JQL Autocomplete
This package allows consumers to retrieve autocomplete suggestions for a given JQL query. JQL
Autocomplete leverages the antlr4-c3 library to provide
code completion candidates using the ANTLR4 generated @atlassianlabs/jql-parser
. Refer to
antlr4-c3 for a more in-depth introduction into code
completion with ANTLR4 grammars.
Usage
For a simple use case you'll want to construct a new JQLAutocomplete
object for a given source
query.
import { JQLAutocomplete } from '@atlassianlabs/jql-autocomplete';
const autocomplete = JQLAutocomplete.fromText("assignee = currentUser()");
You can then specify the (zero-indexed) caret position to retrieve suggestions for, e.g.:
const selectionStart = 11;
const selectionStop = 11;
const suggestions = autocomplete.getJQLSuggestionsForCaretPosition([
selectionStart,
selectionStop,
]);
console.log(suggestions);
Will output the following response:
{
"tokens": {
"matchedText": "",
"replacePosition": [11,11],
"values": ["EMPTY"]
},
"rules": {
"function": {
"matchedText": "",
"replacePosition": [11,11],
"context": {
"field": "assignee",
"operator": "="
}
},
"value": {
"matchedText": "",
"replacePosition": [11,11],
"context": {
"field": "assignee",
"operator": "="
}
}
}
}
This includes a collection of tokens
and rules that are valid for the
given caret position. Rules are essential for deriving more than just keywords from your
autocomplete.
As you can see in the above example a valid query could be produced using:
- An
EMPTY
token, e.g. assignee = EMPTY
- A
function
rule, e.g. assignee = currentUser()
- A
value
rule, e.g. assignee = xxxxxxxxx
Typically, when encountering rules you'll want to enrich this information with Jira data. To do so
you can leverage Jira Cloud REST API's.
Installation
yarn add @atlassianlabs/jql-autocomplete
Documentation
Fine tuning
When constructing an autocomplete object you can specify a collection of ignoredTokens
and
preferredRules
. When not provided, the autocomplete engine specified default arguments which are
suitable for most autocomplete use cases. You can read more about these options in the
antlr4-c3 documentation.
We also extend the antlr4-c3
configuration options with a new argument, delimiterTokens
. By
default JQLAutocomplete
will provide suggestions for the token immediately preceding the caret.
When the caret is positioned at a delimiter token then we'll look for suggestions after the
current token.
For example:
import { JQLLexer, JQLParser } from '@atlassianlabs/jql-parser';
import { JQLAutocomplete } from '@atlassianlabs/jql-autocomplete';
const ignoredTokens = new Set([JQLLexer.NOT_EQUALS]);
const preferredRules = new Set([JQLParser.RULE_jqlField])
const delimiterTokens = new Set([JQLLexer.COMMA])
const autocomplete = JQLAutocomplete.fromText("assignee = currentUser()", ignoredTokens, preferredRules, delimiterTokens);
Support
For developers outside of Atlassian looking for help, or to report issues, please make a post on the community forum.
We will monitor the forums and redirect topics to the appropriate maintainers.
License
Copyright (c) 2021 Atlassian and others. Apache 2.0 licensed, see LICENSE file.