View source with raw comments or as raw
   1/*  Part of SWI-Prolog
   2
   3    Author:        Jan Wielemaker
   4    E-mail:        J.Wielemaker@vu.nl
   5    WWW:           http://www.swi-prolog.org
   6    Copyright (c)  2002-2016, University of Amsterdam,
   7                              VU University Amsterdam
   8    All rights reserved.
   9
  10    Redistribution and use in source and binary forms, with or without
  11    modification, are permitted provided that the following conditions
  12    are met:
  13
  14    1. Redistributions of source code must retain the above copyright
  15       notice, this list of conditions and the following disclaimer.
  16
  17    2. Redistributions in binary form must reproduce the above copyright
  18       notice, this list of conditions and the following disclaimer in
  19       the documentation and/or other materials provided with the
  20       distribution.
  21
  22    THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  23    "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  24    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
  25    FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
  26    COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
  27    INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
  28    BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
  29    LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
  30    CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  31    LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
  32    ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
  33    POSSIBILITY OF SUCH DAMAGE.
  34*/
  35
  36:- module(http_client,
  37          [ http_get/3,                 % +URL, -Reply, +Options
  38            http_delete/3,              % +URL, -Reply, +Options
  39            http_post/4,                % +URL, +In, -Reply, +Options
  40            http_put/4,                 % +URL, +In, -Reply, +Options
  41            http_patch/4,               % +URL, +In, -Reply, +Options
  42            http_read_data/3,           % +Header, -Data, :Options
  43            http_disconnect/1           % +What
  44          ]).
  45:- use_module(library(http/http_open)).
  46:- use_module(library(uri)).
  47:- use_module(http_header).
  48:- use_module(http_stream).
  49:- use_module(library(memfile)).
  50:- use_module(library(lists)).
  51:- use_module(library(error)).
  52:- use_module(library(option)).
  53
  54:- meta_predicate
  55    http_read_data(+, -, :).
  56
  57:- multifile
  58    http_convert_data/4,            % http_read_data plugin-hook
  59    http:post_data_hook/3.
  60
  61:- predicate_options(http_get/3, 3,
  62                     [ pass_to(http_open/3, 3),
  63                       pass_to(http_read_data/3, 3)
  64                     ]).
  65:- predicate_options(http_delete/3, 3, [pass_to(http_get/3, 3)]).
  66:- predicate_options(http_post/4, 4, [pass_to(http_get/3, 3)]).
  67:- predicate_options(http_put/4, 4, [pass_to(http_post/4, 4)]).
  68:- predicate_options(http_read_data/3, 3,
  69                     [ to(any),
  70                       content_type(any),
  71                       form_data(oneof([form,mime])),
  72                       input_encoding(encoding),
  73                       on_filename(callable)
  74                     ]).
  75
  76
  77/** <module> HTTP client library
  78
  79This library provides  the  four  basic   HTTP  client  actions:  =GET=,
  80=DELETE=, =POST= and =PUT=. In   addition, it provides http_read_data/3,
  81which is used by library(http/http_parameters) to  decode =POST= data in
  82server applications.
  83
  84This library is based on http_open/3,  which   opens  a  URL as a Prolog
  85stream. The reply  is  processed   by  http_read_data/3.  The  following
  86content-types are supported. Options passed   to  http_get/3 and friends
  87are passed to  http_read_data/3,  which  in   turn  passes  them  to the
  88conversion predicates. Support for additional content types can be added
  89by extending the multifile predicate http_client:http_convert_data/4.
  90
  91  - 'application/x-www-form-urlencoded'
  92  Built in.  Converts form-data into a list of `Name=Value` terms.
  93  - 'application/x-prolog'
  94  Built in.  Reads a single Prolog term.
  95  - 'multipart/form-data'
  96  Processed if library(http/http_multipart_plugin) is loaded.  This
  97  format should be used to handle web forms that upload a file.
  98  - 'text/html' | 'text/xml'
  99  Processed if library(http/http_sgml_plugin) is loaded. See load_html/3
 100  for details and load_xml/3 for details.  The output is often processed
 101  using xpath/3.
 102  - 'application/json' | 'application/jsonrequest'
 103  Processed if library(http/http_json) is loaded.  The option
 104  json_object(As) can be used to return a term json(Attributes)
 105  (`As` is `term`) or a dict (`As` is `json`).
 106*/
 107
 108                 /*******************************
 109                 *              GET             *
 110                 *******************************/
 111
 112%!  http_get(+URL, -Data, +Options) is det.
 113%
 114%   Get data from a URL server and   convert it to a suitable Prolog
 115%   representation based on the =|Content-Type|= header and plugins.
 116%   This predicate is the common implementation   of the HTTP client
 117%   operations.  The  predicates  http_delete/3,    http_post/4  and
 118%   http_put/4   call   this   predicate     with   an   appropriate
 119%   method(+Method) option and ---for  http_post/4 and http_put/4---
 120%   a post(+Data) option.
 121%
 122%   Options are passed to http_open/3   and  http_read_data/3. Other
 123%   options:
 124%
 125%     - reply_header(-Fields)
 126%     Synonym for headers(Fields) from http_open/3.  Provided for
 127%     backward compatibility.  Note that http_version(Major-Minor)
 128%     is missing in the new version.
 129
 130http_get(URL, Data, Options) :-
 131    headers_option(Options, Options1, Headers),
 132    option(reply_header(Headers), Options, _),
 133    http_open(URL, In, Options1),
 134    delete(Headers, transfer_encoding(_), Headers1),
 135    call_cleanup(
 136        http_read_data(In, Headers1, Data, Options),
 137        close(In)).
 138
 139headers_option(Options, Options1, Headers) :-
 140    option(headers(Headers), Options),
 141    !,
 142    Options1 = Options.
 143headers_option(Options, [headers(Headers)|Options], Headers).
 144
 145
 146%!  http_delete(+URL, -Data, +Options) is det.
 147%
 148%   Execute a =DELETE= method on the  server. Arguments are the same
 149%   as  for  http_get/3.  Typically  one   should  pass  the  option
 150%   status_code(-Code) to assess and evaluate   the  returned status
 151%   code. Without, codes other than 200 are interpreted as an error.
 152%
 153%   @tbd Properly map the 201, 202 and 204 replies.
 154%   @see Implemented on top of http_get/3.
 155
 156http_delete(URL, Data, Options) :-
 157    http_get(URL, Data, [method(delete)|Options]).
 158
 159
 160%!  http_post(+URL, +Data, -Reply, +Options) is det.
 161%
 162%   Issue  an  HTTP   =POST=   request.    Data   is   posted  using
 163%   http_post_data/3. The HTTP server reply   is  returned in Reply,
 164%   using the same rules as for http_get/3.
 165%
 166%   @see Implemented on top of http_get/3.
 167
 168http_post(URL, Data, Reply, Options) :-
 169    http_get(URL, Reply,
 170             [ post(Data)
 171             | Options
 172             ]).
 173
 174%!  http_put(+URL, +Data, -Reply, +Options)
 175%
 176%   Issue an HTTP =PUT=  request.  Arguments   are  the  same as for
 177%   http_post/4.
 178%
 179%   @see Implemented on top of http_post/4.
 180
 181http_put(URL, In, Out, Options) :-
 182    http_post(URL, In, Out, [method(put)|Options]).
 183
 184%!  http_patch(+URL, +Data, -Reply, +Options)
 185%
 186%   Issue an HTTP =PATCH=  request.  Arguments   are  the  same as for
 187%   http_post/4.
 188%
 189%   @see Implemented on top of http_post/4.
 190
 191http_patch(URL, In, Out, Options) :-
 192    http_post(URL, In, Out, [method(patch)|Options]).
 193
 194%!  http_read_data(+Request, -Data, +Options) is det.
 195%
 196%   Read data from an HTTP connection   and  convert it according to
 197%   the supplied to(Format) option or  based on the =|Content-type|=
 198%   in the Request. The following options are supported:
 199%
 200%     * to(Format)
 201%       Convert data into Format.  Values are:
 202%       - stream(+WriteStream))
 203%         Append the content of the message to Stream
 204%       - atom
 205%         Return the reply as an atom
 206%       - string
 207%         Return the reply as a string
 208%       - codes
 209%         Return the reply as a list of codes
 210%     * form_data(AsForm)
 211%     * input_encoding(+Encoding)
 212%     * on_filename(:CallBack)
 213%       These options are implemented by the plugin
 214%       library(http/http_multipart_plugin) and apply to processing
 215%       =|multipart/form-data|= content.
 216%     * content_type(+Type)
 217%       Overrule the content-type that is part of Request as a
 218%       work-around for wrongly configured servers.
 219%
 220%   Without plugins, this predicate handles
 221%
 222%     * 'application/x-www-form-urlencoded'
 223%     Converts form-data into a list of `Name=Value` terms.
 224%     * 'application/x-prolog'
 225%     Converts data into a Prolog term.
 226%
 227%   @param  Request  is  a  parsed  HTTP   request  as  returned  by
 228%   http_read_request/2 or available from the  HTTP server's request
 229%   dispatcher. Request must contain a  term input(In) that provides
 230%   the input stream from the HTTP server.
 231
 232http_read_data(Fields, Data, QOptions) :-
 233    meta_options(is_meta, QOptions, Options),
 234    memberchk(input(In), Fields),
 235    (   http_read_data(In, Fields, Data, Options)
 236    ->  true
 237    ;   throw(error(failed(http_read_data), _))
 238    ).
 239
 240is_meta(on_filename).
 241
 242http_read_data(In, Fields, Data, Options) :-    % Transfer-encoding: chunked
 243    select(transfer_encoding(chunked), Fields, RestFields),
 244    !,
 245    setup_call_cleanup(
 246        http_chunked_open(In, DataStream, []),
 247        http_read_data(DataStream, RestFields, Data, Options),
 248        close(DataStream)).
 249http_read_data(In, Fields, Data, Options) :-
 250    option(to(X), Options),
 251    !,
 252    (   X = stream(Stream)
 253    ->  (   memberchk(content_length(Bytes), Fields)
 254        ->  copy_stream_data(In, Stream, Bytes)
 255        ;   copy_stream_data(In, Stream)
 256        )
 257    ;   must_be(oneof([atom,string,codes]), X),
 258        setup_call_cleanup(
 259            new_memory_file(MemFile),
 260            ( setup_call_cleanup(
 261                  open_memory_file(MemFile, write, Stream,
 262                                   [encoding(octet)]),
 263                  (   memberchk(content_length(Bytes), Fields)
 264                  ->  copy_stream_data(In, Stream, Bytes)
 265                  ;   copy_stream_data(In, Stream)
 266                  ),
 267                  close(Stream)),
 268              encoding(Fields, Encoding),
 269              memory_file_to(X, MemFile, Encoding, Data0)
 270            ),
 271            free_memory_file(MemFile)),
 272        Data = Data0
 273    ).
 274http_read_data(In, Fields, Data, _) :-
 275    option(content_type(ContentType), Fields),
 276    is_content_type(ContentType, 'application/x-www-form-urlencoded'),
 277    !,
 278    http_read_data(In, Fields, Codes, [to(string)]),
 279    uri_query_components(Codes, Data).
 280http_read_data(In, Fields, Data, Options) :-                    % call hook
 281    (   select_option(content_type(Type), Options, Options1)
 282    ->  delete(Fields, content_type(_), Fields1),
 283        http_convert_data(In, [content_type(Type)|Fields1], Data, Options1)
 284    ;   http_convert_data(In, Fields, Data, Options)
 285    ),
 286    !.
 287http_read_data(In, Fields, Data, Options) :-
 288    http_read_data(In, Fields, Data, [to(atom)|Options]).
 289
 290memory_file_to(atom, MemFile, Encoding, Data) :-
 291    memory_file_to_atom(MemFile, Data, Encoding).
 292memory_file_to(string, MemFile, Encoding, Data) :-
 293    memory_file_to_string(MemFile, Data, Encoding).
 294memory_file_to(codes, MemFile, Encoding, Data) :-
 295    memory_file_to_codes(MemFile, Data, Encoding).
 296
 297
 298encoding(Fields, utf8) :-
 299    memberchk(content_type(Type), Fields),
 300    (   sub_atom(Type, _, _, _, 'UTF-8')
 301    ->  true
 302    ;   sub_atom(Type, _, _, _, 'utf-8')
 303    ),
 304    !.
 305encoding(_, octet).
 306
 307is_content_type(ContentType, Check) :-
 308    sub_atom(ContentType, 0, Len, After, Check),
 309    (   After == 0
 310    ->  true
 311    ;   sub_atom(ContentType, Len, 1, _, ';')
 312    ).
 313
 314%!  http_convert_data(+In, +Fields, -Data, +Options) is semidet.
 315%
 316%   Multi-file hook to convert  a  HTTP   payload  according  to the
 317%   _Content-Type_ header. The  default   implementation  deals with
 318%   application/x-prolog.    The    HTTP      framework     provides
 319%   implementations  for  JSON  (library(http/http_json)),  HTML/XML
 320%   (library(http/http_sgml_plugin))
 321
 322http_convert_data(In, Fields, Data, Options) :-
 323    memberchk(content_type(Type), Fields),
 324    is_content_type(Type, 'application/x-prolog'),
 325    !,
 326    (   memberchk(content_length(Bytes), Fields)
 327    ->  setup_call_cleanup(
 328            ( stream_range_open(In, Range, [size(Bytes)]),
 329              set_stream(Range, encoding(utf8)),
 330              set_stream(Range, file_name('HTTP:DATA'))
 331            ),
 332            read_term(Range, Data, Options),
 333            close(Range))
 334    ;   set_stream(In, encoding(utf8)),
 335        read_term(In, Data, Options)
 336    ).
 337
 338%!  http_disconnect(+Connections) is det.
 339%
 340%   Close down some connections. Currently Connections must have the
 341%   value =all=, closing all connections.
 342%
 343%   @deprecated     New code should use http_close_keep_alive/1 from
 344%                   library(http/http_open).
 345
 346http_disconnect(all) :-
 347    http_close_keep_alive(_).
 348
 349%!  http:post_data_hook(+Term, +Out, +Options) is semidet.
 350%
 351%   Hook to extend the datatypes supported  by the post(Data) option
 352%   of   http_open/3.   The   default     implementation    supports
 353%   prolog(Term), sending a Prolog term as =|application/x-prolog|=.
 354
 355http:post_data_hook(prolog(Term), Out, HdrExtra) :-
 356    setup_call_cleanup(
 357        ( new_memory_file(MemFile),
 358          open_memory_file(MemFile, write, Handle)
 359        ),
 360        ( format(Handle,
 361                 'Content-Type: application/x-prolog; charset=UTF-8~n~n',
 362                 []),
 363          write_term(Handle, Term,
 364                     [ quoted(true),
 365                       ignore_ops(true),
 366                       fullstop(true),
 367                       nl(true)
 368                     ])
 369        ),
 370        close(Handle)),
 371    setup_call_cleanup(
 372        open_memory_file(MemFile, read, RdHandle,
 373                         [ free_on_close(true)
 374                         ]),
 375        http_post_data(cgi_stream(RdHandle), Out, HdrExtra),
 376        close(RdHandle)).