This module provides you 2 classes to work with Ctx19.
You can read here about Contextual19.
This is just regular parser that takes rules in object representation.
To initialize this class, just call it and take there rules in object representation:
from ctx19.parsers import Contextual19Parser
data = [
{
"if": ...,
"then": ...
},
...
]
parser = Contextual19Parser(data)
Data you passed will be stored in parser.data
.
bool
saveContext
(default isTrue
)
This parameter set transformation resistance of the rules. Suppose, we have this sentence:
[{a:0}, {a:1}, {a:2}]
,
and this rules (inline-written):
#1:if previous a is 0 then a becomes 2
,
#2:if previous a is 1 then a becomes 3
.
After executing rule #1 sentence becomes:
[{a:0}, {a:2}, {a:2}]
Now the rule #2 is not appliable in this contexts. That is what means "transformation nonresistance". To make the rule #2 work,saveContext
must be set toTrue
and Ctx19 module will remember original sentence before applying rules (viadeepcopy
, so some additional memory will be needed).
apply(sentence: list) -> list
This method apply rules, stored inself.data
to the sentence you passed.
sentence
is the list of dictionaries which represents tokens with their properties. It can look like this:Missing properties will be skipped.[ { "tense": "past", "gender": "neutral", "some_other_stuff": "other_value" }, { "tense": "future", "voice": "active", "another_property": "another_value" } ]
save(filepath: str) -> void
This method will simply save the rules stored inself.data
into file you passed asfilepath
.
There's also some methods that should not be used in your code, but they are widely used in Contextual19Parser.apply
method and they can be useful in some cases.
ruleIsAppliable(rule: dict, sentence: list, token: int) -> bool
Check if the givenrule
is appliable to thetoken
th token in thesentence
. It'll check the context.
This is the children of Contextual19Parser
class with the changed initialization method.
To initialize this parser just pass path to file with rules as the f
parameter:
from ctx19.parsers import Contextual19FileParser
parser = Contextual19FileParser(open("path/to/file.ctx19"))
bool
astext
Set this toTrue
to read text fromf
parameter instead of dealing with it as file. Example of use:data = ''' if ... then ... ... ''' parser = Contextual19FileParser(data, astext=True)
Since this is the Contextual19Parser
children, it'll inherit all the parent methods, but one new private was added.
parseFile(void) -> void
This will parse Contextual19 syntax from the file saved inself.file
. You can change the file and call this method again.
From 1.2 version JSON and YAML parsers was moved in order to make ctx19 independent from other packages, but you can still do it yourself. Here's the example:
import json
from ctx19.parsers import Contextual19Parser
ctx = Contextual19Parser(json.load("path/to/file.json"))