% Copyright 2005-2017 Cisco Systems, Inc. % % Licensed under the Apache License, Version 2.0 (the "License"); % you may not use this file except in compliance with the License. % You may obtain a copy of the License at % % http://www.apache.org/licenses/LICENSE-2.0 % % Unless required by applicable law or agreed to in writing, software % distributed under the License is distributed on an "AS IS" BASIS, % WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. % See the License for the specific language governing permissions and % limitations under the License. \chapter{Compatibility Features\label{CHPTCOMPAT}} This chapter describes several items that are included with current versions of {\ChezScheme} primarily for compatibility with older versions of the system. Section~\ref{SECTCOMPATHASHTABLES} describes a hash-table interface that has since been replaced by the R6RS hashtable interface. Section~\ref{SECTCOMPATEXTENDSYNTAX} describes \scheme{extend-syntax} macros. These features are supported directly by current versions of {\ChezScheme}, but support may be dropped in future versions. New programs should use the standard mechanisms described in \emph{The Scheme Programming Language, 4th Edition}~\cite{Dybvig:tspl4} instead. Section~\ref{SECTCOMPATSTRUCTURES} describes a mechanism for defining record-like structures as vectors instead of new unique types. New programs should use \scheme{define-record}, which is described in Section~\ref{SECTCSV7RECORDS}, instead. Section~\ref{SECTCOMPATOTHER} describes a compatibility file distributed with {\ChezScheme} that contains definitions for forms and procedures no longer supported directly by {\ChezScheme}. % undocumented: % application-expander not bothering... % constant-expander not bothering... % variable-expander not bothering... % syntax-match? not bothering... % extend-syntax/code not bothering... \section{Hash Tables\label{SECTCOMPATHASHTABLES}} The hash table procedures here are obviated by the new hash table procedures listed in Section~\ref{SECTMISCHASHTABLES}. %---------------------------------------------------------------------------- \entryheader \formdef{make-hash-table}{\categoryprocedure}{(make-hash-table)} \formdef{make-hash-table}{\categoryprocedure}{(make-hash-table \var{weak?})} \returns a new hash table \listlibraries \endentryheader If \var{weak?} is provided and is non-false, the hash table is a weak hash table, which means that it does not protect keys from the garbage collector. Keys reclaimed by the garbage collector are removed from the table, and their associated values are dropped the next time the table is modified, if not sooner. %---------------------------------------------------------------------------- \entryheader \formdef{hash-table?}{\categoryprocedure}{(hash-table? \var{obj})} \returns \scheme{#t} if \var{obj} is a hash table, otherwise \scheme{#f} \listlibraries \endentryheader %---------------------------------------------------------------------------- \entryheader \formdef{put-hash-table!}{\categoryprocedure}{(put-hash-table! \var{ht} \var{k} \var{v})} \returns unspecified \listlibraries \endentryheader \var{ht} must be a hash table. \var{k} and \var{v} may be any Scheme values. \scheme{put-hash-table!} associates the value \var{v} with the key \var{k} in \var{ht}. %---------------------------------------------------------------------------- \entryheader \formdef{get-hash-table}{\categoryprocedure}{(get-hash-table \var{ht} \var{k} \var{d})} \returns see below \listlibraries \endentryheader \scheme{get-hash-table} returns the value associated with \var{k} in \var{ht}. If no value is associated with \var{k} in \var{ht}, \scheme{get-hash-table} returns \var{d}. Key comparisons are performed with \var{eq?}. Because objects may be moved by the garbage collector, \scheme{get-hash-table} may need to rehash some objects and therefore cause side effects in the hash table. Thus, it is not safe to perform concurrent accesses of the same hash table from multiple threads using \scheme{get-hash-table}. %---------------------------------------------------------------------------- \entryheader \formdef{remove-hash-table!}{\categoryprocedure}{(remove-hash-table! \var{ht} \var{k})} \returns unspecified \listlibraries \endentryheader \scheme{remove-hash-table!} drops any association for \var{k} from \var{ht}. %---------------------------------------------------------------------------- \entryheader \formdef{hash-table-map}{\categoryprocedure}{(hash-table-map \var{ht} \var{p})} \returns see below \listlibraries \endentryheader \scheme{hash-table-map} applies \var{p} to each key, value association in \var{ht}, in no particular order, and returns a list of the resulting values, again in no particular order. \var{p} should accept two arguments, a key and a value. %---------------------------------------------------------------------------- \entryheader \formdef{hash-table-for-each}{\categoryprocedure}{(hash-table-for-each \var{ht} \var{p})} \returns unspecified \listlibraries \endentryheader \scheme{hash-table-for-each} applies \var{p} to each key, value association in \var{ht}, in no particular order. Unlike \scheme{hash-table-map}, it does not create a list of the values; instead, it's value is unspecified. \var{p} should accept two arguments, a key and a value. \section{Extend-Syntax Macros\label{SECTCOMPATEXTENDSYNTAX}} This section describes \scheme{extend-syntax}, a powerful yet easy to use syntactic extension facility based on \index{pattern matching}pattern matching~\cite{Kohlbecker:phd}. Syntactic transformations written using \scheme{extend-syntax} are similar to those written using a \scheme{define-syntax} with \scheme{syntax-case}, except that the transformations produced by \scheme{extend-syntax} do not automatically respect lexical scoping. It is not typically possible to mix syntactic abstractions written using \scheme{syntax-case} with those written using \scheme{extend-syntax} seamlessly; it is generally preferable to use one or the other wherever possible. Support for \scheme{extend-syntax} within the \scheme{syntax-case} expander is provided only as an aid to migrating to \scheme{syntax-case}. %---------------------------------------------------------------------------- \entryheader \formdef{extend-syntax}{\categorysyntax}{(extend-syntax (\var{name} \var{key} \dots) (\var{pat} \var{fender} \var{template}) \dots)} \returns unspecified \listlibraries \endentryheader \noindent The identifier \var{name} is the name, or syntax keyword, for the syntactic extension to be defined. When the system expander processes any list expression whose car is \var{name}, the syntactic transformation procedure generated by \scheme{extend-syntax} is invoked on this expression. The remaining identifiers \scheme{\var{key} \dots} are additional keywords to be recognized within input expressions during expansion (such as \scheme{else} in \scheme{cond} or \scheme{case}). Each clause after the list of keys consists of a pattern \var{pat}, an optional \index{fenders}\var{fender}, and a \var{template}. The optional \var{fender} is omitted more often than not. The \var{pat} specifies the syntax the input expression must have for the clause to be chosen. Identifiers within the pattern that are not keywords (\emph{pattern variables}) are bound to corresponding pieces of the input expression. If present, the \var{fender} is a Scheme expression that specifies additional constraints on the input expression (accessed through the pattern variables) that must be satisfied in order for the clause to be chosen. The \var{template} specifies what form the output takes, usually in terms of the pattern variables. During expansion, the transformation procedure \scheme{extend-syntax} generates attempts to match the input expression against each pattern in the order the clauses are given. If the input expression matches the pattern, the pattern variables are bound to the corresponding pieces of the input expression and the fender for the clause, if any, is evaluated. If the fender returns a true value, the given expansion is performed. If input does not match the pattern or if the fender returns a false value, the transformation procedure tries the next clause. An exception is raised with condition type \scheme{&assertion} if no clause can be chosen. Within the pattern, \index{\scheme{...}~(ellipses)}\index{ellipses (~\scheme{...}~)}\emph{ellipsis} (\scheme{...}) may be used to specify zero or more occurrences of the preceding pattern fragment, or prototype. Similarly, ellipses may be used in the output to specify the construction of zero or more expansion prototypes. In this case, the expansion prototype must contain part of an input pattern prototype. The use of patterns, templates, ellipses within patterns and templates, and fenders is illustrated in the following sequence of examples. The first example, defining \index{\scheme{rec}}\scheme{rec}, uses a single keyword, a single clause with no fender, and no ellipses. \schemedisplay (extend-syntax (rec) [(rec id val) (let ([id #f]) (set! id val) id)]) \endschemedisplay The second example, defining \index{\scheme{when}}\scheme{when}, shows the use of ellipses. \schemedisplay (extend-syntax (when) [(when test exp1 exp2 ...) (if test (begin exp1 exp2 ...) #f)]) \endschemedisplay The next example shows the definition of \index{\scheme{let}}\scheme{let}. The definition of \scheme{let} shows the use of multiple ellipses, employing one for the identifier/value pairs and one for the expressions in the body. It also shows that the prototype need not be a single identifier, and that pieces of the prototype may be separated from one another in the template. \schemedisplay (extend-syntax (let) [(let ([x e] ...) b1 b2 ...) ((lambda (x ...) b1 b2 ...) e ...)]) \endschemedisplay The next example shows \index{\scheme{let*}}\scheme{let*}, whose syntax is the same as for \scheme{let}, but which is defined recursively in terms of \scheme{let} with two clauses (one for the base case, one for the recursion step) since it must produce a nested structure. \schemedisplay (extend-syntax (let*) [(let* () b1 b2 ...) (let () b1 b2 ...)] [(let* ([x e] more ...) b1 b2 ...) (let ([x e]) (let* (more ...) b1 b2 ...))]) \endschemedisplay The first pattern/template pair matches any \scheme{let*} expression with no identifier/value pairs and maps it into the equivalent \scheme{begin} expression. This is the base case. The second pattern/template pair matches any \scheme{let*} expression with one or more identifier/value pairs and transforms it into a \scheme{let} expression binding the first pair whose body is a \scheme{let*} expression binding the remaining pairs. This is the recursion step, which will eventually lead us to the base case because we remove one identifier/value pair at each step. Notice that the second pattern uses the pattern variable \scheme{more} for the second and later identifier/value pairs; this makes the pattern and template less cluttered and makes it clear that only the first identifier/value pair is dealt with explicitly. The definition for \index{\scheme{and}}\scheme{and} requires three clauses. The first clause is necessary to recognize \scheme{(and)}, and the second two define all other \scheme{and} forms recursively. \schemedisplay (extend-syntax (and) [(and) #t] [(and x) x] [(and x y ...) (if x (and y ...) #f)]) \endschemedisplay The definition for \index{\scheme{cond}}\scheme{cond} requires four clauses. As with \scheme{let*}, \scheme{cond} must be described recursively, partly because it produces nested \scheme{if} expressions, and partly because one ellipsis prototype would not be sufficient to describe all possible \scheme{cond} clauses. The definition of \scheme{cond} also requires that we specify \scheme{else} as a keyword, in addition to \scheme{cond}. Here is the definition: \schemedisplay (extend-syntax (cond else) [(cond) #f] [(cond (else e1 e2 ...)) (begin e1 e2 ...)] [(cond (test) more ...) (or test (cond more ...))] [(cond (test e1 e2 ...) more ...) (if test (begin e1 e2 ...) (cond more ...))]) \endschemedisplay \noindent Two of the clauses are base cases and two are recursion steps. The first base case is an empty \scheme{cond}. The value of \scheme{cond} in this case is unspecified, so the choice of \scheme{#f} is somewhat arbitrary. The second base case is a \scheme{cond} containing only an \scheme{else} clause; this is transformed to the equivalent \scheme{begin} expression. The two recursion steps differ in the number of expressions in the \scheme{cond} clause. The value of \scheme{cond} when the first true clause contains only the test expression is the value of the test. This is similar to what \scheme{or} does, so we expand the \scheme{cond} clause into an \scheme{or} expression. On the other hand, when there are expressions following the test expression, the value of the last expression is returned, so we use \scheme{if} and \scheme{begin}. To be absolutely correct about the syntax of \scheme{let}, we actually must require that the bound identifiers in the input are symbols. If we typed something like \scheme{(let ([3 x]) x)} we would not get an error from \scheme{let} because it does not check to verify that the objects in the identifier positions are symbols. Instead, \scheme{lambda} may complain, or perhaps the system evaluator long after expansion is complete. This is where \index{fenders}fenders are useful. \schemedisplay (extend-syntax (let) [(let ([x e] ...) b1 b2 ...) (andmap symbol? '(x ...)) ((lambda (x ...) b1 b2 ...) e ...)]) \endschemedisplay \noindent The \index{\scheme{andmap}}\scheme{andmap} of \scheme{symbol?} over \scheme{'(x ...)} assures that each bound identifier is a symbol. A fender is simply a Scheme expression. Within that expression, any quoted object is first expanded by the same rules as the template part of the clause. In this case, \scheme{'(x ...)} is expanded to the list of identifiers from the identifier/value pairs. \scheme{extend-syntax} typically handles everything you need it for, but some syntactic extension definitions require the ability to include the result of evaluating an arbitrary Scheme expression. This ability is provided by \scheme{with}. %---------------------------------------------------------------------------- \entryheader \formdef{with}{\categorysyntax}{(with ((\var{pat} \var{expr}) \dots) \var{template})} \returns processed \var{template} \nolistlibraries \endentryheader \noindent \scheme{with} is valid only within an template inside of \scheme{extend-syntax}. \scheme{with} patterns are the same as \scheme{extend-syntax} patterns, \scheme{with} expressions are the same as \scheme{extend-syntax} fenders, and \scheme{with} templates are the same as \scheme{extend-syntax} templates. \scheme{with} can be used to introduce new pattern identifiers bound to expressions produced by arbitrary Scheme expressions within \scheme{extend-syntax} templates. That is, \scheme{with} allows an escape from the declarative style of \scheme{extend-syntax} into the procedural style of full Scheme. One common use of \scheme{with} is the introduction of a temporary identifier or list of temporary identifiers into a template. \scheme{with} is also used to perform complex transformations that might be clumsy or inefficient if performed within the \scheme{extend-syntax} framework. For example, \scheme{or} requires the use of a temporary identifier. We could define \scheme{or} as follows. \schemedisplay (extend-syntax (or) [(or) #f] [(or x) x] [(or x y ...) (let ([temp x]) (if temp temp (or y ...)))]) \endschemedisplay \noindent This would work until we placed an \scheme{or} expression within the scope of an occurrence of \scheme{temp}, in which case strange things could happen, since \scheme{extend-syntax} does not respect lexical scoping. (This is one of the reasons that \scheme{define-syntax} is preferable to \scheme{extend-syntax}.) \schemedisplay (let ([temp #t]) (or #f temp)) ;=> #f \endschemedisplay \noindent One solution is to use \index{\scheme{gensym}}\scheme{gensym} and \scheme{with} to create a temporary identifier, as follows. \schemedisplay (extend-syntax (or) [(or) #f] [(or x) x] [(or x y ...) (with ([temp (gensym)]) (let ([temp x]) (if temp temp (or y ...))))]) \endschemedisplay \noindent Also, \scheme{with} can be used to combine elements of the input pattern in ways not possible directly with \scheme{extend-syntax}, such as the following \scheme{folding-plus} example. \schemedisplay (extend-syntax (folding-plus) [(folding-plus x y) (and (number? 'x) (number? 'y)) (with ([val (+ 'x 'y)]) val)] [(folding-plus x y) (+ x y)]) \endschemedisplay \noindent \scheme{folding-plus} collapses into the value of \scheme{(+ x y)} if both \scheme{x} and \scheme{y} are numeric constants. Otherwise, \scheme{folding-plus} is transformed into \scheme{(+ x y)} for later evaluation. The fender checks that the operands are numbers at expansion time, and the \scheme{with} performs the evaluation. As with fenders, expansion is performed only within a quoted expressions, since \scheme{quote} sets the data apart from the remainder of the Scheme expression. The example below binds a list of pattern variables to a list of temporary symbols, taking advantage of the fact that \scheme{with} allows us to bind patterns to expressions. This list of temporaries helps us to implement the \scheme{sigma} syntactic extension. \scheme{sigma} is similar to \scheme{lambda}, except it assigns the identifiers in the identifier list instead of creating new bindings. It may be used to perform a series of assignments in parallel. \schemedisplay (extend-syntax (sigma) [(sigma (x ...) e1 e2 ...) (with ([(t ...) (map (lambda (x) (gensym)) '(x ...))]) (lambda (t ...) (set! x t) ... e1 e2 ...))]) (let ([x 'a] [y 'b]) ((sigma (x y) (list x y)) y x)) ;=> (b a) \endschemedisplay The final example below uses \scheme{extend-syntax} to implement \scheme{define-structure}, following a similar example using \scheme{syntax-case} in Section~\ref{TSPL:SECTSYNTAXEXAMPLES} of \emph{The Scheme Programming Language, 4th Edition}. The definition of \scheme{define-structure} makes use of two pattern/template clauses. Two clauses are needed to handle the optionality of the second subexpression. The first clause matches the form without the second subexpression and merely converts it into the equivalent form with the second subexpression present, but empty. The definition also makes heavy use of \index{\scheme{with}}\scheme{with} to evaluate Scheme expressions at expansion time. The first four \scheme{with} clauses are used to manufacture the identifiers that name the automatically defined procedures. (The procedure \index{\scheme{format}}\scheme{format} is particularly useful here, but it could be replaced with \scheme{string-append!}, using \scheme{symbol->string} as needed.) The first two clauses yield single identifiers (for the constructor and predicate), while the next two yield lists of identifiers (for the field access and assignment procedures). The fifth \scheme{with} clause (the final clause in the outer \scheme{with}) is used to count the total length vector needed for each instance of the structure, which must include room for the name and all of the fields. The final \scheme{with} clause (the only clause in the inner \scheme{with}) is used to create a list of vector indexes, one for each field (starting at 1, since the structure name occupies position 0). \schemedisplay (extend-syntax (define-structure) [(define-structure (name id1 ...)) (define-structure (name id1 ...) ())] [(define-structure (name id1 ...) ([id2 val] ...)) (with ([constructor (string->symbol (format "make-~a" 'name))] [predicate (string->symbol (format "~a?" 'name))] [(access ...) (map (lambda (x) (string->symbol (format "~a-~a" 'name x))) '(id1 ... id2 ...))] [(assign ...) (map (lambda (x) (string->symbol (format "set-~a-~a!" 'name x))) '(id1 ... id2 ...))] [count (length '(name id1 ... id2 ...))]) (with ([(index ...) (let f ([i 1]) (if (= i 'count) '() (cons i (f (+ i 1)))))]) (begin (define constructor (lambda (id1 ...) (let* ([id2 val] ...) (vector 'name id1 ... id2 ...)))) (define predicate (lambda (obj) (and (vector? obj) (= (vector-length obj) count) (eq? (vector-ref obj 0) 'name)))) (define access (lambda (obj) (vector-ref obj index))) ... (define assign (lambda (obj newval) (vector-set! obj index newval))) ...)))]) \endschemedisplay \section{Structures\label{SECTCOMPATSTRUCTURES}} \index{structures}This section describes a mechanism, similar to the record-defining mechanisms of Section~\ref{SECTCSV7RECORDS}, that permits the creation of data structures with fixed sets of named fields. Unlike record types, structure types are not unique types, but are instead implemented as vectors. Specifically, a structure is implemented as a vector whose length is one more than the number of fields and whose first element contains the symbolic name of the structure. The representation of structures as vectors simplifies reading and printing of structures somewhat as well as extension of the structure definition facility. It does, however, have some drawbacks. One is that structures may be treated as ordinary vectors by mistake in situations where doing so is inappropriate. When dealing with both structures and vectors in a program, care must be taken to look for the more specific structure type before checking for the more generic vector type, e.g., in a series of \scheme{cond} clauses. A similar drawback is that structure instances are easily ``forged,'' either intentionally or by accident. It is also impossible to control how structures are printed and read. Structures are created via \scheme{define-structure}. Each structure definition defines a constructor procedure, a type predicate, an access procedure for each of its fields, and an assignment procedure for each of its fields. \scheme{define-structure} allows the programmer to control which fields are arguments to the generated constructor procedure and which fields are explicitly initialized by the constructor procedure. \scheme{define-structure} is simple yet powerful enough for most applications, and it is easily extended to handle many applications for which it is not sufficient. The definition of \scheme{define-structure} given at the end of this section can serve as a starting point for more complicated variants. %---------------------------------------------------------------------------- \entryheader \formdef{define-structure}{\categorysyntax}{(define-structure (\var{name} \var{id_1} \dots) ((\var{id_2} \var{expr}) \dots))} \returns unspecified \listlibraries \endentryheader \noindent A \scheme{define-structure} form is a definition and may appear anywhere and only where other definitions may appear. \scheme{define-structure} defines a new data structure, \var{name}, and creates a set of procedures for creating and manipulating instances of the structure. The identifiers \scheme{\var{id_1} \dots} and \scheme{\var{id_2} \dots} name the fields of the data structure. The following procedures are defined by \scheme{define-structure}: \begin{itemize} \item a constructor procedure whose name is \scheme{make-\var{name}}, \item a type predicate whose name is \scheme{\var{name}?}, \item an access procedure whose name is \scheme{\var{name}-\var{field}} for each field name \scheme{\var{id_1} \dots} and \scheme{\var{id_2} \dots}, and \item an assignment procedure whose name is \scheme{set-\var{name}-\var{field}!} for each field name \scheme{\var{id_1} \dots} and \scheme{\var{id_2} \dots}. \end{itemize} The fields named by the identifiers \scheme{\var{id_1} \dots} are initialized by the arguments to the constructor procedure. The fields named by the identifiers \scheme{\var{id_2} \dots} are initialized explicitly to the values of the expressions \scheme{\var{expr} \dots}. Each expression is evaluated within the scope of the identifiers \scheme{\var{id_1} \dots} (bound to the corresponding field values) and any of the identifiers \scheme{\var{id_2} \dots} (bound to the corresponding field values) appearing before it (as if within a \scheme{let*}). To clarify, the constructor behaves as if defined as \schemedisplay (define make-\var{name} (lambda (\var{id_1} \dots) (let* ([\var{id_2} \var{expr}] \dots) \var{body}))) \endschemedisplay \noindent where \var{body} builds the structure from the values of the identifiers \scheme{\var{id_1} \dots} and \scheme{\var{id_2} \dots}. If no fields other than those initialized by the arguments to the constructor procedure are needed, the second subexpression, \scheme{((\var{id_2} \var{expr}) \dots)}, may be omitted. \index{pares}\index{\scheme{make-pare}}The following simple example demonstrates how pairs might be defined in Scheme if they did not already exist. Both fields are initialized by the arguments to the constructor procedure. \schemedisplay (define-structure (pare car cdr)) (define p (make-pare 'a 'b)) (pare? p) ;=> #t (pair? p) ;=> #f (pare? '(a . b)) ;=> #f (pare-car p) ;=> a (pare-cdr p) ;=> b (set-pare-cdr! p (make-pare 'b 'c)) (pare-car (pare-cdr p)) ;=> b (pare-cdr (pare-cdr p)) ;=> c \endschemedisplay The following example defines a handy string data structure, called a \index{stretch strings}\emph{stretch-string}, that grows as needed. This example uses a field explicitly initialized to a value that depends on the value of the constructor argument fields. \schemedisplay (define-structure (stretch-string length fill) ([string (make-string length fill)])) (define stretch-string-ref (lambda (s i) (let ([n (stretch-string-length s)]) (when (>= i n) (stretch-stretch-string! s (+ i 1) n)) (string-ref (stretch-string-string s) i)))) (define stretch-string-set! (lambda (s i c) (let ([n (stretch-string-length s)]) (when (>= i n) (stretch-stretch-string! s (+ i 1) n)) (string-set! (stretch-string-string s) i c)))) (define stretch-string-fill! (lambda (s c) (string-fill! (stretch-string-string s) c) (set-stretch-string-fill! s c))) (define stretch-stretch-string! (lambda (s i n) (set-stretch-string-length! s i) (let ([str (stretch-string-string s)] [fill (stretch-string-fill s)]) (let ([xtra (make-string (- i n) fill)]) (set-stretch-string-string! s (string-append str xtra)))))) \endschemedisplay \noindent As often happens, most of the procedures defined automatically are used only to define more specialized procedures, in this case the procedures \scheme{stretch-string-ref} and \scheme{stretch-string-set!}. \scheme{stretch-string-length} and \scheme{stretch-string-string} are the only automatically defined procedures that are likely to be called directly in code that uses stretch strings. \schemedisplay (define ss (make-stretch-string 2 #\X)) (stretch-string-string ss) ;=> "XX" (stretch-string-ref ss 3) ;=> #\X (stretch-string-length ss) ;=> 4 (stretch-string-string ss) ;=> "XXXX" (stretch-string-fill! ss #\@) (stretch-string-string ss) ;=> "@@@@" (stretch-string-ref ss 5) ;=> #\@ (stretch-string-string ss) ;=> "@@@@@@" (stretch-string-set! ss 7 #\=) (stretch-string-length ss) ;=> 8 (stretch-string-string ss) ;=> "@@@@@@@=" \endschemedisplay Section~\ref{TSPL:SECTSYNTAXEXAMPLES} of {\TSPLFOUR} defines a simplified variant of \scheme{define-structure} as an example of the use of \index{\scheme{syntax-case}}\scheme{syntax-case}. The definition given below implements the complete version. \scheme{define-structure} expands into a series of definitions for names generated from the structure name and field names. The generated identifiers are created with \index{\scheme{datum->syntax}}\scheme{datum->syntax} to make the identifiers visible where the \scheme{define-structure} form appears. Since a \scheme{define-structure} form expands into a \scheme{begin} containing definitions, it is itself a definition and can be used wherever definitions are valid. \schemedisplay (define-syntax define-structure (lambda (x) (define gen-id (lambda (template-id . args) (datum->syntax template-id (string->symbol (apply string-append (map (lambda (x) (if (string? x) x (symbol->string (syntax->datum x)))) args)))))) (syntax-case x () ((_ (name field1 ...)) (andmap identifier? #'(name field1 ...)) #'(define-structure (name field1 ...) ())) ((_ (name field1 ...) ((field2 init) ...)) (andmap identifier? #'(name field1 ... field2 ...)) (with-syntax ((constructor (gen-id #'name "make-" #'name)) (predicate (gen-id #'name #'name "?")) ((access ...) (map (lambda (x) (gen-id x #'name "-" x)) #'(field1 ... field2 ...))) ((assign ...) (map (lambda (x) (gen-id x "set-" #'name "-" x "!")) #'(field1 ... field2 ...))) (structure-length (+ (length #'(field1 ... field2 ...)) 1)) ((index ...) (let f ([i 1] [ids #'(field1 ... field2 ...)]) (if (null? ids) '() (cons i (f (+ i 1) (cdr ids))))))) #'(begin (define constructor (lambda (field1 ...) (let* ([field2 init] ...) (vector 'name field1 ... field2 ...)))) (define predicate (lambda (x) (and (vector? x) (#3%fx= (vector-length x) structure-length) (eq? (vector-ref x 0) 'name)))) (define access (lambda (x) (vector-ref x index))) ... (define assign (lambda (x update) (vector-set! x index update))) ...)))))) \endschemedisplay \section{Compatibility File\label{SECTCOMPATOTHER}} Current versions of {\ChezScheme} are distributed with a compatibility file containing definitions of various syntactic forms and procedures supported by earlier versions of {\ChezScheme} for which support has since been dropped. This file, \scheme{compat.ss}, is typically installed in the library subdirectory of the {\ChezScheme} installation directory. In some cases, the forms and procedures found in \scheme{compat.ss} have been dropped because they were infrequently used and easily written directly in Scheme. In other cases, the forms and procedures have been rendered obsolete by improvements in the system. In such cases, new code should be written to use the newer features, and older code should be rewritten if possible to use the newer features as well.