diff options
| -rw-r--r-- | lib/stdlib/src/erl_scan.erl | 301 | ||||
| -rw-r--r-- | lib/stdlib/test/erl_scan_SUITE.erl | 108 |
2 files changed, 138 insertions, 271 deletions
diff --git a/lib/stdlib/src/erl_scan.erl b/lib/stdlib/src/erl_scan.erl index 26d5747ee7..3651f608bc 100644 --- a/lib/stdlib/src/erl_scan.erl +++ b/lib/stdlib/src/erl_scan.erl @@ -84,8 +84,7 @@ -type location() :: line() | {line(),column()}. -type resword_fun() :: fun((atom()) -> boolean()). -type option() :: 'return' | 'return_white_spaces' | 'return_comments' - | 'text' | {'reserved_word_fun', resword_fun()} - | 'unicode'. + | 'text' | {'reserved_word_fun', resword_fun()}. -type options() :: option() | [option()]. -type symbol() :: atom() | float() | integer() | string(). -type info_line() :: integer() | term(). @@ -106,8 +105,7 @@ {resword_fun = fun reserved_word/1 :: resword_fun(), ws = false :: boolean(), comment = false :: boolean(), - text = false :: boolean(), - unicode = true :: boolean()}). + text = false :: boolean()}). %%---------------------------------------------------------------------------- @@ -344,20 +342,12 @@ string_thing(_) -> "string". C > 16#DFFF andalso C < 16#FFFE orelse C > 16#FFFF andalso C =< 16#10FFFF)). -%% When the option 'unicode' is false: return Unicode strings as lists -%% of integers and Unicode characters as integers. For instance, -%% erl_scan:string("\"b\x{aaa}c\".") is equivalent to -%% erl_scan:string("[98,2730,99]."). This is to protect the caller -%% from character codes greater than 255. Search for UNI to find code -%% implementing this "feature". The 'unicode' option is undocumented -%% and will be removed later. --define(NO_UNICODE, 0). --define(UNI255(C), (C =< 16#ff)). +-define(UNI255(C), C >= 0, C =< 16#ff). options(Opts0) when is_list(Opts0) -> Opts = lists:foldr(fun expand_opt/2, [], Opts0), - [RW_fun, Unicode] = - case opts(Opts, [reserved_word_fun, unicode], []) of + [RW_fun] = + case opts(Opts, [reserved_word_fun], []) of badarg -> erlang:error(badarg, [Opts0]); R -> @@ -369,8 +359,7 @@ options(Opts0) when is_list(Opts0) -> #erl_scan{resword_fun = RW_fun, comment = Comment, ws = WS, - text = Txt, - unicode = Unicode}; + text = Txt}; options(Opt) -> options([Opt]). @@ -378,8 +367,6 @@ opts(Options, [Key|Keys], L) -> V = case lists:keyfind(Key, 1, Options) of {reserved_word_fun,F} when ?RESWORDFUN(F) -> {ok,F}; - {unicode, Bool} when is_boolean(Bool) -> - {ok,Bool}; {Key,_} -> badarg; false -> @@ -395,9 +382,7 @@ opts(_Options, [], L) -> lists:reverse(L). default_option(reserved_word_fun) -> - fun reserved_word/1; -default_option(unicode) -> - true. + fun reserved_word/1. expand_opt(return, Os) -> [return_comments,return_white_spaces|Os]; @@ -531,10 +516,10 @@ scan1("."=Cs, _St, Line, Col, Toks) -> scan1([$.=C|Cs], St, Line, Col, Toks) -> scan_dot(Cs, St, Line, Col, Toks, [C]); scan1([$"|Cs], St, Line, Col, Toks) -> %" Emacs - State0 = {[],[],Line,Col,?NO_UNICODE}, + State0 = {[],[],Line,Col}, scan_string(Cs, St, Line, incr_column(Col, 1), Toks, State0); scan1([$'|Cs], St, Line, Col, Toks) -> %' Emacs - State0 = {[],[],Line,Col,?NO_UNICODE}, + State0 = {[],[],Line,Col}, scan_qatom(Cs, St, Line, incr_column(Col, 1), Toks, State0); scan1([$$|Cs], St, Line, Col, Toks) -> scan_char(Cs, St, Line, Col, Toks); @@ -655,7 +640,7 @@ scan1([$~|Cs], St, Line, Col, Toks) -> scan1([$&|Cs], St, Line, Col, Toks) -> tok2(Cs, St, Line, Col, Toks, "&", '&', 1); %% End of optimization. -scan1([C|Cs], St, Line, Col, Toks) when ?CHAR(C), ?UNI255(C) -> +scan1([C|Cs], St, Line, Col, Toks) when ?UNI255(C) -> Str = [C], tok2(Cs, St, Line, Col, Toks, Str, list_to_atom(Str), 1); scan1([C|Cs], _St, Line, Col, _Toks) when ?CHAR(C) -> @@ -718,14 +703,16 @@ scan_name([], Ncs) -> scan_name(Cs, Ncs) -> {lists:reverse(Ncs),Cs}. +-define(STR(St, S), if St#erl_scan.text -> S; true -> [] end). + scan_dot([$%|_]=Cs, St, Line, Col, Toks, Ncs) -> Attrs = attributes(Line, Col, St, Ncs), {ok,[{dot,Attrs}|Toks],Cs,Line,incr_column(Col, 1)}; scan_dot([$\n=C|Cs], St, Line, Col, Toks, Ncs) -> - Attrs = attributes(Line, Col, St, Ncs++[C]), + Attrs = attributes(Line, Col, St, ?STR(St, Ncs++[C])), {ok,[{dot,Attrs}|Toks],Cs,Line+1,new_column(Col, 1)}; scan_dot([C|Cs], St, Line, Col, Toks, Ncs) when ?WHITE_SPACE(C) -> - Attrs = attributes(Line, Col, St, Ncs++[C]), + Attrs = attributes(Line, Col, St, ?STR(St, Ncs++[C])), {ok,[{dot,Attrs}|Toks],Cs,Line,incr_column(Col, 2)}; scan_dot(eof=Cs, St, Line, Col, Toks, Ncs) -> Attrs = attributes(Line, Col, St, Ncs), @@ -858,26 +845,20 @@ scan_char([$\\|Cs]=Cs0, St, Line, Col, Toks) -> {eof,Ncol} -> scan_error(char, Line, Col, Line, Ncol, eof); {nl,Val,Str,Ncs,Ncol} -> - Attrs = attributes(Line, Col, St, "$\\"++Str), %" + Attrs = attributes(Line, Col, St, ?STR(St, "$\\"++Str)), %" Ntoks = [{char,Attrs,Val}|Toks], scan1(Ncs, St, Line+1, Ncol, Ntoks); - {unicode,Val,Str,Ncs,Ncol} -> - Attrs = attributes(Line, Col, St, "$\\"++Str), %" - Tag = char_tag(Val, St), % UNI - Ntoks = [{Tag,Attrs,Val}|Toks], - scan1(Ncs, St, Line, Ncol, Ntoks); {Val,Str,Ncs,Ncol} -> - Attrs = attributes(Line, Col, St, "$\\"++Str), %" + Attrs = attributes(Line, Col, St, ?STR(St, "$\\"++Str)), %" Ntoks = [{char,Attrs,Val}|Toks], scan1(Ncs, St, Line, Ncol, Ntoks) end; scan_char([$\n=C|Cs], St, Line, Col, Toks) -> - Attrs = attributes(Line, Col, St, [$$,C]), + Attrs = attributes(Line, Col, St, ?STR(St, [$$,C])), scan1(Cs, St, Line+1, new_column(Col, 1), [{char,Attrs,C}|Toks]); scan_char([C|Cs], St, Line, Col, Toks) when ?UNICODE(C) -> - Tag = char_tag(C, St), % UNI - Attrs = attributes(Line, Col, St, [$$,C]), - scan1(Cs, St, Line, incr_column(Col, 2), [{Tag,Attrs,C}|Toks]); + Attrs = attributes(Line, Col, St, ?STR(St, [$$,C])), + scan1(Cs, St, Line, incr_column(Col, 2), [{char,Attrs,C}|Toks]); scan_char([C|_Cs], _St, Line, Col, _Toks) when ?CHAR(C) -> scan_error({illegal,character}, Line, Col, Line, incr_column(Col, 1), eof); scan_char([], _St, Line, Col, Toks) -> @@ -885,90 +866,32 @@ scan_char([], _St, Line, Col, Toks) -> scan_char(eof, _St, Line, Col, _Toks) -> scan_error(char, Line, Col, Line, incr_column(Col, 1), eof). --compile({inline,[char_tag/2]}). - -char_tag(C, _St) when ?UNI255(C) -> - char; -char_tag(_C, #erl_scan{unicode = true}) -> - char; -char_tag(_C, _St) -> - integer. - -scan_string(Cs, St, Line, Col, Toks, {Wcs,Str,Line0,Col0,Uni0}) -> - case scan_string0(Cs, St, Line, Col, $\", true, Str, Wcs, Uni0) of %" - {more,Ncs,Nline,Ncol,Nstr,Nwcs,Uni} -> - State = {Nwcs,Nstr,Line0,Col0,Uni}, +scan_string(Cs, St, Line, Col, Toks, {Wcs,Str,Line0,Col0}) -> + case scan_string0(Cs, St, Line, Col, $\", Str, Wcs) of %" + {more,Ncs,Nline,Ncol,Nstr,Nwcs} -> + State = {Nwcs,Nstr,Line0,Col0}, {more,{Ncs,Ncol,Toks,Nline,State,fun scan_string/6}}; {char_error,Ncs,Error,Nline,Ncol,EndCol} -> scan_error(Error, Nline, Ncol, Nline, EndCol, Ncs); {error,Nline,Ncol,Nwcs,Ncs} -> Estr = string:substr(Nwcs, 1, 16), % Expanded escape chars. scan_error({string,$\",Estr}, Line0, Col0, Nline, Ncol, Ncs); %" - {Ncs,Nline,Ncol,Nstr,Nwcs,Uni} when Uni =:= ?NO_UNICODE; - St#erl_scan.unicode -> + {Ncs,Nline,Ncol,Nstr,Nwcs} -> Attrs = attributes(Line0, Col0, St, Nstr), - scan1(Ncs, St, Nline, Ncol, [{string,Attrs,Nwcs}|Toks]); - {Ncs,Nline,Ncol,Nstr,_Nwcs,_Uni} -> - Ntoks = unicode_string_to_list(Line0, Col0, St, Nstr, Toks), - scan1(Ncs, St, Nline, Ncol, Ntoks) + scan1(Ncs, St, Nline, Ncol, [{string,Attrs,Nwcs}|Toks]) end. -%% UNI -unicode_string_to_list(Line, Col, St, [$"=C|Nstr], Toks) -> %" Emacs - Paren = {'[',attributes(Line, Col, St, [C])}, - u2l(Nstr, Line, incr_column(Col, 1), St, [Paren|Toks]). - -u2l([$"]=Cs, Line, Col, St, Toks) -> %" Emacs - [{']',attributes(Line, Col, St, Cs)}|Toks]; -u2l([$\n=C|Cs], Line, Col, St, Toks) -> - Ntoks = unicode_nl_tokens(Line, Col, [C], C, St, Toks, Cs), - u2l(Cs, Line+1, new_column(Col, 1), St, Ntoks); -u2l([$\\|Cs], Line, Col, St, Toks) -> - case scan_escape(Cs, Col) of - {nl,Val,ValStr,Ncs,Ncol} -> - Nstr = [$\\|ValStr], - Ntoks = unicode_nl_tokens(Line, Col, Nstr, Val, St, Toks, Ncs), - u2l(Ncs, Line+1, Ncol, St, Ntoks); - {unicode,Val,ValStr,Ncs,Ncol} -> - Nstr = [$\\|ValStr], - Ntoks = unicode_tokens(Line, Col, Nstr, Val, St, Toks, Ncs), - u2l(Ncs, Line, incr_column(Ncol, 1), St, Ntoks); - {Val,ValStr,Ncs,Ncol} -> - Nstr = [$\\|ValStr], - Ntoks = unicode_tokens(Line, Col, Nstr, Val, St, Toks, Ncs), - u2l(Ncs, Line, incr_column(Ncol, 1), St, Ntoks) - end; -u2l([C|Cs], Line, Col, St, Toks) -> - Ntoks = unicode_tokens(Line, Col, [C], C, St, Toks, Cs), - u2l(Cs, Line, incr_column(Col, 1), St, Ntoks). - -unicode_nl_tokens(Line, Col, Str, Val, St, Toks, Cs) -> - Ccol = new_column(Col, 1), - unicode_tokens(Line, Col, Str, Val, St, Toks, Cs, Line+1, Ccol). - -unicode_tokens(Line, Col, Str, Val, St, Toks, Cs) -> - Ccol = incr_column(Col, length(Str)), - unicode_tokens(Line, Col, Str, Val, St, Toks, Cs, Line, Ccol). - -unicode_tokens(Line, Col, Str, Val, St, Toks, Cs, Cline, Ccol) -> - Attrs = attributes(Line, Col, St, Str), - Tag = if ?UNI255(Val) -> char; true -> integer end, - Token = {Tag,Attrs,Val}, - [{',',attributes(Cline, Ccol, St, "")} || Cs =/= "\""] ++ [Token|Toks]. - -scan_qatom(Cs, St, Line, Col, Toks, {Wcs,Str,Line0,Col0,Uni0}) -> - AllowUni = St#erl_scan.unicode, - case scan_string0(Cs, St, Line, Col, $\', AllowUni, Str, Wcs, Uni0) of %' - {more,Ncs,Nline,Ncol,Nstr,Nwcs,Uni} -> - State = {Nwcs,Nstr,Line0,Col0,Uni}, +scan_qatom(Cs, St, Line, Col, Toks, {Wcs,Str,Line0,Col0}) -> + case scan_string0(Cs, St, Line, Col, $\', Str, Wcs) of %' + {more,Ncs,Nline,Ncol,Nstr,Nwcs} -> + State = {Nwcs,Nstr,Line0,Col0}, {more,{Ncs,Ncol,Toks,Nline,State,fun scan_qatom/6}}; {char_error,Ncs,Error,Nline,Ncol,EndCol} -> scan_error(Error, Nline, Ncol, Nline, EndCol, Ncs); {error,Nline,Ncol,Nwcs,Ncs} -> Estr = string:substr(Nwcs, 1, 16), % Expanded escape chars. scan_error({string,$\',Estr}, Line0, Col0, Nline, Ncol, Ncs); %' - {Ncs,Nline,Ncol,Nstr,Nwcs,Uni} -> - true = Uni =:= ?NO_UNICODE orelse AllowUni, + {Ncs,Nline,Ncol,Nstr,Nwcs} -> case catch list_to_atom(Nwcs) of A when is_atom(A) -> Attrs = attributes(Line0, Col0, St, Nstr), @@ -978,95 +901,75 @@ scan_qatom(Cs, St, Line, Col, Toks, {Wcs,Str,Line0,Col0,Uni0}) -> end end. -scan_string0(Cs, #erl_scan{text=false}, Line, no_col=Col, Q, U, [], Wcs, Uni) -> - scan_string_no_col(Cs, Line, Col, Q, U, Wcs, Uni); -scan_string0(Cs, #erl_scan{text=true}, Line, no_col=Col, Q, U, Str, Wcs, Uni) -> - scan_string1(Cs, Line, Col, Q, U, Str, Wcs, Uni); -scan_string0(Cs, _St, Line, Col, Q, U, [], Wcs, Uni) -> - scan_string_col(Cs, Line, Col, Q, U, Wcs, Uni); -scan_string0(Cs, _St, Line, Col, Q, U, Str, Wcs, Uni) -> - scan_string1(Cs, Line, Col, Q, U, Str, Wcs, Uni). +scan_string0(Cs, #erl_scan{text=false}, Line, no_col=Col, Q, [], Wcs) -> + scan_string_no_col(Cs, Line, Col, Q, Wcs); +scan_string0(Cs, #erl_scan{text=true}, Line, no_col=Col, Q, Str, Wcs) -> + scan_string1(Cs, Line, Col, Q, Str, Wcs); +scan_string0(Cs, St, Line, Col, Q, [], Wcs) -> + scan_string_col(Cs, St, Line, Col, Q, Wcs); +scan_string0(Cs, _St, Line, Col, Q, Str, Wcs) -> + scan_string1(Cs, Line, Col, Q, Str, Wcs). %% Optimization. Col =:= no_col. -scan_string_no_col([Q|Cs], Line, Col, Q, _U, Wcs, Uni) -> - {Cs,Line,Col,_DontCare=[],lists:reverse(Wcs),Uni}; -scan_string_no_col([$\n=C|Cs], Line, Col, Q, U, Wcs, Uni) -> - scan_string_no_col(Cs, Line+1, Col, Q, U, [C|Wcs], Uni); -scan_string_no_col([C|Cs], Line, Col, Q, U, Wcs, Uni) when C =/= $\\, - ?CHAR(C), - ?UNI255(C) -> - scan_string_no_col(Cs, Line, Col, Q, U, [C|Wcs], Uni); -scan_string_no_col(Cs, Line, Col, Q, U, Wcs, Uni) -> - scan_string1(Cs, Line, Col, Q, U, Wcs, Wcs, Uni). +scan_string_no_col([Q|Cs], Line, Col, Q, Wcs) -> + {Cs,Line,Col,_DontCare=[],lists:reverse(Wcs)}; +scan_string_no_col([$\n=C|Cs], Line, Col, Q, Wcs) -> + scan_string_no_col(Cs, Line+1, Col, Q, [C|Wcs]); +scan_string_no_col([C|Cs], Line, Col, Q, Wcs) when C =/= $\\, ?UNICODE(C) -> + scan_string_no_col(Cs, Line, Col, Q, [C|Wcs]); +scan_string_no_col(Cs, Line, Col, Q, Wcs) -> + scan_string1(Cs, Line, Col, Q, Wcs, Wcs). %% Optimization. Col =/= no_col. -scan_string_col([Q|Cs], Line, Col, Q, _U, Wcs0, Uni) -> +scan_string_col([Q|Cs], St, Line, Col, Q, Wcs0) -> Wcs = lists:reverse(Wcs0), - Str = [Q|Wcs++[Q]], - {Cs,Line,Col+1,Str,Wcs,Uni}; -scan_string_col([$\n=C|Cs], Line, _xCol, Q, U, Wcs, Uni) -> - scan_string_col(Cs, Line+1, 1, Q, U, [C|Wcs], Uni); -scan_string_col([C|Cs], Line, Col, Q, U, Wcs, Uni) when C =/= $\\, - ?CHAR(C), - ?UNI255(C) -> - scan_string_col(Cs, Line, Col+1, Q, U, [C|Wcs], Uni); -scan_string_col(Cs, Line, Col, Q, U, Wcs, Uni) -> - scan_string1(Cs, Line, Col, Q, U, Wcs, Wcs, Uni). - -%% UNI_STR is to be replaced by STR when the Unicode-string-to-list -%% workaround is eventually removed. --define(UNI_STR(Col, S), S). + Str = ?STR(St, [Q|Wcs++[Q]]), + {Cs,Line,Col+1,Str,Wcs}; +scan_string_col([$\n=C|Cs], St, Line, _xCol, Q, Wcs) -> + scan_string_col(Cs, St, Line+1, 1, Q, [C|Wcs]); +scan_string_col([C|Cs], St, Line, Col, Q, Wcs) when C =/= $\\, ?UNICODE(C) -> + scan_string_col(Cs, St, Line, Col+1, Q, [C|Wcs]); +scan_string_col(Cs, _St, Line, Col, Q, Wcs) -> + scan_string1(Cs, Line, Col, Q, Wcs, Wcs). %% Note: in those cases when a 'char_error' tuple is returned below it %% is tempting to skip over characters up to the first Q character, %% but then the end location of the error tuple would not correspond %% to the start location of the returned Rest string. (Maybe the end %% location could be modified, but that too is ugly.) -scan_string1([Q|Cs], Line, Col, Q, _U, Str0, Wcs0, Uni) -> +scan_string1([Q|Cs], Line, Col, Q, Str0, Wcs0) -> Wcs = lists:reverse(Wcs0), - Str = ?UNI_STR(Col, [Q|lists:reverse(Str0, [Q])]), - {Cs,Line,incr_column(Col, 1),Str,Wcs,Uni}; -scan_string1([$\n=C|Cs], Line, Col, Q, U, Str, Wcs, Uni) -> + Str = [Q|lists:reverse(Str0, [Q])], + {Cs,Line,incr_column(Col, 1),Str,Wcs}; +scan_string1([$\n=C|Cs], Line, Col, Q, Str, Wcs) -> Ncol = new_column(Col, 1), - scan_string1(Cs, Line+1, Ncol, Q, U, ?UNI_STR(Col, [C|Str]), [C|Wcs], Uni); -scan_string1([$\\|Cs]=Cs0, Line, Col, Q, U, Str, Wcs, Uni) -> + scan_string1(Cs, Line+1, Ncol, Q, [C|Str], [C|Wcs]); +scan_string1([$\\|Cs]=Cs0, Line, Col, Q, Str, Wcs) -> case scan_escape(Cs, Col) of more -> - {more,Cs0,Line,Col,Str,Wcs,Uni}; + {more,Cs0,Line,Col,Str,Wcs}; {error,Ncs,Error,Ncol} -> {char_error,Ncs,Error,Line,Col,incr_column(Ncol, 1)}; {eof,Ncol} -> {error,Line,incr_column(Ncol, 1),lists:reverse(Wcs),eof}; {nl,Val,ValStr,Ncs,Ncol} -> - Nstr = ?UNI_STR(Ncol, lists:reverse(ValStr, [$\\|Str])), + Nstr = lists:reverse(ValStr, [$\\|Str]), Nwcs = [Val|Wcs], - scan_string1(Ncs, Line+1, Ncol, Q, U, Nstr, Nwcs, Uni); - {unicode,_Val,_ValStr,Ncs,Ncol} when not U -> %' Emacs - {char_error,Ncs,{illegal,character},Line,Col,incr_column(Ncol, 1)}; - {unicode,Val,ValStr,Ncs,Ncol} -> % UNI. Uni is set to Val. - Nstr = ?UNI_STR(Ncol, lists:reverse(ValStr, [$\\|Str])), - Nwcs = [Val|Wcs], % not used - scan_string1(Ncs, Line, incr_column(Ncol, 1), Q, U, Nstr, Nwcs, Val); + scan_string1(Ncs, Line+1, Ncol, Q, Nstr, Nwcs); {Val,ValStr,Ncs,Ncol} -> - Nstr = ?UNI_STR(Ncol, lists:reverse(ValStr, [$\\|Str])), + Nstr = lists:reverse(ValStr, [$\\|Str]), Nwcs = [Val|Wcs], - scan_string1(Ncs, Line, incr_column(Ncol, 1), Q, U, Nstr, Nwcs, Uni) + scan_string1(Ncs, Line, incr_column(Ncol, 1), Q, Nstr, Nwcs) end; -scan_string1([C|Cs], Line, no_col=Col, Q, U, Str, Wcs, Uni) when ?CHAR(C), - ?UNI255(C) -> - %% scan_string1(Cs, Line, Col, Q, U, Str, [C|Wcs], Uni); - scan_string1(Cs, Line, Col, Q, U, [C|Str], [C|Wcs], Uni); % UNI -scan_string1([C|Cs], Line, Col, Q, U, Str, Wcs, Uni) when ?CHAR(C), ?UNI255(C) -> - scan_string1(Cs, Line, Col+1, Q, U, [C|Str], [C|Wcs], Uni); -scan_string1([C|Cs], Line, Col, _Q, false, _Str, _Wcs, _Uni) when ?CHAR(C) -> %' UNI - {char_error,Cs,{illegal,character},Line,Col,incr_column(Col, 1)}; -scan_string1([C|Cs], Line, Col, Q, U, Str, Wcs, _Uni) when ?UNICODE(C) -> - scan_string1(Cs, Line, incr_column(Col, 1), Q, U, [C|Str], [C|Wcs], C); -scan_string1([C|Cs], Line, Col, _Q, _U, _Str, _Wcs, _Uni) when ?CHAR(C) -> % UNI +scan_string1([C|Cs], Line, no_col=Col, Q, Str, Wcs) when ?UNICODE(C) -> + scan_string1(Cs, Line, Col, Q, [C|Str], [C|Wcs]); +scan_string1([C|Cs], Line, Col, Q, Str, Wcs) when ?UNICODE(C) -> + scan_string1(Cs, Line, Col+1, Q, [C|Str], [C|Wcs]); +scan_string1([C|Cs], Line, Col, _Q, _Str, _Wcs) when ?CHAR(C) -> {char_error,Cs,{illegal,character},Line,Col,incr_column(Col, 1)}; -scan_string1([]=Cs, Line, Col, _Q, _U, Str, Wcs, Uni) -> - {more,Cs,Line,Col,Str,Wcs,Uni}; -scan_string1(eof, Line, Col, _Q, _U, _Str, Wcs, _Uni) -> +scan_string1([]=Cs, Line, Col, _Q, Str, Wcs) -> + {more,Cs,Line,Col,Str,Wcs}; +scan_string1(eof, Line, Col, _Q, _Str, Wcs) -> {error,Line,Col,lists:reverse(Wcs),eof}. -define(OCT(C), C >= $0, C =< $7). @@ -1077,16 +980,16 @@ scan_string1(eof, Line, Col, _Q, _U, _Str, Wcs, _Uni) -> %% \<1-3> octal digits scan_escape([O1,O2,O3|Cs], Col) when ?OCT(O1), ?OCT(O2), ?OCT(O3) -> Val = (O1*8 + O2)*8 + O3 - 73*$0, - {Val,?UNI_STR(Col, [O1,O2,O3]),Cs,incr_column(Col, 3)}; + {Val,[O1,O2,O3],Cs,incr_column(Col, 3)}; scan_escape([O1,O2], _Col) when ?OCT(O1), ?OCT(O2) -> more; scan_escape([O1,O2|Cs], Col) when ?OCT(O1), ?OCT(O2) -> Val = (O1*8 + O2) - 9*$0, - {Val,?UNI_STR(Col, [O1,O2]),Cs,incr_column(Col, 2)}; + {Val,[O1,O2],Cs,incr_column(Col, 2)}; scan_escape([O1], _Col) when ?OCT(O1) -> more; scan_escape([O1|Cs], Col) when ?OCT(O1) -> - {O1 - $0,?UNI_STR(Col, [O1]),Cs,incr_column(Col, 1)}; + {O1 - $0,[O1],Cs,incr_column(Col, 1)}; %% \x{<hex digits>} scan_escape([$x,${|Cs], Col) -> scan_hex(Cs, incr_column(Col, 2), []); @@ -1097,29 +1000,27 @@ scan_escape([$x|eof], Col) -> %% \x<2> hexadecimal digits scan_escape([$x,H1,H2|Cs], Col) when ?HEX(H1), ?HEX(H2) -> Val = erlang:list_to_integer([H1,H2], 16), - {Val,?UNI_STR(Col, [$x,H1,H2]),Cs,incr_column(Col, 3)}; + {Val,[$x,H1,H2],Cs,incr_column(Col, 3)}; scan_escape([$x,H1], _Col) when ?HEX(H1) -> more; scan_escape([$x|Cs], Col) -> {error,Cs,{illegal,character},incr_column(Col, 1)}; %% \^X -> CTL-X scan_escape([$^=C0,$\n=C|Cs], Col) -> - {nl,C,?UNI_STR(Col, [C0,C]),Cs,new_column(Col, 1)}; + {nl,C,[C0,C],Cs,new_column(Col, 1)}; scan_escape([$^=C0,C|Cs], Col) when ?CHAR(C) -> Val = C band 31, - {Val,?UNI_STR(Col, [C0,C]),Cs,incr_column(Col, 2)}; + {Val,[C0,C],Cs,incr_column(Col, 2)}; scan_escape([$^], _Col) -> more; scan_escape([$^|eof], Col) -> {eof,incr_column(Col, 1)}; scan_escape([$\n=C|Cs], Col) -> - {nl,C,?UNI_STR(Col, [C]),Cs,new_column(Col, 1)}; -scan_escape([C0|Cs], Col) when ?CHAR(C0), ?UNI255(C0) -> + {nl,C,[C],Cs,new_column(Col, 1)}; +scan_escape([C0|Cs], Col) when ?UNICODE(C0) -> C = escape_char(C0), - {C,?UNI_STR(Col, [C0]),Cs,incr_column(Col, 1)}; -scan_escape([C|Cs], Col) when ?UNICODE(C) -> - {unicode,C,?UNI_STR(Col, [C]),Cs,incr_column(Col, 1)}; -scan_escape([C|Cs], Col) when ?CHAR(C) -> % UNI + {C,[C0],Cs,incr_column(Col, 1)}; +scan_escape([C|Cs], Col) when ?CHAR(C) -> {error,Cs,{illegal,character},incr_column(Col, 1)}; scan_escape([], _Col) -> more; @@ -1136,10 +1037,8 @@ scan_hex(Cs, Col, Wcs) -> scan_esc_end([$}|Cs], Col, Wcs0, B, Str0) -> Wcs = lists:reverse(Wcs0), case catch erlang:list_to_integer(Wcs, B) of - Val when Val =< 16#FF -> - {Val,?UNI_STR(Col, Str0++Wcs++[$}]),Cs,incr_column(Col, 1)}; Val when ?UNICODE(Val) -> - {unicode,Val,?UNI_STR(Col, Str0++Wcs++[$}]),Cs,incr_column(Col,1)}; + {Val,Str0++Wcs++[$}],Cs,incr_column(Col, 1)}; _ -> {error,Cs,{illegal,character},incr_column(Col, 1)} end; @@ -1171,7 +1070,7 @@ scan_number([$#|Cs]=Cs0, St, Line, Col, Toks, Ncs0) -> Ncs = lists:reverse(Ncs0), case catch list_to_integer(Ncs) of B when B >= 2, B =< 1+$Z-$A+10 -> - Bcs = Ncs++[$#], + Bcs = ?STR(St, Ncs++[$#]), scan_based_int(Cs, St, Line, Col, Toks, {B,[],Bcs}); B -> Len = length(Ncs), @@ -1204,7 +1103,7 @@ scan_based_int(Cs, St, Line, Col, Toks, {B,Ncs0,Bcs}) -> Ncs = lists:reverse(Ncs0), case catch erlang:list_to_integer(Ncs, B) of N when is_integer(N) -> - tok3(Cs, St, Line, Col, Toks, integer, Bcs++Ncs, N); + tok3(Cs, St, Line, Col, Toks, integer, ?STR(St, Bcs++Ncs), N); _ -> Len = length(Bcs)+length(Ncs), Ncol = incr_column(Col, Len), @@ -1244,36 +1143,30 @@ float_end(Cs, St, Line, Col, Toks, Ncs0) -> scan_error({illegal,float}, Line, Col, Line, Ncol, Cs) end. -skip_comment(Cs, St, Line, Col, Toks, N) -> - skip_comment(Cs, St, Line, Col, Toks, N, St#erl_scan.unicode). - -skip_comment([C|Cs], St, Line, Col, Toks, N, U) when C =/= $\n, ?CHAR(C) -> - case ?UNI255(C) orelse U andalso ?UNICODE(C) of +skip_comment([C|Cs], St, Line, Col, Toks, N) when C =/= $\n, ?CHAR(C) -> + case ?UNICODE(C) of true -> - skip_comment(Cs, St, Line, Col, Toks, N+1, U); + skip_comment(Cs, St, Line, Col, Toks, N+1); false -> Ncol = incr_column(Col, N+1), scan_error({illegal,character}, Line, Col, Line, Ncol, Cs) end; -skip_comment([]=Cs, _St, Line, Col, Toks, N, _U) -> +skip_comment([]=Cs, _St, Line, Col, Toks, N) -> {more,{Cs,Col,Toks,Line,N,fun skip_comment/6}}; -skip_comment(Cs, St, Line, Col, Toks, N, _U) -> +skip_comment(Cs, St, Line, Col, Toks, N) -> scan1(Cs, St, Line, incr_column(Col, N), Toks). -scan_comment(Cs, St, Line, Col, Toks, Ncs) -> - scan_comment(Cs, St, Line, Col, Toks, Ncs, St#erl_scan.unicode). - -scan_comment([C|Cs], St, Line, Col, Toks, Ncs, U) when C =/= $\n, ?CHAR(C) -> - case ?UNI255(C) orelse U andalso ?UNICODE(C) of +scan_comment([C|Cs], St, Line, Col, Toks, Ncs) when C =/= $\n, ?CHAR(C) -> + case ?UNICODE(C) of true -> - scan_comment(Cs, St, Line, Col, Toks, [C|Ncs], U); + scan_comment(Cs, St, Line, Col, Toks, [C|Ncs]); false -> Ncol = incr_column(Col, length(Ncs)+1), scan_error({illegal,character}, Line, Col, Line, Ncol, Cs) end; -scan_comment([]=Cs, _St, Line, Col, Toks, Ncs, _U) -> +scan_comment([]=Cs, _St, Line, Col, Toks, Ncs) -> {more,{Cs,Col,Toks,Line,Ncs,fun scan_comment/6}}; -scan_comment(Cs, St, Line, Col, Toks, Ncs0, _U) -> +scan_comment(Cs, St, Line, Col, Toks, Ncs0) -> Ncs = lists:reverse(Ncs0), tok3(Cs, St, Line, Col, Toks, comment, Ncs, Ncs). diff --git a/lib/stdlib/test/erl_scan_SUITE.erl b/lib/stdlib/test/erl_scan_SUITE.erl index 3f77d40a2e..ecd181e87c 100644 --- a/lib/stdlib/test/erl_scan_SUITE.erl +++ b/lib/stdlib/test/erl_scan_SUITE.erl @@ -118,13 +118,13 @@ check(String) -> %%% (This should be useful for all format_error functions.) check_error({error, Info, EndLine}, Module0) -> - ?line {ErrorLine, Module, Desc} = Info, - ?line true = (Module == Module0), - ?line assert_type(EndLine, integer), - ?line assert_type(ErrorLine, integer), - ?line true = (ErrorLine =< EndLine), - ?line String = lists:flatten(Module0:format_error(Desc)), - ?line true = io_lib:printable_list(String). + {ErrorLine, Module, Desc} = Info, + true = (Module == Module0), + assert_type(EndLine, integer), + assert_type(ErrorLine, integer), + true = (ErrorLine =< EndLine), + String = lists:flatten(Module0:format_error(Desc)), + true = io_lib:printable_list(String). iso88591(doc) -> ["Tests the support for ISO-8859-1 i.e Latin-1"]; iso88591(suite) -> []; @@ -809,77 +809,57 @@ white_spaces() -> unicode() -> ?line {ok,[{char,1,83},{integer,1,45}],1} = - erl_scan:string("$\\12345", 1, [{unicode,false}]), % not unicode + erl_scan:string("$\\12345"), % not unicode ?line {error,{1,erl_scan,{illegal,character}},1} = - erl_scan:string([1089], 1, [{unicode,false}]), + erl_scan:string([1089]), ?line {error,{{1,1},erl_scan,{illegal,character}},{1,2}} = - erl_scan:string([1089], {1,1}, [{unicode,false}]), - ?line {error,{1,erl_scan,{illegal,character}},1} = - %% ?line {error,{1,erl_scan,{illegal,atom}},1} = - erl_scan:string("'a"++[1089]++"b'", 1, [{unicode,false}]), - ?line {error,{{1,3},erl_scan,{illegal,character}},{1,4}} = - erl_scan:string("'a"++[1089]++"b'", {1,1}, [{unicode,false}]), + erl_scan:string([1089], {1,1}), + ?line {error,{1,erl_scan,{illegal,atom}},1} = + erl_scan:string("'a"++[1089]++"b'", 1), + ?line {error,{{1,1},erl_scan,{illegal,atom}},{1,6}} = + erl_scan:string("'a"++[1089]++"b'", {1,1}), ?line test("\"a"++[1089]++"b\""), ?line {ok,[{char,1,1}],1} = - erl_scan:string([$$,$\\,$^,1089], 1, [{unicode,false}]), + erl_scan:string([$$,$\\,$^,1089], 1), ?line {error,{1,erl_scan,Error},1} = - erl_scan:string("\"qa\x{aaa}", 1, [{unicode,false}]), + erl_scan:string("\"qa\x{aaa}", 1), ?line "unterminated string starting with \"qa"++[2730]++"\"" = erl_scan:format_error(Error), ?line {error,{{1,1},erl_scan,_},{1,11}} = - erl_scan:string("\"qa\\x{aaa}",{1,1}, [{unicode,false}]), - ?line {error,{{1,4},erl_scan,{illegal,character}},{1,11}} = - erl_scan:string("'qa\\x{aaa}'",{1,1}, [{unicode,false}]), - - Tags = [category, column, length, line, symbol, text], - - %% Workaround. No character codes greater than 255! To be changed. - %% Note: don't remove these tests, just modify them! + erl_scan:string("\"qa\\x{aaa}",{1,1}), + ?line {error,{{1,1},erl_scan,{illegal,atom}},{1,12}} = + erl_scan:string("'qa\\x{aaa}'",{1,1}), - ?line {ok,[{integer,1,1089}],1} = - erl_scan:string([$$,1089], 1, [{unicode,false}]), - ?line {ok,[{integer,1,1089}],1} = - erl_scan:string([$$,$\\,1089], 1, [{unicode,false}]), + ?line {ok,[{char,1,1089}],1} = + erl_scan:string([$$,1089], 1), + ?line {ok,[{char,1,1089}],1} = + erl_scan:string([$$,$\\,1089], 1), Qs = "$\\x{aaa}", - ?line {ok,[{integer,1,16#aaa}],1} = - erl_scan:string(Qs, 1, [{unicode,false}]), + ?line {ok,[{char,1,$\x{aaa}}],1} = + erl_scan:string(Qs, 1), ?line {ok,[Q2],{1,9}} = - erl_scan:string("$\\x{aaa}", {1,1}, [text,{unicode,false}]), - ?line [{category,integer},{column,1},{length,8}, + erl_scan:string("$\\x{aaa}", {1,1}, [text]), + ?line [{category,char},{column,1},{length,8}, {line,1},{symbol,16#aaa},{text,Qs}] = erl_scan:token_info(Q2), U1 = "\"\\x{aaa}\"", - ?line {ok,[T1,T2,T3],{1,10}} = - erl_scan:string(U1, {1,1}, [text,{unicode,false}]), - ?line [{category,'['},{column,1},{length,1},{line,1}, - {symbol,'['},{text,"\""}] = erl_scan:token_info(T1, Tags), - ?line [{category,integer},{column,2},{length,7}, - {line,1},{symbol,16#aaa},{text,"\\x{aaa}"}] = - erl_scan:token_info(T2, Tags), - ?line [{category,']'},{column,9},{length,1},{line,1}, - {symbol,']'},{text,"\""}] = erl_scan:token_info(T3, Tags), - ?line {ok,[{'[',1},{integer,1,16#aaa},{']',1}],1} = - erl_scan:string(U1, 1, [{unicode,false}]), + {ok, + [{string,[{line,1},{column,1},{text,"\"\\x{aaa}\""}],[2730]}], + {1,10}} = erl_scan:string(U1, {1,1}, [text]), + {ok,[{string,1,[2730]}],1} = erl_scan:string(U1, 1), U2 = "\"\\x41\\x{fff}\\x42\"", - ?line {ok,[{'[',1},{char,1,16#41},{',',1},{integer,1,16#fff}, - {',',1},{char,1,16#42},{']',1}],1} = - erl_scan:string(U2, 1, [{unicode,false}]), + {ok,[{string,1,[$\x41,$\x{fff},$\x42]}],1} = erl_scan:string(U2, 1), U3 = "\"a\n\\x{fff}\n\"", - ?line {ok,[{'[',1},{char,1,$a},{',',1},{char,1,$\n}, - {',',2},{integer,2,16#fff},{',',2},{char,2,$\n}, - {']',3}],3} = - erl_scan:string(U3, 1, [{unicode,false}]), + {ok,[{string,1,[$a,$\n,$\x{fff},$\n]}],3} = erl_scan:string(U3, 1), U4 = "\"\\^\n\\x{aaa}\\^\n\"", - ?line {ok,[{'[',1},{char,1,$\n},{',',2},{integer,2,16#aaa}, - {',',2},{char,2,$\n},{']',3}],3} = - erl_scan:string(U4, 1, [{unicode,false}]), + {ok,[{string,1,[$\n,$\x{aaa},$\n]}],3} = erl_scan:string(U4, 1), %% Keep these tests: ?line test(Qs), @@ -889,21 +869,15 @@ unicode() -> ?line test(U4), Str1 = "\"ab" ++ [1089] ++ "cd\"", - ?line {ok,[{'[',1},{char,1,$a},{',',1},{char,1,$b},{',',1}, - {integer,1,1089},{',',1},{char,1,$c},{',',1}, - {char,1,$d},{']',1}],1} = - erl_scan:string(Str1, 1, [{unicode,false}]), - ?line {ok,[{'[',_},{char,_,$a},{',',_},{char,_,$b},{',',_}, - {integer,_,1089},{',',_},{char,_,$c},{',',_}, - {char,_,$d},{']',_}],{1,8}} = - erl_scan:string(Str1, {1,1}, [{unicode,false}]), + {ok,[{string,1,[$a,$b,1089,$c,$d]}],1} = erl_scan:string(Str1, 1), + {ok,[{string,{1,1},[$a,$b,1089,$c,$d]}],{1,8}} = + erl_scan:string(Str1, {1,1}), ?line test(Str1), Comment = "%% "++[1089], - %% Returned a comment In R15B03: - {error,{1,erl_scan,{illegal,character}},1} = - erl_scan:string(Comment, 1, [return,{unicode,false}]), - {error,{{1,1},erl_scan,{illegal,character}},{1,5}} = - erl_scan:string(Comment, {1,1}, [return,{unicode,false}]), + {ok,[{comment,1,[$%,$%,$\s,1089]}],1} = + erl_scan:string(Comment, 1, [return]), + {ok,[{comment,{1,1},[$%,$%,$\s,1089]}],{1,5}} = + erl_scan:string(Comment, {1,1}, [return]), ok. more_chars() -> |