news()

A

A sequence

X

A sequence/integer; can be understood as to(X) when it is an integer, meaning performing X number of computations on sequence A

x_i

An expression, whose results will be field values; the sign ~ used in the parameter references data from X instead of A

F_i

Field name of the result sequence; when omitted, field names are x_i by default, and use the original field names when x_i is #i

@1

Left join, which creates an empty record when X is empty; here it is number 1 instead of letter l

@m

Use parallel processing to increase performance

A

1

[1,2,3,4,5]

2

=A1.news([10,20]; A1.~:a,~:b,a*b)

Parameter X is a sequence and compute members of A and X one by one, that is, use A1’s members as values of field a, members of [10,20] as values of field b and result values of a*b as values of the 3^rd field; as parameter Fi is absent, use parameter xi as field names

3

=A1.news(2; A1.~:a, ~:b,a*b)

Parameter X is an integer, which is equivalent to to(2), and compute each of A1’s member two times

A

1

=demo.query("select top 5 ID,NAME,BIRTHDAY,DEPT,SALARY from EMPLOYEE")

2

=A1.news(~;NAME,age(BIRTHDAY):AGE)

Parameter X is a table sequence and symbol ~ is understood as member of A1; compute A1 and return records consisting of NAME field and AGE field, where AGE field is computed from A1’s BIRTHDAY field

3

=A1.news(3;EID,NAME,SALARY*~:Salary)

Parameter X is integer 3 and compute each member of A1 three times; ~ is understood as the current number of computations

A

1

=demo.query("select top 10 EID,NAME,DEPT,GENDER,SALARY from EMPLOYEE")

2

=A1.group(DEPT).(~.group(GENDER))

Group table sequence A1 by DEPT and then group the result by GENDER

3

=A2.news(~;A2.~.DEPT,A2.~.~.GENDER,A2.~.~.avg(SALARY):AvgSalary)

Get DEPT and GENDER values from A2’s group and calculate average salary of each GENDER in each DEPT

A

1

=demo.query("select  top 10  EID,NAME,BIRTHDAY,GENDER,SALARY from EMPLOYEE")

2

=A1.group(GENDER;~:GenderGroup,~.avg(SALARY):Avg)

Group A1 by GENDER and calculate average salary in each group

3

=A2.news(GenderGroup;EID,#2,GENDER, age(~.BIRTHDAY):age,SALARY+1000:Nsalary,A2.Avg:AvgSalary)

Compute the expression on each group of A2 and generate records to form a new table sequence, during which #2 represents NAME, the 2^nd field of GenderGroup

A

1

=demo.query("select top 5 EID,NAME,BIRTHDAY,GENDER,SALARY from EMPLOYEE")

2

=A1.group(GENDER;~:GenderGroup)

Group A1’s records by GENDER

3

=demo.query("select top 10 EID,NAME,BIRTHDAY,GENDER,SALARY from EMPLOYEE")

4

=A3.group(GENDER;avg(SALARY):avg)

Group A3 by GENDER and calculate average salary in each group

5

=A4.join(GENDER,A2:GENDER,GenderGroup)

Perform association between A2 and A4 according to GENDER field; since A2 does not have records where GENDER is M, the corresponding GenderGroup field values are represented by nulls after table association

6

=A5.news(GenderGroup;EID,NAME,GENDER,age(~.BIRTHDAY):age,avg:AvgSalary)

Compute the expression on record sequence A5 to generate new records, from which those whose GenderGroup is empty are by default discarded

7

=A5.news@1(GenderGroup;EID,NAME,GENDER,age(~.BIRTHDAY):age,avg:AvgSalary)

Use @1 option to create an empty record when record sequence X is empty

ch

Channel

X

Record sequence

x_i

Expression, whose results will be field values; the sign ~ used in the parameter references data from X instead of A. The sign # is used to represent a field with a ordinal number

F_i

Field name in the given channel; will be automatically identified if the parameter is omitted

A

1

=demo.cursor("select * from EMPLOYEE")

2

=demo.query("select * from EMPLOYEE")

3

=channel()

Create a channel

4

=A2.group(GENDER;~:group)

Group records by GENDER

5

=A1.groupx(GENDER;avg(SALARY):avg)

Group records by GENDER

6

=A5.join(GENDER,A4:GENDER,group)

7

=A3.news(group;EID,NAME,GENDER,#3:surname, age(~.BIRTHDAY):age,SALARY+50: salary,avg:AvgSalary)

A6 is one of the groups; compute field values based on each group and concatenate the results into A3’s channel

8

=A3.fetch()

Fetch and store the existing data in the channel

9

=A6.push(A3)

Be ready to push data in A6’s cursor into A3’s channel, but the action needs to wait

10

=A6.fetch()

Data in A6’s cursor is pushed into the channel and operations are performed as the fetch() operation is performed over A6

11

=A3.result()

cs

Cursor

X

Record sequence

x_i

An expression, whose values are uses as the new field values. It is treated as null if omitted; in that case, : F_i can’t be omitted. The sign # is used to represent a field with a ordinal number

F_i

Filed name of the new cs; use the identifiers parsed from expression x_i if it is omitted

A

1

=demo.cursor("select * from EMPLOYEE")

2

=demo.query("select * from EMPLOYEE")

3

=A2.group(GENDER;~:group)

Perform group by GENDER

4

=A1.groupx(GENDER;avg(SALARY):avg)

Perform group by GENDER

5

=A4.join(GENDER,A3:GENDER,group)

6

=A5.news(group;EID,NAME,GENDER,#3:surname, age(~.BIRTHDAY):age,SALARY+50: salary,avg:AvgSalary)

A4 is composed of data groups; based on each group field values are computed and a cursor is generated; and then cursors are concatenated into a new one.

7

=A6.fetch()

8

=MySQL1.query("select * from EMPLOYEE")

9

=A8.group(GENDER;~:group)

Group GENDER field

10

=A4.join(GENDER,A9:GENDER,group)

11

=A10.news@1(group;EID,NAME,GENDER,age(~.BIRTHDAY):age,salary,avg:AvgSalary)

Compute each GENDER value for each group, generate an empty table for the group where GENDER is M and then concatenate result records into a new table sequence

12

=A11.fetch()

T

A composite table

A/cs

A table sequence/cursor/composite table cursor

K

Field names; when there are K… parameters after A/cs, correspond them with theses specified fields

x

Field names or aggregate function count/sum/max/min/avg

C

Column alias

wi

Filtering condition; retrieve the whole set when this parameter is absent; separate multiple conditions by comma(s) and their relationships are AND. Besides regular filtering expressions, you can also use the following five types of syntax in a filtering condition, where K is a non-key field in entity table T:

1．K=w

w is expression T_i.find(K) or T_i.pfind(K), where T_i is a table sequence. When value of w is null or false, the corresponding record in the entity table will be filtered away; when w is expression T_i.find(K) and the to-be-selected fields C,... contain K, T_i’s referencing field will be assigned to K; when w is expression T_i.pfind(K) and the to-be-selected fields C,... contain K, ordinal numbers of K values in T_i will be assigned to K.

2．(K₁=w₁,…K_n=w_n,w)

w_i is expression T_i.find(K) or T_i.pfind(K), where T_i is a table sequence. When value of wi is null or false, the corresponding record in the entity table will be filtered away; when w_i is expression T_i.find(K) and the to-be-selected fields C,... contain K₁,...K_i, T₁,... T_i’s referencing field will be assigned to K₁,...K_i correspondingly; when w is expression T_i.pfind(K) and the to-be-selected fields C,... contain K, ordinal numbers of K values in T_i will be assigned to K. w is an expression that returns a Boolean value; each of the multiple K_i=w_i and w has a relationship of AND; you can reference K_i in w.

3．K:T_i

T_i is a table sequence. Compare K_i value in the entity table with key values of T_i and discard records whose K_i value does not match; when the to-be-selected fields C,... contain K, T_i’s referencing field will be assigned to K.

4．K:T_i:null

Filter away all records that satisfy K: T_i.

A

1

=file("score.ctx").open()

Open a composite table file

2

=demo.query("select * from students").keys(ID)

Return a table sequence with ID as the key

3

=A1.news(A2:ID,STUDENTID,SUBJECT,SCORE;SUBJECT=="Math",SCORE>70)

Join A1’s composite table and A2’s table sequence where the filteing condition is SUBJECT=Math & SCORE>70 and return eligible records in the composite table

4

=A1.news@r(A2:ID,NAME,sum(SCORE):TotalScore)

Sum scores of all subjects for each student and return result:

A

1

=demo.cursor("select STUDENTID,CLASS,SUBJECT,SCORE from SCORES")

Below is content of the cursor:

2

=file("scores.ctx")

3

=A2.create@y(#STUDENTID,#CLASS,SUBJECT,SCORE )

Create a composite table

4

=A3.append@i(A1)

Append cursor A1’s data to the composite table’s base table

5

=demo.query("select  ID,NAME,GENDER,AGE  from STUDENTS")

Return a table sequence

6

=A3.news(A5:ID,STUDENTID,NAME,SUBJECT,SCORE)

Match the composite table’s key values to table sequence’s fields and retrieve corresponding composite table fields and return a table sequence

7

=create(Subject,Num).record(["PE",01,"Math",02,"History",03]).keys(Subject)

Generate a table sequence using Subject as the key

8

=A3.news(A5:ID,NAME,SUBJECT,SCORE;SUBJECT=A7.find(SUBJECT))

Use K=w filtering mode; in this case w is T_i.find(K) and entity table records making SUBJECT=A7.find(SUBJECT) get null or false are discarded; SUBJECT is the selected field, to which table sequence A7’s referencing field is assigned

9

=A3.news(A5:ID,NAME,SUBJECT,SCORE;SUBJECT=A7.pfind(SUBJECT))

Use K=w filtering mode; in this case w is T_i.pfind(K) and entity table records making SUBJECT=A7.pfind(SUBJECT) get null or false are discarded; SUBJECT is the selected field, to which its ordinal numbers in table sequence A7 are assigned

10

=A3.news(A5:ID,NAME,SUBJECT,SCORE;SUBJECT:A7)

Use K:T_i filtering mode; compare the entity table’s SUBJECT values with the table sequence’s key values and discard entity table records that cannot match

11

=A3.news(A5:ID,NAME,SCORE;SUBJECT:A7)

This is a case where K isn’t selected; SUBJECT isn’t the selected field, so only filtering is performed

12

=A3.news(A5:ID,NAME,SUBJECT,SCORE;SUBJECT:A7:null)

Use K:T_i:null filtering mode; compare the entity table’s SUBJECT values with the table sequence’s key values and discard entity table records that can match

13

=create(Score,Level).record([90,"A",80,"B",70,"C",60,"D"]).keys(Score)

Return a table sequene using Score as the key

14

=A3.news(A5:ID,NAME,SUBJECT,SCORE;(SCORE=A13.find(SCORE),SUBJECT=A7.find(SUBJECT),SCORE!=null&&SUBJECT!=null))

Use (K₁=w₁,…K_n=w_n,w) filtring mode; return records that meet all conditions