Refactoring of ejabberd_http_fileserver.erl

- not a gen_server anymore. Should be way faster now (no more message passing between processes)
- configuration stored in mochiglobal
- support for etag
- support for gzip compression: 
	- static (if a foo.gz is in the same dir as requested foo, it will be served)
	- always (will always gzip, will use static is available)
	- false, don't gzip
- logfile is now in another module.

src/mochiglobal.erl

@@ -0,0 +1,107 @@
+%% @author Bob Ippolito <bob@mochimedia.com>
+%% @copyright 2010 Mochi Media, Inc.
+%% @doc Abuse module constant pools as a "read-only shared heap" (since erts 5.6)
+%%      <a href="http://www.erlang.org/pipermail/erlang-questions/2009-March/042503.html">[1]</a>.
+-module(mochiglobal).
+-author("Bob Ippolito <bob@mochimedia.com>").
+-export([get/1, get/2, put/2, delete/1]).
+
+-spec get(atom()) -> any() | undefined.
+%% @equiv get(K, undefined)
+get(K) ->
+    get(K, undefined).
+
+-spec get(atom(), T) -> any() | T.
+%% @doc Get the term for K or return Default.
+get(K, Default) ->
+    get(K, Default, key_to_module(K)).
+
+get(_K, Default, Mod) ->
+    try Mod:term()
+    catch error:undef ->
+            Default
+    end.
+
+-spec put(atom(), any()) -> ok.
+%% @doc Store term V at K, replaces an existing term if present.
+put(K, V) ->
+    put(K, V, key_to_module(K)).
+
+put(_K, V, Mod) ->
+    Bin = compile(Mod, V),
+    code:purge(Mod),
+    code:load_binary(Mod, atom_to_list(Mod) ++ ".erl", Bin),
+    ok.
+
+-spec delete(atom()) -> boolean().
+%% @doc Delete term stored at K, no-op if non-existent.
+delete(K) ->
+    delete(K, key_to_module(K)).
+
+delete(_K, Mod) ->
+    code:purge(Mod),
+    code:delete(Mod).
+
+-spec key_to_module(atom()) -> atom().
+key_to_module(K) ->
+    list_to_atom("mochiglobal:" ++ atom_to_list(K)).
+
+-spec compile(atom(), any()) -> binary().
+compile(Module, T) ->
+    {ok, Module, Bin} = compile:forms(forms(Module, T),
+                                      [verbose, report_errors]),
+    Bin.
+
+-spec forms(atom(), any()) -> [erl_syntax:syntaxTree()].
+forms(Module, T) ->
+    [erl_syntax:revert(X) || X <- term_to_abstract(Module, term, T)].
+
+-spec term_to_abstract(atom(), atom(), any()) -> [erl_syntax:syntaxTree()].
+term_to_abstract(Module, Getter, T) ->
+    [%% -module(Module).
+     erl_syntax:attribute(
+       erl_syntax:atom(module),
+       [erl_syntax:atom(Module)]),
+     %% -export([Getter/0]).
+     erl_syntax:attribute(
+       erl_syntax:atom(export),
+       [erl_syntax:list(
+         [erl_syntax:arity_qualifier(
+            erl_syntax:atom(Getter),
+            erl_syntax:integer(0))])]),
+     %% Getter() -> T.
+     erl_syntax:function(
+       erl_syntax:atom(Getter),
+       [erl_syntax:clause([], none, [erl_syntax:abstract(T)])])].
+
+%%
+%% Tests
+%%
+-include_lib("eunit/include/eunit.hrl").
+-ifdef(TEST).
+get_put_delete_test() ->
+    K = '$$test$$mochiglobal',
+    delete(K),
+    ?assertEqual(
+       bar,
+       get(K, bar)),
+    try
+        ?MODULE:put(K, baz),
+        ?assertEqual(
+           baz,
+           get(K, bar)),
+        ?MODULE:put(K, wibble),
+        ?assertEqual(
+           wibble,
+           ?MODULE:get(K))
+    after
+        delete(K)
+    end,
+    ?assertEqual(
+       bar,
+       get(K, bar)),
+    ?assertEqual(
+       undefined,
+       ?MODULE:get(K)),
+    ok.
+-endif.

src/web/mod_http_fileserver.erl

@@ -26,45 +26,20 @@
 
 -module(mod_http_fileserver).
 -author('mmirra@process-one.net').
-
+-author('ecestari@process-one.net').
 -behaviour(gen_mod).
--behaviour(gen_server).
 
 %% gen_mod callbacks
 -export([start/2, stop/1]).
 
-%% API
--export([start_link/2]).
-
-%% gen_server callbacks
--export([init/1, handle_call/3, handle_cast/2, handle_info/2,
-	 terminate/2, code_change/3]).
-
 %% request_handlers callbacks
 -export([process/2]).
 
-%% ejabberd_hooks callbacks
--export([reopen_log/1]).
-
 -include("ejabberd.hrl").
 -include("jlib.hrl").
 -include_lib("kernel/include/file.hrl").
 
-%%-include("ejabberd_http.hrl").
+-include("ejabberd_http.hrl").
-%% TODO: When ejabberd-modules SVN gets the new ejabberd_http.hrl, delete this code:
--record(request, {method,
-		  path,
-		  q = [],
-		  us,
-		  auth,
-		  lang = "",
-		  data = "",
-		  ip,
-		  host, % string()
-		  port, % integer()
-		  tp, % transfer protocol = http | https
-		  headers
-		 }).
 
 -ifdef(SSL40).
 -define(STRING2LOWER, string).
@@ -76,11 +51,6 @@
 -endif.
 -endif.
 
--record(state, {host, docroot, accesslog, accesslogfd, directory_indices,
-                custom_headers, default_content_type, content_types = []}).
-
--define(PROCNAME, ejabberd_mod_http_fileserver).
-
 %% Response is {DataSize, Code, [{HeaderKey, HeaderValue}], Data}
 -define(HTTP_ERR_FILE_NOT_FOUND, {-1, 404, [], "Not found"}).
 -define(HTTP_ERR_FORBIDDEN,      {-1, 403, [], "Forbidden"}).
@@ -91,7 +61,7 @@
                                 {".html", "text/html"},
                                 {".jar",  "application/java-archive"},
                                 {".jpeg", "image/jpeg"},
-				{".jpg",  "image/jpeg"},
+				                {".jpg",  "image/jpeg"},
                                 {".js",   "text/javascript"},
                                 {".png",  "image/png"},
                                 {".txt",  "text/plain"},
@@ -101,81 +71,69 @@
 
 -compile(export_all).
 
-%%====================================================================
-%% gen_mod callbacks
-%%====================================================================
 
 start(Host, Opts) ->
-    Proc = get_proc_name(Host),
-    ChildSpec =
-	{Proc,
-	 {?MODULE, start_link, [Host, Opts]},
-	 transient, % if process crashes abruptly, it gets restarted
-	 1000,
-	 worker,
-	 [?MODULE]},
-    supervisor:start_child(ejabberd_sup, ChildSpec).
-
-stop(Host) ->
-    Proc = get_proc_name(Host),
-    gen_server:call(Proc, stop),
-    supervisor:terminate_child(ejabberd_sup, Proc),
-    supervisor:delete_child(ejabberd_sup, Proc).
-
-%%====================================================================
-%% API
-%%====================================================================
-%%--------------------------------------------------------------------
-%% Function: start_link() -> {ok,Pid} | ignore | {error,Error}
-%% Description: Starts the server
-%%--------------------------------------------------------------------
-start_link(Host, Opts) ->
-    Proc = get_proc_name(Host),
-    gen_server:start_link({local, Proc}, ?MODULE, [Host, Opts], []).
-
-%%====================================================================
-%% gen_server callbacks
-%%====================================================================
-%%--------------------------------------------------------------------
-%% Function: init(Args) -> {ok, State} |
-%%                         {ok, State, Timeout} |
-%%                         ignore               |
-%%                         {stop, Reason}
-%% Description: Initiates the server
-%%--------------------------------------------------------------------
-init([Host, Opts]) ->
-    try initialize(Host, Opts) of
-	{DocRoot, AccessLog, AccessLogFD, DirectoryIndices,
-         CustomHeaders, DefaultContentType, ContentTypes} ->
-	    {ok, #state{host = Host,
-			accesslog = AccessLog,
-			accesslogfd = AccessLogFD,
-			docroot = DocRoot,
-                        directory_indices = DirectoryIndices,
-                        custom_headers = CustomHeaders,
-                        default_content_type = DefaultContentType,
-                        content_types = ContentTypes}}
-    catch
-	throw:Reason ->
-	    {stop, Reason}
-    end.
-
-initialize(Host, Opts) ->
     DocRoot = gen_mod:get_opt(docroot, Opts, undefined),
+    set_default_host(Host, Opts),
+    conf_store(Host, docroot, DocRoot),
     check_docroot_defined(DocRoot, Host),
     DRInfo = check_docroot_exists(DocRoot),
     check_docroot_is_dir(DRInfo, DocRoot),
     check_docroot_is_readable(DRInfo, DocRoot),
     AccessLog = gen_mod:get_opt(accesslog, Opts, undefined),
-    AccessLogFD = try_open_log(AccessLog, Host),
+    start_log(Host, AccessLog),
     DirectoryIndices = gen_mod:get_opt(directory_indices, Opts, []),
+    conf_store(Host, directory_indices, DirectoryIndices),
+    ServeStaticGzip = gen_mod:get_opt(serve_gzip, Opts, false),
+    conf_store(Host, serve_gzip, ServeStaticGzip),
     CustomHeaders = gen_mod:get_opt(custom_headers, Opts, []),
+    conf_store(Host, custom_headers, CustomHeaders),
     DefaultContentType = gen_mod:get_opt(default_content_type, Opts,
                                          ?DEFAULT_CONTENT_TYPE),
+    conf_store(Host, default_content_type, DefaultContentType),
     ContentTypes = build_list_content_types(gen_mod:get_opt(content_types, Opts, []), ?DEFAULT_CONTENT_TYPES),
-    ?INFO_MSG("initialize: ~n ~p", [ContentTypes]),%+++
+    conf_store(Host, content_types, ContentTypes),
-    {DocRoot, AccessLog, AccessLogFD, DirectoryIndices,
+    ?INFO_MSG("initialize: ~n ~p", [ContentTypes]),
-     CustomHeaders, DefaultContentType, ContentTypes}.
+    ok.
+
+% Defines host that will answer request if hostname is not recognized.
+% The first configured host will be used.
+set_default_host(Host, _Opts)->
+    case mochiglobal:get(http_default_host) of
+        undefined ->
+            ?DEBUG("Setting default host to ~p", [Host]),
+            mochiglobal:put(http_default_host, Host);
+        _ ->
+            ok
+    end.
+    
+conf_store(Host, Key, Value)->
+    R = case mochiglobal:get(Key) of
+        undefined -> [{Host, Value}];
+        A ->
+            case lists:keymember(Host, 1, A) of
+                true -> lists:keyreplace(Host, 1, A,{Host, Value});
+                false -> [{Host, Value}|A]
+            end
+    end,
+    mochiglobal:put(Key, R).
+    
+conf_get(Host, Key, Default) ->
+    case mochiglobal:get(Key) of
+        undefined-> Default;
+        A ->
+            case lists:keyfind(Host, 1, A) of
+                {Host, Val} -> Val;
+                false -> 
+                    case mochiglobal:get(http_default_host) of
+                        Host -> % stop recursion here
+                            Default;
+                        DefaultHost ->
+                            conf_get(DefaultHost, Key, Default)
+                    end
+            end
+    end.
+
 
 %% @spec (AdminCTs::[CT], Default::[CT]) -> [CT]
 %% where CT = {Extension::string(), Value}
@@ -183,6 +141,12 @@ initialize(Host, Opts) ->
 %% @doc Return a unified list without duplicates.
 %% Elements of AdminCTs have more priority.
 %% If a CT is declared as 'undefined', then it is not included in the result.
+
+start_log(_Host, undefined)->
+    ok;
+start_log(Host, FileName) ->
+    mod_http_fileserver_log:start(Host, FileName).
+    
 build_list_content_types(AdminCTsUnsorted, DefaultCTsUnsorted) ->
     AdminCTs = lists:ukeysort(1, AdminCTsUnsorted),
     DefaultCTs = lists:ukeysort(1, DefaultCTsUnsorted),
@@ -214,78 +178,9 @@ check_docroot_is_readable(DRInfo, DocRoot) ->
 	_ -> throw({docroot_not_readable, DocRoot})
     end.
     
-try_open_log(undefined, _Host) ->
+stop(_Host) ->
-    undefined;
-try_open_log(FN, Host) ->
-    FD = try open_log(FN) of
-	     FD1 -> FD1
-	 catch
-	     throw:{cannot_open_accesslog, FN, Reason} ->
-		 ?ERROR_MSG("Cannot open access log file: ~p~nReason: ~p", [FN, Reason]),
-		 undefined
-	 end,
-    ejabberd_hooks:add(reopen_log_hook, Host, ?MODULE, reopen_log, 50),
-    FD.
-
-%%--------------------------------------------------------------------
-%% Function: handle_call(Request, From, State) -> {reply, Reply, State} |
-%%                                      {reply, Reply, State, Timeout} |
-%%                                      {noreply, State} |
-%%                                      {noreply, State, Timeout} |
-%%                                      {stop, Reason, Reply, State} |
-%%                                      {stop, Reason, State}
-%% Description: Handling call messages
-%%--------------------------------------------------------------------
-handle_call({serve, LocalPath}, _From, State) ->
-    Reply = serve(LocalPath, State#state.docroot, State#state.directory_indices,
-		  State#state.custom_headers,
-                  State#state.default_content_type, State#state.content_types),
-    {reply, Reply, State};
-handle_call(_Request, _From, State) ->
-    {reply, ok, State}.
-
-%%--------------------------------------------------------------------
-%% Function: handle_cast(Msg, State) -> {noreply, State} |
-%%                                      {noreply, State, Timeout} |
-%%                                      {stop, Reason, State}
-%% Description: Handling cast messages
-%%--------------------------------------------------------------------
-handle_cast({add_to_log, FileSize, Code, Request}, State) ->
-    add_to_log(State#state.accesslogfd, FileSize, Code, Request),
-    {noreply, State};
-handle_cast(reopen_log, State) ->
-    FD2 = reopen_log(State#state.accesslog, State#state.accesslogfd),
-    {noreply, State#state{accesslogfd = FD2}};
-handle_cast(_Msg, State) ->
-    {noreply, State}.
-
-%%--------------------------------------------------------------------
-%% Function: handle_info(Info, State) -> {noreply, State} |
-%%                                       {noreply, State, Timeout} |
-%%                                       {stop, Reason, State}
-%% Description: Handling all non call/cast messages
-%%--------------------------------------------------------------------
-handle_info(_Info, State) ->
-    {noreply, State}.
-
-%%--------------------------------------------------------------------
-%% Function: terminate(Reason, State) -> void()
-%% Description: This function is called by a gen_server when it is about to
-%% terminate. It should be the opposite of Module:init/1 and do any necessary
-%% cleaning up. When it returns, the gen_server terminates with Reason.
-%% The return value is ignored.
-%%--------------------------------------------------------------------
-terminate(_Reason, State) ->
-    close_log(State#state.accesslogfd),
-    ejabberd_hooks:delete(reopen_log_hook, State#state.host, ?MODULE, reopen_log, 50),
     ok.
 
-%%--------------------------------------------------------------------
-%% Func: code_change(OldVsn, State, Extra) -> {ok, NewState}
-%% Description: Convert process state when code is changed
-%%--------------------------------------------------------------------
-code_change(_OldVsn, State, _Extra) ->
-    {ok, State}.
 
 %%====================================================================
 %% request_handlers callbacks
@@ -295,114 +190,122 @@ code_change(_OldVsn, State, _Extra) ->
 %% @doc Handle an HTTP request.
 %% LocalPath is the part of the requested URL path that is "local to the module".
 %% Returns the page to be sent back to the client and/or HTTP status code.
+
 process(LocalPath, Request) ->
     ?DEBUG("Requested ~p", [LocalPath]),
-    try gen_server:call(get_proc_name(Request#request.host), {serve, LocalPath}) of
+    Host = Request#request.host,
-	{FileSize, Code, Headers, Contents} ->
+    ClientHeaders = Request#request.headers,
-	    add_to_log(FileSize, Code, Request),
+    DirectoryIndices = conf_get(Host, directory_indices, undefined),
-	    {Code, Headers, Contents}
+    CustomHeaders = conf_get(Host, custom_headers, undefined),
-    catch
+    DefaultContentType = conf_get(Host, default_content_type, undefined),
-	exit:{noproc, _} -> 
+    ContentTypes = conf_get(Host, content_types, undefined),
-	    ?ERROR_MSG("Received an HTTP request with Host ~p, but couldn't find the related "
+    Static = conf_get(Host, serve_gzip, undefined),
-		       "ejabberd virtual host", [Request#request.host]),
+    DocRoot = conf_get(Host, docroot, undefined),
-	    ejabberd_web:error(not_found)
-    end.
-
-serve(LocalPath, DocRoot, DirectoryIndices, CustomHeaders, DefaultContentType, ContentTypes) ->
     FileName = filename:join(filename:split(DocRoot) ++ LocalPath),
-    case file:read_file_info(FileName) of
+    {FileSize, Code, Headers, Contents} = case file:read_file_info(FileName) of
         {error, enoent}                    -> ?HTTP_ERR_FILE_NOT_FOUND;
         {error, eacces}                    -> ?HTTP_ERR_FORBIDDEN;
         {ok, #file_info{type = directory}} -> serve_index(FileName,
                                                           DirectoryIndices,
                                                           CustomHeaders,
                                                           DefaultContentType,
-                                                          ContentTypes);
+                                                          ContentTypes, Static);
-        {ok, FileInfo}                     -> serve_file(FileInfo, FileName,
+        {ok, FileInfo}                     -> 
-                                                         CustomHeaders,
+            case should_serve(FileInfo, ClientHeaders) of
-                                                         DefaultContentType,
+                true ->serve_file(FileInfo, FileName,
-                                                         ContentTypes)
+                                  CustomHeaders,
+                                  DefaultContentType,
+                                  ContentTypes, Static);
+                false ->
+                    {0, 304, [], []}
+            end 
+    end,
+    mod_http_fileserver_log:add_to_log(Host,FileSize, Code, Request),
+    {Code, Headers, Contents}.
+
+should_serve(FileInfo, Headers) ->
+    lists:foldl(fun({Header, Fun}, Acc)->
+        case lists:keyfind(Header, 1, Headers) of
+            {_, Val} ->
+                Fun(FileInfo,Val);
+            _O ->
+                Acc
+        end
+    end, true, [{'If-None-Match',fun etag/2}
+                ]).
+etag(FileInfo, Etag)->
+    case httpd_util:create_etag(FileInfo) of
+        Etag ->
+            false;
+        _ ->
+            true
     end.
+modified(FileInfo, LastModified)->
+    AfterDate = calendar:datetime_to_gregorian_seconds(
+                    httpd_util:convert_request_date(LastModified)),
+    Mtime = calendar:datetime_to_gregorian_seconds(FileInfo#file_info.mtime),
+    ?DEBUG("Modified : ~p > ~p (serving: ~p)", [Mtime, AfterDate,Mtime > AfterDate]),
+    Mtime > AfterDate.
         
 %% Troll through the directory indices attempting to find one which
 %% works, if none can be found, return a 404.
-serve_index(_FileName, [], _CH, _DefaultContentType, _ContentTypes) ->
+serve_index(_FileName, [], _CH, _DefaultContentType, _ContentTypes, _Static) ->
     ?HTTP_ERR_FILE_NOT_FOUND;
-serve_index(FileName, [Index | T], CH, DefaultContentType, ContentTypes) ->
+serve_index(FileName, [Index | T], CH, DefaultContentType, ContentTypes, Static) ->
     IndexFileName = filename:join([FileName] ++ [Index]),
     case file:read_file_info(IndexFileName) of
-        {error, _Error}                    -> serve_index(FileName, T, CH, DefaultContentType, ContentTypes);
+        {error, _Error}                    -> serve_index(FileName, T, CH, DefaultContentType, ContentTypes, Static);
-        {ok, #file_info{type = directory}} -> serve_index(FileName, T, CH, DefaultContentType, ContentTypes);
+        {ok, #file_info{type = directory}} -> serve_index(FileName, T, CH, DefaultContentType, ContentTypes, Static);
-        {ok, FileInfo}                     -> serve_file(FileInfo, IndexFileName, CH, DefaultContentType, ContentTypes)
+        {ok, FileInfo}                     -> serve_file(FileInfo, IndexFileName, CH, DefaultContentType, ContentTypes, Static)
     end.
 
 %% Assume the file exists if we got this far and attempt to read it in
 %% and serve it up.
-serve_file(FileInfo, FileName, CustomHeaders, DefaultContentType, ContentTypes) ->
+     
+serve_file(FileInfo, FileName, CustomHeaders, DefaultContentType, ContentTypes, false) ->
     ?DEBUG("Delivering: ~s", [FileName]),
-    {ok, FileContents} = file:read_file(FileName),
     ContentType = content_type(FileName, DefaultContentType, ContentTypes),
+    {ok, FileContents} = file:read_file(FileName),
     {FileInfo#file_info.size,
      200, [{"Server", "ejabberd"},
            {"Last-Modified", last_modified(FileInfo)},
            {"Content-Type", ContentType} | CustomHeaders],
-     FileContents}.
+     FileContents};
      
-%%----------------------------------------------------------------------
+serve_file(FileInfo, FileName, CustomHeaders, DefaultContentType, ContentTypes, Gzip) ->
-%% Log file
+    ?DEBUG("Delivering: ~s", [FileName]),
-%%----------------------------------------------------------------------
+    ContentType = content_type(FileName, DefaultContentType, ContentTypes),
-
+    CompressedFileName = FileName ++ ".gz",
-open_log(FN) ->
+    case file:read_file_info(CompressedFileName) of
-    case file:open(FN, [append]) of
+        {ok, FileInfoCompressed} -> %Found compressed
-	{ok, FD} ->
+            ?INFO_MSG("Found compressed: ~s", [FileName]),
-	    FD;
+            {ok, FileContents} = file:read_file(CompressedFileName),
-	{error, Reason} ->
+            {FileInfoCompressed#file_info.size,
-	    throw({cannot_open_accesslog, FN, Reason})
+             200, [{"Server", "ejabberd"},
+                   {"Last-Modified", last_modified(FileInfoCompressed)},
+                   {"Content-Type", ContentType},
+                   {"Etag", httpd_util:create_etag(FileInfoCompressed)},
+                   {"Content-Encoding", "gzip"} | CustomHeaders],
+             FileContents};
+        {error, _} ->
+            {FileContents, Size} = case Gzip of
+                static ->
+                    {ok, Content} = file:read_file(FileName),
+                    {Content, FileInfo#file_info.size};
+                always ->
+                    {ok, Content} = file:read_file(FileName),
+                    Compressed = zlib:gzip(Content),
+                    {Compressed, size(Compressed)}
+                end,
+            {Size,
+             200, [{"Server", "ejabberd"},
+                   {"Last-Modified", last_modified(FileInfo)},
+                   {"Etag", httpd_util:create_etag(FileInfo)},
+                   {"Content-Type", ContentType},
+                   {"Content-Encoding", "gzip"} | CustomHeaders],
+             FileContents}
     end.
 
-close_log(FD) ->
-    file:close(FD).
-
-reopen_log(undefined, undefined) ->
-    ok;
-reopen_log(FN, FD) ->
-    close_log(FD),
-    open_log(FN).
-
-reopen_log(Host) ->
-    gen_server:cast(get_proc_name(Host), reopen_log).
-
-add_to_log(FileSize, Code, Request) ->
-    gen_server:cast(get_proc_name(Request#request.host),
-		    {add_to_log, FileSize, Code, Request}).
-
-add_to_log(undefined, _FileSize, _Code, _Request) ->
-    ok;
-add_to_log(File, FileSize, Code, Request) ->
-    {{Year, Month, Day}, {Hour, Minute, Second}} = calendar:local_time(),
-    IP = ip_to_string(element(1, Request#request.ip)),
-    Path = join(Request#request.path, "/"),
-    Query = case join(lists:map(fun(E) -> lists:concat([element(1, E), "=", element(2, E)]) end,
-				Request#request.q), "&") of
-		[] ->
-		    "";
-		String ->
-		    [$? | String]
-	    end,
-    UserAgent = find_header('User-Agent', Request#request.headers, "-"),
-    Referer = find_header('Referer', Request#request.headers, "-"),
-    %% Pseudo Combined Apache log format:
-    %% 127.0.0.1 - - [28/Mar/2007:18:41:55 +0200] "GET / HTTP/1.1" 302 303 "-" "tsung"
-    %% TODO some fields are harcoded/missing:
-    %%   The date/time integers should have always 2 digits. For example day "7" should be "07"
-    %%   Month should be 3*letter, not integer 1..12
-    %%   Missing time zone = (`+' | `-') 4*digit
-    %%   Missing protocol version: HTTP/1.1
-    %% For reference: http://httpd.apache.org/docs/2.2/logs.html
-    io:format(File, "~s - - [~p/~p/~p:~p:~p:~p] \"~s /~s~s\" ~p ~p ~p ~p~n",
-	      [IP, Day, Month, Year, Hour, Minute, Second, Request#request.method, Path, Query, Code,
-               FileSize, Referer, UserAgent]).
-
 find_header(Header, Headers, Default) ->
     case lists:keysearch(Header, 1, Headers) of
         {value, {_, Value}} -> Value;
@@ -413,15 +316,6 @@ find_header(Header, Headers, Default) ->
 %% Utilities
 %%----------------------------------------------------------------------
 
-get_proc_name(Host) -> gen_mod:get_module_proc(Host, ?PROCNAME).
-
-join([], _) ->
-    "";
-join([E], _) ->
-    E;
-join([H | T], Separator) ->
-    lists:foldl(fun(E, Acc) -> lists:concat([Acc, Separator, E]) end, H, T).
-
 content_type(Filename, DefaultContentType, ContentTypes) ->
     Extension = ?STRING2LOWER:to_lower(filename:extension(Filename)),
     case lists:keysearch(Extension, 1, ContentTypes) of
@@ -432,11 +326,3 @@ content_type(Filename, DefaultContentType, ContentTypes) ->
 last_modified(FileInfo) ->
     Then = FileInfo#file_info.mtime,
     httpd_util:rfc1123_date(Then).
-
-%% Convert IP address tuple to string representation. Accepts either
-%% IPv4 or IPv6 address tuples.
-ip_to_string(Address) when size(Address) == 4 ->
-    join(tuple_to_list(Address), ".");
-ip_to_string(Address) when size(Address) == 8 ->
-    Parts = lists:map(fun (Int) -> io_lib:format("~.16B", [Int]) end, tuple_to_list(Address)),
-    ?STRING2LOWER:to_lower(lists:flatten(join(Parts, ":"))).

src/web/mod_http_fileserver_log.erl

@@ -0,0 +1,167 @@
+-module (mod_http_fileserver_log).
+
+-behaviour (gen_server).
+
+-export([init/1, handle_call/3, handle_cast/2, handle_info/2, terminate/2, code_change/3]).
+
+-export ([start_link/2,start/2, stop/1, add_to_log/4,reopen_log/1]).
+
+-include("ejabberd.hrl").
+-include("jlib.hrl").
+-include("ejabberd_http.hrl").
+-include_lib("kernel/include/file.hrl").
+
+-define(PROCNAME, ejabberd_mod_http_fileserver_log).
+
+-record(state, {host,accesslog, accesslogfd}).
+%% Public API
+
+start(Host, Filename) ->
+    Proc =gen_mod:get_module_proc(Host, ?PROCNAME),
+    ChildSpec =
+    {Proc,
+     {?MODULE, start_link, [Host, Filename]},
+     transient, % if process crashes abruptly, it gets restarted
+     1000,
+     worker,
+     [?MODULE]},
+    supervisor:start_child(ejabberd_sup, ChildSpec).
+
+stop(Host) ->
+    Proc = gen_mod:get_module_proc(Host, ?PROCNAME),
+    gen_server:call(Proc, stop),
+    supervisor:terminate_child(ejabberd_sup, Proc),
+    supervisor:delete_child(ejabberd_sup, Proc).
+
+start_link(Host, Filename) ->
+    Proc = gen_mod:get_module_proc(Host, ?PROCNAME),
+    gen_server:start_link({local, Proc}, ?MODULE, [Host, Filename], []).
+
+add_to_log(Host,FileSize, Code, Request) ->
+    gen_server:cast(gen_mod:get_module_proc(Host, ?PROCNAME),
+		    {add_to_log, FileSize, Code, Request}).
+		    
+reopen_log(Host) ->
+    gen_server:cast(gen_mod:get_module_proc(Host, ?PROCNAME), reopen_log).
+
+%% Server implementation, a.k.a.: callbacks
+
+init([Host, Filename]) ->
+    try try_open_log(Filename, Host) of
+	AccessLogFD ->
+	    ?DEBUG("File opened !", []),
+	    {ok, #state{host = Host,
+			accesslog = Filename,
+			accesslogfd = AccessLogFD}}
+    catch
+	throw:Reason ->
+	    {stop, Reason}
+    end.
+
+try_open_log(FN, Host) ->
+    FD = try open_log(FN) of
+	     FD1 -> FD1
+	 catch
+	     throw:{cannot_open_accesslog, FN, Reason} ->
+		 ?ERROR_MSG("Cannot open access log file: ~p~nReason: ~p", [FN, Reason]),
+		 undefined
+	 end,
+    %HostB = list_to_binary(Host),
+    ejabberd_hooks:add(reopen_log_hook, Host, ?MODULE, reopen_log, 50),
+    FD.
+     
+handle_call(_Request, _From, State) ->
+  {reply, ok, State}.
+
+handle_cast({add_to_log, FileSize, Code, Request}, State) ->
+    add_to_log2(State#state.accesslogfd, FileSize, Code, Request),
+    {noreply, State};
+handle_cast(reopen_log, State) ->
+    FD2 = reopen_log(State#state.accesslog, State#state.accesslogfd),
+    {noreply, State#state{accesslogfd = FD2}};
+handle_cast(_Msg, State) ->
+    {noreply, State}.
+
+handle_info(_Info, State) ->
+    {noreply, State}.
+
+terminate(_Reason, State) ->
+    close_log(State#state.accesslogfd),
+    ejabberd_hooks:delete(reopen_log_hook, State#state.host, ?MODULE, reopen_log, 50),
+    ok.
+
+code_change(_OldVsn, State, _Extra) ->
+  {ok, State}.
+  
+  
+%%----------------------------------------------------------------------
+%% Log file
+%%----------------------------------------------------------------------
+
+open_log(FN) ->
+    case file:open(FN, [append]) of
+	{ok, FD} ->
+	    FD;
+	{error, Reason} ->
+	    throw({cannot_open_accesslog, FN, Reason})
+    end.
+
+close_log(FD) ->
+    file:close(FD).
+
+reopen_log(undefined, undefined) ->
+    ok;
+reopen_log(FN, FD) ->
+    ?DEBUG("reopening logs", []),
+    close_log(FD),
+    open_log(FN).
+
+
+
+add_to_log2(undefined, _FileSize, _Code, _Request) ->
+    ok;
+add_to_log2(File, FileSize, Code, Request) ->
+    {{Year, Month, Day}, {Hour, Minute, Second}} = calendar:local_time(),
+    IP = ip_to_string(element(1, Request#request.ip)),
+    Path = join(Request#request.path, "/"),
+    Query = case join(lists:map(fun(E) -> lists:concat([element(1, E), "=", element(2, E)]) end,
+				Request#request.q), "&") of
+		[] ->
+		    "";
+		String ->
+		    [$? | String]
+	    end,
+    UserAgent = find_header('User-Agent', Request#request.headers, "-"),
+    Referer = find_header('Referer', Request#request.headers, "-"),
+    %% Pseudo Combined Apache log format:
+    %% 127.0.0.1 - - [28/Mar/2007:18:41:55 +0200] "GET / HTTP/1.1" 302 303 "-" "tsung"
+    %% TODO some fields are harcoded/missing:
+    %%   The date/time integers should have always 2 digits. For example day "7" should be "07"
+    %%   Month should be 3*letter, not integer 1..12
+    %%   Missing time zone = (`+' | `-') 4*digit
+    %%   Missing protocol version: HTTP/1.1
+    %% For reference: http://httpd.apache.org/docs/2.2/logs.html
+    io:format(File, "~s - - [~p/~p/~p:~p:~p:~p] \"~s /~s~s\" ~p ~p ~p ~p~n",
+	      [IP, Day, Month, Year, Hour, Minute, Second, Request#request.method, Path, Query, Code,
+               FileSize, Referer, UserAgent]).
+               
+find_header(Header, Headers, Default) ->
+    case lists:keysearch(Header, 1, Headers) of
+        {value, {_, Value}} -> Value;
+        false               -> Default
+    end.
+                   
+join([], _) ->
+    "";
+join([E], _) ->
+    E;
+join([H | T], Separator) ->
+    lists:foldl(fun(E, Acc) -> lists:concat([Acc, Separator, E]) end, H, T).
+    
+%% Convert IP address tuple to string representation. Accepts either
+%% IPv4 or IPv6 address tuples.
+ip_to_string(Address) when size(Address) == 4 ->
+    join(tuple_to_list(Address), ".");
+ip_to_string(Address) when size(Address) == 8 ->
+    Parts = lists:map(fun (Int) -> io_lib:format("~.16B", [Int]) end, tuple_to_list(Address)),
+    string:to_lower(lists:flatten(join(Parts, ":"))).