diff options
Diffstat (limited to 'lib/stdlib/src')
| -rw-r--r-- | lib/stdlib/src/binary.erl | 28 | ||||
| -rw-r--r-- | lib/stdlib/src/erl_lint.erl | 8 | ||||
| -rw-r--r-- | lib/stdlib/src/io_lib.erl | 58 | ||||
| -rw-r--r-- | lib/stdlib/src/io_lib_format.erl | 43 | ||||
| -rw-r--r-- | lib/stdlib/src/string.erl | 8 | ||||
| -rw-r--r-- | lib/stdlib/src/uri_string.erl | 65 |
6 files changed, 131 insertions, 79 deletions
diff --git a/lib/stdlib/src/binary.erl b/lib/stdlib/src/binary.erl index 6a64133b45..7d0e42489e 100644 --- a/lib/stdlib/src/binary.erl +++ b/lib/stdlib/src/binary.erl @@ -47,23 +47,39 @@ at(_, _) -> -spec bin_to_list(Subject) -> [byte()] when Subject :: binary(). -bin_to_list(_) -> - erlang:nif_error(undef). +bin_to_list(Subject) -> + binary_to_list(Subject). -spec bin_to_list(Subject, PosLen) -> [byte()] when Subject :: binary(), PosLen :: part(). -bin_to_list(_, _) -> - erlang:nif_error(undef). +bin_to_list(Subject, {Pos, Len}) -> + bin_to_list(Subject, Pos, Len); +bin_to_list(_Subject, _BadArg) -> + erlang:error(badarg). -spec bin_to_list(Subject, Pos, Len) -> [byte()] when Subject :: binary(), Pos :: non_neg_integer(), Len :: integer(). -bin_to_list(_, _, _) -> - erlang:nif_error(undef). +bin_to_list(Subject, Pos, Len) when not is_binary(Subject); + not is_integer(Pos); + not is_integer(Len) -> + %% binary_to_list/3 allows bitstrings as long as the slice fits, and we + %% want to badarg when Pos/Len aren't integers instead of raising badarith + %% when adjusting args for binary_to_list/3. + erlang:error(badarg); +bin_to_list(Subject, Pos, 0) when Pos >= 0, Pos =< byte_size(Subject) -> + %% binary_to_list/3 doesn't handle this case. + []; +bin_to_list(_Subject, _Pos, 0) -> + erlang:error(badarg); +bin_to_list(Subject, Pos, Len) when Len < 0 -> + bin_to_list(Subject, Pos + Len, -Len); +bin_to_list(Subject, Pos, Len) when Len > 0 -> + binary_to_list(Subject, Pos + 1, Pos + Len). -spec compile_pattern(Pattern) -> cp() when Pattern :: binary() | [binary()]. diff --git a/lib/stdlib/src/erl_lint.erl b/lib/stdlib/src/erl_lint.erl index 1930c462e8..9a62d21d34 100644 --- a/lib/stdlib/src/erl_lint.erl +++ b/lib/stdlib/src/erl_lint.erl @@ -3971,6 +3971,8 @@ extract_sequence(3, [$.,_|Fmt], Need) -> extract_sequence(4, Fmt, Need); extract_sequence(3, Fmt, Need) -> extract_sequence(4, Fmt, Need); +extract_sequence(4, [$t, $l | Fmt], Need) -> + extract_sequence(4, [$l, $t | Fmt], Need); extract_sequence(4, [$t, $c | Fmt], Need) -> extract_sequence(5, [$c|Fmt], Need); extract_sequence(4, [$t, $s | Fmt], Need) -> @@ -3987,8 +3989,14 @@ extract_sequence(4, [$t, C | _Fmt], _Need) -> {error,"invalid control ~t" ++ [C]}; extract_sequence(4, [$l, $p | Fmt], Need) -> extract_sequence(5, [$p|Fmt], Need); +extract_sequence(4, [$l, $t, $p | Fmt], Need) -> + extract_sequence(5, [$p|Fmt], Need); extract_sequence(4, [$l, $P | Fmt], Need) -> extract_sequence(5, [$P|Fmt], Need); +extract_sequence(4, [$l, $t, $P | Fmt], Need) -> + extract_sequence(5, [$P|Fmt], Need); +extract_sequence(4, [$l, $t, C | _Fmt], _Need) -> + {error,"invalid control ~lt" ++ [C]}; extract_sequence(4, [$l, C | _Fmt], _Need) -> {error,"invalid control ~l" ++ [C]}; extract_sequence(4, Fmt, Need) -> diff --git a/lib/stdlib/src/io_lib.erl b/lib/stdlib/src/io_lib.erl index cacd9f2524..e37c13093b 100644 --- a/lib/stdlib/src/io_lib.erl +++ b/lib/stdlib/src/io_lib.erl @@ -931,7 +931,7 @@ limit_term(Term, Depth) -> limit(_, 0) -> '...'; limit([H|T]=L, D) -> if - D =:= 1 -> '...'; + D =:= 1 -> ['...']; true -> case printable_list(L) of true -> L; @@ -944,7 +944,7 @@ limit(Term, D) when is_map(Term) -> limit({}=T, _D) -> T; limit(T, D) when is_tuple(T) -> if - D =:= 1 -> '...'; + D =:= 1 -> {'...'}; true -> list_to_tuple([limit(element(1, T), D-1)| limit_tail(tl(tuple_to_list(T)), D-1)]) @@ -961,32 +961,29 @@ limit_tail(Other, D) -> %% Cannot limit maps properly since there is no guarantee that %% maps:from_list() creates a map with the same internal ordering of -%% the selected associations as in Map. +%% the selected associations as in Map. Instead of subtracting one +%% from the depth as the map associations are traversed (as is done +%% for tuples and lists), the same depth is applied to each and every +%% (returned) association. limit_map(Map, D) -> - limit_map(maps:iterator(Map), D, []). + %% Keep one extra association to make sure the final ',...' is included. + limit_map_body(maps:iterator(Map), D + 1, D, []). -limit_map(_I, 0, Acc) -> +limit_map_body(_I, 0, _D0, Acc) -> maps:from_list(Acc); -limit_map(I, D, Acc) -> +limit_map_body(I, D, D0, Acc) -> case maps:next(I) of {K, V, NextI} -> - limit_map(NextI, D-1, [{K,V} | Acc]); + limit_map_body(NextI, D-1, D0, [limit_map_assoc(K, V, D0) | Acc]); none -> maps:from_list(Acc) end. -%% maps:from_list(limit_map_body(erts_internal:maps_to_list(Map, D), D)). +limit_map_assoc(K, V, D) -> + %% Keep keys as are to avoid creating duplicated keys. + {K, limit(V, D - 1)}. -%% limit_map_body(_, 0) -> [{'...', '...'}]; -%% limit_map_body([], _) -> []; -%% limit_map_body([{K,V}], D) -> [limit_map_assoc(K, V, D)]; -%% limit_map_body([{K,V}|KVs], D) -> -%% [limit_map_assoc(K, V, D) | limit_map_body(KVs, D-1)]. - -%% limit_map_assoc(K, V, D) -> -%% {limit(K, D-1), limit(V, D-1)}. - -limit_bitstring(B, _D) -> B. %% Keeps all printable binaries. +limit_bitstring(B, _D) -> B. % Keeps all printable binaries. test_limit(_, 0) -> throw(limit); test_limit([H|T]=L, D) when is_integer(D) -> @@ -1022,18 +1019,21 @@ test_limit_tuple(T, I, Sz, D) -> test_limit(element(I, T), D-1), test_limit_tuple(T, I+1, Sz, D-1). -test_limit_map(_Map, _D) -> ok. -%% test_limit_map_body(erts_internal:maps_to_list(Map, D), D). +test_limit_map(Map, D) -> + test_limit_map_body(maps:iterator(Map), D). -%% test_limit_map_body(_, 0) -> throw(limit); -%% test_limit_map_body([], _) -> ok; -%% test_limit_map_body([{K,V}], D) -> test_limit_map_assoc(K, V, D); -%% test_limit_map_body([{K,V}|KVs], D) -> -%% test_limit_map_assoc(K, V, D), -%% test_limit_map_body(KVs, D-1). +test_limit_map_body(_I, 0) -> throw(limit); % cannot happen +test_limit_map_body(I, D) -> + case maps:next(I) of + {K, V, NextI} -> + test_limit_map_assoc(K, V, D), + test_limit_map_body(NextI, D-1); + none -> + ok + end. -%% test_limit_map_assoc(K, V, D) -> -%% test_limit(K, D-1), -%% test_limit(V, D-1). +test_limit_map_assoc(K, V, D) -> + test_limit(K, D - 1), + test_limit(V, D - 1). test_limit_bitstring(_, _) -> ok. diff --git a/lib/stdlib/src/io_lib_format.erl b/lib/stdlib/src/io_lib_format.erl index e345810ca0..64edbf1824 100644 --- a/lib/stdlib/src/io_lib_format.erl +++ b/lib/stdlib/src/io_lib_format.erl @@ -95,7 +95,7 @@ print([]) -> []. print(C, F, Ad, P, Pad, Encoding, Strings) -> - [$~] ++ print_field_width(F, Ad) ++ print_precision(P) ++ + [$~] ++ print_field_width(F, Ad) ++ print_precision(P, Pad) ++ print_pad_char(Pad) ++ print_encoding(Encoding) ++ print_strings(Strings) ++ [C]. @@ -103,8 +103,9 @@ print_field_width(none, _Ad) -> ""; print_field_width(F, left) -> integer_to_list(-F); print_field_width(F, right) -> integer_to_list(F). -print_precision(none) -> ""; -print_precision(P) -> [$. | integer_to_list(P)]. +print_precision(none, $\s) -> ""; +print_precision(none, _Pad) -> "."; % pad must be second dot +print_precision(P, _Pad) -> [$. | integer_to_list(P)]. print_pad_char($\s) -> ""; % default, no need to make explicit print_pad_char(Pad) -> [$., Pad]. @@ -126,25 +127,23 @@ collect_cseq(Fmt0, Args0) -> {F,Ad,Fmt1,Args1} = field_width(Fmt0, Args0), {P,Fmt2,Args2} = precision(Fmt1, Args1), {Pad,Fmt3,Args3} = pad_char(Fmt2, Args2), - {Encoding,Fmt4,Args4} = encoding(Fmt3, Args3), - {Strings,Fmt5,Args5} = strings(Fmt4, Args4), - {C,As,Fmt6,Args6} = collect_cc(Fmt5, Args5), - FormatSpec = #{control_char => C, args => As, width => F, adjust => Ad, - precision => P, pad_char => Pad, encoding => Encoding, - strings => Strings}, - {FormatSpec,Fmt6,Args6}. - -encoding([$t|Fmt],Args) -> - true = hd(Fmt) =/= $l, - {unicode,Fmt,Args}; -encoding(Fmt,Args) -> - {latin1,Fmt,Args}. - -strings([$l|Fmt],Args) -> - true = hd(Fmt) =/= $t, - {false,Fmt,Args}; -strings(Fmt,Args) -> - {true,Fmt,Args}. + Spec0 = #{width => F, + adjust => Ad, + precision => P, + pad_char => Pad, + encoding => latin1, + strings => true}, + {Spec1,Fmt4} = modifiers(Fmt3, Spec0), + {C,As,Fmt5,Args4} = collect_cc(Fmt4, Args3), + Spec2 = Spec1#{control_char => C, args => As}, + {Spec2,Fmt5,Args4}. + +modifiers([$t|Fmt], Spec) -> + modifiers(Fmt, Spec#{encoding => unicode}); +modifiers([$l|Fmt], Spec) -> + modifiers(Fmt, Spec#{strings => false}); +modifiers(Fmt, Spec) -> + {Spec, Fmt}. field_width([$-|Fmt0], Args0) -> {F,Fmt,Args} = field_value(Fmt0, Args0), diff --git a/lib/stdlib/src/string.erl b/lib/stdlib/src/string.erl index e01bb7d85e..4e89819e41 100644 --- a/lib/stdlib/src/string.erl +++ b/lib/stdlib/src/string.erl @@ -420,10 +420,12 @@ to_number(_, Number, Rest, _, Tail) -> %% Return the remaining string with prefix removed or else nomatch -spec prefix(String::unicode:chardata(), Prefix::unicode:chardata()) -> 'nomatch' | unicode:chardata(). -prefix(Str, []) -> Str; prefix(Str, Prefix0) -> - Prefix = unicode:characters_to_list(Prefix0), - case prefix_1(Str, Prefix) of + Result = case unicode:characters_to_list(Prefix0) of + [] -> Str; + Prefix -> prefix_1(Str, Prefix) + end, + case Result of [] when is_binary(Str) -> <<>>; Res -> Res end. diff --git a/lib/stdlib/src/uri_string.erl b/lib/stdlib/src/uri_string.erl index a84679c595..28d36ea229 100644 --- a/lib/stdlib/src/uri_string.erl +++ b/lib/stdlib/src/uri_string.erl @@ -227,7 +227,7 @@ %% External API %%------------------------------------------------------------------------- -export([compose_query/1, compose_query/2, - dissect_query/1, normalize/1, parse/1, + dissect_query/1, normalize/1, normalize/2, parse/1, recompose/1, transcode/2]). -export_type([error/0, uri_map/0, uri_string/0]). @@ -292,18 +292,36 @@ %%------------------------------------------------------------------------- %% Normalize URIs %%------------------------------------------------------------------------- --spec normalize(URIString) -> NormalizedURI when - URIString :: uri_string(), - NormalizedURI :: uri_string(). -normalize(URIString) -> - %% Percent-encoding normalization and case normalization for - %% percent-encoded triplets are achieved by running parse and - %% recompose on the input URI string. - recompose( - normalize_path_segment( - normalize_scheme_based( - normalize_case( - parse(URIString))))). +-spec normalize(URI) -> NormalizedURI when + URI :: uri_string() | uri_map(), + NormalizedURI :: uri_string() + | error(). +normalize(URIMap) -> + normalize(URIMap, []). + + +-spec normalize(URI, Options) -> NormalizedURI when + URI :: uri_string() | uri_map(), + Options :: [return_map], + NormalizedURI :: uri_string() | uri_map(). +normalize(URIMap, []) when is_map(URIMap) -> + recompose(normalize_map(URIMap)); +normalize(URIMap, [return_map]) when is_map(URIMap) -> + normalize_map(URIMap); +normalize(URIString, []) -> + case parse(URIString) of + Value when is_map(Value) -> + recompose(normalize_map(Value)); + Error -> + Error + end; +normalize(URIString, [return_map]) -> + case parse(URIString) of + Value when is_map(Value) -> + normalize_map(Value); + Error -> + Error + end. %%------------------------------------------------------------------------- @@ -385,7 +403,8 @@ transcode(URIString, Options) when is_list(URIString) -> %%------------------------------------------------------------------------- %% Functions for working with the query part of a URI as a list %% of key/value pairs. -%% HTML5 - 4.10.22.6 URL-encoded form data +%% HTML 5.2 - 4.10.21.6 URL-encoded form data - WHATWG URL (10 Jan 2018) - UTF-8 +%% HTML 5.0 - 4.10.22.6 URL-encoded form data - non UTF-8 %%------------------------------------------------------------------------- %%------------------------------------------------------------------------- @@ -393,7 +412,7 @@ transcode(URIString, Options) when is_list(URIString) -> %% (application/x-www-form-urlencoded encoding algorithm) %%------------------------------------------------------------------------- -spec compose_query(QueryList) -> QueryString when - QueryList :: [{uri_string(), uri_string()}], + QueryList :: [{unicode:chardata(), unicode:chardata()}], QueryString :: uri_string() | error(). compose_query(List) -> @@ -401,7 +420,7 @@ compose_query(List) -> -spec compose_query(QueryList, Options) -> QueryString when - QueryList :: [{uri_string(), uri_string()}], + QueryList :: [{unicode:chardata(), unicode:chardata()}], Options :: [{encoding, atom()}], QueryString :: uri_string() | error(). @@ -432,7 +451,7 @@ compose_query([], _Options, IsList, Acc) -> %%------------------------------------------------------------------------- -spec dissect_query(QueryString) -> QueryList when QueryString :: uri_string(), - QueryList :: [{uri_string(), uri_string()}] + QueryList :: [{unicode:chardata(), unicode:chardata()}] | error(). dissect_query(<<>>) -> []; @@ -1755,7 +1774,8 @@ get_separator(_L) -> <<"&">>. -%% HTML5 - 4.10.22.6 URL-encoded form data - encoding +%% HTML 5.2 - 4.10.21.6 URL-encoded form data - WHATWG URL (10 Jan 2018) - UTF-8 +%% HTML 5.0 - 4.10.22.6 URL-encoded form data - encoding (non UTF-8) form_urlencode(Cs, [{encoding, latin1}]) when is_list(Cs) -> B = convert_to_binary(Cs, utf8, utf8), html5_byte_encode(base10_encode(B)); @@ -1850,7 +1870,8 @@ dissect_query_value(<<>>, IsList, Acc, Key, Value) -> lists:reverse([{K,V}|Acc]). -%% Form-urldecode input based on RFC 1866 [8.2.1] +%% HTML 5.2 - 4.10.21.6 URL-encoded form data - WHATWG URL (10 Jan 2018) - UTF-8 +%% HTML 5.0 - 4.10.22.6 URL-encoded form data - decoding (non UTF-8) form_urldecode(true, B) -> Result = base10_decode(form_urldecode(B, <<>>)), convert_to_list(Result, utf8); @@ -1903,6 +1924,12 @@ base10_decode_unicode(<<H,_/binary>>, _, _) -> %% Helper functions for normalize %%------------------------------------------------------------------------- +normalize_map(URIMap) -> + normalize_path_segment( + normalize_scheme_based( + normalize_case(URIMap))). + + %% 6.2.2.1. Case Normalization normalize_case(#{scheme := Scheme, host := Host} = Map) -> Map#{scheme => to_lower(Scheme), |