diff options
Diffstat (limited to 'lib/xmerl/src')
| -rw-r--r-- | lib/xmerl/src/xmerl_regexp.erl | 4 | ||||
| -rw-r--r-- | lib/xmerl/src/xmerl_sax_parser.erl | 33 | ||||
| -rw-r--r-- | lib/xmerl/src/xmerl_sax_parser.hrl | 11 | ||||
| -rw-r--r-- | lib/xmerl/src/xmerl_sax_parser_base.erlsrc | 144 | ||||
| -rw-r--r-- | lib/xmerl/src/xmerl_sax_parser_latin1.erlsrc | 38 | ||||
| -rw-r--r-- | lib/xmerl/src/xmerl_sax_parser_list.erlsrc | 21 | ||||
| -rw-r--r-- | lib/xmerl/src/xmerl_sax_parser_utf16be.erlsrc | 50 | ||||
| -rw-r--r-- | lib/xmerl/src/xmerl_sax_parser_utf16le.erlsrc | 50 | ||||
| -rw-r--r-- | lib/xmerl/src/xmerl_sax_parser_utf8.erlsrc | 50 | ||||
| -rw-r--r-- | lib/xmerl/src/xmerl_scan.erl | 37 |
10 files changed, 299 insertions, 139 deletions
diff --git a/lib/xmerl/src/xmerl_regexp.erl b/lib/xmerl/src/xmerl_regexp.erl index fc89b80ff1..1bf8496673 100644 --- a/lib/xmerl/src/xmerl_regexp.erl +++ b/lib/xmerl/src/xmerl_regexp.erl @@ -1,7 +1,7 @@ %% %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2006-2016. All Rights Reserved. +%% Copyright Ericsson AB 2006-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -1154,7 +1154,7 @@ comp_crs([], Last) -> [{Last,maxchar}]. %% build_dfa(NFA, NfaStartState) -> {DFA,DfaStartState}. %% Build a DFA from an NFA using "subset construction". The major %% difference from the book is that we keep the marked and unmarked -%% DFA states in seperate lists. New DFA states are added to the +%% DFA states in separate lists. New DFA states are added to the %% unmarked list and states are marked by moving them to the marked %% list. We assume that the NFA accepting state numbers are in %% ascending order for the rules and use ordsets to keep this order. diff --git a/lib/xmerl/src/xmerl_sax_parser.erl b/lib/xmerl/src/xmerl_sax_parser.erl index 318a0cf7f4..1aef6c58c4 100644 --- a/lib/xmerl/src/xmerl_sax_parser.erl +++ b/lib/xmerl/src/xmerl_sax_parser.erl @@ -1,7 +1,7 @@ %%-------------------------------------------------------------------- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2008-2016. All Rights Reserved. +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -33,6 +33,7 @@ %% External exports %%---------------------------------------------------------------------- -export([file/2, + stream/3, stream/2]). %%---------------------------------------------------------------------- @@ -72,11 +73,12 @@ file(Name,Options) -> File = filename:basename(Name), ContinuationFun = fun default_continuation_cb/1, Res = stream(<<>>, - [{continuation_fun, ContinuationFun}, - {continuation_state, FD}, - {current_location, CL}, - {entity, File} - |Options]), + [{continuation_fun, ContinuationFun}, + {continuation_state, FD}, + {current_location, CL}, + {entity, File} + |Options], + file), ok = file:close(FD), Res end. @@ -92,19 +94,22 @@ file(Name,Options) -> %% EventState = term() %% Description: Parse a stream containing an XML document. %%---------------------------------------------------------------------- -stream(Xml, Options) when is_list(Xml), is_list(Options) -> +stream(Xml, Options) -> + stream(Xml, Options, stream). + +stream(Xml, Options, InputType) when is_list(Xml), is_list(Options) -> State = parse_options(Options, initial_state()), - case State#xmerl_sax_parser_state.file_type of + case State#xmerl_sax_parser_state.file_type of dtd -> xmerl_sax_parser_list:parse_dtd(Xml, State#xmerl_sax_parser_state{encoding = list, - input_type = stream}); + input_type = InputType}); normal -> xmerl_sax_parser_list:parse(Xml, State#xmerl_sax_parser_state{encoding = list, - input_type = stream}) + input_type = InputType}) end; -stream(Xml, Options) when is_binary(Xml), is_list(Options) -> +stream(Xml, Options, InputType) when is_binary(Xml), is_list(Options) -> case parse_options(Options, initial_state()) of {error, Reason} -> {error, Reason}; State -> @@ -127,7 +132,7 @@ stream(Xml, Options) when is_binary(Xml), is_list(Options) -> State#xmerl_sax_parser_state.event_state}; {Xml1, State1} -> parse_binary(Xml1, - State1#xmerl_sax_parser_state{input_type = stream}, + State1#xmerl_sax_parser_state{input_type = InputType}, ParseFunction) end end. @@ -226,12 +231,12 @@ check_encoding_option(E) -> %% Description: Detects which character set is used in a binary stream. %%---------------------------------------------------------------------- detect_charset(<<>>, #xmerl_sax_parser_state{continuation_fun = undefined} = _) -> - throw({error, "Can't detect character encoding due to no indata"}); + {error, "Can't detect character encoding due to no indata"}; detect_charset(<<>>, #xmerl_sax_parser_state{continuation_fun = CFun, continuation_state = CState} = State) -> case CFun(CState) of {<<>>, _} -> - throw({error, "Can't detect character encoding due to lack of indata"}); + {error, "Can't detect character encoding due to lack of indata"}; {NewBytes, NewContState} -> detect_charset(NewBytes, State#xmerl_sax_parser_state{continuation_state = NewContState}) end; diff --git a/lib/xmerl/src/xmerl_sax_parser.hrl b/lib/xmerl/src/xmerl_sax_parser.hrl index 932ab0cec5..56a3a42e5f 100644 --- a/lib/xmerl/src/xmerl_sax_parser.hrl +++ b/lib/xmerl/src/xmerl_sax_parser.hrl @@ -1,7 +1,7 @@ %%-------------------------------------------------------------------- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2008-2016. All Rights Reserved. +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -88,14 +88,7 @@ current_location, % Location of the currently parsed XML entity entity, % Parsed XML entity skip_external_dtd = false,% If true the external DTD is skipped during parsing - input_type % Source type: file | stream. - % This field is a preparation for an fix in R17 of a bug in - % the conformance against the standard. - % Today a file which contains two XML documents will be considered - % well-formed and the second is placed in the rest part of the - % return tuple, according to the conformance tests this should fail. - % In the future this will fail if xmerl_sax_aprser:file/2 is used but - % left to the user in the xmerl_sax_aprser:stream/2 case. + input_type % Source type: file | stream }). diff --git a/lib/xmerl/src/xmerl_sax_parser_base.erlsrc b/lib/xmerl/src/xmerl_sax_parser_base.erlsrc index 4d75805b9b..f3470b2809 100644 --- a/lib/xmerl/src/xmerl_sax_parser_base.erlsrc +++ b/lib/xmerl/src/xmerl_sax_parser_base.erlsrc @@ -1,7 +1,7 @@ %%-*-erlang-*- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2008-2016. All Rights Reserved. +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -72,7 +72,12 @@ parse(Xml, State) -> {ok, Rest, State2} -> State3 = event_callback(endDocument, State2), ets:delete(RefTable), - {ok, State3#xmerl_sax_parser_state.event_state, Rest}; + case check_if_rest_ok(State3#xmerl_sax_parser_state.input_type, Rest) of + true -> + {ok, State3#xmerl_sax_parser_state.event_state, Rest}; + false -> + format_error(fatal_error, State3, "Input found after legal document") + end; {fatal_error, {State2, Reason}} -> State3 = event_callback(endDocument, State2), ets:delete(RefTable), @@ -81,10 +86,14 @@ parse(Xml, State) -> State3 = event_callback(endDocument, State2), ets:delete(RefTable), format_error(Tag, State3, Reason); + {endDocument, Rest, State2} -> + State3 = event_callback(endDocument, State2), + ets:delete(RefTable), + {ok, State3#xmerl_sax_parser_state.event_state, Rest}; Other -> _State2 = event_callback(endDocument, State1), ets:delete(RefTable), - throw(Other) + {fatal_error, Other} end. %%---------------------------------------------------------------------- @@ -111,7 +120,7 @@ parse_dtd(Xml, State) -> {Rest, State2} when is_record(State2, xmerl_sax_parser_state) -> State3 = event_callback(endDocument, State2), ets:delete(RefTable), - {ok, State3#xmerl_sax_parser_state.event_state, Rest}; + {ok, State3#xmerl_sax_parser_state.event_state, Rest}; {endDocument, Rest, State2} when is_record(State2, xmerl_sax_parser_state) -> State3 = event_callback(endDocument, State2), ets:delete(RefTable), @@ -119,7 +128,7 @@ parse_dtd(Xml, State) -> Other -> _State2 = event_callback(endDocument, State1), ets:delete(RefTable), - throw(Other) + {fatal_error, Other} end. @@ -136,10 +145,11 @@ parse_dtd(Xml, State) -> %% [1] document ::= prolog element Misc* %%---------------------------------------------------------------------- parse_document(Rest, State) when is_record(State, xmerl_sax_parser_state) -> - {Rest1, State1} = parse_xml_decl(Rest, State), + {Rest1, State1} = parse_byte_order_mark(Rest, State), {Rest2, State2} = parse_misc(Rest1, State1, true), {ok, Rest2, State2}. +?PARSE_BYTE_ORDER_MARK(Bytes, State). %%---------------------------------------------------------------------- %% Function: parse_xml_decl(Rest, State) -> Result @@ -150,15 +160,8 @@ parse_document(Rest, State) when is_record(State, xmerl_sax_parser_state) -> %% [22] prolog ::= XMLDecl? Misc* (doctypedecl Misc*)? %% [23] XMLDecl ::= '<?xml' VersionInfo EncodingDecl? SDDecl? S? '?>' %%---------------------------------------------------------------------- --dialyzer({[no_fail_call, no_match], parse_xml_decl/2}). parse_xml_decl(?STRING_EMPTY, State) -> cf(?STRING_EMPTY, State, fun parse_xml_decl/2); -parse_xml_decl(?BYTE_ORDER_MARK_1, State) -> - cf(?BYTE_ORDER_MARK_1, State, fun parse_xml_decl/2); -parse_xml_decl(?BYTE_ORDER_MARK_2, State) -> - cf(?BYTE_ORDER_MARK_2, State, fun parse_xml_decl/2); -parse_xml_decl(?BYTE_ORDER_MARK_REST(Rest), State) -> - cf(Rest, State, fun parse_xml_decl/2); parse_xml_decl(?STRING("<") = Bytes, State) -> cf(Bytes, State, fun parse_xml_decl/2); parse_xml_decl(?STRING("<?") = Bytes, State) -> @@ -170,31 +173,19 @@ parse_xml_decl(?STRING("<?xm") = Bytes, State) -> parse_xml_decl(?STRING("<?xml") = Bytes, State) -> cf(Bytes, State, fun parse_xml_decl/2); parse_xml_decl(?STRING_REST("<?xml", Rest1), State) -> - parse_xml_decl_1(Rest1, State); -parse_xml_decl(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State) when is_binary(Bytes) -> - case unicode:characters_to_list(Bytes, Enc) of - {incomplete, _, _} -> - cf(Bytes, State, fun parse_xml_decl/2); - {error, _Encoded, _Rest} -> - ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))); - _ -> - parse_prolog(Bytes, State) - end; -parse_xml_decl(Bytes, State) -> - parse_prolog(Bytes, State). - + parse_xml_decl_rest(Rest1, State); +?PARSE_XML_DECL(Bytes, State). -parse_xml_decl_1(?STRING_UNBOUND_REST(C, Rest) = Bytes, State) -> +parse_xml_decl_rest(?STRING_UNBOUND_REST(C, Rest) = Bytes, State) -> if ?is_whitespace(C) -> {_XmlAttributes, Rest1, State1} = parse_version_info(Rest, State, []), - %State2 = event_callback({processingInstruction, "xml", XmlAttributes}, State1),% The XML decl. should not be reported as a PI parse_prolog(Rest1, State1); true -> parse_prolog(?STRING_REST("<?xml", Bytes), State) end; -parse_xml_decl_1(Bytes, State) -> - unicode_incomplete_check([Bytes, State, fun parse_xml_decl_1/2], undefined). +parse_xml_decl_rest(Bytes, State) -> + unicode_incomplete_check([Bytes, State, fun parse_xml_decl_rest/2], undefined). @@ -216,8 +207,6 @@ parse_prolog(?STRING_REST("<?", Rest), State) -> parse_prolog(Rest1, State1); {endDocument, Rest1, State1} -> parse_prolog(Rest1, State1) - % IValue = ?TO_INPUT_FORMAT("<?"), - % {?APPEND_STRING(IValue, Rest1), State1} end; parse_prolog(?STRING_REST("<!", Rest), State) -> parse_prolog_1(Rest, State); @@ -230,7 +219,6 @@ parse_prolog(Bytes, State) -> unicode_incomplete_check([Bytes, State, fun parse_prolog/2], "expecting < or whitespace"). - parse_prolog_1(?STRING_EMPTY, State) -> cf(?STRING_EMPTY, State, fun parse_prolog_1/2); parse_prolog_1(?STRING("D") = Bytes, State) -> @@ -442,6 +430,15 @@ check_if_new_doc_allowed(stream, []) -> check_if_new_doc_allowed(_, _) -> false. +check_if_rest_ok(file, []) -> + true; +check_if_rest_ok(file, <<>>) -> + true; +check_if_rest_ok(stream, _) -> + true; +check_if_rest_ok(_, _) -> + false. + %%---------------------------------------------------------------------- %% Function: parse_pi_1(Rest, State) -> Result %% Input: Rest = string() | binary() @@ -1024,16 +1021,21 @@ parse_etag(Bytes, State) -> unicode_incomplete_check([Bytes, State, fun parse_etag/2], undefined). - parse_etag_1(?STRING_REST(">", Rest), #xmerl_sax_parser_state{end_tags=[{_ETag, Uri, LocalName, QName, OldNsList, NewNsList} - |RestOfETags]} = State, _Tag) -> + |RestOfETags], + input_type=InputType} = State, _Tag) -> State1 = event_callback({endElement, Uri, LocalName, QName}, State), State2 = send_end_prefix_mapping_event(NewNsList, State1), - parse_content(Rest, - State2#xmerl_sax_parser_state{end_tags=RestOfETags, - ns = OldNsList}, - [], true); + case check_if_new_doc_allowed(InputType, RestOfETags) of + true -> + throw({endDocument, Rest, State2#xmerl_sax_parser_state{ns = OldNsList}}); + false -> + parse_content(Rest, + State2#xmerl_sax_parser_state{end_tags=RestOfETags, + ns = OldNsList}, + [], true) + end; parse_etag_1(?STRING_UNBOUND_REST(_C, _), State, Tag) -> {P,TN} = Tag, ?fatal_error(State, "Bad EndTag: " ++ P ++ ":" ++ TN); @@ -1051,21 +1053,26 @@ parse_etag_1(Bytes, State, Tag) -> %% Description: Parsing the content part of tags %% [43] content ::= (element | CharData | Reference | CDSect | PI | Comment)* %%---------------------------------------------------------------------- - parse_content(?STRING_EMPTY, State, Acc, IgnorableWS) -> - case catch cf(?STRING_EMPTY, State, Acc, IgnorableWS, fun parse_content/4) of - {Rest, State1} when is_record(State1, xmerl_sax_parser_state) -> - {Rest, State1}; - {fatal_error, {State1, Msg}} -> - case check_if_document_complete(State1, Msg) of - true -> - State2 = send_character_event(length(Acc), IgnorableWS, lists:reverse(Acc), State1), - {?STRING_EMPTY, State2}; - false -> - ?fatal_error(State1, Msg) - end; - Other -> - throw(Other) + case check_if_document_complete(State, "No more bytes") of + true -> + State1 = send_character_event(length(Acc), IgnorableWS, lists:reverse(Acc), State), + {?STRING_EMPTY, State1}; + false -> + case catch cf(?STRING_EMPTY, State, Acc, IgnorableWS, fun parse_content/4) of + {Rest, State1} when is_record(State1, xmerl_sax_parser_state) -> + {Rest, State1}; + {fatal_error, {State1, Msg}} -> + case check_if_document_complete(State1, Msg) of + true -> + State2 = send_character_event(length(Acc), IgnorableWS, lists:reverse(Acc), State1), + {?STRING_EMPTY, State2}; + false -> + ?fatal_error(State1, Msg) + end; + Other -> + throw(Other) + end end; parse_content(?STRING("\r") = Bytes, State, Acc, IgnorableWS) -> cf(Bytes, State, Acc, IgnorableWS, fun parse_content/4); @@ -1094,7 +1101,7 @@ parse_content(?STRING_REST("<?", Rest), State, Acc, IgnorableWS) -> parse_content(?STRING_REST("<!", Rest1) = Rest, #xmerl_sax_parser_state{end_tags = ET} = State, Acc, IgnorableWS) -> case ET of [] -> - {Rest, State}; %%LATH : Skicka ignorable WS ??? + {Rest, State}; %% Skicka ignorable WS ??? _ -> State1 = send_character_event(length(Acc), IgnorableWS, lists:reverse(Acc), State), parse_cdata(Rest1, State1) @@ -1102,7 +1109,7 @@ parse_content(?STRING_REST("<!", Rest1) = Rest, #xmerl_sax_parser_state{end_tags parse_content(?STRING_REST("<", Rest1) = Rest, #xmerl_sax_parser_state{end_tags = ET} = State, Acc, IgnorableWS) -> case ET of [] -> - {Rest, State}; %%LATH : Skicka ignorable WS ??? + {Rest, State}; %% Skicka ignorable WS ??? _ -> State1 = send_character_event(length(Acc), IgnorableWS, lists:reverse(Acc), State), parse_stag(Rest1, State1) @@ -1204,7 +1211,6 @@ send_character_event(_, true, String, State) -> %% Description: Parse whitespaces. %% [3] S ::= (#x20 | #x9 | #xD | #xA)+ %%---------------------------------------------------------------------- --dialyzer({no_fail_call, whitespace/3}). whitespace(?STRING_EMPTY, State, Acc) -> case cf(?STRING_EMPTY, State, Acc, fun whitespace/3) of {?STRING_EMPTY, State} -> @@ -1230,16 +1236,7 @@ whitespace(?STRING_REST("\r", Rest), State, Acc) -> whitespace(Rest, State#xmerl_sax_parser_state{line_no=N+1}, [?lf |Acc]); whitespace(?STRING_UNBOUND_REST(C, Rest), State, Acc) when ?is_whitespace(C) -> whitespace(Rest, State, [C|Acc]); -whitespace(?STRING_UNBOUND_REST(_C, _) = Bytes, State, Acc) -> - {lists:reverse(Acc), Bytes, State}; -whitespace(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State, Acc) when is_binary(Bytes) -> - case unicode:characters_to_list(Bytes, Enc) of - {incomplete, _, _} -> - cf(Bytes, State, Acc, fun whitespace/3); - {error, _Encoded, _Rest} -> - ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))) - end. - +?WHITESPACE(Bytes, State, Acc). %%---------------------------------------------------------------------- %% Function: parse_reference(Rest, State, HaveToExist) -> Result @@ -1362,7 +1359,6 @@ parse_pe_reference_1(Bytes, State, Name) -> "missing ; after reference " ++ Name). - %%---------------------------------------------------------------------- %% Function: insert_reference(Reference, State) -> Result %% Parameters: Reference = string() @@ -1378,7 +1374,6 @@ insert_reference({Name, Type, Value}, Table) -> end. - %%---------------------------------------------------------------------- %% Function: look_up_reference(Reference, State) -> Result %% Parameters: Reference = string() @@ -1693,7 +1688,7 @@ handle_external_entity({http, Url}, State) -> ++ file:format_error(Reason)); {ok, FD} -> {?STRING_EMPTY, EntityState} = - parse_external_entity_1(<<>>, + parse_external_entity_byte_order_mark(<<>>, State#xmerl_sax_parser_state{continuation_state=FD, current_location=filename:dirname(Url), entity=filename:basename(Url), @@ -1709,6 +1704,8 @@ handle_external_entity({http, Url}, State) -> handle_external_entity({Tag, _Url}, State) -> ?fatal_error(State, "Unsupported URI type: " ++ atom_to_list(Tag)). +?PARSE_EXTERNAL_ENTITY_BYTE_ORDER_MARK(Bytes, State). + %%---------------------------------------------------------------------- %% Function : parse_external_entity_1(Rest, State) -> Result %% Parameters: Rest = string() | binary() @@ -1716,7 +1713,6 @@ handle_external_entity({Tag, _Url}, State) -> %% Result : {Rest, State} %% Description: Parse the external entity. %%---------------------------------------------------------------------- --dialyzer({[no_fail_call, no_match], parse_external_entity_1/2}). parse_external_entity_1(?STRING_EMPTY, #xmerl_sax_parser_state{file_type=Type} = State) -> case catch cf(?STRING_EMPTY, State, fun parse_external_entity_1/2) of {Rest, State1} when is_record(State1, xmerl_sax_parser_state) -> @@ -1726,12 +1722,6 @@ parse_external_entity_1(?STRING_EMPTY, #xmerl_sax_parser_state{file_type=Type} = Other -> throw(Other) end; -parse_external_entity_1(?BYTE_ORDER_MARK_1, State) -> - cf(?BYTE_ORDER_MARK_1, State, fun parse_external_entity_1/2); -parse_external_entity_1(?BYTE_ORDER_MARK_2, State) -> - cf(?BYTE_ORDER_MARK_2, State, fun parse_external_entity_1/2); -parse_external_entity_1(?BYTE_ORDER_MARK_REST(Rest), State) -> - parse_external_entity_1(Rest, State); parse_external_entity_1(?STRING("<") = Bytes, State) -> cf(Bytes, State, fun parse_external_entity_1/2); parse_external_entity_1(?STRING("<?") = Bytes, State) -> @@ -3290,7 +3280,7 @@ cf(Rest, #xmerl_sax_parser_state{continuation_fun = CFun, continuation_state = C catch throw:ErrorTerm -> ?fatal_error(State, ErrorTerm); - exit:Reason -> + exit:Reason -> ?fatal_error(State, {'EXIT', Reason}) end, case Result of diff --git a/lib/xmerl/src/xmerl_sax_parser_latin1.erlsrc b/lib/xmerl/src/xmerl_sax_parser_latin1.erlsrc index 961806bf4c..6e59347fb8 100644 --- a/lib/xmerl/src/xmerl_sax_parser_latin1.erlsrc +++ b/lib/xmerl/src/xmerl_sax_parser_latin1.erlsrc @@ -2,7 +2,7 @@ %%-------------------------------------------------------------------- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2008-2016. All Rights Reserved. +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -34,8 +34,36 @@ -define(APPEND_STRING(Rest, New), <<Rest/binary, New/binary>>). -define(TO_INPUT_FORMAT(Val), unicode:characters_to_binary(Val, unicode, latin1)). -%% STRING_REST and STRING_UNBOUND_REST is only different in the list case -define(STRING_UNBOUND_REST(MatchChar, Rest), <<MatchChar, Rest/binary>>). --define(BYTE_ORDER_MARK_1, undefined_bom1). --define(BYTE_ORDER_MARK_2, undefined_bom2). --define(BYTE_ORDER_MARK_REST(Rest), <<undefined, Rest/binary>>). + +-define(PARSE_BYTE_ORDER_MARK(Bytes, State), + parse_byte_order_mark(Bytes, State) -> + parse_xml_decl(Bytes, State)). + +-define(PARSE_XML_DECL(Bytes, State), + parse_xml_decl(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State) when is_binary(Bytes) -> + case unicode:characters_to_list(Bytes, Enc) of + {incomplete, _, _} -> + cf(Bytes, State, fun parse_xml_decl/2); + {error, _Encoded, _Rest} -> + ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))); + _ -> + parse_prolog(Bytes, State) + end; + parse_xml_decl(Bytes, State) -> + parse_prolog(Bytes, State)). + +-define(WHITESPACE(Bytes, State, Acc), + whitespace(?STRING_UNBOUND_REST(_C, _) = Bytes, State, Acc) -> + {lists:reverse(Acc), Bytes, State}; + whitespace(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State, Acc) when is_binary(Bytes) -> + case unicode:characters_to_list(Bytes, Enc) of + {incomplete, _, _} -> + cf(Bytes, State, Acc, fun whitespace/3); + {error, _Encoded, _Rest} -> + ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))) + end). + +-define(PARSE_EXTERNAL_ENTITY_BYTE_ORDER_MARK(Bytes, State), + parse_external_entity_byte_order_mark(Bytes, State) -> + parse_external_entity_1(Bytes, State)). diff --git a/lib/xmerl/src/xmerl_sax_parser_list.erlsrc b/lib/xmerl/src/xmerl_sax_parser_list.erlsrc index 624a621d92..ac89896215 100644 --- a/lib/xmerl/src/xmerl_sax_parser_list.erlsrc +++ b/lib/xmerl/src/xmerl_sax_parser_list.erlsrc @@ -2,7 +2,7 @@ %%-------------------------------------------------------------------- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2008-2016. All Rights Reserved. +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -36,6 +36,19 @@ %% In the list case we can't use a '++' when matchin against an unbound variable -define(STRING_UNBOUND_REST(MatchChar, Rest), [MatchChar | Rest]). --define(BYTE_ORDER_MARK_1, undefined_bom1). --define(BYTE_ORDER_MARK_2, undefined_bom2). --define(BYTE_ORDER_MARK_REST(Rest), [undefined|Rest]). + +-define(PARSE_BYTE_ORDER_MARK(Bytes, State), + parse_byte_order_mark(Bytes, State) -> + parse_xml_decl(Bytes, State)). + +-define(PARSE_XML_DECL(Bytes, State), + parse_xml_decl(Bytes, State) -> + parse_prolog(Bytes, State)). + +-define(WHITESPACE(Bytes, State, Acc), + whitespace(?STRING_UNBOUND_REST(_C, _) = Bytes, State, Acc) -> + {lists:reverse(Acc), Bytes, State}). + +-define(PARSE_EXTERNAL_ENTITY_BYTE_ORDER_MARK(Bytes, State), + parse_external_entity_byte_order_mark(Bytes, State) -> + parse_external_entity_1(Bytes, State)). diff --git a/lib/xmerl/src/xmerl_sax_parser_utf16be.erlsrc b/lib/xmerl/src/xmerl_sax_parser_utf16be.erlsrc index ff84ece97a..ec89024729 100644 --- a/lib/xmerl/src/xmerl_sax_parser_utf16be.erlsrc +++ b/lib/xmerl/src/xmerl_sax_parser_utf16be.erlsrc @@ -2,7 +2,7 @@ %%-------------------------------------------------------------------- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2008-2016. All Rights Reserved. +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -34,8 +34,50 @@ -define(APPEND_STRING(Rest, New), <<Rest/binary, New/binary>>). -define(TO_INPUT_FORMAT(Val), unicode:characters_to_binary(Val, unicode, {utf16, big})). -%% STRING_REST and STRING_UNBOUND_REST is only different in the list case -define(STRING_UNBOUND_REST(MatchChar, Rest), <<MatchChar/big-utf16, Rest/binary>>). --define(BYTE_ORDER_MARK_1, undefined_bom1). --define(BYTE_ORDER_MARK_2, <<16#FE>>). +-define(BYTE_ORDER_MARK_1, <<16#FE>>). -define(BYTE_ORDER_MARK_REST(Rest), <<16#FE, 16#FF, Rest/binary>>). + +-define(PARSE_BYTE_ORDER_MARK(Bytes, State), + parse_byte_order_mark(?STRING_EMPTY, State) -> + cf(?STRING_EMPTY, State, fun parse_byte_order_mark/2); + parse_byte_order_mark(?BYTE_ORDER_MARK_1, State) -> + cf(?BYTE_ORDER_MARK_1, State, fun parse_byte_order_mark/2); + parse_byte_order_mark(?BYTE_ORDER_MARK_REST(Rest), State) -> + parse_xml_decl(Rest, State); + parse_byte_order_mark(Bytes, State) -> + parse_xml_decl(Bytes, State)). + +-define(PARSE_XML_DECL(Bytes, State), + parse_xml_decl(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State) when is_binary(Bytes) -> + case unicode:characters_to_list(Bytes, Enc) of + {incomplete, _, _} -> + cf(Bytes, State, fun parse_xml_decl/2); + {error, _Encoded, _Rest} -> + ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))); + _ -> + parse_prolog(Bytes, State) + end; + parse_xml_decl(Bytes, State) -> + parse_prolog(Bytes, State)). + +-define(WHITESPACE(Bytes, State, Acc), + whitespace(?STRING_UNBOUND_REST(_C, _) = Bytes, State, Acc) -> + {lists:reverse(Acc), Bytes, State}; + whitespace(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State, Acc) when is_binary(Bytes) -> + case unicode:characters_to_list(Bytes, Enc) of + {incomplete, _, _} -> + cf(Bytes, State, Acc, fun whitespace/3); + {error, _Encoded, _Rest} -> + ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))) + end). + +-define(PARSE_EXTERNAL_ENTITY_BYTE_ORDER_MARK(Bytes, State), + parse_external_entity_byte_order_mark(?STRING_EMPTY, State) -> + cf(?STRING_EMPTY, State, fun parse_external_entity_byte_order_mark/2); + parse_external_entity_byte_order_mark(?BYTE_ORDER_MARK_1, State) -> + cf(?BYTE_ORDER_MARK_1, State, fun parse_external_entity_byte_order_mark/2); + parse_external_entity_byte_order_mark(?BYTE_ORDER_MARK_REST(Rest), State) -> + parse_external_entity_1(Rest, State); + parse_external_entity_byte_order_mark(Bytes, State) -> + parse_external_entity_1(Bytes, State)). diff --git a/lib/xmerl/src/xmerl_sax_parser_utf16le.erlsrc b/lib/xmerl/src/xmerl_sax_parser_utf16le.erlsrc index a330fce8d0..566333a045 100644 --- a/lib/xmerl/src/xmerl_sax_parser_utf16le.erlsrc +++ b/lib/xmerl/src/xmerl_sax_parser_utf16le.erlsrc @@ -2,7 +2,7 @@ %%-------------------------------------------------------------------- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2008-2016. All Rights Reserved. +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -34,8 +34,50 @@ -define(APPEND_STRING(Rest, New), <<Rest/binary, New/binary>>). -define(TO_INPUT_FORMAT(Val), unicode:characters_to_binary(Val, unicode, {utf16, little})). -%% STRING_REST and STRING_UNBOUND_REST is only different in the list case -define(STRING_UNBOUND_REST(MatchChar, Rest), <<MatchChar/little-utf16, Rest/binary>>). --define(BYTE_ORDER_MARK_1, undefined_bom1). --define(BYTE_ORDER_MARK_2, <<16#FF>>). +-define(BYTE_ORDER_MARK_1, <<16#FF>>). -define(BYTE_ORDER_MARK_REST(Rest), <<16#FF, 16#FE, Rest/binary>>). + +-define(PARSE_BYTE_ORDER_MARK(Bytes, State), + parse_byte_order_mark(?STRING_EMPTY, State) -> + cf(?STRING_EMPTY, State, fun parse_byte_order_mark/2); + parse_byte_order_mark(?BYTE_ORDER_MARK_1, State) -> + cf(?BYTE_ORDER_MARK_1, State, fun parse_byte_order_mark/2); + parse_byte_order_mark(?BYTE_ORDER_MARK_REST(Rest), State) -> + parse_xml_decl(Rest, State); + parse_byte_order_mark(Bytes, State) -> + parse_xml_decl(Bytes, State)). + +-define(PARSE_XML_DECL(Bytes, State), + parse_xml_decl(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State) when is_binary(Bytes) -> + case unicode:characters_to_list(Bytes, Enc) of + {incomplete, _, _} -> + cf(Bytes, State, fun parse_xml_decl/2); + {error, _Encoded, _Rest} -> + ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))); + _ -> + parse_prolog(Bytes, State) + end; + parse_xml_decl(Bytes, State) -> + parse_prolog(Bytes, State)). + +-define(WHITESPACE(Bytes, State, Acc), + whitespace(?STRING_UNBOUND_REST(_C, _) = Bytes, State, Acc) -> + {lists:reverse(Acc), Bytes, State}; + whitespace(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State, Acc) when is_binary(Bytes) -> + case unicode:characters_to_list(Bytes, Enc) of + {incomplete, _, _} -> + cf(Bytes, State, Acc, fun whitespace/3); + {error, _Encoded, _Rest} -> + ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))) + end). + +-define(PARSE_EXTERNAL_ENTITY_BYTE_ORDER_MARK(Bytes, State), + parse_external_entity_byte_order_mark(?STRING_EMPTY, State) -> + cf(?STRING_EMPTY, State, fun parse_external_entity_byte_order_mark/2); + parse_external_entity_byte_order_mark(?BYTE_ORDER_MARK_1, State) -> + cf(?BYTE_ORDER_MARK_1, State, fun parse_external_entity_byte_order_mark/2); + parse_external_entity_byte_order_mark(?BYTE_ORDER_MARK_REST(Rest), State) -> + parse_external_entity_1(Rest, State); + parse_external_entity_byte_order_mark(Bytes, State) -> + parse_external_entity_1(Bytes, State)). diff --git a/lib/xmerl/src/xmerl_sax_parser_utf8.erlsrc b/lib/xmerl/src/xmerl_sax_parser_utf8.erlsrc index d46d60d237..f41d06d013 100644 --- a/lib/xmerl/src/xmerl_sax_parser_utf8.erlsrc +++ b/lib/xmerl/src/xmerl_sax_parser_utf8.erlsrc @@ -2,7 +2,7 @@ %%-------------------------------------------------------------------- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2008-2016. All Rights Reserved. +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -34,11 +34,55 @@ -define(APPEND_STRING(Rest, New), <<Rest/binary, New/binary>>). -define(TO_INPUT_FORMAT(Val), unicode:characters_to_binary(Val, unicode, utf8)). - -%% STRING_REST and STRING_UNBOUND_REST is only different in the list case -define(STRING_UNBOUND_REST(MatchChar, Rest), <<MatchChar/utf8, Rest/binary>>). -define(BYTE_ORDER_MARK_1, <<16#EF>>). -define(BYTE_ORDER_MARK_2, <<16#EF, 16#BB>>). -define(BYTE_ORDER_MARK_REST(Rest), <<16#EF, 16#BB, 16#BF, Rest/binary>>). +-define(PARSE_BYTE_ORDER_MARK(Bytes, State), + parse_byte_order_mark(?STRING_EMPTY, State) -> + cf(?STRING_EMPTY, State, fun parse_byte_order_mark/2); + parse_byte_order_mark(?BYTE_ORDER_MARK_1, State) -> + cf(?BYTE_ORDER_MARK_1, State, fun parse_byte_order_mark/2); + parse_byte_order_mark(?BYTE_ORDER_MARK_2, State) -> + cf(?BYTE_ORDER_MARK_2, State, fun parse_byte_order_mark/2); + parse_byte_order_mark(?BYTE_ORDER_MARK_REST(Rest), State) -> + parse_xml_decl(Rest, State); + parse_byte_order_mark(Bytes, State) -> + parse_xml_decl(Bytes, State)). + +-define(PARSE_XML_DECL(Bytes, State), + parse_xml_decl(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State) when is_binary(Bytes) -> + case unicode:characters_to_list(Bytes, Enc) of + {incomplete, _, _} -> + cf(Bytes, State, fun parse_xml_decl/2); + {error, _Encoded, _Rest} -> + ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))); + _ -> + parse_prolog(Bytes, State) + end; + parse_xml_decl(Bytes, State) -> + parse_prolog(Bytes, State)). + +-define(WHITESPACE(Bytes, State, Acc), + whitespace(?STRING_UNBOUND_REST(_C, _) = Bytes, State, Acc) -> + {lists:reverse(Acc), Bytes, State}; + whitespace(Bytes, #xmerl_sax_parser_state{encoding=Enc} = State, Acc) when is_binary(Bytes) -> + case unicode:characters_to_list(Bytes, Enc) of + {incomplete, _, _} -> + cf(Bytes, State, Acc, fun whitespace/3); + {error, _Encoded, _Rest} -> + ?fatal_error(State, lists:flatten(io_lib:format("Bad character, not in ~p\n", [Enc]))) + end). +-define(PARSE_EXTERNAL_ENTITY_BYTE_ORDER_MARK(Bytes, State), + parse_external_entity_byte_order_mark(?STRING_EMPTY, State) -> + cf(?STRING_EMPTY, State, fun parse_external_entity_byte_order_mark/2); + parse_external_entity_byte_order_mark(?BYTE_ORDER_MARK_1, State) -> + cf(?BYTE_ORDER_MARK_1, State, fun parse_external_entity_byte_order_mark/2); + parse_external_entity_byte_order_mark(?BYTE_ORDER_MARK_2, State) -> + cf(?BYTE_ORDER_MARK_2, State, fun parse_external_entity_byte_order_mark/2); + parse_external_entity_byte_order_mark(?BYTE_ORDER_MARK_REST(Rest), State) -> + parse_external_entity_1(Rest, State); + parse_external_entity_byte_order_mark(Bytes, State) -> + parse_external_entity_1(Bytes, State)). diff --git a/lib/xmerl/src/xmerl_scan.erl b/lib/xmerl/src/xmerl_scan.erl index 5e0459ec21..a1f6ad4e2c 100644 --- a/lib/xmerl/src/xmerl_scan.erl +++ b/lib/xmerl/src/xmerl_scan.erl @@ -1,7 +1,7 @@ %% %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2003-2016. All Rights Reserved. +%% Copyright Ericsson AB 2003-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -474,8 +474,8 @@ event(_X, S) -> %% into multiple objects (in which case {Acc',Pos',S'} should be returned.) %% If {Acc',S'} is returned, Pos will be incremented by 1 by default. %% Below is an example of an acceptable operation -acc(X = #xmlText{value = Text}, Acc, S) -> - {[X#xmlText{value = Text}|Acc], S}; +acc(#xmlText{value = Text}, [X = #xmlText{value = AccText}], S) -> + {[X#xmlText{value = AccText ++ Text}], S}; acc(X, Acc, S) -> {[X|Acc], S}. @@ -2225,16 +2225,18 @@ processed_whole_element(S=#xmerl_scanner{hook_fun = _Hook, AllAttrs = case S#xmerl_scanner.default_attrs of true -> - [ #xmlAttribute{name = AttName, - parents = [{Name, Pos} | Parents], - language = Lang, - nsinfo = NSI, - namespace = Namespace, - value = AttValue, - normalized = true} || - {AttName, AttValue} <- get_default_attrs(S, Name), - AttValue =/= no_value, - not lists:keymember(AttName, #xmlAttribute.name, Attrs) ]; + DefaultAttrs = + [ #xmlAttribute{name = AttName, + parents = [{Name, Pos} | Parents], + language = Lang, + nsinfo = NSI, + namespace = Namespace, + value = AttValue, + normalized = true} || + {AttName, AttValue} <- get_default_attrs(S, Name), + AttValue =/= no_value, + not lists:keymember(AttName, #xmlAttribute.name, Attrs) ], + lists:append(Attrs, DefaultAttrs); false -> Attrs end, @@ -2307,7 +2309,9 @@ expanded_name(Name, [], #xmlNamespace{default = URI}, S) -> expanded_name(Name, N = {"xmlns", Local}, #xmlNamespace{nodes = Ns}, S) -> {_, Value} = lists:keyfind(Local, 1, Ns), case Name of - 'xmlns:xml' when Value =/= 'http://www.w3.org/XML/1998/namespace' -> + 'xmlns:xml' when Value =:= 'http://www.w3.org/XML/1998/namespace' -> + N; + 'xmlns:xml' when Value =/= 'http://www.w3.org/XML/1998/namespace' -> ?fatal({xml_prefix_cannot_be_redeclared, Value}, S); 'xmlns:xmlns' -> ?fatal({xmlns_prefix_cannot_be_declared, Value}, S); @@ -2321,6 +2325,8 @@ expanded_name(Name, N = {"xmlns", Local}, #xmlNamespace{nodes = Ns}, S) -> N end end; +expanded_name(_Name, {"xml", Local}, _NS, _S) -> + {'http://www.w3.org/XML/1998/namespace', list_to_atom(Local)}; expanded_name(_Name, {Prefix, Local}, #xmlNamespace{nodes = Ns}, S) -> case lists:keysearch(Prefix, 1, Ns) of {value, {_, URI}} -> @@ -2331,9 +2337,6 @@ expanded_name(_Name, {Prefix, Local}, #xmlNamespace{nodes = Ns}, S) -> ?fatal({namespace_prefix_not_declared, Prefix}, S) end. - - - keyreplaceadd(K, Pos, [H|T], Obj) when K == element(Pos, H) -> [Obj|T]; keyreplaceadd(K, Pos, [H|T], Obj) -> |