% BEGIN LICENSE BLOCK % Version: CMPL 1.1 % % The contents of this file are subject to the Cisco-style Mozilla Public % License Version 1.1 (the "License"); you may not use this file except % in compliance with the License. You may obtain a copy of the License % at www.eclipse-clp.org/license. % % Software distributed under the License is distributed on an "AS IS" % basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See % the License for the specific language governing rights and limitations % under the License. % % The Original Code is The ECLiPSe Constraint Logic Programming System. % The Initial Developer of the Original Code is Cisco Systems, Inc. % Portions created by the Initial Developer are % Copyright (C) 1996 - 2006 Cisco Systems, Inc. All Rights Reserved. % % Contributor(s): % % END LICENSE BLOCK % % @(#)umsmacros.tex 1.9 96/01/08 % % umsmacros.tex % \chapter{{\eclipse} Macros} %HEVEA\cutdef[1]{section} \label{chapmacros} %---------------------------------------------------------------------- \section{Introduction} %---------------------------------------------------------------------- \index{source transformation} \index{macro expansion} {\eclipse} provides a general mechanism to perform macro expansion of Prolog terms. Macro expansion can be performed in 3 situations: \begin{quote} \begin{description} \item[read macros]\index{read macros}\index{macros!read} are expanded just after a Prolog term has been read by the {\eclipse} parser. Note that the parser is not only used during compilation but by all \biptxtref{term-reading}{read/1}{../bips/kernel/ioterm/read-1.html} predicates. \item[compiler macros]\index{compiler macros}\index{macros!compiler} are expanded only during compilation and only when a term occurs in a certain context (clause or goal). \item[write macros]\index{write macros}\index{macros!write} are expanded just before a Prolog term is printed by one of the output predicates \end{description} \end{quote} In addition to transforming a term, macros can also be \emph{source annotation aware}, and provide source annotation information for the transformed term if supplied with source annotation information for the orginal term. Source annotation information is about the source and position of a term, and is provided by the predicate \bipref{read_annotated/3}{../bips/kernel/ioterm/read_annotated-3.html}. Macros are attached to classes of terms specified by their functors or by their type. Macros obey the module system's visibility rules. They may be either \biptxtref{local}{local/1}{../bips/kernel/modules/local-1.html} or \biptxtref{exported}{export/1}{../bips/kernel/modules/export-1.html}. The macro expansion is performed by a user-defined Prolog predicate. %---------------------------------------------------------------------- \section{Using the macros} \label{usingmacros} %---------------------------------------------------------------------- The following declarations and built-ins control macro expansion: \begin{quote} \begin{description} \item[local macro(+\pattern{TermClass}, +\pattern{TransPred}, +\pattern{Options})] defines a macro for the given \about{TermClass}. The transformation itself will be performed by the predicate \about{TransPred}. \item[export macro(+\pattern{TermClass}, +\pattern{TransPred}, +\pattern{Options})] as above, except that the macro is available to other modules. \item[erase_macro(+\pattern{TermClass}, +\pattern{Options})] erases the macro that is currently defined for \about{TermClass}. Note that this can only be done in the module where the definition was made. \item[current_macro(?\pattern{TermClass}, ?\pattern{TransPred}, ?\pattern{Options}, ?\pattern{Module})] can be used to get information about currently defined visible macros. \end{description} \end{quote} Macros are selectively applied only to terms of the specified class. \about{TermClass} can take two forms: \begin{quote} \begin{description} \item[\pattern{Name}/\pattern{Arity}] transform all terms with the specified functor \item[type(\pattern{Type})]\index{type macros}\index{macros!type} transform all terms of the specified type, where \about{Type} is one of the following: \notation{compound}, \notation{string}, \notation{integer}, \notation{rational}, \notation{float}, \notation{breal}, \notation{atom}, \notation{goal}.\footnote{% \notation{type(goal)} stands for suspensions.} \end{description} \end{quote} The \notation{+\pattern{TransPred}} argument specifies the predicate that will perform the transformation. It has to be either of arity 2 or 3 and should have the form: \begin{quote} \notation{\pattern{trans_function}(\pattern{OldTerm},~\pattern{NewTerm}% ~[,~\pattern{Module}])~:-~...~.} \end{quote} or it can be source annotation aware, and be of arity 4 or 5, as follows: \begin{quote} \notation{\pattern{trans_function}(\pattern{OldTerm},~\pattern{NewTerm},% ~\pattern{OldAnn},~\pattern{NewAnn}~[,~\pattern{Module}])~:-~...~.} \end{quote} At transformation time, the system will call \about{TransPred} in the module where \bipref{macro/3}{../bips/kernel/syntax/macro-3.html} was invoked. The term to transform is passed as the first argument, the second is a free variable which the transformation predicate should bind to the transformed term. In the case of the source annotation aware version of \about{TransPred}, if the term was read in by \predspec{read_annotated/2,3}, the annotated version of the term to transformed is passed in the third argument, and the transformation should bind the fourth argument to the annotated transformed term; otherwise, if no source annotation information is available, the third argument is passed in as a free variable, and the transformation should not bind the fourth argument. In both \pattern{TransPred} cases, the optional last argument is the module where the term was being read in. See section~\ref{annotated} for more details on annotated terms. \about{Options} is a list which may be empty (in this case the macro defaults to a local read term macro) or contain specifications from the following categories: \begin{itemize} \item mode: \begin{quote} \begin{description} \item[read:]\index{macros!read} This is a read macro and shall be applied after reading a term (default). \item[write:]\index{macros!write} This is a write macro and shall be applied before printing a term. \end{description} \end{quote} \item type: \begin{quote} \begin{description} \item[term:]\index{macros!term} Transform all terms (default). \item[clause:]\index{macros!clause} Transform only if the term is a program clause, i.e., inside \bipref{compile/1}{../bips/kernel/compiler/compile-1.html}, etc.\footnote{% Note that clause transformation is {\em not} performed with \bipref{assert/1}{../bips/kernel/dynamic/assert-1.html}, \bipref{retract/1}{../bips/kernel/dynamic/retract-1.html} and \bipref{clause/1}{../bips/kernel/dynamic/clause-1.html}. This is a change from previous versions of \eclipse.} Write macros are applied using the \notation{C} option in the \bipref{printf/2}{../bips/kernel/ioterm/printf-2.html} predicate. \item[goal:]\index{macros!goal} Goal-read-macros are transformed only if the term is a subgoal in the body of a program clause. Goal-write macros are applied using the \notation{G} option in the \bipref{printf/2}{../bips/kernel/ioterm/printf-2.html} predicate. \end{description} \end{quote} \item additional specification: \begin{quote} \begin{description} \item[protect_arg:]\index{macros!protect_arg} Disable transformation of subterms (optional). \item[top_only:]\index{macros!top_only} Consider only the whole term, not subterms (optional). \end{description} \end{quote} \end{itemize} The following shorthands exist: \begin{quote} \begin{description} \item[local/export portray(+\pattern{TermClass}, +\pattern{TransPred}, +\pattern{Options}):] \bipref{portray/3}{../bips/kernel/syntax/portray-3.html} is like \bipref{macro/3}{../bips/kernel/syntax/macro-3.html}, but the write-option is implied. \item[inline(+\pattern{PredSpec}, +\pattern{TransPred}):] \bipref{inline/2}{../bips/kernel/compiler/inline-2.html} is the same as a goal-read-macro. The visibility is inherited from the transformed predicate. \end{description} \end{quote} Here is an example of a conditional read macro: \begin{quote} \begin{verbatim} [eclipse 1]: [user]. trans_a(a(X,Y), b(Y)) :- % transform a/2 into b/1, number(X), % but only under these X > 0. % conditions :- local macro(a/2, trans_a/2, []). user compiled traceable 204 bytes in 0.00 seconds yes. [eclipse 2]: read(X). a(1, hello). X = b(hello) % transformed yes. [eclipse 3]: read(X). a(-1, bye). X = a(-1, bye) % not transformed yes. \end{verbatim} \end{quote} If the transformation function fails, the term is not transformed. Thus, \notation{a(1, zzz)} is transformed into \notation{b(zzz)} but \notation{a(-1, zzz)} is not transformed. The arguments are transformed bottom-up. It is possible to protect the subterms of a transformed term by specifying the flag \notation{protect_arg}. A term can be protected against transformation by quoting it with the ``protecting functor'' (by default it is \predspecidx{no_macro_expansion/1}):\index{macro!no_macro_expansion} \begin{quote} \begin{verbatim} [eclipse 4]: read(X). a(1, no_macro_expansion(a(1, zzz))). X = b(a(1, zzz)). \end{verbatim} \end{quote} Note that the protecting functor is itself defined as a macro: \begin{quote} \begin{verbatim} trprotect(no_macro_expansion(X), X). :- export macro(no_macro_expansion/1, trprotect/2, [protect_arg]). \end{verbatim} \end{quote} A local macro is only visible in the module where it has been defined. When it is defined as exported, then it is copied to all other modules that contain a \bipref{use_module/1}{../bips/kernel/modules/use_module-1.html} or \bipref{import/1}{../bips/kernel/modules/import-1.html} for this module. The transformation function should also be exported in this case. There are a few global macros predefined by the system, e.g., for % the following stops latex2html from turning --> into -> {\tt -}{\tt ->/2} (grammar rules, see below) or \predspec{with/2} and \predspec{of/2} (structure syntax, see section \ref{chapstruct}). These predefined macros can be hidden by local macro definitions. The global flag \notationidx{macro_expansion} can be used to disable macro expansion globally, e.g., for debugging purposes. Use \notation{set_flag(macro_expansion, off)} to do so. The next example shows the use of a type macro. Suppose we want to represent integers as \predspec{s/1} terms: \begin{quote} \begin{verbatim} [eclipse 1]: [user]. tr_int(0, 0). tr_int(N, s(S)) :- N > 0, N1 is N-1, tr_int(N1, S). :- local macro(type(integer), tr_int/2, []). yes. [eclipse 2]: read(X). 3. X = s(s(s(0))) yes. \end{verbatim} \end{quote} When we want to convert the \predspec{s/1} terms back to normal integers so that they are printed in the familiar form, we can use a write macro. Note that we first erase the read macro for integers, otherwise we would get unexpected effects since all integers occurring in the definition of \predspec{tr_s/2} would turn into \predspec{s/1} structures: \begin{quote} \begin{verbatim} [eclipse 3]: erase_macro(type(integer)). yes. [eclipse 4]: [user]. tr_s(0, 0). tr_s(s(S), N) :- tr_s(S, N1), N is N1+1. :- local macro(s/1, tr_s/2, [write]). yes. [eclipse 2]: write(s(s(s(0)))). 3 yes. \end{verbatim} \end{quote} \subsection{Source annotation aware macro transformations} \label{annotated} When the macro transformation predicate has 4 or 5 arguments, it is termed source annotation aware, and the extra arguments are used to specify how source information from the original term should be mapped to the transformed term. An annotated term provides the source information about a term. It is structurally similar to the original term and contains all information about the term, plus additional type information, variable names, and source position annotations for all subterms. The structure of the descriptive terms is as follows: \vfill %<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< \begin{quote} \begin{verbatim} :- export struct(annotated_term( term, % var, atomic or compound type, % term type (see below) file, % source file name (atom) line, % source line (integer) from, to % source position (integers) ... )). \end{verbatim} \end{quote} The type-field describes the type of the original term and provide type information similar to those used in \bipref{type_of/2}{../bips/kernel/typetest/type_of-2.html}, except that they convey additional information about variables and \notation{end_of_file}. In the case of atomic terms and variables, the term-field simply contains the plain original term. For compound terms, the term-field contains a structure whose functor is the functor of the plain term, but whose arguments are annotated versions of the plain term arguments. For example, the annotated term representing the source term \notation{foo(bar, X, _, 3)} is: \begin{quote} \begin{verbatim} annotated_term(foo( annotated_term(bar, atom, ...), annotated_term(X, var('X'), ...), annotated_term(_, anonymous, ...), annotated_term(3, integer, ...) ), compound, ... ) \end{verbatim} \end{quote} The file/line/from/to-fields of an annotated term describe the "source position" of the term, as follows: \begin{description} \item[file] The canonical file name of the source file (an atom), or the empty atom \notation{'}\notation{'} if the source is not a file or is not known. \item[line] The line number in the source stream (positive integer). \item[from, to] The exact term position as integer offsets in the source stream, starting at \about{from} and ending at $\mbox{\about{to}}-1$. \end{description} The extra arguments for the transformation predicate are a pair of annotated terms for the original and transformed term. The predicate will be supplied with the annotated term for the original term if available, and the predicate is responsible for specifying the annotated term for the transformed term---the structure of the transformed annotated term must match the annotated term structure expected for the transformed term. If no annotated information is available, the original annotated term will be a variable, and the predicate must not bind the transformed annotated term. For an example, here is a source annotation aware version of the previous \predspec{trans_a/2} example: \begin{verbatim} [eclipse 1]: [user]. ... trans_a(a(X,Y), b(Y), AnnA, AnnTrans) :- number(X), X > 0, ( var(AnnA) -> true % no source information, leave AnnTrans as var ; AnnA = annotated_term{term:a(_AnnX, AnnY), file:File, line:Line, from:From,to:To}, AnnTrans = annotated_term{term:b(AnnY), type: compound, file:File, line:Line, from:From,to:To} ). :- local macro(a/2, trans_a/4, []). Yes (0.23s cpu) [eclipse 2]: read_annotated(user, X, Y). a(3,bar(X)). X = b(bar(X)) Y = annotated_term(b(annotated_term(bar(annotated_term(X, var('X'), user, 18, 654, 655)), compound, user, 18, 650, 654)), compound, user, 18, 646, 648) \end{verbatim} In the example, the main functor of the transformed predicate, \predspec{b/1}, inherits the annotation information for the original term's principal functor, \predspec{a/2}. The argument Y in the transformed term takes the annotation information from the corresponding argument in the original term. The source annotation aware transformation predicate facility is provided to allow the user to access the details of how the subterms of the original term are mapped to the transformed term. Without this extra information, the whole of the transformed term is given the source information (source position, source file etc.) of the original source term. This extra information is useful when the subterms are goals, because without the extra information, source tracing of these goals during debugging will not be done. %---------------------------------------------------------------------- \section{Definite Clause Grammars --- DCGs} \label{dcg} %---------------------------------------------------------------------- \index{DCG} \index{definite clause grammar} \index{grammar rules} \index{\notation{-}\notation{->/2}} Grammar rules are described in many standard Prolog texts (\cite{clocksin81}). In {\eclipse} they are provided by a predefined global\footnote{% So that the user can redefine it with a local one.} macro for % the following stops latex2html from turning --> into -> {\tt -}{\tt ->/2}. When the parser reads a clause whose main functor is % the following stops latex2html from turning --> into -> {\tt -}{\tt ->/2}, it transforms it according to the standard rules. The syntax for DCGs is as follows: \begin{quote} \begin{verbatim} grammar_rule --> grammar_head, ['-->'], grammar_body. grammar_head --> non_terminal. grammar_head --> non_terminal, [','], terminal. grammar_body --> grammar_body, [','], grammar_body. grammar_body --> grammar_body, [';'], grammar_body. grammar_body --> grammar_body, ['->'], grammar_body. grammar_body --> grammar_body, ['|'], grammar_body. grammar_body --> iteration_spec, ['do'], grammar_body. grammar_body --> ['-?->'], grammar_body. grammar_body --> grammar_body_item. grammar_body_item --> ['!']. grammar_body_item --> ['{'], Prolog_goals, ['}']. grammar_body_item --> non_terminal. grammar_body_item --> terminal. \end{verbatim} \end{quote} The non-terminals are syntactically identical to prolog goals (atom, compound term or variable), the terminals are lists of prolog terms (typically characters or tokens). Every term is transformed, unless it is enclosed in curly brackets. The control constructs like conjunction \predspec{,/2}, disjunction (\predspec{;/2} or \predspec{|/2}), the cut (\predspec{!/0}), the condition (\predspec{\notation{->}/1}) and do-loops need not be enclosed in curly brackets. The grammar can be accessed with the built-in \bipref{phrase/3}{../bips/kernel/control/phrase-3.html}. The first argument of \bipref{phrase/3}{../bips/kernel/control/phrase-3.html} is the name of the grammar to be used, the second argument is a list containing the input to be parsed. If the parsing is successful the built-in will succeed. For instance, with the grammar \begin{quote} \begin{verbatim} a --> [] | [z], a. \end{verbatim} \end{quote} \notation{phrase(a, X, [])} will give on backtracking \begin{quote} \begin{verbatim} X = [z] ; X = [z, z] ; X = [z, z, z] ; .... \end{verbatim} \end{quote} \subsection{Simple DCG example} The following example illustrates a simple grammar declared using the DCGs. \begin{quote} \begin{verbatim} sentence --> imperative, noun_phrase(Number), verb_phrase(Number). imperative, [you] --> []. imperative --> []. noun_phrase(Number) --> determiner, noun(Number). noun_phrase(Number) --> pronoun(Number). verb_phrase(Number) --> verb(Number). verb_phrase(Number) --> verb(Number), noun_phrase(_). determiner --> [the]. noun(singular) --> [man]. noun(singular) --> [apple]. noun(plural) --> [men]. noun(plural) --> [apples]. verb(singular) --> [eats]. verb(singular) --> [sings]. verb(plural) --> [eat]. verb(plural) --> [sing]. pronoun(plural) --> [you]. \end{verbatim} \end{quote} The above grammar may be applied by using \bipref{phrase/3}{../bips/kernel/control/phrase-3.html}. If the predicate succeeds then the input has been parsed successfully. \begin{quote} \begin{verbatim} [eclipse 1]: phrase(sentence, [the,man,eats,the,apple], []). yes. [eclipse 2]: phrase(sentence, [the,men,eat], []). yes. [eclipse 3]: phrase(sentence, [the,men,eats], []). no. [eclipse 4]: phrase(sentence, [eat,the,apples], []). yes. [eclipse 5]: phrase(sentence, [you,eat,the,man], []). yes. \end{verbatim} \end{quote} The predicate \bipref{phrase/3}{../bips/kernel/control/phrase-3.html} may be used to return the point at which parsing of input fails---if the returned list is empty then the input has been successfully parsed. \begin{quote} \begin{verbatim} [eclipse 1]: phrase(sentence, [the,man,eats,something,nasty],X). X = [something, nasty] More? (;) no (more) solution. [eclipse 2]: phrase(sentence, [eat,the,apples],X). X = [the, apples] More? (;) X = [] More? (;) no (more) solution. [eclipse 3]: phrase(sentence, [hello,there],X). no (more) solution. \end{verbatim} \end{quote} \subsection{Mapping to Prolog clauses} A grammar rule is translated to a Prolog clause by adding two arguments which represent the input before and after the nonterminal which is represented by the rule. The effect of the transformation can be observed, e.g., by calling \bipref{expand_clause/2}{../bips/kernel/compiler/expand_clause-2.html}: \begin{quote} \begin{verbatim} [eclipse 1]: expand_clause(p(X) --> q(X), Expanded). X = X Expanded = p(X, _250, _243) :- q(X, _250, _243) Yes (0.00s cpu) [eclipse 2]: expand_clause(p(X) --> [a], Expanded). X = X Expanded = p(X, _251, _244) :- 'C'(_251, a, _244) Yes (0.00s cpu) \end{verbatim} \end{quote} \subsection{Parsing other data structures} DCGs are in principle not limited to the parsing of lists. The predicate \bipref{'C'/3}{../bips/kernel/termmanip/C-3.html} is responsible for reading resp.\ generating the input tokens. The default definition is \begin{quote} \begin{verbatim} 'C'([Token|Rest], Token, Rest). \end{verbatim} \end{quote} The first argument represents the parsing input before consuming \about{Token} and \about{Rest} is the input after consuming \about{Token}. By redefining \predspec{'C'/3}, it is possible to apply a DCG to input sources other than a list, e.g., to parse directly from an I/O stream: \begin{quote} \begin{verbatim} :- local 'C'/3. 'C'(Stream-Pos0, Token, Stream-Pos1) :- seek(Stream, Pos0), read_string(Stream, " ", _, TokenString), atom_string(Token, TokenString), at(Stream, Pos1). sentence --> noun, [is], adjective. noun --> [prolog] ; [lisp]. adjective --> [boring] ; [great]. \end{verbatim} \end{quote} This can then be applied to a string as follows: \begin{quote} \begin{verbatim} [eclipse 1]: String = "prolog is great", open(String, string, S), phrase(sentence, S-0, S-End). ... End = 15 yes. \end{verbatim} \end{quote} Here is another redefinition of \predspec{'C'/3}, using a similar idea, which allows direct parsing of {\eclipse} strings as sequences of characters: \begin{quote} \begin{verbatim} :- local 'C'/3. 'C'(String-Pos0, Char, String-Pos1) :- Pos0 =< string_length(String), string_code(String, Pos0, Char), Pos1 is Pos0+1. anagram --> []. anagram --> [_]. anagram --> [C], anagram, [C]. \end{verbatim} \end{quote} This can then be applied to a string as follows: \begin{quote} \begin{verbatim} [eclipse 1]: phrase(anagram, "abba"-1, "abba"-5). yes. [eclipse 2]: phrase(anagram, "abca"-1, "abca"-5). no (more) solution. \end{verbatim} \end{quote} Unlike the default definition, these redefinitions of \predspec{'C'/3} are not bi-directional. Consequently, the grammar rules using them can only be used for parsing, not for generating sentences. Note that every grammar rule uses that definition of \predspec{'C'/3} which is visible in the module where the grammar rule itself is defined. %HEVEA\cutend