@lix-js/sdk 0.6.0-preview.1 → 0.6.0-preview.3
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/SKILL.md +304 -320
- package/dist/engine-wasm/wasm/lix_engine.d.ts +5 -0
- package/dist/engine-wasm/wasm/lix_engine.js +9 -13
- package/dist/engine-wasm/wasm/lix_engine.wasm +0 -0
- package/dist/engine-wasm/wasm/lix_engine.wasm.d.ts +1 -0
- package/dist/generated/builtin-schemas.d.ts +87 -162
- package/dist/generated/builtin-schemas.js +139 -236
- package/dist/open-lix.d.ts +103 -14
- package/dist/open-lix.js +3 -0
- package/dist/sqlite/index.js +99 -22
- package/dist-engine-src/README.md +18 -0
- package/dist-engine-src/src/backend/kv.rs +358 -0
- package/dist-engine-src/src/backend/mod.rs +12 -0
- package/dist-engine-src/src/backend/testing.rs +658 -0
- package/dist-engine-src/src/backend/types.rs +96 -0
- package/dist-engine-src/src/binary_cas/chunking.rs +31 -0
- package/dist-engine-src/src/binary_cas/codec.rs +346 -0
- package/dist-engine-src/src/binary_cas/context.rs +139 -0
- package/dist-engine-src/src/binary_cas/kv.rs +1063 -0
- package/dist-engine-src/src/binary_cas/mod.rs +11 -0
- package/dist-engine-src/src/binary_cas/types.rs +121 -0
- package/dist-engine-src/src/catalog/context.rs +412 -0
- package/dist-engine-src/src/catalog/mod.rs +10 -0
- package/dist-engine-src/src/catalog/schema.rs +4 -0
- package/dist-engine-src/src/catalog/snapshot.rs +1114 -0
- package/dist-engine-src/src/cel/context.rs +86 -0
- package/dist-engine-src/src/cel/error.rs +19 -0
- package/dist-engine-src/src/cel/mod.rs +8 -0
- package/dist-engine-src/src/cel/provider.rs +9 -0
- package/dist-engine-src/src/cel/runtime.rs +167 -0
- package/dist-engine-src/src/cel/value.rs +50 -0
- package/dist-engine-src/src/commit_graph/context.rs +901 -0
- package/dist-engine-src/src/commit_graph/mod.rs +11 -0
- package/dist-engine-src/src/commit_graph/types.rs +109 -0
- package/dist-engine-src/src/commit_graph/walker.rs +756 -0
- package/dist-engine-src/src/commit_store/codec.rs +887 -0
- package/dist-engine-src/src/commit_store/context.rs +944 -0
- package/dist-engine-src/src/commit_store/materialization.rs +84 -0
- package/dist-engine-src/src/commit_store/mod.rs +16 -0
- package/dist-engine-src/src/commit_store/storage.rs +600 -0
- package/dist-engine-src/src/commit_store/types.rs +215 -0
- package/dist-engine-src/src/common/error.rs +313 -0
- package/dist-engine-src/src/common/fingerprint.rs +3 -0
- package/dist-engine-src/src/common/fs_path.rs +1336 -0
- package/dist-engine-src/src/common/identity.rs +145 -0
- package/dist-engine-src/src/common/json_pointer.rs +67 -0
- package/dist-engine-src/src/common/metadata.rs +40 -0
- package/dist-engine-src/src/common/mod.rs +23 -0
- package/dist-engine-src/src/common/types.rs +105 -0
- package/dist-engine-src/src/common/wire.rs +222 -0
- package/dist-engine-src/src/domain.rs +324 -0
- package/dist-engine-src/src/engine.rs +225 -0
- package/dist-engine-src/src/entity_identity.rs +405 -0
- package/dist-engine-src/src/functions/context.rs +292 -0
- package/dist-engine-src/src/functions/deterministic.rs +113 -0
- package/dist-engine-src/src/functions/mod.rs +18 -0
- package/dist-engine-src/src/functions/provider.rs +130 -0
- package/dist-engine-src/src/functions/state.rs +336 -0
- package/dist-engine-src/src/functions/types.rs +37 -0
- package/dist-engine-src/src/init.rs +558 -0
- package/dist-engine-src/src/json_store/compression.rs +77 -0
- package/dist-engine-src/src/json_store/context.rs +423 -0
- package/dist-engine-src/src/json_store/encoded.rs +15 -0
- package/dist-engine-src/src/json_store/mod.rs +12 -0
- package/dist-engine-src/src/json_store/store.rs +1109 -0
- package/dist-engine-src/src/json_store/types.rs +217 -0
- package/dist-engine-src/src/lib.rs +62 -0
- package/dist-engine-src/src/live_state/context.rs +2019 -0
- package/dist-engine-src/src/live_state/mod.rs +15 -0
- package/dist-engine-src/src/live_state/overlay.rs +75 -0
- package/dist-engine-src/src/live_state/reader.rs +23 -0
- package/dist-engine-src/src/live_state/types.rs +222 -0
- package/dist-engine-src/src/live_state/visibility.rs +223 -0
- package/dist-engine-src/src/plugin/archive.rs +438 -0
- package/dist-engine-src/src/plugin/component.rs +183 -0
- package/dist-engine-src/src/plugin/install.rs +619 -0
- package/dist-engine-src/src/plugin/manifest.rs +516 -0
- package/dist-engine-src/src/plugin/materializer.rs +477 -0
- package/dist-engine-src/src/plugin/mod.rs +33 -0
- package/dist-engine-src/src/plugin/plugin_manifest.json +118 -0
- package/dist-engine-src/src/plugin/storage.rs +74 -0
- package/dist-engine-src/src/schema/annotations/defaults.rs +275 -0
- package/dist-engine-src/src/schema/annotations/mod.rs +1 -0
- package/dist-engine-src/src/schema/builtin/lix_account.json +21 -0
- package/dist-engine-src/src/schema/builtin/lix_active_account.json +29 -0
- package/dist-engine-src/src/schema/builtin/lix_binary_blob_ref.json +29 -0
- package/dist-engine-src/src/schema/builtin/lix_change.json +63 -0
- package/dist-engine-src/src/schema/builtin/lix_change_author.json +45 -0
- package/dist-engine-src/src/schema/builtin/lix_commit.json +24 -0
- package/dist-engine-src/src/schema/builtin/lix_commit_edge.json +53 -0
- package/dist-engine-src/src/schema/builtin/lix_directory_descriptor.json +52 -0
- package/dist-engine-src/src/schema/builtin/lix_file_descriptor.json +52 -0
- package/dist-engine-src/src/schema/builtin/lix_key_value.json +40 -0
- package/dist-engine-src/src/schema/builtin/lix_label.json +29 -0
- package/dist-engine-src/src/schema/builtin/lix_label_assignment.json +74 -0
- package/dist-engine-src/src/schema/builtin/lix_registered_schema.json +25 -0
- package/dist-engine-src/src/schema/builtin/lix_version_descriptor.json +34 -0
- package/dist-engine-src/src/schema/builtin/lix_version_ref.json +48 -0
- package/dist-engine-src/src/schema/builtin/mod.rs +222 -0
- package/dist-engine-src/src/schema/compatibility.rs +787 -0
- package/dist-engine-src/src/schema/definition.json +187 -0
- package/dist-engine-src/src/schema/definition.rs +742 -0
- package/dist-engine-src/src/schema/key.rs +138 -0
- package/dist-engine-src/src/schema/mod.rs +20 -0
- package/dist-engine-src/src/schema/seed.rs +14 -0
- package/dist-engine-src/src/schema/tests.rs +780 -0
- package/dist-engine-src/src/session/context.rs +364 -0
- package/dist-engine-src/src/session/create_version.rs +88 -0
- package/dist-engine-src/src/session/execute.rs +478 -0
- package/dist-engine-src/src/session/merge/analysis.rs +102 -0
- package/dist-engine-src/src/session/merge/apply.rs +23 -0
- package/dist-engine-src/src/session/merge/conflicts.rs +63 -0
- package/dist-engine-src/src/session/merge/mod.rs +11 -0
- package/dist-engine-src/src/session/merge/stats.rs +65 -0
- package/dist-engine-src/src/session/merge/version.rs +427 -0
- package/dist-engine-src/src/session/mod.rs +27 -0
- package/dist-engine-src/src/session/optimization9_sql2_bench.rs +100 -0
- package/dist-engine-src/src/session/switch_version.rs +109 -0
- package/dist-engine-src/src/sql2/change_provider.rs +331 -0
- package/dist-engine-src/src/sql2/classify.rs +182 -0
- package/dist-engine-src/src/sql2/context.rs +311 -0
- package/dist-engine-src/src/sql2/directory_history_provider.rs +631 -0
- package/dist-engine-src/src/sql2/directory_provider.rs +2453 -0
- package/dist-engine-src/src/sql2/dml.rs +148 -0
- package/dist-engine-src/src/sql2/entity_history_provider.rs +440 -0
- package/dist-engine-src/src/sql2/entity_provider.rs +3211 -0
- package/dist-engine-src/src/sql2/error.rs +216 -0
- package/dist-engine-src/src/sql2/execute.rs +3440 -0
- package/dist-engine-src/src/sql2/file_history_provider.rs +910 -0
- package/dist-engine-src/src/sql2/file_provider.rs +3679 -0
- package/dist-engine-src/src/sql2/filesystem_planner.rs +1490 -0
- package/dist-engine-src/src/sql2/filesystem_predicates.rs +159 -0
- package/dist-engine-src/src/sql2/filesystem_visibility.rs +383 -0
- package/dist-engine-src/src/sql2/history_projection.rs +56 -0
- package/dist-engine-src/src/sql2/history_provider.rs +412 -0
- package/dist-engine-src/src/sql2/history_route.rs +657 -0
- package/dist-engine-src/src/sql2/lix_state_provider.rs +2512 -0
- package/dist-engine-src/src/sql2/mod.rs +46 -0
- package/dist-engine-src/src/sql2/predicate_typecheck.rs +246 -0
- package/dist-engine-src/src/sql2/public_bind/assignment.rs +46 -0
- package/dist-engine-src/src/sql2/public_bind/capability.rs +41 -0
- package/dist-engine-src/src/sql2/public_bind/dml.rs +166 -0
- package/dist-engine-src/src/sql2/public_bind/mod.rs +25 -0
- package/dist-engine-src/src/sql2/public_bind/table.rs +168 -0
- package/dist-engine-src/src/sql2/read_only.rs +63 -0
- package/dist-engine-src/src/sql2/record_batch.rs +17 -0
- package/dist-engine-src/src/sql2/result_metadata.rs +29 -0
- package/dist-engine-src/src/sql2/runtime.rs +60 -0
- package/dist-engine-src/src/sql2/session.rs +132 -0
- package/dist-engine-src/src/sql2/udfs/common.rs +295 -0
- package/dist-engine-src/src/sql2/udfs/lix_active_version_commit_id.rs +53 -0
- package/dist-engine-src/src/sql2/udfs/lix_empty_blob.rs +47 -0
- package/dist-engine-src/src/sql2/udfs/lix_json.rs +100 -0
- package/dist-engine-src/src/sql2/udfs/lix_json_get.rs +99 -0
- package/dist-engine-src/src/sql2/udfs/lix_json_get_text.rs +99 -0
- package/dist-engine-src/src/sql2/udfs/lix_text_decode.rs +82 -0
- package/dist-engine-src/src/sql2/udfs/lix_text_encode.rs +85 -0
- package/dist-engine-src/src/sql2/udfs/lix_timestamp.rs +76 -0
- package/dist-engine-src/src/sql2/udfs/lix_uuid_v7.rs +76 -0
- package/dist-engine-src/src/sql2/udfs/mod.rs +89 -0
- package/dist-engine-src/src/sql2/udfs/public_call.rs +211 -0
- package/dist-engine-src/src/sql2/version_provider.rs +1202 -0
- package/dist-engine-src/src/sql2/version_scope.rs +394 -0
- package/dist-engine-src/src/sql2/write_normalization.rs +345 -0
- package/dist-engine-src/src/storage/context.rs +356 -0
- package/dist-engine-src/src/storage/mod.rs +14 -0
- package/dist-engine-src/src/storage/read_scope.rs +88 -0
- package/dist-engine-src/src/storage/types.rs +501 -0
- package/dist-engine-src/src/storage_bench.rs +4863 -0
- package/dist-engine-src/src/test_support.rs +228 -0
- package/dist-engine-src/src/tracked_state/by_file_index.rs +98 -0
- package/dist-engine-src/src/tracked_state/codec.rs +2085 -0
- package/dist-engine-src/src/tracked_state/context.rs +1867 -0
- package/dist-engine-src/src/tracked_state/diff.rs +686 -0
- package/dist-engine-src/src/tracked_state/materialization.rs +403 -0
- package/dist-engine-src/src/tracked_state/materializer.rs +488 -0
- package/dist-engine-src/src/tracked_state/merge.rs +492 -0
- package/dist-engine-src/src/tracked_state/mod.rs +32 -0
- package/dist-engine-src/src/tracked_state/storage.rs +375 -0
- package/dist-engine-src/src/tracked_state/tree.rs +3187 -0
- package/dist-engine-src/src/tracked_state/types.rs +231 -0
- package/dist-engine-src/src/transaction/commit.rs +1484 -0
- package/dist-engine-src/src/transaction/context.rs +1548 -0
- package/dist-engine-src/src/transaction/live_state_overlay.rs +35 -0
- package/dist-engine-src/src/transaction/mod.rs +13 -0
- package/dist-engine-src/src/transaction/normalization.rs +890 -0
- package/dist-engine-src/src/transaction/prep.rs +37 -0
- package/dist-engine-src/src/transaction/schema_resolver.rs +149 -0
- package/dist-engine-src/src/transaction/staging.rs +1731 -0
- package/dist-engine-src/src/transaction/types.rs +460 -0
- package/dist-engine-src/src/transaction/validation.rs +5830 -0
- package/dist-engine-src/src/untracked_state/codec.rs +307 -0
- package/dist-engine-src/src/untracked_state/context.rs +98 -0
- package/dist-engine-src/src/untracked_state/materialization.rs +63 -0
- package/dist-engine-src/src/untracked_state/mod.rs +15 -0
- package/dist-engine-src/src/untracked_state/storage.rs +396 -0
- package/dist-engine-src/src/untracked_state/types.rs +146 -0
- package/dist-engine-src/src/version/context.rs +40 -0
- package/dist-engine-src/src/version/lifecycle.rs +221 -0
- package/dist-engine-src/src/version/mod.rs +13 -0
- package/dist-engine-src/src/version/refs.rs +330 -0
- package/dist-engine-src/src/version/stage_rows.rs +67 -0
- package/dist-engine-src/src/version/types.rs +21 -0
- package/dist-engine-src/src/wasm/mod.rs +60 -0
- package/package.json +68 -64
|
@@ -0,0 +1,82 @@
|
|
|
1
|
+
use std::any::Any;
|
|
2
|
+
use std::sync::Arc;
|
|
3
|
+
|
|
4
|
+
use datafusion::arrow::array::StringArray;
|
|
5
|
+
use datafusion::arrow::datatypes::DataType;
|
|
6
|
+
use datafusion::common::{plan_err, Result, ScalarValue};
|
|
7
|
+
use datafusion::logical_expr::{
|
|
8
|
+
ColumnarValue, ScalarFunctionArgs, ScalarUDFImpl, Signature, Volatility,
|
|
9
|
+
};
|
|
10
|
+
|
|
11
|
+
use super::common::{decode_utf8_value, scalar_inputs, validate_utf8_encoding_arg};
|
|
12
|
+
|
|
13
|
+
#[derive(Debug, Clone, PartialEq, Eq, Hash)]
|
|
14
|
+
pub(super) struct LixTextDecode {
|
|
15
|
+
signature: Signature,
|
|
16
|
+
}
|
|
17
|
+
|
|
18
|
+
impl LixTextDecode {
|
|
19
|
+
pub(super) fn new() -> Self {
|
|
20
|
+
Self {
|
|
21
|
+
signature: Signature::one_of(
|
|
22
|
+
vec![Signature::any(1, Volatility::Immutable).type_signature],
|
|
23
|
+
Volatility::Immutable,
|
|
24
|
+
),
|
|
25
|
+
}
|
|
26
|
+
}
|
|
27
|
+
}
|
|
28
|
+
|
|
29
|
+
impl ScalarUDFImpl for LixTextDecode {
|
|
30
|
+
fn as_any(&self) -> &dyn Any {
|
|
31
|
+
self
|
|
32
|
+
}
|
|
33
|
+
|
|
34
|
+
fn name(&self) -> &str {
|
|
35
|
+
"lix_text_decode"
|
|
36
|
+
}
|
|
37
|
+
|
|
38
|
+
fn signature(&self) -> &Signature {
|
|
39
|
+
&self.signature
|
|
40
|
+
}
|
|
41
|
+
|
|
42
|
+
fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
|
|
43
|
+
Ok(DataType::Utf8)
|
|
44
|
+
}
|
|
45
|
+
|
|
46
|
+
fn invoke_with_args(&self, args: ScalarFunctionArgs) -> Result<ColumnarValue> {
|
|
47
|
+
if !(1..=2).contains(&args.args.len()) {
|
|
48
|
+
return plan_err!("lix_text_decode requires 1 or 2 arguments");
|
|
49
|
+
}
|
|
50
|
+
validate_utf8_encoding_arg(self.name(), args.args.get(1))?;
|
|
51
|
+
|
|
52
|
+
let scalar_inputs = scalar_inputs(&args.args);
|
|
53
|
+
let arrays = ColumnarValue::values_to_arrays(&args.args)?;
|
|
54
|
+
let input = &arrays[0];
|
|
55
|
+
let len = input.len();
|
|
56
|
+
|
|
57
|
+
let mut values = Vec::with_capacity(len);
|
|
58
|
+
for row in 0..len {
|
|
59
|
+
values.push(decode_utf8_value(input.as_ref(), row)?);
|
|
60
|
+
}
|
|
61
|
+
if scalar_inputs {
|
|
62
|
+
Ok(ColumnarValue::Scalar(ScalarValue::Utf8(
|
|
63
|
+
values.into_iter().next().flatten(),
|
|
64
|
+
)))
|
|
65
|
+
} else {
|
|
66
|
+
Ok(ColumnarValue::Array(Arc::new(StringArray::from(values))))
|
|
67
|
+
}
|
|
68
|
+
}
|
|
69
|
+
}
|
|
70
|
+
|
|
71
|
+
#[cfg(test)]
|
|
72
|
+
mod tests {
|
|
73
|
+
use super::super::test_support::single_text;
|
|
74
|
+
|
|
75
|
+
#[tokio::test]
|
|
76
|
+
async fn decodes_utf8_binary_to_text() {
|
|
77
|
+
assert_eq!(
|
|
78
|
+
single_text("SELECT lix_text_decode(X'416461')").await,
|
|
79
|
+
Some("Ada".to_string())
|
|
80
|
+
);
|
|
81
|
+
}
|
|
82
|
+
}
|
|
@@ -0,0 +1,85 @@
|
|
|
1
|
+
use std::any::Any;
|
|
2
|
+
|
|
3
|
+
use datafusion::arrow::datatypes::DataType;
|
|
4
|
+
use datafusion::common::{plan_err, Result, ScalarValue};
|
|
5
|
+
use datafusion::logical_expr::{
|
|
6
|
+
ColumnarValue, ScalarFunctionArgs, ScalarUDFImpl, Signature, Volatility,
|
|
7
|
+
};
|
|
8
|
+
|
|
9
|
+
use super::common::{
|
|
10
|
+
array_ref, binary_array_from_owned, encode_utf8_value, scalar_inputs,
|
|
11
|
+
validate_utf8_encoding_arg,
|
|
12
|
+
};
|
|
13
|
+
|
|
14
|
+
#[derive(Debug, Clone, PartialEq, Eq, Hash)]
|
|
15
|
+
pub(super) struct LixTextEncode {
|
|
16
|
+
signature: Signature,
|
|
17
|
+
}
|
|
18
|
+
|
|
19
|
+
impl LixTextEncode {
|
|
20
|
+
pub(super) fn new() -> Self {
|
|
21
|
+
Self {
|
|
22
|
+
signature: Signature::one_of(
|
|
23
|
+
vec![Signature::any(1, Volatility::Immutable).type_signature],
|
|
24
|
+
Volatility::Immutable,
|
|
25
|
+
),
|
|
26
|
+
}
|
|
27
|
+
}
|
|
28
|
+
}
|
|
29
|
+
|
|
30
|
+
impl ScalarUDFImpl for LixTextEncode {
|
|
31
|
+
fn as_any(&self) -> &dyn Any {
|
|
32
|
+
self
|
|
33
|
+
}
|
|
34
|
+
|
|
35
|
+
fn name(&self) -> &str {
|
|
36
|
+
"lix_text_encode"
|
|
37
|
+
}
|
|
38
|
+
|
|
39
|
+
fn signature(&self) -> &Signature {
|
|
40
|
+
&self.signature
|
|
41
|
+
}
|
|
42
|
+
|
|
43
|
+
fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
|
|
44
|
+
Ok(DataType::Binary)
|
|
45
|
+
}
|
|
46
|
+
|
|
47
|
+
fn invoke_with_args(&self, args: ScalarFunctionArgs) -> Result<ColumnarValue> {
|
|
48
|
+
if !(1..=2).contains(&args.args.len()) {
|
|
49
|
+
return plan_err!("lix_text_encode requires 1 or 2 arguments");
|
|
50
|
+
}
|
|
51
|
+
validate_utf8_encoding_arg(self.name(), args.args.get(1))?;
|
|
52
|
+
|
|
53
|
+
let scalar_inputs = scalar_inputs(&args.args);
|
|
54
|
+
let arrays = ColumnarValue::values_to_arrays(&args.args)?;
|
|
55
|
+
let input = &arrays[0];
|
|
56
|
+
let len = input.len();
|
|
57
|
+
|
|
58
|
+
let mut values = Vec::with_capacity(len);
|
|
59
|
+
for row in 0..len {
|
|
60
|
+
values.push(encode_utf8_value(input.as_ref(), row)?);
|
|
61
|
+
}
|
|
62
|
+
if scalar_inputs {
|
|
63
|
+
Ok(ColumnarValue::Scalar(ScalarValue::Binary(
|
|
64
|
+
values.into_iter().next().flatten(),
|
|
65
|
+
)))
|
|
66
|
+
} else {
|
|
67
|
+
Ok(ColumnarValue::Array(array_ref(binary_array_from_owned(
|
|
68
|
+
&values,
|
|
69
|
+
))))
|
|
70
|
+
}
|
|
71
|
+
}
|
|
72
|
+
}
|
|
73
|
+
|
|
74
|
+
#[cfg(test)]
|
|
75
|
+
mod tests {
|
|
76
|
+
use super::super::test_support::single_binary;
|
|
77
|
+
|
|
78
|
+
#[tokio::test]
|
|
79
|
+
async fn encodes_utf8_text_to_binary() {
|
|
80
|
+
assert_eq!(
|
|
81
|
+
single_binary("SELECT lix_text_encode('Ada')").await,
|
|
82
|
+
Some(b"Ada".to_vec())
|
|
83
|
+
);
|
|
84
|
+
}
|
|
85
|
+
}
|
|
@@ -0,0 +1,76 @@
|
|
|
1
|
+
use std::any::Any;
|
|
2
|
+
|
|
3
|
+
use datafusion::arrow::datatypes::DataType;
|
|
4
|
+
use datafusion::common::{plan_err, Result, ScalarValue};
|
|
5
|
+
use datafusion::logical_expr::{
|
|
6
|
+
ColumnarValue, ScalarFunctionArgs, ScalarUDFImpl, Signature, Volatility,
|
|
7
|
+
};
|
|
8
|
+
|
|
9
|
+
use crate::functions::FunctionProviderHandle;
|
|
10
|
+
|
|
11
|
+
#[derive(Clone)]
|
|
12
|
+
pub(super) struct LixTimestamp {
|
|
13
|
+
pub(super) functions: FunctionProviderHandle,
|
|
14
|
+
}
|
|
15
|
+
|
|
16
|
+
impl PartialEq for LixTimestamp {
|
|
17
|
+
fn eq(&self, _other: &Self) -> bool {
|
|
18
|
+
true
|
|
19
|
+
}
|
|
20
|
+
}
|
|
21
|
+
|
|
22
|
+
impl Eq for LixTimestamp {}
|
|
23
|
+
|
|
24
|
+
impl std::hash::Hash for LixTimestamp {
|
|
25
|
+
fn hash<H: std::hash::Hasher>(&self, state: &mut H) {
|
|
26
|
+
self.name().hash(state);
|
|
27
|
+
}
|
|
28
|
+
}
|
|
29
|
+
|
|
30
|
+
impl std::fmt::Debug for LixTimestamp {
|
|
31
|
+
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
|
|
32
|
+
f.debug_struct("LixTimestamp").finish()
|
|
33
|
+
}
|
|
34
|
+
}
|
|
35
|
+
|
|
36
|
+
impl ScalarUDFImpl for LixTimestamp {
|
|
37
|
+
fn as_any(&self) -> &dyn Any {
|
|
38
|
+
self
|
|
39
|
+
}
|
|
40
|
+
|
|
41
|
+
fn name(&self) -> &str {
|
|
42
|
+
"lix_timestamp"
|
|
43
|
+
}
|
|
44
|
+
|
|
45
|
+
fn signature(&self) -> &Signature {
|
|
46
|
+
static SIGNATURE: std::sync::LazyLock<Signature> =
|
|
47
|
+
std::sync::LazyLock::new(|| Signature::nullary(Volatility::Volatile));
|
|
48
|
+
&SIGNATURE
|
|
49
|
+
}
|
|
50
|
+
|
|
51
|
+
fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
|
|
52
|
+
Ok(DataType::Utf8)
|
|
53
|
+
}
|
|
54
|
+
|
|
55
|
+
fn invoke_with_args(&self, args: ScalarFunctionArgs) -> Result<ColumnarValue> {
|
|
56
|
+
if !args.args.is_empty() {
|
|
57
|
+
return plan_err!("lix_timestamp requires no arguments");
|
|
58
|
+
}
|
|
59
|
+
Ok(ColumnarValue::Scalar(ScalarValue::Utf8(Some(
|
|
60
|
+
self.functions.call_timestamp(),
|
|
61
|
+
))))
|
|
62
|
+
}
|
|
63
|
+
}
|
|
64
|
+
|
|
65
|
+
#[cfg(test)]
|
|
66
|
+
mod tests {
|
|
67
|
+
use super::super::test_support::single_text;
|
|
68
|
+
|
|
69
|
+
#[tokio::test]
|
|
70
|
+
async fn returns_timestamp_text() {
|
|
71
|
+
let value = single_text("SELECT lix_timestamp()")
|
|
72
|
+
.await
|
|
73
|
+
.expect("timestamp should not be null");
|
|
74
|
+
assert!(!value.is_empty());
|
|
75
|
+
}
|
|
76
|
+
}
|
|
@@ -0,0 +1,76 @@
|
|
|
1
|
+
use std::any::Any;
|
|
2
|
+
|
|
3
|
+
use datafusion::arrow::datatypes::DataType;
|
|
4
|
+
use datafusion::common::{plan_err, Result, ScalarValue};
|
|
5
|
+
use datafusion::logical_expr::{
|
|
6
|
+
ColumnarValue, ScalarFunctionArgs, ScalarUDFImpl, Signature, Volatility,
|
|
7
|
+
};
|
|
8
|
+
|
|
9
|
+
use crate::functions::FunctionProviderHandle;
|
|
10
|
+
|
|
11
|
+
#[derive(Clone)]
|
|
12
|
+
pub(super) struct LixUuidV7 {
|
|
13
|
+
pub(super) functions: FunctionProviderHandle,
|
|
14
|
+
}
|
|
15
|
+
|
|
16
|
+
impl PartialEq for LixUuidV7 {
|
|
17
|
+
fn eq(&self, _other: &Self) -> bool {
|
|
18
|
+
true
|
|
19
|
+
}
|
|
20
|
+
}
|
|
21
|
+
|
|
22
|
+
impl Eq for LixUuidV7 {}
|
|
23
|
+
|
|
24
|
+
impl std::hash::Hash for LixUuidV7 {
|
|
25
|
+
fn hash<H: std::hash::Hasher>(&self, state: &mut H) {
|
|
26
|
+
self.name().hash(state);
|
|
27
|
+
}
|
|
28
|
+
}
|
|
29
|
+
|
|
30
|
+
impl std::fmt::Debug for LixUuidV7 {
|
|
31
|
+
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
|
|
32
|
+
f.debug_struct("LixUuidV7").finish()
|
|
33
|
+
}
|
|
34
|
+
}
|
|
35
|
+
|
|
36
|
+
impl ScalarUDFImpl for LixUuidV7 {
|
|
37
|
+
fn as_any(&self) -> &dyn Any {
|
|
38
|
+
self
|
|
39
|
+
}
|
|
40
|
+
|
|
41
|
+
fn name(&self) -> &str {
|
|
42
|
+
"lix_uuid_v7"
|
|
43
|
+
}
|
|
44
|
+
|
|
45
|
+
fn signature(&self) -> &Signature {
|
|
46
|
+
static SIGNATURE: std::sync::LazyLock<Signature> =
|
|
47
|
+
std::sync::LazyLock::new(|| Signature::nullary(Volatility::Volatile));
|
|
48
|
+
&SIGNATURE
|
|
49
|
+
}
|
|
50
|
+
|
|
51
|
+
fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
|
|
52
|
+
Ok(DataType::Utf8)
|
|
53
|
+
}
|
|
54
|
+
|
|
55
|
+
fn invoke_with_args(&self, args: ScalarFunctionArgs) -> Result<ColumnarValue> {
|
|
56
|
+
if !args.args.is_empty() {
|
|
57
|
+
return plan_err!("lix_uuid_v7 requires no arguments");
|
|
58
|
+
}
|
|
59
|
+
Ok(ColumnarValue::Scalar(ScalarValue::Utf8(Some(
|
|
60
|
+
self.functions.call_uuid_v7(),
|
|
61
|
+
))))
|
|
62
|
+
}
|
|
63
|
+
}
|
|
64
|
+
|
|
65
|
+
#[cfg(test)]
|
|
66
|
+
mod tests {
|
|
67
|
+
use super::super::test_support::single_text;
|
|
68
|
+
|
|
69
|
+
#[tokio::test]
|
|
70
|
+
async fn returns_uuid_text() {
|
|
71
|
+
let value = single_text("SELECT lix_uuid_v7()")
|
|
72
|
+
.await
|
|
73
|
+
.expect("uuid should not be null");
|
|
74
|
+
assert!(!value.is_empty());
|
|
75
|
+
}
|
|
76
|
+
}
|
|
@@ -0,0 +1,89 @@
|
|
|
1
|
+
mod common;
|
|
2
|
+
mod lix_active_version_commit_id;
|
|
3
|
+
mod lix_empty_blob;
|
|
4
|
+
mod lix_json;
|
|
5
|
+
mod lix_json_get;
|
|
6
|
+
mod lix_json_get_text;
|
|
7
|
+
mod lix_text_decode;
|
|
8
|
+
mod lix_text_encode;
|
|
9
|
+
mod lix_timestamp;
|
|
10
|
+
mod lix_uuid_v7;
|
|
11
|
+
mod public_call;
|
|
12
|
+
|
|
13
|
+
use datafusion::execution::context::SessionContext;
|
|
14
|
+
use datafusion::logical_expr::ScalarUDF;
|
|
15
|
+
|
|
16
|
+
use crate::functions::FunctionProviderHandle;
|
|
17
|
+
|
|
18
|
+
pub(crate) use public_call::validate_public_udf_calls;
|
|
19
|
+
|
|
20
|
+
#[cfg(test)]
|
|
21
|
+
pub(crate) fn system_sql2_function_provider() -> FunctionProviderHandle {
|
|
22
|
+
use crate::functions::{FunctionProvider, SharedFunctionProvider, SystemFunctionProvider};
|
|
23
|
+
|
|
24
|
+
SharedFunctionProvider::new(Box::new(SystemFunctionProvider) as Box<dyn FunctionProvider + Send>)
|
|
25
|
+
}
|
|
26
|
+
|
|
27
|
+
pub(crate) fn register_sql2_functions(
|
|
28
|
+
ctx: &SessionContext,
|
|
29
|
+
functions: FunctionProviderHandle,
|
|
30
|
+
active_version_commit_id: Option<String>,
|
|
31
|
+
) {
|
|
32
|
+
ctx.register_udf(ScalarUDF::from(
|
|
33
|
+
lix_active_version_commit_id::LixActiveVersionCommitId::new(active_version_commit_id),
|
|
34
|
+
));
|
|
35
|
+
ctx.register_udf(ScalarUDF::from(lix_json_get::LixJsonGet::new()));
|
|
36
|
+
ctx.register_udf(ScalarUDF::from(lix_json_get_text::LixJsonGetText::new()));
|
|
37
|
+
ctx.register_udf(ScalarUDF::from(lix_text_decode::LixTextDecode::new()));
|
|
38
|
+
ctx.register_udf(ScalarUDF::from(lix_text_encode::LixTextEncode::new()));
|
|
39
|
+
ctx.register_udf(ScalarUDF::from(lix_json::LixJson));
|
|
40
|
+
ctx.register_udf(ScalarUDF::from(lix_empty_blob::LixEmptyBlob));
|
|
41
|
+
ctx.register_udf(ScalarUDF::from(lix_uuid_v7::LixUuidV7 {
|
|
42
|
+
functions: functions.clone(),
|
|
43
|
+
}));
|
|
44
|
+
ctx.register_udf(ScalarUDF::from(lix_timestamp::LixTimestamp { functions }));
|
|
45
|
+
}
|
|
46
|
+
|
|
47
|
+
#[cfg(test)]
|
|
48
|
+
pub(super) mod test_support {
|
|
49
|
+
use datafusion::arrow::array::{Array, BinaryArray, StringArray};
|
|
50
|
+
use datafusion::prelude::SessionContext;
|
|
51
|
+
|
|
52
|
+
use super::{register_sql2_functions, system_sql2_function_provider};
|
|
53
|
+
|
|
54
|
+
pub(super) async fn single_text(sql: &str) -> Option<String> {
|
|
55
|
+
let ctx = SessionContext::new();
|
|
56
|
+
register_sql2_functions(&ctx, system_sql2_function_provider(), None);
|
|
57
|
+
let batches = ctx
|
|
58
|
+
.sql(sql)
|
|
59
|
+
.await
|
|
60
|
+
.expect("query should plan")
|
|
61
|
+
.collect()
|
|
62
|
+
.await
|
|
63
|
+
.expect("query should execute");
|
|
64
|
+
let array = batches[0]
|
|
65
|
+
.column(0)
|
|
66
|
+
.as_any()
|
|
67
|
+
.downcast_ref::<StringArray>()
|
|
68
|
+
.expect("first column should be utf8");
|
|
69
|
+
(!array.is_null(0)).then(|| array.value(0).to_string())
|
|
70
|
+
}
|
|
71
|
+
|
|
72
|
+
pub(super) async fn single_binary(sql: &str) -> Option<Vec<u8>> {
|
|
73
|
+
let ctx = SessionContext::new();
|
|
74
|
+
register_sql2_functions(&ctx, system_sql2_function_provider(), None);
|
|
75
|
+
let batches = ctx
|
|
76
|
+
.sql(sql)
|
|
77
|
+
.await
|
|
78
|
+
.expect("query should plan")
|
|
79
|
+
.collect()
|
|
80
|
+
.await
|
|
81
|
+
.expect("query should execute");
|
|
82
|
+
let array = batches[0]
|
|
83
|
+
.column(0)
|
|
84
|
+
.as_any()
|
|
85
|
+
.downcast_ref::<BinaryArray>()
|
|
86
|
+
.expect("first column should be binary");
|
|
87
|
+
(!array.is_null(0)).then(|| array.value(0).to_vec())
|
|
88
|
+
}
|
|
89
|
+
}
|
|
@@ -0,0 +1,211 @@
|
|
|
1
|
+
use std::ops::ControlFlow;
|
|
2
|
+
|
|
3
|
+
use datafusion::sql::sqlparser::ast::{
|
|
4
|
+
Expr, Function, FunctionArg, FunctionArgExpr, FunctionArguments, ObjectNamePart, Statement,
|
|
5
|
+
Value, Visit, Visitor,
|
|
6
|
+
};
|
|
7
|
+
use datafusion::sql::sqlparser::dialect::GenericDialect;
|
|
8
|
+
use datafusion::sql::sqlparser::parser::Parser;
|
|
9
|
+
|
|
10
|
+
use crate::LixError;
|
|
11
|
+
|
|
12
|
+
pub(crate) fn validate_public_udf_calls(sql: &str) -> Result<(), LixError> {
|
|
13
|
+
let statements = Parser::parse_sql(&GenericDialect {}, sql).map_err(|error| {
|
|
14
|
+
LixError::new(
|
|
15
|
+
LixError::CODE_PARSE_ERROR,
|
|
16
|
+
format!("sql2 SQL parse error: {error}"),
|
|
17
|
+
)
|
|
18
|
+
})?;
|
|
19
|
+
|
|
20
|
+
let mut visitor = PublicUdfCallVisitor;
|
|
21
|
+
match statements.visit(&mut visitor) {
|
|
22
|
+
ControlFlow::Continue(()) => Ok(()),
|
|
23
|
+
ControlFlow::Break(error) => Err(*error),
|
|
24
|
+
}
|
|
25
|
+
}
|
|
26
|
+
|
|
27
|
+
struct PublicUdfCallVisitor;
|
|
28
|
+
|
|
29
|
+
impl Visitor for PublicUdfCallVisitor {
|
|
30
|
+
type Break = Box<LixError>;
|
|
31
|
+
|
|
32
|
+
fn pre_visit_expr(&mut self, expr: &Expr) -> ControlFlow<Self::Break> {
|
|
33
|
+
let Expr::Function(function) = expr else {
|
|
34
|
+
return ControlFlow::Continue(());
|
|
35
|
+
};
|
|
36
|
+
|
|
37
|
+
match validate_public_function_call(function) {
|
|
38
|
+
Ok(()) => ControlFlow::Continue(()),
|
|
39
|
+
Err(error) => ControlFlow::Break(Box::new(error)),
|
|
40
|
+
}
|
|
41
|
+
}
|
|
42
|
+
|
|
43
|
+
fn pre_visit_statement(&mut self, statement: &Statement) -> ControlFlow<Self::Break> {
|
|
44
|
+
match statement {
|
|
45
|
+
Statement::CreateFunction(_) | Statement::DropFunction(_) => ControlFlow::Continue(()),
|
|
46
|
+
_ => ControlFlow::Continue(()),
|
|
47
|
+
}
|
|
48
|
+
}
|
|
49
|
+
}
|
|
50
|
+
|
|
51
|
+
fn validate_public_function_call(function: &Function) -> Result<(), LixError> {
|
|
52
|
+
let Some(name) = public_lix_function_name(function) else {
|
|
53
|
+
return Ok(());
|
|
54
|
+
};
|
|
55
|
+
let arity = function_arity(&function.args);
|
|
56
|
+
|
|
57
|
+
match name {
|
|
58
|
+
"lix_json" => expect_exact_arity(name, arity, 1),
|
|
59
|
+
"lix_empty_blob" => expect_exact_arity(name, arity, 0),
|
|
60
|
+
"lix_timestamp" => expect_exact_arity(name, arity, 0),
|
|
61
|
+
"lix_uuid_v7" => expect_exact_arity(name, arity, 0),
|
|
62
|
+
"lix_active_version_commit_id" => expect_exact_arity(name, arity, 0),
|
|
63
|
+
"lix_text_encode" | "lix_text_decode" => {
|
|
64
|
+
expect_arity_range(name, arity, 1, 2)?;
|
|
65
|
+
validate_literal_utf8_encoding(name, &function.args)
|
|
66
|
+
}
|
|
67
|
+
_ => Ok(()),
|
|
68
|
+
}
|
|
69
|
+
}
|
|
70
|
+
|
|
71
|
+
fn public_lix_function_name(function: &Function) -> Option<&'static str> {
|
|
72
|
+
let part = function.name.0.last()?;
|
|
73
|
+
let ident = match part {
|
|
74
|
+
ObjectNamePart::Identifier(ident) => ident.value.as_str(),
|
|
75
|
+
ObjectNamePart::Function(_) => return None,
|
|
76
|
+
};
|
|
77
|
+
match ident.to_ascii_lowercase().as_str() {
|
|
78
|
+
"lix_json" => Some("lix_json"),
|
|
79
|
+
"lix_empty_blob" => Some("lix_empty_blob"),
|
|
80
|
+
"lix_timestamp" => Some("lix_timestamp"),
|
|
81
|
+
"lix_uuid_v7" => Some("lix_uuid_v7"),
|
|
82
|
+
"lix_active_version_commit_id" => Some("lix_active_version_commit_id"),
|
|
83
|
+
"lix_text_encode" => Some("lix_text_encode"),
|
|
84
|
+
"lix_text_decode" => Some("lix_text_decode"),
|
|
85
|
+
_ => None,
|
|
86
|
+
}
|
|
87
|
+
}
|
|
88
|
+
|
|
89
|
+
fn function_arity(args: &FunctionArguments) -> usize {
|
|
90
|
+
match args {
|
|
91
|
+
FunctionArguments::None => 0,
|
|
92
|
+
FunctionArguments::Subquery(_) => 1,
|
|
93
|
+
FunctionArguments::List(list) => list.args.len(),
|
|
94
|
+
}
|
|
95
|
+
}
|
|
96
|
+
|
|
97
|
+
fn expect_exact_arity(name: &str, actual: usize, expected: usize) -> Result<(), LixError> {
|
|
98
|
+
if actual == expected {
|
|
99
|
+
return Ok(());
|
|
100
|
+
}
|
|
101
|
+
|
|
102
|
+
let expectation = if expected == 0 {
|
|
103
|
+
"no arguments".to_string()
|
|
104
|
+
} else if expected == 1 {
|
|
105
|
+
"exactly 1 argument".to_string()
|
|
106
|
+
} else {
|
|
107
|
+
format!("exactly {expected} arguments")
|
|
108
|
+
};
|
|
109
|
+
Err(invalid_param(format!("{name} requires {expectation}")))
|
|
110
|
+
}
|
|
111
|
+
|
|
112
|
+
fn expect_arity_range(name: &str, actual: usize, min: usize, max: usize) -> Result<(), LixError> {
|
|
113
|
+
if (min..=max).contains(&actual) {
|
|
114
|
+
return Ok(());
|
|
115
|
+
}
|
|
116
|
+
Err(invalid_param(format!(
|
|
117
|
+
"{name} requires {min} or {max} arguments"
|
|
118
|
+
)))
|
|
119
|
+
}
|
|
120
|
+
|
|
121
|
+
fn validate_literal_utf8_encoding(name: &str, args: &FunctionArguments) -> Result<(), LixError> {
|
|
122
|
+
let Some(encoding) = function_arg(args, 1) else {
|
|
123
|
+
return Ok(());
|
|
124
|
+
};
|
|
125
|
+
let Some(value) = string_literal_arg(encoding) else {
|
|
126
|
+
return Ok(());
|
|
127
|
+
};
|
|
128
|
+
let normalized = value.trim().to_ascii_uppercase().replace('-', "");
|
|
129
|
+
if normalized == "UTF8" {
|
|
130
|
+
Ok(())
|
|
131
|
+
} else {
|
|
132
|
+
Err(invalid_param(format!(
|
|
133
|
+
"{name}() only supports UTF8 encoding, got '{value}'"
|
|
134
|
+
)))
|
|
135
|
+
}
|
|
136
|
+
}
|
|
137
|
+
|
|
138
|
+
fn function_arg(args: &FunctionArguments, index: usize) -> Option<&FunctionArg> {
|
|
139
|
+
match args {
|
|
140
|
+
FunctionArguments::List(list) => list.args.get(index),
|
|
141
|
+
_ => None,
|
|
142
|
+
}
|
|
143
|
+
}
|
|
144
|
+
|
|
145
|
+
fn string_literal_arg(arg: &FunctionArg) -> Option<&str> {
|
|
146
|
+
let expr = match arg {
|
|
147
|
+
FunctionArg::Unnamed(FunctionArgExpr::Expr(expr))
|
|
148
|
+
| FunctionArg::Named {
|
|
149
|
+
arg: FunctionArgExpr::Expr(expr),
|
|
150
|
+
..
|
|
151
|
+
}
|
|
152
|
+
| FunctionArg::ExprNamed {
|
|
153
|
+
arg: FunctionArgExpr::Expr(expr),
|
|
154
|
+
..
|
|
155
|
+
} => expr,
|
|
156
|
+
_ => return None,
|
|
157
|
+
};
|
|
158
|
+
let Expr::Value(value) = expr else {
|
|
159
|
+
return None;
|
|
160
|
+
};
|
|
161
|
+
match &value.value {
|
|
162
|
+
Value::SingleQuotedString(value)
|
|
163
|
+
| Value::DoubleQuotedString(value)
|
|
164
|
+
| Value::TripleSingleQuotedString(value)
|
|
165
|
+
| Value::TripleDoubleQuotedString(value)
|
|
166
|
+
| Value::EscapedStringLiteral(value)
|
|
167
|
+
| Value::UnicodeStringLiteral(value)
|
|
168
|
+
| Value::NationalStringLiteral(value)
|
|
169
|
+
| Value::SingleQuotedRawStringLiteral(value)
|
|
170
|
+
| Value::DoubleQuotedRawStringLiteral(value)
|
|
171
|
+
| Value::TripleSingleQuotedRawStringLiteral(value)
|
|
172
|
+
| Value::TripleDoubleQuotedRawStringLiteral(value) => Some(value.as_str()),
|
|
173
|
+
Value::DollarQuotedString(value) => Some(value.value.as_str()),
|
|
174
|
+
_ => None,
|
|
175
|
+
}
|
|
176
|
+
}
|
|
177
|
+
|
|
178
|
+
fn invalid_param(message: impl Into<String>) -> LixError {
|
|
179
|
+
LixError::new(LixError::CODE_INVALID_PARAM, message)
|
|
180
|
+
}
|
|
181
|
+
|
|
182
|
+
#[cfg(test)]
|
|
183
|
+
mod tests {
|
|
184
|
+
use super::validate_public_udf_calls;
|
|
185
|
+
|
|
186
|
+
#[test]
|
|
187
|
+
fn rejects_lix_udf_wrong_arity_as_public_invalid_param() {
|
|
188
|
+
let error = validate_public_udf_calls("SELECT lix_uuid_v7('extra')")
|
|
189
|
+
.expect_err("wrong arity should be rejected");
|
|
190
|
+
assert_eq!(error.code, "LIX_INVALID_PARAM");
|
|
191
|
+
assert!(error.message.contains("lix_uuid_v7 requires no arguments"));
|
|
192
|
+
}
|
|
193
|
+
|
|
194
|
+
#[test]
|
|
195
|
+
fn rejects_unsupported_literal_encoding_as_public_invalid_param() {
|
|
196
|
+
let error = validate_public_udf_calls("SELECT lix_text_encode('Ada', 'base64')")
|
|
197
|
+
.expect_err("unsupported encoding should be rejected");
|
|
198
|
+
assert_eq!(error.code, "LIX_INVALID_PARAM");
|
|
199
|
+
assert!(error
|
|
200
|
+
.message
|
|
201
|
+
.contains("lix_text_encode() only supports UTF8 encoding"));
|
|
202
|
+
}
|
|
203
|
+
|
|
204
|
+
#[test]
|
|
205
|
+
fn accepts_valid_public_lix_udf_calls() {
|
|
206
|
+
validate_public_udf_calls(
|
|
207
|
+
"SELECT lix_json('{\"x\":1}'), lix_text_decode(X'416461', 'utf-8')",
|
|
208
|
+
)
|
|
209
|
+
.expect("valid calls should pass public validation");
|
|
210
|
+
}
|
|
211
|
+
}
|