umgeher/jsx
0
Fork 0
Code Issues Pull requests Activity

fix bad utf8 encodings by default (convert to u+FFFD)

Browse source 
this (temporarily, i hope) removes the fast path for handling
strings internally without escaping or replacing bad utf8
This commit is contained in:
alisdair sullivan 2013-10-20 03:16:56 +00:00
parent 7b05d0e1b7
commit f137ff484b
6 changed files with 220 additions and 378 deletions
Download patch file Download diff file Expand all files Collapse all files

16
src/jsx_config.erl
View file

@ -40,8 +40,8 @@ parse_config(Config) ->
parse_config([], Config) -> parse_config([], Config) ->
Config; Config;
parse_config([replaced_bad_utf8|Rest], Config) -> parse_config([strict_utf8|Rest], Config) ->
parse_config(Rest, Config#config{replaced_bad_utf8=true}); parse_config(Rest, Config#config{strict_utf8=true});
parse_config([escaped_forward_slashes|Rest], Config) -> parse_config([escaped_forward_slashes|Rest], Config) ->
parse_config(Rest, Config#config{escaped_forward_slashes=true}); parse_config(Rest, Config#config{escaped_forward_slashes=true});
parse_config([stream|Rest], Config) -> parse_config([stream|Rest], Config) ->
@ -60,7 +60,6 @@ parse_config([ignored_bad_escapes|Rest], Config) ->
parse_config(Rest, Config#config{ignored_bad_escapes=true}); parse_config(Rest, Config#config{ignored_bad_escapes=true});
parse_config([relax|Rest], Config) -> parse_config([relax|Rest], Config) ->
parse_config(Rest, Config#config{ parse_config(Rest, Config#config{
replaced_bad_utf8 = true,
single_quoted_strings = true, single_quoted_strings = true,
comments = true, comments = true,
ignored_bad_escapes = true ignored_bad_escapes = true
@ -95,7 +94,7 @@ config_to_list(Config) ->
valid_flags() -> valid_flags() ->
[ [
replaced_bad_utf8, strict_utf8,
escaped_forward_slashes, escaped_forward_slashes,
single_quoted_strings, single_quoted_strings,
unescaped_jsonp, unescaped_jsonp,
@ -136,7 +135,7 @@ config_test_() ->
{"all flags", {"all flags",
?_assertEqual( ?_assertEqual(
#config{ #config{
replaced_bad_utf8=true, strict_utf8=true,
escaped_forward_slashes=true, escaped_forward_slashes=true,
stream=true, stream=true,
single_quoted_strings=true, single_quoted_strings=true,
@ -146,7 +145,7 @@ config_test_() ->
ignored_bad_escapes=true ignored_bad_escapes=true
}, },
parse_config([ parse_config([
replaced_bad_utf8, strict_utf8,
escaped_forward_slashes, escaped_forward_slashes,
stream, stream,
single_quoted_strings, single_quoted_strings,
@ -160,7 +159,6 @@ config_test_() ->
{"relax flag", {"relax flag",
?_assertEqual( ?_assertEqual(
#config{ #config{
replaced_bad_utf8=true,
single_quoted_strings=true, single_quoted_strings=true,
comments=true, comments=true,
ignored_bad_escapes=true ignored_bad_escapes=true
@ -202,7 +200,7 @@ config_to_list_test_() ->
)}, )},
{"all flags", ?_assertEqual( {"all flags", ?_assertEqual(
[ [
replaced_bad_utf8, strict_utf8,
escaped_forward_slashes, escaped_forward_slashes,
single_quoted_strings, single_quoted_strings,
unescaped_jsonp, unescaped_jsonp,
@ -213,7 +211,7 @@ config_to_list_test_() ->
], ],
config_to_list( config_to_list(
#config{ #config{
replaced_bad_utf8=true, strict_utf8=true,
escaped_forward_slashes=true, escaped_forward_slashes=true,
stream=true, stream=true,
single_quoted_strings=true, single_quoted_strings=true,

2
src/jsx_config.hrl
View file

@ -1,5 +1,5 @@
-record(config, { -record(config, {
replaced_bad_utf8 = false, strict_utf8 = false,
escaped_forward_slashes = false, escaped_forward_slashes = false,
single_quoted_strings = false, single_quoted_strings = false,
unescaped_jsonp = false, unescaped_jsonp = false,

165
src/jsx_decoder.erl
View file

@ -550,34 +550,33 @@ string(<<X, Y, Z>>, Handler, Acc, Stack, Config)
Z >= 16#80, Z =< 16#bf -> Z >= 16#80, Z =< 16#bf ->
incomplete(string, <<X, Y, Z>>, Handler, Acc, Stack, Config); incomplete(string, <<X, Y, Z>>, Handler, Acc, Stack, Config);
%% surrogates %% surrogates
string(<<237, X, _, Rest/binary>>, Handler, Acc, Stack, Config=#config{replaced_bad_utf8=true}) string(<<237, X, _, Rest/binary>>, Handler, Acc, Stack, Config=#config{strict_utf8=false})
when X >= 160 -> when X >= 160 ->
string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config); string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config);
%% u+xfffe, u+xffff, control codes and other noncharacters %% u+xfffe, u+xffff, control codes and other noncharacters
string(<<_/utf8, Rest/binary>>, Handler, Acc, Stack, Config=#config{replaced_bad_utf8=true}) -> string(<<_/utf8, Rest/binary>>, Handler, Acc, Stack, Config=#config{strict_utf8=false}) ->
string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config); string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config);
%% u+fffe and u+ffff for R14BXX (subsequent runtimes will happily match the %% u+fffe and u+ffff for R14BXX (subsequent runtimes will happily match the
%% preceeding clause %% preceeding clause
string(<<239, 191, X, Rest/binary>>, Handler, Acc, Stack, Config=#config{replaced_bad_utf8=true}) string(<<239, 191, X, Rest/binary>>, Handler, Acc, Stack, Config=#config{strict_utf8=false})
when X == 190; X == 191 -> when X == 190; X == 191 ->
string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config); string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config);
%% overlong encodings and missing continuations of a 2 byte sequence %% overlong encodings and missing continuations of a 2 byte sequence
string(<<X, Rest/binary>>, Handler, Acc, Stack, Config=#config{replaced_bad_utf8=true}) string(<<X, Rest/binary>>, Handler, Acc, Stack, Config=#config{strict_utf8=false})
when X >= 192, X =< 223 -> when X >= 192, X =< 223 ->
strip_continuations(Rest, Handler, Acc, Stack, Config, 1); strip_continuations(Rest, Handler, Acc, Stack, Config, 1);
%% overlong encodings and missing continuations of a 3 byte sequence %% overlong encodings and missing continuations of a 3 byte sequence
string(<<X, Rest/binary>>, Handler, Acc, Stack, Config=#config{replaced_bad_utf8=true}) string(<<X, Rest/binary>>, Handler, Acc, Stack, Config=#config{strict_utf8=false})
when X >= 224, X =< 239 -> when X >= 224, X =< 239 ->
strip_continuations(Rest, Handler, Acc, Stack, Config, 2); strip_continuations(Rest, Handler, Acc, Stack, Config, 2);
%% overlong encodings and missing continuations of a 4 byte sequence %% overlong encodings and missing continuations of a 4 byte sequence
string(<<X, Rest/binary>>, Handler, Acc, Stack, Config=#config{replaced_bad_utf8=true}) string(<<X, Rest/binary>>, Handler, Acc, Stack, Config=#config{strict_utf8=false})
when X >= 240, X =< 247 -> when X >= 240, X =< 247 ->
strip_continuations(Rest, Handler, Acc, Stack, Config, 3); strip_continuations(Rest, Handler, Acc, Stack, Config, 3);
%% incompletes and unexpected bytes, including orphan continuations %% incompletes and unexpected bytes, including orphan continuations
string(<<_, Rest/binary>>, Handler, Acc, Stack, Config=#config{replaced_bad_utf8=true}) -> string(<<_, Rest/binary>>, Handler, Acc, Stack, Config=#config{strict_utf8=false}) ->
string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config); string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config);
string(Bin, Handler, Acc, Stack, Config) -> string(Bin, Handler, Acc, Stack, Config) -> ?error(string, Bin, Handler, Acc, Stack, Config).
?error(string, Bin, Handler, Acc, Stack, Config).
doublequote(<<Rest/binary>>, Handler, Acc, [key|_] = Stack, Config) -> doublequote(<<Rest/binary>>, Handler, Acc, [key|_] = Stack, Config) ->
@ -658,9 +657,9 @@ unescape(<<$u, $d, A, B, C, ?rsolidus, $u, W, X, Y, Z, Rest/binary>>, Handler, A
when (A == $8 orelse A == $9 orelse A == $a orelse A == $b), when (A == $8 orelse A == $9 orelse A == $a orelse A == $b),
?is_hex(B), ?is_hex(C), ?is_hex(W), ?is_hex(X), ?is_hex(Y), ?is_hex(Z) ?is_hex(B), ?is_hex(C), ?is_hex(W), ?is_hex(X), ?is_hex(Y), ?is_hex(Z)
-> ->
case Config#config.replaced_bad_utf8 of case Config#config.strict_utf8 of
true -> string(Rest, Handler, acc_seq(Acc, [16#fffd, 16#fffd]), Stack, Config); true -> ?error(<<$u, $d, A, B, C, ?rsolidus, $u, W, X, Y, Z, Rest/binary>>, Handler, Acc, Stack, Config);
false -> ?error(<<$u, $d, A, B, C, ?rsolidus, $u, W, X, Y, Z, Rest/binary>>, Handler, Acc, Stack, Config) false -> string(Rest, Handler, acc_seq(Acc, [16#fffd, 16#fffd]), Stack, Config)
end; end;
unescape(<<$u, $d, A, B, C, ?rsolidus, Rest/binary>>, Handler, Acc, Stack, Config) unescape(<<$u, $d, A, B, C, ?rsolidus, Rest/binary>>, Handler, Acc, Stack, Config)
when (A == $8 orelse A == $9 orelse A == $a orelse A == $b) andalso when (A == $8 orelse A == $9 orelse A == $a orelse A == $b) andalso
@ -677,9 +676,9 @@ unescape(<<$u, A, B, C, D, Rest/binary>>, Handler, Acc, Stack, Config)
case erlang:list_to_integer([A, B, C, D], 16) of case erlang:list_to_integer([A, B, C, D], 16) of
Codepoint when Codepoint < 16#d800; Codepoint > 16#dfff -> Codepoint when Codepoint < 16#d800; Codepoint > 16#dfff ->
string(Rest, Handler, acc_seq(Acc, maybe_replace(Codepoint, Config)), Stack, Config); string(Rest, Handler, acc_seq(Acc, maybe_replace(Codepoint, Config)), Stack, Config);
_ when Config#config.replaced_bad_utf8 -> _ when Config#config.strict_utf8 ->
string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config); ?error(string, <<?rsolidus, $u, A, B, C, D, Rest/binary>>, Handler, Acc, Stack, Config);
_ -> ?error(string, <<?rsolidus, $u, A, B, C, D, Rest/binary>>, Handler, Acc, Stack, Config) _ -> string(Rest, Handler, acc_seq(Acc, 16#fffd), Stack, Config)
end; end;
unescape(Bin, Handler, Acc, Stack, Config=#config{ignored_bad_escapes=true}) -> unescape(Bin, Handler, Acc, Stack, Config=#config{ignored_bad_escapes=true}) ->
string(Bin, Handler, acc_seq(Acc, ?rsolidus), Stack, Config); string(Bin, Handler, acc_seq(Acc, ?rsolidus), Stack, Config);
@ -904,7 +903,7 @@ comment(<<?star>>, Handler, Resume, [multicomment|_] = Stack, Config) ->
incomplete(comment, <<?star>>, Handler, Resume, Stack, Config); incomplete(comment, <<?star>>, Handler, Resume, Stack, Config);
comment(<<_/utf8, Rest/binary>>, Handler, Resume, Stack, Config) -> comment(<<_/utf8, Rest/binary>>, Handler, Resume, Stack, Config) ->
comment(Rest, Handler, Resume, Stack, Config); comment(Rest, Handler, Resume, Stack, Config);
comment(<<_, Rest/binary>>, Handler, Resume, Stack, Config=#config{replaced_bad_utf8=true}) -> comment(<<_, Rest/binary>>, Handler, Resume, Stack, Config=#config{strict_utf8=false}) ->
comment(Rest, Handler, Resume, Stack, Config); comment(Rest, Handler, Resume, Stack, Config);
comment(<<>>, Handler, done, [Comment], Config=#config{stream=false}) comment(<<>>, Handler, done, [Comment], Config=#config{stream=false})
when Comment == comment; Comment == multicomment -> when Comment == comment; Comment == multicomment ->
@ -1250,15 +1249,15 @@ comments_test_() ->
)}, )},
{"// comment with badutf", ?_assertEqual( {"// comment with badutf", ?_assertEqual(
[start_array, {literal, true}, end_array, end_json], [start_array, {literal, true}, end_array, end_json],
decode(<<"[ // comment ", 16#00c0, " ", ?newline, "true]">>, [comments, replaced_bad_utf8]) decode(<<"[ // comment ", 16#00c0, " ", ?newline, "true]">>, [comments])
)}, )},
{"/**/ comment with badutf", ?_assertEqual( {"/**/ comment with badutf", ?_assertEqual(
[start_array, {literal, true}, end_array, end_json], [start_array, {literal, true}, end_array, end_json],
decode(<<"[ /* comment ", 16#00c0, " */ true]">>, [comments, replaced_bad_utf8]) decode(<<"[ /* comment ", 16#00c0, " */ true]">>, [comments])
)}, )},
{"/**/ comment with badutf preceeded by /", ?_assertEqual( {"/**/ comment with badutf preceeded by /", ?_assertEqual(
[start_array, {literal, true}, end_array, end_json], [start_array, {literal, true}, end_array, end_json],
decode(<<"[ /* comment /", 16#00c0, " */ true]">>, [comments, replaced_bad_utf8]) decode(<<"[ /* comment /", 16#00c0, " */ true]">>, [comments])
)} )}
]. ].
@ -1326,35 +1325,35 @@ clean_string_test_() ->
)}, )},
{"error reserved space", ?_assertEqual( {"error reserved space", ?_assertEqual(
lists:duplicate(length(reserved_space()), {error, badarg}), lists:duplicate(length(reserved_space()), {error, badarg}),
lists:map(fun(Codepoint) -> decode(Codepoint, []) end, reserved_space()) lists:map(fun(Codepoint) -> decode(Codepoint, [strict_utf8]) end, reserved_space())
)}, )},
{"error surrogates", ?_assertEqual( {"error surrogates", ?_assertEqual(
lists:duplicate(length(surrogates()), {error, badarg}), lists:duplicate(length(surrogates()), {error, badarg}),
lists:map(fun(Codepoint) -> decode(Codepoint, []) end, surrogates()) lists:map(fun(Codepoint) -> decode(Codepoint, [strict_utf8]) end, surrogates())
)}, )},
{"error noncharacters", ?_assertEqual( {"error noncharacters", ?_assertEqual(
lists:duplicate(length(noncharacters()), {error, badarg}), lists:duplicate(length(noncharacters()), {error, badarg}),
lists:map(fun(Codepoint) -> decode(Codepoint, []) end, noncharacters()) lists:map(fun(Codepoint) -> decode(Codepoint, [strict_utf8]) end, noncharacters())
)}, )},
{"error extended noncharacters", ?_assertEqual( {"error extended noncharacters", ?_assertEqual(
lists:duplicate(length(extended_noncharacters()), {error, badarg}), lists:duplicate(length(extended_noncharacters()), {error, badarg}),
lists:map(fun(Codepoint) -> decode(Codepoint, []) end, extended_noncharacters()) lists:map(fun(Codepoint) -> decode(Codepoint, [strict_utf8]) end, extended_noncharacters())
)}, )},
{"clean reserved space", ?_assertEqual( {"clean reserved space", ?_assertEqual(
lists:duplicate(length(reserved_space()), [{string, <<16#fffd/utf8>>}, end_json]), lists:duplicate(length(reserved_space()), [{string, <<16#fffd/utf8>>}, end_json]),
lists:map(fun(Codepoint) -> decode(Codepoint, [replaced_bad_utf8]) end, reserved_space()) lists:map(fun(Codepoint) -> decode(Codepoint, []) end, reserved_space())
)}, )},
{"clean surrogates", ?_assertEqual( {"clean surrogates", ?_assertEqual(
lists:duplicate(length(surrogates()), [{string, <<16#fffd/utf8>>}, end_json]), lists:duplicate(length(surrogates()), [{string, <<16#fffd/utf8>>}, end_json]),
lists:map(fun(Codepoint) -> decode(Codepoint, [replaced_bad_utf8]) end, surrogates()) lists:map(fun(Codepoint) -> decode(Codepoint, []) end, surrogates())
)}, )},
{"clean noncharacters", ?_assertEqual( {"clean noncharacters", ?_assertEqual(
lists:duplicate(length(noncharacters()), [{string, <<16#fffd/utf8>>}, end_json]), lists:duplicate(length(noncharacters()), [{string, <<16#fffd/utf8>>}, end_json]),
lists:map(fun(Codepoint) -> decode(Codepoint, [replaced_bad_utf8]) end, noncharacters()) lists:map(fun(Codepoint) -> decode(Codepoint, []) end, noncharacters())
)}, )},
{"clean extended noncharacters", ?_assertEqual( {"clean extended noncharacters", ?_assertEqual(
lists:duplicate(length(extended_noncharacters()), [{string, <<16#fffd/utf8>>}, end_json]), lists:duplicate(length(extended_noncharacters()), [{string, <<16#fffd/utf8>>}, end_json]),
lists:map(fun(Codepoint) -> decode(Codepoint, [replaced_bad_utf8]) end, extended_noncharacters()) lists:map(fun(Codepoint) -> decode(Codepoint, []) end, extended_noncharacters())
)}, )},
{"dirty \\uwxyz", ?_assertEqual( {"dirty \\uwxyz", ?_assertEqual(
[{string, <<"\\uwxyz">>}, end_json], [{string, <<"\\uwxyz">>}, end_json],
@ -1405,190 +1404,190 @@ bad_utf8_test_() ->
[ [
{"noncharacter u+fffe", ?_assertError( {"noncharacter u+fffe", ?_assertError(
badarg, badarg,
decode_bad_utf(<<239, 191, 190>>, []) decode_bad_utf(<<239, 191, 190>>, [strict_utf8])
)}, )},
{"noncharacter u+fffe replaced", ?_assertEqual( {"noncharacter u+fffe replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
decode_bad_utf(<<239, 191, 190>>, [replaced_bad_utf8]) decode_bad_utf(<<239, 191, 190>>, [])
)}, )},
{"noncharacter u+ffff", ?_assertError( {"noncharacter u+ffff", ?_assertError(
badarg, badarg,
decode_bad_utf(<<239, 191, 191>>, []) decode_bad_utf(<<239, 191, 191>>, [strict_utf8])
)}, )},
{"noncharacter u+ffff replaced", ?_assertEqual( {"noncharacter u+ffff replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
decode_bad_utf(<<239, 191, 191>>, [replaced_bad_utf8]) decode_bad_utf(<<239, 191, 191>>, [])
)}, )},
{"orphan continuation byte u+0080", ?_assertError( {"orphan continuation byte u+0080", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#0080>>, []) decode_bad_utf(<<16#0080>>, [strict_utf8])
)}, )},
{"orphan continuation byte u+0080 replaced", ?_assertEqual( {"orphan continuation byte u+0080 replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
decode_bad_utf(<<16#0080>>, [replaced_bad_utf8]) decode_bad_utf(<<16#0080>>, [])
)}, )},
{"orphan continuation byte u+00bf", ?_assertError( {"orphan continuation byte u+00bf", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#00bf>>, []) decode_bad_utf(<<16#00bf>>, [strict_utf8])
)}, )},
{"orphan continuation byte u+00bf replaced", ?_assertEqual( {"orphan continuation byte u+00bf replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
decode_bad_utf(<<16#00bf>>, [replaced_bad_utf8]) decode_bad_utf(<<16#00bf>>, [])
)}, )},
{"2 continuation bytes", ?_assertError( {"2 continuation bytes", ?_assertError(
badarg, badarg,
decode_bad_utf(<<(binary:copy(<<16#0080>>, 2))/binary>>, []) decode_bad_utf(<<(binary:copy(<<16#0080>>, 2))/binary>>, [strict_utf8])
)}, )},
{"2 continuation bytes replaced", ?_assertEqual( {"2 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 2), binary:copy(<<16#fffd/utf8>>, 2),
decode_bad_utf(<<(binary:copy(<<16#0080>>, 2))/binary>>, [replaced_bad_utf8]) decode_bad_utf(<<(binary:copy(<<16#0080>>, 2))/binary>>, [])
)}, )},
{"3 continuation bytes", ?_assertError( {"3 continuation bytes", ?_assertError(
badarg, badarg,
decode_bad_utf(<<(binary:copy(<<16#0080>>, 3))/binary>>, []) decode_bad_utf(<<(binary:copy(<<16#0080>>, 3))/binary>>, [strict_utf8])
)}, )},
{"3 continuation bytes replaced", ?_assertEqual( {"3 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 3), binary:copy(<<16#fffd/utf8>>, 3),
decode_bad_utf(<<(binary:copy(<<16#0080>>, 3))/binary>>, [replaced_bad_utf8]) decode_bad_utf(<<(binary:copy(<<16#0080>>, 3))/binary>>, [])
)}, )},
{"4 continuation bytes", ?_assertError( {"4 continuation bytes", ?_assertError(
badarg, badarg,
decode_bad_utf(<<(binary:copy(<<16#0080>>, 4))/binary>>, []) decode_bad_utf(<<(binary:copy(<<16#0080>>, 4))/binary>>, [strict_utf8])
)}, )},
{"4 continuation bytes replaced", ?_assertEqual( {"4 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 4), binary:copy(<<16#fffd/utf8>>, 4),
decode_bad_utf(<<(binary:copy(<<16#0080>>, 4))/binary>>, [replaced_bad_utf8]) decode_bad_utf(<<(binary:copy(<<16#0080>>, 4))/binary>>, [])
)}, )},
{"5 continuation bytes", ?_assertError( {"5 continuation bytes", ?_assertError(
badarg, badarg,
decode_bad_utf(<<(binary:copy(<<16#0080>>, 5))/binary>>, []) decode_bad_utf(<<(binary:copy(<<16#0080>>, 5))/binary>>, [strict_utf8])
)}, )},
{"5 continuation bytes replaced", ?_assertEqual( {"5 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 5), binary:copy(<<16#fffd/utf8>>, 5),
decode_bad_utf(<<(binary:copy(<<16#0080>>, 5))/binary>>, [replaced_bad_utf8]) decode_bad_utf(<<(binary:copy(<<16#0080>>, 5))/binary>>, [])
)}, )},
{"6 continuation bytes", ?_assertError( {"6 continuation bytes", ?_assertError(
badarg, badarg,
decode_bad_utf(<<(binary:copy(<<16#0080>>, 6))/binary>>, []) decode_bad_utf(<<(binary:copy(<<16#0080>>, 6))/binary>>, [strict_utf8])
)}, )},
{"6 continuation bytes replaced", ?_assertEqual( {"6 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 6), binary:copy(<<16#fffd/utf8>>, 6),
decode_bad_utf(<<(binary:copy(<<16#0080>>, 6))/binary>>, [replaced_bad_utf8]) decode_bad_utf(<<(binary:copy(<<16#0080>>, 6))/binary>>, [])
)}, )},
{"all continuation bytes", ?_assertError( {"all continuation bytes", ?_assertError(
badarg, badarg,
decode_bad_utf(<<(list_to_binary(lists:seq(16#0080, 16#00bf)))/binary>>, []) decode_bad_utf(<<(list_to_binary(lists:seq(16#0080, 16#00bf)))/binary>>, [strict_utf8])
)}, )},
{"all continuation bytes replaced", ?_assertEqual( {"all continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, length(lists:seq(16#0080, 16#00bf))), binary:copy(<<16#fffd/utf8>>, length(lists:seq(16#0080, 16#00bf))),
decode_bad_utf( decode_bad_utf(
<<(list_to_binary(lists:seq(16#0080, 16#00bf)))/binary>>, <<(list_to_binary(lists:seq(16#0080, 16#00bf)))/binary>>,
[replaced_bad_utf8] []
) )
)}, )},
{"lonely start byte", ?_assertError( {"lonely start byte", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#00c0>>, []) decode_bad_utf(<<16#00c0>>, [strict_utf8])
)}, )},
{"lonely start byte replaced", ?_assertEqual( {"lonely start byte replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
decode_bad_utf(<<16#00c0>>, [replaced_bad_utf8]) decode_bad_utf(<<16#00c0>>, [])
)}, )},
{"lonely start bytes (2 byte)", ?_assertError( {"lonely start bytes (2 byte)", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#00c0, 32, 16#00df>>, []) decode_bad_utf(<<16#00c0, 32, 16#00df>>, [strict_utf8])
)}, )},
{"lonely start bytes (2 byte) replaced", ?_assertEqual( {"lonely start bytes (2 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32, 16#fffd/utf8>>, <<16#fffd/utf8, 32, 16#fffd/utf8>>,
decode_bad_utf(<<16#00c0, 32, 16#00df>>, [replaced_bad_utf8]) decode_bad_utf(<<16#00c0, 32, 16#00df>>, [])
)}, )},
{"lonely start bytes (3 byte)", ?_assertError( {"lonely start bytes (3 byte)", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#00e0, 32, 16#00ef>>, []) decode_bad_utf(<<16#00e0, 32, 16#00ef>>, [strict_utf8])
)}, )},
{"lonely start bytes (3 byte) replaced", ?_assertEqual( {"lonely start bytes (3 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32, 16#fffd/utf8>>, <<16#fffd/utf8, 32, 16#fffd/utf8>>,
decode_bad_utf(<<16#00e0, 32, 16#00ef>>, [replaced_bad_utf8]) decode_bad_utf(<<16#00e0, 32, 16#00ef>>, [])
)}, )},
{"lonely start bytes (4 byte)", ?_assertError( {"lonely start bytes (4 byte)", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#00f0, 32, 16#00f7>>, []) decode_bad_utf(<<16#00f0, 32, 16#00f7>>, [strict_utf8])
)}, )},
{"lonely start bytes (4 byte) replaced", ?_assertEqual( {"lonely start bytes (4 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32, 16#fffd/utf8>>, <<16#fffd/utf8, 32, 16#fffd/utf8>>,
decode_bad_utf(<<16#00f0, 32, 16#00f7>>, [replaced_bad_utf8]) decode_bad_utf(<<16#00f0, 32, 16#00f7>>, [])
)}, )},
{"missing continuation byte (3 byte)", ?_assertError( {"missing continuation byte (3 byte)", ?_assertError(
badarg, badarg,
decode_bad_utf(<<224, 160, 32>>, []) decode_bad_utf(<<224, 160, 32>>, [strict_utf8])
)}, )},
{"missing continuation byte (3 byte) replaced", ?_assertEqual( {"missing continuation byte (3 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
decode_bad_utf(<<224, 160, 32>>, [replaced_bad_utf8]) decode_bad_utf(<<224, 160, 32>>, [])
)}, )},
{"missing continuation byte (4 byte missing one)", ?_assertError( {"missing continuation byte (4 byte missing one)", ?_assertError(
badarg, badarg,
decode_bad_utf(<<240, 144, 128, 32>>, []) decode_bad_utf(<<240, 144, 128, 32>>, [strict_utf8])
)}, )},
{"missing continuation byte (4 byte missing one) replaced", ?_assertEqual( {"missing continuation byte (4 byte missing one) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
decode_bad_utf(<<240, 144, 128, 32>>, [replaced_bad_utf8]) decode_bad_utf(<<240, 144, 128, 32>>, [])
)}, )},
{"missing continuation byte (4 byte missing two)", ?_assertError( {"missing continuation byte (4 byte missing two)", ?_assertError(
badarg, badarg,
decode_bad_utf(<<240, 144, 32>>, []) decode_bad_utf(<<240, 144, 32>>, [strict_utf8])
)}, )},
{"missing continuation byte (4 byte missing two) replaced", ?_assertEqual( {"missing continuation byte (4 byte missing two) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
decode_bad_utf(<<240, 144, 32>>, [replaced_bad_utf8]) decode_bad_utf(<<240, 144, 32>>, [])
)}, )},
{"overlong encoding of u+002f (2 byte)", ?_assertError( {"overlong encoding of u+002f (2 byte)", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#c0, 16#af, 32>>, []) decode_bad_utf(<<16#c0, 16#af, 32>>, [strict_utf8])
)}, )},
{"overlong encoding of u+002f (2 byte) replaced", ?_assertEqual( {"overlong encoding of u+002f (2 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
decode_bad_utf(<<16#c0, 16#af, 32>>, [replaced_bad_utf8]) decode_bad_utf(<<16#c0, 16#af, 32>>, [])
)}, )},
{"overlong encoding of u+002f (3 byte)", ?_assertError( {"overlong encoding of u+002f (3 byte)", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#e0, 16#80, 16#af, 32>>, []) decode_bad_utf(<<16#e0, 16#80, 16#af, 32>>, [strict_utf8])
)}, )},
{"overlong encoding of u+002f (3 byte) replaced", ?_assertEqual( {"overlong encoding of u+002f (3 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
decode_bad_utf(<<16#e0, 16#80, 16#af, 32>>, [replaced_bad_utf8]) decode_bad_utf(<<16#e0, 16#80, 16#af, 32>>, [])
)}, )},
{"overlong encoding of u+002f (4 byte)", ?_assertError( {"overlong encoding of u+002f (4 byte)", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#f0, 16#80, 16#80, 16#af, 32>>, []) decode_bad_utf(<<16#f0, 16#80, 16#80, 16#af, 32>>, [strict_utf8])
)}, )},
{"overlong encoding of u+002f (4 byte) replaced", ?_assertEqual( {"overlong encoding of u+002f (4 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
decode_bad_utf(<<16#f0, 16#80, 16#80, 16#af, 32>>, [replaced_bad_utf8]) decode_bad_utf(<<16#f0, 16#80, 16#80, 16#af, 32>>, [])
)}, )},
{"highest overlong 2 byte sequence", ?_assertError( {"highest overlong 2 byte sequence", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#c1, 16#bf, 32>>, []) decode_bad_utf(<<16#c1, 16#bf, 32>>, [strict_utf8])
)}, )},
{"highest overlong 2 byte sequence replaced", ?_assertEqual( {"highest overlong 2 byte sequence replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
decode_bad_utf(<<16#c1, 16#bf, 32>>, [replaced_bad_utf8]) decode_bad_utf(<<16#c1, 16#bf, 32>>, [])
)}, )},
{"highest overlong 3 byte sequence", ?_assertError( {"highest overlong 3 byte sequence", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#e0, 16#9f, 16#bf, 32>>, []) decode_bad_utf(<<16#e0, 16#9f, 16#bf, 32>>, [strict_utf8])
)}, )},
{"highest overlong 3 byte sequence replaced", ?_assertEqual( {"highest overlong 3 byte sequence replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
decode_bad_utf(<<16#e0, 16#9f, 16#bf, 32>>, [replaced_bad_utf8]) decode_bad_utf(<<16#e0, 16#9f, 16#bf, 32>>, [])
)}, )},
{"highest overlong 4 byte sequence", ?_assertError( {"highest overlong 4 byte sequence", ?_assertError(
badarg, badarg,
decode_bad_utf(<<16#f0, 16#8f, 16#bf, 16#bf, 32>>, []) decode_bad_utf(<<16#f0, 16#8f, 16#bf, 16#bf, 32>>, [strict_utf8])
)}, )},
{"highest overlong 4 byte sequence replaced", ?_assertEqual( {"highest overlong 4 byte sequence replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
decode_bad_utf(<<16#f0, 16#8f, 16#bf, 16#bf, 32>>, [replaced_bad_utf8]) decode_bad_utf(<<16#f0, 16#8f, 16#bf, 16#bf, 32>>, [])
)} )}
]. ].
@ -1647,35 +1646,35 @@ unescape_test_() ->
)}, )},
{"replace bad high surrogate", ?_assertEqual( {"replace bad high surrogate", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
unescape(<<"\\udc00"/utf8>>, [replaced_bad_utf8]) unescape(<<"\\udc00"/utf8>>, [])
)}, )},
{"do not unescape bad high surrogate", ?_assertError( {"do not unescape bad high surrogate", ?_assertError(
badarg, badarg,
unescape(<<"\\udc00"/utf8>>, []) unescape(<<"\\udc00"/utf8>>, [strict_utf8])
)}, )},
{"replace naked high surrogate", ?_assertEqual( {"replace naked high surrogate", ?_assertEqual(
<<16#fffd/utf8, "hello world">>, <<16#fffd/utf8, "hello world">>,
unescape(<<"\\ud800hello world"/utf8>>, [replaced_bad_utf8]) unescape(<<"\\ud800hello world"/utf8>>, [])
)}, )},
{"do not unescape naked high surrogate", ?_assertError( {"do not unescape naked high surrogate", ?_assertError(
badarg, badarg,
unescape(<<"\\ud800hello world"/utf8>>, []) unescape(<<"\\ud800hello world"/utf8>>, [strict_utf8])
)}, )},
{"replace naked low surrogate", ?_assertEqual( {"replace naked low surrogate", ?_assertEqual(
<<16#fffd/utf8, "hello world">>, <<16#fffd/utf8, "hello world">>,
unescape(<<"\\udc00hello world"/utf8>>, [replaced_bad_utf8]) unescape(<<"\\udc00hello world"/utf8>>, [])
)}, )},
{"do not unescape naked low surrogate", ?_assertError( {"do not unescape naked low surrogate", ?_assertError(
badarg, badarg,
unescape(<<"\\udc00hello world"/utf8>>, []) unescape(<<"\\udc00hello world"/utf8>>, [strict_utf8])
)}, )},
{"replace bad surrogate pair", ?_assertEqual( {"replace bad surrogate pair", ?_assertEqual(
<<16#fffd/utf8, 16#fffd/utf8>>, <<16#fffd/utf8, 16#fffd/utf8>>,
unescape(<<"\\ud800\\u0000">>, [replaced_bad_utf8]) unescape(<<"\\ud800\\u0000">>, [])
)}, )},
{"do not unescape bad surrogate pair", ?_assertError( {"do not unescape bad surrogate pair", ?_assertError(
badarg, badarg,
unescape(<<"\\ud800\\u0000">>, []) unescape(<<"\\ud800\\u0000">>, [strict_utf8])
)}, )},
{"bad pseudo escape sequence", ?_assertError( {"bad pseudo escape sequence", ?_assertError(
badarg, badarg,
@ -2104,11 +2103,11 @@ custom_error_handler_test_() ->
)}, )},
{"single_comment error", ?_assertEqual( {"single_comment error", ?_assertEqual(
{comment, <<192>>}, {comment, <<192>>},
Decode(<<"[ //"/utf8, 192>>, [{error_handler, Error}, comments]) Decode(<<"[ //"/utf8, 192>>, [{error_handler, Error}, comments, strict_utf8])
)}, )},
{"multi_comment error", ?_assertEqual( {"multi_comment error", ?_assertEqual(
{comment, <<192>>}, {comment, <<192>>},
Decode(<<"[ /*"/utf8, 192>>, [{error_handler, Error}, comments]) Decode(<<"[ /*"/utf8, 192>>, [{error_handler, Error}, comments, strict_utf8])
)} )}
]. ].

4
src/jsx_encoder.erl
View file

@ -78,7 +78,7 @@ err(Term, Opts) -> (jsx:parser(jsx, [], Opts))(Term).
error_test_() -> error_test_() ->
[ [
{"value error", ?_assertError(badarg, err(self(), []))}, {"value error", ?_assertError(badarg, err(self(), []))},
{"string error", ?_assertError(badarg, err(<<239, 191, 191>>, []))} {"string error", ?_assertError(badarg, err(<<239, 191, 191>>, [strict_utf8]))}
]. ].
custom_error_handler_test_() -> custom_error_handler_test_() ->
@ -90,7 +90,7 @@ custom_error_handler_test_() ->
)}, )},
{"string error", ?_assertEqual( {"string error", ?_assertEqual(
{string, [{string, <<239, 191, 191>>}]}, {string, [{string, <<239, 191, 191>>}]},
err(<<239, 191, 191>>, [{error_handler, Error}]) err(<<239, 191, 191>>, [{error_handler, Error}, strict_utf8])
)} )}
]. ].

301
src/jsx_parser.erl
View file

@ -193,209 +193,46 @@ clean_string(Bin, Tokens, Handler, Stack, Config) ->
end. end.
clean_string(Bin, #config{dirty_strings=true}) -> Bin; clean_string(Bin, #config{dirty_strings=true}) -> Bin;
clean_string(Bin, Config) -> clean_string(Bin, Config) -> clean(Bin, [], Config).
case Config#config.replaced_bad_utf8 orelse Config#config.escaped_strings of
true -> clean(Bin, [], Config);
false -> ensure_clean(Bin)
end.
ensure_clean(Bin) ->
case is_clean(Bin) of
ok -> Bin;
{error, badarg} -> {error, badarg}
end.
%% fast path for no escaping and no correcting, throws error if string is 'bad'
is_clean(<<>>) -> ok;
is_clean(<<0, Rest/binary>>) -> is_clean(Rest);
is_clean(<<1, Rest/binary>>) -> is_clean(Rest);
is_clean(<<2, Rest/binary>>) -> is_clean(Rest);
is_clean(<<3, Rest/binary>>) -> is_clean(Rest);
is_clean(<<4, Rest/binary>>) -> is_clean(Rest);
is_clean(<<5, Rest/binary>>) -> is_clean(Rest);
is_clean(<<6, Rest/binary>>) -> is_clean(Rest);
is_clean(<<7, Rest/binary>>) -> is_clean(Rest);
is_clean(<<8, Rest/binary>>) -> is_clean(Rest);
is_clean(<<9, Rest/binary>>) -> is_clean(Rest);
is_clean(<<10, Rest/binary>>) -> is_clean(Rest);
is_clean(<<11, Rest/binary>>) -> is_clean(Rest);
is_clean(<<12, Rest/binary>>) -> is_clean(Rest);
is_clean(<<13, Rest/binary>>) -> is_clean(Rest);
is_clean(<<14, Rest/binary>>) -> is_clean(Rest);
is_clean(<<15, Rest/binary>>) -> is_clean(Rest);
is_clean(<<16, Rest/binary>>) -> is_clean(Rest);
is_clean(<<17, Rest/binary>>) -> is_clean(Rest);
is_clean(<<18, Rest/binary>>) -> is_clean(Rest);
is_clean(<<19, Rest/binary>>) -> is_clean(Rest);
is_clean(<<20, Rest/binary>>) -> is_clean(Rest);
is_clean(<<21, Rest/binary>>) -> is_clean(Rest);
is_clean(<<22, Rest/binary>>) -> is_clean(Rest);
is_clean(<<23, Rest/binary>>) -> is_clean(Rest);
is_clean(<<24, Rest/binary>>) -> is_clean(Rest);
is_clean(<<25, Rest/binary>>) -> is_clean(Rest);
is_clean(<<26, Rest/binary>>) -> is_clean(Rest);
is_clean(<<27, Rest/binary>>) -> is_clean(Rest);
is_clean(<<28, Rest/binary>>) -> is_clean(Rest);
is_clean(<<29, Rest/binary>>) -> is_clean(Rest);
is_clean(<<30, Rest/binary>>) -> is_clean(Rest);
is_clean(<<31, Rest/binary>>) -> is_clean(Rest);
is_clean(<<32, Rest/binary>>) -> is_clean(Rest);
is_clean(<<33, Rest/binary>>) -> is_clean(Rest);
is_clean(<<34, Rest/binary>>) -> is_clean(Rest);
is_clean(<<35, Rest/binary>>) -> is_clean(Rest);
is_clean(<<36, Rest/binary>>) -> is_clean(Rest);
is_clean(<<37, Rest/binary>>) -> is_clean(Rest);
is_clean(<<38, Rest/binary>>) -> is_clean(Rest);
is_clean(<<39, Rest/binary>>) -> is_clean(Rest);
is_clean(<<40, Rest/binary>>) -> is_clean(Rest);
is_clean(<<41, Rest/binary>>) -> is_clean(Rest);
is_clean(<<42, Rest/binary>>) -> is_clean(Rest);
is_clean(<<43, Rest/binary>>) -> is_clean(Rest);
is_clean(<<44, Rest/binary>>) -> is_clean(Rest);
is_clean(<<45, Rest/binary>>) -> is_clean(Rest);
is_clean(<<46, Rest/binary>>) -> is_clean(Rest);
is_clean(<<47, Rest/binary>>) -> is_clean(Rest);
is_clean(<<48, Rest/binary>>) -> is_clean(Rest);
is_clean(<<49, Rest/binary>>) -> is_clean(Rest);
is_clean(<<50, Rest/binary>>) -> is_clean(Rest);
is_clean(<<51, Rest/binary>>) -> is_clean(Rest);
is_clean(<<52, Rest/binary>>) -> is_clean(Rest);
is_clean(<<53, Rest/binary>>) -> is_clean(Rest);
is_clean(<<54, Rest/binary>>) -> is_clean(Rest);
is_clean(<<55, Rest/binary>>) -> is_clean(Rest);
is_clean(<<56, Rest/binary>>) -> is_clean(Rest);
is_clean(<<57, Rest/binary>>) -> is_clean(Rest);
is_clean(<<58, Rest/binary>>) -> is_clean(Rest);
is_clean(<<59, Rest/binary>>) -> is_clean(Rest);
is_clean(<<60, Rest/binary>>) -> is_clean(Rest);
is_clean(<<61, Rest/binary>>) -> is_clean(Rest);
is_clean(<<62, Rest/binary>>) -> is_clean(Rest);
is_clean(<<63, Rest/binary>>) -> is_clean(Rest);
is_clean(<<64, Rest/binary>>) -> is_clean(Rest);
is_clean(<<65, Rest/binary>>) -> is_clean(Rest);
is_clean(<<66, Rest/binary>>) -> is_clean(Rest);
is_clean(<<67, Rest/binary>>) -> is_clean(Rest);
is_clean(<<68, Rest/binary>>) -> is_clean(Rest);
is_clean(<<69, Rest/binary>>) -> is_clean(Rest);
is_clean(<<70, Rest/binary>>) -> is_clean(Rest);
is_clean(<<71, Rest/binary>>) -> is_clean(Rest);
is_clean(<<72, Rest/binary>>) -> is_clean(Rest);
is_clean(<<73, Rest/binary>>) -> is_clean(Rest);
is_clean(<<74, Rest/binary>>) -> is_clean(Rest);
is_clean(<<75, Rest/binary>>) -> is_clean(Rest);
is_clean(<<76, Rest/binary>>) -> is_clean(Rest);
is_clean(<<77, Rest/binary>>) -> is_clean(Rest);
is_clean(<<78, Rest/binary>>) -> is_clean(Rest);
is_clean(<<79, Rest/binary>>) -> is_clean(Rest);
is_clean(<<80, Rest/binary>>) -> is_clean(Rest);
is_clean(<<81, Rest/binary>>) -> is_clean(Rest);
is_clean(<<82, Rest/binary>>) -> is_clean(Rest);
is_clean(<<83, Rest/binary>>) -> is_clean(Rest);
is_clean(<<84, Rest/binary>>) -> is_clean(Rest);
is_clean(<<85, Rest/binary>>) -> is_clean(Rest);
is_clean(<<86, Rest/binary>>) -> is_clean(Rest);
is_clean(<<87, Rest/binary>>) -> is_clean(Rest);
is_clean(<<88, Rest/binary>>) -> is_clean(Rest);
is_clean(<<89, Rest/binary>>) -> is_clean(Rest);
is_clean(<<90, Rest/binary>>) -> is_clean(Rest);
is_clean(<<91, Rest/binary>>) -> is_clean(Rest);
is_clean(<<92, Rest/binary>>) -> is_clean(Rest);
is_clean(<<93, Rest/binary>>) -> is_clean(Rest);
is_clean(<<94, Rest/binary>>) -> is_clean(Rest);
is_clean(<<95, Rest/binary>>) -> is_clean(Rest);
is_clean(<<96, Rest/binary>>) -> is_clean(Rest);
is_clean(<<97, Rest/binary>>) -> is_clean(Rest);
is_clean(<<98, Rest/binary>>) -> is_clean(Rest);
is_clean(<<99, Rest/binary>>) -> is_clean(Rest);
is_clean(<<100, Rest/binary>>) -> is_clean(Rest);
is_clean(<<101, Rest/binary>>) -> is_clean(Rest);
is_clean(<<102, Rest/binary>>) -> is_clean(Rest);
is_clean(<<103, Rest/binary>>) -> is_clean(Rest);
is_clean(<<104, Rest/binary>>) -> is_clean(Rest);
is_clean(<<105, Rest/binary>>) -> is_clean(Rest);
is_clean(<<106, Rest/binary>>) -> is_clean(Rest);
is_clean(<<107, Rest/binary>>) -> is_clean(Rest);
is_clean(<<108, Rest/binary>>) -> is_clean(Rest);
is_clean(<<109, Rest/binary>>) -> is_clean(Rest);
is_clean(<<110, Rest/binary>>) -> is_clean(Rest);
is_clean(<<111, Rest/binary>>) -> is_clean(Rest);
is_clean(<<112, Rest/binary>>) -> is_clean(Rest);
is_clean(<<113, Rest/binary>>) -> is_clean(Rest);
is_clean(<<114, Rest/binary>>) -> is_clean(Rest);
is_clean(<<115, Rest/binary>>) -> is_clean(Rest);
is_clean(<<116, Rest/binary>>) -> is_clean(Rest);
is_clean(<<117, Rest/binary>>) -> is_clean(Rest);
is_clean(<<118, Rest/binary>>) -> is_clean(Rest);
is_clean(<<119, Rest/binary>>) -> is_clean(Rest);
is_clean(<<120, Rest/binary>>) -> is_clean(Rest);
is_clean(<<121, Rest/binary>>) -> is_clean(Rest);
is_clean(<<122, Rest/binary>>) -> is_clean(Rest);
is_clean(<<123, Rest/binary>>) -> is_clean(Rest);
is_clean(<<124, Rest/binary>>) -> is_clean(Rest);
is_clean(<<125, Rest/binary>>) -> is_clean(Rest);
is_clean(<<126, Rest/binary>>) -> is_clean(Rest);
is_clean(<<127, Rest/binary>>) -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X < 16#d800 -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X > 16#dfff, X < 16#fdd0 -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X > 16#fdef, X < 16#fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#10000, X < 16#1fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#20000, X < 16#2fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#30000, X < 16#3fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#40000, X < 16#4fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#50000, X < 16#5fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#60000, X < 16#6fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#70000, X < 16#7fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#80000, X < 16#8fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#90000, X < 16#9fffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#a0000, X < 16#afffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#b0000, X < 16#bfffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#c0000, X < 16#cfffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#d0000, X < 16#dfffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#e0000, X < 16#efffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#f0000, X < 16#ffffe -> is_clean(Rest);
is_clean(<<X/utf8, Rest/binary>>) when X >= 16#100000, X < 16#10fffe -> is_clean(Rest);
is_clean(_Bin) -> {error, badarg}.
%% escape and/or replace bad codepoints if requested %% escape and/or replace bad codepoints if requested
clean(<<>>, Acc, _Config) -> unicode:characters_to_binary(lists:reverse(Acc)); clean(<<>>, Acc, _Config) -> unicode:characters_to_binary(lists:reverse(Acc));
clean(<<0, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(0, Config) ++ Acc, Config); clean(<<0, Rest/binary>>, Acc, Config) -> maybe_replace(0, Rest, Acc, Config);
clean(<<1, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(1, Config) ++ Acc, Config); clean(<<1, Rest/binary>>, Acc, Config) -> maybe_replace(1, Rest, Acc, Config);
clean(<<2, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(2, Config) ++ Acc, Config); clean(<<2, Rest/binary>>, Acc, Config) -> maybe_replace(2, Rest, Acc, Config);
clean(<<3, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(3, Config) ++ Acc, Config); clean(<<3, Rest/binary>>, Acc, Config) -> maybe_replace(3, Rest, Acc, Config);
clean(<<4, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(4, Config) ++ Acc, Config); clean(<<4, Rest/binary>>, Acc, Config) -> maybe_replace(4, Rest, Acc, Config);
clean(<<5, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(5, Config) ++ Acc, Config); clean(<<5, Rest/binary>>, Acc, Config) -> maybe_replace(5, Rest, Acc, Config);
clean(<<6, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(6, Config) ++ Acc, Config); clean(<<6, Rest/binary>>, Acc, Config) -> maybe_replace(6, Rest, Acc, Config);
clean(<<7, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(7, Config) ++ Acc, Config); clean(<<7, Rest/binary>>, Acc, Config) -> maybe_replace(7, Rest, Acc, Config);
clean(<<8, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(8, Config) ++ Acc, Config); clean(<<8, Rest/binary>>, Acc, Config) -> maybe_replace(8, Rest, Acc, Config);
clean(<<9, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(9, Config) ++ Acc, Config); clean(<<9, Rest/binary>>, Acc, Config) -> maybe_replace(9, Rest, Acc, Config);
clean(<<10, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(10, Config) ++ Acc, Config); clean(<<10, Rest/binary>>, Acc, Config) -> maybe_replace(10, Rest, Acc, Config);
clean(<<11, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(11, Config) ++ Acc, Config); clean(<<11, Rest/binary>>, Acc, Config) -> maybe_replace(11, Rest, Acc, Config);
clean(<<12, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(12, Config) ++ Acc, Config); clean(<<12, Rest/binary>>, Acc, Config) -> maybe_replace(12, Rest, Acc, Config);
clean(<<13, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(13, Config) ++ Acc, Config); clean(<<13, Rest/binary>>, Acc, Config) -> maybe_replace(13, Rest, Acc, Config);
clean(<<14, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(14, Config) ++ Acc, Config); clean(<<14, Rest/binary>>, Acc, Config) -> maybe_replace(14, Rest, Acc, Config);
clean(<<15, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(15, Config) ++ Acc, Config); clean(<<15, Rest/binary>>, Acc, Config) -> maybe_replace(15, Rest, Acc, Config);
clean(<<16, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(16, Config) ++ Acc, Config); clean(<<16, Rest/binary>>, Acc, Config) -> maybe_replace(16, Rest, Acc, Config);
clean(<<17, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(17, Config) ++ Acc, Config); clean(<<17, Rest/binary>>, Acc, Config) -> maybe_replace(17, Rest, Acc, Config);
clean(<<18, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(18, Config) ++ Acc, Config); clean(<<18, Rest/binary>>, Acc, Config) -> maybe_replace(18, Rest, Acc, Config);
clean(<<19, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(19, Config) ++ Acc, Config); clean(<<19, Rest/binary>>, Acc, Config) -> maybe_replace(19, Rest, Acc, Config);
clean(<<20, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(20, Config) ++ Acc, Config); clean(<<20, Rest/binary>>, Acc, Config) -> maybe_replace(20, Rest, Acc, Config);
clean(<<21, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(21, Config) ++ Acc, Config); clean(<<21, Rest/binary>>, Acc, Config) -> maybe_replace(21, Rest, Acc, Config);
clean(<<22, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(22, Config) ++ Acc, Config); clean(<<22, Rest/binary>>, Acc, Config) -> maybe_replace(22, Rest, Acc, Config);
clean(<<23, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(23, Config) ++ Acc, Config); clean(<<23, Rest/binary>>, Acc, Config) -> maybe_replace(23, Rest, Acc, Config);
clean(<<24, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(24, Config) ++ Acc, Config); clean(<<24, Rest/binary>>, Acc, Config) -> maybe_replace(24, Rest, Acc, Config);
clean(<<25, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(25, Config) ++ Acc, Config); clean(<<25, Rest/binary>>, Acc, Config) -> maybe_replace(25, Rest, Acc, Config);
clean(<<26, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(26, Config) ++ Acc, Config); clean(<<26, Rest/binary>>, Acc, Config) -> maybe_replace(26, Rest, Acc, Config);
clean(<<27, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(27, Config) ++ Acc, Config); clean(<<27, Rest/binary>>, Acc, Config) -> maybe_replace(27, Rest, Acc, Config);
clean(<<28, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(28, Config) ++ Acc, Config); clean(<<28, Rest/binary>>, Acc, Config) -> maybe_replace(28, Rest, Acc, Config);
clean(<<29, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(29, Config) ++ Acc, Config); clean(<<29, Rest/binary>>, Acc, Config) -> maybe_replace(29, Rest, Acc, Config);
clean(<<30, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(30, Config) ++ Acc, Config); clean(<<30, Rest/binary>>, Acc, Config) -> maybe_replace(30, Rest, Acc, Config);
clean(<<31, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(31, Config) ++ Acc, Config); clean(<<31, Rest/binary>>, Acc, Config) -> maybe_replace(31, Rest, Acc, Config);
clean(<<32, Rest/binary>>, Acc, Config) -> clean(Rest, [32] ++ Acc, Config); clean(<<32, Rest/binary>>, Acc, Config) -> clean(Rest, [32] ++ Acc, Config);
clean(<<33, Rest/binary>>, Acc, Config) -> clean(Rest, [33] ++ Acc, Config); clean(<<33, Rest/binary>>, Acc, Config) -> clean(Rest, [33] ++ Acc, Config);
clean(<<34, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(34, Config) ++ Acc, Config); clean(<<34, Rest/binary>>, Acc, Config) -> maybe_replace(34, Rest, Acc, Config);
clean(<<35, Rest/binary>>, Acc, Config) -> clean(Rest, [35] ++ Acc, Config); clean(<<35, Rest/binary>>, Acc, Config) -> clean(Rest, [35] ++ Acc, Config);
clean(<<36, Rest/binary>>, Acc, Config) -> clean(Rest, [36] ++ Acc, Config); clean(<<36, Rest/binary>>, Acc, Config) -> clean(Rest, [36] ++ Acc, Config);
clean(<<37, Rest/binary>>, Acc, Config) -> clean(Rest, [37] ++ Acc, Config); clean(<<37, Rest/binary>>, Acc, Config) -> clean(Rest, [37] ++ Acc, Config);
@ -408,7 +245,7 @@ clean(<<43, Rest/binary>>, Acc, Config) -> clean(Rest, [43] ++ Acc, Config);
clean(<<44, Rest/binary>>, Acc, Config) -> clean(Rest, [44] ++ Acc, Config); clean(<<44, Rest/binary>>, Acc, Config) -> clean(Rest, [44] ++ Acc, Config);
clean(<<45, Rest/binary>>, Acc, Config) -> clean(Rest, [45] ++ Acc, Config); clean(<<45, Rest/binary>>, Acc, Config) -> clean(Rest, [45] ++ Acc, Config);
clean(<<46, Rest/binary>>, Acc, Config) -> clean(Rest, [46] ++ Acc, Config); clean(<<46, Rest/binary>>, Acc, Config) -> clean(Rest, [46] ++ Acc, Config);
clean(<<47, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(47, Config) ++ Acc, Config); clean(<<47, Rest/binary>>, Acc, Config) -> maybe_replace(47, Rest, Acc, Config);
clean(<<48, Rest/binary>>, Acc, Config) -> clean(Rest, [48] ++ Acc, Config); clean(<<48, Rest/binary>>, Acc, Config) -> clean(Rest, [48] ++ Acc, Config);
clean(<<49, Rest/binary>>, Acc, Config) -> clean(Rest, [49] ++ Acc, Config); clean(<<49, Rest/binary>>, Acc, Config) -> clean(Rest, [49] ++ Acc, Config);
clean(<<50, Rest/binary>>, Acc, Config) -> clean(Rest, [50] ++ Acc, Config); clean(<<50, Rest/binary>>, Acc, Config) -> clean(Rest, [50] ++ Acc, Config);
@ -453,7 +290,7 @@ clean(<<88, Rest/binary>>, Acc, Config) -> clean(Rest, [88] ++ Acc, Config);
clean(<<89, Rest/binary>>, Acc, Config) -> clean(Rest, [89] ++ Acc, Config); clean(<<89, Rest/binary>>, Acc, Config) -> clean(Rest, [89] ++ Acc, Config);
clean(<<90, Rest/binary>>, Acc, Config) -> clean(Rest, [90] ++ Acc, Config); clean(<<90, Rest/binary>>, Acc, Config) -> clean(Rest, [90] ++ Acc, Config);
clean(<<91, Rest/binary>>, Acc, Config) -> clean(Rest, [91] ++ Acc, Config); clean(<<91, Rest/binary>>, Acc, Config) -> clean(Rest, [91] ++ Acc, Config);
clean(<<92, Rest/binary>>, Acc, Config) -> clean(Rest, maybe_replace(92, Config) ++ Acc, Config); clean(<<92, Rest/binary>>, Acc, Config) -> maybe_replace(92, Rest, Acc, Config);
clean(<<93, Rest/binary>>, Acc, Config) -> clean(Rest, [93] ++ Acc, Config); clean(<<93, Rest/binary>>, Acc, Config) -> clean(Rest, [93] ++ Acc, Config);
clean(<<94, Rest/binary>>, Acc, Config) -> clean(Rest, [94] ++ Acc, Config); clean(<<94, Rest/binary>>, Acc, Config) -> clean(Rest, [94] ++ Acc, Config);
clean(<<95, Rest/binary>>, Acc, Config) -> clean(Rest, [95] ++ Acc, Config); clean(<<95, Rest/binary>>, Acc, Config) -> clean(Rest, [95] ++ Acc, Config);
@ -490,7 +327,7 @@ clean(<<125, Rest/binary>>, Acc, Config) -> clean(Rest, [125] ++ Acc, Config);
clean(<<126, Rest/binary>>, Acc, Config) -> clean(Rest, [126] ++ Acc, Config); clean(<<126, Rest/binary>>, Acc, Config) -> clean(Rest, [126] ++ Acc, Config);
clean(<<127, Rest/binary>>, Acc, Config) -> clean(Rest, [127] ++ Acc, Config); clean(<<127, Rest/binary>>, Acc, Config) -> clean(Rest, [127] ++ Acc, Config);
clean(<<X/utf8, Rest/binary>>, Acc, Config) when X == 16#2028; X == 16#2029 -> clean(<<X/utf8, Rest/binary>>, Acc, Config) when X == 16#2028; X == 16#2029 ->
clean(Rest, maybe_replace(X, Config) ++ Acc, Config); maybe_replace(X, Rest, Acc, Config);
clean(<<X/utf8, Rest/binary>>, Acc, Config) when X < 16#d800 -> clean(<<X/utf8, Rest/binary>>, Acc, Config) when X < 16#d800 ->
clean(Rest, [X] ++ Acc, Config); clean(Rest, [X] ++ Acc, Config);
clean(<<X/utf8, Rest/binary>>, Acc, Config) when X > 16#dfff, X < 16#fdd0 -> clean(<<X/utf8, Rest/binary>>, Acc, Config) when X > 16#dfff, X < 16#fdd0 ->
@ -531,24 +368,24 @@ clean(<<X/utf8, Rest/binary>>, Acc, Config) when X >= 16#100000, X < 16#10fffe -
clean(Rest, [X] ++ Acc, Config); clean(Rest, [X] ++ Acc, Config);
%% surrogates %% surrogates
clean(<<237, X, _, Rest/binary>>, Acc, Config) when X >= 160 -> clean(<<237, X, _, Rest/binary>>, Acc, Config) when X >= 160 ->
clean(Rest, maybe_replace(surrogate, Config) ++ Acc, Config); maybe_replace(surrogate, Rest, Acc, Config);
%% noncharacters %% noncharacters
clean(<<_/utf8, Rest/binary>>, Acc, Config) -> clean(<<_/utf8, Rest/binary>>, Acc, Config) ->
clean(Rest, maybe_replace(noncharacter, Config) ++ Acc, Config); maybe_replace(noncharacter, Rest, Acc, Config);
%% u+fffe and u+ffff for R14BXX %% u+fffe and u+ffff for R14BXX
clean(<<239, 191, X, Rest/binary>>, Acc, Config) when X == 190; X == 191 -> clean(<<239, 191, X, Rest/binary>>, Acc, Config) when X == 190; X == 191 ->
clean(Rest, maybe_replace(noncharacter, Config) ++ Acc, Config); maybe_replace(noncharacter, Rest, Acc, Config);
%% overlong encodings and missing continuations of a 2 byte sequence %% overlong encodings and missing continuations of a 2 byte sequence
clean(<<X, Rest/binary>>, Acc, Config) when X >= 192, X =< 223 -> clean(<<X, Rest/binary>>, Acc, Config) when X >= 192, X =< 223 ->
clean(strip_continuations(Rest, 1), maybe_replace(badutf, Config) ++ Acc, Config); maybe_replace(badutf, strip_continuations(Rest, 1), Acc, Config);
%% overlong encodings and missing continuations of a 3 byte sequence %% overlong encodings and missing continuations of a 3 byte sequence
clean(<<X, Rest/binary>>, Acc, Config) when X >= 224, X =< 239 -> clean(<<X, Rest/binary>>, Acc, Config) when X >= 224, X =< 239 ->
clean(strip_continuations(Rest, 2), maybe_replace(badutf, Config) ++ Acc, Config); maybe_replace(badutf, strip_continuations(Rest, 2), Acc, Config);
%% overlong encodings and missing continuations of a 4 byte sequence %% overlong encodings and missing continuations of a 4 byte sequence
clean(<<X, Rest/binary>>, Acc, Config) when X >= 240, X =< 247 -> clean(<<X, Rest/binary>>, Acc, Config) when X >= 240, X =< 247 ->
clean(strip_continuations(Rest, 3), maybe_replace(badutf, Config) ++ Acc, Config); maybe_replace(badutf, strip_continuations(Rest, 3), Acc, Config);
clean(<<_, Rest/binary>>, Acc, Config) -> clean(<<_, Rest/binary>>, Acc, Config) ->
clean(Rest, maybe_replace(badutf, Config) ++ Acc, Config). maybe_replace(badutf, Rest, Acc, Config).
strip_continuations(Bin, 0) -> Bin; strip_continuations(Bin, 0) -> Bin;
@ -558,29 +395,37 @@ strip_continuations(<<X, Rest/binary>>, N) when X >= 128, X =< 191 ->
strip_continuations(Bin, _) -> Bin. strip_continuations(Bin, _) -> Bin.
maybe_replace($\b, #config{escaped_strings=true}) -> [$b, $\\]; maybe_replace($\b, Rest, Acc, Config=#config{escaped_strings=true}) ->
maybe_replace($\t, #config{escaped_strings=true}) -> [$t, $\\]; clean(Rest, [$b, $\\] ++ Acc, Config);
maybe_replace($\n, #config{escaped_strings=true}) -> [$n, $\\]; maybe_replace($\t, Rest, Acc, Config=#config{escaped_strings=true}) ->
maybe_replace($\f, #config{escaped_strings=true}) -> [$f, $\\]; clean(Rest, [$t, $\\] ++ Acc, Config);
maybe_replace($\r, #config{escaped_strings=true}) -> [$r, $\\]; maybe_replace($\n, Rest, Acc, Config=#config{escaped_strings=true}) ->
maybe_replace($\", #config{escaped_strings=true}) -> [$\", $\\]; clean(Rest, [$n, $\\] ++ Acc, Config);
maybe_replace($/, Config=#config{escaped_strings=true}) -> maybe_replace($\f, Rest, Acc, Config=#config{escaped_strings=true}) ->
clean(Rest, [$f, $\\] ++ Acc, Config);
maybe_replace($\r, Rest, Acc, Config=#config{escaped_strings=true}) ->
clean(Rest, [$r, $\\] ++ Acc, Config);
maybe_replace($\", Rest, Acc, Config=#config{escaped_strings=true}) ->
clean(Rest, [$\", $\\] ++ Acc, Config);
maybe_replace($/, Rest, Acc, Config=#config{escaped_strings=true}) ->
case Config#config.escaped_forward_slashes of case Config#config.escaped_forward_slashes of
true -> [$/, $\\]; true -> clean(Rest, [$/, $\\] ++ Acc, Config);
false -> [$/] false -> clean(Rest, [$/] ++ Acc, Config)
end; end;
maybe_replace($\\, #config{escaped_strings=true}) -> [$\\, $\\]; maybe_replace($\\, Rest, Acc, Config=#config{escaped_strings=true}) ->
maybe_replace(X, Config=#config{escaped_strings=true}) when X == 16#2028; X == 16#2029 -> clean(Rest, [$\\, $\\] ++ Acc, Config);
maybe_replace(X, Rest, Acc, Config=#config{escaped_strings=true}) when X == 16#2028; X == 16#2029 ->
case Config#config.unescaped_jsonp of case Config#config.unescaped_jsonp of
true -> [X]; true -> clean(Rest, [X] ++ Acc, Config);
false -> lists:reverse(json_escape_sequence(X)) false -> clean(Rest, lists:reverse(json_escape_sequence(X)) ++ Acc, Config)
end; end;
maybe_replace(X, #config{escaped_strings=true}) when X < 32 -> maybe_replace(X, Rest, Acc, Config=#config{escaped_strings=true}) when X < 32 ->
lists:reverse(json_escape_sequence(X)); clean(Rest, lists:reverse(json_escape_sequence(X)) ++ Acc, Config);
maybe_replace(noncharacter, #config{replaced_bad_utf8=true}) -> [16#fffd]; maybe_replace(Atom, _, _, #config{strict_utf8=true}) when is_atom(Atom) -> {error, badarg};
maybe_replace(surrogate, #config{replaced_bad_utf8=true}) -> [16#fffd]; maybe_replace(noncharacter, Rest, Acc, Config) -> clean(Rest, [16#fffd] ++ Acc, Config);
maybe_replace(badutf, #config{replaced_bad_utf8=true}) -> [16#fffd]; maybe_replace(surrogate, Rest, Acc, Config) -> clean(Rest, [16#fffd] ++ Acc, Config);
maybe_replace(_, _) -> {error, badarg}. maybe_replace(badutf, Rest, Acc, Config) -> clean(Rest, [16#fffd] ++ Acc, Config);
maybe_replace(X, Rest, Acc, Config) -> clean(Rest, [X] ++ Acc, Config).
%% convert a codepoint to it's \uXXXX equiv. %% convert a codepoint to it's \uXXXX equiv.
@ -650,7 +495,7 @@ error_test_() ->
{"value error", ?_assertError(badarg, parse_error([self()], []))}, {"value error", ?_assertError(badarg, parse_error([self()], []))},
{"maybe_done error", ?_assertError(badarg, parse_error([start_array, end_array, start_array, end_json], []))}, {"maybe_done error", ?_assertError(badarg, parse_error([start_array, end_array, start_array, end_json], []))},
{"done error", ?_assertError(badarg, parse_error([{string, <<"">>}, {literal, true}, end_json], []))}, {"done error", ?_assertError(badarg, parse_error([{string, <<"">>}, {literal, true}, end_json], []))},
{"string error", ?_assertError(badarg, parse_error([{string, <<239, 191, 191>>}, end_json], []))} {"string error", ?_assertError(badarg, parse_error([{string, <<239, 191, 191>>}, end_json], [strict_utf8]))}
]. ].
@ -671,7 +516,7 @@ custom_error_handler_test_() ->
)}, )},
{"string error", ?_assertEqual( {"string error", ?_assertEqual(
{string, [{string, <<239, 191, 191>>}, end_json]}, {string, [{string, <<239, 191, 191>>}, end_json]},
parse_error([{string, <<239, 191, 191>>}, end_json], [{error_handler, Error}]) parse_error([{string, <<239, 191, 191>>}, end_json], [{error_handler, Error}, strict_utf8])
)} )}
]. ].

108
src/jsx_tests.hrl
View file

@ -284,35 +284,35 @@ clean_string_test_() ->
)}, )},
{"error reserved space", ?_assertEqual( {"error reserved space", ?_assertEqual(
lists:duplicate(length(reserved_space()), {error, badarg}), lists:duplicate(length(reserved_space()), {error, badarg}),
lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{}) end, reserved_space()) lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{strict_utf8=true}) end, reserved_space())
)}, )},
{"error surrogates", ?_assertEqual( {"error surrogates", ?_assertEqual(
lists:duplicate(length(surrogates()), {error, badarg}), lists:duplicate(length(surrogates()), {error, badarg}),
lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{}) end, surrogates()) lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{strict_utf8=true}) end, surrogates())
)}, )},
{"error noncharacters", ?_assertEqual( {"error noncharacters", ?_assertEqual(
lists:duplicate(length(noncharacters()), {error, badarg}), lists:duplicate(length(noncharacters()), {error, badarg}),
lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{}) end, noncharacters()) lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{strict_utf8=true}) end, noncharacters())
)}, )},
{"error extended noncharacters", ?_assertEqual( {"error extended noncharacters", ?_assertEqual(
lists:duplicate(length(extended_noncharacters()), {error, badarg}), lists:duplicate(length(extended_noncharacters()), {error, badarg}),
lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{}) end, extended_noncharacters()) lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{strict_utf8=true}) end, extended_noncharacters())
)}, )},
{"clean reserved space", ?_assertEqual( {"clean reserved space", ?_assertEqual(
lists:duplicate(length(reserved_space()), <<16#fffd/utf8>>), lists:duplicate(length(reserved_space()), <<16#fffd/utf8>>),
lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{replaced_bad_utf8=true}) end, reserved_space()) lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{}) end, reserved_space())
)}, )},
{"clean surrogates", ?_assertEqual( {"clean surrogates", ?_assertEqual(
lists:duplicate(length(surrogates()), <<16#fffd/utf8>>), lists:duplicate(length(surrogates()), <<16#fffd/utf8>>),
lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{replaced_bad_utf8=true}) end, surrogates()) lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{}) end, surrogates())
)}, )},
{"clean noncharacters", ?_assertEqual( {"clean noncharacters", ?_assertEqual(
lists:duplicate(length(noncharacters()), <<16#fffd/utf8>>), lists:duplicate(length(noncharacters()), <<16#fffd/utf8>>),
lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{replaced_bad_utf8=true}) end, noncharacters()) lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{}) end, noncharacters())
)}, )},
{"clean extended noncharacters", ?_assertEqual( {"clean extended noncharacters", ?_assertEqual(
lists:duplicate(length(extended_noncharacters()), <<16#fffd/utf8>>), lists:duplicate(length(extended_noncharacters()), <<16#fffd/utf8>>),
lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{replaced_bad_utf8=true}) end, extended_noncharacters()) lists:map(fun(Codepoint) -> jsx_parser:clean_string(Codepoint, #config{}) end, extended_noncharacters())
)} )}
]. ].
@ -492,190 +492,190 @@ bad_utf8_test_() ->
[ [
{"noncharacter u+fffe", ?_assertEqual( {"noncharacter u+fffe", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(to_fake_utf8(16#fffe), #config{}) jsx_parser:clean_string(to_fake_utf8(16#fffe), #config{strict_utf8=true})
)}, )},
{"noncharacter u+fffe replaced", ?_assertEqual( {"noncharacter u+fffe replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
jsx_parser:clean_string(to_fake_utf8(16#fffe), #config{replaced_bad_utf8=true}) jsx_parser:clean_string(to_fake_utf8(16#fffe), #config{})
)}, )},
{"noncharacter u+ffff", ?_assertEqual( {"noncharacter u+ffff", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(to_fake_utf8(16#ffff), #config{}) jsx_parser:clean_string(to_fake_utf8(16#ffff), #config{strict_utf8=true})
)}, )},
{"noncharacter u+ffff replaced", ?_assertEqual( {"noncharacter u+ffff replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
jsx_parser:clean_string(to_fake_utf8(16#ffff), #config{replaced_bad_utf8=true}) jsx_parser:clean_string(to_fake_utf8(16#ffff), #config{})
)}, )},
{"orphan continuation byte u+0080", ?_assertEqual( {"orphan continuation byte u+0080", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#0080>>, #config{}) jsx_parser:clean_string(<<16#0080>>, #config{strict_utf8=true})
)}, )},
{"orphan continuation byte u+0080 replaced", ?_assertEqual( {"orphan continuation byte u+0080 replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
jsx_parser:clean_string(<<16#0080>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#0080>>, #config{})
)}, )},
{"orphan continuation byte u+00bf", ?_assertEqual( {"orphan continuation byte u+00bf", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#00bf>>, #config{}) jsx_parser:clean_string(<<16#00bf>>, #config{strict_utf8=true})
)}, )},
{"orphan continuation byte u+00bf replaced", ?_assertEqual( {"orphan continuation byte u+00bf replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
jsx_parser:clean_string(<<16#00bf>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#00bf>>, #config{})
)}, )},
{"2 continuation bytes", ?_assertEqual( {"2 continuation bytes", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 2))/binary>>, #config{}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 2))/binary>>, #config{strict_utf8=true})
)}, )},
{"2 continuation bytes replaced", ?_assertEqual( {"2 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 2), binary:copy(<<16#fffd/utf8>>, 2),
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 2))/binary>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 2))/binary>>, #config{})
)}, )},
{"3 continuation bytes", ?_assertEqual( {"3 continuation bytes", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 3))/binary>>, #config{}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 3))/binary>>, #config{strict_utf8=true})
)}, )},
{"3 continuation bytes replaced", ?_assertEqual( {"3 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 3), binary:copy(<<16#fffd/utf8>>, 3),
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 3))/binary>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 3))/binary>>, #config{})
)}, )},
{"4 continuation bytes", ?_assertEqual( {"4 continuation bytes", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 4))/binary>>, #config{}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 4))/binary>>, #config{strict_utf8=true})
)}, )},
{"4 continuation bytes replaced", ?_assertEqual( {"4 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 4), binary:copy(<<16#fffd/utf8>>, 4),
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 4))/binary>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 4))/binary>>, #config{})
)}, )},
{"5 continuation bytes", ?_assertEqual( {"5 continuation bytes", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 5))/binary>>, #config{}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 5))/binary>>, #config{strict_utf8=true})
)}, )},
{"5 continuation bytes replaced", ?_assertEqual( {"5 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 5), binary:copy(<<16#fffd/utf8>>, 5),
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 5))/binary>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 5))/binary>>, #config{})
)}, )},
{"6 continuation bytes", ?_assertEqual( {"6 continuation bytes", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 6))/binary>>, #config{}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 6))/binary>>, #config{strict_utf8=true})
)}, )},
{"6 continuation bytes replaced", ?_assertEqual( {"6 continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, 6), binary:copy(<<16#fffd/utf8>>, 6),
jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 6))/binary>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<(binary:copy(<<16#0080>>, 6))/binary>>, #config{})
)}, )},
{"all continuation bytes", ?_assertEqual( {"all continuation bytes", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<(list_to_binary(lists:seq(16#0080, 16#00bf)))/binary>>, #config{}) jsx_parser:clean_string(<<(list_to_binary(lists:seq(16#0080, 16#00bf)))/binary>>, #config{strict_utf8=true})
)}, )},
{"all continuation bytes replaced", ?_assertEqual( {"all continuation bytes replaced", ?_assertEqual(
binary:copy(<<16#fffd/utf8>>, length(lists:seq(16#0080, 16#00bf))), binary:copy(<<16#fffd/utf8>>, length(lists:seq(16#0080, 16#00bf))),
jsx_parser:clean_string( jsx_parser:clean_string(
<<(list_to_binary(lists:seq(16#0080, 16#00bf)))/binary>>, <<(list_to_binary(lists:seq(16#0080, 16#00bf)))/binary>>,
#config{replaced_bad_utf8=true} #config{}
) )
)}, )},
{"lonely start byte", ?_assertEqual( {"lonely start byte", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#00c0>>, #config{}) jsx_parser:clean_string(<<16#00c0>>, #config{strict_utf8=true})
)}, )},
{"lonely start byte replaced", ?_assertEqual( {"lonely start byte replaced", ?_assertEqual(
<<16#fffd/utf8>>, <<16#fffd/utf8>>,
jsx_parser:clean_string(<<16#00c0>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#00c0>>, #config{})
)}, )},
{"lonely start bytes (2 byte)", ?_assertEqual( {"lonely start bytes (2 byte)", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#00c0, 32, 16#00df>>, #config{}) jsx_parser:clean_string(<<16#00c0, 32, 16#00df>>, #config{strict_utf8=true})
)}, )},
{"lonely start bytes (2 byte) replaced", ?_assertEqual( {"lonely start bytes (2 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32, 16#fffd/utf8>>, <<16#fffd/utf8, 32, 16#fffd/utf8>>,
jsx_parser:clean_string(<<16#00c0, 32, 16#00df>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#00c0, 32, 16#00df>>, #config{})
)}, )},
{"lonely start bytes (3 byte)", ?_assertEqual( {"lonely start bytes (3 byte)", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#00e0, 32, 16#00ef>>, #config{}) jsx_parser:clean_string(<<16#00e0, 32, 16#00ef>>, #config{strict_utf8=true})
)}, )},
{"lonely start bytes (3 byte) replaced", ?_assertEqual( {"lonely start bytes (3 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32, 16#fffd/utf8>>, <<16#fffd/utf8, 32, 16#fffd/utf8>>,
jsx_parser:clean_string(<<16#00e0, 32, 16#00ef>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#00e0, 32, 16#00ef>>, #config{})
)}, )},
{"lonely start bytes (4 byte)", ?_assertEqual( {"lonely start bytes (4 byte)", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#00f0, 32, 16#00f7>>, #config{}) jsx_parser:clean_string(<<16#00f0, 32, 16#00f7>>, #config{strict_utf8=true})
)}, )},
{"lonely start bytes (4 byte) replaced", ?_assertEqual( {"lonely start bytes (4 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32, 16#fffd/utf8>>, <<16#fffd/utf8, 32, 16#fffd/utf8>>,
jsx_parser:clean_string(<<16#00f0, 32, 16#00f7>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#00f0, 32, 16#00f7>>, #config{})
)}, )},
{"missing continuation byte (3 byte)", ?_assertEqual( {"missing continuation byte (3 byte)", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<224, 160, 32>>, #config{}) jsx_parser:clean_string(<<224, 160, 32>>, #config{strict_utf8=true})
)}, )},
{"missing continuation byte (3 byte) replaced", ?_assertEqual( {"missing continuation byte (3 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
jsx_parser:clean_string(<<224, 160, 32>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<224, 160, 32>>, #config{})
)}, )},
{"missing continuation byte (4 byte missing one)", ?_assertEqual( {"missing continuation byte (4 byte missing one)", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<240, 144, 128, 32>>, #config{}) jsx_parser:clean_string(<<240, 144, 128, 32>>, #config{strict_utf8=true})
)}, )},
{"missing continuation byte (4 byte missing one) replaced", ?_assertEqual( {"missing continuation byte (4 byte missing one) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
jsx_parser:clean_string(<<240, 144, 128, 32>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<240, 144, 128, 32>>, #config{})
)}, )},
{"missing continuation byte (4 byte missing two)", ?_assertEqual( {"missing continuation byte (4 byte missing two)", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<240, 144, 32>>, #config{}) jsx_parser:clean_string(<<240, 144, 32>>, #config{strict_utf8=true})
)}, )},
{"missing continuation byte (4 byte missing two) replaced", ?_assertEqual( {"missing continuation byte (4 byte missing two) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
jsx_parser:clean_string(<<240, 144, 32>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<240, 144, 32>>, #config{})
)}, )},
{"overlong encoding of u+002f (2 byte)", ?_assertEqual( {"overlong encoding of u+002f (2 byte)", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#c0, 16#af, 32>>, #config{}) jsx_parser:clean_string(<<16#c0, 16#af, 32>>, #config{strict_utf8=true})
)}, )},
{"overlong encoding of u+002f (2 byte) replaced", ?_assertEqual( {"overlong encoding of u+002f (2 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
jsx_parser:clean_string(<<16#c0, 16#af, 32>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#c0, 16#af, 32>>, #config{})
)}, )},
{"overlong encoding of u+002f (3 byte)", ?_assertEqual( {"overlong encoding of u+002f (3 byte)", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#e0, 16#80, 16#af, 32>>, #config{}) jsx_parser:clean_string(<<16#e0, 16#80, 16#af, 32>>, #config{strict_utf8=true})
)}, )},
{"overlong encoding of u+002f (3 byte) replaced", ?_assertEqual( {"overlong encoding of u+002f (3 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
jsx_parser:clean_string(<<16#e0, 16#80, 16#af, 32>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#e0, 16#80, 16#af, 32>>, #config{})
)}, )},
{"overlong encoding of u+002f (4 byte)", ?_assertEqual( {"overlong encoding of u+002f (4 byte)", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#f0, 16#80, 16#80, 16#af, 32>>, #config{}) jsx_parser:clean_string(<<16#f0, 16#80, 16#80, 16#af, 32>>, #config{strict_utf8=true})
)}, )},
{"overlong encoding of u+002f (4 byte) replaced", ?_assertEqual( {"overlong encoding of u+002f (4 byte) replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
jsx_parser:clean_string(<<16#f0, 16#80, 16#80, 16#af, 32>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#f0, 16#80, 16#80, 16#af, 32>>, #config{})
)}, )},
{"highest overlong 2 byte sequence", ?_assertEqual( {"highest overlong 2 byte sequence", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#c1, 16#bf, 32>>, #config{}) jsx_parser:clean_string(<<16#c1, 16#bf, 32>>, #config{strict_utf8=true})
)}, )},
{"highest overlong 2 byte sequence replaced", ?_assertEqual( {"highest overlong 2 byte sequence replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
jsx_parser:clean_string(<<16#c1, 16#bf, 32>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#c1, 16#bf, 32>>, #config{})
)}, )},
{"highest overlong 3 byte sequence", ?_assertEqual( {"highest overlong 3 byte sequence", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#e0, 16#9f, 16#bf, 32>>, #config{}) jsx_parser:clean_string(<<16#e0, 16#9f, 16#bf, 32>>, #config{strict_utf8=true})
)}, )},
{"highest overlong 3 byte sequence replaced", ?_assertEqual( {"highest overlong 3 byte sequence replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
jsx_parser:clean_string(<<16#e0, 16#9f, 16#bf, 32>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#e0, 16#9f, 16#bf, 32>>, #config{})
)}, )},
{"highest overlong 4 byte sequence", ?_assertEqual( {"highest overlong 4 byte sequence", ?_assertEqual(
{error, badarg}, {error, badarg},
jsx_parser:clean_string(<<16#f0, 16#8f, 16#bf, 16#bf, 32>>, #config{}) jsx_parser:clean_string(<<16#f0, 16#8f, 16#bf, 16#bf, 32>>, #config{strict_utf8=true})
)}, )},
{"highest overlong 4 byte sequence replaced", ?_assertEqual( {"highest overlong 4 byte sequence replaced", ?_assertEqual(
<<16#fffd/utf8, 32>>, <<16#fffd/utf8, 32>>,
jsx_parser:clean_string(<<16#f0, 16#8f, 16#bf, 16#bf, 32>>, #config{replaced_bad_utf8=true}) jsx_parser:clean_string(<<16#f0, 16#8f, 16#bf, 16#bf, 32>>, #config{})
)} )}
]. ].