aboutsummaryrefslogtreecommitdiffstats
path: root/lib/stdlib/uc_spec/gen_unicode_mod.escript
diff options
context:
space:
mode:
Diffstat (limited to 'lib/stdlib/uc_spec/gen_unicode_mod.escript')
-rwxr-xr-xlib/stdlib/uc_spec/gen_unicode_mod.escript57
1 files changed, 37 insertions, 20 deletions
diff --git a/lib/stdlib/uc_spec/gen_unicode_mod.escript b/lib/stdlib/uc_spec/gen_unicode_mod.escript
index fefd7d3b70..fe5a860d45 100755
--- a/lib/stdlib/uc_spec/gen_unicode_mod.escript
+++ b/lib/stdlib/uc_spec/gen_unicode_mod.escript
@@ -65,7 +65,7 @@ main(_) ->
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
parse_unicode_data(Line0, Acc) ->
- Line = string:strip(Line0, right, $\n),
+ Line = string:chomp(Line0),
[CodePoint,Name,_Cat,Class,_BiDi,Decomp,
_N1,_N2,_N3,_BDMirror,_Uni1,_Iso|Case] = tokens(Line, ";"),
{Dec,Comp} = case to_decomp(Decomp) of
@@ -78,14 +78,14 @@ parse_unicode_data(Line0, Acc) ->
|Acc].
to_class(String) ->
- list_to_integer(string:strip(String, both)).
+ list_to_integer(string:trim(String, both)).
to_decomp("") -> [];
to_decomp("<" ++ Str) ->
- [Tag,Rest] = string:tokens(Str, ">"),
+ [Tag,Rest] = string:lexemes(Str, ">"),
{list_to_atom(Tag), to_decomp(Rest)};
to_decomp(CodePoints) ->
- CPL = string:tokens(CodePoints, " "),
+ CPL = string:lexemes(CodePoints, " "),
[hex_to_int(CP) || CP <- CPL].
to_case(["","",""]) -> [];
@@ -105,20 +105,20 @@ parse_special_casing(Line, Table) ->
array:set(CP, Entry#cp{cs=Case}, Table).
to_scase([Lower,Title,Upper|_]) ->
- {unlist([hex_to_int(CP) || CP <- string:strip(string:tokens(Upper, " "), both)]),
- unlist([hex_to_int(CP) || CP <- string:strip(string:tokens(Lower, " "), both)]),
- unlist([hex_to_int(CP) || CP <- string:strip(string:tokens(Title, " "), both)]),
+ {unlist([hex_to_int(CP) || CP <- string:lexemes(Upper, " ")]),
+ unlist([hex_to_int(CP) || CP <- string:lexemes(Lower, " ")]),
+ unlist([hex_to_int(CP) || CP <- string:lexemes(Title, " ")]),
[]}.
parse_case_folding(Line, Table) ->
[CodePoint, Class0, CaseStr |_Comments] = tokens(Line, ";"),
- Class = string:strip(Class0, both),
+ Class = string:trim(Class0, both),
if Class =:= "T" -> Table; %% Do not support localization yet
Class =:= "S" -> Table; %% Ignore simple
true ->
CP = hex_to_int(CodePoint),
Case = unlist([hex_to_int(CPC) ||
- CPC <- string:strip(string:tokens(CaseStr, " "), both)]),
+ CPC <- string:lexemes(CaseStr, " ")]),
#cp{cs={U,L,T,_}} = Entry = array:get(CP, Table),
array:set(CP, Entry#cp{cs={U,L,T,Case}}, Table)
end.
@@ -170,7 +170,7 @@ gen_header(Fd) ->
io:put_chars(Fd, "-export([spec_version/0, lookup/1, get_case/1]).\n"),
io:put_chars(Fd, "-inline([class/1]).\n"),
io:put_chars(Fd, "-compile(nowarn_unused_vars).\n"),
- io:put_chars(Fd, "-dialyzer({no_improper_lists, [cp/1, gc_prepend/2, gc_e_cont/2]}).\n"),
+ io:put_chars(Fd, "-dialyzer({no_improper_lists, [cp/1, gc/1, gc_prepend/2, gc_e_cont/2]}).\n"),
io:put_chars(Fd, "-type gc() :: char()|[char()].\n\n\n"),
ok.
@@ -186,7 +186,7 @@ gen_static(Fd) ->
" {U,L} -> #{upper=>U,lower=>L,title=>U,fold=>L};\n"
" {U,L,T,F} -> #{upper=>U,lower=>L,title=>T,fold=>F}\n"
" end.\n\n"),
- io:put_chars(Fd, "spec_version() -> {9,0}.\n\n\n"),
+ io:put_chars(Fd, "spec_version() -> {10,0}.\n\n\n"),
io:put_chars(Fd, "class(Codepoint) -> {CCC,_,_} = unicode_table(Codepoint),\n CCC.\n\n"),
io:put_chars(Fd, "-spec uppercase(unicode:chardata()) -> "
"maybe_improper_list(gc(),unicode:chardata()).\n"),
@@ -240,7 +240,7 @@ gen_norm(Fd) ->
"-spec nfd(unicode:chardata()) -> maybe_improper_list(gc(),unicode:chardata()) | {error, unicode:chardata()}.\n"
"nfd(Str0) ->\n"
" case gc(Str0) of\n"
- " [GC|R] when GC < 127 -> [GC|R];\n"
+ " [GC|R] when GC < 128 -> [GC|R];\n"
" [GC|Str] -> [decompose(GC)|Str];\n"
" [] -> [];\n"
" {error,_}=Error -> Error\n end.\n\n"
@@ -250,7 +250,7 @@ gen_norm(Fd) ->
"-spec nfkd(unicode:chardata()) -> maybe_improper_list(gc(),unicode:chardata()) | {error, unicode:chardata()}.\n"
"nfkd(Str0) ->\n"
" case gc(Str0) of\n"
- " [GC|R] when GC < 127 -> [GC|R];\n"
+ " [GC|R] when GC < 128 -> [GC|R];\n"
" [GC|Str] -> [decompose_compat(GC)|Str];\n"
" [] -> [];\n"
" {error,_}=Error -> Error\n end.\n\n"
@@ -260,7 +260,7 @@ gen_norm(Fd) ->
"-spec nfc(unicode:chardata()) -> maybe_improper_list(gc(),unicode:chardata()) | {error, unicode:chardata()}.\n"
"nfc(Str0) ->\n"
" case gc(Str0) of\n"
- " [GC|R] when GC < 255 -> [GC|R];\n"
+ " [GC|R] when GC < 256 -> [GC|R];\n"
" [GC|Str] -> [compose(decompose(GC))|Str];\n"
" [] -> [];\n"
" {error,_}=Error -> Error\n end.\n\n"
@@ -270,7 +270,7 @@ gen_norm(Fd) ->
"-spec nfkc(unicode:chardata()) -> maybe_improper_list(gc(),unicode:chardata()) | {error, unicode:chardata()}.\n"
"nfkc(Str0) ->\n"
" case gc(Str0) of\n"
- " [GC|R] when GC < 127 -> [GC|R];\n"
+ " [GC|R] when GC < 128 -> [GC|R];\n"
" [GC|Str] -> [compose_compat_0(decompose_compat(GC))|Str];\n"
" [] -> [];\n"
" {error,_}=Error -> Error\n end.\n\n"
@@ -476,13 +476,30 @@ gen_gc(Fd, GBP) ->
"-spec gc(String::unicode:chardata()) ->"
" maybe_improper_list() | {error, unicode:chardata()}.\n"),
io:put_chars(Fd,
+ "gc([CP1, CP2|_]=T)\n"
+ " when CP1 < 256, CP2 < 256, CP1 =/= $\r -> %% Ascii Fast path\n"
+ " T;\n"
+ "gc(<<CP1/utf8, Rest/binary>>) ->\n"
+ " if CP1 < 256, CP1 =/= $\r ->\n"
+ " case Rest of\n"
+ " <<CP2/utf8, _/binary>> when CP2 < 256 -> %% Ascii Fast path\n"
+ " [CP1|Rest];\n"
+ " _ -> gc_1([CP1|Rest])\n"
+ " end;\n"
+ " true -> gc_1([CP1|Rest])\n"
+ " end;\n"
"gc(Str) ->\n"
" gc_1(cp(Str)).\n\n"
"gc_1([$\\r|R0] = R) ->\n"
" case cp(R0) of % Don't break CRLF\n"
" [$\\n|R1] -> [[$\\r,$\\n]|R1];\n"
" _ -> R\n"
- " end;\n"),
+ " end;\n"
+ %% "gc_1([CP1, CP2|_]=T) when CP1 < 256, CP2 < 256 ->\n"
+ %% " T; %% Fast path\n"
+ %% "gc_1([CP1|<<CP2/utf8, _/binary>>]=T) when CP1 < 256, CP2 < 256 ->\n"
+ %% " T; %% Fast path\n"
+ ),
io:put_chars(Fd, "%% Handle control\n"),
GenControl = fun(Range) -> io:format(Fd, "gc_1~s R0;\n", [gen_clause(Range)]) end,
@@ -490,7 +507,7 @@ gen_gc(Fd, GBP) ->
[R1,R2,R3|Crs] = CRs0,
[GenControl(CP) || CP <- merge_ranges([R1,R2,R3], split), CP =/= {$\r, undefined}],
%%GenControl(R1),GenControl(R2),GenControl(R3),
- io:format(Fd, "gc_1([CP|R]) when CP < 255 -> gc_extend(R,CP);\n", []),
+ io:format(Fd, "gc_1([CP|R]) when CP < 256 -> gc_extend(R,CP);\n", []),
[GenControl(CP) || CP <- Crs],
%% One clause per CP
%% CRs0 = merge_ranges(maps:get(cr, GBP) ++ maps:get(lf, GBP) ++ maps:get(control, GBP)),
@@ -869,10 +886,10 @@ optimize_ranges_1(Rs) ->
hex_to_int([]) -> [];
hex_to_int(HexStr) ->
- list_to_integer(string:strip(HexStr, both), 16).
+ list_to_integer(string:trim(HexStr, both), 16).
to_atom(Str) ->
- list_to_atom(string:to_lower(string:strip(Str, both))).
+ list_to_atom(string:lowercase(string:trim(Str, both))).
foldl(Fun, Acc, Fd) ->
Get = fun() -> file:read_line(Fd) end,
@@ -892,7 +909,7 @@ foldl_1(Fun, Acc, Get) ->
-%% Differs from string:tokens, it returns empty string as token between two delimiters
+%% Differs from string:lexemes, it returns empty string as token between two delimiters
tokens(S, [C]) ->
tokens(lists:reverse(S), C, []).
generated by cgit v1.2.3 (git 2.25.1) at 2024-09-09 17:11:52 +0000