lib/etorrent-1.0/src/etorrent_http_uri.erl

   1 % ``The contents of this file are subject to the Erlang Public License,
   2 %% Version 1.1, (the "License"); you may not use this file except in
   3 %% compliance with the License. You should have received a copy of the
   4 %% Erlang Public License along with this software. If not, it can be
   5 %% retrieved via the world wide web at http://www.erlang.org/.
   6 %%
   7 %% Software distributed under the License is distributed on an "AS IS"
   8 %% basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See
   9 %% the License for the specific language governing rights and limitations
  10 %% under the License.
  11 %%
  12 %% The Initial Developer of the Original Code is Ericsson Utvecklings AB.
  13 %% Portions created by Ericsson are Copyright 1999, Ericsson Utvecklings
  14 %% AB. All Rights Reserved.''
  15 %%
  16 %%     $Id$
  17 %%
  18 %%--------------------------------------------------------------------
  19 %%  Note: This URI-parsing library is stolen from R12B-3 from the inets
  20 %%    application. I have bastardized it here for our purposes.
  21 %%--------------------------------------------------------------------
  22 -module(etorrent_http_uri).
  23
  24 -export([parse/1]).
  25
  26 %%%=========================================================================
  27 %%%  API
  28 %%%=========================================================================
  29 parse(AbsURI) ->
  30     case parse_scheme(AbsURI) of
  31         {error, Reason} ->
  32             {error, Reason};
  33         {Scheme, Rest} ->
  34             case (catch parse_uri_rest(Scheme, Rest)) of
  35                 {UserInfo, Host, Port, Path, Query} ->
  36                     {Scheme, UserInfo, Host, Port, Path, Query};
  37                 _  ->
  38                     {error, {malformed_url, AbsURI}}
  39             end
  40     end.
  41
  42 %%%========================================================================
  43 %%% Internal functions
  44 %%%========================================================================
  45 parse_scheme(AbsURI) ->
  46     case split_uri(AbsURI, ":", {error, no_scheme}, 1, 1) of
  47         {error, no_scheme} ->
  48             {error, no_scheme};
  49         {StrScheme, Rest} ->
  50             case list_to_atom(http_util:to_lower(StrScheme)) of
  51                 Scheme when Scheme == http; Scheme == https ->
  52                     {Scheme, Rest};
  53                 Scheme ->
  54                     {error, {not_supported_scheme, Scheme}}
  55             end
  56     end.
  57
  58 parse_uri_rest(Scheme, "//" ++ URIPart) ->
  59
  60     {Authority, PathQuery} =
  61         case split_uri(URIPart, "/", URIPart, 1, 0) of
  62             Split = {_, _} ->
  63                 Split;
  64             URIPart ->
  65                 case split_uri(URIPart, "\\?", URIPart, 1, 0) of
  66                     Split = {_, _} ->
  67                         Split;
  68                     URIPart ->
  69                         {URIPart,""}
  70                 end
  71         end,
  72
  73     {UserInfo, HostPort} = split_uri(Authority, "@", {"", Authority}, 1, 1),
  74     {Host, Port} = parse_host_port(Scheme, HostPort),
  75     {Path, Query} = parse_path_query(PathQuery),
  76     {UserInfo, Host, Port, Path, Query}.
  77
  78
  79 parse_path_query(PathQuery) ->
  80     {Path, Query} =  split_uri(PathQuery, "\\?", {PathQuery, ""}, 1, 0),
  81     {path(Path), Query}.
  82
  83
  84 parse_host_port(Scheme,"[" ++ HostPort) -> %ipv6
  85     DefaultPort = default_port(Scheme),
  86     {Host, ColonPort} = split_uri(HostPort, "\\]", {HostPort, ""}, 1, 1),
  87     {_, Port} = split_uri(ColonPort, ":", {"", DefaultPort}, 0, 1),
  88     {Host, int_port(Port)};
  89
  90 parse_host_port(Scheme, HostPort) ->
  91     DefaultPort = default_port(Scheme),
  92     {Host, Port} = split_uri(HostPort, ":", {HostPort, DefaultPort}, 1, 1),
  93     {Host, int_port(Port)}.
  94
  95 split_uri(UriPart, SplitChar, NoMatchResult, SkipLeft, SkipRight) ->
  96     case regexp:first_match(UriPart, SplitChar) of
  97         {match, Match, _} ->
  98             {string:substr(UriPart, 1, Match - SkipLeft),
  99              string:substr(UriPart, Match + SkipRight, length(UriPart))};
 100         nomatch ->
 101             NoMatchResult
 102     end.
 103
 104 default_port(http) ->
 105     80;
 106 default_port(https) ->
 107     443.
 108
 109 int_port(Port) when is_integer(Port) ->
 110     Port;
 111 int_port(Port) when is_list(Port) ->
 112     list_to_integer(Port).
 113
 114 path("") ->
 115     "/";
 116 path(Path) ->
 117     Path.