ALGOL 60(源自ALGOrithmic Language 1960的缩写),是在1960年创建的称为“算法语言”的一种编程语言。它是以后来称为ALGOL 58的“国际代数语言”为基础,其官方后继者是ALGOL 68,它们一起并称为ALGOL语言家族。Algol 60引进了许多新的概念如:块、词法作用域、递归[2]、巴科斯-诺尔范式(BNF),它在编程语言设计和发展演化中有着巨大的影响力。
Quick Facts 编程范型, 语言家族 ...
ALGOL 60编程范型 | 指令式,过程式,结构化 |
---|
语言家族 | ALGOL |
---|
设计者 | Bauer, Rutishauser, Samelson, 巴科斯, Katz, 佩利, Wegstein, 诺尔, Vauquois, 范·韦恩加登, Woodger, J. Green, 麦卡锡 |
---|
发行时间 | 1960年,64年前(1960) |
---|
当前版本 | - ISO 1538:1984(1984年10月15日)[1]
|
---|
型态系统 | 静态, 强类型 |
---|
作用域 | 词法 |
---|
|
ALGOL 58 |
|
ALGOL 68, 所有“类似ALGOL语言”比如: Simula, Pascal, C等, ISWIM, Scheme |
Close
1960年,在巴黎举行的讨论会上,来自欧洲的诺尔、Bauer、Rutishauser、Samelson、范·韦恩加登、Vauquois、Woodger,与来自美国的佩利、巴科斯、麦卡锡、Katz、Wegstein和J. Green,共同发表了《算法语言ALGOL 60报告》[3]。戴克斯特拉实现了ALGOL 60语言的第一个编译器。在1962年罗马会议上,ALGOL 60报告得到了修订,并于1963年出版[4]。
ALGOL 60是程序设计语言发展史上的一个里程碑,影响到其后的Simula、CPL、ALGOL W、ISWIM、BCPL、B、Pascal、C、Scheme等。它标志着程序设计语言成为一门独立的科学学科,并为后来软件自动化及软件可靠性的发展奠定了基础。
ALGOL 60以及COBOL,是最初的企图标准化的编程语言。ALGOL60曾经提出两项ISO标准:
- ISO 1538: 1984年,编程语言ALGOL 60[5],它已稳固化。
- ISO 1672: 1977年,ALGOL基本符号的ISO 7位编码字符集硬件表示[6],它已撤消。
ALGOL 60在语言报告中没有I/O设施;诸多实现以少有相互兼容的方式定义了自己的设施。迄今ALGOL 60已经有了至少70个扩充、扩展、派生和子语言[7]。
除了字母和数字之外,ALGOL定义了运算符:
More information 优先级, 运算符 ...
优先级
|
运算符
|
第一 算术 |
第一 |
↑ (幂)
|
第二 |
× ,/ (实数),÷ (整数)
|
第三 |
+ ,-
|
第二 |
< ,≤ ,= ,≥ ,> ,≠
|
第三 |
¬ (非)
|
第四 |
∧ (与)
|
第五 |
∨ (或)
|
第六 |
⊃ (蕴涵)
|
第七 |
≡ (等价)
|
Close
和分界符:
, |
. |
⏨ |
: |
; |
≔ |
␣
|
( |
) |
[ |
] |
‘ |
’ |
|
在ISO 1672标准中,用:=
表示≔
(U+2254),用*
表示×
;用%
表示÷
,用**
或^
表示↑
,用@
表示科学记数法中指数运算的底数10
所用符号⏨
(U+23E8),用{
表示‘
,并且用}
表示’
,用空格
表示在字符串中的空白字符␣
(U+2423)。
ALGOL 60有24个关键字:
array
begin
Boolean
comment
|
|
|
|
|
|
还包括标准函数名字作为限制标识符:
关键字写法依赖于实现,常见的是一种叫做索绕(stropping)的方法,即将关键字大写并包围在两个'
之间,例如将go to
写为'GOTO'
。在ISO 1672标准中关键字还包括对特定符号的某种文字转写:
More information 符号, 转写 ...
符号 |
转写 |
|
符号 |
转写 |
|
符号 |
转写 |
|
符号 |
转写
|
[ |
'(/' |
] |
'/)' |
‘ |
'(' |
’ |
')'
|
< |
'LT' |
> |
'GT' |
≤ |
'LE' |
≥ |
'GE'
|
= |
'EQ' |
≡ |
'EQV' |
≠ |
'NE' |
¬ |
'NOT'
|
∧ |
'AND' |
∨ |
'OR' |
⊃ |
'IMPL' |
÷ |
'/'
|
Close
在ALGOL 60中,“复合语句”被定义为:包围在语句括号begin
和end
之间的成序列的语句,形成一个复合语句。块被定义为:成序列的声明,跟随着成序列的语句,并被包围在begin
和end
之间,形成一个块;所有声明以这种方式出现在一个块中,并只在这个块中有效。[23]
一个程序是特定的一个块或复合语句,它不包含在另一个语句之中,并且不使用不包含在它之中的其他语句。在1976年的修改版语言报告中,程序只能包含在叫做“环境块”的假定总是存在的一个虚构块之中,除了可以使用在环境块中声明的过程标识符和函数指定符之外,不使用不包含在它之中的其他语句。
量(quantity)被区分出如下种类:简单变量、数组、标签、switch
(switch
列表由标签组成[24])和过程。声明负担定义在程序中用到的量的特定属性,并给它们关联上标识符。声明包括:类型声明、数组声明、switch
声明和过程声明。量的作用域是语句和表达式的集合,在其中关联于这个量的标识符的声明是有效的。所有的块,自动地介入名称目录(nomenclature)的一个新的层级:
- 在这个块内出现的标识符,可以通过合适的声明,而被指定为局部于所论及的这个块。这个标识符在这个块里面的所表示的实体,不存在于它的外面。这个标识符在这个块外面的所表示的任何实体,在这个块里面是不可见的。
- 除了表示标签的标识符之外,一个标识符,如果出现在一个块中,而且并非声明于这个块中,则它非局部于这个块[25],就是说它在这个块里面和在紧邻其外的层级中所表示的是同一个实体。
- 因为块中的语句自身可以是一个块,局部和非局部于一个块的概念,必须递归地去理解,就是说非局部于一个块
A
的一个标识符,可是亦可否地,非局部于A
是其中语句的块B
。
这动态的蕴涵了:在通过begin
进入一个块的时候,所有为这个块声明的标识符,假定了这个给定声明的本性所蕴涵的意义(significance);如果这些标识符已经被外面的其他声明所定义,这时它们被给予新的意义;在另一方面,并非为这个块声明的标识符,保持它们旧有的含义。在通过end
或go to
语句退出一个块的时候,为这个块声明的标识符失去它们的局部意义。声明可以标记上额外的声明符own
,其效果为:在重新进入一个块的时候,自有的这些量的值将不变更而仍是上次退出时的值,然而声明的没有标记上own
的变量的值是未定义的。
在描述算法处理的程序中,主要构成者是算术表达式、布尔表达式,和得到语句标签的指定(designational)表达式。除了特定的分界符之外,表达式的构成者包括:逻辑值、数值、变量、函数指定式,和基本的算术、关系、逻辑运算符(operator)及顺序运算符。用以形成语句的关键字,在ALGOL 60中被归入顺序运算符和分隔符之中。
变量是对一个单一值的指名(designation)。下标(subscripted)变量指定(designate)多维数组的元素的值,这里将数组元素称为“组成元件”(component)[26]。函数指定式(designator)定义单一的数量值或逻辑值,它们是将给定的由过程声明定义的规则集合,应用于固定的实际参数集合的结果。
现在通常将变量定义为抽象的存储位置,它含有了被称为一个值的某种已知或未知的信息量,并且配对了关联的符号名字。在ALGOL 60中,某些语法单位,比如表达式及其构成者和数组标识符,被称为拥有值。各种“类型”即integer
、real
和Boolean
,指称(denote)值的基本的属性。
在左圆括号和匹配的右圆括号之间的表达式(parenthesized expression)自行求值,而这个值被用于后续的计算之中;因此通过适当的圆括号放置,总是可以在表达式之内安排出想要的运算次序。
ALGOL 60将语句分为三类:无条件语句、条件语句和for
语句。赋值语句担负将表达式的值,指派(assign)到一个或多个变量,或者在定义函数指定式的值的过程主体中指派到过程标识符;在作为指派目标的下标变量中出现的任何下标表达式,先于得出所指派之值的表达式,而按从左至右顺序求值。空无的虚设(dummy)语句不执行任何运算。过程语句负担实施调用一个过程主体的执行。
控制流程语句包括:go to
跳转语句、条件语句、for
循环语句。go to
语句结合了无条件go to
和计算go to
二者,goto
语句不能从块外跳转到块内的标签,但可以跳转进入复合语句。条件语句包括if
语句(即if <布尔表达式> then <无条件语句>
),和if
语句经由关键字else
与随后的语句联合在一起的形式(即<if语句> else <语句>
)。ALGOL 60在if
语句和for
语句中介入了子句概念,算术表达式、布尔表达式和指定表达式,可以是条件表达式(即if ~ then ~ else ~
)[27]。
由于变量和函数指定式二者的语法定义都包含表达式,表达式及其构成者的定义必须是递归的。由于成序列的语句,可以被组织成复合语句和块,语句的定义必需是递归的。ALGOL 60采用了左递归的巴科斯范式(BNF)。
在ALGOL 60中,过程声明担负定义过程标识符所关联的过程,其主要构成者是过程主体,它是一个语句或一段代码。过程主体总是表现得如同一个块,不管它是否有着块的形式;故而标记了在过程主体内语句或者主体自身的标签的作用域,永远不能延伸超出这个过程主体[28]。
过程主体关联着一个头部,它规定了出现在过程主体内的代表形式参数的特定标识符。过程的参数传递有两种求值策略:传名调用和传值调用。在过程声明中,通过对形式参数名字前导value
来指定传值调用,缺省为传名调用。在过程主体是用ALGOL语言写的语句的情况下,过程语句执行它的效果,等价于在程序上进行下列操作的效果:
- 声明为传值调用的形式参数,都要被赋值即指派上对应的实际参数的值,这些指派被认为是在进入过程主体之前显式进行的。其效果如同创建了包围着这个过程主体的一个额外的块,在其中所做的这些指派针对的是局部于这个虚构块的变量,它们具有在相应规定中给出的类型。作为结论,传值调用的变量,被认为非局部于过程主体,但是局部于这个虚构块。
- 声明为传名调用的形式参数,在整个过程主体内,要被替代为对应的实际参数,并且只要语法上可能就对这些实际参数包围上圆括号。在通过这个名字替代处理而插入的标识符,和已经存在于过程主体之内的其他标识符,二者之间的可能冲突,将凭借对涉及到的(这个过程主体的)形式标识符或局部标识符的适合的系统性变更来避免[29]。
- 最终经过上述修改后的过程主体,被插入到过程语句的位置上并被执行。如果调用这个过程的位置,处在这个过程主体的任何非局部量[30]的作用域的外面,在通过这个过程主体替代处理而插入的标识符,和在这个过程语句或函数指定式所在位置上其声明有效的标识符,二者之间的可能冲突,将通过(在发起调用的这个层级上)对后者标识符的适合的系统性变更来避免[31]。
对于定义函数指定式的值的过程声明,在其过程主体中,必须出现具有过程标识符在其左侧部分中的一个或多个赋值语句,其中至少有一个必须被执行;并且这个过程标识符所关联的类型,必须通过以类型声明符作为其过程声明的最先符号的样貌来声明。最后那个这样指派的值,被用来继续此函数指定式在其中出现的表达式的求值。在这个过程主体中,这个过程标识符的不在赋值语句左侧部分中的任何出现,指示了这个过程的激活(activation)。
两个传名调用形式参数,其对应的实际参数之间可能存在依赖关系,比如第一个是整数变量i
,而第二个是下标变量A[i]
,从而导致后者形式参数也依赖于前者形式参数,利用传名调用和这种副作用可以实现Jensen设备[32];它典型的用于定义对应于的级数过程Sum(k, l, u, ak)
,它有两个传名调用的形式参数:索引变量k
和通项(general)表达式ak
。
对于交换两个参数的值的swap(x, y)
过程,其过程主体定义为:t:=x; x:=y; y:=t
,这种依赖性副作用会导致可能出现异常行为,由于名字替代机制相当于宏展开(expansion),过程语句swap(i, A[i])
中下标变量A[i]
的下标i
未经求值,对应的过程主体就转换成为:t:=i; i:=A[i]; A[i]:=t
。1964年IFIP工作组2.1制定了《SUBSET ALGOL 60报告》,在这个子集语言中对“完全的名字概念”(full name-concept)增加了一项限制:在名字替代(传名调用)中,实际参数只能是一个标识符或字符串。
在过程的参数列表( … <参数分界符> … )
中,有可选的“) <字母串>: (
”样式的参数分界符[33]。众所周知的传名调用实现采用了thunk[a][35]。Donald Knuth设计了“男人抑或男孩测试”,来区分编译器是否正确的实现了“递归和非局部引用”,这个测试用到了传名调用。
下面是语言报告中过程声明的一个例子[b]:
procedure Absmax(a) Size:(n, m) Result:(y) Subscripts:(i, k);
value n, m; array a; integer n, m, i, k; real y;
comment 矩阵a,其大小为n乘m,其绝对值最大的
元素被传送到y,并且这个元素的下标是i和k。 ;
begin
integer p, q;
y := 0; i := k := 1;
for p := 1 step 1 until n do
for q := 1 step 1 until m do
if abs(a[p, q]) > y then
begin
y := abs(a[p, q]);
i := p; k := q
end
end Absmax
在1976年修改版语言报告的环境块中,定义了输入输出过程:inchar
、outchar
、length
、outstring
、outterminator
、ininteger
、outinteger
、inreal
和outreal
,下面以其中的outinteger
作为演示例子:
procedure outinteger(channel, int);
value channel, int;
integer channel, int;
comment 将表示int的值的那些字符,
加上尾随的终结符传递到channel。 ;
begin
procedure digits(int);
value int; integer int;
begin
integer j;
comment 使用递归从右至左求值数字,
但从左至右打印它们。 ;
j := int ÷ 10;
int := int - 10 × j;
if j ≠ 0 then
digits(j);
outchar(channel, ‘0123456789’, int + 1)
end digits;
if int < 0 then
begin
outchar(channel, ‘-’, 1);
int := -int
end;
digits(int); outterminator(channel)
end outinteger
这里调用到的outchar(channel, str, int)
,将在字符串str
中对应整数int
的值的那个字符,传递到通道channel
;outterminator(channel)
,用于输出在数值之后的终结符(即空格、换行或分号)。此外,IFIP工作组2.1在1964年曾制定《ALGOL 60输入输出过程报告》,其中定义了insymbol
、outsymbol
、length
、inreal
、outreal
、inarray
和outarray
,这里的多维数组采用了横行为主(Row major)次序[36]。
A thunk is an early limited type of closure. The environment passed for the thunk is that of the expression, not that of the called routine.[34]
这里采用了1976年修改版语言报告中的代码,对原版进行的修改是增加了i := k := 1;
,针对的是数组的全部元素都是0
的特殊情况。
Backus, J. W.; Bauer, F. L.; Green, J.; Katz, C.; McCarthy, J.; Perlis, A. J.; Rutishauser, H.; Samelson, K.; Vauquois, B.; Wegstein, J. H.; van Wijngaarden, A.; Woodger, M. Naur, Peter , 编. Report on the Algorithmic Language ALGOL 60. Communications of the ACM (Copenhagen). May 1960, 3 (5): 299–314. ISSN 0001-0782. S2CID 278290. doi:10.1145/367236.367262.
Stephen J. Garland, Anthony W. Knapp, Thomas E. Kurtz, A Manual for SCALP, being a Self Contained ALgol Processor for the General Precision LGP-30 (PDF), CCM-7A, Computation Center, Dartmouth College, Hanover, NH, January 1, 1964 [2023-07-14], (原始内容存档 (PDF)于2023-05-29)
Claude Pair. Description d'un compilateur ALGOL. European Région 1620 Users Group (IBM). 27 April 1965.
Kruseman Aretz, F.E.J. An ALGOL 60 compiler in ALGOL 60. Mathematical Centre Tracts. Amsterdam: Mathematisch Centrum. 1973.
NASE. [2022-05-02]. (原始内容存档于2022-03-30).
J. W. Backus, F. L. Bauer, J. Green, C. Katz, J. McCarthy, P. Naur, A. J. Perlis, H. Rutishauser, K. Samelson, B. Vauquois, J. H. Wegstein, A. van Wijngaarden, M. Woodger. Peter Naur , 编. Revised Report on the Algorithmic Language ALGOL 60. Communications of the ACM, Volume 6, Number 1, pages 1-17. January 1963 [2023-02-20]. (原始内容存档于2023-02-20). A sequence of statements may be enclosed between the statement brackets begin
and end
to form a compound statement. ……
A sequence of declarations followed by a sequence of statements and enclosed between begin
and end
constitutes a block. Every declaration appears in a block in this way and is valid only for that block.
Peter Naur; et al. Revised Report on the Algorithmic Language Algol 60. [2022-04-14]. (原始内容存档于2007-06-25). A switch declaration defines the set of values of the corresponding switch designators. These values are given one by one as the values of the designational expressions entered in the switch list. With each of these designational expressions there is associated a positive integer, 1, 2, ..., obtained by counting the items in the list from left to right. The value of the switch designator corresponding to a given value of the subscript expression ( …… ) is the value of the designational expression in the switch list having this given value as its associated integer.
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22).“25.4.3.
«
|
begin
switch wernik := ariea, aeryl, m17, larix;
goto wernik[k];
arica: ; comment this for k = 1 ;
⋮
goto common;
aeryl: ; comment this for k = 2 ;
⋮
goto common;
m17: ; comment this for k = 3 ;
⋮
goto common;
larix: ; comment this for k = 4 ;
⋮
common:
end
|
».
|
Here, by virtue of switch wernik
, the computation follows one of four possible branches of the program depending on the current value of k
. Afterwards the common course of the calculation (i.e. the statements which would follow «end
») is taken up again.”
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22).“42.5.1. The operands of a block B
are defined as those quantities existing outside the block which are involved in the execution of the block. Obviously the quantities global to B
are operands of B
provided they are actually used inside B
. ……
42.5.2. Hidden operands. A block may have hidden operands: Indeed, if a procedure P
is operand of block B
, …… the global parameters of P
are also involved in the execution of B
, hence operands of B
. We call these hidden operands of B
because they cannot be found by inspection of block B
but only by inspection of the declaration for procedure P
, which is given somewhere outside B
. ……
42.5.3. The operands of a block B
fall into the following four categories:
⒜ Arguments …… ⒝ Results …… ⒞ Transients: …… which have properties of both arguments and results …… ⒟ Exits: Labels referring to destinations located outside B
and switches which are declared outside B
. ……
52.3.1. Considering a mathematical expression, e.g. an integral
it is seen that the variables x
and y
serve entirely different purposes: y
is a variable upon which the value of I
depends; in mathematical logic this is called a free variable (of the expression). This latter term indicates that one is free to substitute a value, e.g. 2.75
, for y
, where-upon one obtains the result
The variable x
, on the other hand, is only an auxiliary object for describing the operation to be performed by the expression. It is called a bound variable since it is not accessible from outside the expression. ……
52.3.2. Comparing these examples with an ALGOL procedure and the terminology used ……, it becomes obvious that the free variables correspond to what are called the operands of a procedure, while the bound variables correspond to the internal quantities.”
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22).“9.1.2. An array is a set of elements, called the components of the array, everyone of which behaves like a simple variable. The components of an array are distinguished by a set of p
integers (subscripts)i1, i2, ……, ip
, where p
is called the dimension of the array. If we interpret the subscripts as coordinates in a p
-dimensional space, then the entire array corresponds to the total of all unit-gridpoints in a p
-dimensional hyperbox
lk ≤ ik ≤ uk
(k = 1, 2, ……, p
),
whose boundaries (i.e. the array bounds l1, l2 ……, lp, u1, u2, ……, up
) are given in the corresponding array declaration ……. ……
14.3.2. A subscripted variable «I[E1, E2, ……, Ep]
», if encountered in an expression, represents also a single value defined as follows: Evaluate the subscript expressions E1, E2, ……, Ep
; if their values are i1, i2, ……, ip
, then the subscripted variable represents the value that has most recently been assigned to the i1, i2, ……, ip
-component of the array I
.”
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22).“
19.7.1. «x + (if t > t1 then 1 else -1) / x
».
We recall that a conditional arithmetic expression cannot be used directly as a primary in a larger expression, but must for this purpose be enclosed in parentheses (the sequence «+ if
» is always illegal).
19.7.2. Selection of a component of an array with safeguards against exceeding the array bounds:
«a[if k > n then n else if k < 1 then 1 else k]
».
19.7.3. Where conditional expressions are intended as comparands of a relation or as alternatives of a conditional expression, they must again be enclosed in parentheses:
«if (if u then x else y) > 0 then (if z = 0 then x + y else x - y) else x × y
».”
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22).“44.2. …… The operands of a procedure, i.e. the quantities involved in its execution, are essentially the operands of the fictitious block which - if S
stands for the procedure body - is defined as the construction
«begin real æ; S end
»1, 2.
……
1 The declaration of the fictitious variable æ
serves solely to make this piece of program a block.
2 In case S
is already an unlabelled block, this artificial construction is unneeded and we could take S
instead. ……”
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22).“45.2.3. Name conflicts. …… Indeed, without this rule, the effect of a call «x(z)
» of a
«
|
procedure x(y);
real y;
begin real z; z := 2 × y; y := y / z end
|
»
|
would erroneously be interpreted as
«begin real z; z := 2 × y; z := z / z end
»,
which certainly was not the intention of the designer of the procedure. With the above amendment, however, the internal z
is changed into zæ
, after which we obtain the equivalence block correctly as
«begin real zæ; zæ := 2 × z; z := z / zæ end
».”
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22).“44.2. Operands of a procedure …… Indeed, execution of a procedure means essentially execution of this fictitious block, …… However, it is one of the most important properties of procedures that their operands - besides being distinguished as arguments, results, transients and exits - fall into three categories, namely
⒜ Those operands of the fictitious block whose identifiers are not quoted in the formal parameter part are called global operands of the procedure. …… ⒝…… formal operands ……⒞…… hidden operands ……
44.3.1. A global parameter - that is, the identifier of a global operand - represents the same quantity inside the procedure body as outside in the environment of the procedure declaration. A global operand is therefore simply the extension of a quantity which exists outside the procedure. As a consequence we have
44.3.2. The environment rule for global parameters:
If the identifier I
is global parameter of a procedure, then a (true or formal) quantity Q
with that identifier must exist in the environment of the procedure declaration, and it is this Q
which in a call of the procedure is meant by the identifier I
.
According to this rule, a global parameter acts like a thread which links the procedure declaration permanently to its environment; indeed, a procedure which has global parameters is only fully defined if it is embedded into an ALGOL program in which the global operands are properly declared.”
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22).“45.2.4. Suppressed global operands. …… Consider, for instance
«
|
begin
integer t;
procedure common(x); real x; t := x;
z: begin
real t;
common(t);
end z
end
|
».
|
Here the integer type variable t
is suppressed in block z
, and therefore the actual parameter of the call «common(t)
» refers to the real type
variable which is local to block z
. …… the t
occurring as global parameter of procedure common
refers to the
suppressed quantity t
. The above rule makes this evident by requiring that the name of the real type variable t
be changed throughout block z
into tee before the substitution rule is applied:
«
|
begin
integer t;
procedure common(x); real x; t := x;
z: begin
real tæ;
common(tæ);
end z
end
|
».
|
Now the substitution rule yields the equivalent block for the call «common(t)
» correctly as (æ
denoting again the hypothetical variable necessary to make this piece of program a block)
«begin real æ; t := tæ end
».
Accordingly, this call accomplishes something which would seem impossible, namely changing the value of a suppressed variable.
45.2.5. ……Consequently no name changes apply where the identifier of a global operand not suppressed at the location of a procedure call coincides with the identifier of an actual operand.”
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22). In view of its ad-hoc character it seems doubtful that the Jensen device (and to some extent even the full name-concept) is the last word in programming language design. Indeed, the dependence of the components of an array upon its subscripts (and likewise the dependence of a function upon its arguments) is more appropriately described by means of CHURCH'S lambda notation rather than through the bound variables of a computing process. Accordingly, we conclude with a sideview to a possibility for introducing this notation in a future ALGOL, but in doing so we strictly adhere to a SUBSET like language-concept, i.e. one in which quantities rather than names play the fundamental role.
Peter Naur; et al. Revised Report on the Algorithmic Language Algol 60. [2022-04-14]. (原始内容存档于2007-06-25). Parameter delimiters. All parameter delimiters are understood to be equivalent. No correspondence between the parameter delimiters used in a procedure statement and those used in the procedure heading is expected beyond their number is the same. Thus the information conveyed by using the elaborate ones is entirely optional.
Ingerman, P. Z. Thunks: a way of compiling procedure statements with some comments on procedure declarations. Communications of the ACM (Association for Computing Machinery (ACM)). 1961-01-01, 4 (1): 55–58. ISSN 0001-0782. S2CID 14646332. doi:10.1145/366062.366084 .
Heinz Rutishauser. Description of ALGOL 60 (PDF). Springer-Verlag New York Inc. 1967 [2023-07-06]. ISBN 978-3-642-86936-5. doi:10.1007/978-3-642-86934-1. (原始内容存档 (PDF)于2022-12-22).“49.3.2. For inarray
and outarray
the order in which the components of the array b
are transferred is defined to be what for matrices (two-dimensional arrays) is usually called "row-wise". More precisely: b[i1, i2, ……, ip]
is transferred before b[j1, j2, ……, jp]
provided we have for some h ≤ p
:
il = jl
for l = 1, 2, ……, h - 1
, but ih < ih
.
Moreover, these procedures always transfer all components of the array appearing as the second actual operand.
As a consequence, a call «outarray(15, p)
», where p
is declared e.g. as «array p[-4:5, 1:50, 0:20]
», is equivalent to
«
|
for j1 := -4 step 1 until 5 do
for j2 := 1 step 1 until 50 do
for j3 := 0 step 1 until 20 do outreal(15, p[j1, j2, j3])
|
».”
|
- Revised Report on the Algorithmic Language Algol 60 (页面存档备份,存于互联网档案馆) by Peter Naur, et al. ALGOL definition
- Modified Report on the Algorithmic Language Algol 60 (页面存档备份,存于互联网档案馆) as modified by R. M. De Morgan, I. D. Hill and B. A. Wichmann under the authority of IFIP Working Group 2.1.
- A BNF syntax summary (页面存档备份,存于互联网档案馆) of ALGOL 60
- "The Emperor's Old Clothes" – Hoare's 1980 ACM Turing Award speech, which discusses ALGOL history and his involvement
- MARST (页面存档备份,存于互联网档案馆), a free Algol-to-C translator
- An Implementation of ALGOL 60 for the FP6000 (页面存档备份,存于互联网档案馆) Discussion of some implementation issues.
- Naur, Peter. The European Side of the Last Phase of the Development of ALGOL 60. ACM SIGPLAN Notices. August 1978, 13 (8): 15–44. S2CID 15552479. doi:10.1145/960118.808370.
- Edinburgh University wrote compilers for Algol60 (later updated for Algol60M) based on their Atlas Autocode compilers initially bootstrapped from the Atlas to the KDF-9. The Edinburgh compilers generated code for the ICL1900, the ICL4/75 (an IBM360 clone), and the ICL2900. Here is the BNF for Algol60 互联网档案馆的存档,存档日期2020-05-15. and the ICL2900 compiler source 互联网档案馆的存档,存档日期2020-05-15., library documentation 互联网档案馆的存档,存档日期2020-05-15., and a considerable test suite 互联网档案馆的存档,存档日期2020-05-15. including Brian Wichmann's tests. 互联网档案馆的存档,存档日期2020-05-15. Also there is a rather superficial Algol60 to Atlas Autocode source-level translator 互联网档案馆的存档,存档日期2020-05-15..
- Eric S. Raymond's Retrocomputing Museum (页面存档备份,存于互联网档案馆), among others a link to the NASE Algol-60 interpreter written in C.
- The NASE interpreter (页面存档备份,存于互联网档案馆)
- Stories of the B5000 and People Who Were There: a dedicated ALGOL computer [1] (页面存档备份,存于互联网档案馆), [2] (页面存档备份,存于互联网档案馆)
- Hermann Bottenbruch. Structure and Use of ALGOL 60. 1961 [2021-04-10]. doi:10.2172/4020495. (原始内容存档于2021-04-24).
- NUMAL (页面存档备份,存于互联网档案馆) A Library of Numerical Procedures in ALGOL 60 developed at The Stichting Centrum Wiskunde & Informatica (legal successor of Stichting Mathematisch Centrum) legal owner.
- Algol 60 resources: translators, documentation, programs (页面存档备份,存于互联网档案馆)
- Algol-60 (页面存档备份,存于互联网档案馆) included in Racket.