📄 r5rs-z-h-4.html
字号:
<!doctype html public "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd"><html><!-- Generated from TeX source by tex2page, v 4o4, (c) Dorai Sitaram, http://www.cs.rice.edu/~dorai/tex2page --><head><title>Revised^5 Report on the Algorithmic Language Scheme</title><link rel="stylesheet" type="text/css" href="r5rs-Z-C.css" title=default><meta name=robots content="noindex,follow"></head><body><p><div class=navigation>[Go to <span><a href="r5rs.html">first</a>, <a href="r5rs-Z-H-3.html">previous</a></span><span>, <a href="r5rs-Z-H-5.html">next</a></span> page<span>; </span><span><a href="r5rs-Z-H-2.html#%_toc_start">contents</a></span><span><span>; </span><a href="r5rs-Z-H-15.html#%_index_start">index</a></span>]</div><p><a name="%_chap_1"></a><h1 class=chapter><div class=chapterheading><a href="r5rs-Z-H-2.html#%_toc_%_chap_1">Chapter 1</a></div><p><a href="r5rs-Z-H-2.html#%_toc_%_chap_1">Overview of Scheme</a></h1><p><a name="%_sec_1.1"></a><h2><a href="r5rs-Z-H-2.html#%_toc_%_sec_1.1">1.1 Semantics</a></h2><p><p>This section gives an overview of Scheme's semantics. Adetailed informal semantics is the subject ofchapters <a href="r5rs-Z-H-6.html#%_chap_3">3</a> through <a href="r5rs-Z-H-9.html#%_chap_6">6</a>. For referencepurposes, section <a href="r5rs-Z-H-10.html#%_sec_7.2">7.2</a> provides a formalsemantics of Scheme.<p>Following Algol, Scheme is a statically scoped programminglanguage. Each use of a variable is associated with a lexicallyapparent binding of that variable.<p>Scheme has latent as opposed to manifest types. Typesare associated with values (also called objects<a name="%_idx_2"></a>) rather thanwith variables. (Some authors refer to languages with latent types asweakly typed or dynamically typed languages.) Other languages withlatent types are APL, Snobol, and other dialects of Lisp. Languageswith manifest types (sometimes referred to as strongly typed orstatically typed languages) include Algol 60, Pascal, and C.<p>All objects created in the course of a Scheme computation, includingprocedures and continuations, have unlimited extent.No Scheme object is ever destroyed. The reason thatimplementations of Scheme do not (usually!) run out of storage is thatthey are permitted to reclaim the storage occupied by an object ifthey can prove that the object cannot possibly matter to any futurecomputation. Other languages in which most objects have unlimitedextent include APL and other Lisp dialects.<p>Implementations of Scheme are required to be properly tail-recursive.This allows the execution of an iterative computation in constant space,even if the iterative computation is described by a syntacticallyrecursive procedure. Thus with a properly tail-recursive implementation,iteration can be expressed using the ordinary procedure-callmechanics, so that special iteration constructs are useful only assyntactic sugar. See section <a href="r5rs-Z-H-6.html#%_sec_3.5">3.5</a>.<p>Scheme procedures are objects in their own right. Procedures can becreated dynamically, stored in data structures, returned as results ofprocedures, and so on. Other languages with these properties includeCommon Lisp and ML. <p>One distinguishing feature of Scheme is that continuations, whichin most other languages only operate behind the scenes, also have``first-class'' status. Continuations are useful for implementing awide variety of advanced control constructs, including non-local exits,backtracking, and coroutines. See section <a href="r5rs-Z-H-9.html#%_sec_6.4">6.4</a>.<p>Arguments to Scheme procedures are always passed by value, whichmeans that the actual argument expressions are evaluated before theprocedure gains control, whether the procedure needs the result of theevaluation or not. ML, C, and APL are three other languages that alwayspass arguments by value.This is distinct from the lazy-evaluation semantics of Haskell,or the call-by-name semantics of Algol 60, where an argumentexpression is not evaluated unless its value is needed by theprocedure.<p><p>Scheme's model of arithmetic is designed to remain as independent aspossible of the particular ways in which numbers are represented within acomputer. In Scheme, every integer is a rational number, every rational is areal, and every real is a complex number. Thus the distinction between integerand real arithmetic, so important to many programming languages, does notappear in Scheme. In its place is a distinction between exact arithmetic,which corresponds to the mathematical ideal, and inexact arithmetic onapproximations. As in Common Lisp, exact arithmetic is not limited tointegers.<p><a name="%_sec_1.2"></a><h2><a href="r5rs-Z-H-2.html#%_toc_%_sec_1.2">1.2 Syntax</a></h2><p>Scheme, like most dialects of Lisp, employs a fully parenthesized prefixnotation for programs and (other) data; the grammar of Scheme generates asublanguage of the language used for data. An importantconsequence of this simple, uniform representation is the susceptibility ofScheme programs and data to uniform treatment by other Scheme programs.For example, the <tt>eval</tt> procedure evaluates a Scheme program expressedas data.<p>The <tt>read</tt> procedure performs syntactic as well as lexical decomposition ofthe data it reads. The <tt>read</tt> procedure parses its input as data(section <a href="r5rs-Z-H-10.html#%_sec_7.1.2">7.1.2</a>), not as program.<p>The formal syntax of Scheme is described in section <a href="r5rs-Z-H-10.html#%_sec_7.1">7.1</a>.<p><a name="%_sec_1.3"></a><h2><a href="r5rs-Z-H-2.html#%_toc_%_sec_1.3">1.3 Notation and terminology</a></h2><p><a name="%_sec_1.3.1"></a><h3><a href="r5rs-Z-H-2.html#%_toc_%_sec_1.3.1">1.3.1 Primitive, library, and optional features</a></h3><p><p>It is required that every implementation of Scheme support allfeatures that are not marked as being <a name="%_idx_4"></a><em>optional</em>. Implementations arefree to omit optional features of Scheme or to add extensions,provided the extensions are not in conflict with the language reportedhere. In particular, implementations must support portable code byproviding a syntactic mode that preempts no lexical conventions of thisreport.<p>To aid in understanding and implementing Scheme, some features are markedas <a name="%_idx_6"></a><em>library</em>. These can be easily implemented in terms of the other,primitive, features. They are redundant in the strict sense ofthe word, but they capture common patterns of usage, and are thereforeprovided as convenient abbreviations.<p><a name="%_sec_1.3.2"></a><h3><a href="r5rs-Z-H-2.html#%_toc_%_sec_1.3.2">1.3.2 Error situations and unspecified behavior</a></h3><p><a name="%_idx_8"></a>When speaking of an error situation, this report uses the phrase ``anerror is signalled'' to indicate that implementations must detect andreport the error. If such wording does not appear in the discussion ofan error, then implementations are not required to detect or report theerror, though they are encouraged to do so. An error situation thatimplementations are not required to detect is usually referred to simplyas ``an error.''<p>For example, it is an error for a procedure to be passed an argument thatthe procedure is not explicitly specified to handle, even though suchdomain errors are seldom mentioned in this report. Implementations mayextend a procedure's domain of definition to include such arguments.<p>This report uses the phrase ``may report a violation of animplementation restriction'' to indicate circumstances under which animplementation is permitted to report that it is unable to continueexecution of a correct program because of some restriction imposed by theimplementation. Implementation restrictions are of course discouraged,but implementations are encouraged to report violations of implementationrestrictions.<a name="%_idx_10"></a><p>For example, an implementation may report a violation of animplementation restriction if it does not have enough storage to run aprogram.<p>If the value of an expression is said to be ``unspecified,'' thenthe expression must evaluate to some object without signalling an error,but the value depends on the implementation; this report explicitly doesnot say what value should be returned. <a name="%_idx_12"></a><p><p><p><a name="%_sec_1.3.3"></a><h3><a href="r5rs-Z-H-2.html#%_toc_%_sec_1.3.3">1.3.3 Entry format</a></h3><p>Chapters <a href="r5rs-Z-H-7.html#%_chap_4">4</a> and <a href="r5rs-Z-H-9.html#%_chap_6">6</a> are organizedinto entries. Each entry describes one language feature or a group ofrelated features, where a feature is either a syntactic construct or abuilt-in procedure. An entry begins with one or more header lines of the form<p><div align=left><u><i>category</i>:</u> <tt><i>template</i></tt> </div><p>for required, primitive features, or<p><div align=left><u><i>qualifier</i> <i>category</i>:</u> <tt><i>template</i></tt> </div><p>where <i>qualifier</i> is either ``library'' or ``optional'' as definedin section <a href="#%_sec_1.3.1">1.3.1</a>.<p>If <i>category</i> is ``syntax'', the entry describes an expressiontype, and the template gives the syntax of the expression type.Components of expressions are designated by syntactic variables, whichare written using angle brackets, for example, <expression>,<variable>. Syntactic variables should be understood to denote segments ofprogram text; for example, <expression> stands for any string ofcharacters which is a syntactically valid expression. The notation<p> <thing<sub>1</sub>> <tt>...</tt><p>indicates zero or more occurrences of a <thing>, and<p> <thing<sub>1</sub>> <thing<sub>2</sub>> <tt>...</tt><p>indicates one or more occurrences of a <thing>.<p>If <i>category</i> is ``procedure'', then the entry describes a procedure, andthe header line gives a template for a call to the procedure. Argumentnames in the template are <i>italicized</i>. Thus the header line<p><div align=left><u>procedure:</u> <tt>(vector-ref <i>vector</i> <i>k</i>)</tt> </div><p>indicates that the built-in procedure <tt>vector-ref</tt> takestwo arguments, a vector <i>vector</i> and an exact non-negative integer<i>k</i> (see below). The header lines<p><div align=left><u>procedure:</u> <tt>(make-vector <i>k</i>)</tt> </div><div align=left><u>procedure:</u> <tt>(make-vector <i>k</i> <i>fill</i>)</tt> </div><p>indicate that the <tt>make-vector</tt> procedure must be defined to takeeither one or two arguments.<p>It is an error for an operation to be presented with an argument that itis not specified to handle. For succinctness, we follow the conventionthat if an argument name is also the name of a type listed insection <a href="r5rs-Z-H-6.html#%_sec_3.2">3.2</a>, then that argument must be of the named type.For example, the header line for <tt>vector-ref</tt> given above dictates that thefirst argument to <tt>vector-ref</tt> must be a vector. The following namingconventions also imply type restrictions:<p><div align=left><table><tr><td>
<table border=0><tr><td valign=top ><i>obj</i></td><td valign=top >any object</td></tr><tr><td valign=top ><em>l</em><em>i</em><em>s</em><em>t</em>, <em>l</em><em>i</em><em>s</em><em>t</em><sub>1</sub>, <tt>...</tt> <em>l</em><em>i</em><em>s</em><em>t</em><sub><em>j</em></sub>, <tt>...</tt></td><td valign=top >list (see section <a href="r5rs-Z-H-9.html#%_sec_6.3.2">6.3.2</a>)</td></tr><tr><td valign=top ><em>z</em>, <em>z</em><sub>1</sub>, <tt>...</tt> <em>z</em><sub><em>j</em></sub>, <tt>...</tt></td><td valign=top >complex number</td></tr><tr><td valign=top ><em>x</em>, <em>x</em><sub>1</sub>, <tt>...</tt> <em>x</em><sub><em>j</em></sub>, <tt>...</tt></td><td valign=top >real number</td></tr><tr><td valign=top ><em>y</em>, <em>y</em><sub>1</sub>, <tt>...</tt> <em>y</em><sub><em>j</em></sub>, <tt>...</tt></td><td valign=top >real number</td></tr><tr><td valign=top ><em>q</em>, <em>q</em><sub>1</sub>, <tt>...</tt> <em>q</em><sub><em>j</em></sub>, <tt>...</tt></td><td valign=top >rational number</td></tr><tr><td valign=top ><em>n</em>, <em>n</em><sub>1</sub>, <tt>...</tt> <em>n</em><sub><em>j</em></sub>, <tt>...</tt></td><td valign=top >integer</td></tr><tr><td valign=top ><em>k</em>, <em>k</em><sub>1</sub>, <tt>...</tt> <em>k</em><sub><em>j</em></sub>, <tt>...</tt></td><td valign=top >exact non-negative integer</td></tr><tr><td valign=top ></td></tr></table></td></tr></table></div><p><p><p><a name="%_sec_1.3.4"></a><h3><a href="r5rs-Z-H-2.html#%_toc_%_sec_1.3.4">1.3.4 Evaluation examples</a></h3><p>The symbol ``===>'' used in program examples should be read``evaluates to.'' For example,<p><tt><p>(* 5 8) ===> 40<p></tt><p>means that the expression <tt>(* 5 8)</tt> evaluates to the object <tt>40</tt>.Or, more precisely: the expression given by the sequence of characters``<tt>(* 5 8)</tt>'' evaluates, in the initial environment, to an objectthat may be represented externally by the sequence of characters ``<tt>40</tt>''. See section <a href="r5rs-Z-H-6.html#%_sec_3.3">3.3</a> for a discussion of externalrepresentations of objects.<p><a name="%_sec_1.3.5"></a><h3><a href="r5rs-Z-H-2.html#%_toc_%_sec_1.3.5">1.3.5 Naming conventions</a></h3><p>By convention, the names of procedures that always return a booleanvalue usually endin ``<tt>?</tt>''. Such procedures are called predicates.<p>By convention, the names of procedures that store values into previouslyallocated locations (see section <a href="r5rs-Z-H-6.html#%_sec_3.4">3.4</a>) usually end in``<tt>!</tt>''.Such procedures are called mutation procedures.By convention, the value returned by a mutation procedure is unspecified.<p>By convention, ``<tt>-></tt>'' appears within the names of procedures thattake an object of one type and return an analogous object of another type.For example, <tt>list->vector</tt> takes a list and returns a vector whoseelements are the same as those of the list.<p> <p><p><div class=navigation>[Go to <span><a href="r5rs.html">first</a>, <a href="r5rs-Z-H-3.html">previous</a></span><span>, <a href="r5rs-Z-H-5.html">next</a></span> page<span>; </span><span><a href="r5rs-Z-H-2.html#%_toc_start">contents</a></span><span><span>; </span><a href="r5rs-Z-H-15.html#%_index_start">index</a></span>]</div><p></body></html>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -