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 { 26: char * access; /* Now known as "scheme" */ 27: char * host; 28: char * absolute; 29: char * relative; 30: char * fragment; 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__ 1.38 cvs 39: void ConvertToLowerCase (char *string) 1.28 cvs 40: #else /* __STDC__ */ 1.38 cvs 41: void ConvertToLowerCase (string) 1.28 cvs 42: char *string; 43: 44: #endif /* __STDC__ */ 45: { 46: int i; 47: 48: if (!string) 49: return; 50: 51: for (i = 0; string[i] != EOS; i++) 52: string[i] = tolower (string[i]); 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 */ 1.9 cvs 93: while ((*curr != 0) && (*curr != SPACE) && (*curr != '\b') && 94: (*curr != '\r') && (*curr != EOL)) 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: 1.32 cvs 146: if (*curr == ':') 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: 1.4 cvs 177: /*---------------------------------------------------------------------- 1.9 cvs 178: IsHTMLName 179: returns TRUE if path points to an HTML resource. 1.4 cvs 180: ----------------------------------------------------------------------*/ 1.3 cvs 181: #ifdef __STDC__ 1.34 cvs 182: boolean IsHTMLName (const char *path) 1.3 cvs 183: #else /* __STDC__ */ 184: boolean IsHTMLName (path) 1.34 cvs 185: const char *path; 1.3 cvs 186: #endif /* __STDC__ */ 187: { 1.5 cvs 188: char temppath[MAX_LENGTH]; 189: char suffix[MAX_LENGTH]; 190: char nsuffix[MAX_LENGTH]; 191: int i; 192: 193: if (!path) 1.37 cvs 194: return (FALSE); 1.5 cvs 195: 196: strcpy (temppath, path); 197: ExtractSuffix (temppath, suffix); 198: 199: /* Normalize the suffix */ 200: i = 0; 1.39 cvs 201: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 202: { 1.25 cvs 203: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 204: i++; 205: } 1.5 cvs 206: nsuffix[i] = EOS; 1.39 cvs 207: if ((!strcmp (nsuffix, "html")) || 208: (!strcmp (nsuffix, "htm")) || 209: (!strcmp (nsuffix, "shtml"))) 210: return (TRUE); 1.22 cvs 211: else if (!strcmp (nsuffix, "gz")) 1.13 cvs 212: { 1.39 cvs 213: /* take into account compressed files */ 1.13 cvs 214: ExtractSuffix (temppath, suffix); 215: /* Normalize the suffix */ 216: i = 0; 1.39 cvs 217: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 218: { 1.25 cvs 219: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 220: i++; 221: } 222: nsuffix[i] = EOS; 1.39 cvs 223: if ((!strcmp (nsuffix, "html")) || 224: (!strcmp (nsuffix, "htm")) || 225: (!strcmp (nsuffix, "shtml"))) 226: return (TRUE); 227: else 1.13 cvs 228: return (FALSE); 229: } 230: else 1.39 cvs 231: return (FALSE); 1.3 cvs 232: } 233: 1.4 cvs 234: /*---------------------------------------------------------------------- 1.9 cvs 235: IsImageName 236: returns TRUE if path points to an image resource. 1.4 cvs 237: ----------------------------------------------------------------------*/ 1.3 cvs 238: #ifdef __STDC__ 1.34 cvs 239: boolean IsImageName (const char *path) 1.3 cvs 240: #else /* __STDC__ */ 241: boolean IsImageName (path) 1.34 cvs 242: const char *path; 1.3 cvs 243: #endif /* __STDC__ */ 244: { 1.5 cvs 245: char temppath[MAX_LENGTH]; 246: char suffix[MAX_LENGTH]; 247: char nsuffix[MAX_LENGTH]; 248: int i; 249: 250: if (!path) 1.13 cvs 251: return (FALSE); 1.5 cvs 252: 253: strcpy (temppath, path); 254: ExtractSuffix (temppath, suffix); 255: 256: /* Normalize the suffix */ 257: i = 0; 1.39 cvs 258: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 259: { 1.25 cvs 260: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 261: i++; 262: } 1.5 cvs 263: nsuffix[i] = EOS; 1.39 cvs 264: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) || 265: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) || 266: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au"))) 267: return (TRUE); 268: return (FALSE); 1.3 cvs 269: } 270: 1.4 cvs 271: /*---------------------------------------------------------------------- 1.9 cvs 272: IsTextName 1.4 cvs 273: ----------------------------------------------------------------------*/ 1.3 cvs 274: #ifdef __STDC__ 1.34 cvs 275: boolean IsTextName (const char *path) 1.3 cvs 276: #else /* __STDC__ */ 277: boolean IsTextName (path) 1.34 cvs 278: const char *path; 1.3 cvs 279: 280: #endif /* __STDC__ */ 281: { 1.5 cvs 282: char temppath[MAX_LENGTH]; 283: char suffix[MAX_LENGTH]; 284: char nsuffix[MAX_LENGTH]; 285: int i; 286: 287: if (!path) 1.13 cvs 288: return (FALSE); 1.5 cvs 289: 290: strcpy (temppath, path); 291: ExtractSuffix (temppath, suffix); 292: 293: /* Normalize the suffix */ 294: i = 0; 1.39 cvs 295: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.5 cvs 296: { 1.25 cvs 297: nsuffix[i] = tolower (suffix[i]); 1.5 cvs 298: i++; 299: } 300: nsuffix[i] = EOS; 301: 1.39 cvs 302: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd"))) 1.13 cvs 303: return (TRUE); 1.22 cvs 304: else if (!strcmp (nsuffix, "gz")) 1.13 cvs 305: { 1.39 cvs 306: /* take into account compressed files */ 1.13 cvs 307: ExtractSuffix (temppath, suffix); 308: /* Normalize the suffix */ 309: i = 0; 1.39 cvs 310: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 311: { 1.25 cvs 312: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 313: i++; 314: } 315: nsuffix[i] = EOS; 1.39 cvs 316: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd"))) 1.13 cvs 317: return (TRUE); 318: else 319: return (FALSE); 320: } 321: else 322: return (FALSE); 1.3 cvs 323: } 324: 1.4 cvs 325: /*---------------------------------------------------------------------- 1.9 cvs 326: IsHTTPPath 327: returns TRUE if path is in fact an http URL. 1.4 cvs 328: ----------------------------------------------------------------------*/ 1.3 cvs 329: #ifdef __STDC__ 1.34 cvs 330: boolean IsHTTPPath (const char *path) 1.3 cvs 331: #else /* __STDC__ */ 332: boolean IsHTTPPath (path) 1.34 cvs 333: const char *path; 1.3 cvs 334: #endif /* __STDC__ */ 335: { 1.5 cvs 336: if (!path) 337: return FALSE; 1.3 cvs 338: 1.5 cvs 339: if (strncmp (path, "http:", 5) != 0) 340: return FALSE; 341: return TRUE; 1.3 cvs 342: } 343: 1.4 cvs 344: /*---------------------------------------------------------------------- 1.9 cvs 345: IsWithParameters 346: returns TRUE if url has a concatenated query string. 1.4 cvs 347: ----------------------------------------------------------------------*/ 1.3 cvs 348: #ifdef __STDC__ 1.34 cvs 349: boolean IsWithParameters (const char *url) 1.3 cvs 350: #else /* __STDC__ */ 1.9 cvs 351: boolean IsWithParameters (url) 1.34 cvs 352: const char *url; 1.3 cvs 353: #endif /* __STDC__ */ 354: { 1.5 cvs 355: int i; 1.3 cvs 356: 1.9 cvs 357: if ((!url) || (url[0] == EOS)) 1.5 cvs 358: return FALSE; 1.3 cvs 359: 1.9 cvs 360: i = strlen (url) - 1; 361: while (i > 0 && url[i--] != '?') 1.5 cvs 362: if (i < 0) 363: return FALSE; 1.3 cvs 364: 1.5 cvs 365: /* There is a parameter */ 366: return TRUE; 1.3 cvs 367: } 368: 1.4 cvs 369: /*---------------------------------------------------------------------- 1.9 cvs 370: IsW3Path 371: returns TRUE if path is in fact a URL. 1.4 cvs 372: ----------------------------------------------------------------------*/ 1.3 cvs 373: #ifdef __STDC__ 1.34 cvs 374: boolean IsW3Path (const char *path) 1.3 cvs 375: #else /* __STDC__ */ 376: boolean IsW3Path (path) 1.34 cvs 377: const char *path; 1.3 cvs 378: #endif /* __STDC__ */ 379: { 1.5 cvs 380: if ((strncmp (path, "http:", 5)) && (strncmp (path, "ftp:", 4)) && 381: (strncmp (path, "telnet:", 7)) && (strncmp (path, "wais:", 5)) && 382: (strncmp (path, "news:", 5)) && (strncmp (path, "gopher:", 7)) && 383: (strncmp (path, "mailto:", 7)) && (strncmp (path, "archie:", 7))) 384: return FALSE; 385: return TRUE; 1.3 cvs 386: } 387: 1.4 cvs 388: /*---------------------------------------------------------------------- 1.9 cvs 389: IsValidProtocol 390: returns true if the url protocol is supported by Amaya. 1.4 cvs 391: ----------------------------------------------------------------------*/ 1.3 cvs 392: #ifdef __STDC__ 1.34 cvs 393: boolean IsValidProtocol (const char *url) 1.3 cvs 394: #else /* __STDC__ */ 1.9 cvs 395: boolean IsValidProtocol (url) 1.34 cvs 396: const char *url; 1.3 cvs 397: #endif /* __STDC__ */ 398: { 1.26 cvs 399: if (!strncmp (url, "http:", 5)) 1.22 cvs 400: /* experimental */ 1.26 cvs 401: /*** || !strncmp (url, "ftp:", 4)) ***/ 1.24 cvs 402: /*** || !strncmp (path, "news:", 5)***/ 1.8 cvs 403: return (TRUE); 1.5 cvs 404: else 1.8 cvs 405: return (FALSE); 1.3 cvs 406: } 407: 1.31 cvs 408: 409: /*---------------------------------------------------------------------- 410: GetBaseURL 411: normalizes orgName according to a base associated with doc, and 412: following the standard URL format rules. 413: The function returns the base used to solve relative URL and SRC: 414: - the base of the document, 415: - or the document path (without document name). 416: ----------------------------------------------------------------------*/ 417: #ifdef __STDC__ 418: char *GetBaseURL (Document doc) 419: #else /* __STDC__ */ 420: char *GetBaseURL (doc) 421: Document doc; 422: #endif /* __STDC__ */ 423: { 424: Element el; 425: ElementType elType; 426: AttributeType attrType; 427: Attribute attr; 428: char *ptr, *basename; 429: int length; 430: 431: basename = TtaGetMemory (MAX_LENGTH); 1.39 cvs 432: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1); 433: basename[MAX_LENGTH-1] = EOS; 1.31 cvs 434: length = MAX_LENGTH -1; 435: /* get the root element */ 436: el = TtaGetMainRoot (doc); 437: /* search the BASE element */ 438: elType.ElSSchema = TtaGetDocumentSSchema (doc); 439: elType.ElTypeNum = HTML_EL_BASE; 440: el = TtaSearchTypedElement (elType, SearchInTree, el); 441: if (el) 442: { 443: /* The document has a BASE element -> Get the HREF attribute */ 444: attrType.AttrSSchema = elType.ElSSchema; 445: attrType.AttrTypeNum = HTML_ATTR_HREF_; 446: attr = TtaGetAttribute (el, attrType); 447: if (attr) 448: { 449: /* Use the base path of the document */ 450: TtaGiveTextAttributeValue (attr, basename, &length); 451: /* base and orgName have to be separated by a DIR_SEP */ 452: length--; 1.43 cvs 453: if (basename[0] != EOS && basename[length] != URL_SEP && basename[length] != DIR_SEP) 1.31 cvs 454: /* verify if the base has the form "protocol://server:port" */ 455: { 1.33 cvs 456: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS | 457: AMAYA_PARSE_HOST | 458: AMAYA_PARSE_PUNCTUATION); 1.31 cvs 459: if (ptr && !strcmp (ptr, basename)) 460: { 1.43 cvs 461: /* it has this form, we complete it by adding a URL_STR */ 462: if (strchr (basename, DIR_SEP)) 463: strcat (basename, DIR_STR); 464: else 465: strcat (basename, URL_STR); 1.31 cvs 466: length++; 467: } 468: if (ptr) 469: TtaFreeMemory (ptr); 470: } 471: } 1.33 cvs 472: } 473: 1.31 cvs 474: /* Remove anything after the last DIR_SEP char. If no such char is found, 475: * then search for the first ":" char, hoping that what's before that is a 476: * protocol. If found, end the string there. If neither char is found, 477: * then discard the whole base element. 478: */ 479: length = strlen (basename) - 1; 480: /* search for the last DIR_SEP char */ 1.43 cvs 481: while (length >= 0 && basename[length] != URL_SEP && basename[length] != DIR_SEP) 1.31 cvs 482: length--; 483: if (length >= 0) 484: /* found the last DIR_SEP char, end the string there */ 485: basename[length + 1] = EOS; 486: else 487: /* search for the first PATH_STR char */ 488: { 1.32 cvs 489: for (length = 0; basename[length] != ':' && 1.31 cvs 490: basename[length] != EOS; length ++); 1.32 cvs 491: if (basename[length] == ':') 1.31 cvs 492: /* found, so end the string there */ 493: basename[length + 1] = EOS; 494: else 495: /* not found, discard the base */ 496: basename[0] = EOS; 497: } 498: return (basename); 499: } 500: 501: 1.4 cvs 502: /*---------------------------------------------------------------------- 1.40 cvs 503: GetLocalPath 504: Allocate and return the local document path associated to the url 505: ----------------------------------------------------------------------*/ 506: #ifdef __STDC__ 507: char *GetLocalPath (Document doc, char *url) 508: #else /* __STDC__ */ 509: char *GetLocalPath (doc, url) 510: Document doc; 511: char *url; 512: #endif /* __STDC__ */ 513: { 1.46 cvs 514: char *ptr, *n; 1.40 cvs 515: char *documentname; 1.41 cvs 516: char url_sep; 1.40 cvs 517: int len; 518: boolean noFile; 519: 520: if (url != NULL) 521: { 522: /* check whether the file name exists */ 523: len = strlen (url) - 1; 1.41 cvs 524: if (IsW3Path) 525: url_sep = '/'; 526: else 527: url_sep = DIR_SEP; 528: noFile = (url[len] == url_sep); 1.40 cvs 529: if (noFile) 530: url[len] = EOS; 1.47 cvs 531: ptr = TtaGetMemory (MAX_LENGTH); 532: documentname = TtaGetMemory (MAX_LENGTH); 1.40 cvs 533: TtaExtractName (url, ptr, documentname); 534: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR); 535: if (!TtaCheckDirectory (ptr)) 536: /* directory did not exist */ 537: mkdir (ptr, S_IRWXU); 1.47 cvs 538: 539: /* don't include the query string within document name */ 540: n = strrchr(documentname, '?'); 541: if (n != NULL) 542: *n = EOS; 1.46 cvs 543: /* don't include ':' within document name */ 544: n = strchr (documentname, ':'); 545: if (n != NULL) 546: *n = EOS; 1.40 cvs 547: strcat (ptr, documentname); 548: TtaFreeMemory (documentname); 549: /* restore the url */ 550: if (noFile) 1.41 cvs 551: url[len] = url_sep; 1.40 cvs 552: return (ptr); 553: } 554: else 555: return (NULL); 556: } 557: 558: 559: /*---------------------------------------------------------------------- 1.9 cvs 560: NormalizeURL 561: normalizes orgName according to a base associated with doc, and 562: following the standard URL format rules. 563: The function returns the new complete and normalized URL 1.12 cvs 564: or file name path (newName) and the name of the document (docName). 1.9 cvs 565: N.B. If the function can't find out what's the docName, it assigns 566: the name "noname.html". 1.4 cvs 567: ----------------------------------------------------------------------*/ 1.3 cvs 568: #ifdef __STDC__ 569: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName) 570: #else /* __STDC__ */ 571: void NormalizeURL (orgName, doc, newName, docName) 572: char *orgName; 573: Document doc; 574: char *newName; 575: char *docName; 576: #endif /* __STDC__ */ 577: { 1.31 cvs 578: char *basename; 1.18 cvs 579: char tempOrgName[MAX_LENGTH]; 1.5 cvs 580: char *ptr; 581: int length; 582: 1.44 cvs 583: # ifdef _WINDOWS 584: int ndx; 585: # endif /* _WINDOWS */ 586: 1.5 cvs 587: if (!newName || !docName) 588: return; 1.18 cvs 589: 1.32 cvs 590: if (doc != 0) 1.33 cvs 591: basename = GetBaseURL (doc); 1.32 cvs 592: else 1.33 cvs 593: basename = (char *) NULL; 1.32 cvs 594: 1.18 cvs 595: /* 1.31 cvs 596: * Clean orgName 597: * Make sure we have a complete orgName, without any leading or trailing 598: * white spaces, or trailinbg new lines 599: */ 1.5 cvs 600: ptr = orgName; 1.18 cvs 601: /* skip leading white space and new line characters */ 1.19 cvs 602: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS); 1.39 cvs 603: strncpy (tempOrgName, ptr, MAX_LENGTH -1); 604: tempOrgName[MAX_LENGTH -1] = EOS; 1.18 cvs 605: /* clean trailing white space */ 1.48 ! cvs 606: # ifndef _WINDOWS 1.18 cvs 607: ptr = strchr (tempOrgName, ' '); 608: if (ptr) 609: *ptr = EOS; 1.48 ! cvs 610: # endif /* !_WINDOWS */ 1.18 cvs 611: /* clean trailing new lines */ 1.19 cvs 612: ptr = strchr (tempOrgName, EOL); 1.5 cvs 613: if (ptr) 614: *ptr = EOS; 615: 1.18 cvs 616: /* 1.31 cvs 617: * Make orgName a complete URL 618: * If the URL does not include a protocol, then try to calculate 619: * one using the doc's base element (if it exists), 620: */ 1.21 cvs 621: if (tempOrgName[0] == EOS) 622: { 623: newName[0] = EOS; 1.36 cvs 624: TtaFreeMemory (basename); 1.21 cvs 625: return; 626: } 627: else if (IsW3Path (tempOrgName)) 628: { 629: /* the name is complete, go to the Sixth Step */ 630: strcpy (newName, tempOrgName); 631: /* verify if the URL has the form "protocol://server:port" */ 1.25 cvs 632: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | 633: AMAYA_PARSE_PUNCTUATION); 1.21 cvs 634: if (ptr && !strcmp (ptr, newName)) 1.43 cvs 635: /* it has this form, we complete it by adding a DIR_STR */ 636: strcat (newName, URL_STR); 637: 1.21 cvs 638: if (ptr) 1.25 cvs 639: TtaFreeMemory (ptr); 1.21 cvs 640: } 641: else if ( doc == 0) 1.19 cvs 642: /* the name is complete, go to the Sixth Step */ 1.18 cvs 643: strcpy (newName, tempOrgName); 1.5 cvs 644: else 645: { 1.31 cvs 646: /* Calculate the absolute URL, using the base or document URL */ 1.44 cvs 647: # ifdef _WINDOWS 1.45 cvs 648: if (!IsW3Path (basename)) { 649: length = strlen (tempOrgName); 650: for (ndx = 0; ndx < length; ndx++) 651: if (tempOrgName [ndx] == '/') 652: tempOrgName [ndx] = '\\'; 653: } 1.44 cvs 654: # endif /* _WINDOWS */ 1.25 cvs 655: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL); 1.14 cvs 656: if (ptr) 657: { 1.29 cvs 658: SimplifyUrl (&ptr); 1.14 cvs 659: strcpy (newName, ptr); 1.25 cvs 660: TtaFreeMemory (ptr); 1.14 cvs 661: } 662: else 1.18 cvs 663: newName[0] = EOS; 1.5 cvs 664: } 1.36 cvs 665: 666: TtaFreeMemory (basename); 1.18 cvs 667: /* 1.31 cvs 668: * Prepare the docname that will refer to this ressource in the 669: * .amaya directory. If the new URL finishes on DIR_SEP, then use 670: * noname.html as a default ressource name 1.18 cvs 671: */ 1.19 cvs 672: if (newName[0] != EOS) 1.5 cvs 673: { 1.19 cvs 674: length = strlen (newName) - 1; 1.43 cvs 675: if (newName[length] == URL_SEP || newName[length] == DIR_SEP) 1.18 cvs 676: { 677: /* docname was not comprised inside the URL, so let's */ 678: /* assign the default ressource name */ 679: strcpy (docName, "noname.html"); 680: /* remove DIR_SEP at the end of complete path */ 1.23 cvs 681: /* newName[length] = EOS; */ 1.18 cvs 682: } 1.14 cvs 683: else 1.18 cvs 684: { 685: /* docname is comprised inside the URL */ 1.43 cvs 686: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP) 1.18 cvs 687: length--; 688: if (length < 0) 689: strcpy (docName, newName); 690: else 691: strcpy (docName, &newName[length+1]); 692: } 1.5 cvs 693: } 1.18 cvs 694: else 695: docName[0] = EOS; 696: } 1.3 cvs 697: 1.4 cvs 698: /*---------------------------------------------------------------------- 1.9 cvs 699: IsSameHost 1.4 cvs 700: ----------------------------------------------------------------------*/ 1.3 cvs 701: #ifdef __STDC__ 1.34 cvs 702: boolean IsSameHost (const char *url1, const char *url2) 1.3 cvs 703: #else /* __STDC__ */ 704: boolean IsSameHost (url1, url2) 1.34 cvs 705: const char *url1; 706: const char *url2; 1.3 cvs 707: #endif /* __STDC__ */ 708: { 1.34 cvs 709: char *basename_ptr1, *basename_ptr2; 710: boolean result; 1.3 cvs 711: 1.25 cvs 712: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 713: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 1.3 cvs 714: 1.5 cvs 715: if (strcmp (basename_ptr1, basename_ptr2)) 1.8 cvs 716: result = FALSE; 1.5 cvs 717: else 1.8 cvs 718: result = TRUE; 1.3 cvs 719: 1.25 cvs 720: TtaFreeMemory (basename_ptr1); 721: TtaFreeMemory (basename_ptr2); 1.5 cvs 722: return (result); 1.3 cvs 723: } 724: 725: 1.4 cvs 726: /*---------------------------------------------------------------------- 1.22 cvs 727: HasKnownFileSuffix 728: returns TRUE if path points to a file ending with a suffix. 729: ----------------------------------------------------------------------*/ 730: #ifdef __STDC__ 1.34 cvs 731: boolean HasKnownFileSuffix (const char *path) 1.22 cvs 732: #else /* __STDC__ */ 733: boolean HasKnownFileSuffix (path) 1.34 cvs 734: const char *path; 1.22 cvs 735: #endif /* __STDC__ */ 736: { 1.29 cvs 737: char *root; 738: char temppath[MAX_LENGTH]; 739: char suffix[MAX_LENGTH]; 1.22 cvs 740: 1.24 cvs 741: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP) 1.22 cvs 742: return (FALSE); 743: 1.29 cvs 744: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION); 1.22 cvs 745: 746: if (root) 747: { 748: strcpy (temppath, root); 1.25 cvs 749: TtaFreeMemory (root); 1.22 cvs 750: /* Get the suffix */ 751: ExtractSuffix (temppath, suffix); 752: 753: if( suffix[0] == EOS) 754: /* no suffix */ 755: return (FALSE); 756: 757: /* Normalize the suffix */ 758: ConvertToLowerCase (suffix); 759: 1.23 cvs 760: if (!strcmp (suffix, "gz")) 1.22 cvs 761: /* skip the compressed suffix */ 762: { 763: ExtractSuffix (temppath, suffix); 764: if(suffix[0] == EOS) 765: /* no suffix */ 766: return (FALSE); 767: /* Normalize the suffix */ 768: ConvertToLowerCase (suffix); 769: } 770: 771: if ((strcmp (suffix, "gif")) && (strcmp (suffix, "xbm")) && 772: (strcmp (suffix, "xpm")) && (strcmp (suffix, "jpg")) && 773: (strcmp (suffix, "pdf")) && (strcmp (suffix, "png")) && 774: (strcmp (suffix, "tgz")) && (strcmp (suffix, "xpg")) && 775: (strcmp (suffix, "xpd")) && (strcmp (suffix, "ps")) && 776: (strcmp (suffix, "au")) && (strcmp (suffix, "html")) && 777: (strcmp (suffix, "htm")) && (strcmp (suffix, "shtml")) && 778: (strcmp (suffix, "txt")) && (strcmp (suffix, "css")) && 779: (strcmp (suffix, "eps"))) 780: return (FALSE); 781: else 782: return (TRUE); 783: } 784: else 785: return (FALSE); 786: } 787: 788: 789: /*---------------------------------------------------------------------- 1.24 cvs 790: ChopURL 791: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL). 792: If inputURL is bigger than that size, outputURL receives 793: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...", 794: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL. 795: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets 796: copied into outputURL. 797: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH 798: chars. 799: ----------------------------------------------------------------------*/ 800: #ifdef __STDC__ 1.34 cvs 801: void ChopURL (char *outputURL, const char *inputURL) 1.24 cvs 802: #else 803: void ChopURL (outputURL, inputURL) 1.34 cvs 804: char *outputURL; 805: const char *inputURL; 1.24 cvs 806: #endif 1.22 cvs 807: 1.24 cvs 808: { 809: int len; 1.9 cvs 810: 1.24 cvs 811: len = strlen (inputURL); 812: if (len <= MAX_PRINT_URL_LENGTH) 1.29 cvs 813: strcpy (outputURL, inputURL); 1.24 cvs 814: else 815: /* make a truncated urlName on the status window */ 816: { 817: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2); 818: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS; 819: strcat (outputURL, "..."); 820: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ])); 821: } 1.25 cvs 822: } 823: 824: 825: /*---------------------------------------------------------------------- 826: scan 1.47 cvs 827: Scan a filename for its constituents 1.25 cvs 828: ----------------------------------- 829: 830: On entry, 831: name points to a document name which may be incomplete. 832: On exit, 833: absolute or relative may be nonzero (but not both). 834: host, fragment and access may be nonzero if they were specified. 835: Any which are nonzero point to zero terminated strings. 836: ----------------------------------------------------------------------*/ 837: #ifdef __STDC__ 838: static void scan (char * name, HTURI * parts) 839: #else /* __STDC__ */ 840: static void scan (name, parts) 841: char *name; 842: HTURI *parts; 843: 844: #endif /* __STDC__ */ 845: { 1.43 cvs 846: char *p; 847: char *after_access = name; 1.32 cvs 848: 1.43 cvs 849: memset (parts, '\0', sizeof (HTURI)); 1.28 cvs 850: /* Look for fragment identifier */ 851: if ((p = strrchr(name, '#')) != NULL) 852: { 853: *p++ = '\0'; 854: parts->fragment = p; 1.25 cvs 855: } 856: 1.28 cvs 857: for (p=name; *p; p++) 858: { 1.43 cvs 859: if (*p == URL_SEP || *p == DIR_SEP || *p=='#' || *p=='?') 1.28 cvs 860: break; 1.32 cvs 861: if (*p==':') 1.28 cvs 862: { 863: *p = 0; 864: parts->access = after_access; /* Scheme has been specified */ 865: 866: /* The combination of gcc, the "-O" flag and the HP platform is 867: unhealthy. The following three lines is a quick & dirty fix, but is 868: not recommended. Rather, turn off "-O". */ 869: 870: /* after_access = p;*/ 871: /* while (*after_access == 0)*/ 872: /* after_access++;*/ 873: after_access = p+1; 1.43 cvs 874: if (!strcasecmp("URL", parts->access)) 1.28 cvs 875: /* Ignore IETF's URL: pre-prefix */ 876: parts->access = NULL; 877: else 1.25 cvs 878: break; 879: } 880: } 881: 882: p = after_access; 1.43 cvs 883: if (*p == URL_SEP || *p == DIR_SEP) 1.28 cvs 884: { 1.43 cvs 885: if (p[1] == URL_SEP) 1.28 cvs 886: { 1.25 cvs 887: parts->host = p+2; /* host has been specified */ 1.28 cvs 888: *p = 0; /* Terminate access */ 889: /* look for end of host name if any */ 1.43 cvs 890: p = strchr (parts->host, URL_SEP); 1.28 cvs 891: if (p) 892: { 1.43 cvs 893: *p = EOS; /* Terminate host */ 1.25 cvs 894: parts->absolute = p+1; /* Root has been found */ 1.28 cvs 895: } 896: } 897: else 898: /* Root found but no host */ 899: parts->absolute = p+1; 900: } 901: else 902: { 1.25 cvs 903: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */ 1.28 cvs 904: } 1.25 cvs 905: } 906: 907: 908: /*---------------------------------------------------------------------- 1.28 cvs 909: AmayaParseUrl: parse a Name relative to another name 910: 911: This returns those parts of a name which are given (and requested) 912: substituting bits from the related name where necessary. 1.25 cvs 913: 1.28 cvs 914: On entry, 1.25 cvs 915: aName A filename given 916: relatedName A name relative to which aName is to be parsed. Give 917: it an empty string if aName is absolute. 918: wanted A mask for the bits which are wanted. 919: 1.28 cvs 920: On exit, 1.25 cvs 921: returns A pointer to a malloc'd string which MUST BE FREED 922: ----------------------------------------------------------------------*/ 923: #ifdef __STDC__ 1.34 cvs 924: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted) 1.25 cvs 925: #else /* __STDC__ */ 1.28 cvs 926: char *AmayaParseUrl (aName, relatedName, wanted) 1.34 cvs 927: const char *aName; 1.28 cvs 928: char *relatedName; 929: int wanted; 1.25 cvs 930: 931: #endif /* __STDC__ */ 932: { 1.29 cvs 933: char *return_value; 934: char result[MAX_LENGTH]; 935: char name[MAX_LENGTH]; 936: char rel[MAX_LENGTH]; 937: char *p, *access; 938: HTURI given, related; 939: int len; 1.33 cvs 940: char used_sep; 941: char *used_str; 1.32 cvs 942: 1.43 cvs 943: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP)) 1.33 cvs 944: { 1.42 cvs 945: used_str = DIR_STR; 946: used_sep = DIR_SEP; 1.33 cvs 947: } 1.32 cvs 948: else 1.33 cvs 949: { 1.42 cvs 950: used_str = URL_STR; 951: used_sep = URL_SEP; 1.33 cvs 952: } 1.32 cvs 953: 1.29 cvs 954: /* Make working copies of input strings to cut up: */ 955: return_value = NULL; 956: result[0] = 0; /* Clear string */ 957: strcpy (name, aName); 958: if (relatedName != NULL) 959: strcpy (rel, relatedName); 960: else 961: relatedName[0] = EOS; 962: 963: scan (name, &given); 964: scan (rel, &related); 965: access = given.access ? given.access : related.access; 966: if (wanted & AMAYA_PARSE_ACCESS) 967: if (access) 968: { 969: strcat (result, access); 970: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.32 cvs 971: strcat (result, ":"); 1.29 cvs 972: } 973: 974: if (given.access && related.access) 975: /* If different, inherit nothing. */ 976: if (strcmp (given.access, related.access) != 0) 977: { 978: related.host = 0; 979: related.absolute = 0; 980: related.relative = 0; 981: related.fragment = 0; 982: } 983: 984: if (wanted & AMAYA_PARSE_HOST) 985: if(given.host || related.host) 986: { 987: if(wanted & AMAYA_PARSE_PUNCTUATION) 988: strcat (result, "//"); 989: strcat (result, given.host ? given.host : related.host); 990: } 991: 992: if (given.host && related.host) 993: /* If different hosts, inherit no path. */ 994: if (strcmp(given.host, related.host) != 0) 995: { 996: related.absolute = 0; 997: related.relative = 0; 998: related.fragment = 0; 999: } 1000: 1001: if (wanted & AMAYA_PARSE_PATH) 1002: { 1003: if (given.absolute) 1004: { 1005: /* All is given */ 1006: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.33 cvs 1007: strcat (result, used_str); 1.29 cvs 1008: strcat (result, given.absolute); 1.25 cvs 1009: } 1.29 cvs 1010: else if (related.absolute) 1011: { 1012: /* Adopt path not name */ 1.33 cvs 1013: strcat (result, used_str); 1.29 cvs 1014: strcat (result, related.absolute); 1015: if (given.relative) 1016: { 1017: /* Search part? */ 1018: p = strchr (result, '?'); 1019: if (!p) 1020: p=result+strlen(result)-1; 1.33 cvs 1021: for (; *p!=used_sep; p--); /* last / */ 1.29 cvs 1022: /* Remove filename */ 1023: p[1]=0; 1024: /* Add given one */ 1025: strcat (result, given.relative); 1026: /*SimplifyUrl (&result);*/ 1.25 cvs 1027: } 1028: } 1.29 cvs 1029: else if (given.relative) 1030: /* what we've got */ 1031: strcat (result, given.relative); 1032: else if (related.relative) 1033: strcat (result, related.relative); 1034: else 1035: /* No inheritance */ 1.33 cvs 1036: strcat (result, used_str); 1.25 cvs 1037: } 1.29 cvs 1038: 1039: if (wanted & AMAYA_PARSE_ANCHOR) 1040: if (given.fragment || related.fragment) 1041: { 1042: if (given.absolute && given.fragment) 1043: { 1044: /*Fixes for relURLs...*/ 1045: if (wanted & AMAYA_PARSE_PUNCTUATION) 1046: strcat (result, "#"); 1047: strcat (result, given.fragment); 1048: } 1049: else if (!(given.absolute) && !(given.fragment)) 1050: strcat (result, ""); 1051: else 1052: { 1053: if (wanted & AMAYA_PARSE_PUNCTUATION) 1054: strcat (result, "#"); 1055: strcat (result, given.fragment ? given.fragment : related.fragment); 1056: } 1057: } 1058: len = strlen (result); 1059: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1060: strcpy (return_value, result); 1061: return (return_value); /* exactly the right length */ 1.25 cvs 1062: } 1063: 1064: /*---------------------------------------------------------------------- 1065: HTCanon 1066: Canonicalizes the URL in the following manner starting from the host 1067: pointer: 1068: 1069: 1) The host name is converted to lowercase 1070: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp) 1071: 1072: Return: OK The position of the current path part of the URL 1073: which might be the old one or a new one. 1074: 1075: ----------------------------------------------------------------------*/ 1076: #ifdef __STDC__ 1.28 cvs 1077: static char *HTCanon (char ** filename, char * host) 1.25 cvs 1078: #else /* __STDC__ */ 1.28 cvs 1079: static char *HTCanon (filename, host) 1080: char **filename; 1081: char *host; 1.25 cvs 1082: #endif /* __STDC__ */ 1083: { 1084: char *newname = NULL; 1085: char *port; 1086: char *strptr; 1087: char *path; 1088: char *access = host-3; 1.33 cvs 1089: char used_sep; 1.32 cvs 1090: 1091: 1.33 cvs 1092: if (*filename && strchr (*filename, URL_SEP)) 1093: { 1094: used_sep = URL_SEP; 1095: } 1096: else 1097: { 1098: used_sep = DIR_SEP; 1099: } 1.32 cvs 1100: 1.33 cvs 1101: while (access>*filename && *(access-1)!= used_sep) /* Find access method */ 1.25 cvs 1102: access--; 1.33 cvs 1103: if ((path = strchr(host, used_sep)) == NULL) /* Find path */ 1.25 cvs 1104: path = host + strlen(host); 1105: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */ 1106: host = strptr; 1.32 cvs 1107: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */ 1.25 cvs 1108: port = NULL; 1109: 1110: strptr = host; /* Convert to lower-case */ 1.33 cvs 1111: while (strptr<path) 1112: { 1.25 cvs 1113: *strptr = tolower(*strptr); 1114: strptr++; 1.33 cvs 1115: } 1.25 cvs 1116: 1117: /* Does the URL contain a full domain name? This also works for a 1118: numerical host name. The domain name is already made lower-case 1119: and without a trailing dot. */ 1120: { 1.33 cvs 1121: char *dot = port ? port : path; 1122: if (dot > *filename && *--dot=='.') 1123: { 1124: char *orig=dot, *dest=dot+1; 1125: while((*orig++ = *dest++)); 1126: if (port) port--; 1127: path--; 1.25 cvs 1128: } 1129: } 1130: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */ 1.33 cvs 1131: if (port) 1132: { 1133: if (!*(port+1) || *(port+1)==used_sep) 1134: { 1135: if (!newname) 1136: { 1.25 cvs 1137: char *orig=port, *dest=port+1; 1138: while((*orig++ = *dest++)); 1.33 cvs 1139: } 1140: } 1141: else if ((!strncmp(access, "http", 4) && 1142: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) || 1143: (!strncmp(access, "gopher", 6) && 1144: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) || 1145: (!strncmp(access, "ftp", 3) && 1146: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) { 1147: if (!newname) 1148: { 1149: char *orig=port, *dest=port+3; 1150: while((*orig++ = *dest++)); 1151: /* Update path position, Henry Minsky */ 1152: path -= 3; 1.25 cvs 1153: } 1.33 cvs 1154: } 1155: else if (newname) 1156: strncat(newname, port, (int) (path-port)); 1157: } 1.25 cvs 1158: 1.33 cvs 1159: if (newname) 1160: { 1.25 cvs 1161: char *newpath = newname+strlen(newname); 1162: strcat(newname, path); 1163: path = newpath; 1.28 cvs 1164: /* Free old copy */ 1165: TtaFreeMemory(*filename); 1.25 cvs 1166: *filename = newname; 1.33 cvs 1167: } 1.25 cvs 1168: return path; 1169: } 1170: 1171: 1172: /*---------------------------------------------------------------------- 1.29 cvs 1173: SimplifyUrl: simplify a URI 1.32 cvs 1174: A URI is allowed to contain the sequence xxx/../ which may be 1175: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR. 1.28 cvs 1176: Simplification helps us recognize duplicate URIs. 1.25 cvs 1177: 1.28 cvs 1178: Thus, /etc/junk/../fred becomes /etc/fred 1179: /etc/junk/./fred becomes /etc/junk/fred 1.25 cvs 1180: 1.28 cvs 1181: but we should NOT change 1182: http://fred.xxx.edu/../.. 1.25 cvs 1183: 1184: or ../../albert.html 1185: 1.28 cvs 1186: In order to avoid empty URLs the following URLs become: 1.25 cvs 1187: 1188: /fred/.. becomes /fred/.. 1189: /fred/././.. becomes /fred/.. 1190: /fred/.././junk/.././ becomes /fred/.. 1191: 1.28 cvs 1192: If more than one set of `://' is found (several proxies in cascade) then 1193: only the part after the last `://' is simplified. 1.25 cvs 1194: 1.28 cvs 1195: Returns: A string which might be the old one or a new one. 1.25 cvs 1196: ----------------------------------------------------------------------*/ 1197: #ifdef __STDC__ 1.29 cvs 1198: void SimplifyUrl (char ** url) 1.25 cvs 1199: #else /* __STDC__ */ 1.29 cvs 1200: void SimplifyUrl (url) 1.28 cvs 1201: char **url; 1.25 cvs 1202: #endif /* __STDC__ */ 1203: { 1.28 cvs 1204: char *path, *p; 1205: char *newptr, *access; 1206: char *orig, *dest, *end; 1207: 1.33 cvs 1208: char used_sep; 1.32 cvs 1209: 1210: 1.28 cvs 1211: if (!url || !*url) 1212: return; 1213: 1.33 cvs 1214: if (strchr (*url, URL_SEP)) 1215: { 1216: used_sep = URL_SEP; 1217: } 1.32 cvs 1218: else 1.33 cvs 1219: { 1220: used_sep = DIR_SEP; 1221: } 1.32 cvs 1222: 1.28 cvs 1223: /* Find any scheme name */ 1224: if ((path = strstr(*url, "://")) != NULL) 1.33 cvs 1225: { 1226: /* Find host name */ 1.28 cvs 1227: access = *url; 1228: while (access<path && (*access=tolower(*access))) 1229: access++; 1230: path += 3; 1231: while ((newptr = strstr(path, "://")) != NULL) 1232: /* For proxies */ 1233: path = newptr+3; 1234: /* We have a host name */ 1235: path = HTCanon(url, path); 1.25 cvs 1236: } 1.28 cvs 1237: else if ((path = strstr(*url, ":/")) != NULL) 1238: path += 2; 1239: else 1240: path = *url; 1.25 cvs 1241: 1.33 cvs 1242: if (*path == used_sep && *(path+1)==used_sep) 1.28 cvs 1243: /* Some URLs start //<foo> */ 1244: path += 1; 1245: else if (!strncmp(path, "news:", 5)) 1246: { 1247: newptr = strchr(path+5, '@'); 1248: if (!newptr) 1249: newptr = path + 5; 1250: while (*newptr) 1251: { 1252: /* Make group or host lower case */ 1253: *newptr = tolower (*newptr); 1254: newptr++; 1.25 cvs 1255: } 1.28 cvs 1256: /* Doesn't need to do any more */ 1257: return; 1.25 cvs 1258: } 1.28 cvs 1259: 1260: if ((p = path)) 1261: { 1262: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) || 1263: (end = strchr (path, '#')))) 1264: end = path + strlen (path); 1265: 1266: /* Parse string second time to simplify */ 1267: p = path; 1268: while (p < end) 1269: { 1.33 cvs 1270: if (*p==used_sep) 1.28 cvs 1271: { 1.33 cvs 1272: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2))) 1.28 cvs 1273: { 1274: orig = p + 1; 1.33 cvs 1275: dest = (*(p+2)!=used_sep) ? p+2 : p+3; 1.28 cvs 1276: while ((*orig++ = *dest++)); /* Remove a slash and a dot */ 1277: end = orig - 1; 1278: } 1.33 cvs 1279: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3))) 1.28 cvs 1280: { 1281: newptr = p; 1.33 cvs 1282: while (newptr>path && *--newptr!=used_sep); /* prev slash */ 1.28 cvs 1283: if (strncmp(newptr, "/../", 4)) 1284: { 1285: orig = newptr + 1; 1.33 cvs 1286: dest = (*(p+3)!=used_sep) ? p+3 : p+4; 1.28 cvs 1287: while ((*orig++ = *dest++)); /* Remove /xxx/.. */ 1288: end = orig-1; 1289: /* Start again with prev slash */ 1290: p = newptr; 1.25 cvs 1291: } 1.28 cvs 1292: else 1.25 cvs 1293: p++; 1.28 cvs 1294: } 1.33 cvs 1295: else if (*(p+1) == used_sep) 1.28 cvs 1296: { 1.33 cvs 1297: while (*(p+1) == used_sep) 1.28 cvs 1298: { 1299: orig = p; 1300: dest = p + 1; 1301: while ((*orig++ = *dest++)); /* Remove multiple /'s */ 1302: end = orig-1; 1303: } 1304: } 1305: else 1.25 cvs 1306: p++; 1.28 cvs 1307: } 1308: else 1309: p++; 1.25 cvs 1310: } 1311: } 1.28 cvs 1312: return; 1313: } 1314: 1315: 1316: /*---------------------------------------------------------------------- 1317: NormalizeFile normalizes local names. 1318: Return TRUE if target and src differ. 1319: ----------------------------------------------------------------------*/ 1320: #ifdef __STDC__ 1321: boolean NormalizeFile (char *src, char *target) 1322: #else 1323: boolean NormalizeFile (src, target) 1324: char *src; 1325: char *target; 1326: 1327: #endif 1328: { 1329: char *s; 1330: boolean change; 1331: 1332: change = FALSE; 1.48 ! cvs 1333: # ifdef _WINDOWS ! 1334: if (src[1] == ':' && src[2] == '\\') { ! 1335: strcpy (target, src); ! 1336: change = TRUE; ! 1337: } ! 1338: # else /* !_WINDOWS */ 1.28 cvs 1339: if (src[0] == '~') 1340: { 1341: /* replace ~ */ 1342: s = (char *) TtaGetEnvString ("HOME"); 1343: strcpy (target, s); 1344: strcat (target, &src[1]); 1345: change = TRUE; 1346: } 1.48 ! cvs 1347: # endif /* _WINDOWS */ 1.28 cvs 1348: else if (strncmp (src, "file:", 5) == 0) 1349: { 1350: /* remove the prefix file: */ 1351: if (src[5] == EOS) 1352: strcpy (target, DIR_STR); 1353: else if (src[0] == '~') 1354: { 1355: /* replace ~ */ 1356: s = (char *) TtaGetEnvString ("HOME"); 1357: strcpy (target, s); 1358: strcat (target, &src[5]); 1359: } 1360: else 1361: strcpy (target, &src[5]); 1362: change = TRUE; 1363: } 1364: else 1365: strcpy (target, src); 1366: 1367: /* remove /../ and /./ */ 1.29 cvs 1368: SimplifyUrl (&target); 1.28 cvs 1369: return (change); 1.25 cvs 1370: } 1371: 1.28 cvs 1372: 1.25 cvs 1373: /*---------------------------------------------------------------------- 1.31 cvs 1374: MakeRelativeURL: make relative name 1.25 cvs 1375: 1.28 cvs 1376: This function creates and returns a string which gives an expression of 1377: one address as related to another. Where there is no relation, an absolute 1378: address is retured. 1.25 cvs 1379: 1.28 cvs 1380: On entry, 1.25 cvs 1381: Both names must be absolute, fully qualified names of nodes 1382: (no fragment bits) 1383: 1.28 cvs 1384: On exit, 1.25 cvs 1385: The return result points to a newly allocated name which, if 1386: parsed by AmayaParseUrl relative to relatedName, will yield aName. 1387: The caller is responsible for freeing the resulting name later. 1388: ----------------------------------------------------------------------*/ 1389: #ifdef __STDC__ 1.31 cvs 1390: char *MakeRelativeURL (char *aName, char *relatedName) 1.25 cvs 1391: #else /* __STDC__ */ 1.31 cvs 1392: char *MakeRelativeURL (aName, relatedName) 1.28 cvs 1393: char *aName; 1394: char *relatedName; 1.25 cvs 1395: #endif /* __STDC__ */ 1396: { 1.39 cvs 1397: char *return_value; 1398: char result[MAX_LENGTH]; 1.29 cvs 1399: char *p; 1.34 cvs 1400: char *q; 1.29 cvs 1401: char *after_access; 1402: char *last_slash = NULL; 1403: int slashes, levels, len; 1404: 1.44 cvs 1405: # ifdef _WINDOWS 1406: int ndx; 1407: # endif /* _WINDOWS */ 1408: 1.29 cvs 1409: if (aName == NULL || relatedName == NULL) 1410: return (NULL); 1411: 1412: slashes = 0; 1413: after_access = NULL; 1414: p = aName; 1415: q = relatedName; 1416: for (; *p && (*p == *q); p++, q++) 1.27 cvs 1417: { 1418: /* Find extent of match */ 1.32 cvs 1419: if (*p == ':') 1.29 cvs 1420: after_access = p + 1; 1.28 cvs 1421: if (*p == DIR_SEP) 1.27 cvs 1422: { 1.29 cvs 1423: /* memorize the last slash position and count them */ 1.27 cvs 1424: last_slash = p; 1425: slashes++; 1.25 cvs 1426: } 1427: } 1428: 1.31 cvs 1429: /* q, p point to the first non-matching character or zero */ 1430: if (*q == EOS) 1431: { 1432: /* New name is a subset of the related name */ 1433: /* exactly the right length */ 1434: len = strlen (p); 1435: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1436: strcpy (return_value, p); 1437: } 1438: else if ((slashes < 2 && after_access == NULL) 1439: || (slashes < 3 && after_access != NULL)) 1440: { 1441: /* Two names whitout common path */ 1442: /* exactly the right length */ 1443: len = strlen (aName); 1444: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1445: strcpy (return_value, aName); 1446: } 1447: else 1448: { 1449: /* Some path in common */ 1450: if (slashes == 3 && strncmp (aName, "http:", 5) == 0) 1451: /* just the same server */ 1452: strcpy (result, last_slash); 1453: else 1454: { 1455: levels= 0; 1456: for (; *q && (*q != '#'); q++) 1457: if (*q == DIR_SEP) 1458: levels++; 1459: 1460: result[0] = 0; 1461: for (;levels; levels--) 1462: strcat (result, "../"); 1463: strcat (result, last_slash+1); 1464: } 1465: 1466: /* exactly the right length */ 1467: len = strlen (result); 1468: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1469: strcpy (return_value, result); 1.25 cvs 1470: } 1.44 cvs 1471: # ifdef _WINDOWS 1472: len = strlen (return_value); 1473: for (ndx = 0; ndx < len; ndx ++) 1474: if (return_value[ndx] == '\\') 1475: return_value[ndx] = '/' ; 1476: # endif /* _WINDOWS */ 1.29 cvs 1477: return (return_value); 1.24 cvs 1478: } 1.35 cvs 1479: 1480: