duckdb 0.9.2-dev5.0 → 0.9.2-dev7.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/lib/duckdb.d.ts +16 -0
- package/lib/duckdb.js +12 -0
- package/package.json +1 -1
- package/src/database.cpp +30 -1
- package/src/duckdb_node.cpp +29 -8
- package/src/duckdb_node.hpp +2 -0
- package/test/tokenize.test.ts +74 -0
package/lib/duckdb.d.ts
CHANGED
@@ -127,6 +127,20 @@ export type ReplacementScanCallback = (
|
|
127
127
|
table: string
|
128
128
|
) => ReplacementScanResult | null;
|
129
129
|
|
130
|
+
export enum TokenType {
|
131
|
+
IDENTIFIER = 0,
|
132
|
+
NUMERIC_CONSTANT = 1,
|
133
|
+
STRING_CONSTANT = 2,
|
134
|
+
OPERATOR = 3,
|
135
|
+
KEYWORD = 4,
|
136
|
+
COMMENT = 5,
|
137
|
+
}
|
138
|
+
|
139
|
+
export interface ScriptTokens {
|
140
|
+
offsets: number[];
|
141
|
+
types: TokenType[];
|
142
|
+
}
|
143
|
+
|
130
144
|
export class Database {
|
131
145
|
constructor(path: string, accessMode?: number | Record<string,string>, callback?: Callback<any>);
|
132
146
|
constructor(path: string, callback?: Callback<any>);
|
@@ -169,6 +183,8 @@ export class Database {
|
|
169
183
|
registerReplacementScan(
|
170
184
|
replacementScan: ReplacementScanCallback
|
171
185
|
): Promise<void>;
|
186
|
+
|
187
|
+
tokenize(text: string): ScriptTokens;
|
172
188
|
}
|
173
189
|
|
174
190
|
export type GenericTypeInfo = {
|
package/lib/duckdb.js
CHANGED
@@ -64,6 +64,10 @@ var Statement = duckdb.Statement;
|
|
64
64
|
* @class
|
65
65
|
*/
|
66
66
|
var QueryResult = duckdb.QueryResult;
|
67
|
+
/**
|
68
|
+
* Types of tokens return by `tokenize`.
|
69
|
+
*/
|
70
|
+
var TokenType = duckdb.TokenType;
|
67
71
|
|
68
72
|
/**
|
69
73
|
* @method
|
@@ -631,6 +635,14 @@ Database.prototype.unregister_udf = function () {
|
|
631
635
|
|
632
636
|
Database.prototype.registerReplacementScan;
|
633
637
|
|
638
|
+
/**
|
639
|
+
* Return positions and types of tokens in given text
|
640
|
+
* @method
|
641
|
+
* @arg text
|
642
|
+
* @return {ScriptTokens}
|
643
|
+
*/
|
644
|
+
Database.prototype.tokenize;
|
645
|
+
|
634
646
|
/**
|
635
647
|
* Not implemented
|
636
648
|
*/
|
package/package.json
CHANGED
package/src/database.cpp
CHANGED
@@ -1,5 +1,6 @@
|
|
1
1
|
#include "duckdb/parser/expression/constant_expression.hpp"
|
2
2
|
#include "duckdb/parser/expression/function_expression.hpp"
|
3
|
+
#include "duckdb/parser/parser.hpp"
|
3
4
|
#include "duckdb/parser/tableref/table_function_ref.hpp"
|
4
5
|
#include "duckdb/storage/buffer_manager.hpp"
|
5
6
|
#include "duckdb_node.hpp"
|
@@ -18,7 +19,8 @@ Napi::FunctionReference Database::Init(Napi::Env env, Napi::Object exports) {
|
|
18
19
|
{InstanceMethod("close_internal", &Database::Close), InstanceMethod("wait", &Database::Wait),
|
19
20
|
InstanceMethod("serialize", &Database::Serialize), InstanceMethod("parallelize", &Database::Parallelize),
|
20
21
|
InstanceMethod("connect", &Database::Connect), InstanceMethod("interrupt", &Database::Interrupt),
|
21
|
-
InstanceMethod("registerReplacementScan", &Database::RegisterReplacementScan)
|
22
|
+
InstanceMethod("registerReplacementScan", &Database::RegisterReplacementScan),
|
23
|
+
InstanceMethod("tokenize", &Database::Tokenize)});
|
22
24
|
|
23
25
|
exports.Set("Database", t);
|
24
26
|
|
@@ -364,4 +366,31 @@ Napi::Value Database::RegisterReplacementScan(const Napi::CallbackInfo &info) {
|
|
364
366
|
return deferred.Promise();
|
365
367
|
}
|
366
368
|
|
369
|
+
Napi::Value Database::Tokenize(const Napi::CallbackInfo &info) {
|
370
|
+
auto env = info.Env();
|
371
|
+
|
372
|
+
if (info.Length() < 1) {
|
373
|
+
throw Napi::TypeError::New(env, "Text argument expected");
|
374
|
+
}
|
375
|
+
|
376
|
+
std::string text = info[0].As<Napi::String>();
|
377
|
+
|
378
|
+
auto tokens = duckdb::Parser::Tokenize(text);
|
379
|
+
auto numTokens = tokens.size();
|
380
|
+
|
381
|
+
auto offsets = Napi::Array::New(env, numTokens);
|
382
|
+
auto types = Napi::Array::New(env, numTokens);
|
383
|
+
|
384
|
+
for (size_t i = 0; i < numTokens; i++) {
|
385
|
+
auto token = tokens[i];
|
386
|
+
offsets.Set(i, token.start);
|
387
|
+
types.Set(i, (uint8_t)token.type);
|
388
|
+
}
|
389
|
+
|
390
|
+
auto result = Napi::Object::New(env);
|
391
|
+
result.Set("offsets", offsets);
|
392
|
+
result.Set("types", types);
|
393
|
+
return result;
|
394
|
+
}
|
395
|
+
|
367
396
|
} // namespace node_duckdb
|
package/src/duckdb_node.cpp
CHANGED
@@ -12,15 +12,36 @@ NodeDuckDB::NodeDuckDB(Napi::Env env, Napi::Object exports) {
|
|
12
12
|
statement_constructor = node_duckdb::Statement::Init(env, exports);
|
13
13
|
query_result_constructor = node_duckdb::QueryResult::Init(env, exports);
|
14
14
|
|
15
|
-
|
16
|
-
|
15
|
+
auto token_type_enum = Napi::Object::New(env);
|
16
|
+
|
17
|
+
token_type_enum.Set("IDENTIFIER", 0);
|
18
|
+
token_type_enum.Set("NUMERIC_CONSTANT", 1);
|
19
|
+
token_type_enum.Set("STRING_CONSTANT", 2);
|
20
|
+
token_type_enum.Set("OPERATOR", 3);
|
21
|
+
token_type_enum.Set("KEYWORD", 4);
|
22
|
+
token_type_enum.Set("COMMENT", 5);
|
23
|
+
|
24
|
+
// TypeScript enums expose an inverse mapping.
|
25
|
+
token_type_enum.Set((uint32_t)0, "IDENTIFIER");
|
26
|
+
token_type_enum.Set((uint32_t)1, "NUMERIC_CONSTANT");
|
27
|
+
token_type_enum.Set((uint32_t)2, "STRING_CONSTANT");
|
28
|
+
token_type_enum.Set((uint32_t)3, "OPERATOR");
|
29
|
+
token_type_enum.Set((uint32_t)4, "KEYWORD");
|
30
|
+
token_type_enum.Set((uint32_t)5, "COMMENT");
|
31
|
+
|
32
|
+
token_type_enum_ref = Napi::ObjectReference::New(token_type_enum);
|
33
|
+
|
34
|
+
exports.DefineProperties(
|
35
|
+
{DEFINE_CONSTANT_INTEGER(exports, node_duckdb::Database::DUCKDB_NODEJS_ERROR, ERROR) DEFINE_CONSTANT_INTEGER(
|
17
36
|
exports, node_duckdb::Database::DUCKDB_NODEJS_READONLY, OPEN_READONLY) // same as SQLite
|
18
|
-
|
19
|
-
|
20
|
-
|
21
|
-
|
22
|
-
|
23
|
-
|
37
|
+
DEFINE_CONSTANT_INTEGER(exports, 0, OPEN_READWRITE) // ignored
|
38
|
+
DEFINE_CONSTANT_INTEGER(exports, 0, OPEN_CREATE) // ignored
|
39
|
+
DEFINE_CONSTANT_INTEGER(exports, 0, OPEN_FULLMUTEX) // ignored
|
40
|
+
DEFINE_CONSTANT_INTEGER(exports, 0, OPEN_SHAREDCACHE) // ignored
|
41
|
+
DEFINE_CONSTANT_INTEGER(exports, 0, OPEN_PRIVATECACHE) // ignored
|
42
|
+
|
43
|
+
Napi::PropertyDescriptor::Value("TokenType", token_type_enum,
|
44
|
+
static_cast<napi_property_attributes>(napi_enumerable | napi_configurable))});
|
24
45
|
}
|
25
46
|
|
26
47
|
NODE_API_ADDON(NodeDuckDB);
|
package/src/duckdb_node.hpp
CHANGED
@@ -23,6 +23,7 @@ public:
|
|
23
23
|
Napi::FunctionReference connection_constructor;
|
24
24
|
Napi::FunctionReference statement_constructor;
|
25
25
|
Napi::FunctionReference query_result_constructor;
|
26
|
+
Napi::ObjectReference token_type_enum_ref;
|
26
27
|
};
|
27
28
|
|
28
29
|
namespace node_duckdb {
|
@@ -109,6 +110,7 @@ public:
|
|
109
110
|
Napi::Value Interrupt(const Napi::CallbackInfo &info);
|
110
111
|
Napi::Value Close(const Napi::CallbackInfo &info);
|
111
112
|
Napi::Value RegisterReplacementScan(const Napi::CallbackInfo &info);
|
113
|
+
Napi::Value Tokenize(const Napi::CallbackInfo &info);
|
112
114
|
|
113
115
|
public:
|
114
116
|
constexpr static int DUCKDB_NODEJS_ERROR = -1;
|
@@ -0,0 +1,74 @@
|
|
1
|
+
import * as assert from 'assert';
|
2
|
+
import * as duckdb from '..';
|
3
|
+
|
4
|
+
describe('tokenize', function () {
|
5
|
+
it('should return correct tokens for a single statement', function () {
|
6
|
+
const db = new duckdb.Database(':memory:');
|
7
|
+
const output = db.tokenize('select 1;');
|
8
|
+
assert.deepStrictEqual(output, {
|
9
|
+
offsets: [0, 7, 8],
|
10
|
+
types: [duckdb.TokenType.KEYWORD, duckdb.TokenType.NUMERIC_CONSTANT, duckdb.TokenType.OPERATOR]
|
11
|
+
});
|
12
|
+
});
|
13
|
+
it('should return correct tokens for a multiple statements', function () {
|
14
|
+
const db = new duckdb.Database(':memory:');
|
15
|
+
const output = db.tokenize('select 1; select 2;');
|
16
|
+
assert.deepStrictEqual(output, {
|
17
|
+
offsets: [0, 7, 8, 10, 17, 18],
|
18
|
+
types: [
|
19
|
+
duckdb.TokenType.KEYWORD, duckdb.TokenType.NUMERIC_CONSTANT, duckdb.TokenType.OPERATOR,
|
20
|
+
duckdb.TokenType.KEYWORD, duckdb.TokenType.NUMERIC_CONSTANT, duckdb.TokenType.OPERATOR
|
21
|
+
]
|
22
|
+
});
|
23
|
+
});
|
24
|
+
it('should return no tokens for an empty string', function () {
|
25
|
+
const db = new duckdb.Database(':memory:');
|
26
|
+
const output = db.tokenize('');
|
27
|
+
assert.deepStrictEqual(output, {
|
28
|
+
offsets: [],
|
29
|
+
types: []
|
30
|
+
});
|
31
|
+
});
|
32
|
+
it('should handle quoted semicolons in string constants', function () {
|
33
|
+
const db = new duckdb.Database(':memory:');
|
34
|
+
const output = db.tokenize(`select ';';`);
|
35
|
+
assert.deepStrictEqual(output, {
|
36
|
+
offsets: [0, 7, 10],
|
37
|
+
types: [duckdb.TokenType.KEYWORD, duckdb.TokenType.STRING_CONSTANT, duckdb.TokenType.OPERATOR]
|
38
|
+
});
|
39
|
+
});
|
40
|
+
it('should handle quoted semicolons in identifiers', function () {
|
41
|
+
const db = new duckdb.Database(':memory:');
|
42
|
+
const output = db.tokenize(`from ";";`);
|
43
|
+
assert.deepStrictEqual(output, {
|
44
|
+
offsets: [0, 5, 8],
|
45
|
+
types: [duckdb.TokenType.KEYWORD, duckdb.TokenType.IDENTIFIER, duckdb.TokenType.OPERATOR]
|
46
|
+
});
|
47
|
+
});
|
48
|
+
it('should handle comments', function () {
|
49
|
+
const db = new duckdb.Database(':memory:');
|
50
|
+
const output = db.tokenize(`select /* comment */ 1`);
|
51
|
+
// Note that the tokenizer doesn't return tokens for comments.
|
52
|
+
assert.deepStrictEqual(output, {
|
53
|
+
offsets: [0, 21],
|
54
|
+
types: [duckdb.TokenType.KEYWORD, duckdb.TokenType.NUMERIC_CONSTANT]
|
55
|
+
});
|
56
|
+
});
|
57
|
+
it('should handle invalid syntax', function () {
|
58
|
+
const db = new duckdb.Database(':memory:');
|
59
|
+
const output = db.tokenize(`selec 1`);
|
60
|
+
// The misspelled keyword is scanned as an identifier.
|
61
|
+
assert.deepStrictEqual(output, {
|
62
|
+
offsets: [0, 6],
|
63
|
+
types: [duckdb.TokenType.IDENTIFIER, duckdb.TokenType.NUMERIC_CONSTANT]
|
64
|
+
});
|
65
|
+
});
|
66
|
+
it('should support inverse TokenType mapping', function () {
|
67
|
+
assert.equal(duckdb.TokenType[duckdb.TokenType.IDENTIFIER], "IDENTIFIER");
|
68
|
+
assert.equal(duckdb.TokenType[duckdb.TokenType.NUMERIC_CONSTANT], "NUMERIC_CONSTANT");
|
69
|
+
assert.equal(duckdb.TokenType[duckdb.TokenType.STRING_CONSTANT], "STRING_CONSTANT");
|
70
|
+
assert.equal(duckdb.TokenType[duckdb.TokenType.OPERATOR], "OPERATOR");
|
71
|
+
assert.equal(duckdb.TokenType[duckdb.TokenType.KEYWORD], "KEYWORD");
|
72
|
+
assert.equal(duckdb.TokenType[duckdb.TokenType.COMMENT], "COMMENT");
|
73
|
+
});
|
74
|
+
});
|