A tool that provides autocompletion for various sql dialects.
In order to parse any language, you need a lexer (tokenizer) and a parser (ast builder)
Resources to research:
- The ANTLR Mega Tutorial: https://tomassetti.me/antlr-mega-tutorial/
- Code Completion with ANTLR4-c3: https://tomassetti.me/code-completion-with-antlr4-c3/
- ANTLR4 grammar and lexer for different
{dialect}
are defined insrc/autocomplete/databases/{dialect}/grammar
directories. - Actual parsers and lexers for different
{dialect}
are generated intosrc/autocomplete/databases/{dialect}/generated
directories. - Dialect specific code, which is required for the core logic to work, is put in
src/autocomplete/databases/{dialect}/{dialect}-autocomplete.ts
files. - The core logic is inside
src/autocomplete/shared/autocomplete.ts
file. It uses dialect specific helpers to generate autocomplete suggestions based on current cursor position.
- Add new dialect files to src/autocomplete/databases/{$dialect}
- Add new dialect to the
typesVersions
,exports
variables in package.json - Add new dialect to the
databases
array in build.js
npm run generate
- Generate parsers and lexers (Java has to be installed on your system for this to work)npm run test
- Run tests