(Updating to match new version of source page) |
(Updating to match new version of source page) |
||
(One intermediate revision by the same user not shown) | |||
Line 1: | Line 1: | ||
− | |||
− | |||
− | |||
<languages /> | <languages /> | ||
Line 214: | Line 211: | ||
lxLENGTH // length of the current lexeme (as presented in the input) | lxLENGTH // length of the current lexeme (as presented in the input) | ||
lxBEGIN_POS // position of the first character of the current lexeme | lxBEGIN_POS // position of the first character of the current lexeme | ||
− | + | lxBEGIN_IDX // corresponding index | |
lxNAMED_TOKEN(⟨name⟩, ⟨type⟩) // create a variable representing named ⟨name⟩ with the token type ⟨type⟩ | lxNAMED_TOKEN(⟨name⟩, ⟨type⟩) // create a variable representing named ⟨name⟩ with the token type ⟨type⟩ | ||
lxTOKEN(⟨type⟩) // create such a variable named “token” | lxTOKEN(⟨type⟩) // create such a variable named “token” |
KDevelop-PG-Qt является генератором кода парсера для KDevplatform и используется некоторыми расширениями поддержки языков программирования в KDevelop (например: Ruby, PHP, Java...).
KDevelop-PG-Qt основывается на классах Qt, а так же оригинальном парсере KDevelop-PG, который использует типы данных из библиотеки STL, однако у него более широкие возможности. Большинство возможностей схожи, хотя возможно что ...-Qt парсер-генератор более современен и функционально богаче, чем обычный генератор в стиле STL. Для создания парсеров, используемых для расширений поддержки языков программирования в KDevelop следует использовать ...-Qt версию парсер-генератора.
Этот документ не подразумевается как полноценное и глубокое описание для всех частей KDevelop-PG. Вместо этого он предназначен для краткого введения и, что более важно, ссылается на разработчиков.
Для получения более детальной информации, прочтите отличную диссертацию бакалавра Якоба Петровитса (Jakob Petsovits). Вы найдёте её поссылке с секции Web-ссылок в нижней части этой страницы (примечание переводчика: на самом деле, к сожалению, этот документ был удалён и ссылка ведёт вникуда. Ведётся поиск этого документа).
KDevelop-PG-Qt можно найти в репозитории git. В исходном коде имеется 4 пакета с примерами.
Чтобы скачать попробуйте:
git clone git://anongit.kde.org/kdevelop-pg-qt.git
или:
git clone kde:kdevelop-pg-qt(если имеете установленный git с настроенным URL префиксом "kde")
При запуске программа требует файл .g, называемый грамматическим:
./kdev-pg-qt --output=prefix syntax.g
Значение параметра --output определяет префикс выходных файлов, а так же пространство имён для генерируемого кода. Программа Kate производит простую подсветку синтаксиса для грамматических файлов KDevelop-PG-Qt.
While evaluating the grammar and generating its parser files, the application will output information about so called conflicts to STDOUT. As said above, the following files will actually be prefixed.
AST stands for Abstract Syntax Tree. It defines the data structure in which the parse tree is saved. Each node is a struct with the postfix Ast, which contains members that point to any possible sub elements.
One important part of parser.h is the definition of the parser tokens, the TokenType enum. The TokenStream of your lexer should to use this. You have to write your own lexer or let one generate by Flex. See also the part about Tokenizers/Lexers below.
Having the token stream available, you create your root item and call the parser on the parse method for the top-level AST item, e.g. DocumentAst* => parseDocument(&root). On success, root will contain the AST.
The parser will have one parse method for each possible node of the AST. This is nice for e.g. an expression parser or parsers that should only parse a sub-element of a full document.
The Visitor class provides an abstract interface to walk the AST. Most of the time you don't need to use this directly, the DefaultVisitor takes some work off your shoulders.
The DefaultVisitor is an implementation of the abstract Visitor interface and automatically visits each node in the AST. Hence, this is probably the best candidate for a base class for your personal visitors. Most language plugins use these in their Builder classes to create the DUChain.
As mentioned, KDevelop-PG-Qt requires a Tokenizer. You can either let KDevelop-PG-Qt generate one for you, write one per hand, as it has been done for C++ and PHP, or you can use external tools like Flex.
The tokenizer's job, in principle, boils down to:
The rest, e.g. actually building the tree and evaluating the semantics, is part of the parser and the AST visitors.
KDevelop-PG-Qt can generate lexers being well integrated into its architecture (you do not have to create a token-stream-class invoking lex or something like that). See examples/foolisp in the code for a simplistic example, there is also an incomplete PHP-Lexer for demonstration purposes.
Regular expressions are used to write rules using the KDevelop-PG-Qt, we use the following syntax (α and β are arbitrary regular expressions, a and b characters):
All regular expressions are case sensitive. Sorry, there is currently no way for insensitivity.
There are several escape sequences which can be used to encode special characters:
Some regexes are predefined and can be used using braces {⟨name⟩}. They get imported from the official Unicode data, some important regexes:
Rules can be written as:
⟨regular expression⟩ TOKEN;
Then the Lexer will generate the token TOKEN for lexemes matching the given regular expression. Which token will be chosen if there are multiple options? We use the first longest match rule: It will take the longest possible match (eating as many characters as possible), if there are multiple of those matches, it will take the first one.
Rules can perform code actions and you can also omit tokens (then no token will be generated):
⟨regular expression⟩ [: ⟨code⟩ :] TOKEN; ⟨regular expression⟩ [: ⟨code⟩ :]; ⟨regular expression⟩ ;
There is rudimentary support for lookahead and so called (our invention) barriers:
⟨regular expression⟩ %la(⟨regular expression⟩); ⟨regular expression⟩ %ba(⟨regular expression⟩);
The first rule will only accept words if they match the first regular expression and are followed by anything matching the expression specified using %la. The second rule will accept words matched by the first regular expression but will never run into a character sequence matching the regex specified by %ba. However, currently only rules with fixed length are allowed in %la and %ba (for example foo|bar, but not qux|garply). When applying the “first longest match” rule the %la/%ba expressions count, too.
You can create your own named regexes using an arrow:
⟨regular expression⟩ -> ⟨identifier⟩;
The first character of the identifier should not be upper case.
Additionally there are two special actions:
⟨regular expression⟩ %fail; ⟨regular expression⟩ %continue;
%fail will stop tokenization. %continue will make the matched characters part of the next token.
A grammar file can contain multiple rulesets. A ruleset is a set of rules, as described in the previous section. It gets declared using:
%lexer "name" -> ⟨rules⟩ ;
For your main-ruleset you omit the name (the name will be “start”).
Usually the start-ruleset will be used. But you can change the ruleset in code actions using the macro lxSET_RULE_SET(⟨name⟩). You can specify code to be executed when entering or leaving a ruleset by using %enter [: ⟨code⟩ :]; or %leave [: ⟨code⟩ :]; respectively inside the definition of the ruleset.
The standard statements %lexer_declaration_header and %lexer_bits_header are available to include files in the generated lexer.h/lexer.cpp.
By using %lexer_base you can specify the baseclass for the lexer-class, by default it is the TokenStream class defined by KDevelop-PG-Qt.
After %lexerclass(bits) you can specify code to be inserted in lexer.cpp.
You have to specify an encoding the lexer should work with internally using %input_encoding "⟨encoding⟩", possible values:
With %input_stream you can specify which class the lexer should use to get the characters to process, there are some predefined classes:
Whether you choose UTF-8, UTF-16 or UTF-32 is irrelevant for functionality, but it may significantly affect compile-time and run-time performance (you may want to test your Lexer with ASCII if compilation takes too long). For example you want to work with a QByteArray containing UTF-8 data, and you want to get full Unicode support: You could either use the QByteArrayIterator and UTF-8 as internal encoding, or the QUtf8ToUtf16Iterator and UTF-16, or the QUtf8ToUcs4Iterator and UTF-32.
You can also choose between %table_lexer; and %sequence_lexer; In the first case transitions between states of the lexer will get represented by big tables while generating the lexer (cases in the generated code). In the second case sequences will get stored in a compressed data structure and transitions will get represented by nested if-statements. For UTF-32 %table_lexer is infeasible, thus there %sequence_lexer is the onliest option.
Inside your actions of the lexer you can use some predefined macros:
lxCURR_POS // position in the input (some kind of iterator or pointer) lxCURR_IDX // index of the position in the input // (it is the index as presented in the input, for example: input is a QByteArray, index incrementation per byte, but the lexer may operate on 32-bit codepoints) lxCONTINUE // like %continue, add the current lexeme to the next token lxLENGTH // length of the current lexeme (as presented in the input) lxBEGIN_POS // position of the first character of the current lexeme lxBEGIN_IDX // corresponding index lxNAMED_TOKEN(⟨name⟩, ⟨type⟩) // create a variable representing named ⟨name⟩ with the token type ⟨type⟩ lxTOKEN(⟨type⟩) // create such a variable named “token” lxDONE // return the token generated before lxRETURN(X) // create a token of type X and return it lxEOF // create the EOF-token lxFINISH // create the EOF-token and return it (will stop tokenization) lxFAIL // raise the tokenization error lxSKIP // continue with the next lexeme (do not return a token, you should not have created one before) lxNEXT_CHR(⟨chr⟩) // set the variable ⟨chr⟩ to the next char in the input yytoken // current token
With the existing examples, it shouldn't be too hard to write such a lexer. Between most languages, especially those "inheriting" C, there are many common syntactic elements. Especially comments and literals can be handled just the same way over and over again. Adding a simple token is trivial:
"special-command" return Parser::Token_SPECIAL_COMMAND;
That's pretty much it, take a look at eg. java.ll for an excellent example. However, it is quite tricky and ugly to handle UTF-8 with Flex.
KDevelop-PG-Qt uses so called context-free grammars using a concept of non-terminals (nodes) and terminals(tokens). While writing the grammar for the basic structure of your language, you should try to mimic the semantics of the language. Lets take a look at an example:
C++-document consists of lots of declarations and definitions, a class definition could be handled e.g. in the following way:
The member-declarations-list is of course not a part of any C++ description, it is just a helper to explain the structure of a given semantic part of your language. The grammar could then define how exactly such helper might look like.
Now let us have a look at a basic example, a declaration in C++, as described in grammar syntax:
struct_declaration
This is called a rule definition. Every lower-case string in the grammar file references such a rule. Our case above defines what a declaration looks like. The |-char stands for a logical or, all rules have to end on two semicolons.
In the example we reference other rules which also have to be defined. Here's for example the class_declaration, note the tokens in all-upper-case:
CLASS IDENTIFIER LBRACE class_declaration* RBRACE SEMICOLON -> class_declaration ;
There is a new char in there: The asterisk has the same meaning as in regular expressions, i.e. that the previous rule can occur arbitrarily often or not at all.
In a grammar 0
stands for an empty token. Using it in addition with parenthesizing and the logical or from above, you can express optional elements:
some_required_rule SOME_TOKEN ( some_optional_stuff | some_other_stuff | 0 ) -> my_rule ;
All symbols never occuring on the left side of a rule are start-symbols. You can use one of them to start parsing.
The simple rule above could be used to parse the token stream, yet no elements would be saved in the parsetree. This can be easily done though:
class_declaration=class_declaration
The DeclarationAst struct now contains pointers to each of these elements. During the parse process the pointer for each found element gets set, all others become NULL. To store lists of elements, prepend the identifier with a hash # :
CLASS IDENTIFIER SEMICOLON
TODO: internal structure of the list, important for Visitors
Identifier and targets can be used in more than one place:
#one=one (#one=one)* -> one_or_more ;
In the example above, all matches to the rule one will be stored in one and the same list one.
Somewhere in the grammar (you should probably put it near the head) you'll have to define a list of available Tokens. From this list, the TokenType enum in parser.h will be created. Additionally to the enum value names you should define an explanation name which will e.g. be used in error messages. Note that the representation of a Token inside the source code is not required for the grammar/parser as it operates on a TokenStream, see Lexer/Tokenizer section above.
%token T1 ("T1-Name"), T2 ("T2-Name"), COMMA (";"), SEMICOLON (";") ;
It is possible to use %token multiple times to group tokens in the grammar. Though all tokens will still be put into the same TokenType enum.
TODO: explain process of using the parser Tokens
Alternatively to the asterisk (*) you can use a plus sign (+) to mark lists of one-or-more elements:
(#one=one)+ -> one_or_more ;
Using the #rule @ TOKEN syntax you can mark a list of rule, separated by TOKEN:
#item=item @ COMMA -> comma_separated_list ;
Alternatively to the above mentioned (item=item | 0) syntax you can use the following to mark optional items:
?item=item -> optional_item ;
Attention |
---|
(0|x) is equivalent to 0 |
Using a colon : instead of the equal sign = you can store the sub-AST in a local variable that will only be available during parsing, and only in the current rule.
TODO: need example
Instead of a name you can also place a dot . before the equal sign = . Then the AST will inherit all class-members from the sub-AST and the parsing-method for the sub-AST will be merged into the parent-AST. An example:
op=PLUS
In SimpleArithmeticsAst there will be the fields val1, val2 (storing the operands) and op (storing the operator-token). parseSimpleArithmetics will not have to call parseOperator. Obviously recursive inlining is not allowed.
Sometimes it is required to integrate hand-written code into the generated parser. Instead of editing the end-result (**never** do that!) you should put this code into the grammar at the correct places. Here are a few examples when you'd need this:
[: // here be dragons^W code ;-) :]
The code will be put into the generated parser.cpp file. If you use it inside a grammar rule, it will be put into the correct position during the parse process. You can access the current node via the variable yynode, it will have the type 'XYZAst**'.
In KDevelop language plugins, you'll see that most grammars start with something like:
[: #include <QtCore/QString> #include <kdebug.h> #include <tokenstream.h> #include <language/interfaces/iproblem.h> #include "phplexer.h" namespace KDevelop { class DUContext; } :]
This is a code section, that will be put at the beginning of ast.h, i.e. into the global context.
But there are also some newer statements available to include header-files:
%ast_header "header.h" %parser_declaration_header "header.h" %parser_bits_header "header.h"
The include-statement will be inserted into ast.h, parser.h respectively parser.cpp.
Also it's very common to define a set of enumerations e.g. for operators, modifiers, etc. pp. Here's an stripped example from PHP, note that the code will again be put into the generated parser.h file:
%namespace [: enum ModifierFlags { ModifierPrivate = 1, ModifierPublic = 1 << 1, ModifierProtected = 1 << 2, ModifierStatic = 1 << 3, ModifierFinal = 1 << 4, ModifierAbstract = 1 << 5 }; ... enum OperationType { OperationPlus = 1, OperationMinus, OperationConcat, OperationMul, OperationDiv, OperationMod, OperationAnd, OperationOr, OperationXor, OperationSl, OperationSr }; :]
When you write code at the end of the grammar-file simply between [: and :], it will be added to parser.cpp.
To add additional members to _every_ AST variable, use the following syntax:
%ast_extra_members [: KDevelop::DUContext* ducontext; :]
You can also specify the base-class for the nodes:
%ast_base symbol_name "BaseClass"
BaseClass has to inherit from AstNode.
Instead of polluting the global context with state tracker variables, and hence destroying the whole advantages of OOP, you can add additional members to the parser class. It's also very convenient to define functions for error reporting etc. pp. Again a stripped example from PHP:
%parserclass (public declaration) [: enum ProblemType { Error, Warning, Info }; void reportProblem( Parser::ProblemType type, const QString& message ); QList<KDevelop::ProblemPointer> problems() { return m_problems; } ... enum InitialLexerState { HtmlState = 0, DefaultState = 1 }; :]
Note, that we used %parserclass (public declaration), we could instead have used private or protected declaration.
protected
There is also a statement to specify a base-class:
%parser_base "ClassName"
When you add member variables to the class, you'll have to initialize and or destroy them as well. Here's how (either use ctor or dtor, of course):
desctructor] ) [: // Code :]
?[: // some bool expression :]
The following rule will only apply if the boolean expression evaluates to true. Here's an advanced example, which also shows that you can use the pipe symbol ('|') as logical or, i.e. essentially this is a if... else...' conditional:
?[: someCondition :] SOMETOKEN ifrule=myVar
This is especially convenient together with lookaheads (see below).
You can set up the grammar to define local variables whenever a rule gets applied:
... -> class_member [: enum { A_PUBLIC, A_PROTECTED, A_PRIVATE } access; :];
This variable is local to the rule class_member.
Similar to the syntax above, you can define members whenever a rule gets applied:
temporary] variable yourName: yourType ];
For example:
... -> class_member [ member variable access : AccessType; ];
Of course AccessType has to be defined somewhere else, see e.g. the Additional parser class members section above.
Using temporary or member is equivalent.
The first time you write a grammar, you'll potentially fail: Since KDevelop-PG-Qt is a LL(1) parser generator, conflicts can occur and have to be solved by hand. Here's an example for a FIRST/FIRST conflict:
class_definition -> class_expression ;
KDevelop-PG-Qt will output:
** WARNING found FIRST/FIRST conflict in "class_expression"
Sometime's it's these warnings can be ignored, but most of the time it will lead to problems in parsing. In our example the class_expression rule won't be evaluated properly: When you try to parse a class_definition, the parser will see that the first part (CLASS) of class_declaration matches, and jumps think's that this rule is to be applied. Since the next part of the rule does not match, an error will be reported. It does _not_ try to evaluate class_definition automatically, but there are different ways to solve this problem:
In theory such behavior might be unexpected: In BNF e.g. the above syntax would be enough and the parser would automatically jump back and retry with the next rule, here class_definition. But in practice such behaviour is - most of the time - not necessary and would slow down the parser. If however such behavior is explicitly what you want, you can use an explicit backtracking syntax in your grammar:
try/rollback(class_declaration) catch(class_definition) -> class_expression ;
In theory, every FIRST/FIRST could be solved this way, but keep in mind that the more you use this feature, the slower your parser gets. If you use it, sort the rules in order of likeliness. Also, there could be cases where the sort order could make the difference between correct and wrong parsing, especially with rules that "extend" others.
KDevelop-PG-Qt has an alternative to the - potentially slow - backtracking mechanism: Look ahead. You can use the LA(qint64) function in embedded C++ code (see sections above). LA(1) will return the current token, you most probably never need to use that. Accordingly, LA(2) returns the next and LA(0) the previous token. (If you wonder where these somewhat uncommon indexes come from: Read the thesis, or make yourself acquainted with LA(k) parser theory.)
class_declaration -> class_expression
Note: The conflict will still be output, but it is manually resolved. You should always document this somewhere with a comment, for future contributors to know that this is a false-positive.
Often you can solve the conflict all-together with an elegant solution. Like in our example:
LBRACE class_content RBRACE -> class_definition ; CLASS IDENTIFIER ?class_definition SEMICOLON -> class_expression ;
No conflicts, fast: everybody is happy!
A FIRST/FOLLOW conflict says, that it is undefined where a symbol ends and the parent symbol continues. A pretty stupid example is:
item* -> item_list ; item_list item* -> items ;
You probably see the glaring error. try/rollback helps with serious problems (e.g. parser doesn't work), though often you can ignore these conflicts. But if you do so, be aware that the parser is greedy: In our example item_list will always contain all item elements, and items will never have an item member. If this is an issue that leads to later conflicts, only try/rollback, custom manual code to check which rule should be used, or a refactoring of your grammar structure.
Alternatively it is sometimes helpful/required to change the greedy behaviour. In lists you can use manual code to force a break at a given position. Here's an example that shows this on a declaration of an array with fixed size:
typed_identifier=typed_identifier LBRACKET UNSIGNED_INTEGER [: count = static_cast<MyTokenStream*>(token)->strForCurrentToken().toUInt(); :] RBRACKET EQUAL LBRACE (#expression=expression [: if(--count == 0) break; :] @ COMMA) ?[: count == 0 :] RBRACE SEMICOLON -> initialized_fixed_array [ temporary variable count: uint; ];
TODO: return true/false or break??
Via return false you can enforce a premature stop (== error) of the evaluation of the current rule. Using return true you stop the evaluation prematurely, but signalize that the parse process was successful.
try/recover(expression)
-> symbol ;;
This is approximately the same as:
[: ParserState *state = copyCurrentState(); :]
try/rollback(expression)
catch( [: restoreState(state); :] )
-> symbol ;
Hence you have to implement the member-functions copyCurrentState and restoreState and you have to define a type called ParseState. You do not have to write the declaration of those functions in the header-file, it is generated automatically if you use try/recover. This concept seems to be useful if there are additional states used while parsing. The Java-parser takes usage from it very often. But I do not know a lot about this feature and it seems unimportant for me. (I guess, it is not) I would be happy when somebody could explain it to me.
TODO