Return to AHTURLTools.c CVS log | Up to [Public] / Amaya / amaya |
1.7 cvs 1: /* 2: * 3: * (c) COPYRIGHT MIT and INRIA, 1996. 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 1.45 cvs 14: * R. Guetari (Stuff related to Windows). 1.10 cvs 15: * 16: */ 1.7 cvs 17: 1.15 cvs 18: #define THOT_EXPORT extern 1.3 cvs 19: #include "amaya.h" 20: 1.8 cvs 21: #include "init_f.h" 22: #include "AHTURLTools_f.h" 23: 1.24 cvs 24: #define MAX_PRINT_URL_LENGTH 50 1.29 cvs 25: typedef struct _HTURI { 1.67 cvs 26: STRING access; /* Now known as "scheme" */ 27: STRING host; 28: STRING absolute; 29: STRING relative; 30: STRING fragment; 1.29 cvs 31: } HTURI; 1.24 cvs 32: 1.28 cvs 33: 34: /*---------------------------------------------------------------------- 35: ConvertToLowerCase 36: Converts a string to lowercase. 37: ----------------------------------------------------------------------*/ 1.22 cvs 38: #ifdef __STDC__ 1.67 cvs 39: void ConvertToLowerCase (STRING string) 1.28 cvs 40: #else /* __STDC__ */ 1.38 cvs 41: void ConvertToLowerCase (string) 1.67 cvs 42: STRING string; 1.28 cvs 43: 44: #endif /* __STDC__ */ 45: { 46: int i; 47: 48: if (!string) 49: return; 50: 51: for (i = 0; string[i] != EOS; i++) 1.67 cvs 52: string[i] = utolower (string[i]); 1.28 cvs 53: } 1.22 cvs 54: 1.8 cvs 55: /*---------------------------------------------------------------------- 1.75 cvs 56: EscapeChar 57: writes the equivalent escape code of a char in a string 58: ----------------------------------------------------------------------*/ 59: #ifdef __STDC__ 60: void EscapeChar (STRING string, UCHAR_T c) 61: #else 62: void EscapeChar (string, c) 63: STRING string; 64: UCHAR_T c; 65: 66: #endif 67: { 68: c &= 0xFF; /* strange behavior under solaris? */ 69: usprintf (string, TEXT("%02x"), (unsigned int) c); 70: } 71: 72: /*---------------------------------------------------------------------- 73: EscapeURL 74: Takes a URL and escapes all protected chars into 75: %xx sequences. Also, removes any leading white spaces 76: Returns either NULL or a new buffer, which must be freed by the caller 77: ----------------------------------------------------------------------*/ 78: #ifdef __STDC__ 79: STRING EscapeURL (const STRING url) 80: #else 81: STRING EscapeURL (url) 82: STRING url; 83: #endif /* __STDC__ */ 84: { 85: STRING buffer; 86: int buffer_len; 87: int buffer_free_mem; 88: PCHAR_T ptr; 89: int new_chars; 90: void *status; 91: 92: if (url && *url) 93: { 1.76 cvs 94: buffer_free_mem = ustrlen (url) + 20; 95: buffer = TtaAllocString (buffer_free_mem + 1); 1.75 cvs 96: ptr = url; 97: buffer_len = 0; 98: 99: while (*ptr) 100: { 101: switch (*ptr) 102: { 103: /* put here below all the chars that need to 104: be escaped into %xx */ 1.81 cvs 105: case 0x27: /* & */ 106: case 0x20: /* space */ 1.75 cvs 107: new_chars = 3; 108: break; 109: 110: default: 111: new_chars = 1; 112: break; 113: } 114: 115: /* see if we need extra room in the buffer */ 116: if (new_chars > buffer_free_mem) 117: { 1.76 cvs 118: buffer_free_mem = 20; 1.75 cvs 119: status = TtaRealloc (buffer, sizeof (CHAR_T) 120: * (buffer_len + buffer_free_mem + 1)); 121: if (status) 122: buffer = (STRING) status; 123: else { 124: /* @@ maybe we should do some other behavior here, like 125: freeing the buffer and return a void thing */ 1.76 cvs 126: buffer[buffer_len] = EOS; 1.75 cvs 127: break; 128: } 129: } 130: /* escape the char */ 131: if (new_chars == 3) 132: { 133: buffer[buffer_len] = TEXT('%'); 134: EscapeChar (&buffer[buffer_len+1], *ptr); 135: } 136: else 137: buffer[buffer_len] = *ptr; 138: 139: /* update the status */ 140: buffer_len += new_chars; 141: buffer_free_mem -= new_chars; 142: /* examine the next char */ 143: ptr++; 144: } 145: buffer[buffer_len] = EOS; 146: } 1.76 cvs 147: else 148: buffer = NULL; 149: 1.75 cvs 150: return (buffer); 151: } 152: 153: 154: /*---------------------------------------------------------------------- 1.11 cvs 155: ExplodeURL 1.8 cvs 156: ----------------------------------------------------------------------*/ 157: #ifdef __STDC__ 158: void ExplodeURL (char *url, char **proto, char **host, char **dir, char **file) 159: #else 160: void ExplodeURL (url, proto, host, dir, file) 161: char *url; 162: char **proto; 163: char **host; 164: char **dir; 165: char **file; 166: 167: #endif 168: { 1.33 cvs 169: char *curr, *temp; 170: char used_sep; 1.32 cvs 171: 1.33 cvs 172: if (url && strchr (url, URL_SEP)) 173: used_sep = URL_SEP; 174: else 175: used_sep = DIR_SEP; 1.8 cvs 176: 177: if ((url == NULL) || (proto == NULL) || (host == NULL) || 178: (dir == NULL) || (file == NULL)) 179: return; 180: 181: /* initialize every pointer */ 182: *proto = *host = *dir = *file = NULL; 183: 184: /* skip any leading space */ 185: while ((*url == SPACE) || (*url == TAB)) 186: url++; 1.9 cvs 187: curr = url; 188: if (*curr == 0) 1.8 cvs 189: goto finished; 190: 191: /* go to the end of the URL */ 1.68 cvs 192: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) && 193: (*curr != __CR__) && (*curr != EOL)) 1.9 cvs 194: curr++; 1.8 cvs 195: 196: /* mark the end of the chain */ 1.9 cvs 197: *curr = EOS; 198: curr--; 199: if (curr <= url) 1.8 cvs 200: goto finished; 201: 202: /* search the next DIR_SEP indicating the beginning of the file name */ 203: do 1.11 cvs 204: curr--; 1.33 cvs 205: while ((curr >= url) && (*curr != used_sep)); 1.11 cvs 206: 1.9 cvs 207: if (curr < url) 1.8 cvs 208: goto finished; 1.9 cvs 209: *file = curr + 1; 1.8 cvs 210: 211: /* mark the end of the dir */ 1.9 cvs 212: *curr = EOS; 213: curr--; 214: if (curr < url) 1.8 cvs 215: goto finished; 216: 1.29 cvs 217: /* search for the DIR_STR indicating the host name start */ 1.33 cvs 218: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep))) 1.9 cvs 219: curr--; 1.8 cvs 220: 221: /* if we found it, separate the host name from the directory */ 1.33 cvs 222: if ((*curr == DIR_SEP) && (*(curr + 1) == used_sep)) 1.8 cvs 223: { 1.9 cvs 224: *host = temp = curr + 2; 1.33 cvs 225: while ((*temp != 0) && (*temp != used_sep)) 1.8 cvs 226: temp++; 1.33 cvs 227: if (*temp == used_sep) 1.8 cvs 228: { 229: *temp = EOS; 230: *dir = temp + 1; 231: } 232: } 233: else 1.11 cvs 234: *dir = curr; 235: 1.9 cvs 236: if (curr <= url) 1.8 cvs 237: goto finished; 238: 239: /* mark the end of the proto */ 1.9 cvs 240: *curr = EOS; 241: curr--; 242: if (curr < url) 1.8 cvs 243: goto finished; 244: 1.67 cvs 245: if (*curr == TEXT(':')) 1.8 cvs 246: { 1.9 cvs 247: *curr = EOS; 248: curr--; 1.8 cvs 249: } 250: else 251: goto finished; 1.11 cvs 252: 1.9 cvs 253: if (curr < url) 1.8 cvs 254: goto finished; 1.9 cvs 255: while ((curr > url) && (isalpha (*curr))) 256: curr--; 257: *proto = curr; 1.8 cvs 258: 259: finished:; 260: 261: #ifdef AMAYA_DEBUG 262: fprintf (stderr, "ExplodeURL(%s)\n\t", url); 263: if (*proto) 264: fprintf (stderr, "proto : %s, ", *proto); 265: if (*host) 266: fprintf (stderr, "host : %s, ", *host); 267: if (*dir) 268: fprintf (stderr, "dir : %s, ", *dir); 269: if (*file) 270: fprintf (stderr, "file : %s ", *file); 271: fprintf (stderr, "\n"); 272: #endif 273: 274: } 1.3 cvs 275: 1.61 cvs 276: 277: /*---------------------------------------------------------------------- 278: ExtractSuffix extract suffix from document nane. 279: ----------------------------------------------------------------------*/ 280: #ifdef __STDC__ 281: void ExtractSuffix (STRING aName, STRING aSuffix) 282: #else 283: void ExtractSuffix (aName, aSuffix) 284: STRING aName; 285: STRING aSuffix; 286: 287: #endif 288: { 289: int lg, i; 290: STRING ptr, oldptr; 291: 292: if (!aSuffix || !aName) 293: /* bad suffix */ 294: return; 295: 296: aSuffix[0] = EOS; 297: lg = ustrlen (aName); 298: if (lg) 299: { 300: /* the name is not empty */ 301: oldptr = ptr = &aName[0]; 302: do 303: { 1.67 cvs 304: ptr = ustrrchr (oldptr, TEXT('.')); 1.61 cvs 305: if (ptr) 306: oldptr = &ptr[1]; 307: } 308: while (ptr); 309: 310: i = (int) (oldptr) - (int) (aName); /* name length */ 311: if (i > 1) 312: { 313: aName[i - 1] = EOS; 314: if (i != lg) 315: ustrcpy (aSuffix, oldptr); 316: } 317: } 318: } 319: 1.4 cvs 320: /*---------------------------------------------------------------------- 1.9 cvs 321: IsHTMLName 322: returns TRUE if path points to an HTML resource. 1.4 cvs 323: ----------------------------------------------------------------------*/ 1.3 cvs 324: #ifdef __STDC__ 1.84 ! cvs 325: ThotBool IsHTMLName (const CHAR_T* path) 1.3 cvs 326: #else /* __STDC__ */ 1.67 cvs 327: ThotBool IsHTMLName (path) 1.84 ! cvs 328: const CHAR_T* path; 1.3 cvs 329: #endif /* __STDC__ */ 330: { 1.84 ! cvs 331: CHAR_T temppath[MAX_LENGTH]; ! 332: CHAR_T suffix[MAX_LENGTH]; ! 333: CHAR_T nsuffix[MAX_LENGTH]; ! 334: int i; 1.5 cvs 335: 336: if (!path) 1.37 cvs 337: return (FALSE); 1.5 cvs 338: 1.84 ! cvs 339: ustrcpy (temppath, path); 1.5 cvs 340: ExtractSuffix (temppath, suffix); 341: 342: /* Normalize the suffix */ 343: i = 0; 1.84 ! cvs 344: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1) 1.13 cvs 345: { 1.84 ! cvs 346: nsuffix[i] = utolower (suffix[i]); 1.13 cvs 347: i++; 348: } 1.84 ! cvs 349: nsuffix[i] = WC_EOS; ! 350: if (!ustrcmp (nsuffix, TEXT("html")) || ! 351: !ustrcmp (nsuffix, TEXT("htm")) || ! 352: !ustrcmp (nsuffix, TEXT("shtml")) || ! 353: !ustrcmp (nsuffix, TEXT("jsp")) || ! 354: !ustrcmp (nsuffix, TEXT("xht")) || ! 355: !ustrcmp (nsuffix, TEXT("xhtm")) || ! 356: !ustrcmp (nsuffix, TEXT("xhtml"))) 1.39 cvs 357: return (TRUE); 1.84 ! cvs 358: else if (!ustrcmp (nsuffix, TEXT("gz"))) 1.13 cvs 359: { 1.39 cvs 360: /* take into account compressed files */ 1.13 cvs 361: ExtractSuffix (temppath, suffix); 362: /* Normalize the suffix */ 363: i = 0; 1.84 ! cvs 364: while (suffix[i] != WC_EOS && i < MAX_LENGTH -1) 1.13 cvs 365: { 1.84 ! cvs 366: nsuffix[i] = utolower (suffix[i]); 1.13 cvs 367: i++; 368: } 1.84 ! cvs 369: nsuffix[i] = WC_EOS; ! 370: if (!ustrcmp (nsuffix, TEXT("html")) || ! 371: !ustrcmp (nsuffix, TEXT("htm")) || ! 372: !ustrcmp (nsuffix, TEXT("shtml")) || ! 373: !ustrcmp (nsuffix, TEXT("jsp")) || ! 374: !ustrcmp (nsuffix, TEXT("xht")) || ! 375: !ustrcmp (nsuffix, TEXT("xhtm")) || ! 376: !ustrcmp (nsuffix, TEXT("xhtml"))) 1.63 cvs 377: 1.39 cvs 378: return (TRUE); 379: else 1.13 cvs 380: return (FALSE); 381: } 382: else 1.39 cvs 383: return (FALSE); 1.3 cvs 384: } 385: 1.4 cvs 386: /*---------------------------------------------------------------------- 1.56 cvs 387: IsXMLName 388: returns TRUE if path points to an XML resource. 389: ----------------------------------------------------------------------*/ 390: #ifdef __STDC__ 1.67 cvs 391: ThotBool IsXMLName (const STRING path) 1.56 cvs 392: #else /* __STDC__ */ 1.67 cvs 393: ThotBool IsXMLName (path) 394: const STRING path; 1.56 cvs 395: #endif /* __STDC__ */ 396: { 1.67 cvs 397: CHAR_T temppath[MAX_LENGTH]; 398: CHAR_T suffix[MAX_LENGTH]; 1.56 cvs 399: 400: if (!path) 401: return (FALSE); 402: 1.67 cvs 403: ustrcpy (temppath, path); 1.56 cvs 404: ExtractSuffix (temppath, suffix); 405: 1.67 cvs 406: if (!ustrcasecmp (suffix, TEXT("xml")) || 407: !ustrcasecmp (suffix, TEXT("xht")) || 408: !ustrcmp (suffix, TEXT("xhtm")) || 409: !ustrcmp (suffix, TEXT("xhtml"))) 1.56 cvs 410: return (TRUE); 1.67 cvs 411: else if (!ustrcmp (suffix, TEXT("gz"))) 1.56 cvs 412: { 413: /* take into account compressed files */ 414: ExtractSuffix (temppath, suffix); 1.67 cvs 415: if (!ustrcasecmp (suffix, TEXT("xml")) || 416: !ustrcasecmp (suffix, TEXT("xht")) || 417: !ustrcmp (suffix, TEXT("xhtm")) || 418: !ustrcmp (suffix, TEXT("xhtml"))) 1.60 cvs 419: return (TRUE); 420: else 421: return (FALSE); 422: } 423: else 424: return (FALSE); 425: } 426: 427: /*---------------------------------------------------------------------- 428: IsCSSName 429: returns TRUE if path points to an XML resource. 430: ----------------------------------------------------------------------*/ 431: #ifdef __STDC__ 1.67 cvs 432: ThotBool IsCSSName (const STRING path) 1.60 cvs 433: #else /* __STDC__ */ 1.67 cvs 434: ThotBool IsCSSName (path) 435: const STRING path; 1.60 cvs 436: #endif /* __STDC__ */ 437: { 1.67 cvs 438: CHAR_T temppath[MAX_LENGTH]; 439: CHAR_T suffix[MAX_LENGTH]; 1.60 cvs 440: 441: if (!path) 442: return (FALSE); 443: 1.67 cvs 444: ustrcpy (temppath, path); 1.60 cvs 445: ExtractSuffix (temppath, suffix); 446: 1.67 cvs 447: if (!ustrcasecmp (suffix, TEXT("css"))) 1.60 cvs 448: return (TRUE); 1.67 cvs 449: else if (!ustrcmp (suffix, TEXT("gz"))) 1.60 cvs 450: { 451: /* take into account compressed files */ 452: ExtractSuffix (temppath, suffix); 1.67 cvs 453: if (!ustrcasecmp (suffix, TEXT("css"))) 1.56 cvs 454: return (TRUE); 455: else 456: return (FALSE); 457: } 458: else 459: return (FALSE); 460: } 461: 462: /*---------------------------------------------------------------------- 1.9 cvs 463: IsImageName 464: returns TRUE if path points to an image resource. 1.4 cvs 465: ----------------------------------------------------------------------*/ 1.3 cvs 466: #ifdef __STDC__ 1.67 cvs 467: ThotBool IsImageName (const STRING path) 1.3 cvs 468: #else /* __STDC__ */ 1.67 cvs 469: ThotBool IsImageName (path) 470: const STRING path; 1.3 cvs 471: #endif /* __STDC__ */ 472: { 1.67 cvs 473: CHAR_T temppath[MAX_LENGTH]; 474: CHAR_T suffix[MAX_LENGTH]; 475: CHAR_T nsuffix[MAX_LENGTH]; 1.5 cvs 476: int i; 477: 478: if (!path) 1.13 cvs 479: return (FALSE); 1.5 cvs 480: 1.67 cvs 481: ustrcpy (temppath, path); 1.5 cvs 482: ExtractSuffix (temppath, suffix); 483: 484: /* Normalize the suffix */ 485: i = 0; 1.39 cvs 486: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 487: { 1.67 cvs 488: nsuffix[i] = utolower (suffix[i]); 1.13 cvs 489: i++; 490: } 1.5 cvs 491: nsuffix[i] = EOS; 1.67 cvs 492: if ((!ustrcmp (nsuffix, TEXT("gif"))) || (!ustrcmp (nsuffix, TEXT("xbm"))) || 493: (!ustrcmp (nsuffix, TEXT("xpm"))) || (!ustrcmp (nsuffix, TEXT("jpg"))) || 494: (!ustrcmp (nsuffix, TEXT("png"))) || (!ustrcmp (nsuffix, TEXT("au")))) 1.39 cvs 495: return (TRUE); 496: return (FALSE); 1.3 cvs 497: } 498: 1.4 cvs 499: /*---------------------------------------------------------------------- 1.58 cvs 500: IsImageType 501: returns TRUE if type points to an image resource. 502: ----------------------------------------------------------------------*/ 503: #ifdef __STDC__ 1.67 cvs 504: ThotBool IsImageType (const STRING type) 1.58 cvs 505: #else /* __STDC__ */ 1.67 cvs 506: ThotBool IsImageType (type) 507: const STRING type; 1.58 cvs 508: #endif /* __STDC__ */ 509: { 1.67 cvs 510: CHAR_T temptype[MAX_LENGTH]; 1.58 cvs 511: int i; 512: 513: if (!type) 514: return (FALSE); 515: 1.67 cvs 516: ustrcpy (temptype, type); 1.58 cvs 517: /* Normalize the type */ 518: i = 0; 519: while (temptype[i] != EOS) 520: { 521: temptype[i] = tolower (temptype[i]); 522: i++; 523: } 1.67 cvs 524: if ((!ustrcmp (temptype, TEXT("gif"))) || (!ustrcmp (temptype, TEXT("x-xbitmap"))) || 525: (!ustrcmp (temptype, TEXT("x-xpixmap"))) || (!ustrcmp (temptype, TEXT("jpeg"))) || 526: (!ustrcmp (temptype, TEXT("png")))) 1.58 cvs 527: return (TRUE); 528: return (FALSE); 529: } 530: 531: /*---------------------------------------------------------------------- 1.9 cvs 532: IsTextName 1.4 cvs 533: ----------------------------------------------------------------------*/ 1.3 cvs 534: #ifdef __STDC__ 1.67 cvs 535: ThotBool IsTextName (const STRING path) 1.3 cvs 536: #else /* __STDC__ */ 1.67 cvs 537: ThotBool IsTextName (path) 538: const STRING path; 1.3 cvs 539: 540: #endif /* __STDC__ */ 541: { 1.67 cvs 542: CHAR_T temppath[MAX_LENGTH]; 543: CHAR_T suffix[MAX_LENGTH]; 544: CHAR_T nsuffix[MAX_LENGTH]; 1.5 cvs 545: int i; 546: 547: if (!path) 1.13 cvs 548: return (FALSE); 1.5 cvs 549: 1.67 cvs 550: ustrcpy (temppath, path); 1.5 cvs 551: ExtractSuffix (temppath, suffix); 552: 553: /* Normalize the suffix */ 554: i = 0; 1.39 cvs 555: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.5 cvs 556: { 1.25 cvs 557: nsuffix[i] = tolower (suffix[i]); 1.5 cvs 558: i++; 559: } 560: nsuffix[i] = EOS; 561: 1.67 cvs 562: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd")))) 1.13 cvs 563: return (TRUE); 1.67 cvs 564: else if (!ustrcmp (nsuffix, TEXT("gz"))) 1.13 cvs 565: { 1.39 cvs 566: /* take into account compressed files */ 1.13 cvs 567: ExtractSuffix (temppath, suffix); 568: /* Normalize the suffix */ 569: i = 0; 1.39 cvs 570: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 571: { 1.25 cvs 572: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 573: i++; 574: } 575: nsuffix[i] = EOS; 1.67 cvs 576: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd")))) 1.13 cvs 577: return (TRUE); 578: else 579: return (FALSE); 580: } 581: else 582: return (FALSE); 1.3 cvs 583: } 584: 1.4 cvs 585: /*---------------------------------------------------------------------- 1.9 cvs 586: IsHTTPPath 587: returns TRUE if path is in fact an http URL. 1.4 cvs 588: ----------------------------------------------------------------------*/ 1.3 cvs 589: #ifdef __STDC__ 1.67 cvs 590: ThotBool IsHTTPPath (const STRING path) 1.3 cvs 591: #else /* __STDC__ */ 1.67 cvs 592: ThotBool IsHTTPPath (path) 593: const STRING path; 1.3 cvs 594: #endif /* __STDC__ */ 595: { 1.5 cvs 596: if (!path) 597: return FALSE; 1.3 cvs 598: 1.67 cvs 599: if ((!ustrncmp (path, TEXT("http:"), 5) != 0) 600: || !ustrncmp (path, TEXT("internal:"), 9)) 1.58 cvs 601: return TRUE; 602: return FALSE; 1.3 cvs 603: } 604: 1.4 cvs 605: /*---------------------------------------------------------------------- 1.9 cvs 606: IsWithParameters 607: returns TRUE if url has a concatenated query string. 1.4 cvs 608: ----------------------------------------------------------------------*/ 1.3 cvs 609: #ifdef __STDC__ 1.66 cvs 610: ThotBool IsWithParameters (const char *url) 1.3 cvs 611: #else /* __STDC__ */ 1.66 cvs 612: ThotBool IsWithParameters (url) 1.34 cvs 613: const char *url; 1.3 cvs 614: #endif /* __STDC__ */ 615: { 1.5 cvs 616: int i; 1.3 cvs 617: 1.9 cvs 618: if ((!url) || (url[0] == EOS)) 1.5 cvs 619: return FALSE; 1.3 cvs 620: 1.9 cvs 621: i = strlen (url) - 1; 622: while (i > 0 && url[i--] != '?') 1.5 cvs 623: if (i < 0) 624: return FALSE; 1.3 cvs 625: 1.5 cvs 626: /* There is a parameter */ 627: return TRUE; 1.3 cvs 628: } 629: 1.4 cvs 630: /*---------------------------------------------------------------------- 1.9 cvs 631: IsW3Path 632: returns TRUE if path is in fact a URL. 1.4 cvs 633: ----------------------------------------------------------------------*/ 1.3 cvs 634: #ifdef __STDC__ 1.84 ! cvs 635: ThotBool IsW3Path (const CHAR_T* path) 1.3 cvs 636: #else /* __STDC__ */ 1.67 cvs 637: ThotBool IsW3Path (path) 1.84 ! cvs 638: const CHAR_T* path; 1.3 cvs 639: #endif /* __STDC__ */ 640: { 1.84 ! cvs 641: if (ustrncmp (path, TEXT("http:"), 5) && ! 642: ustrncmp (path, TEXT("ftp:"), 4) && ! 643: ustrncmp (path, TEXT("telnet:"), 7) && ! 644: ustrncmp (path, TEXT("wais:"), 5) && ! 645: ustrncmp (path, TEXT("news:"), 5) && ! 646: ustrncmp (path, TEXT("gopher:"), 7) && ! 647: ustrncmp (path, TEXT("mailto:"), 7) && ! 648: ustrncmp (path, TEXT("archie:"), 7)) 1.72 cvs 649: return FALSE; 650: return TRUE; 1.3 cvs 651: } 652: 1.4 cvs 653: /*---------------------------------------------------------------------- 1.9 cvs 654: IsValidProtocol 655: returns true if the url protocol is supported by Amaya. 1.4 cvs 656: ----------------------------------------------------------------------*/ 1.3 cvs 657: #ifdef __STDC__ 1.67 cvs 658: ThotBool IsValidProtocol (const STRING url) 1.3 cvs 659: #else /* __STDC__ */ 1.67 cvs 660: ThotBool IsValidProtocol (url) 661: const STRING url; 1.3 cvs 662: #endif /* __STDC__ */ 663: { 1.67 cvs 664: if (!ustrncmp (url, TEXT("http:"), 5) 1.69 cvs 665: || !ustrncmp (url, TEXT("internal:"), 9) 1.70 cvs 666: || !ustrncmp (url, TEXT("ftp:"), 4)) 1.22 cvs 667: /* experimental */ 1.58 cvs 668: /*** || !strncmp (url, "ftp:", 4) ***/ 1.24 cvs 669: /*** || !strncmp (path, "news:", 5)***/ 1.8 cvs 670: return (TRUE); 1.5 cvs 671: else 1.8 cvs 672: return (FALSE); 1.3 cvs 673: } 674: 1.31 cvs 675: 676: /*---------------------------------------------------------------------- 677: GetBaseURL 678: normalizes orgName according to a base associated with doc, and 679: following the standard URL format rules. 680: The function returns the base used to solve relative URL and SRC: 681: - the base of the document, 682: - or the document path (without document name). 683: ----------------------------------------------------------------------*/ 684: #ifdef __STDC__ 1.84 ! cvs 685: CHAR_T* GetBaseURL (Document doc) 1.31 cvs 686: #else /* __STDC__ */ 1.84 ! cvs 687: CHAR_T* GetBaseURL (doc) 1.31 cvs 688: Document doc; 689: #endif /* __STDC__ */ 690: { 691: Element el; 692: ElementType elType; 693: AttributeType attrType; 694: Attribute attr; 1.84 ! cvs 695: CHAR_T *ptr, *basename; 1.31 cvs 696: int length; 697: 1.57 cvs 698: /* @@@ irene */ 699: if (!DocumentURLs[doc]) 700: return NULL; 1.67 cvs 701: basename = TtaAllocString (MAX_LENGTH); 702: ustrncpy (basename, DocumentURLs[doc], MAX_LENGTH-1); 1.84 ! cvs 703: basename[MAX_LENGTH-1] = WC_EOS; 1.31 cvs 704: length = MAX_LENGTH -1; 705: /* get the root element */ 706: el = TtaGetMainRoot (doc); 707: /* search the BASE element */ 708: elType.ElSSchema = TtaGetDocumentSSchema (doc); 1.65 cvs 709: elType.ElTypeNum = HTML_EL_HEAD; 710: el = TtaSearchTypedElement (elType, SearchForward, el); 711: if (el) 712: { 713: elType.ElTypeNum = HTML_EL_BASE; 714: el = TtaSearchTypedElement (elType, SearchInTree, el); 715: } 1.31 cvs 716: if (el) 717: { 718: /* The document has a BASE element -> Get the HREF attribute */ 719: attrType.AttrSSchema = elType.ElSSchema; 720: attrType.AttrTypeNum = HTML_ATTR_HREF_; 721: attr = TtaGetAttribute (el, attrType); 722: if (attr) 723: { 724: /* Use the base path of the document */ 725: TtaGiveTextAttributeValue (attr, basename, &length); 726: /* base and orgName have to be separated by a DIR_SEP */ 727: length--; 1.84 ! cvs 728: if (basename[0] != WC_EOS && basename[length] != WC_URL_SEP && basename[length] != WC_DIR_SEP) 1.31 cvs 729: /* verify if the base has the form "protocol://server:port" */ 730: { 1.84 ! cvs 731: ptr = AmayaParseUrl (basename, TEXT(""), AMAYA_PARSE_ACCESS | 1.33 cvs 732: AMAYA_PARSE_HOST | 733: AMAYA_PARSE_PUNCTUATION); 1.67 cvs 734: if (ptr && !ustrcmp (ptr, basename)) 1.31 cvs 735: { 1.43 cvs 736: /* it has this form, we complete it by adding a URL_STR */ 1.84 ! cvs 737: if (ustrchr (basename, WC_DIR_SEP)) ! 738: ustrcat (basename, WC_DIR_STR); 1.43 cvs 739: else 1.84 ! cvs 740: ustrcat (basename, WC_URL_STR); 1.31 cvs 741: length++; 742: } 743: if (ptr) 744: TtaFreeMemory (ptr); 745: } 746: } 1.33 cvs 747: } 748: 1.31 cvs 749: /* Remove anything after the last DIR_SEP char. If no such char is found, 750: * then search for the first ":" char, hoping that what's before that is a 751: * protocol. If found, end the string there. If neither char is found, 752: * then discard the whole base element. 753: */ 1.67 cvs 754: length = ustrlen (basename) - 1; 1.31 cvs 755: /* search for the last DIR_SEP char */ 1.84 ! cvs 756: while (length >= 0 && basename[length] != WC_URL_SEP && basename[length] != WC_DIR_SEP) 1.31 cvs 757: length--; 758: if (length >= 0) 759: /* found the last DIR_SEP char, end the string there */ 1.84 ! cvs 760: basename[length + 1] = WC_EOS; 1.31 cvs 761: else 762: /* search for the first PATH_STR char */ 763: { 1.67 cvs 764: for (length = 0; basename[length] != TEXT(':') && 1.84 ! cvs 765: basename[length] != WC_EOS; length ++); 1.67 cvs 766: if (basename[length] == TEXT(':')) 1.31 cvs 767: /* found, so end the string there */ 1.84 ! cvs 768: basename[length + 1] = WC_EOS; 1.31 cvs 769: else 770: /* not found, discard the base */ 1.84 ! cvs 771: basename[0] = WC_EOS; 1.31 cvs 772: } 773: return (basename); 774: } 775: 776: 1.4 cvs 777: /*---------------------------------------------------------------------- 1.40 cvs 778: GetLocalPath 779: Allocate and return the local document path associated to the url 780: ----------------------------------------------------------------------*/ 781: #ifdef __STDC__ 1.84 ! cvs 782: CHAR_T* GetLocalPath (Document doc, CHAR_T* url) 1.40 cvs 783: #else /* __STDC__ */ 1.84 ! cvs 784: CHAR_T* GetLocalPath (doc, url) 1.40 cvs 785: Document doc; 1.84 ! cvs 786: CHAR_T* url; 1.40 cvs 787: #endif /* __STDC__ */ 788: { 1.84 ! cvs 789: CHAR_T* ptr; ! 790: CHAR_T* n; ! 791: CHAR_T* documentname; ! 792: CHAR_T url_sep; 1.83 cvs 793: int len; 1.67 cvs 794: ThotBool noFile; 1.40 cvs 795: 796: if (url != NULL) 797: { 798: /* check whether the file name exists */ 1.84 ! cvs 799: len = ustrlen (url) - 1; 1.71 cvs 800: if (IsW3Path (url)) 1.84 ! cvs 801: url_sep = TEXT('/'); 1.41 cvs 802: else 1.84 ! cvs 803: url_sep = WC_DIR_SEP; 1.41 cvs 804: noFile = (url[len] == url_sep); 1.40 cvs 805: if (noFile) 1.84 ! cvs 806: url[len] = WC_EOS; ! 807: ptr = TtaAllocString (MAX_LENGTH); ! 808: documentname = TtaAllocString (MAX_LENGTH); 1.78 cvs 809: TtaExtractName (url, ptr, documentname); 1.84 ! cvs 810: usprintf (ptr, TEXT("%s%s%d%s"), TempFileDirectory, WC_DIR_STR, doc, WC_DIR_STR); 1.40 cvs 811: if (!TtaCheckDirectory (ptr)) 812: /* directory did not exist */ 1.72 cvs 813: TtaMakeDirectory (ptr); 1.47 cvs 814: 815: /* don't include the query string within document name */ 1.84 ! cvs 816: n = ustrrchr (documentname, TEXT('?')); 1.47 cvs 817: if (n != NULL) 1.84 ! cvs 818: *n = WC_EOS; 1.46 cvs 819: /* don't include ':' within document name */ 1.84 ! cvs 820: n = ustrchr (documentname, TEXT(':')); 1.46 cvs 821: if (n != NULL) 1.84 ! cvs 822: *n = WC_EOS; 1.69 cvs 823: /* if after all this operations document name 824: is empty, let's use noname.html instead */ 1.84 ! cvs 825: if (documentname[0] == WC_EOS) ! 826: ustrcat (ptr, TEXT("noname.html")); 1.69 cvs 827: else 1.84 ! cvs 828: ustrcat (ptr, documentname); 1.40 cvs 829: TtaFreeMemory (documentname); 830: /* restore the url */ 831: if (noFile) 1.41 cvs 832: url[len] = url_sep; 1.40 cvs 833: return (ptr); 834: } 835: else 836: return (NULL); 837: } 838: 1.73 cvs 839: /*---------------------------------------------------------------------- 1.79 cvs 840: ExtractTarget extract the target name from document nane. 841: ----------------------------------------------------------------------*/ 842: #ifdef __STDC__ 1.84 ! cvs 843: void ExtractTarget (CHAR_T* aName, CHAR_T* target) 1.79 cvs 844: #else 845: void ExtractTarget (aName, target) 1.84 ! cvs 846: CHAR_T* aName; ! 847: CHAR_T* target; 1.79 cvs 848: #endif 849: { 1.82 cvs 850: int lg, i; 1.84 ! cvs 851: CHAR_T* ptr; ! 852: CHAR_T* oldptr; 1.79 cvs 853: 854: if (!target || !aName) 855: /* bad target */ 856: return; 857: 1.84 ! cvs 858: target[0] = WC_EOS; ! 859: lg = ustrlen (aName); 1.79 cvs 860: if (lg) 861: { 862: /* the name is not empty */ 863: oldptr = ptr = &aName[0]; 864: do 865: { 1.84 ! cvs 866: ptr = ustrrchr (oldptr, TEXT('#')); 1.79 cvs 867: if (ptr) 868: oldptr = &ptr[1]; 869: } 870: while (ptr); 871: 872: i = (int) (oldptr) - (int) (aName); /* name length */ 873: if (i > 1) 874: { 1.84 ! cvs 875: aName[i - 1] = WC_EOS; 1.79 cvs 876: if (i != lg) 1.84 ! cvs 877: ustrcpy (target, oldptr); 1.79 cvs 878: } 879: } 880: } 881: 882: /*---------------------------------------------------------------------- 1.73 cvs 883: ConvertFileURL 884: If the URL starts with file: prefix, it removes the protocol so that we 885: can use it as a local filename 886: ----------------------------------------------------------------------*/ 887: #ifdef __STDC__ 1.84 ! cvs 888: void ConvertFileURL (CHAR_T* url) 1.73 cvs 889: #else 890: void ConvertFileURL (url) 1.84 ! cvs 891: CHAR_T* url 1.73 cvs 892: #endif /* __STDC__ */ 893: { 1.84 ! cvs 894: if (!ustrncasecmp (url, TEXT("file:"), 5)) ! 895: ustrcpy (url, url + 5); 1.73 cvs 896: } 1.40 cvs 897: 898: /*---------------------------------------------------------------------- 1.9 cvs 899: NormalizeURL 900: normalizes orgName according to a base associated with doc, and 901: following the standard URL format rules. 1.53 cvs 902: if doc is 0 and otherPath not NULL, normalizes orgName according to this 903: other path. 1.9 cvs 904: The function returns the new complete and normalized URL 1.12 cvs 905: or file name path (newName) and the name of the document (docName). 1.9 cvs 906: N.B. If the function can't find out what's the docName, it assigns 907: the name "noname.html". 1.4 cvs 908: ----------------------------------------------------------------------*/ 1.3 cvs 909: #ifdef __STDC__ 1.84 ! cvs 910: void NormalizeURL (CHAR_T* orgName, Document doc, CHAR_T* newName, CHAR_T* docName, CHAR_T* otherPath) 1.3 cvs 911: #else /* __STDC__ */ 1.53 cvs 912: void NormalizeURL (orgName, doc, newName, docName, otherPath) 1.84 ! cvs 913: CHAR_T* orgName; 1.3 cvs 914: Document doc; 1.84 ! cvs 915: CHAR_T* newName; ! 916: CHAR_T* docName; ! 917: CHAR_T* otherPath; 1.3 cvs 918: #endif /* __STDC__ */ 919: { 1.84 ! cvs 920: CHAR_T* basename; ! 921: CHAR_T tempOrgName[MAX_LENGTH]; ! 922: CHAR_T* ptr; ! 923: CHAR_T used_sep; ! 924: int length; ! 925: ThotBool check; 1.5 cvs 926: 1.44 cvs 927: # ifdef _WINDOWS 928: int ndx; 929: # endif /* _WINDOWS */ 930: 1.5 cvs 931: if (!newName || !docName) 932: return; 1.18 cvs 933: 1.32 cvs 934: if (doc != 0) 1.53 cvs 935: basename = GetBaseURL (doc); 936: else if (otherPath != NULL) 1.84 ! cvs 937: basename = TtaWCSdup (otherPath); 1.32 cvs 938: else 1.53 cvs 939: basename = NULL; 1.32 cvs 940: 1.18 cvs 941: /* 1.31 cvs 942: * Clean orgName 943: * Make sure we have a complete orgName, without any leading or trailing 944: * white spaces, or trailinbg new lines 945: */ 1.5 cvs 946: ptr = orgName; 1.18 cvs 947: /* skip leading white space and new line characters */ 1.84 ! cvs 948: while ((*ptr == WC_SPACE || *ptr == WC_EOL) && *ptr++ != WC_EOS); ! 949: ustrncpy (tempOrgName, ptr, MAX_LENGTH -1); ! 950: tempOrgName[MAX_LENGTH -1] = WC_EOS; 1.18 cvs 951: /* 1.31 cvs 952: * Make orgName a complete URL 953: * If the URL does not include a protocol, then try to calculate 954: * one using the doc's base element (if it exists), 955: */ 1.84 ! cvs 956: if (tempOrgName[0] == WC_EOS) 1.53 cvs 957: { 1.84 ! cvs 958: newName[0] = WC_EOS; 1.53 cvs 959: TtaFreeMemory (basename); 960: return; 961: } 1.49 cvs 962: 963: /* clean trailing white space */ 1.84 ! cvs 964: length = ustrlen (tempOrgName) - 1; ! 965: while (tempOrgName[length] == WC_SPACE && tempOrgName[length] == WC_EOL) 1.53 cvs 966: { 1.84 ! cvs 967: tempOrgName[length] = WC_EOS; 1.53 cvs 968: length--; 969: } 1.50 cvs 970: 1.55 cvs 971: /* remove extra dot (which dot???) */ 972: /* ugly, but faster than a strcmp */ 1.84 ! cvs 973: if (tempOrgName[length] == TEXT('.') ! 974: && (length == 0 || tempOrgName[length-1] != TEXT('.'))) ! 975: tempOrgName[length] = WC_EOS; 1.50 cvs 976: 1.53 cvs 977: if (IsW3Path (tempOrgName)) 978: { 979: /* the name is complete, go to the Sixth Step */ 1.84 ! cvs 980: ustrcpy (newName, tempOrgName); 1.53 cvs 981: SimplifyUrl (&newName); 982: /* verify if the URL has the form "protocol://server:port" */ 1.84 ! cvs 983: ptr = AmayaParseUrl (newName, TEXT(""), AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); ! 984: if (ptr && !ustrcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */ ! 985: ustrcat (newName, WC_URL_STR); 1.49 cvs 986: 1.53 cvs 987: if (ptr) 1.50 cvs 988: TtaFreeMemory (ptr); 1.53 cvs 989: } 990: else if ( basename == NULL) 991: /* the name is complete, go to the Sixth Step */ 1.84 ! cvs 992: ustrcpy (newName, tempOrgName); 1.53 cvs 993: else 994: { 1.31 cvs 995: /* Calculate the absolute URL, using the base or document URL */ 1.44 cvs 996: # ifdef _WINDOWS 1.53 cvs 997: if (!IsW3Path (basename)) 998: { 1.67 cvs 999: length = ustrlen (tempOrgName); 1.53 cvs 1000: for (ndx = 0; ndx < length; ndx++) 1.67 cvs 1001: if (tempOrgName [ndx] == TEXT('/')) 1002: tempOrgName [ndx] = TEXT('\\'); 1.53 cvs 1003: } 1.44 cvs 1004: # endif /* _WINDOWS */ 1.25 cvs 1005: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL); 1.53 cvs 1006: if (ptr) 1007: { 1008: SimplifyUrl (&ptr); 1.84 ! cvs 1009: ustrcpy (newName, ptr); 1.53 cvs 1010: TtaFreeMemory (ptr); 1011: } 1012: else 1.84 ! cvs 1013: newName[0] = WC_EOS; 1.53 cvs 1014: } 1.36 cvs 1015: 1016: TtaFreeMemory (basename); 1.18 cvs 1017: /* 1.31 cvs 1018: * Prepare the docname that will refer to this ressource in the 1019: * .amaya directory. If the new URL finishes on DIR_SEP, then use 1020: * noname.html as a default ressource name 1.18 cvs 1021: */ 1.84 ! cvs 1022: if (newName[0] != WC_EOS) 1.53 cvs 1023: { 1.84 ! cvs 1024: length = ustrlen (newName) - 1; ! 1025: if (newName[length] == WC_URL_SEP || newName[length] == WC_DIR_SEP) 1.53 cvs 1026: { 1027: used_sep = newName[length]; 1028: check = TRUE; 1029: while (check) 1030: { 1.50 cvs 1031: length--; 1032: while (length >= 0 && newName[length] != used_sep) 1.53 cvs 1033: length--; 1.84 ! cvs 1034: if (!ustrncmp (&newName[length+1], TEXT(".."), 2)) 1.53 cvs 1035: { 1.84 ! cvs 1036: newName[length+1] = WC_EOS; 1.53 cvs 1037: /* remove also previous directory */ 1038: length--; 1039: while (length >= 0 && newName[length] != used_sep) 1040: length--; 1.84 ! cvs 1041: if (ustrncmp (&newName[length+1], TEXT("//"), 2)) 1.53 cvs 1042: /* don't remove server name */ 1.84 ! cvs 1043: newName[length+1] = WC_EOS; 1.53 cvs 1044: } 1.84 ! cvs 1045: else if (!ustrncmp (&newName[length+1], TEXT("."), 1)) ! 1046: newName[length+1] = WC_EOS; 1.50 cvs 1047: else 1.53 cvs 1048: check = FALSE; 1049: } 1.84 ! cvs 1050: ustrcpy (docName, TEXT("noname.html")); 1.53 cvs 1051: /* docname was not comprised inside the URL, so let's */ 1052: /* assign the default ressource name */ 1.84 ! cvs 1053: ustrcpy (docName, TEXT("noname.html")); 1.53 cvs 1054: } 1055: else 1056: { /* docname is comprised inside the URL */ 1.84 ! cvs 1057: while (length >= 0 && newName[length] != WC_URL_SEP && newName[length] != WC_DIR_SEP) 1.53 cvs 1058: length--; 1059: if (length < 0) 1.84 ! cvs 1060: ustrcpy (docName, newName); 1.53 cvs 1061: else 1.84 ! cvs 1062: ustrcpy (docName, &newName[length+1]); 1.53 cvs 1063: } 1064: } 1065: else 1.84 ! cvs 1066: docName[0] = WC_EOS; 1.18 cvs 1067: } 1.3 cvs 1068: 1.4 cvs 1069: /*---------------------------------------------------------------------- 1.9 cvs 1070: IsSameHost 1.4 cvs 1071: ----------------------------------------------------------------------*/ 1.3 cvs 1072: #ifdef __STDC__ 1.67 cvs 1073: ThotBool IsSameHost (const STRING url1, const STRING url2) 1.3 cvs 1074: #else /* __STDC__ */ 1.67 cvs 1075: ThotBool IsSameHost (url1, url2) 1076: const STRING url1; 1077: const STRING url2; 1.3 cvs 1078: #endif /* __STDC__ */ 1079: { 1.67 cvs 1080: STRING basename_ptr1, basename_ptr2; 1081: ThotBool result; 1.3 cvs 1082: 1.84 ! cvs 1083: basename_ptr1 = AmayaParseUrl (url1, TEXT(""), AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 1.67 cvs 1084: basename_ptr2 = AmayaParseUrl (url2, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 1.3 cvs 1085: 1.67 cvs 1086: if (ustrcmp (basename_ptr1, basename_ptr2)) 1.8 cvs 1087: result = FALSE; 1.5 cvs 1088: else 1.8 cvs 1089: result = TRUE; 1.3 cvs 1090: 1.25 cvs 1091: TtaFreeMemory (basename_ptr1); 1092: TtaFreeMemory (basename_ptr2); 1.5 cvs 1093: return (result); 1.3 cvs 1094: } 1095: 1096: 1.4 cvs 1097: /*---------------------------------------------------------------------- 1.22 cvs 1098: HasKnownFileSuffix 1099: returns TRUE if path points to a file ending with a suffix. 1100: ----------------------------------------------------------------------*/ 1101: #ifdef __STDC__ 1.67 cvs 1102: ThotBool HasKnownFileSuffix (const STRING path) 1.22 cvs 1103: #else /* __STDC__ */ 1.67 cvs 1104: ThotBool HasKnownFileSuffix (path) 1105: const STRING path; 1.22 cvs 1106: #endif /* __STDC__ */ 1107: { 1.67 cvs 1108: STRING root; 1109: CHAR_T temppath[MAX_LENGTH]; 1110: CHAR_T suffix[MAX_LENGTH]; 1.22 cvs 1111: 1.67 cvs 1112: if (!path || path[0] == EOS || path[ustrlen(path)] == DIR_SEP) 1.22 cvs 1113: return (FALSE); 1114: 1.84 ! cvs 1115: root = AmayaParseUrl(path, TEXT(""), AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION); 1.22 cvs 1116: 1117: if (root) 1118: { 1.67 cvs 1119: ustrcpy (temppath, root); 1.25 cvs 1120: TtaFreeMemory (root); 1.22 cvs 1121: /* Get the suffix */ 1122: ExtractSuffix (temppath, suffix); 1123: 1124: if( suffix[0] == EOS) 1125: /* no suffix */ 1126: return (FALSE); 1127: 1128: /* Normalize the suffix */ 1129: ConvertToLowerCase (suffix); 1130: 1.67 cvs 1131: if (!ustrcmp (suffix, TEXT("gz"))) 1.22 cvs 1132: /* skip the compressed suffix */ 1133: { 1134: ExtractSuffix (temppath, suffix); 1135: if(suffix[0] == EOS) 1136: /* no suffix */ 1137: return (FALSE); 1138: /* Normalize the suffix */ 1139: ConvertToLowerCase (suffix); 1140: } 1141: 1.67 cvs 1142: if (ustrcmp (suffix, TEXT("gif")) && 1143: ustrcmp (suffix, TEXT("xbm")) && 1144: ustrcmp (suffix, TEXT("xpm")) && 1145: ustrcmp (suffix, TEXT("jpg")) && 1146: ustrcmp (suffix, TEXT("pdf")) && 1147: ustrcmp (suffix, TEXT("png")) && 1148: ustrcmp (suffix, TEXT("tgz")) && 1149: ustrcmp (suffix, TEXT("xpg")) && 1150: ustrcmp (suffix, TEXT("xpd")) && 1151: ustrcmp (suffix, TEXT("ps")) && 1152: ustrcmp (suffix, TEXT("au")) && 1153: ustrcmp (suffix, TEXT("html")) && 1154: ustrcmp (suffix, TEXT("htm")) && 1155: ustrcmp (suffix, TEXT("shtml")) && 1156: ustrcmp (suffix, TEXT("xht")) && 1157: ustrcmp (suffix, TEXT("xhtm")) && 1158: ustrcmp (suffix, TEXT("xhtml")) && 1159: ustrcmp (suffix, TEXT("txt")) && 1160: ustrcmp (suffix, TEXT("css")) && 1161: ustrcmp (suffix, TEXT("eps"))) 1.22 cvs 1162: return (FALSE); 1163: else 1164: return (TRUE); 1165: } 1166: else 1167: return (FALSE); 1168: } 1169: 1170: 1171: /*---------------------------------------------------------------------- 1.24 cvs 1172: ChopURL 1173: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL). 1174: If inputURL is bigger than that size, outputURL receives 1175: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...", 1176: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL. 1177: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets 1178: copied into outputURL. 1179: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH 1180: chars. 1181: ----------------------------------------------------------------------*/ 1182: #ifdef __STDC__ 1.34 cvs 1183: void ChopURL (char *outputURL, const char *inputURL) 1.24 cvs 1184: #else 1185: void ChopURL (outputURL, inputURL) 1.34 cvs 1186: char *outputURL; 1187: const char *inputURL; 1.24 cvs 1188: #endif 1.22 cvs 1189: 1.24 cvs 1190: { 1191: int len; 1.9 cvs 1192: 1.24 cvs 1193: len = strlen (inputURL); 1194: if (len <= MAX_PRINT_URL_LENGTH) 1.29 cvs 1195: strcpy (outputURL, inputURL); 1.24 cvs 1196: else 1197: /* make a truncated urlName on the status window */ 1198: { 1199: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2); 1200: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS; 1201: strcat (outputURL, "..."); 1202: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ])); 1203: } 1.25 cvs 1204: } 1205: 1206: 1207: /*---------------------------------------------------------------------- 1208: scan 1.47 cvs 1209: Scan a filename for its constituents 1.25 cvs 1210: ----------------------------------- 1211: 1212: On entry, 1213: name points to a document name which may be incomplete. 1214: On exit, 1215: absolute or relative may be nonzero (but not both). 1216: host, fragment and access may be nonzero if they were specified. 1217: Any which are nonzero point to zero terminated strings. 1218: ----------------------------------------------------------------------*/ 1219: #ifdef __STDC__ 1.67 cvs 1220: static void scan (STRING name, HTURI * parts) 1.25 cvs 1221: #else /* __STDC__ */ 1222: static void scan (name, parts) 1.67 cvs 1223: STRING name; 1224: HTURI *parts; 1.25 cvs 1225: 1226: #endif /* __STDC__ */ 1227: { 1.67 cvs 1228: STRING p; 1229: STRING after_access = name; 1.32 cvs 1230: 1.43 cvs 1231: memset (parts, '\0', sizeof (HTURI)); 1.28 cvs 1232: /* Look for fragment identifier */ 1.67 cvs 1233: if ((p = ustrchr(name, TEXT('#'))) != NULL) 1.28 cvs 1234: { 1.67 cvs 1235: *p++ = TEXT('\0'); 1.28 cvs 1236: parts->fragment = p; 1.25 cvs 1237: } 1238: 1.28 cvs 1239: for (p=name; *p; p++) 1240: { 1.67 cvs 1241: if (*p == URL_SEP || *p == DIR_SEP || *p == TEXT('#') || *p == TEXT('?')) 1.28 cvs 1242: break; 1.67 cvs 1243: if (*p == TEXT(':')) 1.28 cvs 1244: { 1245: *p = 0; 1246: parts->access = after_access; /* Scheme has been specified */ 1247: 1248: /* The combination of gcc, the "-O" flag and the HP platform is 1249: unhealthy. The following three lines is a quick & dirty fix, but is 1250: not recommended. Rather, turn off "-O". */ 1251: 1252: /* after_access = p;*/ 1253: /* while (*after_access == 0)*/ 1254: /* after_access++;*/ 1255: after_access = p+1; 1.67 cvs 1256: if (!ustrcasecmp(TEXT("URL"), parts->access)) 1.28 cvs 1257: /* Ignore IETF's URL: pre-prefix */ 1258: parts->access = NULL; 1259: else 1.25 cvs 1260: break; 1261: } 1262: } 1263: 1264: p = after_access; 1.43 cvs 1265: if (*p == URL_SEP || *p == DIR_SEP) 1.28 cvs 1266: { 1.43 cvs 1267: if (p[1] == URL_SEP) 1.28 cvs 1268: { 1.25 cvs 1269: parts->host = p+2; /* host has been specified */ 1.28 cvs 1270: *p = 0; /* Terminate access */ 1271: /* look for end of host name if any */ 1.67 cvs 1272: p = ustrchr (parts->host, URL_SEP); 1.28 cvs 1273: if (p) 1274: { 1.43 cvs 1275: *p = EOS; /* Terminate host */ 1.25 cvs 1276: parts->absolute = p+1; /* Root has been found */ 1.28 cvs 1277: } 1278: } 1279: else 1280: /* Root found but no host */ 1281: parts->absolute = p+1; 1282: } 1283: else 1284: { 1.25 cvs 1285: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */ 1.28 cvs 1286: } 1.25 cvs 1287: } 1288: 1289: 1290: /*---------------------------------------------------------------------- 1.28 cvs 1291: AmayaParseUrl: parse a Name relative to another name 1292: 1293: This returns those parts of a name which are given (and requested) 1294: substituting bits from the related name where necessary. 1.25 cvs 1295: 1.28 cvs 1296: On entry, 1.25 cvs 1297: aName A filename given 1298: relatedName A name relative to which aName is to be parsed. Give 1299: it an empty string if aName is absolute. 1300: wanted A mask for the bits which are wanted. 1301: 1.28 cvs 1302: On exit, 1.25 cvs 1303: returns A pointer to a malloc'd string which MUST BE FREED 1304: ----------------------------------------------------------------------*/ 1305: #ifdef __STDC__ 1.84 ! cvs 1306: CHAR_T* AmayaParseUrl (const CHAR_T* aName, CHAR_T* relatedName, int wanted) 1.25 cvs 1307: #else /* __STDC__ */ 1.84 ! cvs 1308: CHAR_T* AmayaParseUrl (aName, relatedName, wanted) ! 1309: const CHAR_T* aName; ! 1310: CHAR_T* relatedName; 1.28 cvs 1311: int wanted; 1.25 cvs 1312: 1313: #endif /* __STDC__ */ 1314: { 1.84 ! cvs 1315: CHAR_T* return_value; 1.67 cvs 1316: CHAR_T result[MAX_LENGTH]; 1317: CHAR_T name[MAX_LENGTH]; 1318: CHAR_T rel[MAX_LENGTH]; 1.84 ! cvs 1319: CHAR_T *p, *access; 1.29 cvs 1320: HTURI given, related; 1321: int len; 1.67 cvs 1322: CHAR_T used_sep; 1.84 ! cvs 1323: CHAR_T* used_str; 1.32 cvs 1324: 1.84 ! cvs 1325: if (ustrchr (aName, WC_DIR_SEP) || ustrchr (relatedName, WC_DIR_SEP)) 1.33 cvs 1326: { 1.84 ! cvs 1327: used_str = WC_DIR_STR; ! 1328: used_sep = WC_DIR_SEP; 1.33 cvs 1329: } 1.32 cvs 1330: else 1.33 cvs 1331: { 1.84 ! cvs 1332: used_str = WC_URL_STR; ! 1333: used_sep = WC_URL_SEP; 1.33 cvs 1334: } 1.32 cvs 1335: 1.29 cvs 1336: /* Make working copies of input strings to cut up: */ 1337: return_value = NULL; 1338: result[0] = 0; /* Clear string */ 1.67 cvs 1339: ustrcpy (name, aName); 1.29 cvs 1340: if (relatedName != NULL) 1.67 cvs 1341: ustrcpy (rel, relatedName); 1.29 cvs 1342: else 1.84 ! cvs 1343: relatedName[0] = WC_EOS; 1.29 cvs 1344: 1345: scan (name, &given); 1346: scan (rel, &related); 1347: access = given.access ? given.access : related.access; 1348: if (wanted & AMAYA_PARSE_ACCESS) 1349: if (access) 1350: { 1.67 cvs 1351: ustrcat (result, access); 1.29 cvs 1352: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1353: ustrcat (result, TEXT(":")); 1.29 cvs 1354: } 1355: 1356: if (given.access && related.access) 1357: /* If different, inherit nothing. */ 1.67 cvs 1358: if (ustrcmp (given.access, related.access) != 0) 1.29 cvs 1359: { 1360: related.host = 0; 1361: related.absolute = 0; 1362: related.relative = 0; 1363: related.fragment = 0; 1364: } 1365: 1366: if (wanted & AMAYA_PARSE_HOST) 1367: if(given.host || related.host) 1368: { 1369: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1370: ustrcat (result, TEXT("//")); 1371: ustrcat (result, given.host ? given.host : related.host); 1.29 cvs 1372: } 1373: 1374: if (given.host && related.host) 1375: /* If different hosts, inherit no path. */ 1.67 cvs 1376: if (ustrcmp (given.host, related.host) != 0) 1.29 cvs 1377: { 1378: related.absolute = 0; 1379: related.relative = 0; 1380: related.fragment = 0; 1381: } 1382: 1383: if (wanted & AMAYA_PARSE_PATH) 1384: { 1385: if (given.absolute) 1386: { 1387: /* All is given */ 1388: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1389: ustrcat (result, used_str); 1390: ustrcat (result, given.absolute); 1.25 cvs 1391: } 1.29 cvs 1392: else if (related.absolute) 1393: { 1394: /* Adopt path not name */ 1.67 cvs 1395: ustrcat (result, used_str); 1396: ustrcat (result, related.absolute); 1.29 cvs 1397: if (given.relative) 1398: { 1399: /* Search part? */ 1.67 cvs 1400: p = ustrchr (result, TEXT('?')); 1.29 cvs 1401: if (!p) 1.67 cvs 1402: p=result+ustrlen(result)-1; 1.33 cvs 1403: for (; *p!=used_sep; p--); /* last / */ 1.29 cvs 1404: /* Remove filename */ 1405: p[1]=0; 1406: /* Add given one */ 1.67 cvs 1407: ustrcat (result, given.relative); 1.25 cvs 1408: } 1409: } 1.29 cvs 1410: else if (given.relative) 1411: /* what we've got */ 1.67 cvs 1412: ustrcat (result, given.relative); 1.29 cvs 1413: else if (related.relative) 1.67 cvs 1414: ustrcat (result, related.relative); 1.29 cvs 1415: else 1416: /* No inheritance */ 1.67 cvs 1417: ustrcat (result, used_str); 1.25 cvs 1418: } 1.29 cvs 1419: 1420: if (wanted & AMAYA_PARSE_ANCHOR) 1421: if (given.fragment || related.fragment) 1422: { 1423: if (given.absolute && given.fragment) 1424: { 1425: /*Fixes for relURLs...*/ 1426: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1427: ustrcat (result, TEXT("#")); 1428: ustrcat (result, given.fragment); 1.29 cvs 1429: } 1430: else if (!(given.absolute) && !(given.fragment)) 1.84 ! cvs 1431: ustrcat (result, TEXT("")); 1.29 cvs 1432: else 1433: { 1434: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1435: ustrcat (result, TEXT("#")); 1436: ustrcat (result, given.fragment ? given.fragment : related.fragment); 1.29 cvs 1437: } 1438: } 1.67 cvs 1439: len = ustrlen (result); 1440: if ((return_value = TtaAllocString (len + 1)) != NULL) 1441: ustrcpy (return_value, result); 1.29 cvs 1442: return (return_value); /* exactly the right length */ 1.25 cvs 1443: } 1444: 1445: /*---------------------------------------------------------------------- 1446: HTCanon 1447: Canonicalizes the URL in the following manner starting from the host 1448: pointer: 1449: 1450: 1) The host name is converted to lowercase 1451: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp) 1452: 1453: Return: OK The position of the current path part of the URL 1454: which might be the old one or a new one. 1455: 1456: ----------------------------------------------------------------------*/ 1457: #ifdef __STDC__ 1.84 ! cvs 1458: static CHAR_T* HTCanon (CHAR_T** filename, CHAR_T* host) 1.25 cvs 1459: #else /* __STDC__ */ 1.67 cvs 1460: static STRING HTCanon (filename, host) 1.84 ! cvs 1461: CHAR_T** filename; ! 1462: CHAR_T* host; 1.25 cvs 1463: #endif /* __STDC__ */ 1464: { 1.84 ! cvs 1465: CHAR_T* newname = NULL; ! 1466: CHAR_T used_sep; ! 1467: CHAR_T* path; ! 1468: CHAR_T* strptr; ! 1469: CHAR_T* port; ! 1470: CHAR_T* access = host-3; 1.32 cvs 1471: 1472: 1.84 ! cvs 1473: if (*filename && ustrchr (*filename, WC_URL_SEP)) 1.33 cvs 1474: { 1.84 ! cvs 1475: used_sep = WC_URL_SEP; 1.33 cvs 1476: } 1477: else 1478: { 1.84 ! cvs 1479: used_sep = WC_DIR_SEP; 1.33 cvs 1480: } 1.32 cvs 1481: 1.82 cvs 1482: while (access > *filename && *(access - 1) != used_sep) /* Find access method */ 1.25 cvs 1483: access--; 1.84 ! cvs 1484: if ((path = ustrchr (host, used_sep)) == NULL) /* Find path */ ! 1485: path = host + ustrlen (host); ! 1486: if ((strptr = ustrchr (host, TEXT('@'))) != NULL && strptr < path) /* UserId */ 1.82 cvs 1487: host = strptr; 1.84 ! cvs 1488: if ((port = ustrchr (host, TEXT(':'))) != NULL && port > path) /* Port number */ 1.82 cvs 1489: port = NULL; 1.25 cvs 1490: 1491: strptr = host; /* Convert to lower-case */ 1.82 cvs 1492: while (strptr < path) 1.33 cvs 1493: { 1.84 ! cvs 1494: *strptr = utolower (*strptr); 1.82 cvs 1495: strptr++; 1.33 cvs 1496: } 1.25 cvs 1497: 1498: /* Does the URL contain a full domain name? This also works for a 1499: numerical host name. The domain name is already made lower-case 1500: and without a trailing dot. */ 1501: { 1.84 ! cvs 1502: CHAR_T* dot = port ? port : path; ! 1503: if (dot > *filename && *--dot == TEXT('.')) 1.33 cvs 1504: { 1.84 ! cvs 1505: CHAR_T* orig = dot; ! 1506: CHAR_T* dest = dot + 1; 1.82 cvs 1507: while ((*orig++ = *dest++)); 1508: if (port) port--; 1.33 cvs 1509: path--; 1.25 cvs 1510: } 1511: } 1512: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */ 1.33 cvs 1513: if (port) 1514: { 1.82 cvs 1515: if (!*(port+1) || *(port+1) == used_sep) 1.33 cvs 1516: { 1517: if (!newname) 1518: { 1.84 ! cvs 1519: CHAR_T* orig = port; ! 1520: CHAR_T* dest = port + 1; 1.82 cvs 1521: while ((*orig++ = *dest++)); 1.33 cvs 1522: } 1523: } 1.84 ! cvs 1524: else if ((!ustrncmp (access, TEXT("http"), 4) && ! 1525: (*(port + 1) == TEXT('8') && ! 1526: *(port+2) == TEXT('0') && 1.82 cvs 1527: (*(port+3) == used_sep || !*(port + 3)))) || 1.84 ! cvs 1528: (!ustrncmp (access, TEXT("gopher"), 6) && ! 1529: (*(port+1) == TEXT('7') && ! 1530: *(port+2) == TEXT('0') && 1.82 cvs 1531: (*(port+3) == used_sep || !*(port+3)))) || 1.84 ! cvs 1532: (!ustrncmp (access, TEXT("ftp"), 3) && ! 1533: (*(port+1) == TEXT('2') && ! 1534: *(port + 2) == TEXT('1') && 1.82 cvs 1535: (*(port+3) == used_sep || !*(port+3))))) { 1.33 cvs 1536: if (!newname) 1537: { 1.84 ! cvs 1538: CHAR_T* orig = port; ! 1539: CHAR_T* dest = port + 3; 1.33 cvs 1540: while((*orig++ = *dest++)); 1541: /* Update path position, Henry Minsky */ 1542: path -= 3; 1.25 cvs 1543: } 1.33 cvs 1544: } 1545: else if (newname) 1.84 ! cvs 1546: ustrncat (newname, port, (int) (path - port)); 1.33 cvs 1547: } 1.25 cvs 1548: 1.33 cvs 1549: if (newname) 1550: { 1.84 ! cvs 1551: CHAR_T* newpath = newname + ustrlen (newname); ! 1552: ustrcat (newname, path); 1.25 cvs 1553: path = newpath; 1.28 cvs 1554: /* Free old copy */ 1555: TtaFreeMemory(*filename); 1.25 cvs 1556: *filename = newname; 1.33 cvs 1557: } 1.25 cvs 1558: return path; 1559: } 1560: 1561: 1562: /*---------------------------------------------------------------------- 1.29 cvs 1563: SimplifyUrl: simplify a URI 1.32 cvs 1564: A URI is allowed to contain the sequence xxx/../ which may be 1565: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR. 1.28 cvs 1566: Simplification helps us recognize duplicate URIs. 1.25 cvs 1567: 1.28 cvs 1568: Thus, /etc/junk/../fred becomes /etc/fred 1569: /etc/junk/./fred becomes /etc/junk/fred 1.25 cvs 1570: 1.28 cvs 1571: but we should NOT change 1572: http://fred.xxx.edu/../.. 1.25 cvs 1573: 1574: or ../../albert.html 1575: 1.28 cvs 1576: In order to avoid empty URLs the following URLs become: 1.25 cvs 1577: 1578: /fred/.. becomes /fred/.. 1579: /fred/././.. becomes /fred/.. 1580: /fred/.././junk/.././ becomes /fred/.. 1581: 1.28 cvs 1582: If more than one set of `://' is found (several proxies in cascade) then 1583: only the part after the last `://' is simplified. 1.25 cvs 1584: 1.28 cvs 1585: Returns: A string which might be the old one or a new one. 1.25 cvs 1586: ----------------------------------------------------------------------*/ 1587: #ifdef __STDC__ 1.84 ! cvs 1588: void SimplifyUrl (CHAR_T** url) 1.25 cvs 1589: #else /* __STDC__ */ 1.29 cvs 1590: void SimplifyUrl (url) 1.84 ! cvs 1591: CHAR_T** url; 1.25 cvs 1592: #endif /* __STDC__ */ 1593: { 1.84 ! cvs 1594: CHAR_T* path; ! 1595: CHAR_T* access; ! 1596: CHAR_T* newptr; ! 1597: CHAR_T* p; ! 1598: CHAR_T *orig, *dest, *end; 1.28 cvs 1599: 1.82 cvs 1600: CharUnit used_sep; 1.77 cvs 1601: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation: 1602: something/../ simplification in relative URLs when they start with a ../ */ 1.32 cvs 1603: 1604: 1.28 cvs 1605: if (!url || !*url) 1606: return; 1607: 1.84 ! cvs 1608: if (ustrchr (*url, WC_URL_SEP)) 1.33 cvs 1609: { 1.84 ! cvs 1610: used_sep = WC_URL_SEP; 1.33 cvs 1611: } 1.32 cvs 1612: else 1.33 cvs 1613: { 1.84 ! cvs 1614: used_sep = WC_DIR_SEP; 1.33 cvs 1615: } 1.32 cvs 1616: 1.77 cvs 1617: /* should we simplify double dot? */ 1618: path = *url; 1.84 ! cvs 1619: if (*path == TEXT('.') && *(path + 1) == TEXT('.')) 1.77 cvs 1620: ddot_simplify = FALSE; 1621: else 1622: ddot_simplify = TRUE; 1623: 1.28 cvs 1624: /* Find any scheme name */ 1.84 ! cvs 1625: if ((path = ustrstr (*url, TEXT("://"))) != NULL) 1.33 cvs 1626: { 1627: /* Find host name */ 1.28 cvs 1628: access = *url; 1.84 ! cvs 1629: while (access < path && (*access = utolower (*access))) 1.82 cvs 1630: access++; 1.28 cvs 1631: path += 3; 1.84 ! cvs 1632: while ((newptr = ustrstr (path, TEXT ("://"))) != NULL) 1.82 cvs 1633: /* For proxies */ 1634: path = newptr+3; 1635: /* We have a host name */ 1.84 ! cvs 1636: path = HTCanon (url, path); 1.25 cvs 1637: } 1.84 ! cvs 1638: else if ((path = ustrstr (*url, TEXT(":/"))) != NULL) 1.28 cvs 1639: path += 2; 1640: else 1641: path = *url; 1.25 cvs 1642: 1.84 ! cvs 1643: if (*path == used_sep && *(path+1) == used_sep) 1.28 cvs 1644: /* Some URLs start //<foo> */ 1645: path += 1; 1.84 ! cvs 1646: else if (!ustrncmp (path, TEXT("news:"), 5)) 1.28 cvs 1647: { 1.84 ! cvs 1648: newptr = ustrchr (path+5, TEXT('@')); 1.28 cvs 1649: if (!newptr) 1650: newptr = path + 5; 1651: while (*newptr) 1652: { 1653: /* Make group or host lower case */ 1.84 ! cvs 1654: *newptr = utolower (*newptr); 1.28 cvs 1655: newptr++; 1.25 cvs 1656: } 1.28 cvs 1657: /* Doesn't need to do any more */ 1658: return; 1.25 cvs 1659: } 1.28 cvs 1660: 1661: if ((p = path)) 1662: { 1.67 cvs 1663: if (!((end = ustrchr (path, TEXT(';'))) || (end = ustrchr (path, TEXT('?'))) || 1664: (end = ustrchr (path, TEXT('#'))))) 1665: end = path + ustrlen (path); 1.28 cvs 1666: 1667: /* Parse string second time to simplify */ 1668: p = path; 1669: while (p < end) 1670: { 1.77 cvs 1671: /* if we're pointing to a char, it's safe to reactivate the ../ convertion */ 1672: if (!ddot_simplify && *p != TEXT('.') && *p != used_sep) 1673: ddot_simplify = TRUE; 1674: 1.33 cvs 1675: if (*p==used_sep) 1.28 cvs 1676: { 1.67 cvs 1677: if (p > *url && *(p+1) == TEXT('.') && (*(p+2) == used_sep || !*(p+2))) 1.28 cvs 1678: { 1679: orig = p + 1; 1.84 ! cvs 1680: dest = (*(p+2) != used_sep) ? p+2 : p+3; 1.52 cvs 1681: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/ 1.28 cvs 1682: end = orig - 1; 1683: } 1.77 cvs 1684: else if (ddot_simplify && *(p+1) == TEXT('.') && *(p+2) == TEXT('.') 1685: && (*(p+3) == used_sep || !*(p+3))) 1.28 cvs 1686: { 1687: newptr = p; 1.52 cvs 1688: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */ 1689: if (*newptr == used_sep) 1690: orig = newptr + 1; 1.28 cvs 1691: else 1.52 cvs 1692: orig = newptr; 1693: 1694: dest = (*(p+3) != used_sep) ? p+3 : p+4; 1695: while ((*orig++ = *dest++)); /* Remove /xxx/.. */ 1696: end = orig-1; 1697: /* Start again with prev slash */ 1698: p = newptr; 1.28 cvs 1699: } 1.33 cvs 1700: else if (*(p+1) == used_sep) 1.28 cvs 1701: { 1.33 cvs 1702: while (*(p+1) == used_sep) 1.28 cvs 1703: { 1704: orig = p; 1705: dest = p + 1; 1706: while ((*orig++ = *dest++)); /* Remove multiple /'s */ 1707: end = orig-1; 1708: } 1709: } 1710: else 1.25 cvs 1711: p++; 1.28 cvs 1712: } 1713: else 1714: p++; 1.25 cvs 1715: } 1716: } 1.51 cvs 1717: 1718: /* 1719: ** Check for host/../.. kind of things 1720: */ 1.77 cvs 1721: if (*path == used_sep && *(path+1) == TEXT('.') && *(path+2) == TEXT('.') 1722: && (!*(path+3) || *(path+3) == used_sep)) 1.51 cvs 1723: *(path+1) = EOS; 1724: 1.28 cvs 1725: return; 1726: } 1727: 1728: 1729: /*---------------------------------------------------------------------- 1730: NormalizeFile normalizes local names. 1731: Return TRUE if target and src differ. 1732: ----------------------------------------------------------------------*/ 1733: #ifdef __STDC__ 1.84 ! cvs 1734: ThotBool NormalizeFile (CHAR_T* src, CHAR_T* target) 1.28 cvs 1735: #else 1.67 cvs 1736: ThotBool NormalizeFile (src, target) 1.84 ! cvs 1737: CHAR_T* src; ! 1738: CHAR_T* target; 1.28 cvs 1739: 1740: #endif 1741: { 1.84 ! cvs 1742: CHAR_T* s; 1.82 cvs 1743: ThotBool change; 1.28 cvs 1744: 1.54 cvs 1745: change = FALSE; 1.84 ! cvs 1746: if (ustrncmp (src, TEXT("file:"), 5) == 0) 1.28 cvs 1747: { 1748: /* remove the prefix file: */ 1.84 ! cvs 1749: if (src[5] == WC_EOS) ! 1750: ustrcpy (target, WC_DIR_STR); ! 1751: else if (src[0] == TEXT('~')) 1.28 cvs 1752: { 1753: /* replace ~ */ 1.74 cvs 1754: s = TtaGetEnvString ("HOME"); 1.84 ! cvs 1755: ustrcpy (target, s); ! 1756: ustrcat (target, &src[5]); 1.28 cvs 1757: } 1758: else 1.84 ! cvs 1759: ustrcpy (target, &src[5]); 1.54 cvs 1760: change = TRUE; 1.28 cvs 1761: } 1.53 cvs 1762: # ifndef _WINDOWS 1.84 ! cvs 1763: else if (src[0] == TEXT('~')) 1.53 cvs 1764: { 1765: /* replace ~ */ 1.82 cvs 1766: s = TtaGetEnvString ("HOME"); 1.84 ! cvs 1767: ustrcpy (target, s); ! 1768: if (src[1] != WC_DIR_SEP) ! 1769: ustrcat (target, WC_DIR_STR); ! 1770: ustrcat (target, &src[1]); 1.54 cvs 1771: change = TRUE; 1.53 cvs 1772: } 1773: # endif /* _WINDOWS */ 1.28 cvs 1774: else 1.84 ! cvs 1775: ustrcpy (target, src); 1.28 cvs 1776: 1777: /* remove /../ and /./ */ 1.29 cvs 1778: SimplifyUrl (&target); 1.54 cvs 1779: if (!change) 1.84 ! cvs 1780: change = ustrcmp (src, target); 1.28 cvs 1781: return (change); 1.25 cvs 1782: } 1783: 1.28 cvs 1784: 1.25 cvs 1785: /*---------------------------------------------------------------------- 1.31 cvs 1786: MakeRelativeURL: make relative name 1.25 cvs 1787: 1.28 cvs 1788: This function creates and returns a string which gives an expression of 1789: one address as related to another. Where there is no relation, an absolute 1790: address is retured. 1.25 cvs 1791: 1.28 cvs 1792: On entry, 1.25 cvs 1793: Both names must be absolute, fully qualified names of nodes 1794: (no fragment bits) 1795: 1.28 cvs 1796: On exit, 1.25 cvs 1797: The return result points to a newly allocated name which, if 1798: parsed by AmayaParseUrl relative to relatedName, will yield aName. 1799: The caller is responsible for freeing the resulting name later. 1800: ----------------------------------------------------------------------*/ 1801: #ifdef __STDC__ 1.67 cvs 1802: STRING MakeRelativeURL (STRING aName, STRING relatedName) 1.25 cvs 1803: #else /* __STDC__ */ 1.67 cvs 1804: STRING MakeRelativeURL (aName, relatedName) 1805: STRING aName; 1806: STRING relatedName; 1.25 cvs 1807: #endif /* __STDC__ */ 1808: { 1.67 cvs 1809: STRING return_value; 1810: CHAR_T result[MAX_LENGTH]; 1811: STRING p; 1812: STRING q; 1813: STRING after_access; 1814: STRING last_slash = NULL; 1.29 cvs 1815: int slashes, levels, len; 1816: 1.44 cvs 1817: # ifdef _WINDOWS 1818: int ndx; 1819: # endif /* _WINDOWS */ 1820: 1.29 cvs 1821: if (aName == NULL || relatedName == NULL) 1822: return (NULL); 1823: 1824: slashes = 0; 1825: after_access = NULL; 1826: p = aName; 1827: q = relatedName; 1828: for (; *p && (*p == *q); p++, q++) 1.27 cvs 1829: { 1830: /* Find extent of match */ 1.67 cvs 1831: if (*p == TEXT(':')) 1.29 cvs 1832: after_access = p + 1; 1.28 cvs 1833: if (*p == DIR_SEP) 1.27 cvs 1834: { 1.29 cvs 1835: /* memorize the last slash position and count them */ 1.27 cvs 1836: last_slash = p; 1837: slashes++; 1.25 cvs 1838: } 1839: } 1840: 1.31 cvs 1841: /* q, p point to the first non-matching character or zero */ 1842: if (*q == EOS) 1843: { 1844: /* New name is a subset of the related name */ 1845: /* exactly the right length */ 1.67 cvs 1846: len = ustrlen (p); 1847: if ((return_value = TtaAllocString (len + 1)) != NULL) 1848: ustrcpy (return_value, p); 1.31 cvs 1849: } 1850: else if ((slashes < 2 && after_access == NULL) 1851: || (slashes < 3 && after_access != NULL)) 1852: { 1853: /* Two names whitout common path */ 1854: /* exactly the right length */ 1.67 cvs 1855: len = ustrlen (aName); 1856: if ((return_value = TtaAllocString (len + 1)) != NULL) 1857: ustrcpy (return_value, aName); 1.31 cvs 1858: } 1859: else 1860: { 1861: /* Some path in common */ 1.67 cvs 1862: if (slashes == 3 && ustrncmp (aName, TEXT("http:"), 5) == 0) 1.31 cvs 1863: /* just the same server */ 1.67 cvs 1864: ustrcpy (result, last_slash); 1.31 cvs 1865: else 1866: { 1867: levels= 0; 1.67 cvs 1868: for (; *q && *q != TEXT('#') && *q != TEXT(';') && *q != TEXT('?'); q++) 1.31 cvs 1869: if (*q == DIR_SEP) 1870: levels++; 1871: 1.52 cvs 1872: result[0] = EOS; 1.31 cvs 1873: for (;levels; levels--) 1.67 cvs 1874: ustrcat (result, TEXT("../")); 1875: ustrcat (result, last_slash+1); 1.31 cvs 1876: } 1.52 cvs 1877: 1878: if (!*result) 1.67 cvs 1879: ustrcat (result, TEXT("./")); 1.52 cvs 1880: 1.31 cvs 1881: /* exactly the right length */ 1.67 cvs 1882: len = ustrlen (result); 1883: if ((return_value = TtaAllocString (len + 1)) != NULL) 1884: ustrcpy (return_value, result); 1.52 cvs 1885: 1.25 cvs 1886: } 1.44 cvs 1887: # ifdef _WINDOWS 1.67 cvs 1888: len = ustrlen (return_value); 1.44 cvs 1889: for (ndx = 0; ndx < len; ndx ++) 1.67 cvs 1890: if (return_value[ndx] == TEXT('\\')) 1891: return_value[ndx] = TEXT('/') ; 1.44 cvs 1892: # endif /* _WINDOWS */ 1.29 cvs 1893: return (return_value); 1.24 cvs 1894: } 1.35 cvs 1895: 1896: