A
sequence is an ordered set consisting of some data, which are called members of
the sequence. A sequence is similar to an array in high-level language, but the
data type of its members not have to be the same. The following will explain
its basic computation through creation, access, operators and functions.
Creation
1. Create with Constant
Bracketing
members with "[]" represents sequence constant, e.g.
A5=[15.2,"b",1] /Members of a sequence can be float, string and integer
B5=[A1:C3] /Members are cell area, i.e.
A6=[1,2,3,3] /Members can be repeated
B6=[] / A empty sequence
C6=[[]] /A non-empty sequence
that its member is an empty sequence
Note:
Members of a sequence can be any data types, including atom type, another
sequences, records, etc. A sequence that all its members are integers is called
an integer sequence.
2.Create with functions
to(2,6) / Numbers of the
integer sequence are [2,3,4,5,6]. A integer sequence beginning with 1 can be
expressed with to(6)
"1,a,b,c".array() /Split a character string into a sequence
[1,a,b,c], a reversed joint can be expressed with [1,a,b,c].string()
periods@y("2012-08-10
12:00:00",now(),1) /Create
a integer sequence for a period of time by dividing it into smaller time
periods whose starting and ending points are in adjacent years, sequential value
is ["2012-08-10 12:00:00","2013-01-01 00:00:00","2014-01-01
00:00:00","2014-07-01 10:10:41"]
file("e:/sales.txt").import@t() /Import
records from structured text files to form a sequence, sequential value is:
A
sequence whose members are records is table sequence, which is often used to
make computations of structured data. It’s not the focus in this article, for
more information please refer to “Basic Computation of esProc Table Sequence
and Record Sequence”.
3. Create by computing
Description:
read the text file sales.txt into table sequence A1, fetch column Client to
create sequence A2; group the records according to Client to create sequence
A3.
Code：
A1=file("e:/sales.txt").import@t()
A2=A1.(Client)
A3=A1.group(Client)
Note：
(1) Value of sequence A2：
(2) Value of sequence A3：
We can see that members of sequence A3 are a number of sequences
whose members are records.
Accessing
4. Access members according to serial numbers
A1=[a,b,c,d,e,f,g] /Sequence
A1
A1(2) /Fetch
the second member whose value is string "b", which equals to A1.m(2)
A1([2,3,4]) /Fetch
members from the second to the fourth, whose value is expressed by the sequence
[b,c,d]. Note that [2,3,4] is also a sequence(integer sequence). Intervals can
be used to rewritten the expression as A1(to(2,4)).
A1.m(-1) /Fetch
the last member. Note that m function must be used when fetching members
backwards, the expression cannot be abbreviated to A1(-1).
5. Assignment and
modification
A1(2)=r /Modify the
second member to r, now value of sequence A1 is [a,r,c,d,e,f,g]
A1([2,4])=["r","s"] /Modify the second and the fourth
member. A1=[a,r,c,s,d,e,f,g]
A1.Modify(2,["r","s"]) /Modify in turn from the second member,
expression A1= [a,r,s,d,e,f,g] equals to A1([2,3])=["r","s"]
6. Add members
A1.Insert(0,"r") Add members at the end of the
sequence, A1=[a,b,c,d,e,f,g,r]
A1.Insert(2,["r","s","t"]) /Insert three members consecutively
before the second member, A1=[ a,r,s,t,b,c,d,e,f,g,r]
7. Delete members
A1.Delete(2) /Delete the second member
A1.Delete([2,4]) /Delete the second and the fourth member
Operators
8. Sets computation
Sets computation include ^
intersection, & union, \complement, and |concatenate, etc. For example:
A1=["a","b",1,2,3,4] /Sequence A1
B1=
["d","b",10,12,3,4] /Sequence
A2
A1^B1 /Intersection, return the
sequence made up of members of both thetwo sequences,
value is ["a","b",3,4]
A1\B1 /Complement, a new sequence
created by successively removing from A1 the members of B1, value is ["a",1,2]
A1&B1 /Union, value is["a","b",1,2,3,4,"d",10,12]
A1|B1 /Concatenate, value is ["a","b",1,2,3,4,"d","b",10,12,3,4]
Note: Both union and concatenate
are created by combining members of two sequences in order. Common members only
appear once in union while, in concatenate, all of them will appear.
9. Alignment arithmetic operation
Two sequences of the same length
can make alignment operation according to members and return the sequence. The
operation includes ++ (add), -- (subtract), ** (multiply), // (divide) and %%
(complementation). For example:
A1=[1,2,3,4] /Sequence A1
B1= [10,12,3,4] /Sequence A2
A1++B1 /Counterpoint addition, value is
[11,14,16,18]
10. Boolean operation
Two sequences can compare in
alignment, the result is a Boolean type.
[1,2,3]==[1,2,3] /Comparative result is true
[1,B,3]<=[1,b,4] /Comparative result is true, because B
is less than b
[1,2,3]<[1,3,4] /Result is true, because the second
member of [1,2,3] is "2" , which is smaller than the second member “3” of
[1,3,4]
Note: "in" function is used to
judge the inclusion relation between sequences.
Functions
11. Aggregate function
Functions for sequences include sum, avg, max, variance,etc. For
example:
A1=[2,4,6] /Sequence
A1.sum() /Summation, result is 12
A1.sum(~*~) /Quadratic sum, which equals to 2*2+4*4+6*6,
result is 56. ~ represents each member of a sequence.
12. Loop function
Loop
function can make computation aiming at every member of a sequence, and express
complex loop statement with simple functions, including loop computation,
filter, locate, look up, rank, sort, etc.
A1=[2,4,-6] /Construct sequence A1
A1=(~+1) /Add 1 to every member,
result is [3,5,-5]
A1.select(~>1) /Filter out members that are greater
than 1, result is [2,4]
A1.pselect@a(~>1) /Locate serial numbers of members that are
greater than 1, result is [1,2]
A1.pos([-6,2]) /Look up serial numbers of members
-6 and 2 in A1, result is [3,1]
A1.rank() /Rank of members of the sequence,
result is [2,1,3]
A1.sort() /Sort in ascending order, result is
[-6,2,4]; [2,4,-6].sort(~:-1) is the expression when sorting in descending
order
没有评论:
发表评论