diff options
| -rw-r--r-- | bin/couchdb.tpl.in | 2 | ||||
| -rw-r--r-- | src/mochiweb/Makefile.am | 8 | ||||
| -rw-r--r-- | src/mochiweb/mochifmt.erl | 426 | ||||
| -rw-r--r-- | src/mochiweb/mochifmt_records.erl | 30 | ||||
| -rw-r--r-- | src/mochiweb/mochifmt_std.erl | 23 | ||||
| -rw-r--r-- | src/mochiweb/mochiweb.erl | 1 | ||||
| -rw-r--r-- | src/mochiweb/mochiweb_html.erl | 89 | ||||
| -rw-r--r-- | src/mochiweb/mochiweb_multipart.erl | 31 | ||||
| -rw-r--r-- | src/mochiweb/mochiweb_request.erl | 2 | ||||
| -rw-r--r-- | src/mochiweb/mochiweb_socket_server.erl | 3 |
10 files changed, 589 insertions, 26 deletions
diff --git a/bin/couchdb.tpl.in b/bin/couchdb.tpl.in index 4002b006..0a7da952 100644 --- a/bin/couchdb.tpl.in +++ b/bin/couchdb.tpl.in @@ -223,7 +223,7 @@ start_couchdb () { command="`%ICU_CONFIG% --invoke` \ %ERL% $interactive_option -sasl errlog_type error \ -pa %erlanglibdir%/couch-%version%/ebin \ - %erlanglibdir%/mochiweb-r64/ebin \ + %erlanglibdir%/mochiweb-r73/ebin \ -eval \"application:load(inets)\" \ -eval \"application:load(crypto)\" \ -eval \"application:load(couch)\" \ diff --git a/src/mochiweb/Makefile.am b/src/mochiweb/Makefile.am index cbd1438a..a4021bb4 100644 --- a/src/mochiweb/Makefile.am +++ b/src/mochiweb/Makefile.am @@ -12,9 +12,12 @@ datarootdir = @prefix@/share -mochiwebebindir = $(erlanglibdir)/mochiweb-r64/ebin +mochiwebebindir = $(erlanglibdir)/mochiweb-r73/ebin mochiweb_file_collection = \ + mochifmt.erl \ + mochifmt_records.erl \ + mochifmt_std.erl \ mochihex.erl \ mochijson.erl \ mochijson2.erl \ @@ -39,6 +42,9 @@ mochiweb_file_collection = \ mochiwebebin_static_file = mochiweb.app mochiwebebin_make_generated_file_list = \ + mochifmt.beam \ + mochifmt_records.beam \ + mochifmt_std.beam \ mochihex.beam \ mochijson.beam \ mochijson2.beam \ diff --git a/src/mochiweb/mochifmt.erl b/src/mochiweb/mochifmt.erl new file mode 100644 index 00000000..da0a133a --- /dev/null +++ b/src/mochiweb/mochifmt.erl @@ -0,0 +1,426 @@ +%% @author Bob Ippolito <bob@mochimedia.com> +%% @copyright 2008 Mochi Media, Inc. + +%% @doc String Formatting for Erlang, inspired by Python 2.6 +%% (<a href="http://www.python.org/dev/peps/pep-3101/">PEP 3101</a>). +%% +-module(mochifmt). +-author('bob@mochimedia.com'). +-export([format/2, format_field/2, convert_field/2, get_value/2, get_field/2]). +-export([tokenize/1, format/3, get_field/3, format_field/3]). +-export([bformat/2, bformat/3]). +-export([f/2, f/3]). +-export([test/0]). + +-record(conversion, {length, precision, ctype, align, fill_char, sign}). + +%% @spec tokenize(S::string()) -> tokens() +%% @doc Tokenize a format string into mochifmt's internal format. +tokenize(S) -> + {?MODULE, tokenize(S, "", [])}. + +%% @spec convert_field(Arg, Conversion::conversion()) -> term() +%% @doc Process Arg according to the given explicit conversion specifier. +convert_field(Arg, "") -> + Arg; +convert_field(Arg, "r") -> + repr(Arg); +convert_field(Arg, "s") -> + str(Arg). + +%% @spec get_value(Key::string(), Args::args()) -> term() +%% @doc Get the Key from Args. If Args is a tuple then convert Key to +%% an integer and get element(1 + Key, Args). If Args is a list and Key +%% can be parsed as an integer then use lists:nth(1 + Key, Args), +%% otherwise try and look for Key in Args as a proplist, converting +%% Key to an atom or binary if necessary. +get_value(Key, Args) when is_tuple(Args) -> + element(1 + list_to_integer(Key), Args); +get_value(Key, Args) when is_list(Args) -> + try lists:nth(1 + list_to_integer(Key), Args) + catch error:_ -> + {_K, V} = proplist_lookup(Key, Args), + V + end. + +%% @spec get_field(Key::string(), Args) -> term() +%% @doc Consecutively call get_value/2 on parts of Key delimited by ".", +%% replacing Args with the result of the previous get_value. This +%% is used to implement formats such as {0.0}. +get_field(Key, Args) -> + get_field(Key, Args, ?MODULE). + +%% @spec get_field(Key::string(), Args, Module) -> term() +%% @doc Consecutively call Module:get_value/2 on parts of Key delimited by ".", +%% replacing Args with the result of the previous get_value. This +%% is used to implement formats such as {0.0}. +get_field(Key, Args, Module) -> + {Name, Next} = lists:splitwith(fun (C) -> C =/= $. end, Key), + Res = try Module:get_value(Name, Args) + catch error:undef -> get_value(Name, Args) end, + case Next of + "" -> + Res; + "." ++ S1 -> + get_field(S1, Res, Module) + end. + +%% @spec format(Format::string(), Args) -> iolist() +%% @doc Format Args with Format. +format(Format, Args) -> + format(Format, Args, ?MODULE). + +%% @spec format(Format::string(), Args, Module) -> iolist() +%% @doc Format Args with Format using Module. +format({?MODULE, Parts}, Args, Module) -> + format2(Parts, Args, Module, []); +format(S, Args, Module) -> + format(tokenize(S), Args, Module). + +%% @spec format_field(Arg, Format) -> iolist() +%% @doc Format Arg with Format. +format_field(Arg, Format) -> + format_field(Arg, Format, ?MODULE). + +%% @spec format_field(Arg, Format, _Module) -> iolist() +%% @doc Format Arg with Format. +format_field(Arg, Format, _Module) -> + F = default_ctype(Arg, parse_std_conversion(Format)), + fix_padding(fix_sign(convert2(Arg, F), F), F). + +%% @spec f(Format::string(), Args) -> string() +%% @doc Format Args with Format and return a string(). +f(Format, Args) -> + f(Format, Args, ?MODULE). + +%% @spec f(Format::string(), Args, Module) -> string() +%% @doc Format Args with Format using Module and return a string(). +f(Format, Args, Module) -> + case lists:member(${, Format) of + true -> + binary_to_list(bformat(Format, Args, Module)); + false -> + Format + end. + +%% @spec bformat(Format::string(), Args) -> binary() +%% @doc Format Args with Format and return a binary(). +bformat(Format, Args) -> + iolist_to_binary(format(Format, Args)). + +%% @spec bformat(Format::string(), Args, Module) -> binary() +%% @doc Format Args with Format using Module and return a binary(). +bformat(Format, Args, Module) -> + iolist_to_binary(format(Format, Args, Module)). + +%% @spec test() -> ok +%% @doc Run tests. +test() -> + ok = test_tokenize(), + ok = test_format(), + ok = test_std(), + ok = test_records(), + ok. + +%% Internal API + +add_raw("", Acc) -> + Acc; +add_raw(S, Acc) -> + [{raw, lists:reverse(S)} | Acc]. + +tokenize([], S, Acc) -> + lists:reverse(add_raw(S, Acc)); +tokenize("{{" ++ Rest, S, Acc) -> + tokenize(Rest, "{" ++ S, Acc); +tokenize("{" ++ Rest, S, Acc) -> + {Format, Rest1} = tokenize_format(Rest), + tokenize(Rest1, "", [{format, make_format(Format)} | add_raw(S, Acc)]); +tokenize("}}" ++ Rest, S, Acc) -> + tokenize(Rest, "}" ++ S, Acc); +tokenize([C | Rest], S, Acc) -> + tokenize(Rest, [C | S], Acc). + +tokenize_format(S) -> + tokenize_format(S, 1, []). + +tokenize_format("}" ++ Rest, 1, Acc) -> + {lists:reverse(Acc), Rest}; +tokenize_format("}" ++ Rest, N, Acc) -> + tokenize_format(Rest, N - 1, "}" ++ Acc); +tokenize_format("{" ++ Rest, N, Acc) -> + tokenize_format(Rest, 1 + N, "{" ++ Acc); +tokenize_format([C | Rest], N, Acc) -> + tokenize_format(Rest, N, [C | Acc]). + +make_format(S) -> + {Name0, Spec} = case lists:splitwith(fun (C) -> C =/= $: end, S) of + {_, ""} -> + {S, ""}; + {SN, ":" ++ SS} -> + {SN, SS} + end, + {Name, Transform} = case lists:splitwith(fun (C) -> C =/= $! end, Name0) of + {_, ""} -> + {Name0, ""}; + {TN, "!" ++ TT} -> + {TN, TT} + end, + {Name, Transform, Spec}. + +proplist_lookup(S, P) -> + A = try list_to_existing_atom(S) + catch error:_ -> make_ref() end, + B = try list_to_binary(S) + catch error:_ -> make_ref() end, + proplist_lookup2({S, A, B}, P). + +proplist_lookup2({KS, KA, KB}, [{K, V} | _]) + when KS =:= K orelse KA =:= K orelse KB =:= K -> + {K, V}; +proplist_lookup2(Keys, [_ | Rest]) -> + proplist_lookup2(Keys, Rest). + +format2([], _Args, _Module, Acc) -> + lists:reverse(Acc); +format2([{raw, S} | Rest], Args, Module, Acc) -> + format2(Rest, Args, Module, [S | Acc]); +format2([{format, {Key, Convert, Format0}} | Rest], Args, Module, Acc) -> + Format = f(Format0, Args, Module), + V = case Module of + ?MODULE -> + V0 = get_field(Key, Args), + V1 = convert_field(V0, Convert), + format_field(V1, Format); + _ -> + V0 = try Module:get_field(Key, Args) + catch error:undef -> get_field(Key, Args, Module) end, + V1 = try Module:convert_field(V0, Convert) + catch error:undef -> convert_field(V0, Convert) end, + try Module:format_field(V1, Format) + catch error:undef -> format_field(V1, Format, Module) end + end, + format2(Rest, Args, Module, [V | Acc]). + +default_ctype(_Arg, C=#conversion{ctype=N}) when N =/= undefined -> + C; +default_ctype(Arg, C) when is_integer(Arg) -> + C#conversion{ctype=decimal}; +default_ctype(Arg, C) when is_float(Arg) -> + C#conversion{ctype=general}; +default_ctype(_Arg, C) -> + C#conversion{ctype=string}. + +fix_padding(Arg, #conversion{length=undefined}) -> + Arg; +fix_padding(Arg, F=#conversion{length=Length, fill_char=Fill0, align=Align0, + ctype=Type}) -> + Padding = Length - iolist_size(Arg), + Fill = case Fill0 of + undefined -> + $\s; + _ -> + Fill0 + end, + Align = case Align0 of + undefined -> + case Type of + string -> + left; + _ -> + right + end; + _ -> + Align0 + end, + case Padding > 0 of + true -> + do_padding(Arg, Padding, Fill, Align, F); + false -> + Arg + end. + +do_padding(Arg, Padding, Fill, right, _F) -> + [lists:duplicate(Padding, Fill), Arg]; +do_padding(Arg, Padding, Fill, center, _F) -> + LPadding = lists:duplicate(Padding div 2, Fill), + RPadding = case Padding band 1 of + 1 -> + [Fill | LPadding]; + _ -> + LPadding + end, + [LPadding, Arg, RPadding]; +do_padding([$- | Arg], Padding, Fill, sign_right, _F) -> + [[$- | lists:duplicate(Padding, Fill)], Arg]; +do_padding(Arg, Padding, Fill, sign_right, #conversion{sign=$-}) -> + [lists:duplicate(Padding, Fill), Arg]; +do_padding([S | Arg], Padding, Fill, sign_right, #conversion{sign=S}) -> + [[S | lists:duplicate(Padding, Fill)], Arg]; +do_padding(Arg, Padding, Fill, sign_right, #conversion{sign=undefined}) -> + [lists:duplicate(Padding, Fill), Arg]; +do_padding(Arg, Padding, Fill, left, _F) -> + [Arg | lists:duplicate(Padding, Fill)]. + +fix_sign(Arg, #conversion{sign=$+}) when Arg >= 0 -> + [$+, Arg]; +fix_sign(Arg, #conversion{sign=$\s}) when Arg >= 0 -> + [$\s, Arg]; +fix_sign(Arg, _F) -> + Arg. + +ctype($\%) -> percent; +ctype($s) -> string; +ctype($b) -> bin; +ctype($o) -> oct; +ctype($X) -> upper_hex; +ctype($x) -> hex; +ctype($c) -> char; +ctype($d) -> decimal; +ctype($g) -> general; +ctype($f) -> fixed; +ctype($e) -> exp. + +align($<) -> left; +align($>) -> right; +align($^) -> center; +align($=) -> sign_right. + +convert2(Arg, F=#conversion{ctype=percent}) -> + [convert2(100.0 * Arg, F#conversion{ctype=fixed}), $\%]; +convert2(Arg, #conversion{ctype=string}) -> + str(Arg); +convert2(Arg, #conversion{ctype=bin}) -> + erlang:integer_to_list(Arg, 2); +convert2(Arg, #conversion{ctype=oct}) -> + erlang:integer_to_list(Arg, 8); +convert2(Arg, #conversion{ctype=upper_hex}) -> + erlang:integer_to_list(Arg, 16); +convert2(Arg, #conversion{ctype=hex}) -> + string:to_lower(erlang:integer_to_list(Arg, 16)); +convert2(Arg, #conversion{ctype=char}) when Arg < 16#80 -> + [Arg]; +convert2(Arg, #conversion{ctype=char}) -> + xmerl_ucs:to_utf8(Arg); +convert2(Arg, #conversion{ctype=decimal}) -> + integer_to_list(Arg); +convert2(Arg, #conversion{ctype=general, precision=undefined}) -> + try mochinum:digits(Arg) + catch error:undef -> io_lib:format("~g", [Arg]) end; +convert2(Arg, #conversion{ctype=fixed, precision=undefined}) -> + io_lib:format("~f", [Arg]); +convert2(Arg, #conversion{ctype=exp, precision=undefined}) -> + io_lib:format("~e", [Arg]); +convert2(Arg, #conversion{ctype=general, precision=P}) -> + io_lib:format("~." ++ integer_to_list(P) ++ "g", [Arg]); +convert2(Arg, #conversion{ctype=fixed, precision=P}) -> + io_lib:format("~." ++ integer_to_list(P) ++ "f", [Arg]); +convert2(Arg, #conversion{ctype=exp, precision=P}) -> + io_lib:format("~." ++ integer_to_list(P) ++ "e", [Arg]). + +str(A) when is_atom(A) -> + atom_to_list(A); +str(I) when is_integer(I) -> + integer_to_list(I); +str(F) when is_float(F) -> + try mochinum:digits(F) + catch error:undef -> io_lib:format("~g", [F]) end; +str(L) when is_list(L) -> + L; +str(B) when is_binary(B) -> + B; +str(P) -> + repr(P). + +repr(P) when is_float(P) -> + try mochinum:digits(P) + catch error:undef -> float_to_list(P) end; +repr(P) -> + io_lib:format("~p", [P]). + +parse_std_conversion(S) -> + parse_std_conversion(S, #conversion{}). + +parse_std_conversion("", Acc) -> + Acc; +parse_std_conversion([Fill, Align | Spec], Acc) + when Align =:= $< orelse Align =:= $> orelse Align =:= $= orelse Align =:= $^ -> + parse_std_conversion(Spec, Acc#conversion{fill_char=Fill, + align=align(Align)}); +parse_std_conversion([Align | Spec], Acc) + when Align =:= $< orelse Align =:= $> orelse Align =:= $= orelse Align =:= $^ -> + parse_std_conversion(Spec, Acc#conversion{align=align(Align)}); +parse_std_conversion([Sign | Spec], Acc) + when Sign =:= $+ orelse Sign =:= $- orelse Sign =:= $\s -> + parse_std_conversion(Spec, Acc#conversion{sign=Sign}); +parse_std_conversion("0" ++ Spec, Acc) -> + Align = case Acc#conversion.align of + undefined -> + sign_right; + A -> + A + end, + parse_std_conversion(Spec, Acc#conversion{fill_char=$0, align=Align}); +parse_std_conversion(Spec=[D|_], Acc) when D >= $0 andalso D =< $9 -> + {W, Spec1} = lists:splitwith(fun (C) -> C >= $0 andalso C =< $9 end, Spec), + parse_std_conversion(Spec1, Acc#conversion{length=list_to_integer(W)}); +parse_std_conversion([$. | Spec], Acc) -> + case lists:splitwith(fun (C) -> C >= $0 andalso C =< $9 end, Spec) of + {"", Spec1} -> + parse_std_conversion(Spec1, Acc); + {P, Spec1} -> + parse_std_conversion(Spec1, + Acc#conversion{precision=list_to_integer(P)}) + end; +parse_std_conversion([Type], Acc) -> + parse_std_conversion("", Acc#conversion{ctype=ctype(Type)}). + +test_tokenize() -> + {?MODULE, [{raw, "ABC"}]} = tokenize("ABC"), + {?MODULE, [{format, {"0", "", ""}}]} = tokenize("{0}"), + {?MODULE, [{raw, "ABC"}, {format, {"1", "", ""}}, {raw, "DEF"}]} = + tokenize("ABC{1}DEF"), + ok. + +test_format() -> + <<" -4">> = bformat("{0:4}", [-4]), + <<" 4">> = bformat("{0:4}", [4]), + <<" 4">> = bformat("{0:{0}}", [4]), + <<"4 ">> = bformat("{0:4}", ["4"]), + <<"4 ">> = bformat("{0:{0}}", ["4"]), + <<"1.2yoDEF">> = bformat("{2}{0}{1}{3}", {yo, "DE", 1.2, <<"F">>}), + <<"cafebabe">> = bformat("{0:x}", {16#cafebabe}), + <<"CAFEBABE">> = bformat("{0:X}", {16#cafebabe}), + <<"CAFEBABE">> = bformat("{0:X}", {16#cafebabe}), + <<"755">> = bformat("{0:o}", {8#755}), + <<"a">> = bformat("{0:c}", {97}), + %% Horizontal ellipsis + <<226, 128, 166>> = bformat("{0:c}", {16#2026}), + <<"11">> = bformat("{0:b}", {3}), + <<"11">> = bformat("{0:b}", [3]), + <<"11">> = bformat("{three:b}", [{three, 3}]), + <<"11">> = bformat("{three:b}", [{"three", 3}]), + <<"11">> = bformat("{three:b}", [{<<"three">>, 3}]), + <<"\"foo\"">> = bformat("{0!r}", {"foo"}), + <<"2008-5-4">> = bformat("{0.0}-{0.1}-{0.2}", {{2008,5,4}}), + <<"2008-05-04">> = bformat("{0.0:04}-{0.1:02}-{0.2:02}", {{2008,5,4}}), + <<"foo6bar-6">> = bformat("foo{1}{0}-{1}", {bar, 6}), + <<"-'atom test'-">> = bformat("-{arg!r}-", [{arg, 'atom test'}]), + <<"2008-05-04">> = bformat("{0.0:0{1.0}}-{0.1:0{1.1}}-{0.2:0{1.2}}", + {{2008,5,4}, {4, 2, 2}}), + ok. + +test_std() -> + M = mochifmt_std:new(), + <<"01">> = bformat("{0}{1}", [0, 1], M), + ok. + +test_records() -> + M = mochifmt_records:new([{conversion, record_info(fields, conversion)}]), + R = #conversion{length=long, precision=hard, sign=peace}, + long = M:get_value("length", R), + hard = M:get_value("precision", R), + peace = M:get_value("sign", R), + <<"long hard">> = bformat("{length} {precision}", R, M), + <<"long hard">> = bformat("{0.length} {0.precision}", [R], M), + ok. diff --git a/src/mochiweb/mochifmt_records.erl b/src/mochiweb/mochifmt_records.erl new file mode 100644 index 00000000..94c77978 --- /dev/null +++ b/src/mochiweb/mochifmt_records.erl @@ -0,0 +1,30 @@ +%% @author Bob Ippolito <bob@mochimedia.com> +%% @copyright 2008 Mochi Media, Inc. + +%% @doc Formatter that understands records. +%% +%% Usage: +%% +%% 1> M = mochifmt_records:new([{rec, record_info(fields, rec)}]), +%% M:format("{0.bar}", [#rec{bar=foo}]). +%% foo + +-module(mochifmt_records, [Recs]). +-author('bob@mochimedia.com'). +-export([get_value/2]). + +get_value(Key, Rec) when is_tuple(Rec) and is_atom(element(1, Rec)) -> + try begin + Atom = list_to_existing_atom(Key), + {_, Fields} = proplists:lookup(element(1, Rec), Recs), + element(get_rec_index(Atom, Fields, 2), Rec) + end + catch error:_ -> mochifmt:get_value(Key, Rec) + end; +get_value(Key, Args) -> + mochifmt:get_value(Key, Args). + +get_rec_index(Atom, [Atom | _], Index) -> + Index; +get_rec_index(Atom, [_ | Rest], Index) -> + get_rec_index(Atom, Rest, 1 + Index). diff --git a/src/mochiweb/mochifmt_std.erl b/src/mochiweb/mochifmt_std.erl new file mode 100644 index 00000000..9442016a --- /dev/null +++ b/src/mochiweb/mochifmt_std.erl @@ -0,0 +1,23 @@ +%% @author Bob Ippolito <bob@mochimedia.com> +%% @copyright 2008 Mochi Media, Inc. + +%% @doc Template module for a mochifmt formatter. + +-module(mochifmt_std, []). +-author('bob@mochimedia.com'). +-export([format/2, get_value/2, format_field/2, get_field/2, convert_field/2]). + +format(Format, Args) -> + mochifmt:format(Format, Args, THIS). + +get_field(Key, Args) -> + mochifmt:get_field(Key, Args, THIS). + +convert_field(Key, Args) -> + mochifmt:convert_field(Key, Args). + +get_value(Key, Args) -> + mochifmt:get_value(Key, Args). + +format_field(Arg, Format) -> + mochifmt:format_field(Arg, Format, THIS). diff --git a/src/mochiweb/mochiweb.erl b/src/mochiweb/mochiweb.erl index baf9c0c5..be30fccf 100644 --- a/src/mochiweb/mochiweb.erl +++ b/src/mochiweb/mochiweb.erl @@ -35,6 +35,7 @@ test() -> mochijson:test(), mochiweb_charref:test(), mochiweb_html:test(), + mochifmt:test(), test_request(), ok. diff --git a/src/mochiweb/mochiweb_html.erl b/src/mochiweb/mochiweb_html.erl index 3c7f9dfc..0ac4826e 100644 --- a/src/mochiweb/mochiweb_html.erl +++ b/src/mochiweb/mochiweb_html.erl @@ -35,7 +35,7 @@ -define(IS_LITERAL_SAFE(C), ((C >= $A andalso C =< $Z) orelse (C >= $a andalso C =< $z) orelse (C >= $0 andalso C =< $9))). - + -record(decoder, {line=1, column=1, offset=0}). @@ -202,7 +202,7 @@ attrs_to_html([{K, V} | Rest], Acc) -> attrs_to_html(Rest, [[<<" ">>, escape(K), <<"=\"">>, escape_attr(V), <<"\"">>] | Acc]). - + test_escape() -> <<"&quot;\"word <<up!&quot;">> = escape(<<""\"word <<up!"">>), @@ -299,6 +299,14 @@ test_tokens() -> tokens(<<"<foo bar=baz wibble='wibble' alice=bob/>">>), [{comment, <<"[if lt IE 7]>\n<style type=\"text/css\">\n.no_ie { display: none; }\n</style>\n<![endif]">>}] = tokens(<<"<!--[if lt IE 7]>\n<style type=\"text/css\">\n.no_ie { display: none; }\n</style>\n<![endif]-->">>), + [{start_tag, <<"script">>, [{<<"type">>, <<"text/javascript">>}], false}, + {data, <<" A= B <= C ">>, false}, + {end_tag, <<"script">>}] = + tokens(<<"<script type=\"text/javascript\"> A= B <= C </script>">>), + [{start_tag, <<"textarea">>, [], false}, + {data, <<"<html></body>">>, false}, + {end_tag, <<"textarea">>}] = + tokens(<<"<textarea><html></body></textarea>">>), ok. tokens(B, S=#decoder{offset=O}, Acc) -> @@ -307,9 +315,30 @@ tokens(B, S=#decoder{offset=O}, Acc) -> lists:reverse(Acc); _ -> {Tag, S1} = tokenize(B, S), - tokens(B, S1, [Tag | Acc]) + case parse_flag(Tag) of + script -> + {Tag2, S2} = tokenize_script(B, S1), + tokens(B, S2, [Tag2, Tag | Acc]); + textarea -> + {Tag2, S2} = tokenize_textarea(B, S1), + tokens(B, S2, [Tag2, Tag | Acc]); + none -> + tokens(B, S1, [Tag | Acc]) + end end. +parse_flag({start_tag, B, _, false}) -> + case string:to_lower(binary_to_list(B)) of + "script" -> + script; + "textarea" -> + textarea; + _ -> + none + end; +parse_flag(_) -> + none. + tokenize(B, S=#decoder{offset=O}) -> case B of <<_:O/binary, "<!--", _/binary>> -> @@ -330,7 +359,7 @@ tokenize(B, S=#decoder{offset=O}) -> {S2, _} = find_gt(B, S1), {{end_tag, Tag}, S2}; <<_:O/binary, "<", C, _/binary>> when ?IS_WHITESPACE(C) -> - %% This isn't really strict HTML but we want this for markdown + %% This isn't really strict HTML tokenize_data(B, ?INC_COL(S)); <<_:O/binary, "<", _/binary>> -> {Tag, S1} = tokenize_literal(B, ?INC_COL(S)), @@ -524,7 +553,7 @@ append_stack_child(StartTag, [{Name, Attrs, Acc} | Stack]) -> destack(TagName, Stack) when is_list(Stack) -> F = fun (X) -> - case X of + case X of {TagName, _, _} -> false; _ -> @@ -542,7 +571,7 @@ destack(TagName, Stack) when is_list(Stack) -> %% Unfurl up to the tag, then accumulate it [{T0, A0, [destack(Pre ++ [T]) | Acc0]} | Post] end. - + destack([{Tag, Attrs, Acc}]) -> {Tag, Attrs, lists:reverse(Acc)}; destack([{T1, A1, Acc1}, {T0, A0, Acc0} | Rest]) -> @@ -655,7 +684,7 @@ find_gt(Bin, S=#decoder{offset=O}, HasSlash) -> tokenize_charref(Bin, S=#decoder{offset=O}) -> tokenize_charref(Bin, S, O). - + tokenize_charref(Bin, S=#decoder{offset=O}, Start) -> case Bin of <<_:O/binary>> -> @@ -758,3 +787,49 @@ tokenize_comment(Bin, S=#decoder{offset=O}, Start) -> <<_:Start/binary, Raw/binary>> -> {{comment, Raw}, S} end. + +tokenize_script(Bin, S=#decoder{offset=O}) -> + tokenize_script(Bin, S, O). + +tokenize_script(Bin, S=#decoder{offset=O}, Start) -> + case Bin of + %% Just a look-ahead, we want the end_tag separately + <<_:O/binary, $<, $/, SS, CC, RR, II, PP, TT, _/binary>> + when (SS =:= $s orelse SS =:= $S) andalso + (CC =:= $c orelse CC =:= $C) andalso + (RR =:= $r orelse RR =:= $R) andalso + (II =:= $i orelse II =:= $I) andalso + (PP =:= $p orelse PP =:= $P) andalso + (TT=:= $t orelse TT =:= $T) -> + Len = O - Start, + <<_:Start/binary, Raw:Len/binary, _/binary>> = Bin, + {{data, Raw, false}, S}; + <<_:O/binary, C, _/binary>> -> + tokenize_script(Bin, ?INC_CHAR(S, C), Start); + <<_:Start/binary, Raw/binary>> -> + {{data, Raw, false}, S} + end. + +tokenize_textarea(Bin, S=#decoder{offset=O}) -> + tokenize_textarea(Bin, S, O). + +tokenize_textarea(Bin, S=#decoder{offset=O}, Start) -> + case Bin of + %% Just a look-ahead, we want the end_tag separately + <<_:O/binary, $<, $/, TT, EE, XX, TT2, AA, RR, EE2, AA2, _/binary>> + when (TT =:= $t orelse TT =:= $T) andalso + (EE =:= $e orelse EE =:= $E) andalso + (XX =:= $x orelse XX =:= $X) andalso + (TT2 =:= $t orelse TT2 =:= $T) andalso + (AA =:= $a orelse AA =:= $A) andalso + (RR =:= $r orelse RR =:= $R) andalso + (EE2 =:= $e orelse EE2 =:= $E) andalso + (AA2 =:= $a orelse AA2 =:= $A) -> + Len = O - Start, + <<_:Start/binary, Raw:Len/binary, _/binary>> = Bin, + {{data, Raw, false}, S}; + <<_:O/binary, C, _/binary>> -> + tokenize_textarea(Bin, ?INC_CHAR(S, C), Start); + <<_:Start/binary, Raw/binary>> -> + {{data, Raw, false}, S} + end. diff --git a/src/mochiweb/mochiweb_multipart.erl b/src/mochiweb/mochiweb_multipart.erl index cac35a91..9eb4badd 100644 --- a/src/mochiweb/mochiweb_multipart.erl +++ b/src/mochiweb/mochiweb_multipart.erl @@ -6,7 +6,7 @@ -module(mochiweb_multipart). -author('bob@mochimedia.com'). --export([parse_form/2]). +-export([parse_form/1, parse_form/2]). -export([parse_multipart_request/2]). -export([test/0]). @@ -16,6 +16,9 @@ %% TODO: DOCUMENT THIS MODULE. +parse_form(Req) -> + parse_form(Req, fun default_file_handler/2). + parse_form(Req, FileHandler) -> Callback = fun (Next) -> parse_form_outer(Next, FileHandler, []) end, {_, _, Res} = parse_multipart_request(Req, Callback), @@ -56,6 +59,17 @@ parse_form_file({body, Data}, {Name, Handler}, FileHandler, State) -> H1 = Handler(Data), fun (Next) -> parse_form_file(Next, {Name, H1}, FileHandler, State) end. +default_file_handler(Filename, ContentType) -> + default_file_handler_1(Filename, ContentType, []). + +default_file_handler_1(Filename, ContentType, Acc) -> + fun(eof) -> + Value = iolist_to_binary(lists:reverse(Acc)), + {Filename, ContentType, Value}; + (Next) -> + default_file_handler_1(Filename, ContentType, [Next | Acc]) + end. + parse_multipart_request(Req, Callback) -> %% TODO: Support chunked? Length = list_to_integer(Req:get_header_value("content-length")), @@ -295,19 +309,6 @@ test_parse2() -> ok = with_socket_server(ServerFun, ClientFun), ok. -handler_test(Filename, ContentType) -> - fun (Next) -> - handler_test_read(Next, {Filename, ContentType}, []) - end. - -handler_test_read(eof, {Filename, ContentType}, Acc) -> - Value = iolist_to_binary(lists:reverse(Acc)), - {Filename, ContentType, Value}; -handler_test_read(Data, H, Acc) -> - Acc1 = [Data | Acc], - fun (Next) -> handler_test_read(Next, H, Acc1) end. - - test_parse_form() -> ContentType = "multipart/form-data; boundary=AaB03x", "AaB03x" = get_boundary(ContentType), @@ -334,7 +335,7 @@ test_parse_form() -> ClientFun = fun (Socket) -> Req = fake_request(Socket, ContentType, size(BinContent)), - Res = parse_form(Req, fun handler_test/2), + Res = parse_form(Req), [{"submit-name", "Larry"}, {"files", {"file1.txt", {"text/plain",[]}, <<"... contents of file1.txt ...">>} diff --git a/src/mochiweb/mochiweb_request.erl b/src/mochiweb/mochiweb_request.erl index d9ca7eb0..b4a7bcf7 100644 --- a/src/mochiweb/mochiweb_request.erl +++ b/src/mochiweb/mochiweb_request.erl @@ -386,7 +386,7 @@ parse_post() -> []; Binary -> case get_primary_header_value("content-type") of - "application/x-www-form-urlencoded" -> + "application/x-www-form-urlencoded" ++ _ -> mochiweb_util:parse_qs(Binary); _ -> [] diff --git a/src/mochiweb/mochiweb_socket_server.erl b/src/mochiweb/mochiweb_socket_server.erl index a88ace76..0d538ac9 100644 --- a/src/mochiweb/mochiweb_socket_server.erl +++ b/src/mochiweb/mochiweb_socket_server.erl @@ -103,7 +103,8 @@ init(State=#mochiweb_socket_server{ip=Ip, port=Port, backlog=Backlog}) -> {packet, 0}, {backlog, Backlog}, {recbuf, 8192}, - {active, false}], + {active, false}, + {nodelay, true}], Opts = case Ip of any -> BaseOpts; |