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__ 1.28 cvs 38: static void ConvertToLowerCase (char *string) 39: #else /* __STDC__ */ 40: static void ConvertToLowerCase (string) 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) 1.13 cvs 193: return (FALSE); 1.5 cvs 194: 195: strcpy (temppath, path); 196: ExtractSuffix (temppath, suffix); 197: 198: /* Normalize the suffix */ 199: i = 0; 200: while (suffix[i] != EOS) 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; 206: if ((strcmp (nsuffix, "html")) && 207: (strcmp (nsuffix, "htm")) && 208: (strcmp (nsuffix, "shtml"))) 1.13 cvs 209: return (FALSE); 1.22 cvs 210: else if (!strcmp (nsuffix, "gz")) 1.13 cvs 211: { 212: /* take in account compressed files */ 213: ExtractSuffix (temppath, suffix); 214: /* Normalize the suffix */ 215: i = 0; 216: while (suffix[i] != EOS) 217: { 1.25 cvs 218: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 219: i++; 220: } 221: nsuffix[i] = EOS; 222: if ((strcmp (nsuffix, "html")) && 223: (strcmp (nsuffix, "htm")) && 224: (strcmp (nsuffix, "shtml"))) 225: return (FALSE); 226: else 227: return (TRUE); 228: } 229: else 230: return (TRUE); 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; 257: while (suffix[i] != EOS) 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; 263: if ((strcmp (nsuffix, "gif")) && (strcmp (nsuffix, "xbm")) && 264: (strcmp (nsuffix, "xpm")) && (strcmp (nsuffix, "jpg")) && 265: (strcmp (nsuffix, "png")) && (strcmp (nsuffix, "au"))) 1.13 cvs 266: return (FALSE); 267: return (TRUE); 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; 294: while (suffix[i] != EOS) 295: { 1.25 cvs 296: nsuffix[i] = tolower (suffix[i]); 1.5 cvs 297: i++; 298: } 299: nsuffix[i] = EOS; 300: 301: if ((strcmp (nsuffix, "gif")) && (strcmp (nsuffix, "xbm")) && 302: (strcmp (nsuffix, "xpm")) && (strcmp (nsuffix, "jpg")) && 303: (strcmp (nsuffix, "pdf")) && (strcmp (nsuffix, "png")) && 1.22 cvs 304: (strcmp (nsuffix, "tgz")) && (strcmp (nsuffix, "tar")) && 305: (strcmp (nsuffix, "xpg")) && (strcmp (nsuffix, "xpd")) && 306: (strcmp (nsuffix, "ps")) && (strcmp (nsuffix, "au"))) 1.13 cvs 307: return (TRUE); 1.22 cvs 308: else if (!strcmp (nsuffix, "gz")) 1.13 cvs 309: { 310: /* take in account compressed files */ 311: ExtractSuffix (temppath, suffix); 312: /* Normalize the suffix */ 313: i = 0; 314: while (suffix[i] != EOS) 315: { 1.25 cvs 316: nsuffix[i] = tolower (suffix[i]); 1.13 cvs 317: i++; 318: } 319: nsuffix[i] = EOS; 320: if ((!strcmp (nsuffix, "html")) || 321: (!strcmp (nsuffix, "htm")) || 322: (!strcmp (nsuffix, "shtml"))) 323: return (TRUE); 324: else 325: return (FALSE); 326: } 327: else 328: return (FALSE); 1.3 cvs 329: } 330: 1.4 cvs 331: /*---------------------------------------------------------------------- 1.9 cvs 332: IsHTTPPath 333: returns TRUE if path is in fact an http URL. 1.4 cvs 334: ----------------------------------------------------------------------*/ 1.3 cvs 335: #ifdef __STDC__ 1.34 ! cvs 336: boolean IsHTTPPath (const char *path) 1.3 cvs 337: #else /* __STDC__ */ 338: boolean IsHTTPPath (path) 1.34 ! cvs 339: const char *path; 1.3 cvs 340: #endif /* __STDC__ */ 341: { 1.5 cvs 342: if (!path) 343: return FALSE; 1.3 cvs 344: 1.5 cvs 345: if (strncmp (path, "http:", 5) != 0) 346: return FALSE; 347: return TRUE; 1.3 cvs 348: } 349: 1.4 cvs 350: /*---------------------------------------------------------------------- 1.9 cvs 351: IsWithParameters 352: returns TRUE if url has a concatenated query string. 1.4 cvs 353: ----------------------------------------------------------------------*/ 1.3 cvs 354: #ifdef __STDC__ 1.34 ! cvs 355: boolean IsWithParameters (const char *url) 1.3 cvs 356: #else /* __STDC__ */ 1.9 cvs 357: boolean IsWithParameters (url) 1.34 ! cvs 358: const char *url; 1.3 cvs 359: #endif /* __STDC__ */ 360: { 1.5 cvs 361: int i; 1.3 cvs 362: 1.9 cvs 363: if ((!url) || (url[0] == EOS)) 1.5 cvs 364: return FALSE; 1.3 cvs 365: 1.9 cvs 366: i = strlen (url) - 1; 367: while (i > 0 && url[i--] != '?') 1.5 cvs 368: if (i < 0) 369: return FALSE; 1.3 cvs 370: 1.5 cvs 371: /* There is a parameter */ 372: return TRUE; 1.3 cvs 373: } 374: 1.4 cvs 375: /*---------------------------------------------------------------------- 1.9 cvs 376: IsW3Path 377: returns TRUE if path is in fact a URL. 1.4 cvs 378: ----------------------------------------------------------------------*/ 1.3 cvs 379: #ifdef __STDC__ 1.34 ! cvs 380: boolean IsW3Path (const char *path) 1.3 cvs 381: #else /* __STDC__ */ 382: boolean IsW3Path (path) 1.34 ! cvs 383: const char *path; 1.3 cvs 384: #endif /* __STDC__ */ 385: { 1.5 cvs 386: if ((strncmp (path, "http:", 5)) && (strncmp (path, "ftp:", 4)) && 387: (strncmp (path, "telnet:", 7)) && (strncmp (path, "wais:", 5)) && 388: (strncmp (path, "news:", 5)) && (strncmp (path, "gopher:", 7)) && 389: (strncmp (path, "mailto:", 7)) && (strncmp (path, "archie:", 7))) 390: return FALSE; 391: return TRUE; 1.3 cvs 392: } 393: 1.4 cvs 394: /*---------------------------------------------------------------------- 1.9 cvs 395: IsValidProtocol 396: returns true if the url protocol is supported by Amaya. 1.4 cvs 397: ----------------------------------------------------------------------*/ 1.3 cvs 398: #ifdef __STDC__ 1.34 ! cvs 399: boolean IsValidProtocol (const char *url) 1.3 cvs 400: #else /* __STDC__ */ 1.9 cvs 401: boolean IsValidProtocol (url) 1.34 ! cvs 402: const char *url; 1.3 cvs 403: #endif /* __STDC__ */ 404: { 1.26 cvs 405: if (!strncmp (url, "http:", 5)) 1.22 cvs 406: /* experimental */ 1.26 cvs 407: /*** || !strncmp (url, "ftp:", 4)) ***/ 1.24 cvs 408: /*** || !strncmp (path, "news:", 5)***/ 1.8 cvs 409: return (TRUE); 1.5 cvs 410: else 1.8 cvs 411: return (FALSE); 1.3 cvs 412: } 413: 1.31 cvs 414: 415: /*---------------------------------------------------------------------- 416: GetBaseURL 417: normalizes orgName according to a base associated with doc, and 418: following the standard URL format rules. 419: The function returns the base used to solve relative URL and SRC: 420: - the base of the document, 421: - or the document path (without document name). 422: ----------------------------------------------------------------------*/ 423: #ifdef __STDC__ 424: char *GetBaseURL (Document doc) 425: #else /* __STDC__ */ 426: char *GetBaseURL (doc) 427: Document doc; 428: #endif /* __STDC__ */ 429: { 430: Element el; 431: ElementType elType; 432: AttributeType attrType; 433: Attribute attr; 1.34 ! cvs 434: char used_sep = URL_SEP; 1.33 cvs 435: char *used_str; 1.31 cvs 436: char *ptr, *basename; 437: int length; 438: 439: basename = TtaGetMemory (MAX_LENGTH); 440: strcpy (basename, DocumentURLs[doc]); 441: length = MAX_LENGTH -1; 442: /* get the root element */ 443: el = TtaGetMainRoot (doc); 444: /* search the BASE element */ 445: elType.ElSSchema = TtaGetDocumentSSchema (doc); 446: elType.ElTypeNum = HTML_EL_BASE; 447: el = TtaSearchTypedElement (elType, SearchInTree, el); 448: if (el) 449: { 450: /* The document has a BASE element -> Get the HREF attribute */ 451: attrType.AttrSSchema = elType.ElSSchema; 452: attrType.AttrTypeNum = HTML_ATTR_HREF_; 453: attr = TtaGetAttribute (el, attrType); 454: if (attr) 455: { 456: /* Use the base path of the document */ 457: TtaGiveTextAttributeValue (attr, basename, &length); 458: /* base and orgName have to be separated by a DIR_SEP */ 459: length--; 1.32 cvs 460: 1.33 cvs 461: if (basename [0] != EOS && strchr (basename, URL_SEP)) 462: { 463: used_str = URL_STR; 464: used_sep = URL_SEP; 465: } 466: else 467: { 468: used_str = DIR_STR; 469: used_sep = DIR_SEP; 470: } 1.32 cvs 471: 1.33 cvs 472: if (basename[0] != EOS && basename[length] != used_sep) 1.31 cvs 473: /* verify if the base has the form "protocol://server:port" */ 474: { 1.33 cvs 475: ptr = AmayaParseUrl (basename, "", AMAYA_PARSE_ACCESS | 476: AMAYA_PARSE_HOST | 477: AMAYA_PARSE_PUNCTUATION); 1.31 cvs 478: if (ptr && !strcmp (ptr, basename)) 479: { 480: /* it has this form, we complete it by adding a DIR_STR */ 1.33 cvs 481: strcat (basename, used_str); 1.31 cvs 482: length++; 483: } 484: if (ptr) 485: TtaFreeMemory (ptr); 486: } 487: } 1.33 cvs 488: } 1.32 cvs 489: else 1.33 cvs 490: { 491: if (basename [0] != EOS && strchr (basename, URL_SEP)) 492: { 493: used_str = URL_STR; 494: used_sep = URL_SEP; 495: } 496: else 497: { 498: used_str = DIR_STR; 499: used_sep = DIR_SEP; 500: } 501: } 502: 1.31 cvs 503: /* Remove anything after the last DIR_SEP char. If no such char is found, 504: * then search for the first ":" char, hoping that what's before that is a 505: * protocol. If found, end the string there. If neither char is found, 506: * then discard the whole base element. 507: */ 508: length = strlen (basename) - 1; 509: /* search for the last DIR_SEP char */ 1.33 cvs 510: while (length >= 0 && basename[length] != used_sep) 1.31 cvs 511: length--; 512: if (length >= 0) 513: /* found the last DIR_SEP char, end the string there */ 514: basename[length + 1] = EOS; 515: else 516: /* search for the first PATH_STR char */ 517: { 1.32 cvs 518: for (length = 0; basename[length] != ':' && 1.31 cvs 519: basename[length] != EOS; length ++); 1.32 cvs 520: if (basename[length] == ':') 1.31 cvs 521: /* found, so end the string there */ 522: basename[length + 1] = EOS; 523: else 524: /* not found, discard the base */ 525: basename[0] = EOS; 526: } 527: return (basename); 528: } 529: 530: 1.4 cvs 531: /*---------------------------------------------------------------------- 1.9 cvs 532: NormalizeURL 533: normalizes orgName according to a base associated with doc, and 534: following the standard URL format rules. 535: The function returns the new complete and normalized URL 1.12 cvs 536: or file name path (newName) and the name of the document (docName). 1.9 cvs 537: N.B. If the function can't find out what's the docName, it assigns 538: the name "noname.html". 1.4 cvs 539: ----------------------------------------------------------------------*/ 1.3 cvs 540: #ifdef __STDC__ 541: void NormalizeURL (char *orgName, Document doc, char *newName, char *docName) 542: #else /* __STDC__ */ 543: void NormalizeURL (orgName, doc, newName, docName) 544: char *orgName; 545: Document doc; 546: char *newName; 547: char *docName; 548: #endif /* __STDC__ */ 549: { 1.31 cvs 550: char *basename; 1.18 cvs 551: char tempOrgName[MAX_LENGTH]; 1.5 cvs 552: char *ptr; 553: int length; 1.33 cvs 554: char used_sep; 555: char *used_str; 1.5 cvs 556: 557: if (!newName || !docName) 558: return; 1.18 cvs 559: 1.32 cvs 560: if (doc != 0) 1.33 cvs 561: basename = GetBaseURL (doc); 1.32 cvs 562: else 1.33 cvs 563: basename = (char *) NULL; 1.32 cvs 564: 1.33 cvs 565: if (strchr (orgName, URL_SEP) || (basename && strchr (basename, URL_SEP))) 566: { 567: used_str = URL_STR; 568: used_sep = URL_SEP; 569: } 570: else 571: { 572: used_str = DIR_STR; 573: used_sep = DIR_SEP; 574: } 1.32 cvs 575: 1.18 cvs 576: /* 1.31 cvs 577: * Clean orgName 578: * Make sure we have a complete orgName, without any leading or trailing 579: * white spaces, or trailinbg new lines 580: */ 1.18 cvs 581: 1.5 cvs 582: ptr = orgName; 1.18 cvs 583: /* skip leading white space and new line characters */ 1.19 cvs 584: while ((*ptr == ' ' || *ptr == EOL) && *ptr++ != EOS); 1.18 cvs 585: strcpy (tempOrgName, ptr); 586: /* clean trailing white space */ 587: ptr = strchr (tempOrgName, ' '); 588: if (ptr) 589: *ptr = EOS; 590: /* clean trailing new lines */ 1.19 cvs 591: ptr = strchr (tempOrgName, EOL); 1.5 cvs 592: if (ptr) 593: *ptr = EOS; 594: 1.18 cvs 595: /* 1.31 cvs 596: * Make orgName a complete URL 597: * If the URL does not include a protocol, then try to calculate 598: * one using the doc's base element (if it exists), 599: */ 1.21 cvs 600: if (tempOrgName[0] == EOS) 601: { 602: newName[0] = EOS; 603: return; 604: } 605: else if (IsW3Path (tempOrgName)) 606: { 607: /* the name is complete, go to the Sixth Step */ 608: strcpy (newName, tempOrgName); 609: /* verify if the URL has the form "protocol://server:port" */ 1.25 cvs 610: ptr = AmayaParseUrl (newName, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | 611: AMAYA_PARSE_PUNCTUATION); 1.21 cvs 612: if (ptr && !strcmp (ptr, newName)) 613: { 1.29 cvs 614: /* it has this form, we complete it by adding a DIR_STR */ 1.33 cvs 615: strcat (newName, used_str); 1.21 cvs 616: } 617: if (ptr) 1.25 cvs 618: TtaFreeMemory (ptr); 1.21 cvs 619: } 620: else if ( doc == 0) 1.19 cvs 621: /* the name is complete, go to the Sixth Step */ 1.18 cvs 622: strcpy (newName, tempOrgName); 1.5 cvs 623: else 624: { 1.22 cvs 625: 1.31 cvs 626: /* Calculate the absolute URL, using the base or document URL */ 1.32 cvs 627: 1.25 cvs 628: ptr = AmayaParseUrl (tempOrgName, basename, AMAYA_PARSE_ALL); 1.31 cvs 629: TtaFreeMemory (basename); 1.14 cvs 630: if (ptr) 631: { 1.29 cvs 632: SimplifyUrl (&ptr); 1.14 cvs 633: strcpy (newName, ptr); 1.25 cvs 634: TtaFreeMemory (ptr); 1.14 cvs 635: } 636: else 1.18 cvs 637: newName[0] = EOS; 1.5 cvs 638: } 639: 1.18 cvs 640: /* 1.31 cvs 641: * Prepare the docname that will refer to this ressource in the 642: * .amaya directory. If the new URL finishes on DIR_SEP, then use 643: * noname.html as a default ressource name 1.18 cvs 644: */ 1.19 cvs 645: if (newName[0] != EOS) 1.5 cvs 646: { 1.19 cvs 647: length = strlen (newName) - 1; 1.33 cvs 648: if (newName[length] == used_sep) 1.18 cvs 649: { 650: /* docname was not comprised inside the URL, so let's */ 651: /* assign the default ressource name */ 652: strcpy (docName, "noname.html"); 653: /* remove DIR_SEP at the end of complete path */ 1.23 cvs 654: /* newName[length] = EOS; */ 1.18 cvs 655: } 1.14 cvs 656: else 1.18 cvs 657: { 658: /* docname is comprised inside the URL */ 1.33 cvs 659: while (length >= 0 && newName[length] != used_sep) 1.18 cvs 660: length--; 661: if (length < 0) 662: strcpy (docName, newName); 663: else 664: strcpy (docName, &newName[length+1]); 665: } 1.19 cvs 666: 1.5 cvs 667: } 1.18 cvs 668: else 669: docName[0] = EOS; 670: } 1.3 cvs 671: 1.4 cvs 672: /*---------------------------------------------------------------------- 1.9 cvs 673: IsSameHost 1.4 cvs 674: ----------------------------------------------------------------------*/ 1.3 cvs 675: #ifdef __STDC__ 1.34 ! cvs 676: boolean IsSameHost (const char *url1, const char *url2) 1.3 cvs 677: #else /* __STDC__ */ 678: boolean IsSameHost (url1, url2) 1.34 ! cvs 679: const char *url1; ! 680: const char *url2; 1.3 cvs 681: #endif /* __STDC__ */ 682: { 1.34 ! cvs 683: char *basename_ptr1, *basename_ptr2; ! 684: boolean result; 1.3 cvs 685: 1.25 cvs 686: basename_ptr1 = AmayaParseUrl (url1, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 687: basename_ptr2 = AmayaParseUrl (url2, "", AMAYA_PARSE_ACCESS | AMAYA_PARSE_HOST | AMAYA_PARSE_PUNCTUATION); 1.3 cvs 688: 1.5 cvs 689: if (strcmp (basename_ptr1, basename_ptr2)) 1.8 cvs 690: result = FALSE; 1.5 cvs 691: else 1.8 cvs 692: result = TRUE; 1.3 cvs 693: 1.25 cvs 694: TtaFreeMemory (basename_ptr1); 695: TtaFreeMemory (basename_ptr2); 1.5 cvs 696: return (result); 1.3 cvs 697: } 698: 699: 1.4 cvs 700: /*---------------------------------------------------------------------- 1.22 cvs 701: HasKnownFileSuffix 702: returns TRUE if path points to a file ending with a suffix. 703: ----------------------------------------------------------------------*/ 704: #ifdef __STDC__ 1.34 ! cvs 705: boolean HasKnownFileSuffix (const char *path) 1.22 cvs 706: #else /* __STDC__ */ 707: boolean HasKnownFileSuffix (path) 1.34 ! cvs 708: const char *path; 1.22 cvs 709: #endif /* __STDC__ */ 710: { 1.29 cvs 711: char *root; 712: char temppath[MAX_LENGTH]; 713: char suffix[MAX_LENGTH]; 1.22 cvs 714: 1.24 cvs 715: if (!path || path[0] == EOS || path[strlen(path)] == DIR_SEP) 1.22 cvs 716: return (FALSE); 717: 1.29 cvs 718: root = AmayaParseUrl(path, "", AMAYA_PARSE_PATH | AMAYA_PARSE_PUNCTUATION); 1.22 cvs 719: 720: if (root) 721: { 722: strcpy (temppath, root); 1.25 cvs 723: TtaFreeMemory (root); 1.22 cvs 724: /* Get the suffix */ 725: ExtractSuffix (temppath, suffix); 726: 727: if( suffix[0] == EOS) 728: /* no suffix */ 729: return (FALSE); 730: 731: /* Normalize the suffix */ 732: ConvertToLowerCase (suffix); 733: 1.23 cvs 734: if (!strcmp (suffix, "gz")) 1.22 cvs 735: /* skip the compressed suffix */ 736: { 737: ExtractSuffix (temppath, suffix); 738: if(suffix[0] == EOS) 739: /* no suffix */ 740: return (FALSE); 741: /* Normalize the suffix */ 742: ConvertToLowerCase (suffix); 743: } 744: 745: if ((strcmp (suffix, "gif")) && (strcmp (suffix, "xbm")) && 746: (strcmp (suffix, "xpm")) && (strcmp (suffix, "jpg")) && 747: (strcmp (suffix, "pdf")) && (strcmp (suffix, "png")) && 748: (strcmp (suffix, "tgz")) && (strcmp (suffix, "xpg")) && 749: (strcmp (suffix, "xpd")) && (strcmp (suffix, "ps")) && 750: (strcmp (suffix, "au")) && (strcmp (suffix, "html")) && 751: (strcmp (suffix, "htm")) && (strcmp (suffix, "shtml")) && 752: (strcmp (suffix, "txt")) && (strcmp (suffix, "css")) && 753: (strcmp (suffix, "eps"))) 754: return (FALSE); 755: else 756: return (TRUE); 757: } 758: else 759: return (FALSE); 760: } 761: 762: 763: /*---------------------------------------------------------------------- 1.24 cvs 764: ChopURL 765: Gives back a URL no longer than MAX_PRINT_URL_LENGTH chars (outputURL). 766: If inputURL is bigger than that size, outputURL receives 767: MAX_PRINT_URL_LENGTH / 2 chars from the beginning of inputURL, "...", 768: and MAX_PRINT_URL_LENGTH / 2 chars from the end of inputURL. 769: If inputURL is not longer than MAX_PRINT_URL_LENGTH chars, it gets 770: copied into outputURL. 771: N.B.: outputURL must point to a memory block of MAX_PRINT_URL_LENGTH 772: chars. 773: ----------------------------------------------------------------------*/ 774: #ifdef __STDC__ 1.34 ! cvs 775: void ChopURL (char *outputURL, const char *inputURL) 1.24 cvs 776: #else 777: void ChopURL (outputURL, inputURL) 1.34 ! cvs 778: char *outputURL; ! 779: const char *inputURL; 1.24 cvs 780: #endif 1.22 cvs 781: 1.24 cvs 782: { 783: int len; 1.9 cvs 784: 1.24 cvs 785: len = strlen (inputURL); 786: if (len <= MAX_PRINT_URL_LENGTH) 1.29 cvs 787: strcpy (outputURL, inputURL); 1.24 cvs 788: else 789: /* make a truncated urlName on the status window */ 790: { 791: strncpy (outputURL, inputURL, MAX_PRINT_URL_LENGTH / 2); 792: outputURL [MAX_PRINT_URL_LENGTH / 2] = EOS; 793: strcat (outputURL, "..."); 794: strcat (outputURL, &(inputURL[len - MAX_PRINT_URL_LENGTH / 2 ])); 795: } 1.25 cvs 796: } 797: 798: 799: /*---------------------------------------------------------------------- 800: scan 801: Scan a filename for its consituents 802: ----------------------------------- 803: 804: On entry, 805: name points to a document name which may be incomplete. 806: On exit, 807: absolute or relative may be nonzero (but not both). 808: host, fragment and access may be nonzero if they were specified. 809: Any which are nonzero point to zero terminated strings. 810: ----------------------------------------------------------------------*/ 811: #ifdef __STDC__ 812: static void scan (char * name, HTURI * parts) 813: #else /* __STDC__ */ 814: static void scan (name, parts) 815: char *name; 816: HTURI *parts; 817: 818: #endif /* __STDC__ */ 819: { 1.28 cvs 820: char * p; 821: char * after_access = name; 1.33 cvs 822: char used_sep; 1.32 cvs 823: char my_path_sep; 824: 1.33 cvs 825: if (name && strchr (name, URL_SEP)) 826: { 827: used_sep = URL_SEP; 828: my_path_sep = ':'; 829: } 830: else 831: { 832: used_sep = DIR_SEP; 833: my_path_sep = ':'; 834: } 1.32 cvs 835: 1.28 cvs 836: memset(parts, '\0', sizeof(HTURI)); 837: /* Look for fragment identifier */ 838: if ((p = strrchr(name, '#')) != NULL) 839: { 840: *p++ = '\0'; 841: parts->fragment = p; 1.25 cvs 842: } 843: 1.28 cvs 844: for (p=name; *p; p++) 845: { 1.33 cvs 846: if (*p==used_sep || *p=='#' || *p=='?') 1.28 cvs 847: break; 1.32 cvs 848: if (*p==':') 1.28 cvs 849: { 850: *p = 0; 851: parts->access = after_access; /* Scheme has been specified */ 852: 853: /* The combination of gcc, the "-O" flag and the HP platform is 854: unhealthy. The following three lines is a quick & dirty fix, but is 855: not recommended. Rather, turn off "-O". */ 856: 857: /* after_access = p;*/ 858: /* while (*after_access == 0)*/ 859: /* after_access++;*/ 860: after_access = p+1; 861: if (0==strcasecmp("URL", parts->access)) 862: /* Ignore IETF's URL: pre-prefix */ 863: parts->access = NULL; 864: else 1.25 cvs 865: break; 866: } 867: } 868: 869: p = after_access; 1.33 cvs 870: if (*p==used_sep) 1.28 cvs 871: { 1.33 cvs 872: if (p[1]==used_sep) 1.28 cvs 873: { 1.25 cvs 874: parts->host = p+2; /* host has been specified */ 1.28 cvs 875: *p = 0; /* Terminate access */ 876: /* look for end of host name if any */ 1.33 cvs 877: p = strchr(parts->host,used_sep); 1.28 cvs 878: if (p) 879: { 1.25 cvs 880: *p=0; /* Terminate host */ 881: parts->absolute = p+1; /* Root has been found */ 1.28 cvs 882: } 883: } 884: else 885: /* Root found but no host */ 886: parts->absolute = p+1; 887: } 888: else 889: { 1.25 cvs 890: parts->relative = (*after_access) ? after_access : 0; /* zero for "" */ 1.28 cvs 891: } 1.25 cvs 892: } 893: 894: 895: /*---------------------------------------------------------------------- 1.28 cvs 896: AmayaParseUrl: parse a Name relative to another name 897: 898: This returns those parts of a name which are given (and requested) 899: substituting bits from the related name where necessary. 1.25 cvs 900: 1.28 cvs 901: On entry, 1.25 cvs 902: aName A filename given 903: relatedName A name relative to which aName is to be parsed. Give 904: it an empty string if aName is absolute. 905: wanted A mask for the bits which are wanted. 906: 1.28 cvs 907: On exit, 1.25 cvs 908: returns A pointer to a malloc'd string which MUST BE FREED 909: ----------------------------------------------------------------------*/ 910: #ifdef __STDC__ 1.34 ! cvs 911: char *AmayaParseUrl (const char *aName, char *relatedName, int wanted) 1.25 cvs 912: #else /* __STDC__ */ 1.28 cvs 913: char *AmayaParseUrl (aName, relatedName, wanted) 1.34 ! cvs 914: const char *aName; 1.28 cvs 915: char *relatedName; 916: int wanted; 1.25 cvs 917: 918: #endif /* __STDC__ */ 919: { 1.29 cvs 920: char *return_value; 921: char result[MAX_LENGTH]; 922: char name[MAX_LENGTH]; 923: char rel[MAX_LENGTH]; 924: char *p, *access; 925: HTURI given, related; 926: int len; 1.33 cvs 927: char used_sep; 928: char *used_str; 1.32 cvs 929: 1.33 cvs 930: if (strchr (aName, URL_SEP) || strchr (relatedName, URL_SEP) ) 931: { 932: used_str = URL_STR; 933: used_sep = URL_SEP; 934: } 1.32 cvs 935: else 1.33 cvs 936: { 937: used_str = DIR_STR; 938: used_sep = DIR_SEP; 939: } 1.32 cvs 940: 1.29 cvs 941: /* Make working copies of input strings to cut up: */ 942: return_value = NULL; 943: result[0] = 0; /* Clear string */ 944: strcpy (name, aName); 945: if (relatedName != NULL) 946: strcpy (rel, relatedName); 947: else 948: relatedName[0] = EOS; 949: 950: scan (name, &given); 951: scan (rel, &related); 952: access = given.access ? given.access : related.access; 953: if (wanted & AMAYA_PARSE_ACCESS) 954: if (access) 955: { 956: strcat (result, access); 957: if(wanted & AMAYA_PARSE_PUNCTUATION) 1.32 cvs 958: strcat (result, ":"); 1.29 cvs 959: } 960: 961: if (given.access && related.access) 962: /* If different, inherit nothing. */ 963: if (strcmp (given.access, related.access) != 0) 964: { 965: related.host = 0; 966: related.absolute = 0; 967: related.relative = 0; 968: related.fragment = 0; 969: } 970: 971: if (wanted & AMAYA_PARSE_HOST) 972: if(given.host || related.host) 973: { 974: if(wanted & AMAYA_PARSE_PUNCTUATION) 975: strcat (result, "//"); 976: strcat (result, given.host ? given.host : related.host); 977: } 978: 979: if (given.host && related.host) 980: /* If different hosts, inherit no path. */ 981: if (strcmp(given.host, related.host) != 0) 982: { 983: related.absolute = 0; 984: related.relative = 0; 985: related.fragment = 0; 986: } 987: 988: if (wanted & AMAYA_PARSE_PATH) 989: { 990: if (given.absolute) 991: { 992: /* All is given */ 993: if (wanted & AMAYA_PARSE_PUNCTUATION) 1.33 cvs 994: strcat (result, used_str); 1.29 cvs 995: strcat (result, given.absolute); 1.25 cvs 996: } 1.29 cvs 997: else if (related.absolute) 998: { 999: /* Adopt path not name */ 1.33 cvs 1000: strcat (result, used_str); 1.29 cvs 1001: strcat (result, related.absolute); 1002: if (given.relative) 1003: { 1004: /* Search part? */ 1005: p = strchr (result, '?'); 1006: if (!p) 1007: p=result+strlen(result)-1; 1.33 cvs 1008: for (; *p!=used_sep; p--); /* last / */ 1.29 cvs 1009: /* Remove filename */ 1010: p[1]=0; 1011: /* Add given one */ 1012: strcat (result, given.relative); 1013: /*SimplifyUrl (&result);*/ 1.25 cvs 1014: } 1015: } 1.29 cvs 1016: else if (given.relative) 1017: /* what we've got */ 1018: strcat (result, given.relative); 1019: else if (related.relative) 1020: strcat (result, related.relative); 1021: else 1022: /* No inheritance */ 1.33 cvs 1023: strcat (result, used_str); 1.25 cvs 1024: } 1.29 cvs 1025: 1026: if (wanted & AMAYA_PARSE_ANCHOR) 1027: if (given.fragment || related.fragment) 1028: { 1029: if (given.absolute && given.fragment) 1030: { 1031: /*Fixes for relURLs...*/ 1032: if (wanted & AMAYA_PARSE_PUNCTUATION) 1033: strcat (result, "#"); 1034: strcat (result, given.fragment); 1035: } 1036: else if (!(given.absolute) && !(given.fragment)) 1037: strcat (result, ""); 1038: else 1039: { 1040: if (wanted & AMAYA_PARSE_PUNCTUATION) 1041: strcat (result, "#"); 1042: strcat (result, given.fragment ? given.fragment : related.fragment); 1043: } 1044: } 1045: len = strlen (result); 1046: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1047: strcpy (return_value, result); 1048: return (return_value); /* exactly the right length */ 1.25 cvs 1049: } 1050: 1051: /*---------------------------------------------------------------------- 1052: HTCanon 1053: Canonicalizes the URL in the following manner starting from the host 1054: pointer: 1055: 1056: 1) The host name is converted to lowercase 1057: 2) Chop off port if `:80' (http), `:70' (gopher), or `:21' (ftp) 1058: 1059: Return: OK The position of the current path part of the URL 1060: which might be the old one or a new one. 1061: 1062: ----------------------------------------------------------------------*/ 1063: #ifdef __STDC__ 1.28 cvs 1064: static char *HTCanon (char ** filename, char * host) 1.25 cvs 1065: #else /* __STDC__ */ 1.28 cvs 1066: static char *HTCanon (filename, host) 1067: char **filename; 1068: char *host; 1.25 cvs 1069: #endif /* __STDC__ */ 1070: { 1071: char *newname = NULL; 1072: char *port; 1073: char *strptr; 1074: char *path; 1075: char *access = host-3; 1.33 cvs 1076: char used_sep; 1077: char *used_str; 1.32 cvs 1078: 1079: 1.33 cvs 1080: if (*filename && strchr (*filename, URL_SEP)) 1081: { 1082: used_str = URL_STR; 1083: used_sep = URL_SEP; 1084: } 1085: else 1086: { 1087: used_str = DIR_STR; 1088: used_sep = DIR_SEP; 1089: } 1.32 cvs 1090: 1.33 cvs 1091: while (access>*filename && *(access-1)!= used_sep) /* Find access method */ 1.25 cvs 1092: access--; 1.33 cvs 1093: if ((path = strchr(host, used_sep)) == NULL) /* Find path */ 1.25 cvs 1094: path = host + strlen(host); 1095: if ((strptr = strchr(host, '@')) != NULL && strptr<path) /* UserId */ 1096: host = strptr; 1.32 cvs 1097: if ((port = strchr(host, ':')) != NULL && port>path) /* Port number */ 1.25 cvs 1098: port = NULL; 1099: 1100: strptr = host; /* Convert to lower-case */ 1.33 cvs 1101: while (strptr<path) 1102: { 1.25 cvs 1103: *strptr = tolower(*strptr); 1104: strptr++; 1.33 cvs 1105: } 1.25 cvs 1106: 1107: /* Does the URL contain a full domain name? This also works for a 1108: numerical host name. The domain name is already made lower-case 1109: and without a trailing dot. */ 1110: { 1.33 cvs 1111: char *dot = port ? port : path; 1112: if (dot > *filename && *--dot=='.') 1113: { 1114: char *orig=dot, *dest=dot+1; 1115: while((*orig++ = *dest++)); 1116: if (port) port--; 1117: path--; 1.25 cvs 1118: } 1119: } 1120: /* Chop off port if `:', `:80' (http), `:70' (gopher), or `:21' (ftp) */ 1.33 cvs 1121: if (port) 1122: { 1123: if (!*(port+1) || *(port+1)==used_sep) 1124: { 1125: if (!newname) 1126: { 1.25 cvs 1127: char *orig=port, *dest=port+1; 1128: while((*orig++ = *dest++)); 1.33 cvs 1129: } 1130: } 1131: else if ((!strncmp(access, "http", 4) && 1132: (*(port+1)=='8'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) || 1133: (!strncmp(access, "gopher", 6) && 1134: (*(port+1)=='7'&&*(port+2)=='0'&&(*(port+3)==used_sep||!*(port+3)))) || 1135: (!strncmp(access, "ftp", 3) && 1136: (*(port+1)=='2'&&*(port+2)=='1'&&(*(port+3)==used_sep||!*(port+3))))) { 1137: if (!newname) 1138: { 1139: char *orig=port, *dest=port+3; 1140: while((*orig++ = *dest++)); 1141: /* Update path position, Henry Minsky */ 1142: path -= 3; 1.25 cvs 1143: } 1.33 cvs 1144: } 1145: else if (newname) 1146: strncat(newname, port, (int) (path-port)); 1147: } 1.25 cvs 1148: 1.33 cvs 1149: if (newname) 1150: { 1.25 cvs 1151: char *newpath = newname+strlen(newname); 1152: strcat(newname, path); 1153: path = newpath; 1.28 cvs 1154: /* Free old copy */ 1155: TtaFreeMemory(*filename); 1.25 cvs 1156: *filename = newname; 1.33 cvs 1157: } 1.25 cvs 1158: return path; 1159: } 1160: 1161: 1162: /*---------------------------------------------------------------------- 1.29 cvs 1163: SimplifyUrl: simplify a URI 1.32 cvs 1164: A URI is allowed to contain the sequence xxx/../ which may be 1165: replaced by "" , and the sequence "/./" which may be replaced by DIR_STR. 1.28 cvs 1166: Simplification helps us recognize duplicate URIs. 1.25 cvs 1167: 1.28 cvs 1168: Thus, /etc/junk/../fred becomes /etc/fred 1169: /etc/junk/./fred becomes /etc/junk/fred 1.25 cvs 1170: 1.28 cvs 1171: but we should NOT change 1172: http://fred.xxx.edu/../.. 1.25 cvs 1173: 1174: or ../../albert.html 1175: 1.28 cvs 1176: In order to avoid empty URLs the following URLs become: 1.25 cvs 1177: 1178: /fred/.. becomes /fred/.. 1179: /fred/././.. becomes /fred/.. 1180: /fred/.././junk/.././ becomes /fred/.. 1181: 1.28 cvs 1182: If more than one set of `://' is found (several proxies in cascade) then 1183: only the part after the last `://' is simplified. 1.25 cvs 1184: 1.28 cvs 1185: Returns: A string which might be the old one or a new one. 1.25 cvs 1186: ----------------------------------------------------------------------*/ 1187: #ifdef __STDC__ 1.29 cvs 1188: void SimplifyUrl (char ** url) 1.25 cvs 1189: #else /* __STDC__ */ 1.29 cvs 1190: void SimplifyUrl (url) 1.28 cvs 1191: char **url; 1.25 cvs 1192: #endif /* __STDC__ */ 1193: { 1.28 cvs 1194: char *path, *p; 1195: char *newptr, *access; 1196: char *orig, *dest, *end; 1197: 1.33 cvs 1198: char used_sep; 1199: char *used_str; 1.32 cvs 1200: 1201: 1.28 cvs 1202: if (!url || !*url) 1203: return; 1204: 1.33 cvs 1205: if (strchr (*url, URL_SEP)) 1206: { 1207: used_str = URL_STR; 1208: used_sep = URL_SEP; 1209: } 1.32 cvs 1210: else 1.33 cvs 1211: { 1212: used_str = DIR_STR; 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: { 1.29 cvs 1383: char *return_value; 1384: char result[MAX_LENGTH]; 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: 1391: if (aName == NULL || relatedName == NULL) 1392: return (NULL); 1393: 1394: slashes = 0; 1395: after_access = NULL; 1396: p = aName; 1397: q = relatedName; 1398: for (; *p && (*p == *q); p++, q++) 1.27 cvs 1399: { 1400: /* Find extent of match */ 1.32 cvs 1401: if (*p == ':') 1.29 cvs 1402: after_access = p + 1; 1.28 cvs 1403: if (*p == DIR_SEP) 1.27 cvs 1404: { 1.29 cvs 1405: /* memorize the last slash position and count them */ 1.27 cvs 1406: last_slash = p; 1407: slashes++; 1.25 cvs 1408: } 1409: } 1410: 1.31 cvs 1411: /* q, p point to the first non-matching character or zero */ 1412: if (*q == EOS) 1413: { 1414: /* New name is a subset of the related name */ 1415: /* exactly the right length */ 1416: len = strlen (p); 1417: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1418: strcpy (return_value, p); 1419: } 1420: else if ((slashes < 2 && after_access == NULL) 1421: || (slashes < 3 && after_access != NULL)) 1422: { 1423: /* Two names whitout common path */ 1424: /* exactly the right length */ 1425: len = strlen (aName); 1426: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1427: strcpy (return_value, aName); 1428: } 1429: else 1430: { 1431: /* Some path in common */ 1432: if (slashes == 3 && strncmp (aName, "http:", 5) == 0) 1433: /* just the same server */ 1434: strcpy (result, last_slash); 1435: else 1436: { 1437: levels= 0; 1438: for (; *q && (*q != '#'); q++) 1439: if (*q == DIR_SEP) 1440: levels++; 1441: 1442: result[0] = 0; 1443: for (;levels; levels--) 1444: strcat (result, "../"); 1445: strcat (result, last_slash+1); 1446: } 1447: 1448: /* exactly the right length */ 1449: len = strlen (result); 1450: if ((return_value = (char *) TtaGetMemory (len + 1)) != NULL) 1451: strcpy (return_value, result); 1.25 cvs 1452: } 1.29 cvs 1453: return (return_value); 1.24 cvs 1454: }