1 unstable release
0.1.0 | Mar 7, 2021 |
---|
#34 in #expression-language
Used in ter
38KB
1.5K
SLoC
The Text Expression Language
This is a crate contains a parser, execution engine and specification for a super simple language to write readable and memorizable text processing expressions.
This document should be used as specification for language to clarify the development of this package.
Data Types
This Language holds only two data types; strings and integers. Both are used as literals in arguments of attribute statements.
str
A str
must use double quotes. So "foo"
and "!=$)(j0j0802"
are both valid strings. If a double quote occurs inside the string, it can be
escaped using \
. So strings with escaped double quotes look like this: "\"Quoted Text\""
. This is actually all there is to strings. They cant contain
any other escaped chars (such as new lines etc.)
int
An ìnt
is either a single 0
or 1-9
followed by 0-9
as often as wanted. There are no signed integers.
Queries
Queries indicate the format of a string which gets tested against it. Currently there are 9 Attributes which are specified:
Attribute | Resolve to true if the tested string |
---|---|
starts <str> |
starts with the given string |
ends <str> |
ends with the given string |
contains <str> |
contains a substring equal to the given string |
equals <str> |
exactly equals the given string |
length <int> |
has the given length |
numeric |
contains only numeric chars |
alpha |
contains only alphabetic chars |
alphanumeric |
contains only alphanumeric chars |
special |
contains only special chars |
Logical Operators
Currently there are only two binary logical operations: and
and or
Operator | Precedence | Associativity |
---|---|---|
and |
2 | Right |
or |
1 | Right |
Examples
1 or 2
parses as (1 or 2)
1 and 2
parses as (1 and 2)
1 and 2 or 3
parses as ((1 and 2) or 3)
1 or 2 and 3
parses as (1 or (2 and 3))
1 and 2 and 3
parses as (1 and (2 and 3))
1 or 2 or 3
parses as (1 or (2 or 3))
1 or 2 or 3 and 4 or 5
parses as (1 or (2 or ((3 and 4) or 5)))))
Groups
Currently not implemented. May come in future versions to enable more complex text processing;