Annotation of libwww/Library/src/HTAccess.html, revision 2.46

2.9       timbl       1: <HTML>
                      2: <HEAD>
2.45      frystyk     3: <TITLE>Access manager  for libwww</TITLE>
2.19      timbl       4: <NEXTID N="z11">
2.9       timbl       5: </HEAD>
2.5       timbl       6: <BODY>
2.39      frystyk     7: 
2.33      frystyk     8: <H1>Access Manager</H1>
2.39      frystyk     9: 
2.41      frystyk    10: <PRE>
                     11: /*
                     12: **     (c) COPYRIGHT CERN 1994.
                     13: **     Please first read the full copyright statement in the file COPYRIGH.
                     14: */
                     15: </PRE>
                     16: 
2.39      frystyk    17: This module keeps a list of valid protocol (naming scheme) specifiers
                     18: with associated access code.  It allows documents to be loaded given
                     19: various combinations of parameters.  New access protocols may be
                     20: registered at any time.<P>
                     21: 
                     22: This module is implemented by <A HREF="HTAccess.c">HTAccess.c</A>, and
                     23: it is a part of the <A NAME="z10"
2.46    ! frystyk    24: HREF="http://www.w3.org/hypertext/WWW/Library/User/Guide/Guide.html">Library
2.39      frystyk    25: of Common Code</A>. <P>
                     26: 
                     27: The module contains a lot of stuff but the main topics are:
                     28: 
                     29: <UL>
2.41      frystyk    30: <LI><A HREF="#Library">Initializing and Terminating the Library</A>
2.46    ! frystyk    31: <LI><A HREF="#log">Logging of requests</A>
2.39      frystyk    32: <LI><A HREF="#Methods">Management of access methods</A>
                     33: <LI><A HREF="#Addresses">A lot of hard coded addresses</A>
                     34: <LI><A HREF="#z1">The HTRequest structure</A>
                     35: <LI><A HREF="#z100">Management of the HTRequest structure</A>
                     36: <LI><A HREF="#LoadDoc">Functions for loading a document</A>
                     37: <LI><A HREF="#ClientHelp">Help functions for clients to get started</A>
                     38: <LI><A HREF="#PostDoc">Functions for posting a document</A>
                     39: <LI><A HREF="#ProtReg">Access Method Registration</H2></A>
                     40: </UL>
                     41: 
2.33      frystyk    42: 
                     43: <PRE>
                     44: #ifndef HTACCESS_H
1.1       timbl      45: #define HTACCESS_H
2.44      roeber     46: 
2.16      luotonen   47: #include "HTList.h"
2.39      frystyk    48: #include "HTChunk.h"
2.35      frystyk    49: </PRE>
1.1       timbl      50: 
2.46    ! frystyk    51: <H2>Global Flags</H2>
        !            52: 
        !            53: Flags and variables which may be set to control the Library
        !            54: 
        !            55: <H3>Redirections</H3>
        !            56: 
        !            57: The maximum number of redirections is pr. default 10. This prevents
        !            58: the library from going into an infinite loop which is kind of nice :-)
        !            59: 
2.35      frystyk    60: <PRE>
2.46    ! frystyk    61: extern int HTMaxRedirections;
2.33      frystyk    62: </PRE>
1.1       timbl      63: 
2.46    ! frystyk    64: <H3>Other Flags</H3>
2.36      frystyk    65: 
                     66: <PRE>
                     67: extern char * HTClientHost;            /* Name or number of telnetting host */
                     68: extern BOOL HTSecure;                  /* Disable security holes? */
2.46    ! frystyk    69: 
2.36      frystyk    70: extern char * HTImServer;              /* If I'm cern_httpd */
                     71: extern BOOL HTImProxy;                 /* If I'm cern_httpd as a proxy */
                     72: </PRE>
                     73: 
2.41      frystyk    74: <A NAME="Library"><H2>Initializing and Terminating the Library</H2></A>
                     75: 
2.46    ! frystyk    76: <IMG SRC="http://www.w3.org/hypertext/WWW/Icons/32x32/warning.gif">
2.41      frystyk    77: These two functions initiates memory and settings for the Library and
                     78: cleans up memory kept by the Library when about to exit the
                     79: application. It is highly recommended that they are used!
                     80: 
                     81: <PRE>
                     82: extern BOOL HTLibInit NOPARAMS;
                     83: extern BOOL HTLibTerminate NOPARAMS;
                     84: </PRE>
                     85: 
2.39      frystyk    86: <A NAME="Methods"><H2>Method Management</H2></A>
2.33      frystyk    87: 
2.41      frystyk    88: These are the valid methods, see <A
2.46    ! frystyk    89: HREF="http://www.w3.org/hypertext/WWW/Protocols/HTTP/Methods.html">HTTP
        !            90: Methods</A>. <P>
        !            91: 
        !            92: <B>NOTE:</B> the anchor list of allowed methods are not a bitflag, not
        !            93: at list.
2.33      frystyk    94: 
                     95: <PRE>
2.16      luotonen   96: typedef enum {
2.46    ! frystyk    97:        METHOD_INVALID  = 0x0,
        !            98:        METHOD_GET      = 0x1,
        !            99:        METHOD_HEAD     = 0x2,    
        !           100:        METHOD_POST     = 0x4,    
        !           101:        METHOD_PUT      = 0x8,    
        !           102:        METHOD_DELETE   = 0x10,
        !           103:        METHOD_LINK     = 0x20,
        !           104:        METHOD_UNLINK   = 0x40
2.16      luotonen  105: } HTMethod;
2.33      frystyk   106: </PRE>
                    107: 
                    108: <H3>Get Method Enumeration</H3>
2.16      luotonen  109: 
2.33      frystyk   110: Gives the enumeration value of the method as a function of the (char *) name.
2.16      luotonen  111: 
2.33      frystyk   112: <PRE>
2.46    ! frystyk   113: extern HTMethod HTMethod_enum PARAMS((CONST char * name));
2.16      luotonen  114: </PRE>
                    115: 
2.33      frystyk   116: <H3>Get Method String</H3>
2.16      luotonen  117: 
2.33      frystyk   118: The reverse of <I>HTMethod_enum()</I>
2.16      luotonen  119: 
2.33      frystyk   120: <PRE>
2.46    ! frystyk   121: extern CONST char * HTMethod_name PARAMS((HTMethod method));
2.16      luotonen  122: </PRE>
2.33      frystyk   123: 
2.35      frystyk   124: <HR>
                    125: <EM>This section might be move to the Access Authentication Module</EM>
                    126: 
2.36      frystyk   127: <H4>Match Template Against Filename</H4>
2.16      luotonen  128: <PRE>
2.40      frystyk   129: /* extern                                              HTAA_templateMatch()
2.16      luotonen  130: **             STRING COMPARISON FUNCTION FOR FILE NAMES
                    131: **                WITH ONE WILDCARD * IN THE TEMPLATE
                    132: ** NOTE:
                    133: **     This is essentially the same code as in HTRules.c, but it
                    134: **     cannot be used because it is embedded in between other code.
                    135: **     (In fact, HTRules.c should use this routine, but then this
                    136: **      routine would have to be more sophisticated... why is life
                    137: **      sometimes so hard...)
                    138: **
                    139: ** ON ENTRY:
2.37      frystyk   140: **     tmplate         is a template string to match the file name
2.16      luotonen  141: **                     agaist, may contain a single wildcard
                    142: **                     character * which matches zero or more
                    143: **                     arbitrary characters.
                    144: **     filename        is the filename (or pathname) to be matched
                    145: **                     agaist the template.
                    146: **
                    147: ** ON EXIT:
                    148: **     returns         YES, if filename matches the template.
                    149: **                     NO, otherwise.
                    150: */
2.40      frystyk   151: extern BOOL HTAA_templateMatch PARAMS((CONST char * tmplate, 
2.16      luotonen  152:                                       CONST char * filename));
2.35      frystyk   153: </PRE>
                    154: <HR>
2.16      luotonen  155: 
2.46    ! frystyk   156: The following have to be defined in advance of the other include files
2.10      timbl     157: because of circular references.
2.46    ! frystyk   158: 
2.33      frystyk   159: <PRE>
                    160: typedef struct _HTRequest HTRequest;
2.39      frystyk   161: typedef struct _HTNetInfo HTNetInfo;
2.10      timbl     162: 
2.14      luotonen  163: /*
2.39      frystyk   164: ** Callback to a protocol module
2.14      luotonen  165: */
2.39      frystyk   166: typedef int (*HTLoadCallBack)  PARAMS((HTRequest *     req));
2.14      luotonen  167: 
2.10      timbl     168: #include "HTAnchor.h"
                    169: #include <A
                    170: NAME="z3" HREF="HTFormat.html">"HTFormat.h"</A>
2.15      luotonen  171: #include "HTAAUtil.h"          /* HTAAScheme, HTAAFailReason */
2.14      luotonen  172: #include "HTAABrow.h"          /* HTAASetup */
2.33      frystyk   173: </PRE>
2.10      timbl     174: 
2.46    ! frystyk   175: <A NAME="HeaderMask"><H2>General HTTP Header Mask</H2></A>
        !           176: 
        !           177: This mask enables and disables which automaticly generated HTTP
        !           178: headers are to be sent in a request. If the bit is not turned on they
        !           179: are not sent. Extra header can be generated when initializing the <A
        !           180: HREF="#ExtraHeaders">ExtraHeaders field</A>. The first set is general
        !           181: headers and the second is request headers.
        !           182: 
        !           183: <PRE>
        !           184: typedef enum _HeaderEnum {
        !           185:     HT_DATE            = 0x1,
        !           186:     HT_MESSAGE_ID      = 0x2,
        !           187:     HT_MIME            = 0x4,                               /* MIME-Version */
        !           188: 
        !           189:     HT_ACCEPT          = 0x8,
        !           190:     HT_FROM            = 0x10,
        !           191:     HT_PRAGMA          = 0x20,
        !           192:     HT_REFERER         = 0x40,
        !           193:     HT_USER_AGENT      = 0x80
        !           194: } HeaderEnum;
        !           195: 
        !           196: #define <A NAME="DEF_HEAD">DEFAULT_HEADERS</A> HT_ACCEPT+HT_REFERER+HT_USER_AGENT
        !           197: </PRE>
        !           198: 
        !           199: <H2>Entity Header Mask</H2>
        !           200: 
        !           201: The entity headers contain information about the object sent in the
        !           202: HTTP transaction. See the <A HREF="HTAnchor.html">Anchor module</A>,
        !           203: for the storage of entity headers. This flag defines which headers are
        !           204: to be sent in a request together with an entity body (the <B>O</B>
        !           205: stands for <EM>object</EM>)
        !           206: 
        !           207: <PRE>
        !           208: typedef enum _EntityHeaderEnum {
        !           209:     HT_ALLOW           = 0x1,
        !           210:     HT_CONTENT_ENCODING        = 0x2,
        !           211:     HT_CONTENT_LANGUAGE        = 0x4,
        !           212:     HT_CONTENT_LENGTH  = 0x8,
        !           213:     HT_CTE             = 0x10,                 /* Content-Transfer-Encoding */
        !           214:     HT_CONTENT_TYPE    = 0x20,
        !           215:     HT_DERIVED_FROM    = 0x40,
        !           216:     HT_EXPIRES         = 0x80,
        !           217:     HT_LAST_MODIFIED   = 0x200,
        !           218:     HT_LINK            = 0x400,
        !           219:     HT_TITLE           = 0x800,
        !           220:     HT_URI             = 0x1000,
        !           221:     HT_VERSION         = 0x2000
        !           222: } EntityHeaderEnum;
        !           223: 
        !           224: #define <A NAME="DEF_ENTITY">DEFAULT_ENTITY_HEADERS</A> HT_CONTENT_TYPE+HT_CONTENT_LENGTH \
        !           225:        +HT_DERIVED_FROM+HT_VERSION
        !           226: </PRE>
        !           227: 
2.39      frystyk   228: <A NAME="Addresses"><H2>Default WWW Addresses</H2></A>
2.10      timbl     229: 
2.33      frystyk   230: These control the home page selection. To mess with these for normal browses
2.6       timbl     231: is asking for user confusion.
2.33      frystyk   232: <PRE>
                    233: #define LOGICAL_DEFAULT "WWW_HOME"           /* Defined to be the home page */
1.1       timbl     234: 
2.6       timbl     235: #ifndef PERSONAL_DEFAULT
2.33      frystyk   236: #define PERSONAL_DEFAULT "WWW/default.html"            /* in home directory */
2.6       timbl     237: #endif
2.33      frystyk   238: 
2.46    ! frystyk   239: /* If the home page isn't found, use this file: */
        !           240: #ifndef LAST_RESORT
        !           241: #define LAST_RESORT    "http://www.w3.org/"
2.6       timbl     242: #endif
2.33      frystyk   243: 
                    244: /* If one telnets to an access point it will look in this file for home page */
2.7       timbl     245: #ifndef REMOTE_POINTER
2.33      frystyk   246: #define REMOTE_POINTER  "/etc/www-remote.url"              /* can't be file */
2.7       timbl     247: #endif
2.33      frystyk   248: 
2.7       timbl     249: /* and if that fails it will use this. */
2.6       timbl     250: #ifndef REMOTE_ADDRESS
2.46    ! frystyk   251: #define REMOTE_ADDRESS  "http://www.w3.org/remote.html"   /* can't be file */
1.1       timbl     252: #endif
                    253: 
2.46    ! frystyk   254: /* Default log file name */
1.1       timbl     255: #ifndef DEFAULT_LOGFILE
2.46    ! frystyk   256: #define DEFAULT_LOGFILE                "www-log"
1.1       timbl     257: #endif
                    258: 
2.46    ! frystyk   259: #ifndef LOCAL_DEFAULT_FILE
        !           260: #define LOCAL_DEFAULT_FILE "/usr/local/lib/WWW/default.html"
1.1       timbl     261: #endif
                    262: 
2.33      frystyk   263: /* This is the default cache directory: */
2.46    ! frystyk   264: #ifndef HT_CACHE_ROOT
        !           265: #define HT_CACHE_ROOT          "/tmp"
2.23      frystyk   266: #endif
                    267: 
2.33      frystyk   268: /* The default directory for "save locally" and "save and execute" files: */
2.46    ! frystyk   269: #ifndef HT_TMP_ROOT
        !           270: #define HT_TMP_ROOT            "/tmp"
2.23      frystyk   271: #endif
2.33      frystyk   272: </PRE>
2.10      timbl     273: 
2.34      frystyk   274: <H2><A NAME="HTNetInfo">Protocol Specific Information</A></H2>
                    275: 
2.45      frystyk   276: This structure contains information about socket number, input buffer
                    277: for reading from the network etc. The structure is used through out
                    278: the protocol modules and is the refenrence point for introducing multi
                    279: threaded execution into the library, see specifications on <A
2.46    ! frystyk   280: HREF="http://www.w3.org/hypertext/WWW/Library/User/Features/multithread.html">Multiple
2.45      frystyk   281: Threads</A>.
2.34      frystyk   282: 
                    283: <PRE>
2.46    ! frystyk   284: typedef enum _SocAction {
        !           285:     SOC_INVALID = -1,
        !           286:     SOC_WRITE = 0,                             /* By default ready to write */
        !           287:     SOC_READ,
        !           288:     SOC_INTERRUPT
        !           289: } SocAction;
        !           290: 
2.39      frystyk   291: struct _HTNetInfo {
2.45      frystyk   292:     SOCKFD             sockfd;                         /* Socket descripter */
2.39      frystyk   293:     SockA              sock_addr;              /* SockA is defined in tcp.h */
2.36      frystyk   294:     HTInputSocket *    isoc;                                /* Input buffer */
2.46    ! frystyk   295:     SocAction          action;                 /* Result of the select call */
        !           296:     HTStream *         target;                             /* Target stream */
2.36      frystyk   297:     int                addressCount;        /* Attempts if multi-homed host */
2.39      frystyk   298:     time_t             connecttime;             /* Used on multihomed hosts */
2.36      frystyk   299:     struct _HTRequest *        request;           /* Link back to request structure */
2.39      frystyk   300: };
2.34      frystyk   301: </PRE>
                    302: 
2.36      frystyk   303: <EM><B>Note:</B> The AddressCount varaible is used to count the number
                    304: of attempt to connect to a multi-homed host so we know when to stop
                    305: trying new IP-addresses.</EM>
                    306: 
2.46    ! frystyk   307: <H2><A NAME="z1">The Request structure</A></H2>
        !           308: 
        !           309: When a request is handled, all kinds of things about it need to be
        !           310: passed along.  These are all put into a HTRequest structure. This is
        !           311: the most essential structure in the library. It contains two main
        !           312: categories of information regarding a request:
        !           313: 
        !           314: <UL>
        !           315: <LI>Application dependent information
        !           316: <LI>Library dependent information
        !           317: </UL>
        !           318: 
        !           319: Applications using the Library should <EM>never</EM> use the internal
        !           320: library dependent information. It's only because we dont have real
        !           321: classes that we can't hide it. <P>
        !           322: 
        !           323: <B>Note:</B> If you reuse the request structure for more than one
        !           324: request then make sure that the request is re-initialized, so that no
        !           325: `old' data is reused, see <A HREF="#z100">functions to manipulate
        !           326: HTRequest Structure</A>. The library handles its own internal
        !           327: information from request to request but the information set by the
        !           328: caller is untouched. <P>
2.31      frystyk   329: 
2.46    ! frystyk   330: The elements of the request structure are as follows:
2.39      frystyk   331: 
                    332: <PRE>
                    333: struct _HTRequest {
                    334: </PRE>
2.19      timbl     335: 
2.46    ! frystyk   336: <H3>Application Dependent - Set by the caller of HTAccess</H3>
2.19      timbl     337: 
2.39      frystyk   338: <PRE>
                    339:     <A HREF="#Methods">HTMethod</A>    method;
                    340: </PRE>
                    341: 
                    342: An enum used to specify the HTTP <A NAME="z7"
2.46    ! frystyk   343: HREF="../../Protocols/HTTP/Methods.html">method</A> used for the
        !           344: actual request. The default value is <A
        !           345: HREF="#Methods"><CODE>GET</CODE></A>.
        !           346: 
        !           347: <H4>HTTP Header Information</H4>
2.39      frystyk   348: 
                    349: <PRE>
                    350:     HTList *   conversions;
                    351: </PRE>
                    352: 
2.46    ! frystyk   353: NULL, or a <EM>local</EM> list of specific conversions which the
        !           354: format manager can do in order to fulfill the request.  It typically
        !           355: points to a list set up on initialisation time for example by <A
        !           356: HREF="HTInit.html">HTInit()</A>. There is also a <A
        !           357: HREF="HTFormat.html#z17"><EM>global</EM></A> list of conversions which
        !           358: contains a generic set of possible conversions.
2.39      frystyk   359: 
                    360: <PRE>
                    361:     HTList *   encodings;
                    362: </PRE>
                    363: 
                    364: The list of encodings acceptable in the output stream.
                    365: 
                    366: <PRE>
                    367:     HTList *   languages;
                    368: </PRE>
                    369: 
                    370: The list of (human) language values acceptable in the response. The default
                    371: is all languages.
                    372: 
                    373: <PRE>
2.46    ! frystyk   374:     HeaderEnum         HeaderMask;
        !           375:     EntityHeaderEnum   EntityMask;
        !           376: </PRE>
        !           377: 
        !           378: These two masks defines which headers to include in a HTTP request (or
        !           379: any other MIME-like protocol). the first mask defines what
        !           380: autogenerated HTTP header fields should be sent in the HTTP
        !           381: request. The default value is <A
        !           382: HREF="DEF_HEAD">DEFAULT_HEADERS</A>. The second defines what entity
        !           383: headers should be sent along with an entity (using PUT or POST).
        !           384: 
        !           385: <PRE>
        !           386:     HTParentAnchor *parentAnchor;
        !           387: </PRE>
        !           388: 
        !           389: If this parameter is set then a `Referer: &lt;parent address&gt; can
        !           390: be generated in the request to the server, see <A
        !           391: HREF="http://www.w3.org/hypertext/WWW/Protocols/HTTP/HTRQ_Headers.html#z14">
        !           392: Referer field in a HTTP Request</A>
        !           393: 
        !           394: <PRE>
        !           395:    <A NAME="#ExtraHeaders">char * ExtraHeaders;</A>
2.39      frystyk   396: </PRE>
                    397: 
2.46    ! frystyk   398: Extra header information can be send along with a request using this
        !           399: variable. The text is sent as is so it must be preformatted with
        !           400: &lt;CRLF&gt; line terminators.
        !           401: 
        !           402: <H4>Streams From Network to Application</H4>
2.39      frystyk   403: 
                    404: <PRE>
2.46    ! frystyk   405:     HTStream * output_stream; 
2.39      frystyk   406: </PRE>
                    407: 
2.46    ! frystyk   408: The output stream is to be used to put data down to as they come in
        !           409: <B>from</B> the network and back to the application. The default value
        !           410: is <CODE>NULL</CODE> which means that the stream goes to the user
        !           411: (display).
2.39      frystyk   412: 
                    413: <PRE>
2.46    ! frystyk   414:     HTAtom *   output_format;
2.39      frystyk   415: </PRE>
                    416: 
2.46    ! frystyk   417: The desired format of the output stream. This can be used to get
        !           418: unconverted data etc. from the library. If <CODE>NULL</CODE>, then <A
        !           419: HREF="HTFormat.html#FormatTypes">WWW_PRESENT</A> is default value.
2.45      frystyk   420: 
                    421: <PRE>
2.46    ! frystyk   422:     HTStream*  error_stream;
2.45      frystyk   423: </PRE>
                    424: 
                    425: All object bodies sent from the server with status codes different
                    426: from <CODE>200 OK</CODE> will be put down this stream. This can be
                    427: used as a debug window etc. If the value is NULL (default) then the
                    428: stream used is <A HREF="HTFormat.html#BlackHole">HTBlackHole</A>.
2.39      frystyk   429: 
2.46    ! frystyk   430: <H4>Streams From Application to Network</H4>
        !           431: 
2.39      frystyk   432: <PRE>
2.46    ! frystyk   433:     HTStream * input_stream; 
        !           434: </PRE>
        !           435: 
        !           436: The input stream is to be used by the <CODE>PostCallBack</CODE>
        !           437: function to put data out on the network. The user should not
        !           438: initialize this field.
        !           439: 
        !           440: <PRE>
        !           441:     HTAtom *   input_format;
        !           442: </PRE>
        !           443: 
        !           444: The desired format of the output stream. This can be used to upload
        !           445: converted data to a remote server. If <CODE>NULL</CODE>, then <A
        !           446: HREF="HTFormat.html#FormatTypes">WWW_SOURCE</A> is default value.
        !           447: 
        !           448: <PRE>
        !           449:     int (*PostCallBack)                PARAMS((struct _HTRequest *     request,
        !           450:                                        HTStream *              target));
2.39      frystyk   451: </PRE>
                    452: 
2.46    ! frystyk   453: The call back function which is called when the current request is
        !           454: ready for sending (posting) the data object. The request is the
        !           455: current request so that the application knows which post we are
        !           456: handling. The function must have the same return values as the other
        !           457: <A HREF="#LoadDoc">Load functions</A>.
        !           458: 
        !           459: <H4>Other Flags</H4>
2.39      frystyk   460: 
                    461: <PRE>
2.43      frystyk   462:     BOOL BlockingIO;
2.46    ! frystyk   463:     BOOL ForceReload;
2.43      frystyk   464: </PRE>
                    465: 
                    466: This flag can be set to override if a protocol module is registered as
                    467: using non-blocking IO.
                    468: 
                    469: <PRE>
2.46    ! frystyk   470:     BOOL (*<A NAME="z9"> callback</A> ) PARAMS((struct _HTRequest* request,
        !           471:                                                void *param));
        !           472: </PRE>
        !           473: 
        !           474: A function to be called back in the event that a file has been saved
        !           475: to disk by HTSaveAndCallBack for example.
        !           476: 
        !           477: <PRE>
        !           478:     void *     context;
2.39      frystyk   479: </PRE>
                    480: 
2.46    ! frystyk   481: An arbitrary pointer passed to HTAccess and passed back as a parameter
        !           482: to the <A NAME="z10" HREF="#z9">callback</A>.
2.39      frystyk   483: 
2.46    ! frystyk   484: <H3>Library Dependent - Set by Library</H3>
2.39      frystyk   485: 
2.46    ! frystyk   486: None of the bits below may be looked at by a WWW application. The
        !           487: Library handles the cleanup by itself.
2.39      frystyk   488: 
                    489: <PRE>
                    490:     HTParentAnchor*    anchor;
                    491: </PRE>
                    492: 
                    493: The anchor for the object in question.  Set immediately by HTAcesss.
                    494: Used by the protocol and parsing modules.  Valid thoughout the access.
                    495: 
                    496: <PRE>
                    497:     HTChildAnchor *    childAnchor;    /* For element within the object  */
                    498: </PRE>
                    499: 
                    500: The anchor for the sub object if any.  The object builder should
2.46    ! frystyk   501: ensure that is is selected, highlighted, etc when the object is
2.39      frystyk   502: loaded.
                    503: 
                    504: <PRE>
2.46    ! frystyk   505:     struct _HTRequest *        CopyRequest;
        !           506: </PRE>
        !           507: 
        !           508: We need to know if we have a remote request sending data along with
        !           509: this request.
        !           510: 
        !           511: <PRE>
        !           512:     void *     using_cache;
        !           513:     BOOL       using_proxy;
2.39      frystyk   514: </PRE>
                    515: 
2.46    ! frystyk   516: Pointer to cache element if cache hit anfd if using proxy
2.19      timbl     517: 
2.25      luotonen  518: <PRE>
2.46    ! frystyk   519:     BOOL       error_block;            /* YES if stream has been used    */
        !           520:     HTList *   error_stack;            /* List of errors                 */
        !           521: </PRE>
2.25      luotonen  522: 
2.46    ! frystyk   523: These two fields are used by the error reporting system to keep a
        !           524: stack of messages.
        !           525: 
        !           526: <PRE>
        !           527:     HTNetInfo *        net_info;               /* Information about socket etc. */
        !           528:     int                redirections;           /* Number of redirections */
2.34      frystyk   529: </PRE>
2.46    ! frystyk   530: 
        !           531: Protocol specific information, socket number etc.
        !           532: 
2.34      frystyk   533: <PRE>
2.46    ! frystyk   534:     char *     redirect;               /* Location or URI */
        !           535:     char *     WWWAAScheme;            /* WWW-Authenticate scheme */
        !           536:     char *     WWWAARealm;             /* WWW-Authenticate realm */
        !           537:     char *     WWWprotection;          /* WWW-Protection-Template */
2.25      luotonen  538: </PRE>
2.39      frystyk   539: 
2.46    ! frystyk   540: Information taken from the MIME header specifically oriented towards
        !           541: the request (not the object itself)
        !           542: 
2.39      frystyk   543: <PRE>
2.46    ! frystyk   544:     char *     authorization;          /* Authorization: field           */
        !           545:     HTAAScheme scheme;                 /* Authentication scheme used     */
        !           546:     HTInputSocket *    isoc;           /* InputSocket object for reading */
        !           547:     HTAtom *   content_type;           /* Content-Type:                  */
        !           548:     HTAtom *   content_language;       /* Language                       */
        !           549:     HTAtom *   content_encoding;       /* Encoding                       */
        !           550:     int                content_length;         /* Content-Length:                */
2.39      frystyk   551: </PRE>
                    552: 
2.46    ! frystyk   553: These header fields are only used by the server and will be removed at some
        !           554: point.
        !           555: 
2.21      luotonen  556: <PRE>
2.46    ! frystyk   557:     HTList *   valid_schemes;          /* Valid auth.schemes             */
        !           558:     HTAssocList **     scheme_specifics;/* Scheme-specific parameters    */
        !           559:     char *     authenticate;           /* WWW-authenticate: field */
        !           560:     char *     prot_template;          /* WWW-Protection-Template: field */
        !           561:     HTAASetup *        setup;                  /* Doc protection info            */
        !           562:     HTAARealm *        realm;                  /* Password realm                 */
        !           563:     char *     dialog_msg;             /* Authentication prompt (client) */
2.19      timbl     564: </PRE>
2.46    ! frystyk   565: 
        !           566: These fields are used by the HTTP access authentication used by a
        !           567: client application.
        !           568: 
2.19      timbl     569: <PRE>
2.10      timbl     570: };
2.31      frystyk   571: </PRE>
                    572: 
                    573: <H2><A NAME="z100">Functions to Manipulate a HTRequest Structure</A></H2>
                    574: 
                    575: Just to make things easier especially for clients, here are some functions to
                    576: manipulate the request structure:
                    577: 
                    578: <H3>Create blank request</H3>This request has defaults in -- in
2.9       timbl     579: most cases it will need some information
                    580: added before being passed to HTAccess,
                    581: but it will work as is for a simple
                    582: request.
2.14      luotonen  583: <PRE>
2.40      frystyk   584: extern HTRequest * HTRequest_new NOPARAMS;
2.31      frystyk   585: </PRE>
2.14      luotonen  586: 
2.31      frystyk   587: <H3>Delete request structure</H3>Frees also conversion list hanging
2.19      timbl     588: from req->conversions.
2.14      luotonen  589: <PRE>
2.40      frystyk   590: extern void HTRequest_delete PARAMS((HTRequest * req));
2.31      frystyk   591: </PRE>
1.1       timbl     592: 
2.31      frystyk   593: <H3>Clear a request structure</H3>
2.46    ! frystyk   594: 
        !           595: Clears a request structure so that it can be reused. The only thing
        !           596: that differs from using free/new is that the list of conversions is
        !           597: kept. <P>
        !           598: 
        !           599: <B>NOTE:</B> It is <B>NOT</B> recommended to reuse a request structure!!!
        !           600: 
2.31      frystyk   601: <PRE>
                    602: extern void HTRequest_clear PARAMS((HTRequest * req));
                    603: </PRE>
2.9       timbl     604: 
2.39      frystyk   605: <A NAME="LoadDoc"><H2>Functions for Loading a Document</H2></A>
                    606: 
                    607: There are several different ways of loading a document. However, the
                    608: major difference between them is whether the document is referenced by
                    609: 
                    610: <UL>
                    611: <LI><A HREF="#Relative">Relative URI</A>
                    612: <LI><A HREF="#Absolute">Absolute URI</A>
                    613: <LI><A HREF="#Anchor">Anchor element</A> or
                    614: <LI>Contains keywords for <A HREF="#RelSearch">searching an relative URI</A>
                    615: <LI>Contains keywords for <A HREF="#AbsSearch">searching an absolute URI</A>
                    616: </UL>
                    617: 
                    618: <B>NOTE:</B> From release 3.0 of the Library, the return codes from
                    619: the loading functions are no mode <CODE>BOOL</CODE>, that is
                    620: <CODE>YES</CODE> or <CODE>NO</CODE>. Insted they have been replaced
                    621: with the following set of return codes defined in the <A
                    622: HREF="HTUtils.html#ReturnCodes">Utility module</A>:
                    623: 
2.5       timbl     624: <DL>
2.39      frystyk   625: <DT>HT_WOULD_BLOCK
                    626: <DD>An I/O operation would block
                    627: 
                    628: <DT>HT_ERROR
                    629: <DD>Error has occured
                    630: 
                    631: <DT>HT_LOADED
                    632: <DD>Success
                    633: 
                    634: <DT>HT_NO_DATA
                    635: <DD>Success, but no document loaded. This might be the situation when a 
                    636: telnet sesssion is started etc.
2.5       timbl     637: </DL>
                    638: 
2.39      frystyk   639: However, a general rule about the return codes is that <B>ERRORS</B>
                    640: have a <EM>negative</EM> value whereas <B>SUCCESS</B> has a
                    641: <EM>positive</EM> value. <P>
1.1       timbl     642: 
2.39      frystyk   643: There are also some functions to help the client getting started with
                    644: <A HREF="#ClientHelp">the first URI</A>.
1.1       timbl     645: 
2.46    ! frystyk   646: <A NAME="Relative"><H3>Load a document from relative URL</H3></A>
1.1       timbl     647: 
2.39      frystyk   648: <PRE>
                    649: extern int HTLoadRelative      PARAMS((CONST char *    relative_name,
                    650:                                        HTParentAnchor* here,
                    651:                                        HTRequest *     request));
2.5       timbl     652: </PRE>
2.39      frystyk   653: 
2.46    ! frystyk   654: <A NAME="Absolute"></A><H3>Load a document from absolute URL</H3>
1.1       timbl     655: 
2.5       timbl     656: <PRE>
2.39      frystyk   657: extern int HTLoadAbsolute      PARAMS((CONST char *    addr,
                    658:                                        HTRequest *     request));
2.5       timbl     659: </PRE>
2.39      frystyk   660: 
                    661: <H3>Load a document from absolute name to a stream</H3>
                    662: 
2.5       timbl     663: <PRE>
2.39      frystyk   664: extern int HTLoadToStream      PARAMS((CONST char *    addr,
                    665:                                        BOOL            filter,
                    666:                                        HTRequest *     request));
2.5       timbl     667: </PRE>
1.1       timbl     668: 
2.46    ! frystyk   669: <A NAME="Anchor"><H3>Load a document from anchor</H3></A>
1.1       timbl     670: 
2.39      frystyk   671: The anchor parameter may be a child anchor. The anchor in the request
                    672: is set to the parent anchor. The recursive function keeps the error
                    673: stack in the request structure so that no information is lost having
                    674: more than one call. See also <A HREF="#BindAnchor">HTBindAnchor()</A>.
1.1       timbl     675: 
2.39      frystyk   676: <PRE>
2.41      frystyk   677: extern int HTLoadAnchor                PARAMS((HTAnchor  *     a,
                    678:                                        HTRequest *     request));
2.39      frystyk   679: extern int HTLoadAnchorRecursive PARAMS((HTAnchor *    a,
                    680:                                        HTRequest *     request));
2.5       timbl     681: </PRE>
                    682: 
2.39      frystyk   683: <H3>Load a Document</H3>
                    684: 
2.46    ! frystyk   685: These are two internal routines for loading a document which has an
2.41      frystyk   686: address AND a matching anchor.  (The public routines are called with
                    687: one OR the other.)  This is recursively called from file load module
                    688: to try ftp (though this will be obsolete in the next major
                    689: release).<P>
1.1       timbl     690: 
2.39      frystyk   691: If <CODE>keep_error_stack</CODE> is YES then the error (or info) stack
                    692: is not cleared from the previous call.
1.1       timbl     693: 
2.39      frystyk   694: <PRE>
2.41      frystyk   695: extern int HTLoad              PARAMS((HTRequest * request,
                    696:                                        BOOL keep_error_stack));
                    697: </PRE>
                    698: 
                    699: <PRE>
                    700: extern BOOL HTLoadTerminate    PARAMS((HTRequest * request, int status));
2.5       timbl     701: </PRE>
                    702: 
2.46    ! frystyk   703: <A NAME="RelSearch"><H3>Search Using Relative URL</H3></A>
2.39      frystyk   704: 
                    705: Performs a search on word given by the user. Adds the search words to
                    706: the end of the current address and attempts to open the new address.
                    707: 
2.5       timbl     708: <PRE>
2.39      frystyk   709: extern int HTSearch            PARAMS((CONST char *    keywords,
                    710:                                        HTParentAnchor* here,
                    711:                                        HTRequest *     request));
2.5       timbl     712: </PRE>
2.39      frystyk   713: 
2.46    ! frystyk   714: <A NAME="AbsSearch"><H3>Search using Absolute URL</H3></A>
2.39      frystyk   715: 
                    716: Performs a keyword search on word given by the user. Adds the keyword
                    717: to the end of the current address and attempts to open the new
                    718: address.
2.5       timbl     719: 
2.33      frystyk   720: <PRE>
2.40      frystyk   721: extern int HTSearchAbsolute    PARAMS((CONST char *    keywords,
2.39      frystyk   722:                                        CONST char *    indexname,
                    723:                                        HTRequest *     request));
2.5       timbl     724: </PRE>
2.24      luotonen  725: 
2.39      frystyk   726: 
                    727: <A NAME="ClientHelp"><H2>Help Function for Clients to get started</H2></A>
                    728: 
                    729: These function helps the client to load the first document. They are
                    730: not mandatory to use - but they make life easier!
                    731: 
                    732: <A NAME="BindAnchor"><H3>Bind an anchor to a request structure without
                    733: loading</H3></A>
2.24      luotonen  734: 
                    735: <PRE>
2.39      frystyk   736: extern BOOL HTBindAnchor PARAMS((HTAnchor *anchor, HTRequest *request));
2.24      luotonen  737: </PRE>
                    738: 
2.39      frystyk   739: <A NAME="HomePage"><H3>Generate the Anchor for the Home Page</H3></A>
2.24      luotonen  740: 
2.39      frystyk   741: As it involves file access, this should only be done once when the
                    742: program first runs. This is a default algorithm using the
                    743: <CODE>WWW_HOME</CODE> environment variable.
2.20      frystyk   744: 
                    745: <PRE>
2.40      frystyk   746: extern HTParentAnchor * HTHomeAnchor NOPARAMS;
2.39      frystyk   747: </PRE>
                    748: 
                    749: <H3>Find Related Name</H3>
                    750: 
                    751: Creates a local file URI that can be used as a relative name when
                    752: calling HTParse() to expand a relative file name to an absolute
                    753: one. <P>
                    754: 
2.46    ! frystyk   755: The code for this routine originates from the Line Mode Browser and
        !           756: was moved here by <EM>howcome@dxcern.cern.ch</EM> in order for all
        !           757: clients to take advantage.<P>
2.39      frystyk   758: 
2.20      frystyk   759: <PRE>
2.39      frystyk   760: extern char *  HTFindRelatedName NOPARAMS;
                    761: </PRE>
                    762: 
                    763: <A NAME="PostDoc"><H2>Functions for Posting a Document</H2></A>
                    764: 
2.46    ! frystyk   765: <B>NOTE:</B> The Posting functions are used to send a data object
        !           766: along with the request. The functions have the same set of return
        !           767: codes as for the <A HREF="#LoadDoc">Load Functions</A>.
2.20      frystyk   768: 
2.39      frystyk   769: <H3>Get a Save Stream</H3>
2.20      frystyk   770: 
2.39      frystyk   771: <H4>On Entry,</H4>
2.5       timbl     772: <DL>
2.9       timbl     773: <DT>request->anchor
                    774: <DD> is valid anchor which
                    775: has previously beeing loaded
2.5       timbl     776: </DL>
                    777: 
2.39      frystyk   778: <H4>On exit,</H4>
2.5       timbl     779: <DL>
                    780: <DT>returns
2.6       timbl     781: <DD> 0 if error else a stream
                    782: to save the object to.
2.5       timbl     783: </DL>
                    784: 
                    785: <PRE>
2.46    ! frystyk   786: extern HTStream * HTSaveStream PARAMS((HTRequest * request));
        !           787: </PRE>
        !           788: 
        !           789: <H3>Copy an Anchor</H3>
        !           790: 
        !           791: Fetch the URL (possibly local file URL) and send it using either
        !           792: <B>PUT</B> or <B>POST</B> directly to the remote destination using
        !           793: HTTP, that is remote copy of object <EM>O</EM> from <EM>A</EM> to
        !           794: <EM>B</EM> where <EM>A</EM> might be the host of the application. The
        !           795: caller can decide the exact method used and which HTTP header fields
        !           796: to transmit by setting the user fields in the destination request
        !           797: structure.
        !           798: 
        !           799: <PRE>
        !           800: extern int HTCopyAnchor                PARAMS((HTAnchor *      src_anchor,
        !           801:                                        HTRequest *     src_req,
        !           802:                                        HTParentAnchor *dest_anchor,
        !           803:                                        HTRequest *     dest_req));
        !           804: </PRE>
        !           805: 
        !           806: 
        !           807: <H3>Upload an Anchor</H3>
        !           808: 
        !           809: Send the contents (in hyperdoc) of the source anchor using either
        !           810: <B>PUT</B> or <B>POST</B> to the remote destination using HTTP. The
        !           811: caller can decide the exact method used and which HTTP header fields
        !           812: to transmit by setting the user fields in the request structure.
        !           813: <EM>Format conversion</EM> can be made on the fly by setting the <A
        !           814: HREF="#input_format">input_format field</A> in the destination request
        !           815: structure. If the content-length is unknown (-1) then a <A
        !           816: HREF="HTConLen.html">content-length counter</A> is automaticly put
        !           817: into the stream pipe.
        !           818: 
        !           819: 
        !           820: <PRE>
        !           821: extern int HTUploadAnchor      PARAMS((HTAnchor *      src_anchor,
        !           822:                                        HTParentAnchor *dest_anchor,
        !           823:                                        HTRequest *     dest_req));
2.39      frystyk   824: </PRE>
2.5       timbl     825: 
1.1       timbl     826: 
2.39      frystyk   827: <A NAME="ProtReg"><H2>Access Method Registration</H2></A>
1.1       timbl     828: 
2.39      frystyk   829: An access method is defined by an HTProtocol structure which point to
                    830: the routines for performing the various logical operations on an
                    831: object: in HTTP terms, GET, PUT, and POST. The access methods
                    832: supported in the Library are initiated automaticly using the private
                    833: function <CODE>HTAccessInit()</CODE> <B>if not</B> defined
2.46    ! frystyk   834: <CODE>HT_NO_INIT</CODE> <P>
2.39      frystyk   835: 
                    836: Each of these routine takes as a parameter a <A NAME="z2"
                    837: HREF="#z1">request structure</A> containing details of the request.
                    838: When the protocol class routine is called, the anchor element in the
                    839: request is already valid (made valid by HTAccess).
                    840: 
                    841: <PRE>
                    842: typedef enum _HTSocBlock {
                    843:     SOC_BLOCK,
2.42      frystyk   844:     SOC_NON_BLOCK
2.39      frystyk   845: } HTSocBlock;
                    846: 
                    847: typedef struct _HTProtocol {
                    848:     char *     name;
                    849:     HTSocBlock block;  
                    850:     int                (*load)         PARAMS((HTRequest *     request));
                    851:     HTStream*  (*saveStream)   PARAMS((HTRequest *     request));
                    852:     HTStream*  (*postStream)   PARAMS((HTRequest *     request,
                    853:                                        HTParentAnchor* postTo));
                    854: } HTProtocol;
1.1       timbl     855: 
2.40      frystyk   856: extern BOOL HTRegisterProtocol PARAMS((HTProtocol * protocol));
2.42      frystyk   857: extern void HTDisposeProtocols NOPARAMS;
2.5       timbl     858: </PRE>
1.1       timbl     859: 
2.39      frystyk   860: <H3>Uses Protocol Blocking IO</H3>
1.1       timbl     861: 
2.39      frystyk   862: A small function to make life easier. Returns <CODE>YES</CODE> or
2.42      frystyk   863: <CODE>NO</CODE>. If the Library is run in NON-INTERACTIVE MODE then
                    864: the function always returns YES;
2.38      howcome   865: 
                    866: <PRE>
2.40      frystyk   867: extern BOOL HTProtocolBlocking PARAMS((HTRequest *     request));
2.38      howcome   868: </PRE>
                    869: 
2.39      frystyk   870: end
2.25      luotonen  871: 
                    872: <PRE>
1.1       timbl     873: #endif /* HTACCESS_H */
2.25      luotonen  874: </PRE>
                    875: end of HTAccess
                    876: </BODY>
2.9       timbl     877: </HTML>

Webmaster