lava-oushudb-dt-sql-parser/test/parser/hive/suggestion/suggestionWithEntity.test.ts

311 lines
13 KiB
TypeScript
Raw Normal View History

feat: collect entity (#265) * feat: add text and word utils * feat: add entity collector class * refactor: rename SyntaxContextType to EntityContextType * refactor: improve EntityCollector * feat: improve mysql parser grammar * feat: add mysql entity collector * test: mysql entity collector tests * feat: remove useless method * feat: improve spark grammar file * feat: add spark entity collector * test: spark entity collector unit tests * feat: remove useless code * feat: add queryStatement label * feat: add crateDatabaseStmt * feat: add trino entity collector * feat: rename trinosql to trino * test: trino collect entity unit tests * test: fix spark test * feat(impala): support impale entity collector (#256) * Feat/collect entity hive (#263) * feat(hive): support hive collect entity * feat(hive): update tableAllColumns * feat: replace antlr4ts with antlr4ng * feat(pgsql): pgsql collect entity (#268) * feat(pgsql): pgsql collect entity * feat(pgsql): optimize some name --------- Co-authored-by: zhaoge <> * feat: get word text by token.text * feat: supprt collect db/function and add splitListener (#270) * feat: supprt collect db/function and add splitListner * feat: remove SplitListener interface in baseParser to use SplitListener in root * fix(mysql): fix show create xxx not celloct as createXXXEntity type * test: fix pgsql unit tests * Feat/error recover predicate (#274) * feat: optimize pgsql grammar * feat: add sql parser base * feat: apply SQLParserBase * feat: add geAllEntities method * test: test collect table when missing column * feat: compose collect and suggestion (#276) * feat: mark stmt which contain caret * test: correct name of getAllEntities * test: remove misscolumn unit tests * test: add suggestionWithEntity tests * feat: flink collect entity (#277) * feat: improve flink sql parser * feat: support flink entity collector * test: flink entity collect unit test * feat: move combine entities to parent class --------- Co-authored-by: 霜序 <976060700@qq.com> Co-authored-by: XCynthia <942884029@qq.com>
2024-03-26 14:28:27 +08:00
import fs from 'fs';
import path from 'path';
import { HiveSQL } from 'src/parser/hive';
import { CaretPosition, EntityContextType } from 'src/parser/common/types';
feat: collect entity (#265) * feat: add text and word utils * feat: add entity collector class * refactor: rename SyntaxContextType to EntityContextType * refactor: improve EntityCollector * feat: improve mysql parser grammar * feat: add mysql entity collector * test: mysql entity collector tests * feat: remove useless method * feat: improve spark grammar file * feat: add spark entity collector * test: spark entity collector unit tests * feat: remove useless code * feat: add queryStatement label * feat: add crateDatabaseStmt * feat: add trino entity collector * feat: rename trinosql to trino * test: trino collect entity unit tests * test: fix spark test * feat(impala): support impale entity collector (#256) * Feat/collect entity hive (#263) * feat(hive): support hive collect entity * feat(hive): update tableAllColumns * feat: replace antlr4ts with antlr4ng * feat(pgsql): pgsql collect entity (#268) * feat(pgsql): pgsql collect entity * feat(pgsql): optimize some name --------- Co-authored-by: zhaoge <> * feat: get word text by token.text * feat: supprt collect db/function and add splitListener (#270) * feat: supprt collect db/function and add splitListner * feat: remove SplitListener interface in baseParser to use SplitListener in root * fix(mysql): fix show create xxx not celloct as createXXXEntity type * test: fix pgsql unit tests * Feat/error recover predicate (#274) * feat: optimize pgsql grammar * feat: add sql parser base * feat: apply SQLParserBase * feat: add geAllEntities method * test: test collect table when missing column * feat: compose collect and suggestion (#276) * feat: mark stmt which contain caret * test: correct name of getAllEntities * test: remove misscolumn unit tests * test: add suggestionWithEntity tests * feat: flink collect entity (#277) * feat: improve flink sql parser * feat: support flink entity collector * test: flink entity collect unit test * feat: move combine entities to parent class --------- Co-authored-by: 霜序 <976060700@qq.com> Co-authored-by: XCynthia <942884029@qq.com>
2024-03-26 14:28:27 +08:00
import { commentOtherLine } from 'test/helper';
const syntaxSql = fs.readFileSync(
path.join(__dirname, 'fixtures', 'suggestionWithEntity.sql'),
'utf-8'
);
describe('Hive SQL Syntax Suggestion with collect entity', () => {
const hive = new HiveSQL();
test('Validate Syntax SQL', () => {
expect(hive.validate(syntaxSql).length).not.toBe(0);
});
test('select with no columns', () => {
const pos: CaretPosition = {
lineNumber: 1,
column: 8,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(1);
expect(entities[0].text).toBe('tb1');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.isContainCaret).toBeTruthy();
});
test('select with columns with columns and trailing comma', () => {
const pos: CaretPosition = {
lineNumber: 3,
column: 20,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(1);
expect(entities[0].text).toBe('tb');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.isContainCaret).toBeTruthy();
});
test('from table select with no column', () => {
const pos: CaretPosition = {
lineNumber: 5,
column: 26,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(1);
expect(entities[0].text).toBe('table_name_1');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.rootStmt.isContainCaret).toBeTruthy();
});
test('from table select with with columns and trailing comma', () => {
const pos: CaretPosition = {
lineNumber: 7,
column: 38,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(1);
expect(entities[0].text).toBe('table_name_1');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.rootStmt.isContainCaret).toBeTruthy();
});
test('from joined table select with no column', () => {
const pos: CaretPosition = {
lineNumber: 9,
column: 71,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(2);
expect(entities[0].text).toBe('a');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.isContainCaret).toBeFalsy();
expect(entities[0].belongStmt.rootStmt.isContainCaret).toBeTruthy();
expect(entities[1].text).toBe('b');
expect(entities[1].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[1].belongStmt.isContainCaret).toBeFalsy();
expect(entities[1].belongStmt.rootStmt.isContainCaret).toBeTruthy();
});
test('from joined table select with columns and trailing comma', () => {
const pos: CaretPosition = {
lineNumber: 11,
column: 76,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(2);
expect(entities[0].text).toBe('a');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.isContainCaret).toBeFalsy();
expect(entities[0].belongStmt.rootStmt.isContainCaret).toBeTruthy();
expect(entities[1].text).toBe('b');
expect(entities[1].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[1].belongStmt.isContainCaret).toBeFalsy();
expect(entities[1].belongStmt.rootStmt.isContainCaret).toBeTruthy();
});
test('from table insert into table select no columns', () => {
const pos: CaretPosition = {
lineNumber: 13,
column: 100,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(2);
expect(entities[0].text).toBe('page_view_stg');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.isContainCaret).toBeFalsy();
expect(entities[0].belongStmt.rootStmt.isContainCaret).toBeTruthy();
expect(entities[1].text).toBe('page_view');
expect(entities[1].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[1].belongStmt.isContainCaret).toBeFalsy();
expect(entities[1].belongStmt.rootStmt.isContainCaret).toBeTruthy();
});
test('from table insert into table select with column and trailing comma', () => {
const pos: CaretPosition = {
lineNumber: 15,
column: 104,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(2);
expect(entities[0].text).toBe('page_view_stg');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.isContainCaret).toBeFalsy();
expect(entities[0].belongStmt.rootStmt.isContainCaret).toBeTruthy();
expect(entities[1].text).toBe('page_view');
expect(entities[1].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[1].belongStmt.isContainCaret).toBeFalsy();
expect(entities[1].belongStmt.rootStmt.isContainCaret).toBeTruthy();
});
test('insert into from nested query with no column', () => {
const pos: CaretPosition = {
lineNumber: 17,
column: 98,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(2);
expect(entities[0].text).toBe('insert_tb');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.isContainCaret).toBeTruthy();
expect(entities[1].text).toBe('inside_tb');
expect(entities[1].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[1].belongStmt.isContainCaret).toBeTruthy();
});
test('insert into from nested query with columns and trailing comma', () => {
const pos: CaretPosition = {
lineNumber: 19,
column: 102,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(2);
expect(entities[0].text).toBe('insert_tb');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[0].belongStmt.isContainCaret).toBeTruthy();
expect(entities[1].text).toBe('inside_tb');
expect(entities[1].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[1].belongStmt.isContainCaret).toBeTruthy();
});
test('create table as select with no column', () => {
const pos: CaretPosition = {
lineNumber: 21,
column: 52,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(2);
expect(entities[0].text).toBe('derived_table');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE_CREATE);
expect(entities[0].belongStmt.isContainCaret).toBeTruthy();
expect(entities[1].text).toBe('origin_table');
expect(entities[1].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[1].belongStmt.isContainCaret).toBeTruthy();
});
test('create table as select with columns and trailing comma', () => {
const pos: CaretPosition = {
lineNumber: 23,
column: 56,
};
const sql = commentOtherLine(syntaxSql, pos.lineNumber);
const syntaxes = hive.getSuggestionAtCaretPosition(sql, pos)?.syntax;
const suggestion = syntaxes?.find(
(syn) => syn.syntaxContextType === EntityContextType.COLUMN
);
expect(suggestion).not.toBeUndefined();
expect(suggestion?.wordRanges.map((token) => token.text)).toEqual([]);
const entities = hive.getAllEntities(sql, pos);
expect(entities.length).toBe(2);
expect(entities[0].text).toBe('derived_table');
expect(entities[0].entityContextType).toBe(EntityContextType.TABLE_CREATE);
expect(entities[0].belongStmt.isContainCaret).toBeTruthy();
expect(entities[1].text).toBe('origin_table');
expect(entities[1].entityContextType).toBe(EntityContextType.TABLE);
expect(entities[1].belongStmt.isContainCaret).toBeTruthy();
});
});