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: /*---------------------------------------------------------------------- 1.11 cvs 56: ExplodeURL 1.8 cvs 57: ----------------------------------------------------------------------*/ 58: #ifdef __STDC__ 59: void ExplodeURL (char *url, char **proto, char **host, char **dir, char **file) 60: #else 61: void ExplodeURL (url, proto, host, dir, file) 62: char *url; 63: char **proto; 64: char **host; 65: char **dir; 66: char **file; 67: 68: #endif 69: { 1.33 cvs 70: char *curr, *temp; 71: char used_sep; 1.32 cvs 72: 1.33 cvs 73: if (url && strchr (url, URL_SEP)) 74: used_sep = URL_SEP; 75: else 76: used_sep = DIR_SEP; 1.8 cvs 77: 78: if ((url == NULL) || (proto == NULL) || (host == NULL) || 79: (dir == NULL) || (file == NULL)) 80: return; 81: 82: /* initialize every pointer */ 83: *proto = *host = *dir = *file = NULL; 84: 85: /* skip any leading space */ 86: while ((*url == SPACE) || (*url == TAB)) 87: url++; 1.9 cvs 88: curr = url; 89: if (*curr == 0) 1.8 cvs 90: goto finished; 91: 92: /* go to the end of the URL */ 1.68 cvs 93: while ((*curr != EOS) && (*curr != SPACE) && (*curr != BSPACE) && 94: (*curr != __CR__) && (*curr != EOL)) 1.9 cvs 95: curr++; 1.8 cvs 96: 97: /* mark the end of the chain */ 1.9 cvs 98: *curr = EOS; 99: curr--; 100: if (curr <= url) 1.8 cvs 101: goto finished; 102: 103: /* search the next DIR_SEP indicating the beginning of the file name */ 104: do 1.11 cvs 105: curr--; 1.33 cvs 106: while ((curr >= url) && (*curr != used_sep)); 1.11 cvs 107: 1.9 cvs 108: if (curr < url) 1.8 cvs 109: goto finished; 1.9 cvs 110: *file = curr + 1; 1.8 cvs 111: 112: /* mark the end of the dir */ 1.9 cvs 113: *curr = EOS; 114: curr--; 115: if (curr < url) 1.8 cvs 116: goto finished; 117: 1.29 cvs 118: /* search for the DIR_STR indicating the host name start */ 1.33 cvs 119: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep))) 1.9 cvs 120: curr--; 1.8 cvs 121: 122: /* if we found it, separate the host name from the directory */ 1.33 cvs 123: if ((*curr == DIR_SEP) && (*(curr + 1) == used_sep)) 1.8 cvs 124: { 1.9 cvs 125: *host = temp = curr + 2; 1.33 cvs 126: while ((*temp != 0) && (*temp != used_sep)) 1.8 cvs 127: temp++; 1.33 cvs 128: if (*temp == used_sep) 1.8 cvs 129: { 130: *temp = EOS; 131: *dir = temp + 1; 132: } 133: } 134: else 1.11 cvs 135: *dir = curr; 136: 1.9 cvs 137: if (curr <= url) 1.8 cvs 138: goto finished; 139: 140: /* mark the end of the proto */ 1.9 cvs 141: *curr = EOS; 142: curr--; 143: if (curr < url) 1.8 cvs 144: goto finished; 145: 1.67 cvs 146: if (*curr == TEXT(':')) 1.8 cvs 147: { 1.9 cvs 148: *curr = EOS; 149: curr--; 1.8 cvs 150: } 151: else 152: goto finished; 1.11 cvs 153: 1.9 cvs 154: if (curr < url) 1.8 cvs 155: goto finished; 1.9 cvs 156: while ((curr > url) && (isalpha (*curr))) 157: curr--; 158: *proto = curr; 1.8 cvs 159: 160: finished:; 161: 162: #ifdef AMAYA_DEBUG 163: fprintf (stderr, "ExplodeURL(%s)\n\t", url); 164: if (*proto) 165: fprintf (stderr, "proto : %s, ", *proto); 166: if (*host) 167: fprintf (stderr, "host : %s, ", *host); 168: if (*dir) 169: fprintf (stderr, "dir : %s, ", *dir); 170: if (*file) 171: fprintf (stderr, "file : %s ", *file); 172: fprintf (stderr, "\n"); 173: #endif 174: 175: } 1.3 cvs 176: 1.61 cvs 177: 178: /*---------------------------------------------------------------------- 179: ExtractSuffix extract suffix from document nane. 180: ----------------------------------------------------------------------*/ 181: #ifdef __STDC__ 182: void ExtractSuffix (STRING aName, STRING aSuffix) 183: #else 184: void ExtractSuffix (aName, aSuffix) 185: STRING aName; 186: STRING aSuffix; 187: 188: #endif 189: { 190: int lg, i; 191: STRING ptr, oldptr; 192: 193: if (!aSuffix || !aName) 194: /* bad suffix */ 195: return; 196: 197: aSuffix[0] = EOS; 198: lg = ustrlen (aName); 199: if (lg) 200: { 201: /* the name is not empty */ 202: oldptr = ptr = &aName[0]; 203: do 204: { 1.67 cvs 205: ptr = ustrrchr (oldptr, TEXT('.')); 1.61 cvs 206: if (ptr) 207: oldptr = &ptr[1]; 208: } 209: while (ptr); 210: 211: i = (int) (oldptr) - (int) (aName); /* name length */ 212: if (i > 1) 213: { 214: aName[i - 1] = EOS; 215: if (i != lg) 216: ustrcpy (aSuffix, oldptr); 217: } 218: } 219: } 220: 1.4 cvs 221: /*---------------------------------------------------------------------- 1.9 cvs 222: IsHTMLName 223: returns TRUE if path points to an HTML resource. 1.4 cvs 224: ----------------------------------------------------------------------*/ 1.3 cvs 225: #ifdef __STDC__ 1.67 cvs 226: ThotBool IsHTMLName (const STRING path) 1.3 cvs 227: #else /* __STDC__ */ 1.67 cvs 228: ThotBool IsHTMLName (path) 229: const STRING path; 1.3 cvs 230: #endif /* __STDC__ */ 231: { 1.67 cvs 232: CHAR_T temppath[MAX_LENGTH]; 233: CHAR_T suffix[MAX_LENGTH]; 234: CHAR_T nsuffix[MAX_LENGTH]; 1.5 cvs 235: int i; 236: 237: if (!path) 1.37 cvs 238: return (FALSE); 1.5 cvs 239: 1.67 cvs 240: ustrcpy (temppath, path); 1.5 cvs 241: ExtractSuffix (temppath, suffix); 242: 243: /* Normalize the suffix */ 244: i = 0; 1.39 cvs 245: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 246: { 1.67 cvs 247: nsuffix[i] = utolower (suffix[i]); 1.13 cvs 248: i++; 249: } 1.5 cvs 250: nsuffix[i] = EOS; 1.67 cvs 251: if (!ustrcmp (nsuffix, TEXT("html")) || 252: !ustrcmp (nsuffix, TEXT("htm")) || 253: !ustrcmp (nsuffix, TEXT("shtml")) || 254: !ustrcmp (nsuffix, TEXT("xht")) || 255: !ustrcmp (nsuffix, TEXT("xhtm")) || 256: !ustrcmp (nsuffix, TEXT("xhtml"))) 1.39 cvs 257: return (TRUE); 1.67 cvs 258: else if (!ustrcmp (nsuffix, TEXT("gz"))) 1.13 cvs 259: { 1.39 cvs 260: /* take into account compressed files */ 1.13 cvs 261: ExtractSuffix (temppath, suffix); 262: /* Normalize the suffix */ 263: i = 0; 1.39 cvs 264: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 265: { 1.25 cvs 266: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 267: i++; 268: } 269: nsuffix[i] = EOS; 1.67 cvs 270: if (!ustrcmp (nsuffix, TEXT("html")) || 271: !ustrcmp (nsuffix, TEXT("htm")) || 272: !ustrcmp (nsuffix, TEXT("shtml")) || 273: !ustrcmp (nsuffix, TEXT("xht")) || 274: !ustrcmp (nsuffix, TEXT("xhtm")) || 275: !ustrcmp (nsuffix, TEXT("xhtml"))) 1.63 cvs 276: 1.39 cvs 277: return (TRUE); 278: else 1.13 cvs 279: return (FALSE); 280: } 281: else 1.39 cvs 282: return (FALSE); 1.3 cvs 283: } 284: 1.4 cvs 285: /*---------------------------------------------------------------------- 1.56 cvs 286: IsXMLName 287: returns TRUE if path points to an XML resource. 288: ----------------------------------------------------------------------*/ 289: #ifdef __STDC__ 1.67 cvs 290: ThotBool IsXMLName (const STRING path) 1.56 cvs 291: #else /* __STDC__ */ 1.67 cvs 292: ThotBool IsXMLName (path) 293: const STRING path; 1.56 cvs 294: #endif /* __STDC__ */ 295: { 1.67 cvs 296: CHAR_T temppath[MAX_LENGTH]; 297: CHAR_T suffix[MAX_LENGTH]; 1.56 cvs 298: 299: if (!path) 300: return (FALSE); 301: 1.67 cvs 302: ustrcpy (temppath, path); 1.56 cvs 303: ExtractSuffix (temppath, suffix); 304: 1.67 cvs 305: if (!ustrcasecmp (suffix, TEXT("xml")) || 306: !ustrcasecmp (suffix, TEXT("xht")) || 307: !ustrcmp (suffix, TEXT("xhtm")) || 308: !ustrcmp (suffix, TEXT("xhtml"))) 1.56 cvs 309: return (TRUE); 1.67 cvs 310: else if (!ustrcmp (suffix, TEXT("gz"))) 1.56 cvs 311: { 312: /* take into account compressed files */ 313: ExtractSuffix (temppath, suffix); 1.67 cvs 314: if (!ustrcasecmp (suffix, TEXT("xml")) || 315: !ustrcasecmp (suffix, TEXT("xht")) || 316: !ustrcmp (suffix, TEXT("xhtm")) || 317: !ustrcmp (suffix, TEXT("xhtml"))) 1.60 cvs 318: return (TRUE); 319: else 320: return (FALSE); 321: } 322: else 323: return (FALSE); 324: } 325: 326: /*---------------------------------------------------------------------- 327: IsCSSName 328: returns TRUE if path points to an XML resource. 329: ----------------------------------------------------------------------*/ 330: #ifdef __STDC__ 1.67 cvs 331: ThotBool IsCSSName (const STRING path) 1.60 cvs 332: #else /* __STDC__ */ 1.67 cvs 333: ThotBool IsCSSName (path) 334: const STRING path; 1.60 cvs 335: #endif /* __STDC__ */ 336: { 1.67 cvs 337: CHAR_T temppath[MAX_LENGTH]; 338: CHAR_T suffix[MAX_LENGTH]; 1.60 cvs 339: 340: if (!path) 341: return (FALSE); 342: 1.67 cvs 343: ustrcpy (temppath, path); 1.60 cvs 344: ExtractSuffix (temppath, suffix); 345: 1.67 cvs 346: if (!ustrcasecmp (suffix, TEXT("css"))) 1.60 cvs 347: return (TRUE); 1.67 cvs 348: else if (!ustrcmp (suffix, TEXT("gz"))) 1.60 cvs 349: { 350: /* take into account compressed files */ 351: ExtractSuffix (temppath, suffix); 1.67 cvs 352: if (!ustrcasecmp (suffix, TEXT("css"))) 1.56 cvs 353: return (TRUE); 354: else 355: return (FALSE); 356: } 357: else 358: return (FALSE); 359: } 360: 361: /*---------------------------------------------------------------------- 1.9 cvs 362: IsImageName 363: returns TRUE if path points to an image resource. 1.4 cvs 364: ----------------------------------------------------------------------*/ 1.3 cvs 365: #ifdef __STDC__ 1.67 cvs 366: ThotBool IsImageName (const STRING path) 1.3 cvs 367: #else /* __STDC__ */ 1.67 cvs 368: ThotBool IsImageName (path) 369: const STRING path; 1.3 cvs 370: #endif /* __STDC__ */ 371: { 1.67 cvs 372: CHAR_T temppath[MAX_LENGTH]; 373: CHAR_T suffix[MAX_LENGTH]; 374: CHAR_T nsuffix[MAX_LENGTH]; 1.5 cvs 375: int i; 376: 377: if (!path) 1.13 cvs 378: return (FALSE); 1.5 cvs 379: 1.67 cvs 380: ustrcpy (temppath, path); 1.5 cvs 381: ExtractSuffix (temppath, suffix); 382: 383: /* Normalize the suffix */ 384: i = 0; 1.39 cvs 385: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 386: { 1.67 cvs 387: nsuffix[i] = utolower (suffix[i]); 1.13 cvs 388: i++; 389: } 1.5 cvs 390: nsuffix[i] = EOS; 1.67 cvs 391: if ((!ustrcmp (nsuffix, TEXT("gif"))) || (!ustrcmp (nsuffix, TEXT("xbm"))) || 392: (!ustrcmp (nsuffix, TEXT("xpm"))) || (!ustrcmp (nsuffix, TEXT("jpg"))) || 393: (!ustrcmp (nsuffix, TEXT("png"))) || (!ustrcmp (nsuffix, TEXT("au")))) 1.39 cvs 394: return (TRUE); 395: return (FALSE); 1.3 cvs 396: } 397: 1.4 cvs 398: /*---------------------------------------------------------------------- 1.58 cvs 399: IsImageType 400: returns TRUE if type points to an image resource. 401: ----------------------------------------------------------------------*/ 402: #ifdef __STDC__ 1.67 cvs 403: ThotBool IsImageType (const STRING type) 1.58 cvs 404: #else /* __STDC__ */ 1.67 cvs 405: ThotBool IsImageType (type) 406: const STRING type; 1.58 cvs 407: #endif /* __STDC__ */ 408: { 1.67 cvs 409: CHAR_T temptype[MAX_LENGTH]; 1.58 cvs 410: int i; 411: 412: if (!type) 413: return (FALSE); 414: 1.67 cvs 415: ustrcpy (temptype, type); 1.58 cvs 416: /* Normalize the type */ 417: i = 0; 418: while (temptype[i] != EOS) 419: { 420: temptype[i] = tolower (temptype[i]); 421: i++; 422: } 1.67 cvs 423: if ((!ustrcmp (temptype, TEXT("gif"))) || (!ustrcmp (temptype, TEXT("x-xbitmap"))) || 424: (!ustrcmp (temptype, TEXT("x-xpixmap"))) || (!ustrcmp (temptype, TEXT("jpeg"))) || 425: (!ustrcmp (temptype, TEXT("png")))) 1.58 cvs 426: return (TRUE); 427: return (FALSE); 428: } 429: 430: /*---------------------------------------------------------------------- 1.9 cvs 431: IsTextName 1.4 cvs 432: ----------------------------------------------------------------------*/ 1.3 cvs 433: #ifdef __STDC__ 1.67 cvs 434: ThotBool IsTextName (const STRING path) 1.3 cvs 435: #else /* __STDC__ */ 1.67 cvs 436: ThotBool IsTextName (path) 437: const STRING path; 1.3 cvs 438: 439: #endif /* __STDC__ */ 440: { 1.67 cvs 441: CHAR_T temppath[MAX_LENGTH]; 442: CHAR_T suffix[MAX_LENGTH]; 443: CHAR_T nsuffix[MAX_LENGTH]; 1.5 cvs 444: int i; 445: 446: if (!path) 1.13 cvs 447: return (FALSE); 1.5 cvs 448: 1.67 cvs 449: ustrcpy (temppath, path); 1.5 cvs 450: ExtractSuffix (temppath, suffix); 451: 452: /* Normalize the suffix */ 453: i = 0; 1.39 cvs 454: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.5 cvs 455: { 1.25 cvs 456: nsuffix[i] = tolower (suffix[i]); 1.5 cvs 457: i++; 458: } 459: nsuffix[i] = EOS; 460: 1.67 cvs 461: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd")))) 1.13 cvs 462: return (TRUE); 1.67 cvs 463: else if (!ustrcmp (nsuffix, TEXT("gz"))) 1.13 cvs 464: { 1.39 cvs 465: /* take into account compressed files */ 1.13 cvs 466: ExtractSuffix (temppath, suffix); 467: /* Normalize the suffix */ 468: i = 0; 1.39 cvs 469: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 470: { 1.25 cvs 471: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 472: i++; 473: } 474: nsuffix[i] = EOS; 1.67 cvs 475: if ((!ustrcmp (nsuffix, TEXT("txt"))) || (!ustrcmp (nsuffix, TEXT("dtd")))) 1.13 cvs 476: return (TRUE); 477: else 478: return (FALSE); 479: } 480: else 481: return (FALSE); 1.3 cvs 482: } 483: 1.4 cvs 484: /*---------------------------------------------------------------------- 1.9 cvs 485: IsHTTPPath 486: returns TRUE if path is in fact an http URL. 1.4 cvs 487: ----------------------------------------------------------------------*/ 1.3 cvs 488: #ifdef __STDC__ 1.67 cvs 489: ThotBool IsHTTPPath (const STRING path) 1.3 cvs 490: #else /* __STDC__ */ 1.67 cvs 491: ThotBool IsHTTPPath (path) 492: const STRING path; 1.3 cvs 493: #endif /* __STDC__ */ 494: { 1.5 cvs 495: if (!path) 496: return FALSE; 1.3 cvs 497: 1.67 cvs 498: if ((!ustrncmp (path, TEXT("http:"), 5) != 0) 499: || !ustrncmp (path, TEXT("internal:"), 9)) 1.58 cvs 500: return TRUE; 501: return FALSE; 1.3 cvs 502: } 503: 1.4 cvs 504: /*---------------------------------------------------------------------- 1.9 cvs 505: IsWithParameters 506: returns TRUE if url has a concatenated query string. 1.4 cvs 507: ----------------------------------------------------------------------*/ 1.3 cvs 508: #ifdef __STDC__ 1.66 cvs 509: ThotBool IsWithParameters (const char *url) 1.3 cvs 510: #else /* __STDC__ */ 1.66 cvs 511: ThotBool IsWithParameters (url) 1.34 cvs 512: const char *url; 1.3 cvs 513: #endif /* __STDC__ */ 514: { 1.5 cvs 515: int i; 1.3 cvs 516: 1.9 cvs 517: if ((!url) || (url[0] == EOS)) 1.5 cvs 518: return FALSE; 1.3 cvs 519: 1.9 cvs 520: i = strlen (url) - 1; 521: while (i > 0 && url[i--] != '?') 1.5 cvs 522: if (i < 0) 523: return FALSE; 1.3 cvs 524: 1.5 cvs 525: /* There is a parameter */ 526: return TRUE; 1.3 cvs 527: } 528: 1.4 cvs 529: /*---------------------------------------------------------------------- 1.9 cvs 530: IsW3Path 531: returns TRUE if path is in fact a URL. 1.4 cvs 532: ----------------------------------------------------------------------*/ 1.3 cvs 533: #ifdef __STDC__ 1.67 cvs 534: ThotBool IsW3Path (const STRING path) 1.3 cvs 535: #else /* __STDC__ */ 1.67 cvs 536: ThotBool IsW3Path (path) 537: const STRING path; 1.3 cvs 538: #endif /* __STDC__ */ 539: { 1.67 cvs 540: if ((ustrncmp (path, TEXT("http:"), 5)) && (ustrncmp (path, TEXT("ftp:"), 4)) && 541: (ustrncmp (path, TEXT("telnet:"), 7)) && (ustrncmp (path, TEXT("wais:"), 5)) && 542: (ustrncmp (path, TEXT("news:"), 5)) && (ustrncmp (path, TEXT("gopher:"), 7)) && 543: (ustrncmp (path, TEXT("mailto:"), 7)) && (ustrncmp (path, TEXT("archie:"), 7))) 1.5 cvs 544: return FALSE; 545: return TRUE; 1.3 cvs 546: } 547: 1.4 cvs 548: /*---------------------------------------------------------------------- 1.9 cvs 549: IsValidProtocol 550: returns true if the url protocol is supported by Amaya. 1.4 cvs 551: ----------------------------------------------------------------------*/ 1.3 cvs 552: #ifdef __STDC__ 1.67 cvs 553: ThotBool IsValidProtocol (const STRING url) 1.3 cvs 554: #else /* __STDC__ */ 1.67 cvs 555: ThotBool IsValidProtocol (url) 556: const STRING url; 1.3 cvs 557: #endif /* __STDC__ */ 558: { 1.67 cvs 559: if (!ustrncmp (url, TEXT("http:"), 5) 1.69 ! cvs 560: || !ustrncmp (url, TEXT("internal:"), 9) ! 561: || !strncmp (url, "ftp:", 4)) 1.22 cvs 562: /* experimental */ 1.58 cvs 563: /*** || !strncmp (url, "ftp:", 4) ***/ 1.24 cvs 564: /*** || !strncmp (path, "news:", 5)***/ 1.8 cvs 565: return (TRUE); 1.5 cvs 566: else 1.8 cvs 567: return (FALSE); 1.3 cvs 568: } 569: 1.31 cvs 570: 571: /*---------------------------------------------------------------------- 572: GetBaseURL 573: normalizes orgName according to a base associated with doc, and 574: following the standard URL format rules. 575: The function returns the base used to solve relative URL and SRC: 576: - the base of the document, 577: - or the document path (without document name). 578: ----------------------------------------------------------------------*/ 579: #ifdef __STDC__ 1.67 cvs 580: STRING GetBaseURL (Document doc) 1.31 cvs 581: #else /* __STDC__ */ 1.67 cvs 582: STRING GetBaseURL (doc) 1.31 cvs 583: Document doc; 584: #endif /* __STDC__ */ 585: { 586: Element el; 587: ElementType elType; 588: AttributeType attrType; 589: Attribute attr; 1.67 cvs 590: STRING ptr, basename; 1.31 cvs 591: int length; 592: 1.57 cvs 593: /* @@@ irene */ 594: if (!DocumentURLs[doc]) 595: return NULL; 1.67 cvs 596: basename = TtaAllocString (MAX_LENGTH); 597: ustrncpy (basename, DocumentURLs[doc], MAX_LENGTH-1); 1.39 cvs 598: basename[MAX_LENGTH-1] = EOS; 1.31 cvs 599: length = MAX_LENGTH -1; 600: /* get the root element */ 601: el = TtaGetMainRoot (doc); 602: /* search the BASE element */ 603: elType.ElSSchema = TtaGetDocumentSSchema (doc); 1.65 cvs 604: elType.ElTypeNum = HTML_EL_HEAD; 605: el = TtaSearchTypedElement (elType, SearchForward, el); 606: if (el) 607: { 608: elType.ElTypeNum = HTML_EL_BASE; 609: el = TtaSearchTypedElement (elType, SearchInTree, el); 610: } 1.31 cvs 611: if (el) 612: { 613: /* The document has a BASE element -> Get the HREF attribute */ 614: attrType.AttrSSchema = elType.ElSSchema; 615: attrType.AttrTypeNum = HTML_ATTR_HREF_; 616: attr = TtaGetAttribute (el, attrType); 617: if (attr) 618: { 619: /* Use the base path of the document */ 620: TtaGiveTextAttributeValue (attr, basename, &length); 621: /* base and orgName have to be separated by a DIR_SEP */ 622: length--; 1.43 cvs 623: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP) 1.31 cvs 624: /* verify if the base has the form "protocol://server:port" */ 625: { 1.67 cvs 626: ptr = AmayaParseUrl (basename, _EMPTYSTR_, AMAYA_PARSE_ACCESS | 1.33 cvs 627: AMAYA_PARSE_HOST | 628: AMAYA_PARSE_PUNCTUATION); 1.67 cvs 629: if (ptr && !ustrcmp (ptr, basename)) 1.31 cvs 630: { 1.43 cvs 631: /* it has this form, we complete it by adding a URL_STR */ 1.67 cvs 632: if (ustrchr (basename, DIR_SEP)) 633: ustrcat (basename, DIR_STR); 1.43 cvs 634: else 1.67 cvs 635: ustrcat (basename, URL_STR); 1.31 cvs 636: length++; 637: } 638: if (ptr) 639: TtaFreeMemory (ptr); 640: } 641: } 1.33 cvs 642: } 643: 1.31 cvs 644: /* Remove anything after the last DIR_SEP char. If no such char is found, 645: * then search for the first ":" char, hoping that what's before that is a 646: * protocol. If found, end the string there. If neither char is found, 647: * then discard the whole base element. 648: */ 1.67 cvs 649: length = ustrlen (basename) - 1; 1.31 cvs 650: /* search for the last DIR_SEP char */ 1.43 cvs 651: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP) 1.31 cvs 652: length--; 653: if (length >= 0) 654: /* found the last DIR_SEP char, end the string there */ 655: basename[length + 1] = EOS; 656: else 657: /* search for the first PATH_STR char */ 658: { 1.67 cvs 659: for (length = 0; basename[length] != TEXT(':') && 1.31 cvs 660: basename[length] != EOS; length ++); 1.67 cvs 661: if (basename[length] == TEXT(':')) 1.31 cvs 662: /* found, so end the string there */ 663: basename[length + 1] = EOS; 664: else 665: /* not found, discard the base */ 666: basename[0] = EOS; 667: } 668: return (basename); 669: } 670: 671: 1.4 cvs 672: /*---------------------------------------------------------------------- 1.40 cvs 673: GetLocalPath 674: Allocate and return the local document path associated to the url 675: ----------------------------------------------------------------------*/ 676: #ifdef __STDC__ 1.67 cvs 677: STRING GetLocalPath (Document doc, STRING url) 1.40 cvs 678: #else /* __STDC__ */ 1.67 cvs 679: STRING GetLocalPath (doc, url) 1.40 cvs 680: Document doc; 1.67 cvs 681: STRING url; 1.40 cvs 682: #endif /* __STDC__ */ 683: { 1.67 cvs 684: STRING ptr, n; 685: STRING documentname; 686: CHAR_T url_sep; 1.40 cvs 687: int len; 1.67 cvs 688: STRING tmpDir, tmpName; 689: ThotBool noFile; 1.40 cvs 690: 691: if (url != NULL) 692: { 693: /* check whether the file name exists */ 1.67 cvs 694: len = ustrlen (url) - 1; 1.41 cvs 695: if (IsW3Path) 1.67 cvs 696: url_sep = TEXT('/'); 1.41 cvs 697: else 698: url_sep = DIR_SEP; 699: noFile = (url[len] == url_sep); 1.40 cvs 700: if (noFile) 701: url[len] = EOS; 1.67 cvs 702: /* ptr = TtaAllocString (MAX_LENGTH); 703: documentname = TtaAllocString (MAX_LENGTH); */ 704: tmpDir = TtaAllocString (MAX_LENGTH ); 705: tmpName = TtaAllocString(MAX_LENGTH ); 706: TtaExtractName (url, tmpDir, tmpName); 707: ptr = tmpDir; 708: documentname = tmpName; 709: 710: usprintf (ptr, TEXT("%s%s%d%s"), TempFileDirectory, DIR_STR, doc, DIR_STR); 1.40 cvs 711: if (!TtaCheckDirectory (ptr)) 712: /* directory did not exist */ 1.59 cvs 713: # ifdef _WINDOWS 1.67 cvs 714: umkdir (ptr); 1.59 cvs 715: # else /* !_WINDOWS */ 1.67 cvs 716: umkdir (ptr, S_IRWXU); 1.59 cvs 717: # endif /* !_WINDOWS */ 1.47 cvs 718: 719: /* don't include the query string within document name */ 1.67 cvs 720: n = ustrrchr(documentname, TEXT('?')); 1.47 cvs 721: if (n != NULL) 722: *n = EOS; 1.46 cvs 723: /* don't include ':' within document name */ 1.67 cvs 724: n = ustrchr (documentname, TEXT(':')); 1.46 cvs 725: if (n != NULL) 726: *n = EOS; 1.69 ! cvs 727: /* if after all this operations document name ! 728: is empty, let's use noname.html instead */ ! 729: if (documentname[0] == EOS) ! 730: ustrcat (ptr, TEXT("noname.html")); ! 731: else ! 732: ustrcat (ptr, documentname); 1.40 cvs 733: TtaFreeMemory (documentname); 734: /* restore the url */ 735: if (noFile) 1.41 cvs 736: url[len] = url_sep; 1.40 cvs 737: return (ptr); 738: } 739: else 740: return (NULL); 741: } 742: 743: 744: /*---------------------------------------------------------------------- 1.9 cvs 745: NormalizeURL 746: normalizes orgName according to a base associated with doc, and 747: following the standard URL format rules. 1.53 cvs 748: if doc is 0 and otherPath not NULL, normalizes orgName according to this 749: other path. 1.9 cvs 750: The function returns the new complete and normalized URL 1.12 cvs 751: or file name path (newName) and the name of the document (docName). 1.9 cvs 752: N.B. If the function can't find out what's the docName, it assigns 753: the name "noname.html". 1.4 cvs 754: ----------------------------------------------------------------------*/ 1.3 cvs 755: #ifdef __STDC__ 1.67 cvs 756: void NormalizeURL (STRING orgName, Document doc, STRING newName, STRING docName, STRING otherPath) 1.3 cvs 757: #else /* __STDC__ */ 1.53 cvs 758: void NormalizeURL (orgName, doc, newName, docName, otherPath) 1.67 cvs 759: STRING orgName; 1.3 cvs 760: Document doc; 1.67 cvs 761: STRING newName; 762: STRING docName; 763: STRING otherPath; 1.3 cvs 764: #endif /* __STDC__ */ 765: { 1.67 cvs 766: STRING basename; 767: CHAR_T tempOrgName[MAX_LENGTH]; 768: STRING ptr; 769: CHAR_T used_sep; 770: int length; 1.66 cvs 771: ThotBool check; 1.5 cvs 772: 1.44 cvs 773: # ifdef _WINDOWS 774: int ndx; 775: # endif /* _WINDOWS */ 776: 1.5 cvs 777: if (!newName || !docName) 778: return; 1.18 cvs 779: 1.32 cvs 780: if (doc != 0) 1.53 cvs 781: basename = GetBaseURL (doc); 782: else if (otherPath != NULL) 783: basename = TtaStrdup (otherPath); 1.32 cvs 784: else 1.53 cvs 785: basename = NULL; 1.32 cvs 786: 1.18 cvs 787: /* 1.31 cvs 788: * Clean orgName 789: * Make sure we have a complete orgName, without any leading or trailing 790: * white spaces, or trailinbg new lines 791: */ 1.5 cvs 792: ptr = orgName; 1.18 cvs 793: /* skip leading white space and new line characters */ 1.67 cvs 794: while ((*ptr == SPACE || *ptr == EOL) && *ptr++ != EOS); 795: ustrncpy (tempOrgName, ptr, MAX_LENGTH -1); 1.39 cvs 796: tempOrgName[MAX_LENGTH -1] = EOS; 1.18 cvs 797: /* 1.31 cvs 798: * Make orgName a complete URL 799: * If the URL does not include a protocol, then try to calculate 800: * one using the doc's base element (if it exists), 801: */ 1.53 cvs 802: if (tempOrgName[0] == EOS) 803: { 804: newName[0] = EOS; 805: TtaFreeMemory (basename); 806: return; 807: } 1.49 cvs 808: 809: /* clean trailing white space */ 1.67 cvs 810: length = ustrlen (tempOrgName) - 1; 1.53 cvs 811: while (tempOrgName[length] == SPACE && tempOrgName[length] == EOL) 812: { 813: tempOrgName[length] = EOS; 814: length--; 815: } 1.50 cvs 816: 1.55 cvs 817: /* remove extra dot (which dot???) */ 818: /* ugly, but faster than a strcmp */ 1.67 cvs 819: if (tempOrgName[length] == TEXT('.') 820: && (length == 0 || tempOrgName[length-1] != TEXT('.'))) 1.55 cvs 821: tempOrgName[length] = EOS; 1.50 cvs 822: 1.53 cvs 823: if (IsW3Path (tempOrgName)) 824: { 825: /* the name is complete, go to the Sixth Step */ 1.67 cvs 826: ustrcpy (newName, tempOrgName); 1.53 cvs 827: SimplifyUrl (&newName); 828: /* verify if the URL has the form "protocol://server:port" */ 1.67 cvs 829: ptr = AmayaParseUrl (newName, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 830: if (ptr && !ustrcmp (ptr, newName)) /* it has this form, we complete it by adding a DIR_STR */ 831: ustrcat (newName, URL_STR); 1.49 cvs 832: 1.53 cvs 833: if (ptr) 1.50 cvs 834: TtaFreeMemory (ptr); 1.53 cvs 835: } 836: else if ( basename == NULL) 837: /* the name is complete, go to the Sixth Step */ 1.67 cvs 838: ustrcpy (newName, tempOrgName); 1.53 cvs 839: else 840: { 1.31 cvs 841: /* Calculate the absolute URL, using the base or document URL */ 1.44 cvs 842: # ifdef _WINDOWS 1.53 cvs 843: if (!IsW3Path (basename)) 844: { 1.67 cvs 845: length = ustrlen (tempOrgName); 1.53 cvs 846: for (ndx = 0; ndx < length; ndx++) 1.67 cvs 847: if (tempOrgName [ndx] == TEXT('/')) 848: tempOrgName [ndx] = TEXT('\\'); 1.53 cvs 849: } 1.44 cvs 850: # endif /* _WINDOWS */ 1.25 cvs 851: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL); 1.53 cvs 852: if (ptr) 853: { 854: SimplifyUrl (&ptr); 1.67 cvs 855: ustrcpy (newName, ptr); 1.53 cvs 856: TtaFreeMemory (ptr); 857: } 858: else 859: newName[0] = EOS; 860: } 1.36 cvs 861: 862: TtaFreeMemory (basename); 1.18 cvs 863: /* 1.31 cvs 864: * Prepare the docname that will refer to this ressource in the 865: * .amaya directory. If the new URL finishes on DIR_SEP, then use 866: * noname.html as a default ressource name 1.18 cvs 867: */ 1.53 cvs 868: if (newName[0] != EOS) 869: { 1.67 cvs 870: length = ustrlen (newName) - 1; 1.53 cvs 871: if (newName[length] == URL_SEP || newName[length] == DIR_SEP) 872: { 873: used_sep = newName[length]; 874: check = TRUE; 875: while (check) 876: { 1.50 cvs 877: length--; 878: while (length >= 0 && newName[length] != used_sep) 1.53 cvs 879: length--; 1.67 cvs 880: if (!ustrncmp (&newName[length+1], TEXT(".."), 2)) 1.53 cvs 881: { 882: newName[length+1] = EOS; 883: /* remove also previous directory */ 884: length--; 885: while (length >= 0 && newName[length] != used_sep) 886: length--; 1.67 cvs 887: if (ustrncmp (&newName[length+1], TEXT("//"), 2)) 1.53 cvs 888: /* don't remove server name */ 1.50 cvs 889: newName[length+1] = EOS; 1.53 cvs 890: } 1.67 cvs 891: else if (!ustrncmp (&newName[length+1], TEXT("."), 1)) 1.53 cvs 892: newName[length+1] = EOS; 1.50 cvs 893: else 1.53 cvs 894: check = FALSE; 895: } 1.67 cvs 896: ustrcpy (docName, TEXT("noname.html")); 1.53 cvs 897: /* docname was not comprised inside the URL, so let's */ 898: /* assign the default ressource name */ 1.67 cvs 899: ustrcpy (docName, TEXT("noname.html")); 1.53 cvs 900: } 901: else 902: { /* docname is comprised inside the URL */ 1.50 cvs 903: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP) 1.53 cvs 904: length--; 905: if (length < 0) 1.67 cvs 906: ustrcpy (docName, newName); 1.53 cvs 907: else 1.67 cvs 908: ustrcpy (docName, &newName[length+1]); 1.53 cvs 909: } 910: } 911: else 912: docName[0] = EOS; 1.18 cvs 913: } 1.3 cvs 914: 1.4 cvs 915: /*---------------------------------------------------------------------- 1.9 cvs 916: IsSameHost 1.4 cvs 917: ----------------------------------------------------------------------*/ 1.3 cvs 918: #ifdef __STDC__ 1.67 cvs 919: ThotBool IsSameHost (const STRING url1, const STRING url2) 1.3 cvs 920: #else /* __STDC__ */ 1.67 cvs 921: ThotBool IsSameHost (url1, url2) 922: const STRING url1; 923: const STRING url2; 1.3 cvs 924: #endif /* __STDC__ */ 925: { 1.67 cvs 926: STRING basename_ptr1, basename_ptr2; 927: ThotBool result; 1.3 cvs 928: 1.67 cvs 929: basename_ptr1 = AmayaParseUrl (url1, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 930: basename_ptr2 = AmayaParseUrl (url2, _EMPTYSTR_, AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 1.3 cvs 931: 1.67 cvs 932: if (ustrcmp (basename_ptr1, basename_ptr2)) 1.8 cvs 933: result = FALSE; 1.5 cvs 934: else 1.8 cvs 935: result = TRUE; 1.3 cvs 936: 1.25 cvs 937: TtaFreeMemory (basename_ptr1); 938: TtaFreeMemory (basename_ptr2); 1.5 cvs 939: return (result); 1.3 cvs 940: } 941: 942: 1.4 cvs 943: /*---------------------------------------------------------------------- 1.22 cvs 944: HasKnownFileSuffix 945: returns TRUE if path points to a file ending with a suffix. 946: ----------------------------------------------------------------------*/ 947: #ifdef __STDC__ 1.67 cvs 948: ThotBool HasKnownFileSuffix (const STRING path) 1.22 cvs 949: #else /* __STDC__ */ 1.67 cvs 950: ThotBool HasKnownFileSuffix (path) 951: const STRING path; 1.22 cvs 952: #endif /* __STDC__ */ 953: { 1.67 cvs 954: STRING root; 955: CHAR_T temppath[MAX_LENGTH]; 956: CHAR_T suffix[MAX_LENGTH]; 1.22 cvs 957: 1.67 cvs 958: if (!path || path[0] == EOS || path[ustrlen(path)] == DIR_SEP) 1.22 cvs 959: return (FALSE); 960: 1.67 cvs 961: root = AmayaParseUrl(path, _EMPTYSTR_, AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION); 1.22 cvs 962: 963: if (root) 964: { 1.67 cvs 965: ustrcpy (temppath, root); 1.25 cvs 966: TtaFreeMemory (root); 1.22 cvs 967: /* Get the suffix */ 968: ExtractSuffix (temppath, suffix); 969: 970: if( suffix[0] == EOS) 971: /* no suffix */ 972: return (FALSE); 973: 974: /* Normalize the suffix */ 975: ConvertToLowerCase (suffix); 976: 1.67 cvs 977: if (!ustrcmp (suffix, TEXT("gz"))) 1.22 cvs 978: /* skip the compressed suffix */ 979: { 980: ExtractSuffix (temppath, suffix); 981: if(suffix[0] == EOS) 982: /* no suffix */ 983: return (FALSE); 984: /* Normalize the suffix */ 985: ConvertToLowerCase (suffix); 986: } 987: 1.67 cvs 988: if (ustrcmp (suffix, TEXT("gif")) && 989: ustrcmp (suffix, TEXT("xbm")) && 990: ustrcmp (suffix, TEXT("xpm")) && 991: ustrcmp (suffix, TEXT("jpg")) && 992: ustrcmp (suffix, TEXT("pdf")) && 993: ustrcmp (suffix, TEXT("png")) && 994: ustrcmp (suffix, TEXT("tgz")) && 995: ustrcmp (suffix, TEXT("xpg")) && 996: ustrcmp (suffix, TEXT("xpd")) && 997: ustrcmp (suffix, TEXT("ps")) && 998: ustrcmp (suffix, TEXT("au")) && 999: ustrcmp (suffix, TEXT("html")) && 1000: ustrcmp (suffix, TEXT("htm")) && 1001: ustrcmp (suffix, TEXT("shtml")) && 1002: ustrcmp (suffix, TEXT("xht")) && 1003: ustrcmp (suffix, TEXT("xhtm")) && 1004: ustrcmp (suffix, TEXT("xhtml")) && 1005: ustrcmp (suffix, TEXT("txt")) && 1006: ustrcmp (suffix, TEXT("css")) && 1007: ustrcmp (suffix, TEXT("eps"))) 1.22 cvs 1008: return (FALSE); 1009: else 1010: return (TRUE); 1011: } 1012: else 1013: return (FALSE); 1014: } 1015: 1016: 1017: /*---------------------------------------------------------------------- 1.24 cvs 1018: ChopURL 1019: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL). 1020: If inputURL is bigger than that size, outputURL receives 1021: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...", 1022: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL. 1023: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets 1024: copied into outputURL. 1025: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH 1026: chars. 1027: ----------------------------------------------------------------------*/ 1028: #ifdef __STDC__ 1.34 cvs 1029: void ChopURL (char *outputURL, const char *inputURL) 1.24 cvs 1030: #else 1031: void ChopURL (outputURL, inputURL) 1.34 cvs 1032: char *outputURL; 1033: const char *inputURL; 1.24 cvs 1034: #endif 1.22 cvs 1035: 1.24 cvs 1036: { 1037: int len; 1.9 cvs 1038: 1.24 cvs 1039: len = strlen (inputURL); 1040: if (len <= MAX_PRINT_URL_LENGTH) 1.29 cvs 1041: strcpy (outputURL, inputURL); 1.24 cvs 1042: else 1043: /* make a truncated urlName on the status window */ 1044: { 1045: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2); 1046: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS; 1047: strcat (outputURL, "..."); 1048: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ])); 1049: } 1.25 cvs 1050: } 1051: 1052: 1053: /*---------------------------------------------------------------------- 1054: scan 1.47 cvs 1055: Scan a filename for its constituents 1.25 cvs 1056: ----------------------------------- 1057: 1058: On entry, 1059: name points to a document name which may be incomplete. 1060: On exit, 1061: absolute or relative may be nonzero (but not both). 1062: host, fragment and access may be nonzero if they were specified. 1063: Any which are nonzero point to zero terminated strings. 1064: ----------------------------------------------------------------------*/ 1065: #ifdef __STDC__ 1.67 cvs 1066: static void scan (STRING name, HTURI * parts) 1.25 cvs 1067: #else /* __STDC__ */ 1068: static void scan (name, parts) 1.67 cvs 1069: STRING name; 1070: HTURI *parts; 1.25 cvs 1071: 1072: #endif /* __STDC__ */ 1073: { 1.67 cvs 1074: STRING p; 1075: STRING after_access = name; 1.32 cvs 1076: 1.43 cvs 1077: memset (parts, '\0', sizeof (HTURI)); 1.28 cvs 1078: /* Look for fragment identifier */ 1.67 cvs 1079: if ((p = ustrchr(name, TEXT('#'))) != NULL) 1.28 cvs 1080: { 1.67 cvs 1081: *p++ = TEXT('\0'); 1.28 cvs 1082: parts->fragment = p; 1.25 cvs 1083: } 1084: 1.28 cvs 1085: for (p=name; *p; p++) 1086: { 1.67 cvs 1087: if (*p == URL_SEP || *p == DIR_SEP || *p == TEXT('#') || *p == TEXT('?')) 1.28 cvs 1088: break; 1.67 cvs 1089: if (*p == TEXT(':')) 1.28 cvs 1090: { 1091: *p = 0; 1092: parts->access = after_access; /* Scheme has been specified */ 1093: 1094: /* The combination of gcc, the "-O" flag and the HP platform is 1095: unhealthy. The following three lines is a quick & dirty fix, but is 1096: not recommended. Rather, turn off "-O". */ 1097: 1098: /* after_access = p;*/ 1099: /* while (*after_access == 0)*/ 1100: /* after_access++;*/ 1101: after_access = p+1; 1.67 cvs 1102: if (!ustrcasecmp(TEXT("URL"), parts->access)) 1.28 cvs 1103: /* Ignore IETF's URL: pre-prefix */ 1104: parts->access = NULL; 1105: else 1.25 cvs 1106: break; 1107: } 1108: } 1109: 1110: p = after_access; 1.43 cvs 1111: if (*p == URL_SEP || *p == DIR_SEP) 1.28 cvs 1112: { 1.43 cvs 1113: if (p[1] == URL_SEP) 1.28 cvs 1114: { 1.25 cvs 1115: parts->host = p+2; /* host has been specified */ 1.28 cvs 1116: *p = 0; /* Terminate access */ 1117: /* look for end of host name if any */ 1.67 cvs 1118: p = ustrchr (parts->host, URL_SEP); 1.28 cvs 1119: if (p) 1120: { 1.43 cvs 1121: *p = EOS; /* Terminate host */ 1.25 cvs 1122: parts->absolute = p+1; /* Root has been found */ 1.28 cvs 1123: } 1124: } 1125: else 1126: /* Root found but no host */ 1127: parts->absolute = p+1; 1128: } 1129: else 1130: { 1.25 cvs 1131: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */ 1.28 cvs 1132: } 1.25 cvs 1133: } 1134: 1135: 1136: /*---------------------------------------------------------------------- 1.28 cvs 1137: AmayaParseUrl: parse a Name relative to another name 1138: 1139: This returns those parts of a name which are given (and requested) 1140: substituting bits from the related name where necessary. 1.25 cvs 1141: 1.28 cvs 1142: On entry, 1.25 cvs 1143: aName A filename given 1144: relatedName A name relative to which aName is to be parsed. Give 1145: it an empty string if aName is absolute. 1146: wanted A mask for the bits which are wanted. 1147: 1.28 cvs 1148: On exit, 1.25 cvs 1149: returns A pointer to a malloc'd string which MUST BE FREED 1150: ----------------------------------------------------------------------*/ 1151: #ifdef __STDC__ 1.67 cvs 1152: STRING AmayaParseUrl (const STRING aName, STRING relatedName, int wanted) 1.25 cvs 1153: #else /* __STDC__ */ 1.67 cvs 1154: STRING AmayaParseUrl (aName, relatedName, wanted) 1155: const STRING aName; 1156: STRING relatedName; 1.28 cvs 1157: int wanted; 1.25 cvs 1158: 1159: #endif /* __STDC__ */ 1160: { 1.67 cvs 1161: STRING return_value; 1162: CHAR_T result[MAX_LENGTH]; 1163: CHAR_T name[MAX_LENGTH]; 1164: CHAR_T rel[MAX_LENGTH]; 1165: STRING p, access; 1.29 cvs 1166: HTURI given, related; 1167: int len; 1.67 cvs 1168: CHAR_T used_sep; 1169: STRING used_str; 1.32 cvs 1170: 1.67 cvs 1171: if (ustrchr (aName, DIR_SEP) || ustrchr (relatedName, DIR_SEP)) 1.33 cvs 1172: { 1.42 cvs 1173: used_str = DIR_STR; 1174: used_sep = DIR_SEP; 1.33 cvs 1175: } 1.32 cvs 1176: else 1.33 cvs 1177: { 1.42 cvs 1178: used_str = URL_STR; 1179: used_sep = URL_SEP; 1.33 cvs 1180: } 1.32 cvs 1181: 1.29 cvs 1182: /* Make working copies of input strings to cut up: */ 1183: return_value = NULL; 1184: result[0] = 0; /* Clear string */ 1.67 cvs 1185: ustrcpy (name, aName); 1.29 cvs 1186: if (relatedName != NULL) 1.67 cvs 1187: ustrcpy (rel, relatedName); 1.29 cvs 1188: else 1189: relatedName[0] = EOS; 1190: 1191: scan (name, &given); 1192: scan (rel, &related); 1193: access = given.access ? given.access : related.access; 1194: if (wanted & AMAYA_PARSE_ACCESS) 1195: if (access) 1196: { 1.67 cvs 1197: ustrcat (result, access); 1.29 cvs 1198: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1199: ustrcat (result, TEXT(":")); 1.29 cvs 1200: } 1201: 1202: if (given.access && related.access) 1203: /* If different, inherit nothing. */ 1.67 cvs 1204: if (ustrcmp (given.access, related.access) != 0) 1.29 cvs 1205: { 1206: related.host = 0; 1207: related.absolute = 0; 1208: related.relative = 0; 1209: related.fragment = 0; 1210: } 1211: 1212: if (wanted & AMAYA_PARSE_HOST) 1213: if(given.host || related.host) 1214: { 1215: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1216: ustrcat (result, TEXT("//")); 1217: ustrcat (result, given.host ? given.host : related.host); 1.29 cvs 1218: } 1219: 1220: if (given.host && related.host) 1221: /* If different hosts, inherit no path. */ 1.67 cvs 1222: if (ustrcmp (given.host, related.host) != 0) 1.29 cvs 1223: { 1224: related.absolute = 0; 1225: related.relative = 0; 1226: related.fragment = 0; 1227: } 1228: 1229: if (wanted & AMAYA_PARSE_PATH) 1230: { 1231: if (given.absolute) 1232: { 1233: /* All is given */ 1234: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1235: ustrcat (result, used_str); 1236: ustrcat (result, given.absolute); 1.25 cvs 1237: } 1.29 cvs 1238: else if (related.absolute) 1239: { 1240: /* Adopt path not name */ 1.67 cvs 1241: ustrcat (result, used_str); 1242: ustrcat (result, related.absolute); 1.29 cvs 1243: if (given.relative) 1244: { 1245: /* Search part? */ 1.67 cvs 1246: p = ustrchr (result, TEXT('?')); 1.29 cvs 1247: if (!p) 1.67 cvs 1248: p=result+ustrlen(result)-1; 1.33 cvs 1249: for (; *p!=used_sep; p--); /* last / */ 1.29 cvs 1250: /* Remove filename */ 1251: p[1]=0; 1252: /* Add given one */ 1.67 cvs 1253: ustrcat (result, given.relative); 1.25 cvs 1254: } 1255: } 1.29 cvs 1256: else if (given.relative) 1257: /* what we've got */ 1.67 cvs 1258: ustrcat (result, given.relative); 1.29 cvs 1259: else if (related.relative) 1.67 cvs 1260: ustrcat (result, related.relative); 1.29 cvs 1261: else 1262: /* No inheritance */ 1.67 cvs 1263: ustrcat (result, used_str); 1.25 cvs 1264: } 1.29 cvs 1265: 1266: if (wanted & AMAYA_PARSE_ANCHOR) 1267: if (given.fragment || related.fragment) 1268: { 1269: if (given.absolute && given.fragment) 1270: { 1271: /*Fixes for relURLs...*/ 1272: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1273: ustrcat (result, TEXT("#")); 1274: ustrcat (result, given.fragment); 1.29 cvs 1275: } 1276: else if (!(given.absolute) && !(given.fragment)) 1.67 cvs 1277: ustrcat (result, _EMPTYSTR_); 1.29 cvs 1278: else 1279: { 1280: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.67 cvs 1281: ustrcat (result, TEXT("#")); 1282: ustrcat (result, given.fragment ? given.fragment : related.fragment); 1.29 cvs 1283: } 1284: } 1.67 cvs 1285: len = ustrlen (result); 1286: if ((return_value = TtaAllocString (len + 1)) != NULL) 1287: ustrcpy (return_value, result); 1.29 cvs 1288: return (return_value); /* exactly the right length */ 1.25 cvs 1289: } 1290: 1291: /*---------------------------------------------------------------------- 1292: HTCanon 1293: Canonicalizes the URL in the following manner starting from the host 1294: pointer: 1295: 1296: 1) The host name is converted to lowercase 1297: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp) 1298: 1299: Return: OK The position of the current path part of the URL 1300: which might be the old one or a new one. 1301: 1302: ----------------------------------------------------------------------*/ 1303: #ifdef __STDC__ 1.67 cvs 1304: static STRING HTCanon (STRING* filename, STRING host) 1.25 cvs 1305: #else /* __STDC__ */ 1.67 cvs 1306: static STRING HTCanon (filename, host) 1307: STRING *filename; 1308: STRING host; 1.25 cvs 1309: #endif /* __STDC__ */ 1310: { 1.67 cvs 1311: STRING newname = NULL; 1312: STRING port; 1313: STRING strptr; 1314: STRING path; 1315: STRING access = host-3; 1316: CHAR_T used_sep; 1.32 cvs 1317: 1318: 1.67 cvs 1319: if (*filename && ustrchr (*filename, URL_SEP)) 1.33 cvs 1320: { 1321: used_sep = URL_SEP; 1322: } 1323: else 1324: { 1325: used_sep = DIR_SEP; 1326: } 1.32 cvs 1327: 1.33 cvs 1328: while (access>*filename && *(access-1)!= used_sep) /* Find access method */ 1.25 cvs 1329: access--; 1.67 cvs 1330: if ((path = ustrchr(host, used_sep)) == NULL) /* Find path */ 1331: path = host + ustrlen(host); 1332: if ((strptr = ustrchr(host, TEXT('@'))) != NULL && strptr<path) /* UserId */ 1.25 cvs 1333: host = strptr; 1.67 cvs 1334: if ((port = ustrchr(host, TEXT(':'))) != NULL && port>path) /* Port number */ 1.25 cvs 1335: port = NULL; 1336: 1337: strptr = host; /* Convert to lower-case */ 1.33 cvs 1338: while (strptr<path) 1339: { 1.25 cvs 1340: *strptr = tolower(*strptr); 1341: strptr++; 1.33 cvs 1342: } 1.25 cvs 1343: 1344: /* Does the URL contain a full domain name? This also works for a 1345: numerical host name. The domain name is already made lower-case 1346: and without a trailing dot. */ 1347: { 1.67 cvs 1348: STRING dot = port ? port : path; 1349: if (dot > *filename && *--dot == TEXT('.')) 1.33 cvs 1350: { 1.67 cvs 1351: STRING orig=dot, dest=dot+1; 1.33 cvs 1352: while((*orig++ = *dest++)); 1353: if (port) port--; 1354: path--; 1.25 cvs 1355: } 1356: } 1357: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */ 1.33 cvs 1358: if (port) 1359: { 1360: if (!*(port+1) || *(port+1)==used_sep) 1361: { 1362: if (!newname) 1363: { 1.67 cvs 1364: STRING orig=port, dest=port+1; 1.25 cvs 1365: while((*orig++ = *dest++)); 1.33 cvs 1366: } 1367: } 1.67 cvs 1368: else if ((!ustrncmp(access, TEXT("http"), 4) && 1369: (*(port+1) == TEXT('8') && *(port+2) == TEXT('0') && (*(port+3) == used_sep || !*(port + 3)))) || 1370: (!ustrncmp (access, TEXT("gopher"), 6) && 1371: (*(port+1) == TEXT('7') && *(port+2) == TEXT('0') && (*(port+3) == used_sep || !*(port+3)))) || 1372: (!ustrncmp(access, TEXT("ftp"), 3) && 1373: (*(port+1) == TEXT('2') && *(port + 2) == TEXT('1') && (*(port+3) == used_sep || !*(port+3))))) { 1.33 cvs 1374: if (!newname) 1375: { 1.67 cvs 1376: STRING orig=port, dest=port+3; 1.33 cvs 1377: while((*orig++ = *dest++)); 1378: /* Update path position, Henry Minsky */ 1379: path -= 3; 1.25 cvs 1380: } 1.33 cvs 1381: } 1382: else if (newname) 1.67 cvs 1383: ustrncat(newname, port, (int) (path-port)); 1.33 cvs 1384: } 1.25 cvs 1385: 1.33 cvs 1386: if (newname) 1387: { 1.67 cvs 1388: STRING newpath = newname + ustrlen (newname); 1389: ustrcat(newname, path); 1.25 cvs 1390: path = newpath; 1.28 cvs 1391: /* Free old copy */ 1392: TtaFreeMemory(*filename); 1.25 cvs 1393: *filename = newname; 1.33 cvs 1394: } 1.25 cvs 1395: return path; 1396: } 1397: 1398: 1399: /*---------------------------------------------------------------------- 1.29 cvs 1400: SimplifyUrl: simplify a URI 1.32 cvs 1401: A URI is allowed to contain the sequence xxx/../ which may be 1402: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR. 1.28 cvs 1403: Simplification helps us recognize duplicate URIs. 1.25 cvs 1404: 1.28 cvs 1405: Thus, /etc/junk/../fred becomes /etc/fred 1406: /etc/junk/./fred becomes /etc/junk/fred 1.25 cvs 1407: 1.28 cvs 1408: but we should NOT change 1409: http://fred.xxx.edu/../.. 1.25 cvs 1410: 1411: or ../../albert.html 1412: 1.28 cvs 1413: In order to avoid empty URLs the following URLs become: 1.25 cvs 1414: 1415: /fred/.. becomes /fred/.. 1416: /fred/././.. becomes /fred/.. 1417: /fred/.././junk/.././ becomes /fred/.. 1418: 1.28 cvs 1419: If more than one set of `://' is found (several proxies in cascade) then 1420: only the part after the last `://' is simplified. 1.25 cvs 1421: 1.28 cvs 1422: Returns: A string which might be the old one or a new one. 1.25 cvs 1423: ----------------------------------------------------------------------*/ 1424: #ifdef __STDC__ 1.67 cvs 1425: void SimplifyUrl (STRING* url) 1.25 cvs 1426: #else /* __STDC__ */ 1.29 cvs 1427: void SimplifyUrl (url) 1.67 cvs 1428: STRING *url; 1.25 cvs 1429: #endif /* __STDC__ */ 1430: { 1.67 cvs 1431: STRING path, p; 1432: STRING newptr, access; 1433: STRING orig, dest, end; 1.28 cvs 1434: 1.67 cvs 1435: CHAR_T used_sep; 1.32 cvs 1436: 1437: 1.28 cvs 1438: if (!url || !*url) 1439: return; 1440: 1.67 cvs 1441: if (ustrchr (*url, URL_SEP)) 1.33 cvs 1442: { 1443: used_sep = URL_SEP; 1444: } 1.32 cvs 1445: else 1.33 cvs 1446: { 1447: used_sep = DIR_SEP; 1448: } 1.32 cvs 1449: 1.28 cvs 1450: /* Find any scheme name */ 1.67 cvs 1451: if ((path = ustrstr(*url, TEXT("://"))) != NULL) 1.33 cvs 1452: { 1453: /* Find host name */ 1.28 cvs 1454: access = *url; 1455: while (access<path && (*access=tolower(*access))) 1456: access++; 1457: path += 3; 1.67 cvs 1458: while ((newptr = ustrstr(path, TEXT("://"))) != NULL) 1.28 cvs 1459: /* For proxies */ 1460: path = newptr+3; 1461: /* We have a host name */ 1462: path = HTCanon(url, path); 1.25 cvs 1463: } 1.67 cvs 1464: else if ((path = ustrstr(*url, TEXT(":/"))) != NULL) 1.28 cvs 1465: path += 2; 1466: else 1467: path = *url; 1.25 cvs 1468: 1.33 cvs 1469: if (*path == used_sep && *(path+1)==used_sep) 1.28 cvs 1470: /* Some URLs start //<foo> */ 1471: path += 1; 1.67 cvs 1472: else if (!ustrncmp(path, TEXT("news:"), 5)) 1.28 cvs 1473: { 1.67 cvs 1474: newptr = ustrchr(path+5, TEXT('@')); 1.28 cvs 1475: if (!newptr) 1476: newptr = path + 5; 1477: while (*newptr) 1478: { 1479: /* Make group or host lower case */ 1480: *newptr = tolower (*newptr); 1481: newptr++; 1.25 cvs 1482: } 1.28 cvs 1483: /* Doesn't need to do any more */ 1484: return; 1.25 cvs 1485: } 1.28 cvs 1486: 1487: if ((p = path)) 1488: { 1.67 cvs 1489: if (!((end = ustrchr (path, TEXT(';'))) || (end = ustrchr (path, TEXT('?'))) || 1490: (end = ustrchr (path, TEXT('#'))))) 1491: end = path + ustrlen (path); 1.28 cvs 1492: 1493: /* Parse string second time to simplify */ 1494: p = path; 1495: while (p < end) 1496: { 1.33 cvs 1497: if (*p==used_sep) 1.28 cvs 1498: { 1.67 cvs 1499: if (p > *url && *(p+1) == TEXT('.') && (*(p+2) == used_sep || !*(p+2))) 1.28 cvs 1500: { 1501: orig = p + 1; 1.33 cvs 1502: dest = (*(p+2)!=used_sep) ? p+2 : p+3; 1.52 cvs 1503: while ((*orig++ = *dest++)); /* Remove a used_sep and a dot*/ 1.28 cvs 1504: end = orig - 1; 1505: } 1.67 cvs 1506: else if (*(p+1) == TEXT('.') && *(p+2) == TEXT('.') && (*(p+3) == used_sep || !*(p+3))) 1.28 cvs 1507: { 1508: newptr = p; 1.52 cvs 1509: while (newptr>path && *--newptr!=used_sep); /* prev used_sep */ 1510: if (*newptr == used_sep) 1511: orig = newptr + 1; 1.28 cvs 1512: else 1.52 cvs 1513: orig = newptr; 1514: 1515: dest = (*(p+3) != used_sep) ? p+3 : p+4; 1516: while ((*orig++ = *dest++)); /* Remove /xxx/.. */ 1517: end = orig-1; 1518: /* Start again with prev slash */ 1519: p = newptr; 1.28 cvs 1520: } 1.33 cvs 1521: else if (*(p+1) == used_sep) 1.28 cvs 1522: { 1.33 cvs 1523: while (*(p+1) == used_sep) 1.28 cvs 1524: { 1525: orig = p; 1526: dest = p + 1; 1527: while ((*orig++ = *dest++)); /* Remove multiple /'s */ 1528: end = orig-1; 1529: } 1530: } 1531: else 1.25 cvs 1532: p++; 1.28 cvs 1533: } 1534: else 1535: p++; 1.25 cvs 1536: } 1537: } 1.51 cvs 1538: 1539: /* 1540: ** Check for host/../.. kind of things 1541: */ 1.67 cvs 1542: if (*path == used_sep && *(path+1) == TEXT('.') && *(path+2) == TEXT('.') && (!*(path+3) || *(path+3) == used_sep)) 1.51 cvs 1543: *(path+1) = EOS; 1544: 1.28 cvs 1545: return; 1546: } 1547: 1548: 1549: /*---------------------------------------------------------------------- 1550: NormalizeFile normalizes local names. 1551: Return TRUE if target and src differ. 1552: ----------------------------------------------------------------------*/ 1553: #ifdef __STDC__ 1.67 cvs 1554: ThotBool NormalizeFile (STRING src, STRING target) 1.28 cvs 1555: #else 1.67 cvs 1556: ThotBool NormalizeFile (src, target) 1557: STRING src; 1558: STRING target; 1.28 cvs 1559: 1560: #endif 1561: { 1.67 cvs 1562: STRING s; 1.66 cvs 1563: ThotBool change; 1.28 cvs 1564: 1.54 cvs 1565: change = FALSE; 1.67 cvs 1566: if (ustrncmp (src, TEXT("file:"), 5) == 0) 1.28 cvs 1567: { 1568: /* remove the prefix file: */ 1569: if (src[5] == EOS) 1.67 cvs 1570: ustrcpy (target, DIR_STR); 1571: else if (src[0] == TEXT('~')) 1.28 cvs 1572: { 1573: /* replace ~ */ 1.67 cvs 1574: s = TtaGetEnvString (TEXT("HOME")); 1575: ustrcpy (target, s); 1576: ustrcat (target, &src[5]); 1.28 cvs 1577: } 1578: else 1.67 cvs 1579: ustrcpy (target, &src[5]); 1.54 cvs 1580: change = TRUE; 1.28 cvs 1581: } 1.53 cvs 1582: # ifndef _WINDOWS 1.67 cvs 1583: else if (src[0] == TEXT('~')) 1.53 cvs 1584: { 1585: /* replace ~ */ 1586: s = (char *) TtaGetEnvString ("HOME"); 1587: strcpy (target, s); 1588: if (src[1] != DIR_SEP) 1589: strcat (target, DIR_STR); 1590: strcat (target, &src[1]); 1.54 cvs 1591: change = TRUE; 1.53 cvs 1592: } 1593: # endif /* _WINDOWS */ 1.28 cvs 1594: else 1.67 cvs 1595: ustrcpy (target, src); 1.28 cvs 1596: 1597: /* remove /../ and /./ */ 1.29 cvs 1598: SimplifyUrl (&target); 1.54 cvs 1599: if (!change) 1.67 cvs 1600: change = ustrcmp (src, target); 1.28 cvs 1601: return (change); 1.25 cvs 1602: } 1603: 1.28 cvs 1604: 1.25 cvs 1605: /*---------------------------------------------------------------------- 1.31 cvs 1606: MakeRelativeURL: make relative name 1.25 cvs 1607: 1.28 cvs 1608: This function creates and returns a string which gives an expression of 1609: one address as related to another. Where there is no relation, an absolute 1610: address is retured. 1.25 cvs 1611: 1.28 cvs 1612: On entry, 1.25 cvs 1613: Both names must be absolute, fully qualified names of nodes 1614: (no fragment bits) 1615: 1.28 cvs 1616: On exit, 1.25 cvs 1617: The return result points to a newly allocated name which, if 1618: parsed by AmayaParseUrl relative to relatedName, will yield aName. 1619: The caller is responsible for freeing the resulting name later. 1620: ----------------------------------------------------------------------*/ 1621: #ifdef __STDC__ 1.67 cvs 1622: STRING MakeRelativeURL (STRING aName, STRING relatedName) 1.25 cvs 1623: #else /* __STDC__ */ 1.67 cvs 1624: STRING MakeRelativeURL (aName, relatedName) 1625: STRING aName; 1626: STRING relatedName; 1.25 cvs 1627: #endif /* __STDC__ */ 1628: { 1.67 cvs 1629: STRING return_value; 1630: CHAR_T result[MAX_LENGTH]; 1631: STRING p; 1632: STRING q; 1633: STRING after_access; 1634: STRING last_slash = NULL; 1.29 cvs 1635: int slashes, levels, len; 1636: 1.44 cvs 1637: # ifdef _WINDOWS 1638: int ndx; 1639: # endif /* _WINDOWS */ 1640: 1.29 cvs 1641: if (aName == NULL || relatedName == NULL) 1642: return (NULL); 1643: 1644: slashes = 0; 1645: after_access = NULL; 1646: p = aName; 1647: q = relatedName; 1648: for (; *p && (*p == *q); p++, q++) 1.27 cvs 1649: { 1650: /* Find extent of match */ 1.67 cvs 1651: if (*p == TEXT(':')) 1.29 cvs 1652: after_access = p + 1; 1.28 cvs 1653: if (*p == DIR_SEP) 1.27 cvs 1654: { 1.29 cvs 1655: /* memorize the last slash position and count them */ 1.27 cvs 1656: last_slash = p; 1657: slashes++; 1.25 cvs 1658: } 1659: } 1660: 1.31 cvs 1661: /* q, p point to the first non-matching character or zero */ 1662: if (*q == EOS) 1663: { 1664: /* New name is a subset of the related name */ 1665: /* exactly the right length */ 1.67 cvs 1666: len = ustrlen (p); 1667: if ((return_value = TtaAllocString (len + 1)) != NULL) 1668: ustrcpy (return_value, p); 1.31 cvs 1669: } 1670: else if ((slashes < 2 && after_access == NULL) 1671: || (slashes < 3 && after_access != NULL)) 1672: { 1673: /* Two names whitout common path */ 1674: /* exactly the right length */ 1.67 cvs 1675: len = ustrlen (aName); 1676: if ((return_value = TtaAllocString (len + 1)) != NULL) 1677: ustrcpy (return_value, aName); 1.31 cvs 1678: } 1679: else 1680: { 1681: /* Some path in common */ 1.67 cvs 1682: if (slashes == 3 && ustrncmp (aName, TEXT("http:"), 5) == 0) 1.31 cvs 1683: /* just the same server */ 1.67 cvs 1684: ustrcpy (result, last_slash); 1.31 cvs 1685: else 1686: { 1687: levels= 0; 1.67 cvs 1688: for (; *q && *q != TEXT('#') && *q != TEXT(';') && *q != TEXT('?'); q++) 1.31 cvs 1689: if (*q == DIR_SEP) 1690: levels++; 1691: 1.52 cvs 1692: result[0] = EOS; 1.31 cvs 1693: for (;levels; levels--) 1.67 cvs 1694: ustrcat (result, TEXT("../")); 1695: ustrcat (result, last_slash+1); 1.31 cvs 1696: } 1.52 cvs 1697: 1698: if (!*result) 1.67 cvs 1699: ustrcat (result, TEXT("./")); 1.52 cvs 1700: 1.31 cvs 1701: /* exactly the right length */ 1.67 cvs 1702: len = ustrlen (result); 1703: if ((return_value = TtaAllocString (len + 1)) != NULL) 1704: ustrcpy (return_value, result); 1.52 cvs 1705: 1.25 cvs 1706: } 1.44 cvs 1707: # ifdef _WINDOWS 1.67 cvs 1708: len = ustrlen (return_value); 1.44 cvs 1709: for (ndx = 0; ndx < len; ndx ++) 1.67 cvs 1710: if (return_value[ndx] == TEXT('\\')) 1711: return_value[ndx] = TEXT('/') ; 1.44 cvs 1712: # endif /* _WINDOWS */ 1.29 cvs 1713: return (return_value); 1.24 cvs 1714: } 1.35 cvs 1715: 1716: