fqs - file queries

fqs is a command line tool for writing queries over files. Query syntax is inspired by SQL. In many ways fqs is similar to one (or combination) of the following: cut, paste, bc, grep, and awk, but fqs provides SQL-like declarative query language.

Here is an example command that prints the values from the second column for each row if a value in the first column is greater than 1000.

fqs "select str(@1) from path/to/file where int(@0) > 1000"

fqs uses ' ' as a delimiter of columns in the given file. (This is equivalent to cut -d' '.)

Examples

This section provides several examples. Let's say that we have the following file (demo.txt).

33 77.0 true text
44 88.98 false more
77 123. true next

The next command prints all the values from the first column of the file. Note that we use @0 as a column reference.

fqs "select int(@0) from demo.txt"

The next command prints values from the first column doubled.

fqs "select int(@0) * 2 from demo.txt"

The next command prints values from the first column only if the third column is true.

fqs "select int(@0) from demo.txt where bool(@2) = true"

The next command reorders columns, changes one of the columns from lower-case letters to upper-case letters and processes only rows in which the second column has value greater than 80.

fqs "select upper(str(@3)), sin(int(@0)) from demo.txt where float(@1) > 80"

The next command shows the number of rows, the sum of all values in the first column, and the max value in the second column.

fqs "select count(1), sum(int(@0)), max(float(@1)) from demo.txt"

Query language

At the moment, fqs supports the select statement. In many ways the select statement is similar to the one you might know from SQL, but there are differences due to the nature of the data. Below is the (approximate) grammar of the language:

Query ::= "select" Columns "from" Path ["where" Condition]
Columns ::= Aggs | Exprs
Aggs ::= AggFunc [,AggFunc]*
AggFunc ::= Id "(" CExpr ")" # see the list of functions later in this document
Exprs ::= CExpr [, CExpr]*
CExpr ::= ScaFunc | AExprs
AExprs ::= MExprs [Aop AExprs]
MExprs ::= MExprs [Mop Operand] | Operand
ScaFunc ::= Id "(" CExpr ")" # see the list of functions later in this document
Path ::= path to a file that contains data to process
Condition ::= WExp
WExp ::= Operand Lop Operand
Operand ::= Cast | Int | Float | Bool | String
Cast ::= Type "(" ColRef ")"
Type ::= "int" | "float" | "bool" | "str"
ColRef ::= "@"Int
Int ::= int constant
Float ::= float constant
Bool ::= "true" | "false"
String ::= "'"string"'"
Aop ::= "+" | "-"
Mop ::= "*" | "/"
Lop ::= "<" | ">" | "<=" | ">=" | "=" | "!="
Id ::= an identifier

Keywords

This section contains the list of keywords.

select, from, limit, where, int, float, str, bool, true, false.

Scalar functions

This section contains the list of scalar functions. All functions in this section report an error if the given argument has an incorrect type.

upper(str)

• Returns a string in which all lower-case characters are converted to their upper-case equivalent. It returns null if the argument is null.

lower(str)

• Returns a string in which all upper-case characters are converted to their lower-case equivalent. It returns null if the argument is null.

length(str)

• Returns the number of characters in the string argument. It returns null if the argument is null.

rev(str)

• Reverses the given argument string. It returns null if the argument is null.

abs(int|float)

• Computes the absolute value of the argument. It returns null if the argument is null.

sign(int|float)

• Returns the sign of the given numerical argument. It returns null if the argument is null.

ceil(int|float)

• Round the given number to an integer greater than or equal to the input number. It returns null if the argument is null.

floor(int|float)

• Returns integer value less than or equal to the given argument. It returns null if the argument is null.

round(int|float)

• Rounds the given argument numeric value to integer. It returns null if the argument is null.

cos(int|float)

• Returns the cosine of the given numeric argument in radians. It returns null if the argument is null.

sin(int|float)

• Returns the sine of the given numeric argument in radians. It returns null if the argument is null.

Aggregate functions

This section contains the list of aggregate functions.

sum(int|float)

• Returns the sum of non-null values.

count(any)

• Returns the number non-null values.

max(int|float)

• Finds the max numerical value. Null values are ignored.

min(int|float)

• Finds the min numerical value. Null values are ignored.

avg(int|float)

• Computes the average value. Null values are ignored.

Contributing

Please check this page.

License

BSD-3-Clause license.

Contact

Feel free to get in touch if you have any comments: Milos Gligoric <[email protected]>.