esker
v0.1.1
Published
Fluent interface query strings
Downloads
3
Readme
esker
Fluent query strings
companions/name.is.Amelia_and_age.is.greater.than.22
Queries
Esker looks at query strings as a sequence of one or more sentences. Sentences are conjuncted using the logical conjunctions _and_
and _or_
.
Example:
year.is.2015_and_style.is.soul
Sentences
Sentences basically follow the structure <field><comparison><value>
.
The <field>
is the name of the property / field to filter against.
The <value>
is the value the field should be compared against
The <comparison>
is a verb than can be mixed with modificators (not
) and optional conjunctions (than
). Supported verbs are:
- equality:
is
,equals
- partial equality, membership:
contains
,has
- greater than:
greater
,larger
,more
- smaller than:
smaller
,less
Note: when using greater / smaller than comparisons, you are free to still use the is
, equals
verbs. E.g. age.is.less.than.10
To make queries more readable, the optional than
conjunctions can be added.
To inverse comparison, the not
modificator can be added.
Here a few examples
// items that have the name 'Amelia'
name.is.Amelia
name.equals.Amelia
// items that do NOT have the name 'Amelia'
name.is.not.Amelia
// items whose name contains 'eli'
name.contains.eli
// items whose age is more than 20
age.is.more.than.20
age.greater.than.20
age.greater.20
// items whose price is less than 10
price.is.less.than.10
price.is.smaller.than.10
price.smaller.10
// items whose price is less than 10 (using not)
price.is.not.more.than.9
Usage
Esker's central (and currently only exposed) method is parse(<query-string>)
. To use Esker, first install it:
npm install esker --save
Import it as a module then use the parse()
method on your query string.
const esker = require('esker')
let result = esker.parse('name.is.Amelia')
The result
will contain an object containing an array of object (one for each sentence passed).
Each sentence is represented as an object with the properties field
, verb
, not
, value
, operatorLeft
, operatorRight
. The above example ('name.is.Amelia'
), results in:
{
'field': 'name',
'verb': 'is',
'not': false,
'value': 'Amelia,
'operatorLeft': 'none',
'operatorRight': 'none'
}
The verb
field may contain one of the following values:
'is'
- equality'has'
- partial equality, membership'>'
- greater than'<'
- less than
The not
field is set to true if the sentence contains the modifier 'not'
.
The operatorLeft
and operatorRight
express and sentence's relation to its neighbours (see Logical Operators). It may contain one of the following values:
'none'
'and'
'or'
Logical Operators
Sentences can be combined using the '_and_'
and '_or_'
operators. Here an example:
color.is.blue_and_price.is.10_or_color.is.red
Note that Esker currently does not support defining any operator precedence by means of brackets.
Logical operators are simply added to the result's sentences
items as operatorLeft
and operatorRight
properties.
How these operators are interpreted depends on the extension used (see section on Extensions)
Extensions
The parse()
method takes a second (optional) attribute, an extension method. This can be any method take takes and array of sentences and returns an object.
The returned object is then added to the result as the value of the output
property.
Currently Esker comes with a MongoDB extension. Here is how you use it:
const esker = require('esker')
let query = 'color.is.blue_and_price.is.10_or_color.is.red'
let result = esker.parse(query, esker.mongo)
// result.output contains an object that you now can use as the filter in a MongoDB find() call
let filter = result.output