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: ptr = TtaGetMemory (MAX_LENGTH); 523: documentname = TtaGetMemory (MAX_LENGTH); 524: /* check whether the file name exists */ 525: len = strlen (url) - 1; 1.41 cvs 526: if (IsW3Path) 527: url_sep = '/'; 528: else 529: url_sep = DIR_SEP; 530: noFile = (url[len] == url_sep); 1.40 cvs 531: if (noFile) 532: url[len] = EOS; 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); 1.46 ! cvs 538: /* don't include ':' within document name */ ! 539: n = strchr (documentname, ':'); ! 540: if (n != NULL) ! 541: *n = EOS; 1.40 cvs 542: strcat (ptr, documentname); 543: TtaFreeMemory (documentname); 544: /* restore the url */ 545: if (noFile) 1.41 cvs 546: url[len] = url_sep; 1.40 cvs 547: return (ptr); 548: } 549: else 550: return (NULL); 551: } 552: 553: 554: /*---------------------------------------------------------------------- 1.9 cvs 555: NormalizeURL 556: normalizes orgName according to a base associated with doc, and 557: following the standard URL format rules. 558: The function returns the new complete and normalized URL 1.12 cvs 559: or file name path (newName) and the name of the document (docName). 1.9 cvs 560: N.B. If the function can't find out what's the docName, it assigns 561: the name "noname.html". 1.4 cvs 562: ----------------------------------------------------------------------*/ 1.3 cvs 563: #ifdef __STDC__ 564: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName) 565: #else /* __STDC__ */ 566: void NormalizeURL (orgName, doc, newName, docName) 567: char *orgName; 568: Document doc; 569: char *newName; 570: char *docName; 571: #endif /* __STDC__ */ 572: { 1.31 cvs 573: char *basename; 1.18 cvs 574: char tempOrgName[MAX_LENGTH]; 1.5 cvs 575: char *ptr; 576: int length; 577: 1.44 cvs 578: # ifdef _WINDOWS 579: int ndx; 580: # endif /* _WINDOWS */ 581: 1.5 cvs 582: if (!newName || !docName) 583: return; 1.18 cvs 584: 1.32 cvs 585: if (doc != 0) 1.33 cvs 586: basename = GetBaseURL (doc); 1.32 cvs 587: else 1.33 cvs 588: basename = (char *) NULL; 1.32 cvs 589: 1.18 cvs 590: /* 1.31 cvs 591: * Clean orgName 592: * Make sure we have a complete orgName, without any leading or trailing 593: * white spaces, or trailinbg new lines 594: */ 1.5 cvs 595: ptr = orgName; 1.18 cvs 596: /* skip leading white space and new line characters */ 1.19 cvs 597: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS); 1.39 cvs 598: strncpy (tempOrgName, ptr, MAX_LENGTH -1); 599: tempOrgName[MAX_LENGTH -1] = EOS; 1.18 cvs 600: /* clean trailing white space */ 601: ptr = strchr (tempOrgName, ' '); 602: if (ptr) 603: *ptr = EOS; 604: /* clean trailing new lines */ 1.19 cvs 605: ptr = strchr (tempOrgName, EOL); 1.5 cvs 606: if (ptr) 607: *ptr = EOS; 608: 1.18 cvs 609: /* 1.31 cvs 610: * Make orgName a complete URL 611: * If the URL does not include a protocol, then try to calculate 612: * one using the doc's base element (if it exists), 613: */ 1.21 cvs 614: if (tempOrgName[0] == EOS) 615: { 616: newName[0] = EOS; 1.36 cvs 617: TtaFreeMemory (basename); 1.21 cvs 618: return; 619: } 620: else if (IsW3Path (tempOrgName)) 621: { 622: /* the name is complete, go to the Sixth Step */ 623: strcpy (newName, tempOrgName); 624: /* verify if the URL has the form "protocol://server:port" */ 1.25 cvs 625: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | 626: AMAYA_PARSE_PUNCTUATION); 1.21 cvs 627: if (ptr && !strcmp (ptr, newName)) 1.43 cvs 628: /* it has this form, we complete it by adding a DIR_STR */ 629: strcat (newName, URL_STR); 630: 1.21 cvs 631: if (ptr) 1.25 cvs 632: TtaFreeMemory (ptr); 1.21 cvs 633: } 634: else if ( doc == 0) 1.19 cvs 635: /* the name is complete, go to the Sixth Step */ 1.18 cvs 636: strcpy (newName, tempOrgName); 1.5 cvs 637: else 638: { 1.31 cvs 639: /* Calculate the absolute URL, using the base or document URL */ 1.44 cvs 640: # ifdef _WINDOWS 1.45 cvs 641: if (!IsW3Path (basename)) { 642: length = strlen (tempOrgName); 643: for (ndx = 0; ndx < length; ndx++) 644: if (tempOrgName [ndx] == '/') 645: tempOrgName [ndx] = '\\'; 646: } 1.44 cvs 647: # endif /* _WINDOWS */ 1.25 cvs 648: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL); 1.14 cvs 649: if (ptr) 650: { 1.29 cvs 651: SimplifyUrl (&ptr); 1.14 cvs 652: strcpy (newName, ptr); 1.25 cvs 653: TtaFreeMemory (ptr); 1.14 cvs 654: } 655: else 1.18 cvs 656: newName[0] = EOS; 1.5 cvs 657: } 1.36 cvs 658: 659: TtaFreeMemory (basename); 1.18 cvs 660: /* 1.31 cvs 661: * Prepare the docname that will refer to this ressource in the 662: * .amaya directory. If the new URL finishes on DIR_SEP, then use 663: * noname.html as a default ressource name 1.18 cvs 664: */ 1.19 cvs 665: if (newName[0] != EOS) 1.5 cvs 666: { 1.19 cvs 667: length = strlen (newName) - 1; 1.43 cvs 668: if (newName[length] == URL_SEP || newName[length] == DIR_SEP) 1.18 cvs 669: { 670: /* docname was not comprised inside the URL, so let's */ 671: /* assign the default ressource name */ 672: strcpy (docName, "noname.html"); 673: /* remove DIR_SEP at the end of complete path */ 1.23 cvs 674: /* newName[length] = EOS; */ 1.18 cvs 675: } 1.14 cvs 676: else 1.18 cvs 677: { 678: /* docname is comprised inside the URL */ 1.43 cvs 679: while (length >= 0 && newName[length] != URL_SEP && newName[length] != DIR_SEP) 1.18 cvs 680: length--; 681: if (length < 0) 682: strcpy (docName, newName); 683: else 684: strcpy (docName, &newName[length+1]); 685: } 1.5 cvs 686: } 1.18 cvs 687: else 688: docName[0] = EOS; 689: } 1.3 cvs 690: 1.4 cvs 691: /*---------------------------------------------------------------------- 1.9 cvs 692: IsSameHost 1.4 cvs 693: ----------------------------------------------------------------------*/ 1.3 cvs 694: #ifdef __STDC__ 1.34 cvs 695: boolean IsSameHost (const char *url1, const char *url2) 1.3 cvs 696: #else /* __STDC__ */ 697: boolean IsSameHost (url1, url2) 1.34 cvs 698: const char *url1; 699: const char *url2; 1.3 cvs 700: #endif /* __STDC__ */ 701: { 1.34 cvs 702: char *basename_ptr1, *basename_ptr2; 703: boolean result; 1.3 cvs 704: 1.25 cvs 705: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 706: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 1.3 cvs 707: 1.5 cvs 708: if (strcmp (basename_ptr1, basename_ptr2)) 1.8 cvs 709: result = FALSE; 1.5 cvs 710: else 1.8 cvs 711: result = TRUE; 1.3 cvs 712: 1.25 cvs 713: TtaFreeMemory (basename_ptr1); 714: TtaFreeMemory (basename_ptr2); 1.5 cvs 715: return (result); 1.3 cvs 716: } 717: 718: 1.4 cvs 719: /*---------------------------------------------------------------------- 1.22 cvs 720: HasKnownFileSuffix 721: returns TRUE if path points to a file ending with a suffix. 722: ----------------------------------------------------------------------*/ 723: #ifdef __STDC__ 1.34 cvs 724: boolean HasKnownFileSuffix (const char *path) 1.22 cvs 725: #else /* __STDC__ */ 726: boolean HasKnownFileSuffix (path) 1.34 cvs 727: const char *path; 1.22 cvs 728: #endif /* __STDC__ */ 729: { 1.29 cvs 730: char *root; 731: char temppath[MAX_LENGTH]; 732: char suffix[MAX_LENGTH]; 1.22 cvs 733: 1.24 cvs 734: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP) 1.22 cvs 735: return (FALSE); 736: 1.29 cvs 737: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION); 1.22 cvs 738: 739: if (root) 740: { 741: strcpy (temppath, root); 1.25 cvs 742: TtaFreeMemory (root); 1.22 cvs 743: /* Get the suffix */ 744: ExtractSuffix (temppath, suffix); 745: 746: if( suffix[0] == EOS) 747: /* no suffix */ 748: return (FALSE); 749: 750: /* Normalize the suffix */ 751: ConvertToLowerCase (suffix); 752: 1.23 cvs 753: if (!strcmp (suffix, "gz")) 1.22 cvs 754: /* skip the compressed suffix */ 755: { 756: ExtractSuffix (temppath, suffix); 757: if(suffix[0] == EOS) 758: /* no suffix */ 759: return (FALSE); 760: /* Normalize the suffix */ 761: ConvertToLowerCase (suffix); 762: } 763: 764: if ((strcmp (suffix, "gif")) && (strcmp (suffix, "xbm")) && 765: (strcmp (suffix, "xpm")) && (strcmp (suffix, "jpg")) && 766: (strcmp (suffix, "pdf")) && (strcmp (suffix, "png")) && 767: (strcmp (suffix, "tgz")) && (strcmp (suffix, "xpg")) && 768: (strcmp (suffix, "xpd")) && (strcmp (suffix, "ps")) && 769: (strcmp (suffix, "au")) && (strcmp (suffix, "html")) && 770: (strcmp (suffix, "htm")) && (strcmp (suffix, "shtml")) && 771: (strcmp (suffix, "txt")) && (strcmp (suffix, "css")) && 772: (strcmp (suffix, "eps"))) 773: return (FALSE); 774: else 775: return (TRUE); 776: } 777: else 778: return (FALSE); 779: } 780: 781: 782: /*---------------------------------------------------------------------- 1.24 cvs 783: ChopURL 784: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL). 785: If inputURL is bigger than that size, outputURL receives 786: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...", 787: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL. 788: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets 789: copied into outputURL. 790: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH 791: chars. 792: ----------------------------------------------------------------------*/ 793: #ifdef __STDC__ 1.34 cvs 794: void ChopURL (char *outputURL, const char *inputURL) 1.24 cvs 795: #else 796: void ChopURL (outputURL, inputURL) 1.34 cvs 797: char *outputURL; 798: const char *inputURL; 1.24 cvs 799: #endif 1.22 cvs 800: 1.24 cvs 801: { 802: int len; 1.9 cvs 803: 1.24 cvs 804: len = strlen (inputURL); 805: if (len <= MAX_PRINT_URL_LENGTH) 1.29 cvs 806: strcpy (outputURL, inputURL); 1.24 cvs 807: else 808: /* make a truncated urlName on the status window */ 809: { 810: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2); 811: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS; 812: strcat (outputURL, "..."); 813: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ])); 814: } 1.25 cvs 815: } 816: 817: 818: /*---------------------------------------------------------------------- 819: scan 820: Scan a filename for its consituents 821: ----------------------------------- 822: 823: On entry, 824: name points to a document name which may be incomplete. 825: On exit, 826: absolute or relative may be nonzero (but not both). 827: host, fragment and access may be nonzero if they were specified. 828: Any which are nonzero point to zero terminated strings. 829: ----------------------------------------------------------------------*/ 830: #ifdef __STDC__ 831: static void scan (char * name, HTURI * parts) 832: #else /* __STDC__ */ 833: static void scan (name, parts) 834: char *name; 835: HTURI *parts; 836: 837: #endif /* __STDC__ */ 838: { 1.43 cvs 839: char *p; 840: char *after_access = name; 1.32 cvs 841: 1.43 cvs 842: memset (parts, '\0', sizeof (HTURI)); 1.28 cvs 843: /* Look for fragment identifier */ 844: if ((p = strrchr(name, '#')) != NULL) 845: { 846: *p++ = '\0'; 847: parts->fragment = p; 1.25 cvs 848: } 849: 1.28 cvs 850: for (p=name; *p; p++) 851: { 1.43 cvs 852: if (*p == URL_SEP || *p == DIR_SEP || *p=='#' || *p=='?') 1.28 cvs 853: break; 1.32 cvs 854: if (*p==':') 1.28 cvs 855: { 856: *p = 0; 857: parts->access = after_access; /* Scheme has been specified */ 858: 859: /* The combination of gcc, the "-O" flag and the HP platform is 860: unhealthy. The following three lines is a quick & dirty fix, but is 861: not recommended. Rather, turn off "-O". */ 862: 863: /* after_access = p;*/ 864: /* while (*after_access == 0)*/ 865: /* after_access++;*/ 866: after_access = p+1; 1.43 cvs 867: if (!strcasecmp("URL", parts->access)) 1.28 cvs 868: /* Ignore IETF's URL: pre-prefix */ 869: parts->access = NULL; 870: else 1.25 cvs 871: break; 872: } 873: } 874: 875: p = after_access; 1.43 cvs 876: if (*p == URL_SEP || *p == DIR_SEP) 1.28 cvs 877: { 1.43 cvs 878: if (p[1] == URL_SEP) 1.28 cvs 879: { 1.25 cvs 880: parts->host = p+2; /* host has been specified */ 1.28 cvs 881: *p = 0; /* Terminate access */ 882: /* look for end of host name if any */ 1.43 cvs 883: p = strchr (parts->host, URL_SEP); 1.28 cvs 884: if (p) 885: { 1.43 cvs 886: *p = EOS; /* Terminate host */ 1.25 cvs 887: parts->absolute = p+1; /* Root has been found */ 1.28 cvs 888: } 889: } 890: else 891: /* Root found but no host */ 892: parts->absolute = p+1; 893: } 894: else 895: { 1.25 cvs 896: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */ 1.28 cvs 897: } 1.25 cvs 898: } 899: 900: 901: /*---------------------------------------------------------------------- 1.28 cvs 902: AmayaParseUrl: parse a Name relative to another name 903: 904: This returns those parts of a name which are given (and requested) 905: substituting bits from the related name where necessary. 1.25 cvs 906: 1.28 cvs 907: On entry, 1.25 cvs 908: aName A filename given 909: relatedName A name relative to which aName is to be parsed. Give 910: it an empty string if aName is absolute. 911: wanted A mask for the bits which are wanted. 912: 1.28 cvs 913: On exit, 1.25 cvs 914: returns A pointer to a malloc'd string which MUST BE FREED 915: ----------------------------------------------------------------------*/ 916: #ifdef __STDC__ 1.34 cvs 917: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted) 1.25 cvs 918: #else /* __STDC__ */ 1.28 cvs 919: char *AmayaParseUrl (aName, relatedName, wanted) 1.34 cvs 920: const char *aName; 1.28 cvs 921: char *relatedName; 922: int wanted; 1.25 cvs 923: 924: #endif /* __STDC__ */ 925: { 1.29 cvs 926: char *return_value; 927: char result[MAX_LENGTH]; 928: char name[MAX_LENGTH]; 929: char rel[MAX_LENGTH]; 930: char *p, *access; 931: HTURI given, related; 932: int len; 1.33 cvs 933: char used_sep; 934: char *used_str; 1.32 cvs 935: 1.43 cvs 936: if (strchr (aName, DIR_SEP) || strchr (relatedName, DIR_SEP)) 1.33 cvs 937: { 1.42 cvs 938: used_str = DIR_STR; 939: used_sep = DIR_SEP; 1.33 cvs 940: } 1.32 cvs 941: else 1.33 cvs 942: { 1.42 cvs 943: used_str = URL_STR; 944: used_sep = URL_SEP; 1.33 cvs 945: } 1.32 cvs 946: 1.29 cvs 947: /* Make working copies of input strings to cut up: */ 948: return_value = NULL; 949: result[0] = 0; /* Clear string */ 950: strcpy (name, aName); 951: if (relatedName != NULL) 952: strcpy (rel, relatedName); 953: else 954: relatedName[0] = EOS; 955: 956: scan (name, &given); 957: scan (rel, &related); 958: access = given.access ? given.access : related.access; 959: if (wanted & AMAYA_PARSE_ACCESS) 960: if (access) 961: { 962: strcat (result, access); 963: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.32 cvs 964: strcat (result, ":"); 1.29 cvs 965: } 966: 967: if (given.access && related.access) 968: /* If different, inherit nothing. */ 969: if (strcmp (given.access, related.access) != 0) 970: { 971: related.host = 0; 972: related.absolute = 0; 973: related.relative = 0; 974: related.fragment = 0; 975: } 976: 977: if (wanted & AMAYA_PARSE_HOST) 978: if(given.host || related.host) 979: { 980: if(wanted & AMAYA_PARSE_PUNCTUATION) 981: strcat (result, "//"); 982: strcat (result, given.host ? given.host : related.host); 983: } 984: 985: if (given.host && related.host) 986: /* If different hosts, inherit no path. */ 987: if (strcmp(given.host, related.host) != 0) 988: { 989: related.absolute = 0; 990: related.relative = 0; 991: related.fragment = 0; 992: } 993: 994: if (wanted & AMAYA_PARSE_PATH) 995: { 996: if (given.absolute) 997: { 998: /* All is given */ 999: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.33 cvs 1000: strcat (result, used_str); 1.29 cvs 1001: strcat (result, given.absolute); 1.25 cvs 1002: } 1.29 cvs 1003: else if (related.absolute) 1004: { 1005: /* Adopt path not name */ 1.33 cvs 1006: strcat (result, used_str); 1.29 cvs 1007: strcat (result, related.absolute); 1008: if (given.relative) 1009: { 1010: /* Search part? */ 1011: p = strchr (result, '?'); 1012: if (!p) 1013: p=result+strlen(result)-1; 1.33 cvs 1014: for (; *p!=used_sep; p--); /* last / */ 1.29 cvs 1015: /* Remove filename */ 1016: p[1]=0; 1017: /* Add given one */ 1018: strcat (result, given.relative); 1019: /*SimplifyUrl (&result);*/ 1.25 cvs 1020: } 1021: } 1.29 cvs 1022: else if (given.relative) 1023: /* what we've got */ 1024: strcat (result, given.relative); 1025: else if (related.relative) 1026: strcat (result, related.relative); 1027: else 1028: /* No inheritance */ 1.33 cvs 1029: strcat (result, used_str); 1.25 cvs 1030: } 1.29 cvs 1031: 1032: if (wanted & AMAYA_PARSE_ANCHOR) 1033: if (given.fragment || related.fragment) 1034: { 1035: if (given.absolute && given.fragment) 1036: { 1037: /*Fixes for relURLs...*/ 1038: if (wanted & AMAYA_PARSE_PUNCTUATION) 1039: strcat (result, "#"); 1040: strcat (result, given.fragment); 1041: } 1042: else if (!(given.absolute) && !(given.fragment)) 1043: strcat (result, ""); 1044: else 1045: { 1046: if (wanted & AMAYA_PARSE_PUNCTUATION) 1047: strcat (result, "#"); 1048: strcat (result, given.fragment ? given.fragment : related.fragment); 1049: } 1050: } 1051: len = strlen (result); 1052: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1053: strcpy (return_value, result); 1054: return (return_value); /* exactly the right length */ 1.25 cvs 1055: } 1056: 1057: /*---------------------------------------------------------------------- 1058: HTCanon 1059: Canonicalizes the URL in the following manner starting from the host 1060: pointer: 1061: 1062: 1) The host name is converted to lowercase 1063: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp) 1064: 1065: Return: OK The position of the current path part of the URL 1066: which might be the old one or a new one. 1067: 1068: ----------------------------------------------------------------------*/ 1069: #ifdef __STDC__ 1.28 cvs 1070: static char *HTCanon (char ** filename, char * host) 1.25 cvs 1071: #else /* __STDC__ */ 1.28 cvs 1072: static char *HTCanon (filename, host) 1073: char **filename; 1074: char *host; 1.25 cvs 1075: #endif /* __STDC__ */ 1076: { 1077: char *newname = NULL; 1078: char *port; 1079: char *strptr; 1080: char *path; 1081: char *access = host-3; 1.33 cvs 1082: char used_sep; 1.32 cvs 1083: 1084: 1.33 cvs 1085: if (*filename && strchr (*filename, URL_SEP)) 1086: { 1087: used_sep = URL_SEP; 1088: } 1089: else 1090: { 1091: used_sep = DIR_SEP; 1092: } 1.32 cvs 1093: 1.33 cvs 1094: while (access>*filename && *(access-1)!= used_sep) /* Find access method */ 1.25 cvs 1095: access--; 1.33 cvs 1096: if ((path = strchr(host, used_sep)) == NULL) /* Find path */ 1.25 cvs 1097: path = host + strlen(host); 1098: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */ 1099: host = strptr; 1.32 cvs 1100: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */ 1.25 cvs 1101: port = NULL; 1102: 1103: strptr = host; /* Convert to lower-case */ 1.33 cvs 1104: while (strptr<path) 1105: { 1.25 cvs 1106: *strptr = tolower(*strptr); 1107: strptr++; 1.33 cvs 1108: } 1.25 cvs 1109: 1110: /* Does the URL contain a full domain name? This also works for a 1111: numerical host name. The domain name is already made lower-case 1112: and without a trailing dot. */ 1113: { 1.33 cvs 1114: char *dot = port ? port : path; 1115: if (dot > *filename && *--dot=='.') 1116: { 1117: char *orig=dot, *dest=dot+1; 1118: while((*orig++ = *dest++)); 1119: if (port) port--; 1120: path--; 1.25 cvs 1121: } 1122: } 1123: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */ 1.33 cvs 1124: if (port) 1125: { 1126: if (!*(port+1) || *(port+1)==used_sep) 1127: { 1128: if (!newname) 1129: { 1.25 cvs 1130: char *orig=port, *dest=port+1; 1131: while((*orig++ = *dest++)); 1.33 cvs 1132: } 1133: } 1134: else if ((!strncmp(access, "http", 4) && 1135: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) || 1136: (!strncmp(access, "gopher", 6) && 1137: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) || 1138: (!strncmp(access, "ftp", 3) && 1139: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) { 1140: if (!newname) 1141: { 1142: char *orig=port, *dest=port+3; 1143: while((*orig++ = *dest++)); 1144: /* Update path position, Henry Minsky */ 1145: path -= 3; 1.25 cvs 1146: } 1.33 cvs 1147: } 1148: else if (newname) 1149: strncat(newname, port, (int) (path-port)); 1150: } 1.25 cvs 1151: 1.33 cvs 1152: if (newname) 1153: { 1.25 cvs 1154: char *newpath = newname+strlen(newname); 1155: strcat(newname, path); 1156: path = newpath; 1.28 cvs 1157: /* Free old copy */ 1158: TtaFreeMemory(*filename); 1.25 cvs 1159: *filename = newname; 1.33 cvs 1160: } 1.25 cvs 1161: return path; 1162: } 1163: 1164: 1165: /*---------------------------------------------------------------------- 1.29 cvs 1166: SimplifyUrl: simplify a URI 1.32 cvs 1167: A URI is allowed to contain the sequence xxx/../ which may be 1168: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR. 1.28 cvs 1169: Simplification helps us recognize duplicate URIs. 1.25 cvs 1170: 1.28 cvs 1171: Thus, /etc/junk/../fred becomes /etc/fred 1172: /etc/junk/./fred becomes /etc/junk/fred 1.25 cvs 1173: 1.28 cvs 1174: but we should NOT change 1175: http://fred.xxx.edu/../.. 1.25 cvs 1176: 1177: or ../../albert.html 1178: 1.28 cvs 1179: In order to avoid empty URLs the following URLs become: 1.25 cvs 1180: 1181: /fred/.. becomes /fred/.. 1182: /fred/././.. becomes /fred/.. 1183: /fred/.././junk/.././ becomes /fred/.. 1184: 1.28 cvs 1185: If more than one set of `://' is found (several proxies in cascade) then 1186: only the part after the last `://' is simplified. 1.25 cvs 1187: 1.28 cvs 1188: Returns: A string which might be the old one or a new one. 1.25 cvs 1189: ----------------------------------------------------------------------*/ 1190: #ifdef __STDC__ 1.29 cvs 1191: void SimplifyUrl (char ** url) 1.25 cvs 1192: #else /* __STDC__ */ 1.29 cvs 1193: void SimplifyUrl (url) 1.28 cvs 1194: char **url; 1.25 cvs 1195: #endif /* __STDC__ */ 1196: { 1.28 cvs 1197: char *path, *p; 1198: char *newptr, *access; 1199: char *orig, *dest, *end; 1200: 1.33 cvs 1201: char used_sep; 1.32 cvs 1202: 1203: 1.28 cvs 1204: if (!url || !*url) 1205: return; 1206: 1.33 cvs 1207: if (strchr (*url, URL_SEP)) 1208: { 1209: used_sep = URL_SEP; 1210: } 1.32 cvs 1211: else 1.33 cvs 1212: { 1213: used_sep = DIR_SEP; 1214: } 1.32 cvs 1215: 1.28 cvs 1216: /* Find any scheme name */ 1217: if ((path = strstr(*url, "://")) != NULL) 1.33 cvs 1218: { 1219: /* Find host name */ 1.28 cvs 1220: access = *url; 1221: while (access<path && (*access=tolower(*access))) 1222: access++; 1223: path += 3; 1224: while ((newptr = strstr(path, "://")) != NULL) 1225: /* For proxies */ 1226: path = newptr+3; 1227: /* We have a host name */ 1228: path = HTCanon(url, path); 1.25 cvs 1229: } 1.28 cvs 1230: else if ((path = strstr(*url, ":/")) != NULL) 1231: path += 2; 1232: else 1233: path = *url; 1.25 cvs 1234: 1.33 cvs 1235: if (*path == used_sep && *(path+1)==used_sep) 1.28 cvs 1236: /* Some URLs start //<foo> */ 1237: path += 1; 1238: else if (!strncmp(path, "news:", 5)) 1239: { 1240: newptr = strchr(path+5, '@'); 1241: if (!newptr) 1242: newptr = path + 5; 1243: while (*newptr) 1244: { 1245: /* Make group or host lower case */ 1246: *newptr = tolower (*newptr); 1247: newptr++; 1.25 cvs 1248: } 1.28 cvs 1249: /* Doesn't need to do any more */ 1250: return; 1.25 cvs 1251: } 1.28 cvs 1252: 1253: if ((p = path)) 1254: { 1255: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) || 1256: (end = strchr (path, '#')))) 1257: end = path + strlen (path); 1258: 1259: /* Parse string second time to simplify */ 1260: p = path; 1261: while (p < end) 1262: { 1.33 cvs 1263: if (*p==used_sep) 1.28 cvs 1264: { 1.33 cvs 1265: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2))) 1.28 cvs 1266: { 1267: orig = p + 1; 1.33 cvs 1268: dest = (*(p+2)!=used_sep) ? p+2 : p+3; 1.28 cvs 1269: while ((*orig++ = *dest++)); /* Remove a slash and a dot */ 1270: end = orig - 1; 1271: } 1.33 cvs 1272: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3))) 1.28 cvs 1273: { 1274: newptr = p; 1.33 cvs 1275: while (newptr>path && *--newptr!=used_sep); /* prev slash */ 1.28 cvs 1276: if (strncmp(newptr, "/../", 4)) 1277: { 1278: orig = newptr + 1; 1.33 cvs 1279: dest = (*(p+3)!=used_sep) ? p+3 : p+4; 1.28 cvs 1280: while ((*orig++ = *dest++)); /* Remove /xxx/.. */ 1281: end = orig-1; 1282: /* Start again with prev slash */ 1283: p = newptr; 1.25 cvs 1284: } 1.28 cvs 1285: else 1.25 cvs 1286: p++; 1.28 cvs 1287: } 1.33 cvs 1288: else if (*(p+1) == used_sep) 1.28 cvs 1289: { 1.33 cvs 1290: while (*(p+1) == used_sep) 1.28 cvs 1291: { 1292: orig = p; 1293: dest = p + 1; 1294: while ((*orig++ = *dest++)); /* Remove multiple /'s */ 1295: end = orig-1; 1296: } 1297: } 1298: else 1.25 cvs 1299: p++; 1.28 cvs 1300: } 1301: else 1302: p++; 1.25 cvs 1303: } 1304: } 1.28 cvs 1305: return; 1306: } 1307: 1308: 1309: /*---------------------------------------------------------------------- 1310: NormalizeFile normalizes local names. 1311: Return TRUE if target and src differ. 1312: ----------------------------------------------------------------------*/ 1313: #ifdef __STDC__ 1314: boolean NormalizeFile (char *src, char *target) 1315: #else 1316: boolean NormalizeFile (src, target) 1317: char *src; 1318: char *target; 1319: 1320: #endif 1321: { 1322: char *s; 1323: boolean change; 1324: 1325: change = FALSE; 1326: if (src[0] == '~') 1327: { 1328: /* replace ~ */ 1329: s = (char *) TtaGetEnvString ("HOME"); 1330: strcpy (target, s); 1331: strcat (target, &src[1]); 1332: change = TRUE; 1333: } 1334: else if (strncmp (src, "file:", 5) == 0) 1335: { 1336: /* remove the prefix file: */ 1337: if (src[5] == EOS) 1338: strcpy (target, DIR_STR); 1339: else if (src[0] == '~') 1340: { 1341: /* replace ~ */ 1342: s = (char *) TtaGetEnvString ("HOME"); 1343: strcpy (target, s); 1344: strcat (target, &src[5]); 1345: } 1346: else 1347: strcpy (target, &src[5]); 1348: change = TRUE; 1349: } 1350: else 1351: strcpy (target, src); 1352: 1353: /* remove /../ and /./ */ 1.29 cvs 1354: SimplifyUrl (&target); 1.28 cvs 1355: return (change); 1.25 cvs 1356: } 1357: 1.28 cvs 1358: 1.25 cvs 1359: /*---------------------------------------------------------------------- 1.31 cvs 1360: MakeRelativeURL: make relative name 1.25 cvs 1361: 1.28 cvs 1362: This function creates and returns a string which gives an expression of 1363: one address as related to another. Where there is no relation, an absolute 1364: address is retured. 1.25 cvs 1365: 1.28 cvs 1366: On entry, 1.25 cvs 1367: Both names must be absolute, fully qualified names of nodes 1368: (no fragment bits) 1369: 1.28 cvs 1370: On exit, 1.25 cvs 1371: The return result points to a newly allocated name which, if 1372: parsed by AmayaParseUrl relative to relatedName, will yield aName. 1373: The caller is responsible for freeing the resulting name later. 1374: ----------------------------------------------------------------------*/ 1375: #ifdef __STDC__ 1.31 cvs 1376: char *MakeRelativeURL (char *aName, char *relatedName) 1.25 cvs 1377: #else /* __STDC__ */ 1.31 cvs 1378: char *MakeRelativeURL (aName, relatedName) 1.28 cvs 1379: char *aName; 1380: char *relatedName; 1.25 cvs 1381: #endif /* __STDC__ */ 1382: { 1.39 cvs 1383: char *return_value; 1384: char result[MAX_LENGTH]; 1.29 cvs 1385: char *p; 1.34 cvs 1386: char *q; 1.29 cvs 1387: char *after_access; 1388: char *last_slash = NULL; 1389: int slashes, levels, len; 1390: 1.44 cvs 1391: # ifdef _WINDOWS 1392: int ndx; 1393: # endif /* _WINDOWS */ 1394: 1.29 cvs 1395: if (aName == NULL || relatedName == NULL) 1396: return (NULL); 1397: 1398: slashes = 0; 1399: after_access = NULL; 1400: p = aName; 1401: q = relatedName; 1402: for (; *p && (*p == *q); p++, q++) 1.27 cvs 1403: { 1404: /* Find extent of match */ 1.32 cvs 1405: if (*p == ':') 1.29 cvs 1406: after_access = p + 1; 1.28 cvs 1407: if (*p == DIR_SEP) 1.27 cvs 1408: { 1.29 cvs 1409: /* memorize the last slash position and count them */ 1.27 cvs 1410: last_slash = p; 1411: slashes++; 1.25 cvs 1412: } 1413: } 1414: 1.31 cvs 1415: /* q, p point to the first non-matching character or zero */ 1416: if (*q == EOS) 1417: { 1418: /* New name is a subset of the related name */ 1419: /* exactly the right length */ 1420: len = strlen (p); 1421: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1422: strcpy (return_value, p); 1423: } 1424: else if ((slashes < 2 && after_access == NULL) 1425: || (slashes < 3 && after_access != NULL)) 1426: { 1427: /* Two names whitout common path */ 1428: /* exactly the right length */ 1429: len = strlen (aName); 1430: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1431: strcpy (return_value, aName); 1432: } 1433: else 1434: { 1435: /* Some path in common */ 1436: if (slashes == 3 && strncmp (aName, "http:", 5) == 0) 1437: /* just the same server */ 1438: strcpy (result, last_slash); 1439: else 1440: { 1441: levels= 0; 1442: for (; *q && (*q != '#'); q++) 1443: if (*q == DIR_SEP) 1444: levels++; 1445: 1446: result[0] = 0; 1447: for (;levels; levels--) 1448: strcat (result, "../"); 1449: strcat (result, last_slash+1); 1450: } 1451: 1452: /* exactly the right length */ 1453: len = strlen (result); 1454: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1455: strcpy (return_value, result); 1.25 cvs 1456: } 1.44 cvs 1457: # ifdef _WINDOWS 1458: len = strlen (return_value); 1459: for (ndx = 0; ndx < len; ndx ++) 1460: if (return_value[ndx] == '\\') 1461: return_value[ndx] = '/' ; 1462: # endif /* _WINDOWS */ 1.29 cvs 1463: return (return_value); 1.24 cvs 1464: } 1.35 cvs 1465: 1466: