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 */ 105: case TEXT(0x27): /* & */ 106: case TEXT(0x20): /* space */ 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__ 1.67 cvs 325: ThotBool IsHTMLName (const STRING path) 1.3 cvs 326: #else /* __STDC__ */ 1.67 cvs 327: ThotBool IsHTMLName (path) 328: const STRING path; 1.3 cvs 329: #endif /* __STDC__ */ 330: { 1.67 cvs 331: CHAR_T temppath[MAX_LENGTH]; 332: CHAR_T suffix[MAX_LENGTH]; 333: CHAR_T nsuffix[MAX_LENGTH]; 1.5 cvs 334: int i; 335: 336: if (!path) 1.37 cvs 337: return (FALSE); 1.5 cvs 338: 1.67 cvs 339: ustrcpy (temppath, path); 1.5 cvs 340: ExtractSuffix (temppath, suffix); 341: 342: /* Normalize the suffix */ 343: i = 0; 1.39 cvs 344: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 345: { 1.67 cvs 346: nsuffix[i] = utolower (suffix[i]); 1.13 cvs 347: i++; 348: } 1.5 cvs 349: nsuffix[i] = EOS; 1.67 cvs 350: if (!ustrcmp (nsuffix, TEXT("html")) || 351: !ustrcmp (nsuffix, TEXT("htm")) || 352: !ustrcmp (nsuffix, TEXT("shtml")) || 353: !ustrcmp (nsuffix, TEXT("xht")) || 354: !ustrcmp (nsuffix, TEXT("xhtm")) || 355: !ustrcmp (nsuffix, TEXT("xhtml"))) 1.39 cvs 356: return (TRUE); 1.67 cvs 357: else if (!ustrcmp (nsuffix, TEXT("gz"))) 1.13 cvs 358: { 1.39 cvs 359: /* take into account compressed files */ 1.13 cvs 360: ExtractSuffix (temppath, suffix); 361: /* Normalize the suffix */ 362: i = 0; 1.39 cvs 363: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 364: { 1.25 cvs 365: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 366: i++; 367: } 368: nsuffix[i] = EOS; 1.67 cvs 369: if (!ustrcmp (nsuffix, TEXT("html")) || 370: !ustrcmp (nsuffix, TEXT("htm")) || 371: !ustrcmp (nsuffix, TEXT("shtml")) || 372: !ustrcmp (nsuffix, TEXT("xht")) || 373: !ustrcmp (nsuffix, TEXT("xhtm")) || 374: !ustrcmp (nsuffix, TEXT("xhtml"))) 1.63 cvs 375: 1.39 cvs 376: return (TRUE); 377: else 1.13 cvs 378: return (FALSE); 379: } 380: else 1.39 cvs 381: return (FALSE); 1.3 cvs 382: } 383: 1.4 cvs 384: /*---------------------------------------------------------------------- 1.56 cvs 385: IsXMLName 386: returns TRUE if path points to an XML resource. 387: ----------------------------------------------------------------------*/ 388: #ifdef __STDC__ 1.67 cvs 389: ThotBool IsXMLName (const STRING path) 1.56 cvs 390: #else /* __STDC__ */ 1.67 cvs 391: ThotBool IsXMLName (path) 392: const STRING path; 1.56 cvs 393: #endif /* __STDC__ */ 394: { 1.67 cvs 395: CHAR_T temppath[MAX_LENGTH]; 396: CHAR_T suffix[MAX_LENGTH]; 1.56 cvs 397: 398: if (!path) 399: return (FALSE); 400: 1.67 cvs 401: ustrcpy (temppath, path); 1.56 cvs 402: ExtractSuffix (temppath, suffix); 403: 1.67 cvs 404: if (!ustrcasecmp (suffix, TEXT("xml")) || 405: !ustrcasecmp (suffix, TEXT("xht")) || 406: !ustrcmp (suffix, TEXT("xhtm")) || 407: !ustrcmp (suffix, TEXT("xhtml"))) 1.56 cvs 408: return (TRUE); 1.67 cvs 409: else if (!ustrcmp (suffix, TEXT("gz"))) 1.56 cvs 410: { 411: /* take into account compressed files */ 412: ExtractSuffix (temppath, suffix); 1.67 cvs 413: if (!ustrcasecmp (suffix, TEXT("xml")) || 414: !ustrcasecmp (suffix, TEXT("xht")) || 415: !ustrcmp (suffix, TEXT("xhtm")) || 416: !ustrcmp (suffix, TEXT("xhtml"))) 1.60 cvs 417: return (TRUE); 418: else 419: return (FALSE); 420: } 421: else 422: return (FALSE); 423: } 424: 425: /*---------------------------------------------------------------------- 426: IsCSSName 427: returns TRUE if path points to an XML resource. 428: ----------------------------------------------------------------------*/ 429: #ifdef __STDC__ 1.67 cvs 430: ThotBool IsCSSName (const STRING path) 1.60 cvs 431: #else /* __STDC__ */ 1.67 cvs 432: ThotBool IsCSSName (path) 433: const STRING path; 1.60 cvs 434: #endif /* __STDC__ */ 435: { 1.67 cvs 436: CHAR_T temppath[MAX_LENGTH]; 437: CHAR_T suffix[MAX_LENGTH]; 1.60 cvs 438: 439: if (!path) 440: return (FALSE); 441: 1.67 cvs 442: ustrcpy (temppath, path); 1.60 cvs 443: ExtractSuffix (temppath, suffix); 444: 1.67 cvs 445: if (!ustrcasecmp (suffix, TEXT("css"))) 1.60 cvs 446: return (TRUE); 1.67 cvs 447: else if (!ustrcmp (suffix, TEXT("gz"))) 1.60 cvs 448: { 449: /* take into account compressed files */ 450: ExtractSuffix (temppath, suffix); 1.67 cvs 451: if (!ustrcasecmp (suffix, TEXT("css"))) 1.56 cvs 452: return (TRUE); 453: else 454: return (FALSE); 455: } 456: else 457: return (FALSE); 458: } 459: 460: /*---------------------------------------------------------------------- 1.9 cvs 461: IsImageName 462: returns TRUE if path points to an image resource. 1.4 cvs 463: ----------------------------------------------------------------------*/ 1.3 cvs 464: #ifdef __STDC__ 1.67 cvs 465: ThotBool IsImageName (const STRING path) 1.3 cvs 466: #else /* __STDC__ */ 1.67 cvs 467: ThotBool IsImageName (path) 468: const STRING path; 1.3 cvs 469: #endif /* __STDC__ */ 470: { 1.67 cvs 471: CHAR_T temppath[MAX_LENGTH]; 472: CHAR_T suffix[MAX_LENGTH]; 473: CHAR_T nsuffix[MAX_LENGTH]; 1.5 cvs 474: int i; 475: 476: if (!path) 1.13 cvs 477: return (FALSE); 1.5 cvs 478: 1.67 cvs 479: ustrcpy (temppath, path); 1.5 cvs 480: ExtractSuffix (temppath, suffix); 481: 482: /* Normalize the suffix */ 483: i = 0; 1.39 cvs 484: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 485: { 1.67 cvs 486: nsuffix[i] = utolower (suffix[i]); 1.13 cvs 487: i++; 488: } 1.5 cvs 489: nsuffix[i] = EOS; 1.67 cvs 490: if ((!ustrcmp (nsuffix, TEXT("gif"))) || (!ustrcmp (nsuffix, TEXT("xbm"))) || 491: (!ustrcmp (nsuffix, TEXT("xpm"))) || (!ustrcmp (nsuffix, TEXT("jpg"))) || 492: (!ustrcmp (nsuffix, TEXT("png"))) || (!ustrcmp (nsuffix, TEXT("au")))) 1.39 cvs 493: return (TRUE); 494: return (FALSE); 1.3 cvs 495: } 496: 1.4 cvs 497: /*---------------------------------------------------------------------- 1.58 cvs 498: IsImageType 499: returns TRUE if type points to an image resource. 500: ----------------------------------------------------------------------*/ 501: #ifdef __STDC__ 1.67 cvs 502: ThotBool IsImageType (const STRING type) 1.58 cvs 503: #else /* __STDC__ */ 1.67 cvs 504: ThotBool IsImageType (type) 505: const STRING type; 1.58 cvs 506: #endif /* __STDC__ */ 507: { 1.67 cvs 508: CHAR_T temptype[MAX_LENGTH]; 1.58 cvs 509: int i; 510: 511: if (!type) 512: return (FALSE); 513: 1.67 cvs 514: ustrcpy (temptype, type); 1.58 cvs 515: /* Normalize the type */ 516: i = 0; 517: while (temptype[i] != EOS) 518: { 519: temptype[i] = tolower (temptype[i]); 520: i++; 521: } 1.67 cvs 522: if ((!ustrcmp (temptype, TEXT("gif"))) || (!ustrcmp (temptype, TEXT("x-xbitmap"))) || 523: (!ustrcmp (temptype, TEXT("x-xpixmap"))) || (!ustrcmp (temptype, TEXT("jpeg"))) || 524: (!ustrcmp (temptype, TEXT("png")))) 1.58 cvs 525: return (TRUE); 526: return (FALSE); 527: } 528: 529: /*---------------------------------------------------------------------- 1.9 cvs 530: IsTextName 1.4 cvs 531: ----------------------------------------------------------------------*/ 1.3 cvs 532: #ifdef __STDC__ 1.67 cvs 533: ThotBool IsTextName (const STRING path) 1.3 cvs 534: #else /* __STDC__ */ 1.67 cvs 535: ThotBool IsTextName (path) 536: const STRING path; 1.3 cvs 537: 538: #endif /* __STDC__ */ 539: { 1.67 cvs 540: CHAR_T temppath[MAX_LENGTH]; 541: CHAR_T suffix[MAX_LENGTH]; 542: CHAR_T nsuffix[MAX_LENGTH]; 1.5 cvs 543: int i; 544: 545: if (!path) 1.13 cvs 546: return (FALSE); 1.5 cvs 547: 1.67 cvs 548: ustrcpy (temppath, path); 1.5 cvs 549: ExtractSuffix (temppath, suffix); 550: 551: /* Normalize the suffix */ 552: i = 0; 1.39 cvs 553: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.5 cvs 554: { 1.25 cvs 555: nsuffix[i] = tolower (suffix[i]); 1.5 cvs 556: i++; 557: } 558: nsuffix[i] = EOS; 559: 1.67 cvs 560: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd")))) 1.13 cvs 561: return (TRUE); 1.67 cvs 562: else if (!ustrcmp (nsuffix, TEXT("gz"))) 1.13 cvs 563: { 1.39 cvs 564: /* take into account compressed files */ 1.13 cvs 565: ExtractSuffix (temppath, suffix); 566: /* Normalize the suffix */ 567: i = 0; 1.39 cvs 568: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 569: { 1.25 cvs 570: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 571: i++; 572: } 573: nsuffix[i] = EOS; 1.67 cvs 574: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd")))) 1.13 cvs 575: return (TRUE); 576: else 577: return (FALSE); 578: } 579: else 580: return (FALSE); 1.3 cvs 581: } 582: 1.4 cvs 583: /*---------------------------------------------------------------------- 1.9 cvs 584: IsHTTPPath 585: returns TRUE if path is in fact an http URL. 1.4 cvs 586: ----------------------------------------------------------------------*/ 1.3 cvs 587: #ifdef __STDC__ 1.67 cvs 588: ThotBool IsHTTPPath (const STRING path) 1.3 cvs 589: #else /* __STDC__ */ 1.67 cvs 590: ThotBool IsHTTPPath (path) 591: const STRING path; 1.3 cvs 592: #endif /* __STDC__ */ 593: { 1.5 cvs 594: if (!path) 595: return FALSE; 1.3 cvs 596: 1.67 cvs 597: if ((!ustrncmp (path, TEXT("http:"), 5) != 0) 598: || !ustrncmp (path, TEXT("internal:"), 9)) 1.58 cvs 599: return TRUE; 600: return FALSE; 1.3 cvs 601: } 602: 1.4 cvs 603: /*---------------------------------------------------------------------- 1.9 cvs 604: IsWithParameters 605: returns TRUE if url has a concatenated query string. 1.4 cvs 606: ----------------------------------------------------------------------*/ 1.3 cvs 607: #ifdef __STDC__ 1.66 cvs 608: ThotBool IsWithParameters (const char *url) 1.3 cvs 609: #else /* __STDC__ */ 1.66 cvs 610: ThotBool IsWithParameters (url) 1.34 cvs 611: const char *url; 1.3 cvs 612: #endif /* __STDC__ */ 613: { 1.5 cvs 614: int i; 1.3 cvs 615: 1.9 cvs 616: if ((!url) || (url[0] == EOS)) 1.5 cvs 617: return FALSE; 1.3 cvs 618: 1.9 cvs 619: i = strlen (url) - 1; 620: while (i > 0 && url[i--] != '?') 1.5 cvs 621: if (i < 0) 622: return FALSE; 1.3 cvs 623: 1.5 cvs 624: /* There is a parameter */ 625: return TRUE; 1.3 cvs 626: } 627: 1.4 cvs 628: /*---------------------------------------------------------------------- 1.9 cvs 629: IsW3Path 630: returns TRUE if path is in fact a URL. 1.4 cvs 631: ----------------------------------------------------------------------*/ 1.3 cvs 632: #ifdef __STDC__ 1.67 cvs 633: ThotBool IsW3Path (const STRING path) 1.3 cvs 634: #else /* __STDC__ */ 1.67 cvs 635: ThotBool IsW3Path (path) 636: const STRING path; 1.3 cvs 637: #endif /* __STDC__ */ 638: { 1.72 cvs 639: if (ustrncmp (path, TEXT("http:"), 5) && ustrncmp (path, TEXT("ftp:"), 4) && 640: ustrncmp (path, TEXT("telnet:"), 7) && ustrncmp (path, TEXT("wais:"), 5) && 641: ustrncmp (path, TEXT("news:"), 5) && ustrncmp (path, TEXT("gopher:"), 7) && 642: ustrncmp (path, TEXT("mailto:"), 7) && ustrncmp (path, TEXT("archie:"), 7)) 643: return FALSE; 644: return TRUE; 1.3 cvs 645: } 646: 1.4 cvs 647: /*---------------------------------------------------------------------- 1.9 cvs 648: IsValidProtocol 649: returns true if the url protocol is supported by Amaya. 1.4 cvs 650: ----------------------------------------------------------------------*/ 1.3 cvs 651: #ifdef __STDC__ 1.67 cvs 652: ThotBool IsValidProtocol (const STRING url) 1.3 cvs 653: #else /* __STDC__ */ 1.67 cvs 654: ThotBool IsValidProtocol (url) 655: const STRING url; 1.3 cvs 656: #endif /* __STDC__ */ 657: { 1.67 cvs 658: if (!ustrncmp (url, TEXT("http:"), 5) 1.69 cvs 659: || !ustrncmp (url, TEXT("internal:"), 9) 1.70 cvs 660: || !ustrncmp (url, TEXT("ftp:"), 4)) 1.22 cvs 661: /* experimental */ 1.58 cvs 662: /*** || !strncmp (url, "ftp:", 4) ***/ 1.24 cvs 663: /*** || !strncmp (path, "news:", 5)***/ 1.8 cvs 664: return (TRUE); 1.5 cvs 665: else 1.8 cvs 666: return (FALSE); 1.3 cvs 667: } 668: 1.31 cvs 669: 670: /*---------------------------------------------------------------------- 671: GetBaseURL 672: normalizes orgName according to a base associated with doc, and 673: following the standard URL format rules. 674: The function returns the base used to solve relative URL and SRC: 675: - the base of the document, 676: - or the document path (without document name). 677: ----------------------------------------------------------------------*/ 678: #ifdef __STDC__ 1.67 cvs 679: STRING GetBaseURL (Document doc) 1.31 cvs 680: #else /* __STDC__ */ 1.67 cvs 681: STRING GetBaseURL (doc) 1.31 cvs 682: Document doc; 683: #endif /* __STDC__ */ 684: { 685: Element el; 686: ElementType elType; 687: AttributeType attrType; 688: Attribute attr; 1.67 cvs 689: STRING ptr, basename; 1.31 cvs 690: int length; 691: 1.57 cvs 692: /* @@@ irene */ 693: if (!DocumentURLs[doc]) 694: return NULL; 1.67 cvs 695: basename = TtaAllocString (MAX_LENGTH); 696: ustrncpy (basename, DocumentURLs[doc], MAX_LENGTH-1); 1.39 cvs 697: basename[MAX_LENGTH-1] = EOS; 1.31 cvs 698: length = MAX_LENGTH -1; 699: /* get the root element */ 700: el = TtaGetMainRoot (doc); 701: /* search the BASE element */ 702: elType.ElSSchema = TtaGetDocumentSSchema (doc); 1.65 cvs 703: elType.ElTypeNum = HTML_EL_HEAD; 704: el = TtaSearchTypedElement (elType, SearchForward, el); 705: if (el) 706: { 707: elType.ElTypeNum = HTML_EL_BASE; 708: el = TtaSearchTypedElement (elType, SearchInTree, el); 709: } 1.31 cvs 710: if (el) 711: { 712: /* The document has a BASE element -> Get the HREF attribute */ 713: attrType.AttrSSchema = elType.ElSSchema; 714: attrType.AttrTypeNum = HTML_ATTR_HREF_; 715: attr = TtaGetAttribute (el, attrType); 716: if (attr) 717: { 718: /* Use the base path of the document */ 719: TtaGiveTextAttributeValue (attr, basename, &length); 720: /* base and orgName have to be separated by a DIR_SEP */ 721: length--; 1.43 cvs 722: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP) 1.31 cvs 723: /* verify if the base has the form "protocol://server:port" */ 724: { 1.67 cvs 725: ptr = AmayaParseUrl (basename, _EMPTYSTR_, AMAYA_PARSE_ACCESS | 1.33 cvs 726: AMAYA_PARSE_HOST | 727: AMAYA_PARSE_PUNCTUATION); 1.67 cvs 728: if (ptr && !ustrcmp (ptr, basename)) 1.31 cvs 729: { 1.43 cvs 730: /* it has this form, we complete it by adding a URL_STR */ 1.67 cvs 731: if (ustrchr (basename, DIR_SEP)) 732: ustrcat (basename, DIR_STR); 1.43 cvs 733: else 1.67 cvs 734: ustrcat (basename, URL_STR); 1.31 cvs 735: length++; 736: } 737: if (ptr) 738: TtaFreeMemory (ptr); 739: } 740: } 1.33 cvs 741: } 742: 1.31 cvs 743: /* Remove anything after the last DIR_SEP char. If no such char is found, 744: * then search for the first ":" char, hoping that what's before that is a 745: * protocol. If found, end the string there. If neither char is found, 746: * then discard the whole base element. 747: */ 1.67 cvs 748: length = ustrlen (basename) - 1; 1.31 cvs 749: /* search for the last DIR_SEP char */ 1.43 cvs 750: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP) 1.31 cvs 751: length--; 752: if (length >= 0) 753: /* found the last DIR_SEP char, end the string there */ 754: basename[length + 1] = EOS; 755: else 756: /* search for the first PATH_STR char */ 757: { 1.67 cvs 758: for (length = 0; basename[length] != TEXT(':') && 1.31 cvs 759: basename[length] != EOS; length ++); 1.67 cvs 760: if (basename[length] == TEXT(':')) 1.31 cvs 761: /* found, so end the string there */ 762: basename[length + 1] = EOS; 763: else 764: /* not found, discard the base */ 765: basename[0] = EOS; 766: } 767: return (basename); 768: } 769: 770: 1.4 cvs 771: /*---------------------------------------------------------------------- 1.40 cvs 772: GetLocalPath 773: Allocate and return the local document path associated to the url 774: ----------------------------------------------------------------------*/ 775: #ifdef __STDC__ 1.67 cvs 776: STRING GetLocalPath (Document doc, STRING url) 1.40 cvs 777: #else /* __STDC__ */ 1.67 cvs 778: STRING GetLocalPath (doc, url) 1.40 cvs 779: Document doc; 1.67 cvs 780: STRING url; 1.40 cvs 781: #endif /* __STDC__ */ 782: { 1.67 cvs 783: STRING ptr, n; 784: STRING documentname; 785: CHAR_T url_sep; 1.40 cvs 786: int len; 1.67 cvs 787: ThotBool noFile; 1.40 cvs 788: 789: if (url != NULL) 790: { 791: /* check whether the file name exists */ 1.67 cvs 792: len = ustrlen (url) - 1; 1.71 cvs 793: if (IsW3Path (url)) 1.67 cvs 794: url_sep = TEXT('/'); 1.41 cvs 795: else 796: url_sep = DIR_SEP; 797: noFile = (url[len] == url_sep); 1.40 cvs 798: if (noFile) 799: url[len] = EOS; 1.78 ! cvs 800: ptr = TtaAllocString (MAX_LENGTH); ! 801: documentname = TtaAllocString (MAX_LENGTH); ! 802: TtaExtractName (url, ptr, documentname); 1.67 cvs 803: usprintf (ptr, TEXT("%s%s%d%s"), TempFileDirectory, DIR_STR, doc, DIR_STR); 1.40 cvs 804: if (!TtaCheckDirectory (ptr)) 805: /* directory did not exist */ 1.72 cvs 806: TtaMakeDirectory (ptr); 1.47 cvs 807: 808: /* don't include the query string within document name */ 1.67 cvs 809: n = ustrrchr(documentname, TEXT('?')); 1.47 cvs 810: if (n != NULL) 811: *n = EOS; 1.46 cvs 812: /* don't include ':' within document name */ 1.67 cvs 813: n = ustrchr (documentname, TEXT(':')); 1.46 cvs 814: if (n != NULL) 815: *n = EOS; 1.69 cvs 816: /* if after all this operations document name 817: is empty, let's use noname.html instead */ 818: if (documentname[0] == EOS) 819: ustrcat (ptr, TEXT("noname.html")); 820: else 821: ustrcat (ptr, documentname); 1.40 cvs 822: TtaFreeMemory (documentname); 823: /* restore the url */ 824: if (noFile) 1.41 cvs 825: url[len] = url_sep; 1.40 cvs 826: return (ptr); 827: } 828: else 829: return (NULL); 830: } 831: 1.73 cvs 832: /*---------------------------------------------------------------------- 833: ConvertFileURL 834: If the URL starts with file: prefix, it removes the protocol so that we 835: can use it as a local filename 836: ----------------------------------------------------------------------*/ 837: #ifdef __STDC__ 838: void ConvertFileURL (STRING url) 839: #else 840: void ConvertFileURL (url) 841: STRING url 842: #endif /* __STDC__ */ 843: { 844: if (!ustrncasecmp (url, TEXT("file:"), 5)) 845: ustrcpy (url, url + 5); 846: } 1.40 cvs 847: 848: /*---------------------------------------------------------------------- 1.9 cvs 849: NormalizeURL 850: normalizes orgName according to a base associated with doc, and 851: following the standard URL format rules. 1.53 cvs 852: if doc is 0 and otherPath not NULL, normalizes orgName according to this 853: other path. 1.9 cvs 854: The function returns the new complete and normalized URL 1.12 cvs 855: or file name path (newName) and the name of the document (docName). 1.9 cvs 856: N.B. If the function can't find out what's the docName, it assigns 857: the name "noname.html". 1.4 cvs 858: ----------------------------------------------------------------------*/ 1.3 cvs 859: #ifdef __STDC__ 1.67 cvs 860: void NormalizeURL (STRING orgName, Document doc, STRING newName, STRING docName, STRING otherPath) 1.3 cvs 861: #else /* __STDC__ */ 1.53 cvs 862: void NormalizeURL (orgName, doc, newName, docName, otherPath) 1.67 cvs 863: STRING orgName; 1.3 cvs 864: Document doc; 1.67 cvs 865: STRING newName; 866: STRING docName; 867: STRING otherPath; 1.3 cvs 868: #endif /* __STDC__ */ 869: { 1.67 cvs 870: STRING basename; 871: CHAR_T tempOrgName[MAX_LENGTH]; 872: STRING ptr; 873: CHAR_T used_sep; 874: int length; 1.66 cvs 875: ThotBool check; 1.5 cvs 876: 1.44 cvs 877: # ifdef _WINDOWS 878: int ndx; 879: # endif /* _WINDOWS */ 880: 1.5 cvs 881: if (!newName || !docName) 882: return; 1.18 cvs 883: 1.32 cvs 884: if (doc != 0) 1.53 cvs 885: basename = GetBaseURL (doc); 886: else if (otherPath != NULL) 887: basename = TtaStrdup (otherPath); 1.32 cvs 888: else 1.53 cvs 889: basename = NULL; 1.32 cvs 890: 1.18 cvs 891: /* 1.31 cvs 892: * Clean orgName 893: * Make sure we have a complete orgName, without any leading or trailing 894: * white spaces, or trailinbg new lines 895: */ 1.5 cvs 896: ptr = orgName; 1.18 cvs 897: /* skip leading white space and new line characters */ 1.67 cvs 898: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS); 899: ustrncpy (tempOrgName, ptr, MAX_LENGTH -1); 1.39 cvs 900: tempOrgName[MAX_LENGTH -1] = EOS; 1.18 cvs 901: /* 1.31 cvs 902: * Make orgName a complete URL 903: * If the URL does not include a protocol, then try to calculate 904: * one using the doc's base element (if it exists), 905: */ 1.53 cvs 906: if (tempOrgName[0] == EOS) 907: { 908: newName[0] = EOS; 909: TtaFreeMemory (basename); 910: return; 911: } 1.49 cvs 912: 913: /* clean trailing white space */ 1.67 cvs 914: length = ustrlen (tempOrgName) - 1; 1.53 cvs 915: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL) 916: { 917: tempOrgName[length] = EOS; 918: length--; 919: } 1.50 cvs 920: 1.55 cvs 921: /* remove extra dot (which dot???) */ 922: /* ugly, but faster than a strcmp */ 1.67 cvs 923: if (tempOrgName[length] == TEXT('.') 924: && (length == 0 || tempOrgName[length-1] != TEXT('.'))) 1.55 cvs 925: tempOrgName[length] = EOS; 1.50 cvs 926: 1.53 cvs 927: if (IsW3Path (tempOrgName)) 928: { 929: /* the name is complete, go to the Sixth Step */ 1.67 cvs 930: ustrcpy (newName, tempOrgName); 1.53 cvs 931: SimplifyUrl (&newName); 932: /* verify if the URL has the form "protocol://server:port" */ 1.67 cvs 933: ptr = AmayaParseUrl (newName, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 934: if (ptr && !ustrcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */ 935: ustrcat (newName, URL_STR); 1.49 cvs 936: 1.53 cvs 937: if (ptr) 1.50 cvs 938: TtaFreeMemory (ptr); 1.53 cvs 939: } 940: else if ( basename == NULL) 941: /* the name is complete, go to the Sixth Step */ 1.67 cvs 942: ustrcpy (newName, tempOrgName); 1.53 cvs 943: else 944: { 1.31 cvs 945: /* Calculate the absolute URL, using the base or document URL */ 1.44 cvs 946: # ifdef _WINDOWS 1.53 cvs 947: if (!IsW3Path (basename)) 948: { 1.67 cvs 949: length = ustrlen (tempOrgName); 1.53 cvs 950: for (ndx = 0; ndx < length; ndx++) 1.67 cvs 951: if (tempOrgName [ndx] == TEXT('/')) 952: tempOrgName [ndx] = TEXT('\\'); 1.53 cvs 953: } 1.44 cvs 954: # endif /* _WINDOWS */ 1.25 cvs 955: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL); 1.53 cvs 956: if (ptr) 957: { 958: SimplifyUrl (&ptr); 1.67 cvs 959: ustrcpy (newName, ptr); 1.53 cvs 960: TtaFreeMemory (ptr); 961: } 962: else 963: newName[0] = EOS; 964: } 1.36 cvs 965: 966: TtaFreeMemory (basename); 1.18 cvs 967: /* 1.31 cvs 968: * Prepare the docname that will refer to this ressource in the 969: * .amaya directory. If the new URL finishes on DIR_SEP, then use 970: * noname.html as a default ressource name 1.18 cvs 971: */ 1.53 cvs 972: if (newName[0] != EOS) 973: { 1.67 cvs 974: length = ustrlen (newName) - 1; 1.53 cvs 975: if (newName[length] == URL_SEP || newName[length] == DIR_SEP) 976: { 977: used_sep = newName[length]; 978: check = TRUE; 979: while (check) 980: { 1.50 cvs 981: length--; 982: while (length >= 0 && newName[length] != used_sep) 1.53 cvs 983: length--; 1.67 cvs 984: if (!ustrncmp (&newName[length+1], TEXT(".."), 2)) 1.53 cvs 985: { 986: newName[length+1] = EOS; 987: /* remove also previous directory */ 988: length--; 989: while (length >= 0 && newName[length] != used_sep) 990: length--; 1.67 cvs 991: if (ustrncmp (&newName[length+1], TEXT("//"), 2)) 1.53 cvs 992: /* don't remove server name */ 1.50 cvs 993: newName[length+1] = EOS; 1.53 cvs 994: } 1.67 cvs 995: else if (!ustrncmp (&newName[length+1], TEXT("."), 1)) 1.53 cvs 996: newName[length+1] = EOS; 1.50 cvs 997: else 1.53 cvs 998: check = FALSE; 999: } 1.67 cvs 1000: ustrcpy (docName, TEXT("noname.html")); 1.53 cvs 1001: /* docname was not comprised inside the URL, so let's */ 1002: /* assign the default ressource name */ 1.67 cvs 1003: ustrcpy (docName, TEXT("noname.html")); 1.53 cvs 1004: } 1005: else 1006: { /* docname is comprised inside the URL */ 1.50 cvs 1007: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP) 1.53 cvs 1008: length--; 1009: if (length < 0) 1.67 cvs 1010: ustrcpy (docName, newName); 1.53 cvs 1011: else 1.67 cvs 1012: ustrcpy (docName, &newName[length+1]); 1.53 cvs 1013: } 1014: } 1015: else 1016: docName[0] = EOS; 1.18 cvs 1017: } 1.3 cvs 1018: 1.4 cvs 1019: /*---------------------------------------------------------------------- 1.9 cvs 1020: IsSameHost 1.4 cvs 1021: ----------------------------------------------------------------------*/ 1.3 cvs 1022: #ifdef __STDC__ 1.67 cvs 1023: ThotBool IsSameHost (const STRING url1, const STRING url2) 1.3 cvs 1024: #else /* __STDC__ */ 1.67 cvs 1025: ThotBool IsSameHost (url1, url2) 1026: const STRING url1; 1027: const STRING url2; 1.3 cvs 1028: #endif /* __STDC__ */ 1029: { 1.67 cvs 1030: STRING basename_ptr1, basename_ptr2; 1031: ThotBool result; 1.3 cvs 1032: 1.67 cvs 1033: basename_ptr1 = AmayaParseUrl (url1, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 1034: basename_ptr2 = AmayaParseUrl (url2, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 1.3 cvs 1035: 1.67 cvs 1036: if (ustrcmp (basename_ptr1, basename_ptr2)) 1.8 cvs 1037: result = FALSE; 1.5 cvs 1038: else 1.8 cvs 1039: result = TRUE; 1.3 cvs 1040: 1.25 cvs 1041: TtaFreeMemory (basename_ptr1); 1042: TtaFreeMemory (basename_ptr2); 1.5 cvs 1043: return (result); 1.3 cvs 1044: } 1045: 1046: 1.4 cvs 1047: /*---------------------------------------------------------------------- 1.22 cvs 1048: HasKnownFileSuffix 1049: returns TRUE if path points to a file ending with a suffix. 1050: ----------------------------------------------------------------------*/ 1051: #ifdef __STDC__ 1.67 cvs 1052: ThotBool HasKnownFileSuffix (const STRING path) 1.22 cvs 1053: #else /* __STDC__ */ 1.67 cvs 1054: ThotBool HasKnownFileSuffix (path) 1055: const STRING path; 1.22 cvs 1056: #endif /* __STDC__ */ 1057: { 1.67 cvs 1058: STRING root; 1059: CHAR_T temppath[MAX_LENGTH]; 1060: CHAR_T suffix[MAX_LENGTH]; 1.22 cvs 1061: 1.67 cvs 1062: if (!path || path[0] == EOS || path[ustrlen(path)] == DIR_SEP) 1.22 cvs 1063: return (FALSE); 1064: 1.67 cvs 1065: root = AmayaParseUrl(path, _EMPTYSTR_, AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION); 1.22 cvs 1066: 1067: if (root) 1068: { 1.67 cvs 1069: ustrcpy (temppath, root); 1.25 cvs 1070: TtaFreeMemory (root); 1.22 cvs 1071: /* Get the suffix */ 1072: ExtractSuffix (temppath, suffix); 1073: 1074: if( suffix[0] == EOS) 1075: /* no suffix */ 1076: return (FALSE); 1077: 1078: /* Normalize the suffix */ 1079: ConvertToLowerCase (suffix); 1080: 1.67 cvs 1081: if (!ustrcmp (suffix, TEXT("gz"))) 1.22 cvs 1082: /* skip the compressed suffix */ 1083: { 1084: ExtractSuffix (temppath, suffix); 1085: if(suffix[0] == EOS) 1086: /* no suffix */ 1087: return (FALSE); 1088: /* Normalize the suffix */ 1089: ConvertToLowerCase (suffix); 1090: } 1091: 1.67 cvs 1092: if (ustrcmp (suffix, TEXT("gif")) && 1093: ustrcmp (suffix, TEXT("xbm")) && 1094: ustrcmp (suffix, TEXT("xpm")) && 1095: ustrcmp (suffix, TEXT("jpg")) && 1096: ustrcmp (suffix, TEXT("pdf")) && 1097: ustrcmp (suffix, TEXT("png")) && 1098: ustrcmp (suffix, TEXT("tgz")) && 1099: ustrcmp (suffix, TEXT("xpg")) && 1100: ustrcmp (suffix, TEXT("xpd")) && 1101: ustrcmp (suffix, TEXT("ps")) && 1102: ustrcmp (suffix, TEXT("au")) && 1103: ustrcmp (suffix, TEXT("html")) && 1104: ustrcmp (suffix, TEXT("htm")) && 1105: ustrcmp (suffix, TEXT("shtml")) && 1106: ustrcmp (suffix, TEXT("xht")) && 1107: ustrcmp (suffix, TEXT("xhtm")) && 1108: ustrcmp (suffix, TEXT("xhtml")) && 1109: ustrcmp (suffix, TEXT("txt")) && 1110: ustrcmp (suffix, TEXT("css")) && 1111: ustrcmp (suffix, TEXT("eps"))) 1.22 cvs 1112: return (FALSE); 1113: else 1114: return (TRUE); 1115: } 1116: else 1117: return (FALSE); 1118: } 1119: 1120: 1121: /*---------------------------------------------------------------------- 1.24 cvs 1122: ChopURL 1123: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL). 1124: If inputURL is bigger than that size, outputURL receives 1125: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...", 1126: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL. 1127: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets 1128: copied into outputURL. 1129: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH 1130: chars. 1131: ----------------------------------------------------------------------*/ 1132: #ifdef __STDC__ 1.34 cvs 1133: void ChopURL (char *outputURL, const char *inputURL) 1.24 cvs 1134: #else 1135: void ChopURL (outputURL, inputURL) 1.34 cvs 1136: char *outputURL; 1137: const char *inputURL; 1.24 cvs 1138: #endif 1.22 cvs 1139: 1.24 cvs 1140: { 1141: int len; 1.9 cvs 1142: 1.24 cvs 1143: len = strlen (inputURL); 1144: if (len <= MAX_PRINT_URL_LENGTH) 1.29 cvs 1145: strcpy (outputURL, inputURL); 1.24 cvs 1146: else 1147: /* make a truncated urlName on the status window */ 1148: { 1149: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2); 1150: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS; 1151: strcat (outputURL, "..."); 1152: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ])); 1153: } 1.25 cvs 1154: } 1155: 1156: 1157: /*---------------------------------------------------------------------- 1158: scan 1.47 cvs 1159: Scan a filename for its constituents 1.25 cvs 1160: ----------------------------------- 1161: 1162: On entry, 1163: name points to a document name which may be incomplete. 1164: On exit, 1165: absolute or relative may be nonzero (but not both). 1166: host, fragment and access may be nonzero if they were specified. 1167: Any which are nonzero point to zero terminated strings. 1168: ----------------------------------------------------------------------*/ 1169: #ifdef __STDC__ 1.67 cvs 1170: static void scan (STRING name, HTURI * parts) 1.25 cvs 1171: #else /* __STDC__ */ 1172: static void scan (name, parts) 1.67 cvs 1173: STRING name; 1174: HTURI *parts; 1.25 cvs 1175: 1176: #endif /* __STDC__ */ 1177: { 1.67 cvs 1178: STRING p; 1179: STRING after_access = name; 1.32 cvs 1180: 1.43 cvs 1181: memset (parts, '\0', sizeof (HTURI)); 1.28 cvs 1182: /* Look for fragment identifier */ 1.67 cvs 1183: if ((p = ustrchr(name, TEXT('#'))) != NULL) 1.28 cvs 1184: { 1.67 cvs 1185: *p++ = TEXT('\0'); 1.28 cvs 1186: parts->fragment = p; 1.25 cvs 1187: } 1188: 1.28 cvs 1189: for (p=name; *p; p++) 1190: { 1.67 cvs 1191: if (*p == URL_SEP || *p == DIR_SEP || *p == TEXT('#') || *p == TEXT('?')) 1.28 cvs 1192: break; 1.67 cvs 1193: if (*p == TEXT(':')) 1.28 cvs 1194: { 1195: *p = 0; 1196: parts->access = after_access; /* Scheme has been specified */ 1197: 1198: /* The combination of gcc, the "-O" flag and the HP platform is 1199: unhealthy. The following three lines is a quick & dirty fix, but is 1200: not recommended. Rather, turn off "-O". */ 1201: 1202: /* after_access = p;*/ 1203: /* while (*after_access == 0)*/ 1204: /* after_access++;*/ 1205: after_access = p+1; 1.67 cvs 1206: if (!ustrcasecmp(TEXT("URL"), parts->access)) 1.28 cvs 1207: /* Ignore IETF's URL: pre-prefix */ 1208: parts->access = NULL; 1209: else 1.25 cvs 1210: break; 1211: } 1212: } 1213: 1214: p = after_access; 1.43 cvs 1215: if (*p == URL_SEP || *p == DIR_SEP) 1.28 cvs 1216: { 1.43 cvs 1217: if (p[1] == URL_SEP) 1.28 cvs 1218: { 1.25 cvs 1219: parts->host = p+2; /* host has been specified */ 1.28 cvs 1220: *p = 0; /* Terminate access */ 1221: /* look for end of host name if any */ 1.67 cvs 1222: p = ustrchr (parts->host, URL_SEP); 1.28 cvs 1223: if (p) 1224: { 1.43 cvs 1225: *p = EOS; /* Terminate host */ 1.25 cvs 1226: parts->absolute = p+1; /* Root has been found */ 1.28 cvs 1227: } 1228: } 1229: else 1230: /* Root found but no host */ 1231: parts->absolute = p+1; 1232: } 1233: else 1234: { 1.25 cvs 1235: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */ 1.28 cvs 1236: } 1.25 cvs 1237: } 1238: 1239: 1240: /*---------------------------------------------------------------------- 1.28 cvs 1241: AmayaParseUrl: parse a Name relative to another name 1242: 1243: This returns those parts of a name which are given (and requested) 1244: substituting bits from the related name where necessary. 1.25 cvs 1245: 1.28 cvs 1246: On entry, 1.25 cvs 1247: aName A filename given 1248: relatedName A name relative to which aName is to be parsed. Give 1249: it an empty string if aName is absolute. 1250: wanted A mask for the bits which are wanted. 1251: 1.28 cvs 1252: On exit, 1.25 cvs 1253: returns A pointer to a malloc'd string which MUST BE FREED 1254: ----------------------------------------------------------------------*/ 1255: #ifdef __STDC__ 1.67 cvs 1256: STRING AmayaParseUrl (const STRING aName, STRING relatedName, int wanted) 1.25 cvs 1257: #else /* __STDC__ */ 1.67 cvs 1258: STRING AmayaParseUrl (aName, relatedName, wanted) 1259: const STRING aName; 1260: STRING relatedName; 1.28 cvs 1261: int wanted; 1.25 cvs 1262: 1263: #endif /* __STDC__ */ 1264: { 1.67 cvs 1265: STRING return_value; 1266: CHAR_T result[MAX_LENGTH]; 1267: CHAR_T name[MAX_LENGTH]; 1268: CHAR_T rel[MAX_LENGTH]; 1269: STRING p, access; 1.29 cvs 1270: HTURI given, related; 1271: int len; 1.67 cvs 1272: CHAR_T used_sep; 1273: STRING used_str; 1.32 cvs 1274: 1.67 cvs 1275: if (ustrchr (aName, DIR_SEP) || ustrchr (relatedName, DIR_SEP)) 1.33 cvs 1276: { 1.42 cvs 1277: used_str = DIR_STR; 1278: used_sep = DIR_SEP; 1.33 cvs 1279: } 1.32 cvs 1280: else 1.33 cvs 1281: { 1.42 cvs 1282: used_str = URL_STR; 1283: used_sep = URL_SEP; 1.33 cvs 1284: } 1.32 cvs 1285: 1.29 cvs 1286: /* Make working copies of input strings to cut up: */ 1287: return_value = NULL; 1288: result[0] = 0; /* Clear string */ 1.67 cvs 1289: ustrcpy (name, aName); 1.29 cvs 1290: if (relatedName != NULL) 1.67 cvs 1291: ustrcpy (rel, relatedName); 1.29 cvs 1292: else 1293: relatedName[0] = EOS; 1294: 1295: scan (name, &given); 1296: scan (rel, &related); 1297: access = given.access ? given.access : related.access; 1298: if (wanted & AMAYA_PARSE_ACCESS) 1299: if (access) 1300: { 1.67 cvs 1301: ustrcat (result, access); 1.29 cvs 1302: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1303: ustrcat (result, TEXT(":")); 1.29 cvs 1304: } 1305: 1306: if (given.access && related.access) 1307: /* If different, inherit nothing. */ 1.67 cvs 1308: if (ustrcmp (given.access, related.access) != 0) 1.29 cvs 1309: { 1310: related.host = 0; 1311: related.absolute = 0; 1312: related.relative = 0; 1313: related.fragment = 0; 1314: } 1315: 1316: if (wanted & AMAYA_PARSE_HOST) 1317: if(given.host || related.host) 1318: { 1319: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1320: ustrcat (result, TEXT("//")); 1321: ustrcat (result, given.host ? given.host : related.host); 1.29 cvs 1322: } 1323: 1324: if (given.host && related.host) 1325: /* If different hosts, inherit no path. */ 1.67 cvs 1326: if (ustrcmp (given.host, related.host) != 0) 1.29 cvs 1327: { 1328: related.absolute = 0; 1329: related.relative = 0; 1330: related.fragment = 0; 1331: } 1332: 1333: if (wanted & AMAYA_PARSE_PATH) 1334: { 1335: if (given.absolute) 1336: { 1337: /* All is given */ 1338: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1339: ustrcat (result, used_str); 1340: ustrcat (result, given.absolute); 1.25 cvs 1341: } 1.29 cvs 1342: else if (related.absolute) 1343: { 1344: /* Adopt path not name */ 1.67 cvs 1345: ustrcat (result, used_str); 1346: ustrcat (result, related.absolute); 1.29 cvs 1347: if (given.relative) 1348: { 1349: /* Search part? */ 1.67 cvs 1350: p = ustrchr (result, TEXT('?')); 1.29 cvs 1351: if (!p) 1.67 cvs 1352: p=result+ustrlen(result)-1; 1.33 cvs 1353: for (; *p!=used_sep; p--); /* last / */ 1.29 cvs 1354: /* Remove filename */ 1355: p[1]=0; 1356: /* Add given one */ 1.67 cvs 1357: ustrcat (result, given.relative); 1.25 cvs 1358: } 1359: } 1.29 cvs 1360: else if (given.relative) 1361: /* what we've got */ 1.67 cvs 1362: ustrcat (result, given.relative); 1.29 cvs 1363: else if (related.relative) 1.67 cvs 1364: ustrcat (result, related.relative); 1.29 cvs 1365: else 1366: /* No inheritance */ 1.67 cvs 1367: ustrcat (result, used_str); 1.25 cvs 1368: } 1.29 cvs 1369: 1370: if (wanted & AMAYA_PARSE_ANCHOR) 1371: if (given.fragment || related.fragment) 1372: { 1373: if (given.absolute && given.fragment) 1374: { 1375: /*Fixes for relURLs...*/ 1376: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1377: ustrcat (result, TEXT("#")); 1378: ustrcat (result, given.fragment); 1.29 cvs 1379: } 1380: else if (!(given.absolute) && !(given.fragment)) 1.67 cvs 1381: ustrcat (result, _EMPTYSTR_); 1.29 cvs 1382: else 1383: { 1384: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1385: ustrcat (result, TEXT("#")); 1386: ustrcat (result, given.fragment ? given.fragment : related.fragment); 1.29 cvs 1387: } 1388: } 1.67 cvs 1389: len = ustrlen (result); 1390: if ((return_value = TtaAllocString (len + 1)) != NULL) 1391: ustrcpy (return_value, result); 1.29 cvs 1392: return (return_value); /* exactly the right length */ 1.25 cvs 1393: } 1394: 1395: /*---------------------------------------------------------------------- 1396: HTCanon 1397: Canonicalizes the URL in the following manner starting from the host 1398: pointer: 1399: 1400: 1) The host name is converted to lowercase 1401: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp) 1402: 1403: Return: OK The position of the current path part of the URL 1404: which might be the old one or a new one. 1405: 1406: ----------------------------------------------------------------------*/ 1407: #ifdef __STDC__ 1.67 cvs 1408: static STRING HTCanon (STRING* filename, STRING host) 1.25 cvs 1409: #else /* __STDC__ */ 1.67 cvs 1410: static STRING HTCanon (filename, host) 1411: STRING *filename; 1412: STRING host; 1.25 cvs 1413: #endif /* __STDC__ */ 1414: { 1.67 cvs 1415: STRING newname = NULL; 1416: STRING port; 1417: STRING strptr; 1418: STRING path; 1419: STRING access = host-3; 1420: CHAR_T used_sep; 1.32 cvs 1421: 1422: 1.67 cvs 1423: if (*filename && ustrchr (*filename, URL_SEP)) 1.33 cvs 1424: { 1425: used_sep = URL_SEP; 1426: } 1427: else 1428: { 1429: used_sep = DIR_SEP; 1430: } 1.32 cvs 1431: 1.33 cvs 1432: while (access>*filename && *(access-1)!= used_sep) /* Find access method */ 1.25 cvs 1433: access--; 1.67 cvs 1434: if ((path = ustrchr(host, used_sep)) == NULL) /* Find path */ 1435: path = host + ustrlen(host); 1436: if ((strptr = ustrchr(host, TEXT('@'))) != NULL && strptr<path) /* UserId */ 1.25 cvs 1437: host = strptr; 1.67 cvs 1438: if ((port = ustrchr(host, TEXT(':'))) != NULL && port>path) /* Port number */ 1.25 cvs 1439: port = NULL; 1440: 1441: strptr = host; /* Convert to lower-case */ 1.33 cvs 1442: while (strptr<path) 1443: { 1.25 cvs 1444: *strptr = tolower(*strptr); 1445: strptr++; 1.33 cvs 1446: } 1.25 cvs 1447: 1448: /* Does the URL contain a full domain name? This also works for a 1449: numerical host name. The domain name is already made lower-case 1450: and without a trailing dot. */ 1451: { 1.67 cvs 1452: STRING dot = port ? port : path; 1453: if (dot > *filename && *--dot == TEXT('.')) 1.33 cvs 1454: { 1.67 cvs 1455: STRING orig=dot, dest=dot+1; 1.33 cvs 1456: while((*orig++ = *dest++)); 1457: if (port) port--; 1458: path--; 1.25 cvs 1459: } 1460: } 1461: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */ 1.33 cvs 1462: if (port) 1463: { 1464: if (!*(port+1) || *(port+1)==used_sep) 1465: { 1466: if (!newname) 1467: { 1.67 cvs 1468: STRING orig=port, dest=port+1; 1.25 cvs 1469: while((*orig++ = *dest++)); 1.33 cvs 1470: } 1471: } 1.67 cvs 1472: else if ((!ustrncmp(access, TEXT("http"), 4) && 1473: (*(port+1) == TEXT('8') && *(port+2) == TEXT('0') && (*(port+3) == used_sep || !*(port + 3)))) || 1474: (!ustrncmp (access, TEXT("gopher"), 6) && 1475: (*(port+1) == TEXT('7') && *(port+2) == TEXT('0') && (*(port+3) == used_sep || !*(port+3)))) || 1476: (!ustrncmp(access, TEXT("ftp"), 3) && 1477: (*(port+1) == TEXT('2') && *(port + 2) == TEXT('1') && (*(port+3) == used_sep || !*(port+3))))) { 1.33 cvs 1478: if (!newname) 1479: { 1.67 cvs 1480: STRING orig=port, dest=port+3; 1.33 cvs 1481: while((*orig++ = *dest++)); 1482: /* Update path position, Henry Minsky */ 1483: path -= 3; 1.25 cvs 1484: } 1.33 cvs 1485: } 1486: else if (newname) 1.67 cvs 1487: ustrncat(newname, port, (int) (path-port)); 1.33 cvs 1488: } 1.25 cvs 1489: 1.33 cvs 1490: if (newname) 1491: { 1.67 cvs 1492: STRING newpath = newname + ustrlen (newname); 1493: ustrcat(newname, path); 1.25 cvs 1494: path = newpath; 1.28 cvs 1495: /* Free old copy */ 1496: TtaFreeMemory(*filename); 1.25 cvs 1497: *filename = newname; 1.33 cvs 1498: } 1.25 cvs 1499: return path; 1500: } 1501: 1502: 1503: /*---------------------------------------------------------------------- 1.29 cvs 1504: SimplifyUrl: simplify a URI 1.32 cvs 1505: A URI is allowed to contain the sequence xxx/../ which may be 1506: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR. 1.28 cvs 1507: Simplification helps us recognize duplicate URIs. 1.25 cvs 1508: 1.28 cvs 1509: Thus, /etc/junk/../fred becomes /etc/fred 1510: /etc/junk/./fred becomes /etc/junk/fred 1.25 cvs 1511: 1.28 cvs 1512: but we should NOT change 1513: http://fred.xxx.edu/../.. 1.25 cvs 1514: 1515: or ../../albert.html 1516: 1.28 cvs 1517: In order to avoid empty URLs the following URLs become: 1.25 cvs 1518: 1519: /fred/.. becomes /fred/.. 1520: /fred/././.. becomes /fred/.. 1521: /fred/.././junk/.././ becomes /fred/.. 1522: 1.28 cvs 1523: If more than one set of `://' is found (several proxies in cascade) then 1524: only the part after the last `://' is simplified. 1.25 cvs 1525: 1.28 cvs 1526: Returns: A string which might be the old one or a new one. 1.25 cvs 1527: ----------------------------------------------------------------------*/ 1528: #ifdef __STDC__ 1.67 cvs 1529: void SimplifyUrl (STRING* url) 1.25 cvs 1530: #else /* __STDC__ */ 1.29 cvs 1531: void SimplifyUrl (url) 1.67 cvs 1532: STRING *url; 1.25 cvs 1533: #endif /* __STDC__ */ 1534: { 1.67 cvs 1535: STRING path, p; 1536: STRING newptr, access; 1537: STRING orig, dest, end; 1.28 cvs 1538: 1.67 cvs 1539: CHAR_T used_sep; 1.77 cvs 1540: ThotBool ddot_simplify; /* used to desactivate the double dot simplifcation: 1541: something/../ simplification in relative URLs when they start with a ../ */ 1.32 cvs 1542: 1543: 1.28 cvs 1544: if (!url || !*url) 1545: return; 1546: 1.67 cvs 1547: if (ustrchr (*url, URL_SEP)) 1.33 cvs 1548: { 1549: used_sep = URL_SEP; 1550: } 1.32 cvs 1551: else 1.33 cvs 1552: { 1553: used_sep = DIR_SEP; 1554: } 1.32 cvs 1555: 1.77 cvs 1556: /* should we simplify double dot? */ 1557: path = *url; 1558: if (*path == TEXT('.') && *(path + 1) == TEXT('.')) 1559: ddot_simplify = FALSE; 1560: else 1561: ddot_simplify = TRUE; 1562: 1.28 cvs 1563: /* Find any scheme name */ 1.67 cvs 1564: if ((path = ustrstr(*url, TEXT("://"))) != NULL) 1.33 cvs 1565: { 1566: /* Find host name */ 1.28 cvs 1567: access = *url; 1568: while (access<path && (*access=tolower(*access))) 1569: access++; 1570: path += 3; 1.67 cvs 1571: while ((newptr = ustrstr(path, TEXT("://"))) != NULL) 1.28 cvs 1572: /* For proxies */ 1573: path = newptr+3; 1574: /* We have a host name */ 1575: path = HTCanon(url, path); 1.25 cvs 1576: } 1.67 cvs 1577: else if ((path = ustrstr(*url, TEXT(":/"))) != NULL) 1.28 cvs 1578: path += 2; 1579: else 1580: path = *url; 1.25 cvs 1581: 1.33 cvs 1582: if (*path == used_sep && *(path+1)==used_sep) 1.28 cvs 1583: /* Some URLs start //<foo> */ 1584: path += 1; 1.67 cvs 1585: else if (!ustrncmp(path, TEXT("news:"), 5)) 1.28 cvs 1586: { 1.67 cvs 1587: newptr = ustrchr(path+5, TEXT('@')); 1.28 cvs 1588: if (!newptr) 1589: newptr = path + 5; 1590: while (*newptr) 1591: { 1592: /* Make group or host lower case */ 1593: *newptr = tolower (*newptr); 1594: newptr++; 1.25 cvs 1595: } 1.28 cvs 1596: /* Doesn't need to do any more */ 1597: return; 1.25 cvs 1598: } 1.28 cvs 1599: 1600: if ((p = path)) 1601: { 1.67 cvs 1602: if (!((end = ustrchr (path, TEXT(';'))) || (end = ustrchr (path, TEXT('?'))) || 1603: (end = ustrchr (path, TEXT('#'))))) 1604: end = path + ustrlen (path); 1.28 cvs 1605: 1606: /* Parse string second time to simplify */ 1607: p = path; 1608: while (p < end) 1609: { 1.77 cvs 1610: /* if we're pointing to a char, it's safe to reactivate the ../ convertion */ 1611: if (!ddot_simplify && *p != TEXT('.') && *p != used_sep) 1612: ddot_simplify = TRUE; 1613: 1.33 cvs 1614: if (*p==used_sep) 1.28 cvs 1615: { 1.67 cvs 1616: if (p > *url && *(p+1) == TEXT('.') && (*(p+2) == used_sep || !*(p+2))) 1.28 cvs 1617: { 1618: orig = p + 1; 1.33 cvs 1619: dest = (*(p+2)!=used_sep) ? p+2 : p+3; 1.52 cvs 1620: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/ 1.28 cvs 1621: end = orig - 1; 1622: } 1.77 cvs 1623: else if (ddot_simplify && *(p+1) == TEXT('.') && *(p+2) == TEXT('.') 1624: && (*(p+3) == used_sep || !*(p+3))) 1.28 cvs 1625: { 1626: newptr = p; 1.52 cvs 1627: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */ 1628: if (*newptr == used_sep) 1629: orig = newptr + 1; 1.28 cvs 1630: else 1.52 cvs 1631: orig = newptr; 1632: 1633: dest = (*(p+3) != used_sep) ? p+3 : p+4; 1634: while ((*orig++ = *dest++)); /* Remove /xxx/.. */ 1635: end = orig-1; 1636: /* Start again with prev slash */ 1637: p = newptr; 1.28 cvs 1638: } 1.33 cvs 1639: else if (*(p+1) == used_sep) 1.28 cvs 1640: { 1.33 cvs 1641: while (*(p+1) == used_sep) 1.28 cvs 1642: { 1643: orig = p; 1644: dest = p + 1; 1645: while ((*orig++ = *dest++)); /* Remove multiple /'s */ 1646: end = orig-1; 1647: } 1648: } 1649: else 1.25 cvs 1650: p++; 1.28 cvs 1651: } 1652: else 1653: p++; 1.25 cvs 1654: } 1655: } 1.51 cvs 1656: 1657: /* 1658: ** Check for host/../.. kind of things 1659: */ 1.77 cvs 1660: if (*path == used_sep && *(path+1) == TEXT('.') && *(path+2) == TEXT('.') 1661: && (!*(path+3) || *(path+3) == used_sep)) 1.51 cvs 1662: *(path+1) = EOS; 1663: 1.28 cvs 1664: return; 1665: } 1666: 1667: 1668: /*---------------------------------------------------------------------- 1669: NormalizeFile normalizes local names. 1670: Return TRUE if target and src differ. 1671: ----------------------------------------------------------------------*/ 1672: #ifdef __STDC__ 1.67 cvs 1673: ThotBool NormalizeFile (STRING src, STRING target) 1.28 cvs 1674: #else 1.67 cvs 1675: ThotBool NormalizeFile (src, target) 1676: STRING src; 1677: STRING target; 1.28 cvs 1678: 1679: #endif 1680: { 1.67 cvs 1681: STRING s; 1.66 cvs 1682: ThotBool change; 1.28 cvs 1683: 1.54 cvs 1684: change = FALSE; 1.67 cvs 1685: if (ustrncmp (src, TEXT("file:"), 5) == 0) 1.28 cvs 1686: { 1687: /* remove the prefix file: */ 1688: if (src[5] == EOS) 1.67 cvs 1689: ustrcpy (target, DIR_STR); 1690: else if (src[0] == TEXT('~')) 1.28 cvs 1691: { 1692: /* replace ~ */ 1.74 cvs 1693: s = TtaGetEnvString ("HOME"); 1.67 cvs 1694: ustrcpy (target, s); 1695: ustrcat (target, &src[5]); 1.28 cvs 1696: } 1697: else 1.67 cvs 1698: ustrcpy (target, &src[5]); 1.54 cvs 1699: change = TRUE; 1.28 cvs 1700: } 1.53 cvs 1701: # ifndef _WINDOWS 1.67 cvs 1702: else if (src[0] == TEXT('~')) 1.53 cvs 1703: { 1704: /* replace ~ */ 1705: s = (char *) TtaGetEnvString ("HOME"); 1706: strcpy (target, s); 1707: if (src[1] != DIR_SEP) 1708: strcat (target, DIR_STR); 1709: strcat (target, &src[1]); 1.54 cvs 1710: change = TRUE; 1.53 cvs 1711: } 1712: # endif /* _WINDOWS */ 1.28 cvs 1713: else 1.67 cvs 1714: ustrcpy (target, src); 1.28 cvs 1715: 1716: /* remove /../ and /./ */ 1.29 cvs 1717: SimplifyUrl (&target); 1.54 cvs 1718: if (!change) 1.67 cvs 1719: change = ustrcmp (src, target); 1.28 cvs 1720: return (change); 1.25 cvs 1721: } 1722: 1.28 cvs 1723: 1.25 cvs 1724: /*---------------------------------------------------------------------- 1.31 cvs 1725: MakeRelativeURL: make relative name 1.25 cvs 1726: 1.28 cvs 1727: This function creates and returns a string which gives an expression of 1728: one address as related to another. Where there is no relation, an absolute 1729: address is retured. 1.25 cvs 1730: 1.28 cvs 1731: On entry, 1.25 cvs 1732: Both names must be absolute, fully qualified names of nodes 1733: (no fragment bits) 1734: 1.28 cvs 1735: On exit, 1.25 cvs 1736: The return result points to a newly allocated name which, if 1737: parsed by AmayaParseUrl relative to relatedName, will yield aName. 1738: The caller is responsible for freeing the resulting name later. 1739: ----------------------------------------------------------------------*/ 1740: #ifdef __STDC__ 1.67 cvs 1741: STRING MakeRelativeURL (STRING aName, STRING relatedName) 1.25 cvs 1742: #else /* __STDC__ */ 1.67 cvs 1743: STRING MakeRelativeURL (aName, relatedName) 1744: STRING aName; 1745: STRING relatedName; 1.25 cvs 1746: #endif /* __STDC__ */ 1747: { 1.67 cvs 1748: STRING return_value; 1749: CHAR_T result[MAX_LENGTH]; 1750: STRING p; 1751: STRING q; 1752: STRING after_access; 1753: STRING last_slash = NULL; 1.29 cvs 1754: int slashes, levels, len; 1755: 1.44 cvs 1756: # ifdef _WINDOWS 1757: int ndx; 1758: # endif /* _WINDOWS */ 1759: 1.29 cvs 1760: if (aName == NULL || relatedName == NULL) 1761: return (NULL); 1762: 1763: slashes = 0; 1764: after_access = NULL; 1765: p = aName; 1766: q = relatedName; 1767: for (; *p && (*p == *q); p++, q++) 1.27 cvs 1768: { 1769: /* Find extent of match */ 1.67 cvs 1770: if (*p == TEXT(':')) 1.29 cvs 1771: after_access = p + 1; 1.28 cvs 1772: if (*p == DIR_SEP) 1.27 cvs 1773: { 1.29 cvs 1774: /* memorize the last slash position and count them */ 1.27 cvs 1775: last_slash = p; 1776: slashes++; 1.25 cvs 1777: } 1778: } 1779: 1.31 cvs 1780: /* q, p point to the first non-matching character or zero */ 1781: if (*q == EOS) 1782: { 1783: /* New name is a subset of the related name */ 1784: /* exactly the right length */ 1.67 cvs 1785: len = ustrlen (p); 1786: if ((return_value = TtaAllocString (len + 1)) != NULL) 1787: ustrcpy (return_value, p); 1.31 cvs 1788: } 1789: else if ((slashes < 2 && after_access == NULL) 1790: || (slashes < 3 && after_access != NULL)) 1791: { 1792: /* Two names whitout common path */ 1793: /* exactly the right length */ 1.67 cvs 1794: len = ustrlen (aName); 1795: if ((return_value = TtaAllocString (len + 1)) != NULL) 1796: ustrcpy (return_value, aName); 1.31 cvs 1797: } 1798: else 1799: { 1800: /* Some path in common */ 1.67 cvs 1801: if (slashes == 3 && ustrncmp (aName, TEXT("http:"), 5) == 0) 1.31 cvs 1802: /* just the same server */ 1.67 cvs 1803: ustrcpy (result, last_slash); 1.31 cvs 1804: else 1805: { 1806: levels= 0; 1.67 cvs 1807: for (; *q && *q != TEXT('#') && *q != TEXT(';') && *q != TEXT('?'); q++) 1.31 cvs 1808: if (*q == DIR_SEP) 1809: levels++; 1810: 1.52 cvs 1811: result[0] = EOS; 1.31 cvs 1812: for (;levels; levels--) 1.67 cvs 1813: ustrcat (result, TEXT("../")); 1814: ustrcat (result, last_slash+1); 1.31 cvs 1815: } 1.52 cvs 1816: 1817: if (!*result) 1.67 cvs 1818: ustrcat (result, TEXT("./")); 1.52 cvs 1819: 1.31 cvs 1820: /* exactly the right length */ 1.67 cvs 1821: len = ustrlen (result); 1822: if ((return_value = TtaAllocString (len + 1)) != NULL) 1823: ustrcpy (return_value, result); 1.52 cvs 1824: 1.25 cvs 1825: } 1.44 cvs 1826: # ifdef _WINDOWS 1.67 cvs 1827: len = ustrlen (return_value); 1.44 cvs 1828: for (ndx = 0; ndx < len; ndx ++) 1.67 cvs 1829: if (return_value[ndx] == TEXT('\\')) 1830: return_value[ndx] = TEXT('/') ; 1.44 cvs 1831: # endif /* _WINDOWS */ 1.29 cvs 1832: return (return_value); 1.24 cvs 1833: } 1.35 cvs 1834: 1835: