of the document */
1127: TtaGiveTextAttributeValue (attr, basename, &length);
1.176 vatton 1128: utf8path = basename;
1129: basename = (char *)TtaConvertMbsToByte ((unsigned char *)utf8path,
1130: TtaGetDefaultCharset ());
1131: TtaFreeMemory (utf8path);
1.151 kahan 1132: }
1133: }
1134: }
1135:
1136: /* there was no BASE. Do we have a location header? */
1137: if (!hasDocBase && DocumentMeta[doc] && DocumentMeta[doc]->full_content_location
1.148 kahan 1138: && DocumentMeta[doc]->full_content_location[0] != EOS)
1.65 cvs 1139: {
1.148 kahan 1140: strncpy (basename, DocumentMeta[doc]->full_content_location, MAX_LENGTH-1);
1141: basename[MAX_LENGTH-1] = EOS;
1142: length = strlen (basename);
1143: }
1144:
1145: if (basename[0] != EOS)
1146: {
1147: /* base and orgName have to be separated by a DIR_SEP */
1148: length--;
1149: if (basename[0] != EOS && basename[length] != URL_SEP &&
1150: basename[length] != DIR_SEP)
1151: /* verify if the base has the form "protocol://server:port" */
1.31 cvs 1152: {
1.148 kahan 1153: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS |
1154: AMAYA_PARSE_HOST |
1155: AMAYA_PARSE_PUNCTUATION);
1156: if (ptr && !strcmp (ptr, basename))
1.31 cvs 1157: {
1.148 kahan 1158: /* it has this form, complete it by adding a URL_STR */
1159: if (strchr (basename, DIR_SEP))
1160: strcat (basename, DIR_STR);
1161: else
1162: strcat (basename, URL_STR);
1163: length++;
1.31 cvs 1164: }
1.149 kahan 1165: else if (!ptr || ptr[0] == EOS)
1166: {
1167: /* no host was detected, we may have a relative URL. We test
1168: if it begins with a URL_SEP, DIR_SEP or period. If yes, it's
1169: relative. */
1170: if (! (basename[0] == '.' || basename[0] == URL_SEP
1171: || basename[0] == DIR_SEP))
1172: basename[0] = EOS;
1173: }
1.148 kahan 1174: if (ptr)
1175: TtaFreeMemory (ptr);
1.31 cvs 1176: }
1.113 cvs 1177: }
1178:
1.148 kahan 1179: /* there was no base element and no location header, we use the DocumentURL */
1180: if (basename[0] == EOS)
1181: {
1182: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1);
1183: basename[MAX_LENGTH-1] = EOS;
1184: }
1185:
1.31 cvs 1186: /* Remove anything after the last DIR_SEP char. If no such char is found,
1187: * then search for the first ":" char, hoping that what's before that is a
1188: * protocol. If found, end the string there. If neither char is found,
1189: * then discard the whole base element.
1190: */
1.106 cvs 1191: length = strlen (basename) - 1;
1.31 cvs 1192: /* search for the last DIR_SEP char */
1.106 cvs 1193: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP)
1.31 cvs 1194: length--;
1195: if (length >= 0)
1196: /* found the last DIR_SEP char, end the string there */
1.106 cvs 1197: basename[length + 1] = EOS;
1.31 cvs 1198: else
1199: /* search for the first PATH_STR char */
1200: {
1.106 cvs 1201: for (length = 0; basename[length] != ':' &&
1202: basename[length] != EOS; length ++);
1203: if (basename[length] == ':')
1.31 cvs 1204: /* found, so end the string there */
1.106 cvs 1205: basename[length + 1] = EOS;
1.31 cvs 1206: else
1207: /* not found, discard the base */
1.106 cvs 1208: basename[0] = EOS;
1.31 cvs 1209: }
1210: return (basename);
1211: }
1212:
1213:
1.4 cvs 1214: /*----------------------------------------------------------------------
1.40 cvs 1215: GetLocalPath
1216: Allocate and return the local document path associated to the url
1217: ----------------------------------------------------------------------*/
1.150 vatton 1218: char *GetLocalPath (Document doc, char *url)
1.106 cvs 1219: {
1220: char *ptr;
1221: char *n;
1222: char *documentname;
1223: char url_sep;
1.83 cvs 1224: int len;
1.67 cvs 1225: ThotBool noFile;
1.40 cvs 1226:
1.153 vatton 1227: if (url)
1.40 cvs 1228: {
1229: /* check whether the file name exists */
1.106 cvs 1230: len = strlen (url) - 1;
1.71 cvs 1231: if (IsW3Path (url))
1.157 kahan 1232: url_sep = URL_SEP;
1.41 cvs 1233: else
1.106 cvs 1234: url_sep = DIR_SEP;
1.41 cvs 1235: noFile = (url[len] == url_sep);
1.40 cvs 1236: if (noFile)
1.106 cvs 1237: url[len] = EOS;
1.171 gully 1238: ptr = (char *)TtaGetMemory (MAX_LENGTH);
1239: documentname = (char *)TtaGetMemory (MAX_LENGTH);
1.78 cvs 1240: TtaExtractName (url, ptr, documentname);
1.106 cvs 1241: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR);
1.40 cvs 1242: if (!TtaCheckDirectory (ptr))
1243: /* directory did not exist */
1.72 cvs 1244: TtaMakeDirectory (ptr);
1.47 cvs 1245:
1.153 vatton 1246: if (doc == 0)
1247: {
1248: n = strrchr (documentname, '.');
1249: if (n)
1250: *n = EOS;
1251: if (documentname[0] == EOS)
1252: strcpy (documentname, "noname");
1253: n = GetTempName (ptr, documentname);
1254: TtaFreeMemory (ptr);
1255: ptr = n;
1256: }
1.69 cvs 1257: else
1.153 vatton 1258: {
1259: /* don't include the query string within document name */
1260: n = strrchr (documentname, '?');
1261: if (n)
1262: *n = EOS;
1263: /* don't include ':' within document name */
1264: n = strchr (documentname, ':');
1265: if (n)
1266: *n = EOS;
1267: /* if after all this operations document name
1268: is empty, let's use noname.html instead */
1269: if (documentname[0] == EOS)
1270: strcat (ptr, "noname.html");
1271: else
1272: strcat (ptr, documentname);
1273: }
1.40 cvs 1274: TtaFreeMemory (documentname);
1.157 kahan 1275: /* substitute invalid chars in file names by a _ */
1.188 gully 1276: n = strrchr(ptr,DIR_SEP);
1.157 kahan 1277: while (*n)
1278: {
1279: if (*n == '*' || *n == ',')
1280: *n = '_';
1.187 cvs 1281: if ((unsigned char)*n >= 0x80) /* avoid non-ASCII */
1282: *n = 'A' + ((unsigned char)*n % 26);
1.157 kahan 1283: n++;
1284: }
1.40 cvs 1285: /* restore the url */
1286: if (noFile)
1.41 cvs 1287: url[len] = url_sep;
1.40 cvs 1288: return (ptr);
1289: }
1290: else
1291: return (NULL);
1292: }
1293:
1.73 cvs 1294: /*----------------------------------------------------------------------
1.79 cvs 1295: ExtractTarget extract the target name from document nane.
1296: ----------------------------------------------------------------------*/
1.150 vatton 1297: void ExtractTarget (char *aName, char *target)
1.79 cvs 1298: {
1.194 ! cvs 1299: long int lg, i;
1.106 cvs 1300: char *ptr;
1301: char *oldptr;
1.79 cvs 1302:
1303: if (!target || !aName)
1304: /* bad target */
1305: return;
1306:
1.106 cvs 1307: target[0] = EOS;
1308: lg = strlen (aName);
1.79 cvs 1309: if (lg)
1310: {
1311: /* the name is not empty */
1312: oldptr = ptr = &aName[0];
1313: do
1314: {
1.106 cvs 1315: ptr = strrchr (oldptr, '#');
1.79 cvs 1316: if (ptr)
1317: oldptr = &ptr[1];
1318: }
1319: while (ptr);
1320:
1.194 ! cvs 1321: i = (long int) (oldptr) - (long int) (aName); /* name length */
1.79 cvs 1322: if (i > 1)
1323: {
1.106 cvs 1324: aName[i - 1] = EOS;
1.79 cvs 1325: if (i != lg)
1.106 cvs 1326: strcpy (target, oldptr);
1.79 cvs 1327: }
1328: }
1329: }
1330:
1331: /*----------------------------------------------------------------------
1.90 cvs 1332: RemoveNewLines (text)
1333: Removes any '\n' chars that are found in text.
1334: Returns TRUE if it did the operation, FALSE otherwise.
1.73 cvs 1335: ----------------------------------------------------------------------*/
1.106 cvs 1336: ThotBool RemoveNewLines (char *text)
1337: {
1338: ThotBool change = FALSE;
1339: char *src;
1340: char *dest;
1.90 cvs 1341:
1342: src = text;
1343: dest = text;
1.115 kahan 1344:
1345: /* remove any preceding whitespace */
1346: while (*src && *src == ' ')
1347: {
1348: src++;
1349: change = 1;
1350: }
1351:
1.90 cvs 1352: while (*src)
1353: {
1354: switch (*src)
1355: {
1.106 cvs 1356: case '\n':
1.90 cvs 1357: /* don't copy the newline */
1358: change = 1;
1359: break;
1360: default:
1361: *dest = *src;
1362: dest++;
1363: break;
1364: }
1365: src++;
1366: }
1367: /* copy the last EOS char */
1368: *dest = *src;
1369:
1370: return (change);
1371: }
1372:
1373: /*----------------------------------------------------------------------
1374: CleanCopyFileURL
1375: Copies a file url from a src string to destination string.
1.97 cvs 1376: convertion says which type of convertion (none, %xx, URL_SEP into DIR_SEP
1377: we want to do).
1.90 cvs 1378: ----------------------------------------------------------------------*/
1.106 cvs 1379: static void CleanCopyFileURL (char *dest, char *src,
1380: ConvertionType convertion)
1.90 cvs 1381: {
1382: while (*src)
1.89 cvs 1383: {
1.90 cvs 1384: switch (*src)
1.89 cvs 1385: {
1.184 gully 1386: #ifdef _WINDOWS
1.106 cvs 1387: case URL_SEP:
1.96 cvs 1388: /* make DIR_SEP transformation */
1.97 cvs 1389: if (convertion & AM_CONV_URL_SEP)
1.106 cvs 1390: *dest = DIR_SEP;
1.96 cvs 1391: else
1392: *dest = *src;
1.90 cvs 1393: dest++;
1.96 cvs 1394: src++;
1.90 cvs 1395: break;
1.184 gully 1396: #endif /* _WINDOWS */
1.96 cvs 1397:
1.106 cvs 1398: case '%':
1.97 cvs 1399: if (convertion & AM_CONV_PERCENT)
1.96 cvs 1400: {
1.97 cvs 1401: /* (code adapted from libwww's HTUnEscape function */
1.96 cvs 1402: src++;
1.106 cvs 1403: if (*src != EOS)
1.97 cvs 1404: {
1405: *dest = UnEscapeChar (*src) * 16;
1406: src++;
1407: }
1.106 cvs 1408: if (*src != EOS)
1.97 cvs 1409: {
1410: *dest = *dest + UnEscapeChar (*src);
1411: src++;
1412: }
1413: dest++;
1.96 cvs 1414: }
1.97 cvs 1415: else
1.96 cvs 1416: {
1.97 cvs 1417: *dest = *src;
1418: dest++;
1.96 cvs 1419: src++;
1420: }
1421: break;
1422:
1.90 cvs 1423: default:
1424: *dest = *src;
1.89 cvs 1425: dest++;
1.96 cvs 1426: src++;
1.90 cvs 1427: break;
1.89 cvs 1428: }
1429: }
1.90 cvs 1430: /* copy the EOS char */
1431: *dest = *src;
1.73 cvs 1432: }
1.40 cvs 1433:
1434: /*----------------------------------------------------------------------
1.9 cvs 1435: NormalizeURL
1436: normalizes orgName according to a base associated with doc, and
1437: following the standard URL format rules.
1.113 cvs 1438: if doc is < 0, use as a base the URL of the document that contains
1439: (or contained) the elements that are now in the copy/cut buffer.
1.53 cvs 1440: if doc is 0 and otherPath not NULL, normalizes orgName according to this
1441: other path.
1.9 cvs 1442: The function returns the new complete and normalized URL
1.12 cvs 1443: or file name path (newName) and the name of the document (docName).
1.9 cvs 1444: N.B. If the function can't find out what's the docName, it assigns
1445: the name "noname.html".
1.4 cvs 1446: ----------------------------------------------------------------------*/
1.106 cvs 1447: void NormalizeURL (char *orgName, Document doc, char *newName,
1448: char *docName, char *otherPath)
1449: {
1450: char *basename;
1451: char tempOrgName[MAX_LENGTH];
1452: char *ptr;
1453: char used_sep;
1.84 cvs 1454: int length;
1455: ThotBool check;
1.5 cvs 1456:
1.184 gully 1457: #ifdef _WINDOWS
1.44 cvs 1458: int ndx;
1.184 gully 1459: #endif /* _WINDOWS */
1.44 cvs 1460:
1.5 cvs 1461: if (!newName || !docName)
1462: return;
1.18 cvs 1463:
1.113 cvs 1464: if (doc < 0)
1465: basename = TtaStrdup (SavedDocumentURL);
1466: else if (doc > 0)
1.53 cvs 1467: basename = GetBaseURL (doc);
1468: else if (otherPath != NULL)
1.108 cvs 1469: basename = TtaStrdup (otherPath);
1.32 cvs 1470: else
1.53 cvs 1471: basename = NULL;
1.32 cvs 1472:
1.18 cvs 1473: /*
1.31 cvs 1474: * Clean orgName
1475: * Make sure we have a complete orgName, without any leading or trailing
1476: * white spaces, or trailinbg new lines
1477: */
1.5 cvs 1478: ptr = orgName;
1.18 cvs 1479: /* skip leading white space and new line characters */
1.106 cvs 1480: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS);
1481: strncpy (tempOrgName, ptr, MAX_LENGTH -1);
1482: tempOrgName[MAX_LENGTH -1] = EOS;
1.18 cvs 1483: /*
1.31 cvs 1484: * Make orgName a complete URL
1485: * If the URL does not include a protocol, then try to calculate
1486: * one using the doc's base element (if it exists),
1487: */
1.106 cvs 1488: if (tempOrgName[0] == EOS)
1.53 cvs 1489: {
1.106 cvs 1490: newName[0] = EOS;
1491: docName[0] = EOS;
1.53 cvs 1492: TtaFreeMemory (basename);
1493: return;
1494: }
1.49 cvs 1495:
1496: /* clean trailing white space */
1.106 cvs 1497: length = strlen (tempOrgName) - 1;
1498: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL)
1.53 cvs 1499: {
1.106 cvs 1500: tempOrgName[length] = EOS;
1.53 cvs 1501: length--;
1502: }
1.50 cvs 1503:
1.55 cvs 1504: /* remove extra dot (which dot???) */
1505: /* ugly, but faster than a strcmp */
1.106 cvs 1506: if (tempOrgName[length] == '.'
1507: && (length == 0 || tempOrgName[length-1] != '.'))
1508: tempOrgName[length] = EOS;
1.50 cvs 1509:
1.94 cvs 1510: if (IsW3Path (tempOrgName))
1.53 cvs 1511: {
1512: /* the name is complete, go to the Sixth Step */
1.106 cvs 1513: strcpy (newName, tempOrgName);
1.53 cvs 1514: SimplifyUrl (&newName);
1515: /* verify if the URL has the form "protocol://server:port" */
1.110 cvs 1516: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS |
1517: AMAYA_PARSE_HOST |
1518: AMAYA_PARSE_PUNCTUATION);
1519: if (ptr && !strcmp (ptr, newName))
1520: /* it has this form, we complete it by adding a DIR_STR */
1.106 cvs 1521: strcat (newName, URL_STR);
1.49 cvs 1522:
1.53 cvs 1523: if (ptr)
1.50 cvs 1524: TtaFreeMemory (ptr);
1.53 cvs 1525: }
1.113 cvs 1526: else if (basename == NULL)
1.53 cvs 1527: /* the name is complete, go to the Sixth Step */
1.106 cvs 1528: strcpy (newName, tempOrgName);
1.53 cvs 1529: else
1530: {
1.31 cvs 1531: /* Calculate the absolute URL, using the base or document URL */
1.184 gully 1532: #ifdef _WINDOWS
1.53 cvs 1533: if (!IsW3Path (basename))
1534: {
1.106 cvs 1535: length = strlen (tempOrgName);
1.53 cvs 1536: for (ndx = 0; ndx < length; ndx++)
1.106 cvs 1537: if (tempOrgName [ndx] == '/')
1538: tempOrgName [ndx] = '\\';
1.53 cvs 1539: }
1.184 gully 1540: #endif /* _WINDOWS */
1.25 cvs 1541: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL);
1.53 cvs 1542: if (ptr)
1543: {
1544: SimplifyUrl (&ptr);
1.106 cvs 1545: strcpy (newName, ptr);
1.53 cvs 1546: TtaFreeMemory (ptr);
1547: }
1548: else
1.106 cvs 1549: newName[0] = EOS;
1.53 cvs 1550: }
1.36 cvs 1551:
1552: TtaFreeMemory (basename);
1.18 cvs 1553: /*
1.31 cvs 1554: * Prepare the docname that will refer to this ressource in the
1555: * .amaya directory. If the new URL finishes on DIR_SEP, then use
1556: * noname.html as a default ressource name
1.18 cvs 1557: */
1.106 cvs 1558: if (newName[0] != EOS)
1.53 cvs 1559: {
1.106 cvs 1560: length = strlen (newName) - 1;
1561: if (newName[length] == URL_SEP || newName[length] == DIR_SEP)
1.53 cvs 1562: {
1563: used_sep = newName[length];
1564: check = TRUE;
1565: while (check)
1566: {
1.50 cvs 1567: length--;
1568: while (length >= 0 && newName[length] != used_sep)
1.53 cvs 1569: length--;
1.106 cvs 1570: if (!strncmp (&newName[length+1], "..", 2))
1.53 cvs 1571: {
1.106 cvs 1572: newName[length+1] = EOS;
1.53 cvs 1573: /* remove also previous directory */
1574: length--;
1575: while (length >= 0 && newName[length] != used_sep)
1576: length--;
1.106 cvs 1577: if (strncmp (&newName[length+1], "//", 2))
1.131 cheyroul 1578: /* don't remove server name */
1.106 cvs 1579: newName[length+1] = EOS;
1.53 cvs 1580: }
1.106 cvs 1581: else if (!strncmp (&newName[length+1], ".", 1))
1582: newName[length+1] = EOS;
1.50 cvs 1583: else
1.53 cvs 1584: check = FALSE;
1585: }
1586: /* docname was not comprised inside the URL, so let's */
1587: /* assign the default ressource name */
1.106 cvs 1588: strcpy (docName, "noname.html");
1.53 cvs 1589: }
1590: else
1591: { /* docname is comprised inside the URL */
1.110 cvs 1592: while (length >= 0 && newName[length] != URL_SEP &&
1593: newName[length] != DIR_SEP)
1.53 cvs 1594: length--;
1595: if (length < 0)
1.106 cvs 1596: strcpy (docName, newName);
1.53 cvs 1597: else
1.106 cvs 1598: strcpy (docName, &newName[length+1]);
1.53 cvs 1599: }
1600: }
1601: else
1.106 cvs 1602: docName[0] = EOS;
1.18 cvs 1603: }
1.3 cvs 1604:
1.4 cvs 1605: /*----------------------------------------------------------------------
1.9 cvs 1606: IsSameHost
1.4 cvs 1607: ----------------------------------------------------------------------*/
1.106 cvs 1608: ThotBool IsSameHost (const char *url1, const char *url2)
1.3 cvs 1609: {
1.106 cvs 1610: char *basename_ptr1, *basename_ptr2;
1611: ThotBool result;
1.3 cvs 1612:
1.106 cvs 1613: basename_ptr1 = AmayaParseUrl (url1, "",
1614: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1615: basename_ptr2 = AmayaParseUrl (url2, "",
1616: AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION);
1.3 cvs 1617:
1.106 cvs 1618: if (strcmp (basename_ptr1, basename_ptr2))
1619: result = FALSE;
1620: else
1621: result = TRUE;
1622: TtaFreeMemory (basename_ptr1);
1623: TtaFreeMemory (basename_ptr2);
1624: return (result);
1.3 cvs 1625: }
1626:
1627:
1.4 cvs 1628: /*----------------------------------------------------------------------
1.22 cvs 1629: HasKnownFileSuffix
1630: returns TRUE if path points to a file ending with a suffix.
1631: ----------------------------------------------------------------------*/
1.153 vatton 1632: ThotBool HasKnownFileSuffix (const char *path)
1.106 cvs 1633: {
1634: char *root;
1635: char temppath[MAX_LENGTH];
1636: char suffix[MAX_LENGTH];
1.22 cvs 1637:
1.106 cvs 1638: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP)
1.22 cvs 1639: return (FALSE);
1640:
1.106 cvs 1641: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION);
1.22 cvs 1642:
1643: if (root)
1644: {
1.106 cvs 1645: strcpy (temppath, root);
1.25 cvs 1646: TtaFreeMemory (root);
1.22 cvs 1647: /* Get the suffix */
1.124 vatton 1648: TtaExtractSuffix (temppath, suffix);
1.22 cvs 1649:
1.106 cvs 1650: if( suffix[0] == EOS)
1.22 cvs 1651: /* no suffix */
1652: return (FALSE);
1653:
1654: /* Normalize the suffix */
1655: ConvertToLowerCase (suffix);
1656:
1.106 cvs 1657: if (!strcmp (suffix, "gz"))
1.22 cvs 1658: /* skip the compressed suffix */
1659: {
1.124 vatton 1660: TtaExtractSuffix (temppath, suffix);
1.106 cvs 1661: if(suffix[0] == EOS)
1.22 cvs 1662: /* no suffix */
1663: return (FALSE);
1664: /* Normalize the suffix */
1665: ConvertToLowerCase (suffix);
1666: }
1667:
1.106 cvs 1668: if (strcmp (suffix, "gif") &&
1669: strcmp (suffix, "xbm") &&
1670: strcmp (suffix, "xpm") &&
1671: strcmp (suffix, "jpg") &&
1672: strcmp (suffix, "pdf") &&
1673: strcmp (suffix, "png") &&
1674: strcmp (suffix, "tgz") &&
1675: strcmp (suffix, "xpg") &&
1676: strcmp (suffix, "xpd") &&
1677: strcmp (suffix, "ps") &&
1678: strcmp (suffix, "au") &&
1679: strcmp (suffix, "html") &&
1680: strcmp (suffix, "htm") &&
1681: strcmp (suffix, "shtml") &&
1682: strcmp (suffix, "xht") &&
1683: strcmp (suffix, "xhtm") &&
1684: strcmp (suffix, "xhtml") &&
1685: strcmp (suffix, "txt") &&
1686: strcmp (suffix, "css") &&
1687: strcmp (suffix, "eps"))
1.22 cvs 1688: return (FALSE);
1689: else
1690: return (TRUE);
1691: }
1692: else
1693: return (FALSE);
1694: }
1695:
1696:
1697: /*----------------------------------------------------------------------
1.24 cvs 1698: ChopURL
1699: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL).
1700: If inputURL is bigger than that size, outputURL receives
1701: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...",
1702: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL.
1703: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets
1704: copied into outputURL.
1705: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH
1706: chars.
1707: ----------------------------------------------------------------------*/
1.106 cvs 1708: void ChopURL (char *outputURL, const char *inputURL)
1.24 cvs 1709: {
1710: int len;
1.9 cvs 1711:
1.106 cvs 1712: len = strlen (inputURL);
1.24 cvs 1713: if (len <= MAX_PRINT_URL_LENGTH)
1.106 cvs 1714: strcpy (outputURL, inputURL);
1.24 cvs 1715: else
1716: /* make a truncated urlName on the status window */
1717: {
1.106 cvs 1718: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2);
1719: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS;
1720: strcat (outputURL, "...");
1721: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ]));
1.24 cvs 1722: }
1.25 cvs 1723: }
1724:
1725:
1726: /*----------------------------------------------------------------------
1727: scan
1.47 cvs 1728: Scan a filename for its constituents
1.25 cvs 1729: -----------------------------------
1730:
1731: On entry,
1732: name points to a document name which may be incomplete.
1733: On exit,
1734: absolute or relative may be nonzero (but not both).
1735: host, fragment and access may be nonzero if they were specified.
1736: Any which are nonzero point to zero terminated strings.
1737: ----------------------------------------------------------------------*/
1.106 cvs 1738: static void scan (char *name, HTURI *parts)
1.25 cvs 1739: {
1.106 cvs 1740: char * p;
1741: char * after_access = name;
1.32 cvs 1742:
1.43 cvs 1743: memset (parts, '\0', sizeof (HTURI));
1.28 cvs 1744: /* Look for fragment identifier */
1.106 cvs 1745: if ((p = strchr(name, '#')) != NULL)
1.28 cvs 1746: {
1.106 cvs 1747: *p++ = '\0';
1.28 cvs 1748: parts->fragment = p;
1.25 cvs 1749: }
1750:
1.28 cvs 1751: for (p=name; *p; p++)
1752: {
1.106 cvs 1753: if (*p == URL_SEP || *p == DIR_SEP || *p == '#' || *p == '?')
1.28 cvs 1754: break;
1.106 cvs 1755: if (*p == ':')
1.28 cvs 1756: {
1757: *p = 0;
1758: parts->access = after_access; /* Scheme has been specified */
1759:
1760: /* The combination of gcc, the "-O" flag and the HP platform is
1761: unhealthy. The following three lines is a quick & dirty fix, but is
1762: not recommended. Rather, turn off "-O". */
1763:
1764: /* after_access = p;*/
1765: /* while (*after_access == 0)*/
1766: /* after_access++;*/
1767: after_access = p+1;
1.106 cvs 1768: if (!strcasecmp("URL", parts->access))
1.28 cvs 1769: /* Ignore IETF's URL: pre-prefix */
1770: parts->access = NULL;
1771: else
1.25 cvs 1772: break;
1773: }
1774: }
1775:
1776: p = after_access;
1.43 cvs 1777: if (*p == URL_SEP || *p == DIR_SEP)
1.28 cvs 1778: {
1.43 cvs 1779: if (p[1] == URL_SEP)
1.28 cvs 1780: {
1.25 cvs 1781: parts->host = p+2; /* host has been specified */
1.28 cvs 1782: *p = 0; /* Terminate access */
1783: /* look for end of host name if any */
1.106 cvs 1784: p = strchr (parts->host, URL_SEP);
1.28 cvs 1785: if (p)
1786: {
1.106 cvs 1787: *p = EOS; /* Terminate host */
1.25 cvs 1788: parts->absolute = p+1; /* Root has been found */
1.28 cvs 1789: }
1790: }
1791: else
1792: /* Root found but no host */
1793: parts->absolute = p+1;
1794: }
1795: else
1796: {
1.25 cvs 1797: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */
1.28 cvs 1798: }
1.25 cvs 1799: }
1800:
1801:
1802: /*----------------------------------------------------------------------
1.28 cvs 1803: AmayaParseUrl: parse a Name relative to another name
1804:
1805: This returns those parts of a name which are given (and requested)
1806: substituting bits from the related name where necessary.
1.25 cvs 1807:
1.28 cvs 1808: On entry,
1.25 cvs 1809: aName A filename given
1810: relatedName A name relative to which aName is to be parsed. Give
1811: it an empty string if aName is absolute.
1812: wanted A mask for the bits which are wanted.
1813:
1.28 cvs 1814: On exit,
1.25 cvs 1815: returns A pointer to a malloc'd string which MUST BE FREED
1816: ----------------------------------------------------------------------*/
1.106 cvs 1817: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted)
1818: {
1819: char *return_value;
1820: char result[MAX_LENGTH];
1821: char name[MAX_LENGTH];
1822: char rel[MAX_LENGTH];
1823: char *p, *access;
1.29 cvs 1824: HTURI given, related;
1825: int len;
1.106 cvs 1826: char used_sep;
1827: char *used_str;
1.32 cvs 1828:
1.106 cvs 1829: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP))
1.33 cvs 1830: {
1.106 cvs 1831: used_str = DIR_STR;
1832: used_sep = DIR_SEP;
1.33 cvs 1833: }
1.32 cvs 1834: else
1.33 cvs 1835: {
1.106 cvs 1836: used_str = URL_STR;
1837: used_sep = URL_SEP;
1.33 cvs 1838: }
1.32 cvs 1839:
1.29 cvs 1840: /* Make working copies of input strings to cut up: */
1841: return_value = NULL;
1842: result[0] = 0; /* Clear string */
1.169 quint 1843: rel[0] = EOS;
1844: strncpy (name, aName, MAX_LENGTH - 1);
1845: name[MAX_LENGTH - 1] = EOS;
1846: if (relatedName != NULL)
1847: {
1848: strncpy (rel, relatedName, MAX_LENGTH - 1);
1849: rel[MAX_LENGTH - 1] = EOS;
1850: }
1.29 cvs 1851: else
1.106 cvs 1852: relatedName[0] = EOS;
1.29 cvs 1853:
1854: scan (name, &given);
1855: scan (rel, &related);
1856: access = given.access ? given.access : related.access;
1857: if (wanted & AMAYA_PARSE_ACCESS)
1858: if (access)
1859: {
1.106 cvs 1860: strcat (result, access);
1.29 cvs 1861: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1862: strcat (result, ":");
1.29 cvs 1863: }
1864:
1865: if (given.access && related.access)
1866: /* If different, inherit nothing. */
1.106 cvs 1867: if (strcmp (given.access, related.access) != 0)
1.29 cvs 1868: {
1869: related.host = 0;
1870: related.absolute = 0;
1871: related.relative = 0;
1872: related.fragment = 0;
1873: }
1874:
1875: if (wanted & AMAYA_PARSE_HOST)
1876: if(given.host || related.host)
1877: {
1878: if(wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1879: strcat (result, "//");
1880: strcat (result, given.host ? given.host : related.host);
1.29 cvs 1881: }
1882:
1883: if (given.host && related.host)
1884: /* If different hosts, inherit no path. */
1.106 cvs 1885: if (strcmp (given.host, related.host) != 0)
1.29 cvs 1886: {
1887: related.absolute = 0;
1888: related.relative = 0;
1889: related.fragment = 0;
1890: }
1891:
1892: if (wanted & AMAYA_PARSE_PATH)
1893: {
1894: if (given.absolute)
1895: {
1896: /* All is given */
1897: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1898: strcat (result, used_str);
1899: strcat (result, given.absolute);
1.25 cvs 1900: }
1.29 cvs 1901: else if (related.absolute)
1902: {
1903: /* Adopt path not name */
1.106 cvs 1904: strcat (result, used_str);
1905: strcat (result, related.absolute);
1.29 cvs 1906: if (given.relative)
1907: {
1908: /* Search part? */
1.106 cvs 1909: p = strchr (result, '?');
1.29 cvs 1910: if (!p)
1.106 cvs 1911: p=result+strlen(result)-1;
1.33 cvs 1912: for (; *p!=used_sep; p--); /* last / */
1.29 cvs 1913: /* Remove filename */
1914: p[1]=0;
1915: /* Add given one */
1.106 cvs 1916: strcat (result, given.relative);
1.25 cvs 1917: }
1918: }
1.29 cvs 1919: else if (given.relative)
1920: /* what we've got */
1.106 cvs 1921: strcat (result, given.relative);
1.29 cvs 1922: else if (related.relative)
1.106 cvs 1923: strcat (result, related.relative);
1.29 cvs 1924: else
1925: /* No inheritance */
1.106 cvs 1926: strcat (result, used_str);
1.25 cvs 1927: }
1.29 cvs 1928:
1929: if (wanted & AMAYA_PARSE_ANCHOR)
1930: if (given.fragment || related.fragment)
1931: {
1932: if (given.absolute && given.fragment)
1933: {
1934: /*Fixes for relURLs...*/
1935: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1936: strcat (result, "#");
1937: strcat (result, given.fragment);
1.29 cvs 1938: }
1939: else if (!(given.absolute) && !(given.fragment))
1.106 cvs 1940: strcat (result, "");
1.29 cvs 1941: else
1942: {
1.110 cvs 1943: if (wanted & AMAYA_PARSE_PUNCTUATION)
1.106 cvs 1944: strcat (result, "#");
1.110 cvs 1945: strcat (result, given.fragment ? given.fragment : related.fragment);
1.29 cvs 1946: }
1947: }
1.106 cvs 1948: len = strlen (result);
1.171 gully 1949: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 1950: strcpy (return_value, result);
1.29 cvs 1951: return (return_value); /* exactly the right length */
1.25 cvs 1952: }
1953:
1954: /*----------------------------------------------------------------------
1955: HTCanon
1956: Canonicalizes the URL in the following manner starting from the host
1957: pointer:
1958:
1959: 1) The host name is converted to lowercase
1960: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp)
1961:
1962: Return: OK The position of the current path part of the URL
1963: which might be the old one or a new one.
1964:
1965: ----------------------------------------------------------------------*/
1.106 cvs 1966: static char *HTCanon (char **filename, char *host)
1967: {
1968: char *newname = NULL;
1969: char used_sep;
1970: char *path;
1971: char *strptr;
1972: char *port;
1973: char *access = host-3;
1974:
1975: if (*filename && strchr (*filename, URL_SEP))
1976: used_sep = URL_SEP;
1.33 cvs 1977: else
1.106 cvs 1978: used_sep = DIR_SEP;
1.32 cvs 1979:
1.110 cvs 1980: while (access > *filename && *(access - 1) != used_sep) /* Find access method */
1.25 cvs 1981: access--;
1.110 cvs 1982: if ((path = strchr (host, used_sep)) == NULL) /* Find path */
1.106 cvs 1983: path = host + strlen (host);
1984: if ((strptr = strchr (host, '@')) != NULL && strptr < path) /* UserId */
1.82 cvs 1985: host = strptr;
1.110 cvs 1986: if ((port = strchr (host, ':')) != NULL && port > path) /* Port number */
1.82 cvs 1987: port = NULL;
1.25 cvs 1988:
1989: strptr = host; /* Convert to lower-case */
1.82 cvs 1990: while (strptr < path)
1.33 cvs 1991: {
1.123 vatton 1992: *strptr = tolower (*strptr);
1.82 cvs 1993: strptr++;
1.33 cvs 1994: }
1.25 cvs 1995:
1996: /* Does the URL contain a full domain name? This also works for a
1997: numerical host name. The domain name is already made lower-case
1998: and without a trailing dot. */
1999: {
1.106 cvs 2000: char *dot = port ? port : path;
2001: if (dot > *filename && *--dot == '.')
1.33 cvs 2002: {
1.106 cvs 2003: char *orig = dot;
2004: char *dest = dot + 1;
1.82 cvs 2005: while ((*orig++ = *dest++));
2006: if (port) port--;
1.33 cvs 2007: path--;
1.25 cvs 2008: }
2009: }
2010: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */
1.33 cvs 2011: if (port)
2012: {
1.82 cvs 2013: if (!*(port+1) || *(port+1) == used_sep)
1.33 cvs 2014: {
2015: if (!newname)
2016: {
1.106 cvs 2017: char *orig = port;
2018: char *dest = port + 1;
1.82 cvs 2019: while ((*orig++ = *dest++));
1.33 cvs 2020: }
2021: }
1.106 cvs 2022: else if ((!strncmp (access, "http", 4) &&
2023: (*(port + 1) == '8' &&
2024: *(port+2) == '0' &&
1.82 cvs 2025: (*(port+3) == used_sep || !*(port + 3)))) ||
1.106 cvs 2026: (!strncmp (access, "gopher", 6) &&
2027: (*(port+1) == '7' &&
2028: *(port+2) == '0' &&
1.82 cvs 2029: (*(port+3) == used_sep || !*(port+3)))) ||
1.106 cvs 2030: (!strncmp (access, "ftp", 3) &&
2031: (*(port+1) == '2' &&
2032: *(port + 2) == '1' &&
1.82 cvs 2033: (*(port+3) == used_sep || !*(port+3))))) {
1.33 cvs 2034: if (!newname)
2035: {
1.106 cvs 2036: char *orig = port;
2037: char *dest = port + 3;
1.33 cvs 2038: while((*orig++ = *dest++));
2039: /* Update path position, Henry Minsky */
2040: path -= 3;
1.25 cvs 2041: }
1.33 cvs 2042: }
2043: else if (newname)
1.106 cvs 2044: strncat (newname, port, (int) (path - port));
1.33 cvs 2045: }
1.25 cvs 2046:
1.33 cvs 2047: if (newname)
2048: {
1.106 cvs 2049: char *newpath = newname + strlen (newname);
2050: strcat (newname, path);
1.25 cvs 2051: path = newpath;
1.28 cvs 2052: /* Free old copy */
2053: TtaFreeMemory(*filename);
1.25 cvs 2054: *filename = newname;
1.33 cvs 2055: }
1.25 cvs 2056: return path;
2057: }
2058:
2059:
2060: /*----------------------------------------------------------------------
1.29 cvs 2061: SimplifyUrl: simplify a URI
1.32 cvs 2062: A URI is allowed to contain the sequence xxx/../ which may be
2063: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR.
1.28 cvs 2064: Simplification helps us recognize duplicate URIs.
1.25 cvs 2065:
1.28 cvs 2066: Thus, /etc/junk/../fred becomes /etc/fred
2067: /etc/junk/./fred becomes /etc/junk/fred
1.25 cvs 2068:
1.28 cvs 2069: but we should NOT change
2070: http://fred.xxx.edu/../..
1.25 cvs 2071:
2072: or ../../albert.html
2073:
1.28 cvs 2074: In order to avoid empty URLs the following URLs become:
1.25 cvs 2075:
2076: /fred/.. becomes /fred/..
2077: /fred/././.. becomes /fred/..
2078: /fred/.././junk/.././ becomes /fred/..
2079:
1.28 cvs 2080: If more than one set of `://' is found (several proxies in cascade) then
2081: only the part after the last `://' is simplified.
1.25 cvs 2082:
1.28 cvs 2083: Returns: A string which might be the old one or a new one.
1.25 cvs 2084: ----------------------------------------------------------------------*/
1.106 cvs 2085: void SimplifyUrl (char **url)
2086: {
2087: char *path;
2088: char *access;
2089: char *newptr;
2090: char *p;
2091: char *orig, *dest, *end;
1.28 cvs 2092:
1.106 cvs 2093: char used_sep;
1.77 cvs 2094: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation:
2095: something/../ simplification in relative URLs when they start with a ../ */
1.32 cvs 2096:
1.28 cvs 2097: if (!url || !*url)
2098: return;
2099:
1.106 cvs 2100: if (strchr (*url, URL_SEP))
2101: used_sep = URL_SEP;
1.32 cvs 2102: else
1.106 cvs 2103: used_sep = DIR_SEP;
1.32 cvs 2104:
1.77 cvs 2105: /* should we simplify double dot? */
2106: path = *url;
1.106 cvs 2107: if (*path == '.' && *(path + 1) == '.')
1.77 cvs 2108: ddot_simplify = FALSE;
2109: else
2110: ddot_simplify = TRUE;
2111:
1.28 cvs 2112: /* Find any scheme name */
1.106 cvs 2113: if ((path = strstr (*url, "://")) != NULL)
1.33 cvs 2114: {
2115: /* Find host name */
1.28 cvs 2116: access = *url;
1.123 vatton 2117: while (access < path && (*access = tolower (*access)))
1.82 cvs 2118: access++;
1.28 cvs 2119: path += 3;
1.106 cvs 2120: while ((newptr = strstr (path, "://")) != NULL)
1.82 cvs 2121: /* For proxies */
1.106 cvs 2122: path = newptr + 3;
1.82 cvs 2123: /* We have a host name */
1.84 cvs 2124: path = HTCanon (url, path);
1.25 cvs 2125: }
1.106 cvs 2126: else if ((path = strstr (*url, ":/")) != NULL)
1.28 cvs 2127: path += 2;
2128: else
2129: path = *url;
1.84 cvs 2130: if (*path == used_sep && *(path+1) == used_sep)
1.28 cvs 2131: /* Some URLs start //<foo> */
2132: path += 1;
1.94 cvs 2133: else if (IsFilePath (path))
2134: {
2135: /* doesn't need to do anything more */
2136: return;
2137: }
1.106 cvs 2138: else if (!strncmp (path, "news:", 5))
1.28 cvs 2139: {
1.106 cvs 2140: newptr = strchr (path+5, '@');
1.28 cvs 2141: if (!newptr)
2142: newptr = path + 5;
2143: while (*newptr)
2144: {
2145: /* Make group or host lower case */
1.123 vatton 2146: *newptr = tolower (*newptr);
1.28 cvs 2147: newptr++;
1.25 cvs 2148: }
1.28 cvs 2149: /* Doesn't need to do any more */
2150: return;
1.25 cvs 2151: }
1.126 cheyroul 2152:
1.189 vatton 2153: /* remove initial multiple /'s */
2154: p = path - 1;
2155: if (path != *url && *p == used_sep && *path == used_sep)
2156: {
2157: while (*path == used_sep)
2158: {
2159: orig = path;
2160: dest = path + 1;
2161: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2162: end = orig-1;
2163: }
2164: }
2165:
2166: if (path)
1.28 cvs 2167: {
1.106 cvs 2168: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) ||
2169: (end = strchr (path, '#'))))
2170: end = path + strlen (path);
1.28 cvs 2171:
2172: /* Parse string second time to simplify */
2173: p = path;
2174: while (p < end)
2175: {
1.110 cvs 2176: /* if we're pointing to a char, it's safe to reactivate the
2177: ../ convertion */
1.106 cvs 2178: if (!ddot_simplify && *p != '.' && *p != used_sep)
1.77 cvs 2179: ddot_simplify = TRUE;
2180:
1.33 cvs 2181: if (*p==used_sep)
1.28 cvs 2182: {
1.106 cvs 2183: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2)))
1.28 cvs 2184: {
2185: orig = p + 1;
1.84 cvs 2186: dest = (*(p+2) != used_sep) ? p+2 : p+3;
1.52 cvs 2187: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/
1.28 cvs 2188: end = orig - 1;
2189: }
1.106 cvs 2190: else if (ddot_simplify && *(p+1) == '.' && *(p+2) == '.'
1.77 cvs 2191: && (*(p+3) == used_sep || !*(p+3)))
1.28 cvs 2192: {
2193: newptr = p;
1.52 cvs 2194: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */
2195: if (*newptr == used_sep)
2196: orig = newptr + 1;
1.28 cvs 2197: else
1.52 cvs 2198: orig = newptr;
2199:
2200: dest = (*(p+3) != used_sep) ? p+3 : p+4;
2201: while ((*orig++ = *dest++)); /* Remove /xxx/.. */
2202: end = orig-1;
2203: /* Start again with prev slash */
2204: p = newptr;
1.28 cvs 2205: }
1.33 cvs 2206: else if (*(p+1) == used_sep)
1.28 cvs 2207: {
1.33 cvs 2208: while (*(p+1) == used_sep)
1.28 cvs 2209: {
2210: orig = p;
2211: dest = p + 1;
2212: while ((*orig++ = *dest++)); /* Remove multiple /'s */
2213: end = orig-1;
2214: }
2215: }
2216: else
1.25 cvs 2217: p++;
1.28 cvs 2218: }
2219: else
2220: p++;
1.25 cvs 2221: }
2222: }
1.51 cvs 2223: /*
2224: ** Check for host/../.. kind of things
2225: */
1.106 cvs 2226: if (*path == used_sep && *(path+1) == '.' && *(path+2) == '.'
1.77 cvs 2227: && (!*(path+3) || *(path+3) == used_sep))
1.106 cvs 2228: *(path+1) = EOS;
1.28 cvs 2229: return;
2230: }
2231:
2232:
2233: /*----------------------------------------------------------------------
1.96 cvs 2234: NormalizeFile normalizes local names.
1.28 cvs 2235: Return TRUE if target and src differ.
2236: ----------------------------------------------------------------------*/
1.106 cvs 2237: ThotBool NormalizeFile (char *src, char *target, ConvertionType convertion)
1.28 cvs 2238: {
1.184 gully 2239: #ifndef _WINDOWS
1.106 cvs 2240: char *s;
1.93 cvs 2241: int i;
1.184 gully 2242: #endif /* _WINDOWS */
1.82 cvs 2243: ThotBool change;
1.90 cvs 2244: int start_index; /* the first char that we'll copy */
1.28 cvs 2245:
1.54 cvs 2246: change = FALSE;
1.90 cvs 2247: start_index = 0;
2248:
1.106 cvs 2249: if (!src || src[0] == EOS)
1.96 cvs 2250: {
1.106 cvs 2251: target[0] = EOS;
1.96 cvs 2252: return FALSE;
2253: }
1.90 cvs 2254:
2255: /* @@ do I need file: or file:/ here? */
1.106 cvs 2256: if (strncmp (src, "file:", 5) == 0)
1.28 cvs 2257: {
1.90 cvs 2258: /* remove the prefix file: */
2259: start_index += 5;
2260:
2261: /* remove the localhost prefix */
1.106 cvs 2262: if (strncmp (&src[start_index], "//localhost/", 12) == 0)
1.94 cvs 2263: start_index += 11;
2264:
2265: /* remove the first two slashes in / / /path */
2266: while (src[start_index] &&
1.106 cvs 2267: src[start_index] == '/'
2268: && src[start_index + 1] == '/')
1.94 cvs 2269: start_index++;
2270:
1.184 gully 2271: #ifdef _WINDOWS
1.94 cvs 2272: /* remove any extra slash before the drive name */
1.106 cvs 2273: if (src[start_index] == '/'
2274: &&src[start_index+2] == ':')
1.94 cvs 2275: start_index++;
1.184 gully 2276: #endif /* _WINDOWS */
1.90 cvs 2277:
1.106 cvs 2278: if (src[start_index] == EOS)
1.90 cvs 2279: /* if there's nothing afterwards, add a DIR_STR */
1.106 cvs 2280: strcpy (target, DIR_STR);
1.90 cvs 2281: else
1.97 cvs 2282: /* as we're inside a file: URL, we'll apply all the convertions
2283: we know */
2284: CleanCopyFileURL (target, &src[start_index], AM_CONV_ALL);
1.96 cvs 2285:
2286: change = TRUE;
2287: }
1.97 cvs 2288: else if (convertion != AM_CONV_NONE)
1.96 cvs 2289: {
2290: /* we are following a "local" relative link, we do all the
2291: convertions except for the HOME_DIR ~ one */
1.97 cvs 2292: CleanCopyFileURL (target, src, convertion);
1.28 cvs 2293: }
1.184 gully 2294: #ifndef _WINDOWS
1.106 cvs 2295: else if (src[0] == '~')
1.53 cvs 2296: {
1.96 cvs 2297: /* it must be a URL typed in a text input field */
2298: /* do the HOME_DIR ~ substitution */
1.82 cvs 2299: s = TtaGetEnvString ("HOME");
1.106 cvs 2300: strcpy (target, s);
1.90 cvs 2301: #if 0
1.96 cvs 2302: /* JK: invalidated this part of the code as it's simpler
2303: to add the DIR_SEP whenever we have something to add
2304: to the path rather than adding it systematically */
1.106 cvs 2305: if (src[1] != DIR_SEP)
2306: strcat (target, DIR_STR);
1.90 cvs 2307: #endif
1.106 cvs 2308: i = strlen (target);
2309: strcpy (&target[i], &src[1]);
1.54 cvs 2310: change = TRUE;
1.53 cvs 2311: }
1.184 gully 2312: #endif /* _WINDOWS */
1.28 cvs 2313: else
1.96 cvs 2314: /* leave it as it is */
1.106 cvs 2315: strcpy (target, src);
1.96 cvs 2316:
1.28 cvs 2317: /* remove /../ and /./ */
1.29 cvs 2318: SimplifyUrl (&target);
1.54 cvs 2319: if (!change)
1.191 gully 2320: change = (strcmp (src, target) != 0);
1.28 cvs 2321: return (change);
1.25 cvs 2322: }
2323:
1.28 cvs 2324:
1.25 cvs 2325: /*----------------------------------------------------------------------
1.31 cvs 2326: MakeRelativeURL: make relative name
1.25 cvs 2327:
1.28 cvs 2328: This function creates and returns a string which gives an expression of
2329: one address as related to another. Where there is no relation, an absolute
2330: address is retured.
1.25 cvs 2331:
1.28 cvs 2332: On entry,
1.25 cvs 2333: Both names must be absolute, fully qualified names of nodes
2334: (no fragment bits)
2335:
1.28 cvs 2336: On exit,
1.25 cvs 2337: The return result points to a newly allocated name which, if
2338: parsed by AmayaParseUrl relative to relatedName, will yield aName.
2339: The caller is responsible for freeing the resulting name later.
2340: ----------------------------------------------------------------------*/
1.106 cvs 2341: char *MakeRelativeURL (char *aName, char *relatedName)
2342: {
2343: char *return_value;
2344: char result[MAX_LENGTH];
2345: char *p;
2346: char *q;
2347: char *after_access;
2348: char *last_slash = NULL;
2349: int slashes, levels, len;
1.184 gully 2350: #ifdef _WINDOWS
1.44 cvs 2351: int ndx;
1.184 gully 2352: #endif /* _WINDOWS */
1.44 cvs 2353:
1.29 cvs 2354: if (aName == NULL || relatedName == NULL)
2355: return (NULL);
2356:
2357: slashes = 0;
2358: after_access = NULL;
2359: p = aName;
2360: q = relatedName;
1.147 vatton 2361: len = 0;
2362: for (; *p && !strncasecmp (p, q, 1); p++, q++, len++)
1.27 cvs 2363: {
2364: /* Find extent of match */
1.106 cvs 2365: if (*p == ':')
1.146 cvs 2366: {
1.168 cvs 2367: after_access = p + 1;
1.184 gully 2368: #ifdef _WINDOWS
1.168 cvs 2369: if (len == 1)
2370: {
2371: /* it's a local Windows path like c:... */
2372: slashes+=2;
2373: }
1.184 gully 2374: #endif /* _WINDOWS */
1.168 cvs 2375: }
2376: if (*p == DIR_SEP)
2377: {
2378: /* memorize the last slash position and count them */
2379: last_slash = p;
1.147 vatton 2380: slashes++;
1.146 cvs 2381: }
1.25 cvs 2382: }
2383:
1.31 cvs 2384: /* q, p point to the first non-matching character or zero */
1.106 cvs 2385: if (*q == EOS)
1.31 cvs 2386: {
2387: /* New name is a subset of the related name */
2388: /* exactly the right length */
1.106 cvs 2389: len = strlen (p);
1.171 gully 2390: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 2391: strcpy (return_value, p);
1.31 cvs 2392: }
2393: else if ((slashes < 2 && after_access == NULL)
2394: || (slashes < 3 && after_access != NULL))
1.168 cvs 2395: {
1.31 cvs 2396: /* Two names whitout common path */
2397: /* exactly the right length */
1.106 cvs 2398: len = strlen (aName);
1.171 gully 2399: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 2400: strcpy (return_value, aName);
1.31 cvs 2401: }
2402: else
2403: {
2404: /* Some path in common */
1.106 cvs 2405: if (slashes == 3 && strncmp (aName, "http:", 5) == 0)
1.31 cvs 2406: /* just the same server */
1.106 cvs 2407: strcpy (result, last_slash);
1.31 cvs 2408: else
2409: {
2410: levels= 0;
1.106 cvs 2411: for (; *q && *q != '#' && *q != ';' && *q != '?'; q++)
1.31 cvs 2412: if (*q == DIR_SEP)
2413: levels++;
2414:
1.106 cvs 2415: result[0] = EOS;
1.31 cvs 2416: for (;levels; levels--)
1.106 cvs 2417: strcat (result, "../");
2418: strcat (result, last_slash+1);
1.31 cvs 2419: }
1.52 cvs 2420:
2421: if (!*result)
1.106 cvs 2422: strcat (result, "./");
1.52 cvs 2423:
1.31 cvs 2424: /* exactly the right length */
1.106 cvs 2425: len = strlen (result);
1.171 gully 2426: if ((return_value = (char *)TtaGetMemory (len + 1)) != NULL)
1.106 cvs 2427: strcpy (return_value, result);
1.52 cvs 2428:
1.25 cvs 2429: }
1.184 gully 2430: #ifdef _WINDOWS
1.106 cvs 2431: len = strlen (return_value);
1.168 cvs 2432: for (ndx = 0; ndx < len; ndx ++)
1.106 cvs 2433: if (return_value[ndx] == '\\')
2434: return_value[ndx] = '/' ;
1.184 gully 2435: #endif /* _WINDOWS */
1.29 cvs 2436: return (return_value);
1.24 cvs 2437: }
1.35 cvs 2438:
1.104 kahan 2439: /*----------------------------------------------------------------------
2440: AM_GetFileSize
2441: Returns TRUE and the filesize in the 2nd parameter.
2442: Otherwise, in case of a system error, returns FALSE, with a
2443: filesize of 0L.
2444: ---------------------------------------------------------------------*/
1.106 cvs 2445: ThotBool AM_GetFileSize (char *filename, unsigned long *file_size)
1.104 kahan 2446: {
2447: if (!TtaFileExist (filename))
2448: return FALSE;
1.185 vatton 2449: *file_size = TtaGetFileSize (filename);
2450: return TRUE;
1.104 kahan 2451: }
1.139 kahan 2452:
2453: /*----------------------------------------------------------------------
2454: AM_UseXHTMLMimeType
2455: Returns TRUE if the user has configured Amaya to use this MIME type,
2456: FALSE otherwise.
2457: ---------------------------------------------------------------------*/
2458: ThotBool AM_UseXHTMLMimeType (void)
2459: {
2460: ThotBool xhtml_mimetype;
2461:
2462: /* does the user wants to use the new MIME type? */
2463: TtaGetEnvBoolean ("ENABLE_XHTML_MIMETYPE", &xhtml_mimetype);
2464:
2465: return (xhtml_mimetype);
1.152 kahan 2466: }
2467:
1.154 kahan 2468:
2469: /********************************************
2470: The following routines were adapted from the GNU libc functions
2471: for generating a tmpnam.
2472: *********************************************/
2473:
2474: /* These are the characters used in temporary filenames. */
2475: static const char letters[] =
2476: "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
2477:
2478: /* Generate a temporary file name based on TMPL. TMPL must match the
2479: rules for mk[s]temp (i.e. end in "XXXXXX"). The name constructed
2480: does not exist at the time of the call to __gen_tempname. TMPL is
2481: overwritten with the result.
2482:
2483: We use a clever algorithm to get hard-to-predict names. */
2484: void
2485: AM_gen_tempname (char *tmpl)
2486: {
2487: int len;
2488: char *XXXXXX;
2489: static uint64_t value;
2490: uint64_t random_time_bits;
2491: unsigned int count;
2492: int save_errno = errno;
2493: struct stat st;
2494:
2495: /* A lower bound on the number of temporary files to attempt to
2496: generate. The maximum total number of temporary file names that
2497: can exist for a given template is 62**6. It should never be
2498: necessary to try all these combinations. Instead if a reasonable
2499: number of names is tried (we define reasonable as 62**3) fail to
2500: give the system administrator the chance to remove the problems. */
2501: unsigned int attempts_min = 62 * 62 * 62;
2502:
2503: /* The number of times to attempt to generate a temporary file. To
2504: conform to POSIX, this must be no smaller than TMP_MAX. */
2505: unsigned int attempts = attempts_min < TMP_MAX ? TMP_MAX : attempts_min;
2506:
2507: len = strlen (tmpl);
2508: if (len < 6 || strcmp (&tmpl[len - 6], "XXXXXX"))
2509: {
2510: /* @@ JK ? */
2511: errno = EINVAL;
2512: return;
2513: }
2514:
2515: /* This is where the Xs start. */
2516: XXXXXX = &tmpl[len - 6];
2517:
2518: /* Get some more or less random data. */
2519: #ifdef RANDOM_BITS
2520: RANDOM_BITS (random_time_bits);
2521: #else
2522: # if HAVE_GETTIMEOFDAY || _LIBC
2523: {
2524: struct timeval tv;
2525: gettimeofday (&tv, NULL);
2526: random_time_bits = ((uint64_t) tv.tv_usec << 16) ^ tv.tv_sec;
2527: }
2528: # else
2529: random_time_bits = time (NULL);
2530: # endif
2531: #endif
2532: value += random_time_bits ^ getpid ();
2533:
2534: for (count = 0; count < attempts; value += 7777, ++count)
2535: {
2536: uint64_t v = value;
2537:
2538: /* Fill in the random bits. */
2539: XXXXXX[0] = letters[v % 62];
2540: v /= 62;
2541: XXXXXX[1] = letters[v % 62];
2542: v /= 62;
2543: XXXXXX[2] = letters[v % 62];
2544: v /= 62;
2545: XXXXXX[3] = letters[v % 62];
2546: v /= 62;
2547: XXXXXX[4] = letters[v % 62];
2548: v /= 62;
2549: XXXXXX[5] = letters[v % 62];
2550:
2551: /* This case is backward from the other three. AM_gen_tempname
2552: succeeds if __xstat fails because the name does not exist.
2553: Note the continue to bypass the common logic at the bottom
2554: of the loop. */
2555: if (stat (tmpl, &st) < 0)
2556: break;
2557:
2558: continue;
2559: }
2560:
2561: if (count == attempts || errno != ENOENT)
2562: tmpl[0] = EOS;
2563: else
2564: errno = save_errno;
2565:
2566: return;
2567: }
2568:
2569: #define JOSE 1
2570:
1.152 kahan 2571: /*-----------------------------------------------------------------------
2572: GetTempName
2573: Front end to the Unix tempnam function, which is independent of the
2574: value of the TMPDIR env value
2575: Returns a dynamically allocated string with a tempname. The user
2576: must free this memory.
2577: -----------------------------------------------------------------------*/
2578: char *GetTempName (const char *dir, const char *prefix)
2579: {
1.154 kahan 2580: #ifdef JOSE
2581:
1.162 kahan 2582: static char tmpbufmem[PATH_MAX + 1];
1.154 kahan 2583: int len;
2584: int i;
2585:
1.155 cvs 2586: if (!dir || *dir == EOS || !TtaDirExists (dir))
1.154 kahan 2587: return NULL;
2588:
1.162 kahan 2589: /* make sure that the name is no bigger than PATH_MAX + the 6 tempname chars we
2590: will add */
1.154 kahan 2591:
1.156 cvs 2592: len = strlen (dir);
1.162 kahan 2593: if (len + 6 > PATH_MAX)
1.154 kahan 2594: return NULL;
2595:
2596: /* copy the dir name, and add a DIR_SEP if it's missing */
2597: if (dir[strlen (dir) - 1] == DIR_SEP)
2598: strcpy (tmpbufmem, dir);
2599: else
1.156 cvs 2600: {
1.154 kahan 2601: sprintf (tmpbufmem, "%s%c", dir, DIR_SEP);
1.156 cvs 2602: len++;
2603: }
1.154 kahan 2604:
1.161 kahan 2605: /* copy the prefix (no more than L_tmpnam chars, to respect POSIX). Save
1.156 cvs 2606: space for the 6 X and EOS chars that will become the random bits */
2607: if (prefix)
2608: {
2609: i = 0;
1.167 cheyroul 2610: while (prefix[i] != EOS && i < L_tmpnam - 8)
1.156 cvs 2611: tmpbufmem[len++] = prefix[i++];
2612: tmpbufmem[len] = EOS;
2613: }
2614:
2615: /* Add the 6 X chars */
2616: len = strlen (tmpbufmem);
2617: i = 0;
2618: while (i < 6)
2619: {
2620: tmpbufmem[len++] = 'X';
2621: i++;
2622: }
2623: tmpbufmem[len] = EOS;
1.154 kahan 2624:
2625: AM_gen_tempname (tmpbufmem);
2626:
2627: if (tmpbufmem[0] == EOS)
2628: return NULL;
2629: else
2630: return (TtaStrdup (tmpbufmem));
2631:
2632: #else
1.152 kahan 2633: char *tmpdir;
1.181 vatton 2634: char *tmp = NULL;
1.152 kahan 2635: char *name = NULL;
2636:
2637: /* save the value of TMPDIR */
1.181 vatton 2638: tmpdir = getenv ("TMPDIR");
1.152 kahan 2639: if (tmpdir)
2640: {
1.181 vatton 2641: /* remove TMPDIR from the environment */
2642: tmp = TtaGetMemory (strlen (tmpdir) + 20);
2643: sprintf (tmp, "TMPDIR=");
1.184 gully 2644: #ifdef _WINDOWS
1.152 kahan 2645: _putenv (tmp);
2646: #else
2647: putenv (tmp);
1.184 gully 2648: #endif /* _WINDOWS */
1.181 vatton 2649: /* prepare the string to restore the value of TMPDIR */
2650: strrcat (tmp, tmpdir);
1.152 kahan 2651: }
2652:
2653: /* create the tempname */
1.184 gully 2654: #ifdef _WINDOWS
1.152 kahan 2655: /* Under Windows, _tempnam returns the same name until the file is created */
2656: {
2657: char *altprefix;
2658: name = tmpnam (NULL); /* get a possibly unique string */
1.181 vatton 2659: altprefix = TtaGetMemory (strlen (prefix) + strlen(name) + 1);
1.152 kahan 2660: sprintf (altprefix, "%s%s", prefix, name + strlen(_P_tmpdir));
2661: name = _tempnam (dir, altprefix); /* get a name that isn't yet in use */
2662: TtaFreeMemory (altprefix);
2663: }
2664: #else
2665: name = tempnam (dir, prefix);
1.184 gully 2666: #endif /* _WINDOWS */
1.152 kahan 2667:
2668: if (tmpdir)
2669: {
1.181 vatton 2670: /* restore the value of TMPDIR */
1.184 gully 2671: #ifdef _WINDOWS
1.152 kahan 2672: _putenv (tmpdir);
2673: #else
2674: putenv (tmpdir);
1.184 gully 2675: #endif /* _WINDOWS */
1.181 vatton 2676: TtaFreeMemory (tmpdir);
1.152 kahan 2677: }
2678: return (name);
1.154 kahan 2679: #endif
1.139 kahan 2680: }
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