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 14: * 15: */ 1.7 cvs 16: 1.15 cvs 17: #define THOT_EXPORT extern 1.3 cvs 18: #include "amaya.h" 19: 1.8 cvs 20: #include "init_f.h" 21: #include "AHTURLTools_f.h" 22: 1.24 cvs 23: #define MAX_PRINT_URL_LENGTH 50 1.29 cvs 24: typedef struct _HTURI { 25: char * access; /* Now known as "scheme" */ 26: char * host; 27: char * absolute; 28: char * relative; 29: char * fragment; 30: } HTURI; 1.24 cvs 31: 1.28 cvs 32: 33: /*---------------------------------------------------------------------- 34: ConvertToLowerCase 35: Converts a string to lowercase. 36: ----------------------------------------------------------------------*/ 1.22 cvs 37: #ifdef __STDC__ 1.38 cvs 38: void ConvertToLowerCase (char *string) 1.28 cvs 39: #else /* __STDC__ */ 1.38 cvs 40: void ConvertToLowerCase (string) 1.28 cvs 41: char *string; 42: 43: #endif /* __STDC__ */ 44: { 45: int i; 46: 47: if (!string) 48: return; 49: 50: for (i = 0; string[i] != EOS; i++) 51: string[i] = tolower (string[i]); 52: } 1.22 cvs 53: 1.8 cvs 54: /*---------------------------------------------------------------------- 1.11 cvs 55: ExplodeURL 1.8 cvs 56: ----------------------------------------------------------------------*/ 57: #ifdef __STDC__ 58: void ExplodeURL (char *url, char **proto, char **host, char **dir, char **file) 59: #else 60: void ExplodeURL (url, proto, host, dir, file) 61: char *url; 62: char **proto; 63: char **host; 64: char **dir; 65: char **file; 66: 67: #endif 68: { 1.33 cvs 69: char *curr, *temp; 70: char used_sep; 1.32 cvs 71: 1.33 cvs 72: if (url && strchr (url, URL_SEP)) 73: used_sep = URL_SEP; 74: else 75: used_sep = DIR_SEP; 1.8 cvs 76: 77: if ((url == NULL) || (proto == NULL) || (host == NULL) || 78: (dir == NULL) || (file == NULL)) 79: return; 80: 81: /* initialize every pointer */ 82: *proto = *host = *dir = *file = NULL; 83: 84: /* skip any leading space */ 85: while ((*url == SPACE) || (*url == TAB)) 86: url++; 1.9 cvs 87: curr = url; 88: if (*curr == 0) 1.8 cvs 89: goto finished; 90: 91: /* go to the end of the URL */ 1.9 cvs 92: while ((*curr != 0) && (*curr != SPACE) && (*curr != '\b') && 93: (*curr != '\r') && (*curr != EOL)) 94: curr++; 1.8 cvs 95: 96: /* mark the end of the chain */ 1.9 cvs 97: *curr = EOS; 98: curr--; 99: if (curr <= url) 1.8 cvs 100: goto finished; 101: 102: /* search the next DIR_SEP indicating the beginning of the file name */ 103: do 1.11 cvs 104: curr--; 1.33 cvs 105: while ((curr >= url) && (*curr != used_sep)); 1.11 cvs 106: 1.9 cvs 107: if (curr < url) 1.8 cvs 108: goto finished; 1.9 cvs 109: *file = curr + 1; 1.8 cvs 110: 111: /* mark the end of the dir */ 1.9 cvs 112: *curr = EOS; 113: curr--; 114: if (curr < url) 1.8 cvs 115: goto finished; 116: 1.29 cvs 117: /* search for the DIR_STR indicating the host name start */ 1.33 cvs 118: while ((curr > url) && ((*curr != used_sep) || (*(curr + 1) != used_sep))) 1.9 cvs 119: curr--; 1.8 cvs 120: 121: /* if we found it, separate the host name from the directory */ 1.33 cvs 122: if ((*curr == DIR_SEP) && (*(curr + 1) == used_sep)) 1.8 cvs 123: { 1.9 cvs 124: *host = temp = curr + 2; 1.33 cvs 125: while ((*temp != 0) && (*temp != used_sep)) 1.8 cvs 126: temp++; 1.33 cvs 127: if (*temp == used_sep) 1.8 cvs 128: { 129: *temp = EOS; 130: *dir = temp + 1; 131: } 132: } 133: else 1.11 cvs 134: *dir = curr; 135: 1.9 cvs 136: if (curr <= url) 1.8 cvs 137: goto finished; 138: 139: /* mark the end of the proto */ 1.9 cvs 140: *curr = EOS; 141: curr--; 142: if (curr < url) 1.8 cvs 143: goto finished; 144: 1.32 cvs 145: if (*curr == ':') 1.8 cvs 146: { 1.9 cvs 147: *curr = EOS; 148: curr--; 1.8 cvs 149: } 150: else 151: goto finished; 1.11 cvs 152: 1.9 cvs 153: if (curr < url) 1.8 cvs 154: goto finished; 1.9 cvs 155: while ((curr > url) && (isalpha (*curr))) 156: curr--; 157: *proto = curr; 1.8 cvs 158: 159: finished:; 160: 161: #ifdef AMAYA_DEBUG 162: fprintf (stderr, "ExplodeURL(%s)\n\t", url); 163: if (*proto) 164: fprintf (stderr, "proto : %s, ", *proto); 165: if (*host) 166: fprintf (stderr, "host : %s, ", *host); 167: if (*dir) 168: fprintf (stderr, "dir : %s, ", *dir); 169: if (*file) 170: fprintf (stderr, "file : %s ", *file); 171: fprintf (stderr, "\n"); 172: #endif 173: 174: } 1.3 cvs 175: 1.4 cvs 176: /*---------------------------------------------------------------------- 1.9 cvs 177: IsHTMLName 178: returns TRUE if path points to an HTML resource. 1.4 cvs 179: ----------------------------------------------------------------------*/ 1.3 cvs 180: #ifdef __STDC__ 1.34 cvs 181: boolean IsHTMLName (const char *path) 1.3 cvs 182: #else /* __STDC__ */ 183: boolean IsHTMLName (path) 1.34 cvs 184: const char *path; 1.3 cvs 185: #endif /* __STDC__ */ 186: { 1.5 cvs 187: char temppath[MAX_LENGTH]; 188: char suffix[MAX_LENGTH]; 189: char nsuffix[MAX_LENGTH]; 190: int i; 191: 192: if (!path) 1.37 cvs 193: return (FALSE); 1.5 cvs 194: 195: strcpy (temppath, path); 196: ExtractSuffix (temppath, suffix); 197: 198: /* Normalize the suffix */ 199: i = 0; 1.39 cvs 200: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 201: { 1.25 cvs 202: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 203: i++; 204: } 1.5 cvs 205: nsuffix[i] = EOS; 1.39 cvs 206: if ((!strcmp (nsuffix, "html")) || 207: (!strcmp (nsuffix, "htm")) || 208: (!strcmp (nsuffix, "shtml"))) 209: return (TRUE); 1.22 cvs 210: else if (!strcmp (nsuffix, "gz")) 1.13 cvs 211: { 1.39 cvs 212: /* take into account compressed files */ 1.13 cvs 213: ExtractSuffix (temppath, suffix); 214: /* Normalize the suffix */ 215: i = 0; 1.39 cvs 216: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 217: { 1.25 cvs 218: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 219: i++; 220: } 221: nsuffix[i] = EOS; 1.39 cvs 222: if ((!strcmp (nsuffix, "html")) || 223: (!strcmp (nsuffix, "htm")) || 224: (!strcmp (nsuffix, "shtml"))) 225: return (TRUE); 226: else 1.13 cvs 227: return (FALSE); 228: } 229: else 1.39 cvs 230: return (FALSE); 1.3 cvs 231: } 232: 1.4 cvs 233: /*---------------------------------------------------------------------- 1.9 cvs 234: IsImageName 235: returns TRUE if path points to an image resource. 1.4 cvs 236: ----------------------------------------------------------------------*/ 1.3 cvs 237: #ifdef __STDC__ 1.34 cvs 238: boolean IsImageName (const char *path) 1.3 cvs 239: #else /* __STDC__ */ 240: boolean IsImageName (path) 1.34 cvs 241: const char *path; 1.3 cvs 242: #endif /* __STDC__ */ 243: { 1.5 cvs 244: char temppath[MAX_LENGTH]; 245: char suffix[MAX_LENGTH]; 246: char nsuffix[MAX_LENGTH]; 247: int i; 248: 249: if (!path) 1.13 cvs 250: return (FALSE); 1.5 cvs 251: 252: strcpy (temppath, path); 253: ExtractSuffix (temppath, suffix); 254: 255: /* Normalize the suffix */ 256: i = 0; 1.39 cvs 257: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 258: { 1.25 cvs 259: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 260: i++; 261: } 1.5 cvs 262: nsuffix[i] = EOS; 1.39 cvs 263: if ((!strcmp (nsuffix, "gif")) || (!strcmp (nsuffix, "xbm")) || 264: (!strcmp (nsuffix, "xpm")) || (!strcmp (nsuffix, "jpg")) || 265: (!strcmp (nsuffix, "png")) || (!strcmp (nsuffix, "au"))) 266: return (TRUE); 267: return (FALSE); 1.3 cvs 268: } 269: 1.4 cvs 270: /*---------------------------------------------------------------------- 1.9 cvs 271: IsTextName 1.4 cvs 272: ----------------------------------------------------------------------*/ 1.3 cvs 273: #ifdef __STDC__ 1.34 cvs 274: boolean IsTextName (const char *path) 1.3 cvs 275: #else /* __STDC__ */ 276: boolean IsTextName (path) 1.34 cvs 277: const char *path; 1.3 cvs 278: 279: #endif /* __STDC__ */ 280: { 1.5 cvs 281: char temppath[MAX_LENGTH]; 282: char suffix[MAX_LENGTH]; 283: char nsuffix[MAX_LENGTH]; 284: int i; 285: 286: if (!path) 1.13 cvs 287: return (FALSE); 1.5 cvs 288: 289: strcpy (temppath, path); 290: ExtractSuffix (temppath, suffix); 291: 292: /* Normalize the suffix */ 293: i = 0; 1.39 cvs 294: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.5 cvs 295: { 1.25 cvs 296: nsuffix[i] = tolower (suffix[i]); 1.5 cvs 297: i++; 298: } 299: nsuffix[i] = EOS; 300: 1.39 cvs 301: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd"))) 1.13 cvs 302: return (TRUE); 1.22 cvs 303: else if (!strcmp (nsuffix, "gz")) 1.13 cvs 304: { 1.39 cvs 305: /* take into account compressed files */ 1.13 cvs 306: ExtractSuffix (temppath, suffix); 307: /* Normalize the suffix */ 308: i = 0; 1.39 cvs 309: while (suffix[i] != EOS && i < MAX_LENGTH -1) 1.13 cvs 310: { 1.25 cvs 311: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 312: i++; 313: } 314: nsuffix[i] = EOS; 1.39 cvs 315: if ((!strcmp (nsuffix, "txt")) || (!strcmp (nsuffix, "dtd"))) 1.13 cvs 316: return (TRUE); 317: else 318: return (FALSE); 319: } 320: else 321: return (FALSE); 1.3 cvs 322: } 323: 1.4 cvs 324: /*---------------------------------------------------------------------- 1.9 cvs 325: IsHTTPPath 326: returns TRUE if path is in fact an http URL. 1.4 cvs 327: ----------------------------------------------------------------------*/ 1.3 cvs 328: #ifdef __STDC__ 1.34 cvs 329: boolean IsHTTPPath (const char *path) 1.3 cvs 330: #else /* __STDC__ */ 331: boolean IsHTTPPath (path) 1.34 cvs 332: const char *path; 1.3 cvs 333: #endif /* __STDC__ */ 334: { 1.5 cvs 335: if (!path) 336: return FALSE; 1.3 cvs 337: 1.5 cvs 338: if (strncmp (path, "http:", 5) != 0) 339: return FALSE; 340: return TRUE; 1.3 cvs 341: } 342: 1.4 cvs 343: /*---------------------------------------------------------------------- 1.9 cvs 344: IsWithParameters 345: returns TRUE if url has a concatenated query string. 1.4 cvs 346: ----------------------------------------------------------------------*/ 1.3 cvs 347: #ifdef __STDC__ 1.34 cvs 348: boolean IsWithParameters (const char *url) 1.3 cvs 349: #else /* __STDC__ */ 1.9 cvs 350: boolean IsWithParameters (url) 1.34 cvs 351: const char *url; 1.3 cvs 352: #endif /* __STDC__ */ 353: { 1.5 cvs 354: int i; 1.3 cvs 355: 1.9 cvs 356: if ((!url) || (url[0] == EOS)) 1.5 cvs 357: return FALSE; 1.3 cvs 358: 1.9 cvs 359: i = strlen (url) - 1; 360: while (i > 0 && url[i--] != '?') 1.5 cvs 361: if (i < 0) 362: return FALSE; 1.3 cvs 363: 1.5 cvs 364: /* There is a parameter */ 365: return TRUE; 1.3 cvs 366: } 367: 1.4 cvs 368: /*---------------------------------------------------------------------- 1.9 cvs 369: IsW3Path 370: returns TRUE if path is in fact a URL. 1.4 cvs 371: ----------------------------------------------------------------------*/ 1.3 cvs 372: #ifdef __STDC__ 1.34 cvs 373: boolean IsW3Path (const char *path) 1.3 cvs 374: #else /* __STDC__ */ 375: boolean IsW3Path (path) 1.34 cvs 376: const char *path; 1.3 cvs 377: #endif /* __STDC__ */ 378: { 1.5 cvs 379: if ((strncmp (path, "http:", 5)) && (strncmp (path, "ftp:", 4)) && 380: (strncmp (path, "telnet:", 7)) && (strncmp (path, "wais:", 5)) && 381: (strncmp (path, "news:", 5)) && (strncmp (path, "gopher:", 7)) && 382: (strncmp (path, "mailto:", 7)) && (strncmp (path, "archie:", 7))) 383: return FALSE; 384: return TRUE; 1.3 cvs 385: } 386: 1.4 cvs 387: /*---------------------------------------------------------------------- 1.9 cvs 388: IsValidProtocol 389: returns true if the url protocol is supported by Amaya. 1.4 cvs 390: ----------------------------------------------------------------------*/ 1.3 cvs 391: #ifdef __STDC__ 1.34 cvs 392: boolean IsValidProtocol (const char *url) 1.3 cvs 393: #else /* __STDC__ */ 1.9 cvs 394: boolean IsValidProtocol (url) 1.34 cvs 395: const char *url; 1.3 cvs 396: #endif /* __STDC__ */ 397: { 1.26 cvs 398: if (!strncmp (url, "http:", 5)) 1.22 cvs 399: /* experimental */ 1.26 cvs 400: /*** || !strncmp (url, "ftp:", 4)) ***/ 1.24 cvs 401: /*** || !strncmp (path, "news:", 5)***/ 1.8 cvs 402: return (TRUE); 1.5 cvs 403: else 1.8 cvs 404: return (FALSE); 1.3 cvs 405: } 406: 1.31 cvs 407: 408: /*---------------------------------------------------------------------- 409: GetBaseURL 410: normalizes orgName according to a base associated with doc, and 411: following the standard URL format rules. 412: The function returns the base used to solve relative URL and SRC: 413: - the base of the document, 414: - or the document path (without document name). 415: ----------------------------------------------------------------------*/ 416: #ifdef __STDC__ 417: char *GetBaseURL (Document doc) 418: #else /* __STDC__ */ 419: char *GetBaseURL (doc) 420: Document doc; 421: #endif /* __STDC__ */ 422: { 423: Element el; 424: ElementType elType; 425: AttributeType attrType; 426: Attribute attr; 1.34 cvs 427: char used_sep = URL_SEP; 1.33 cvs 428: char *used_str; 1.31 cvs 429: char *ptr, *basename; 430: int length; 431: 432: basename = TtaGetMemory (MAX_LENGTH); 1.39 cvs 433: strncpy (basename, DocumentURLs[doc], MAX_LENGTH-1); 434: basename[MAX_LENGTH-1] = EOS; 1.31 cvs 435: length = MAX_LENGTH -1; 436: /* get the root element */ 437: el = TtaGetMainRoot (doc); 438: /* search the BASE element */ 439: elType.ElSSchema = TtaGetDocumentSSchema (doc); 440: elType.ElTypeNum = HTML_EL_BASE; 441: el = TtaSearchTypedElement (elType, SearchInTree, el); 442: if (el) 443: { 444: /* The document has a BASE element -> Get the HREF attribute */ 445: attrType.AttrSSchema = elType.ElSSchema; 446: attrType.AttrTypeNum = HTML_ATTR_HREF_; 447: attr = TtaGetAttribute (el, attrType); 448: if (attr) 449: { 450: /* Use the base path of the document */ 451: TtaGiveTextAttributeValue (attr, basename, &length); 452: /* base and orgName have to be separated by a DIR_SEP */ 453: length--; 1.32 cvs 454: 1.33 cvs 455: if (basename [0] != EOS && strchr (basename, URL_SEP)) 456: { 457: used_str = URL_STR; 458: used_sep = URL_SEP; 459: } 460: else 461: { 462: used_str = DIR_STR; 463: used_sep = DIR_SEP; 464: } 1.32 cvs 465: 1.33 cvs 466: if (basename[0] != EOS && basename[length] != used_sep) 1.31 cvs 467: /* verify if the base has the form "protocol://server:port" */ 468: { 1.33 cvs 469: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS | 470: AMAYA_PARSE_HOST | 471: AMAYA_PARSE_PUNCTUATION); 1.31 cvs 472: if (ptr && !strcmp (ptr, basename)) 473: { 474: /* it has this form, we complete it by adding a DIR_STR */ 1.33 cvs 475: strcat (basename, used_str); 1.31 cvs 476: length++; 477: } 478: if (ptr) 479: TtaFreeMemory (ptr); 480: } 481: } 1.33 cvs 482: } 1.32 cvs 483: else 1.33 cvs 484: { 485: if (basename [0] != EOS && strchr (basename, URL_SEP)) 486: { 487: used_str = URL_STR; 488: used_sep = URL_SEP; 489: } 490: else 491: { 492: used_str = DIR_STR; 493: used_sep = DIR_SEP; 494: } 495: } 496: 1.31 cvs 497: /* Remove anything after the last DIR_SEP char. If no such char is found, 498: * then search for the first ":" char, hoping that what's before that is a 499: * protocol. If found, end the string there. If neither char is found, 500: * then discard the whole base element. 501: */ 502: length = strlen (basename) - 1; 503: /* search for the last DIR_SEP char */ 1.33 cvs 504: while (length >= 0 && basename[length] != used_sep) 1.31 cvs 505: length--; 506: if (length >= 0) 507: /* found the last DIR_SEP char, end the string there */ 508: basename[length + 1] = EOS; 509: else 510: /* search for the first PATH_STR char */ 511: { 1.32 cvs 512: for (length = 0; basename[length] != ':' && 1.31 cvs 513: basename[length] != EOS; length ++); 1.32 cvs 514: if (basename[length] == ':') 1.31 cvs 515: /* found, so end the string there */ 516: basename[length + 1] = EOS; 517: else 518: /* not found, discard the base */ 519: basename[0] = EOS; 520: } 521: return (basename); 522: } 523: 524: 1.4 cvs 525: /*---------------------------------------------------------------------- 1.40 ! cvs 526: GetLocalPath ! 527: Allocate and return the local document path associated to the url ! 528: ----------------------------------------------------------------------*/ ! 529: #ifdef __STDC__ ! 530: char *GetLocalPath (Document doc, char *url) ! 531: #else /* __STDC__ */ ! 532: char *GetLocalPath (doc, url) ! 533: Document doc; ! 534: char *url; ! 535: #endif /* __STDC__ */ ! 536: { ! 537: char *ptr; ! 538: char *documentname; ! 539: int len; ! 540: boolean noFile; ! 541: ! 542: if (url != NULL) ! 543: { ! 544: ptr = TtaGetMemory (MAX_LENGTH); ! 545: documentname = TtaGetMemory (MAX_LENGTH); ! 546: /* check whether the file name exists */ ! 547: len = strlen (url) - 1; ! 548: noFile = (url[len] == DIR_SEP); ! 549: if (noFile) ! 550: url[len] = EOS; ! 551: TtaExtractName (url, ptr, documentname); ! 552: sprintf (ptr, "%s%s%d%s", TempFileDirectory, DIR_STR, doc, DIR_STR); ! 553: if (!TtaCheckDirectory (ptr)) ! 554: /* directory did not exist */ ! 555: mkdir (ptr, S_IRWXU); ! 556: strcat (ptr, documentname); ! 557: TtaFreeMemory (documentname); ! 558: /* restore the url */ ! 559: if (noFile) ! 560: url[len] = DIR_SEP; ! 561: return (ptr); ! 562: } ! 563: else ! 564: return (NULL); ! 565: } ! 566: ! 567: ! 568: /*---------------------------------------------------------------------- 1.9 cvs 569: NormalizeURL 570: normalizes orgName according to a base associated with doc, and 571: following the standard URL format rules. 572: The function returns the new complete and normalized URL 1.12 cvs 573: or file name path (newName) and the name of the document (docName). 1.9 cvs 574: N.B. If the function can't find out what's the docName, it assigns 575: the name "noname.html". 1.4 cvs 576: ----------------------------------------------------------------------*/ 1.3 cvs 577: #ifdef __STDC__ 578: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName) 579: #else /* __STDC__ */ 580: void NormalizeURL (orgName, doc, newName, docName) 581: char *orgName; 582: Document doc; 583: char *newName; 584: char *docName; 585: #endif /* __STDC__ */ 586: { 1.31 cvs 587: char *basename; 1.18 cvs 588: char tempOrgName[MAX_LENGTH]; 1.5 cvs 589: char *ptr; 590: int length; 1.33 cvs 591: char used_sep; 592: char *used_str; 1.5 cvs 593: 594: if (!newName || !docName) 595: return; 1.18 cvs 596: 1.32 cvs 597: if (doc != 0) 1.33 cvs 598: basename = GetBaseURL (doc); 1.32 cvs 599: else 1.33 cvs 600: basename = (char *) NULL; 1.32 cvs 601: 1.33 cvs 602: if (strchr (orgName, URL_SEP) || (basename && strchr (basename, URL_SEP))) 603: { 604: used_str = URL_STR; 605: used_sep = URL_SEP; 606: } 607: else 608: { 609: used_str = DIR_STR; 610: used_sep = DIR_SEP; 611: } 1.32 cvs 612: 1.18 cvs 613: /* 1.31 cvs 614: * Clean orgName 615: * Make sure we have a complete orgName, without any leading or trailing 616: * white spaces, or trailinbg new lines 617: */ 1.18 cvs 618: 1.5 cvs 619: ptr = orgName; 1.18 cvs 620: /* skip leading white space and new line characters */ 1.19 cvs 621: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS); 1.39 cvs 622: strncpy (tempOrgName, ptr, MAX_LENGTH -1); 623: tempOrgName[MAX_LENGTH -1] = EOS; 1.18 cvs 624: /* clean trailing white space */ 625: ptr = strchr (tempOrgName, ' '); 626: if (ptr) 627: *ptr = EOS; 628: /* clean trailing new lines */ 1.19 cvs 629: ptr = strchr (tempOrgName, EOL); 1.5 cvs 630: if (ptr) 631: *ptr = EOS; 632: 1.18 cvs 633: /* 1.31 cvs 634: * Make orgName a complete URL 635: * If the URL does not include a protocol, then try to calculate 636: * one using the doc's base element (if it exists), 637: */ 1.21 cvs 638: if (tempOrgName[0] == EOS) 639: { 640: newName[0] = EOS; 1.36 cvs 641: TtaFreeMemory (basename); 1.21 cvs 642: return; 643: } 644: else if (IsW3Path (tempOrgName)) 645: { 646: /* the name is complete, go to the Sixth Step */ 647: strcpy (newName, tempOrgName); 648: /* verify if the URL has the form "protocol://server:port" */ 1.25 cvs 649: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | 650: AMAYA_PARSE_PUNCTUATION); 1.21 cvs 651: if (ptr && !strcmp (ptr, newName)) 652: { 1.29 cvs 653: /* it has this form, we complete it by adding a DIR_STR */ 1.33 cvs 654: strcat (newName, used_str); 1.21 cvs 655: } 656: if (ptr) 1.25 cvs 657: TtaFreeMemory (ptr); 1.21 cvs 658: } 659: else if ( doc == 0) 1.19 cvs 660: /* the name is complete, go to the Sixth Step */ 1.18 cvs 661: strcpy (newName, tempOrgName); 1.5 cvs 662: else 663: { 1.22 cvs 664: 1.31 cvs 665: /* Calculate the absolute URL, using the base or document URL */ 1.32 cvs 666: 1.25 cvs 667: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL); 1.14 cvs 668: if (ptr) 669: { 1.29 cvs 670: SimplifyUrl (&ptr); 1.14 cvs 671: strcpy (newName, ptr); 1.25 cvs 672: TtaFreeMemory (ptr); 1.14 cvs 673: } 674: else 1.18 cvs 675: newName[0] = EOS; 1.5 cvs 676: } 1.36 cvs 677: 678: TtaFreeMemory (basename); 1.5 cvs 679: 1.18 cvs 680: /* 1.31 cvs 681: * Prepare the docname that will refer to this ressource in the 682: * .amaya directory. If the new URL finishes on DIR_SEP, then use 683: * noname.html as a default ressource name 1.18 cvs 684: */ 1.19 cvs 685: if (newName[0] != EOS) 1.5 cvs 686: { 1.19 cvs 687: length = strlen (newName) - 1; 1.33 cvs 688: if (newName[length] == used_sep) 1.18 cvs 689: { 690: /* docname was not comprised inside the URL, so let's */ 691: /* assign the default ressource name */ 692: strcpy (docName, "noname.html"); 693: /* remove DIR_SEP at the end of complete path */ 1.23 cvs 694: /* newName[length] = EOS; */ 1.18 cvs 695: } 1.14 cvs 696: else 1.18 cvs 697: { 698: /* docname is comprised inside the URL */ 1.33 cvs 699: while (length >= 0 && newName[length] != used_sep) 1.18 cvs 700: length--; 701: if (length < 0) 702: strcpy (docName, newName); 703: else 704: strcpy (docName, &newName[length+1]); 705: } 1.19 cvs 706: 1.5 cvs 707: } 1.18 cvs 708: else 709: docName[0] = EOS; 710: } 1.3 cvs 711: 1.4 cvs 712: /*---------------------------------------------------------------------- 1.9 cvs 713: IsSameHost 1.4 cvs 714: ----------------------------------------------------------------------*/ 1.3 cvs 715: #ifdef __STDC__ 1.34 cvs 716: boolean IsSameHost (const char *url1, const char *url2) 1.3 cvs 717: #else /* __STDC__ */ 718: boolean IsSameHost (url1, url2) 1.34 cvs 719: const char *url1; 720: const char *url2; 1.3 cvs 721: #endif /* __STDC__ */ 722: { 1.34 cvs 723: char *basename_ptr1, *basename_ptr2; 724: boolean result; 1.3 cvs 725: 1.25 cvs 726: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 727: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 1.3 cvs 728: 1.5 cvs 729: if (strcmp (basename_ptr1, basename_ptr2)) 1.8 cvs 730: result = FALSE; 1.5 cvs 731: else 1.8 cvs 732: result = TRUE; 1.3 cvs 733: 1.25 cvs 734: TtaFreeMemory (basename_ptr1); 735: TtaFreeMemory (basename_ptr2); 1.5 cvs 736: return (result); 1.3 cvs 737: } 738: 739: 1.4 cvs 740: /*---------------------------------------------------------------------- 1.22 cvs 741: HasKnownFileSuffix 742: returns TRUE if path points to a file ending with a suffix. 743: ----------------------------------------------------------------------*/ 744: #ifdef __STDC__ 1.34 cvs 745: boolean HasKnownFileSuffix (const char *path) 1.22 cvs 746: #else /* __STDC__ */ 747: boolean HasKnownFileSuffix (path) 1.34 cvs 748: const char *path; 1.22 cvs 749: #endif /* __STDC__ */ 750: { 1.29 cvs 751: char *root; 752: char temppath[MAX_LENGTH]; 753: char suffix[MAX_LENGTH]; 1.22 cvs 754: 1.24 cvs 755: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP) 1.22 cvs 756: return (FALSE); 757: 1.29 cvs 758: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION); 1.22 cvs 759: 760: if (root) 761: { 762: strcpy (temppath, root); 1.25 cvs 763: TtaFreeMemory (root); 1.22 cvs 764: /* Get the suffix */ 765: ExtractSuffix (temppath, suffix); 766: 767: if( suffix[0] == EOS) 768: /* no suffix */ 769: return (FALSE); 770: 771: /* Normalize the suffix */ 772: ConvertToLowerCase (suffix); 773: 1.23 cvs 774: if (!strcmp (suffix, "gz")) 1.22 cvs 775: /* skip the compressed suffix */ 776: { 777: ExtractSuffix (temppath, suffix); 778: if(suffix[0] == EOS) 779: /* no suffix */ 780: return (FALSE); 781: /* Normalize the suffix */ 782: ConvertToLowerCase (suffix); 783: } 784: 785: if ((strcmp (suffix, "gif")) && (strcmp (suffix, "xbm")) && 786: (strcmp (suffix, "xpm")) && (strcmp (suffix, "jpg")) && 787: (strcmp (suffix, "pdf")) && (strcmp (suffix, "png")) && 788: (strcmp (suffix, "tgz")) && (strcmp (suffix, "xpg")) && 789: (strcmp (suffix, "xpd")) && (strcmp (suffix, "ps")) && 790: (strcmp (suffix, "au")) && (strcmp (suffix, "html")) && 791: (strcmp (suffix, "htm")) && (strcmp (suffix, "shtml")) && 792: (strcmp (suffix, "txt")) && (strcmp (suffix, "css")) && 793: (strcmp (suffix, "eps"))) 794: return (FALSE); 795: else 796: return (TRUE); 797: } 798: else 799: return (FALSE); 800: } 801: 802: 803: /*---------------------------------------------------------------------- 1.24 cvs 804: ChopURL 805: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL). 806: If inputURL is bigger than that size, outputURL receives 807: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...", 808: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL. 809: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets 810: copied into outputURL. 811: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH 812: chars. 813: ----------------------------------------------------------------------*/ 814: #ifdef __STDC__ 1.34 cvs 815: void ChopURL (char *outputURL, const char *inputURL) 1.24 cvs 816: #else 817: void ChopURL (outputURL, inputURL) 1.34 cvs 818: char *outputURL; 819: const char *inputURL; 1.24 cvs 820: #endif 1.22 cvs 821: 1.24 cvs 822: { 823: int len; 1.9 cvs 824: 1.24 cvs 825: len = strlen (inputURL); 826: if (len <= MAX_PRINT_URL_LENGTH) 1.29 cvs 827: strcpy (outputURL, inputURL); 1.24 cvs 828: else 829: /* make a truncated urlName on the status window */ 830: { 831: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2); 832: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS; 833: strcat (outputURL, "..."); 834: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ])); 835: } 1.25 cvs 836: } 837: 838: 839: /*---------------------------------------------------------------------- 840: scan 841: Scan a filename for its consituents 842: ----------------------------------- 843: 844: On entry, 845: name points to a document name which may be incomplete. 846: On exit, 847: absolute or relative may be nonzero (but not both). 848: host, fragment and access may be nonzero if they were specified. 849: Any which are nonzero point to zero terminated strings. 850: ----------------------------------------------------------------------*/ 851: #ifdef __STDC__ 852: static void scan (char * name, HTURI * parts) 853: #else /* __STDC__ */ 854: static void scan (name, parts) 855: char *name; 856: HTURI *parts; 857: 858: #endif /* __STDC__ */ 859: { 1.28 cvs 860: char * p; 861: char * after_access = name; 1.33 cvs 862: char used_sep; 1.32 cvs 863: 1.33 cvs 864: if (name && strchr (name, URL_SEP)) 865: { 866: used_sep = URL_SEP; 867: } 868: else 869: { 870: used_sep = DIR_SEP; 871: } 1.32 cvs 872: 1.28 cvs 873: memset(parts, '\0', sizeof(HTURI)); 874: /* Look for fragment identifier */ 875: if ((p = strrchr(name, '#')) != NULL) 876: { 877: *p++ = '\0'; 878: parts->fragment = p; 1.25 cvs 879: } 880: 1.28 cvs 881: for (p=name; *p; p++) 882: { 1.33 cvs 883: if (*p==used_sep || *p=='#' || *p=='?') 1.28 cvs 884: break; 1.32 cvs 885: if (*p==':') 1.28 cvs 886: { 887: *p = 0; 888: parts->access = after_access; /* Scheme has been specified */ 889: 890: /* The combination of gcc, the "-O" flag and the HP platform is 891: unhealthy. The following three lines is a quick & dirty fix, but is 892: not recommended. Rather, turn off "-O". */ 893: 894: /* after_access = p;*/ 895: /* while (*after_access == 0)*/ 896: /* after_access++;*/ 897: after_access = p+1; 898: if (0==strcasecmp("URL", parts->access)) 899: /* Ignore IETF's URL: pre-prefix */ 900: parts->access = NULL; 901: else 1.25 cvs 902: break; 903: } 904: } 905: 906: p = after_access; 1.33 cvs 907: if (*p==used_sep) 1.28 cvs 908: { 1.33 cvs 909: if (p[1]==used_sep) 1.28 cvs 910: { 1.25 cvs 911: parts->host = p+2; /* host has been specified */ 1.28 cvs 912: *p = 0; /* Terminate access */ 913: /* look for end of host name if any */ 1.33 cvs 914: p = strchr(parts->host,used_sep); 1.28 cvs 915: if (p) 916: { 1.25 cvs 917: *p=0; /* Terminate host */ 918: parts->absolute = p+1; /* Root has been found */ 1.28 cvs 919: } 920: } 921: else 922: /* Root found but no host */ 923: parts->absolute = p+1; 924: } 925: else 926: { 1.25 cvs 927: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */ 1.28 cvs 928: } 1.25 cvs 929: } 930: 931: 932: /*---------------------------------------------------------------------- 1.28 cvs 933: AmayaParseUrl: parse a Name relative to another name 934: 935: This returns those parts of a name which are given (and requested) 936: substituting bits from the related name where necessary. 1.25 cvs 937: 1.28 cvs 938: On entry, 1.25 cvs 939: aName A filename given 940: relatedName A name relative to which aName is to be parsed. Give 941: it an empty string if aName is absolute. 942: wanted A mask for the bits which are wanted. 943: 1.28 cvs 944: On exit, 1.25 cvs 945: returns A pointer to a malloc'd string which MUST BE FREED 946: ----------------------------------------------------------------------*/ 947: #ifdef __STDC__ 1.34 cvs 948: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted) 1.25 cvs 949: #else /* __STDC__ */ 1.28 cvs 950: char *AmayaParseUrl (aName, relatedName, wanted) 1.34 cvs 951: const char *aName; 1.28 cvs 952: char *relatedName; 953: int wanted; 1.25 cvs 954: 955: #endif /* __STDC__ */ 956: { 1.29 cvs 957: char *return_value; 958: char result[MAX_LENGTH]; 959: char name[MAX_LENGTH]; 960: char rel[MAX_LENGTH]; 961: char *p, *access; 962: HTURI given, related; 963: int len; 1.33 cvs 964: char used_sep; 965: char *used_str; 1.32 cvs 966: 1.33 cvs 967: if (strchr (aName, URL_SEP) || strchr (relatedName, URL_SEP) ) 968: { 969: used_str = URL_STR; 970: used_sep = URL_SEP; 971: } 1.32 cvs 972: else 1.33 cvs 973: { 974: used_str = DIR_STR; 975: used_sep = DIR_SEP; 976: } 1.32 cvs 977: 1.29 cvs 978: /* Make working copies of input strings to cut up: */ 979: return_value = NULL; 980: result[0] = 0; /* Clear string */ 981: strcpy (name, aName); 982: if (relatedName != NULL) 983: strcpy (rel, relatedName); 984: else 985: relatedName[0] = EOS; 986: 987: scan (name, &given); 988: scan (rel, &related); 989: access = given.access ? given.access : related.access; 990: if (wanted & AMAYA_PARSE_ACCESS) 991: if (access) 992: { 993: strcat (result, access); 994: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.32 cvs 995: strcat (result, ":"); 1.29 cvs 996: } 997: 998: if (given.access && related.access) 999: /* If different, inherit nothing. */ 1000: if (strcmp (given.access, related.access) != 0) 1001: { 1002: related.host = 0; 1003: related.absolute = 0; 1004: related.relative = 0; 1005: related.fragment = 0; 1006: } 1007: 1008: if (wanted & AMAYA_PARSE_HOST) 1009: if(given.host || related.host) 1010: { 1011: if(wanted & AMAYA_PARSE_PUNCTUATION) 1012: strcat (result, "//"); 1013: strcat (result, given.host ? given.host : related.host); 1014: } 1015: 1016: if (given.host && related.host) 1017: /* If different hosts, inherit no path. */ 1018: if (strcmp(given.host, related.host) != 0) 1019: { 1020: related.absolute = 0; 1021: related.relative = 0; 1022: related.fragment = 0; 1023: } 1024: 1025: if (wanted & AMAYA_PARSE_PATH) 1026: { 1027: if (given.absolute) 1028: { 1029: /* All is given */ 1030: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.33 cvs 1031: strcat (result, used_str); 1.29 cvs 1032: strcat (result, given.absolute); 1.25 cvs 1033: } 1.29 cvs 1034: else if (related.absolute) 1035: { 1036: /* Adopt path not name */ 1.33 cvs 1037: strcat (result, used_str); 1.29 cvs 1038: strcat (result, related.absolute); 1039: if (given.relative) 1040: { 1041: /* Search part? */ 1042: p = strchr (result, '?'); 1043: if (!p) 1044: p=result+strlen(result)-1; 1.33 cvs 1045: for (; *p!=used_sep; p--); /* last / */ 1.29 cvs 1046: /* Remove filename */ 1047: p[1]=0; 1048: /* Add given one */ 1049: strcat (result, given.relative); 1050: /*SimplifyUrl (&result);*/ 1.25 cvs 1051: } 1052: } 1.29 cvs 1053: else if (given.relative) 1054: /* what we've got */ 1055: strcat (result, given.relative); 1056: else if (related.relative) 1057: strcat (result, related.relative); 1058: else 1059: /* No inheritance */ 1.33 cvs 1060: strcat (result, used_str); 1.25 cvs 1061: } 1.29 cvs 1062: 1063: if (wanted & AMAYA_PARSE_ANCHOR) 1064: if (given.fragment || related.fragment) 1065: { 1066: if (given.absolute && given.fragment) 1067: { 1068: /*Fixes for relURLs...*/ 1069: if (wanted & AMAYA_PARSE_PUNCTUATION) 1070: strcat (result, "#"); 1071: strcat (result, given.fragment); 1072: } 1073: else if (!(given.absolute) && !(given.fragment)) 1074: strcat (result, ""); 1075: else 1076: { 1077: if (wanted & AMAYA_PARSE_PUNCTUATION) 1078: strcat (result, "#"); 1079: strcat (result, given.fragment ? given.fragment : related.fragment); 1080: } 1081: } 1082: len = strlen (result); 1083: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1084: strcpy (return_value, result); 1085: return (return_value); /* exactly the right length */ 1.25 cvs 1086: } 1087: 1088: /*---------------------------------------------------------------------- 1089: HTCanon 1090: Canonicalizes the URL in the following manner starting from the host 1091: pointer: 1092: 1093: 1) The host name is converted to lowercase 1094: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp) 1095: 1096: Return: OK The position of the current path part of the URL 1097: which might be the old one or a new one. 1098: 1099: ----------------------------------------------------------------------*/ 1100: #ifdef __STDC__ 1.28 cvs 1101: static char *HTCanon (char ** filename, char * host) 1.25 cvs 1102: #else /* __STDC__ */ 1.28 cvs 1103: static char *HTCanon (filename, host) 1104: char **filename; 1105: char *host; 1.25 cvs 1106: #endif /* __STDC__ */ 1107: { 1108: char *newname = NULL; 1109: char *port; 1110: char *strptr; 1111: char *path; 1112: char *access = host-3; 1.33 cvs 1113: char used_sep; 1.32 cvs 1114: 1115: 1.33 cvs 1116: if (*filename && strchr (*filename, URL_SEP)) 1117: { 1118: used_sep = URL_SEP; 1119: } 1120: else 1121: { 1122: used_sep = DIR_SEP; 1123: } 1.32 cvs 1124: 1.33 cvs 1125: while (access>*filename && *(access-1)!= used_sep) /* Find access method */ 1.25 cvs 1126: access--; 1.33 cvs 1127: if ((path = strchr(host, used_sep)) == NULL) /* Find path */ 1.25 cvs 1128: path = host + strlen(host); 1129: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */ 1130: host = strptr; 1.32 cvs 1131: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */ 1.25 cvs 1132: port = NULL; 1133: 1134: strptr = host; /* Convert to lower-case */ 1.33 cvs 1135: while (strptr<path) 1136: { 1.25 cvs 1137: *strptr = tolower(*strptr); 1138: strptr++; 1.33 cvs 1139: } 1.25 cvs 1140: 1141: /* Does the URL contain a full domain name? This also works for a 1142: numerical host name. The domain name is already made lower-case 1143: and without a trailing dot. */ 1144: { 1.33 cvs 1145: char *dot = port ? port : path; 1146: if (dot > *filename && *--dot=='.') 1147: { 1148: char *orig=dot, *dest=dot+1; 1149: while((*orig++ = *dest++)); 1150: if (port) port--; 1151: path--; 1.25 cvs 1152: } 1153: } 1154: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */ 1.33 cvs 1155: if (port) 1156: { 1157: if (!*(port+1) || *(port+1)==used_sep) 1158: { 1159: if (!newname) 1160: { 1.25 cvs 1161: char *orig=port, *dest=port+1; 1162: while((*orig++ = *dest++)); 1.33 cvs 1163: } 1164: } 1165: else if ((!strncmp(access, "http", 4) && 1166: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) || 1167: (!strncmp(access, "gopher", 6) && 1168: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) || 1169: (!strncmp(access, "ftp", 3) && 1170: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) { 1171: if (!newname) 1172: { 1173: char *orig=port, *dest=port+3; 1174: while((*orig++ = *dest++)); 1175: /* Update path position, Henry Minsky */ 1176: path -= 3; 1.25 cvs 1177: } 1.33 cvs 1178: } 1179: else if (newname) 1180: strncat(newname, port, (int) (path-port)); 1181: } 1.25 cvs 1182: 1.33 cvs 1183: if (newname) 1184: { 1.25 cvs 1185: char *newpath = newname+strlen(newname); 1186: strcat(newname, path); 1187: path = newpath; 1.28 cvs 1188: /* Free old copy */ 1189: TtaFreeMemory(*filename); 1.25 cvs 1190: *filename = newname; 1.33 cvs 1191: } 1.25 cvs 1192: return path; 1193: } 1194: 1195: 1196: /*---------------------------------------------------------------------- 1.29 cvs 1197: SimplifyUrl: simplify a URI 1.32 cvs 1198: A URI is allowed to contain the sequence xxx/../ which may be 1199: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR. 1.28 cvs 1200: Simplification helps us recognize duplicate URIs. 1.25 cvs 1201: 1.28 cvs 1202: Thus, /etc/junk/../fred becomes /etc/fred 1203: /etc/junk/./fred becomes /etc/junk/fred 1.25 cvs 1204: 1.28 cvs 1205: but we should NOT change 1206: http://fred.xxx.edu/../.. 1.25 cvs 1207: 1208: or ../../albert.html 1209: 1.28 cvs 1210: In order to avoid empty URLs the following URLs become: 1.25 cvs 1211: 1212: /fred/.. becomes /fred/.. 1213: /fred/././.. becomes /fred/.. 1214: /fred/.././junk/.././ becomes /fred/.. 1215: 1.28 cvs 1216: If more than one set of `://' is found (several proxies in cascade) then 1217: only the part after the last `://' is simplified. 1.25 cvs 1218: 1.28 cvs 1219: Returns: A string which might be the old one or a new one. 1.25 cvs 1220: ----------------------------------------------------------------------*/ 1221: #ifdef __STDC__ 1.29 cvs 1222: void SimplifyUrl (char ** url) 1.25 cvs 1223: #else /* __STDC__ */ 1.29 cvs 1224: void SimplifyUrl (url) 1.28 cvs 1225: char **url; 1.25 cvs 1226: #endif /* __STDC__ */ 1227: { 1.28 cvs 1228: char *path, *p; 1229: char *newptr, *access; 1230: char *orig, *dest, *end; 1231: 1.33 cvs 1232: char used_sep; 1.32 cvs 1233: 1234: 1.28 cvs 1235: if (!url || !*url) 1236: return; 1237: 1.33 cvs 1238: if (strchr (*url, URL_SEP)) 1239: { 1240: used_sep = URL_SEP; 1241: } 1.32 cvs 1242: else 1.33 cvs 1243: { 1244: used_sep = DIR_SEP; 1245: } 1.32 cvs 1246: 1.28 cvs 1247: /* Find any scheme name */ 1248: if ((path = strstr(*url, "://")) != NULL) 1.33 cvs 1249: { 1250: /* Find host name */ 1.28 cvs 1251: access = *url; 1252: while (access<path && (*access=tolower(*access))) 1253: access++; 1254: path += 3; 1255: while ((newptr = strstr(path, "://")) != NULL) 1256: /* For proxies */ 1257: path = newptr+3; 1258: /* We have a host name */ 1259: path = HTCanon(url, path); 1.25 cvs 1260: } 1.28 cvs 1261: else if ((path = strstr(*url, ":/")) != NULL) 1262: path += 2; 1263: else 1264: path = *url; 1.25 cvs 1265: 1.33 cvs 1266: if (*path == used_sep && *(path+1)==used_sep) 1.28 cvs 1267: /* Some URLs start //<foo> */ 1268: path += 1; 1269: else if (!strncmp(path, "news:", 5)) 1270: { 1271: newptr = strchr(path+5, '@'); 1272: if (!newptr) 1273: newptr = path + 5; 1274: while (*newptr) 1275: { 1276: /* Make group or host lower case */ 1277: *newptr = tolower (*newptr); 1278: newptr++; 1.25 cvs 1279: } 1.28 cvs 1280: /* Doesn't need to do any more */ 1281: return; 1.25 cvs 1282: } 1.28 cvs 1283: 1284: if ((p = path)) 1285: { 1286: if (!((end = strchr (path, ';')) || (end = strchr (path, '?')) || 1287: (end = strchr (path, '#')))) 1288: end = path + strlen (path); 1289: 1290: /* Parse string second time to simplify */ 1291: p = path; 1292: while (p < end) 1293: { 1.33 cvs 1294: if (*p==used_sep) 1.28 cvs 1295: { 1.33 cvs 1296: if (p > *url && *(p+1) == '.' && (*(p+2) == used_sep || !*(p+2))) 1.28 cvs 1297: { 1298: orig = p + 1; 1.33 cvs 1299: dest = (*(p+2)!=used_sep) ? p+2 : p+3; 1.28 cvs 1300: while ((*orig++ = *dest++)); /* Remove a slash and a dot */ 1301: end = orig - 1; 1302: } 1.33 cvs 1303: else if (*(p+1)=='.' && *(p+2)=='.' && (*(p+3)==used_sep || !*(p+3))) 1.28 cvs 1304: { 1305: newptr = p; 1.33 cvs 1306: while (newptr>path && *--newptr!=used_sep); /* prev slash */ 1.28 cvs 1307: if (strncmp(newptr, "/../", 4)) 1308: { 1309: orig = newptr + 1; 1.33 cvs 1310: dest = (*(p+3)!=used_sep) ? p+3 : p+4; 1.28 cvs 1311: while ((*orig++ = *dest++)); /* Remove /xxx/.. */ 1312: end = orig-1; 1313: /* Start again with prev slash */ 1314: p = newptr; 1.25 cvs 1315: } 1.28 cvs 1316: else 1.25 cvs 1317: p++; 1.28 cvs 1318: } 1.33 cvs 1319: else if (*(p+1) == used_sep) 1.28 cvs 1320: { 1.33 cvs 1321: while (*(p+1) == used_sep) 1.28 cvs 1322: { 1323: orig = p; 1324: dest = p + 1; 1325: while ((*orig++ = *dest++)); /* Remove multiple /'s */ 1326: end = orig-1; 1327: } 1328: } 1329: else 1.25 cvs 1330: p++; 1.28 cvs 1331: } 1332: else 1333: p++; 1.25 cvs 1334: } 1335: } 1.28 cvs 1336: return; 1337: } 1338: 1339: 1340: /*---------------------------------------------------------------------- 1341: NormalizeFile normalizes local names. 1342: Return TRUE if target and src differ. 1343: ----------------------------------------------------------------------*/ 1344: #ifdef __STDC__ 1345: boolean NormalizeFile (char *src, char *target) 1346: #else 1347: boolean NormalizeFile (src, target) 1348: char *src; 1349: char *target; 1350: 1351: #endif 1352: { 1353: char *s; 1354: boolean change; 1355: 1356: change = FALSE; 1357: if (src[0] == '~') 1358: { 1359: /* replace ~ */ 1360: s = (char *) TtaGetEnvString ("HOME"); 1361: strcpy (target, s); 1362: strcat (target, &src[1]); 1363: change = TRUE; 1364: } 1365: else if (strncmp (src, "file:", 5) == 0) 1366: { 1367: /* remove the prefix file: */ 1368: if (src[5] == EOS) 1369: strcpy (target, DIR_STR); 1370: else if (src[0] == '~') 1371: { 1372: /* replace ~ */ 1373: s = (char *) TtaGetEnvString ("HOME"); 1374: strcpy (target, s); 1375: strcat (target, &src[5]); 1376: } 1377: else 1378: strcpy (target, &src[5]); 1379: change = TRUE; 1380: } 1381: else 1382: strcpy (target, src); 1383: 1384: /* remove /../ and /./ */ 1.29 cvs 1385: SimplifyUrl (&target); 1.28 cvs 1386: return (change); 1.25 cvs 1387: } 1388: 1.28 cvs 1389: 1.25 cvs 1390: /*---------------------------------------------------------------------- 1.31 cvs 1391: MakeRelativeURL: make relative name 1.25 cvs 1392: 1.28 cvs 1393: This function creates and returns a string which gives an expression of 1394: one address as related to another. Where there is no relation, an absolute 1395: address is retured. 1.25 cvs 1396: 1.28 cvs 1397: On entry, 1.25 cvs 1398: Both names must be absolute, fully qualified names of nodes 1399: (no fragment bits) 1400: 1.28 cvs 1401: On exit, 1.25 cvs 1402: The return result points to a newly allocated name which, if 1403: parsed by AmayaParseUrl relative to relatedName, will yield aName. 1404: The caller is responsible for freeing the resulting name later. 1405: ----------------------------------------------------------------------*/ 1406: #ifdef __STDC__ 1.31 cvs 1407: char *MakeRelativeURL (char *aName, char *relatedName) 1.25 cvs 1408: #else /* __STDC__ */ 1.31 cvs 1409: char *MakeRelativeURL (aName, relatedName) 1.28 cvs 1410: char *aName; 1411: char *relatedName; 1.25 cvs 1412: #endif /* __STDC__ */ 1413: { 1.39 cvs 1414: char *return_value; 1415: char result[MAX_LENGTH]; 1.29 cvs 1416: char *p; 1.34 cvs 1417: char *q; 1.29 cvs 1418: char *after_access; 1419: char *last_slash = NULL; 1420: int slashes, levels, len; 1421: 1422: if (aName == NULL || relatedName == NULL) 1423: return (NULL); 1424: 1425: slashes = 0; 1426: after_access = NULL; 1427: p = aName; 1428: q = relatedName; 1429: for (; *p && (*p == *q); p++, q++) 1.27 cvs 1430: { 1431: /* Find extent of match */ 1.32 cvs 1432: if (*p == ':') 1.29 cvs 1433: after_access = p + 1; 1.28 cvs 1434: if (*p == DIR_SEP) 1.27 cvs 1435: { 1.29 cvs 1436: /* memorize the last slash position and count them */ 1.27 cvs 1437: last_slash = p; 1438: slashes++; 1.25 cvs 1439: } 1440: } 1441: 1.31 cvs 1442: /* q, p point to the first non-matching character or zero */ 1443: if (*q == EOS) 1444: { 1445: /* New name is a subset of the related name */ 1446: /* exactly the right length */ 1447: len = strlen (p); 1448: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1449: strcpy (return_value, p); 1450: } 1451: else if ((slashes < 2 && after_access == NULL) 1452: || (slashes < 3 && after_access != NULL)) 1453: { 1454: /* Two names whitout common path */ 1455: /* exactly the right length */ 1456: len = strlen (aName); 1457: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1458: strcpy (return_value, aName); 1459: } 1460: else 1461: { 1462: /* Some path in common */ 1463: if (slashes == 3 && strncmp (aName, "http:", 5) == 0) 1464: /* just the same server */ 1465: strcpy (result, last_slash); 1466: else 1467: { 1468: levels= 0; 1469: for (; *q && (*q != '#'); q++) 1470: if (*q == DIR_SEP) 1471: levels++; 1472: 1473: result[0] = 0; 1474: for (;levels; levels--) 1475: strcat (result, "../"); 1476: strcat (result, last_slash+1); 1477: } 1478: 1479: /* exactly the right length */ 1480: len = strlen (result); 1481: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1482: strcpy (return_value, result); 1.25 cvs 1483: } 1.29 cvs 1484: return (return_value); 1.24 cvs 1485: } 1.35 cvs 1486: 1487: