@lix-js/sdk 0.6.0-preview.4 → 0.6.0

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.
Files changed (223) hide show
  1. package/README.md +76 -4
  2. package/dist/errors.d.ts +7 -0
  3. package/dist/errors.js +19 -0
  4. package/dist/index.d.ts +4 -5
  5. package/dist/index.js +3 -3
  6. package/dist/native.d.ts +1 -0
  7. package/dist/native.js +47 -0
  8. package/dist/open-lix.d.ts +39 -201
  9. package/dist/open-lix.js +59 -284
  10. package/dist/result.d.ts +18 -0
  11. package/dist/result.js +48 -0
  12. package/dist/types.d.ts +114 -1
  13. package/dist/value.d.ts +28 -0
  14. package/dist/value.js +245 -0
  15. package/package.json +20 -50
  16. package/SKILL.md +0 -506
  17. package/dist/builtin-schemas.d.ts +0 -1
  18. package/dist/builtin-schemas.js +0 -1
  19. package/dist/engine-wasm/index.d.ts +0 -87
  20. package/dist/engine-wasm/index.js +0 -339
  21. package/dist/engine-wasm/wasm/lix_engine.d.ts +0 -79
  22. package/dist/engine-wasm/wasm/lix_engine.js +0 -821
  23. package/dist/engine-wasm/wasm/lix_engine.wasm +0 -0
  24. package/dist/engine-wasm/wasm/lix_engine.wasm.d.ts +0 -26
  25. package/dist/generated/builtin-schemas.d.ts +0 -427
  26. package/dist/generated/builtin-schemas.js +0 -643
  27. package/dist/sqlite/index.d.ts +0 -12
  28. package/dist/sqlite/index.js +0 -303
  29. package/dist-engine-src/README.md +0 -18
  30. package/dist-engine-src/src/backend/kv.rs +0 -358
  31. package/dist-engine-src/src/backend/mod.rs +0 -12
  32. package/dist-engine-src/src/backend/testing.rs +0 -658
  33. package/dist-engine-src/src/backend/types.rs +0 -96
  34. package/dist-engine-src/src/binary_cas/chunking.rs +0 -31
  35. package/dist-engine-src/src/binary_cas/codec.rs +0 -346
  36. package/dist-engine-src/src/binary_cas/context.rs +0 -139
  37. package/dist-engine-src/src/binary_cas/kv.rs +0 -1063
  38. package/dist-engine-src/src/binary_cas/mod.rs +0 -11
  39. package/dist-engine-src/src/binary_cas/types.rs +0 -121
  40. package/dist-engine-src/src/catalog/context.rs +0 -412
  41. package/dist-engine-src/src/catalog/mod.rs +0 -10
  42. package/dist-engine-src/src/catalog/schema.rs +0 -4
  43. package/dist-engine-src/src/catalog/snapshot.rs +0 -1114
  44. package/dist-engine-src/src/cel/context.rs +0 -86
  45. package/dist-engine-src/src/cel/error.rs +0 -19
  46. package/dist-engine-src/src/cel/mod.rs +0 -8
  47. package/dist-engine-src/src/cel/provider.rs +0 -9
  48. package/dist-engine-src/src/cel/runtime.rs +0 -167
  49. package/dist-engine-src/src/cel/value.rs +0 -50
  50. package/dist-engine-src/src/commit_graph/context.rs +0 -901
  51. package/dist-engine-src/src/commit_graph/mod.rs +0 -11
  52. package/dist-engine-src/src/commit_graph/types.rs +0 -109
  53. package/dist-engine-src/src/commit_graph/walker.rs +0 -756
  54. package/dist-engine-src/src/commit_store/codec.rs +0 -887
  55. package/dist-engine-src/src/commit_store/context.rs +0 -944
  56. package/dist-engine-src/src/commit_store/materialization.rs +0 -84
  57. package/dist-engine-src/src/commit_store/mod.rs +0 -16
  58. package/dist-engine-src/src/commit_store/storage.rs +0 -600
  59. package/dist-engine-src/src/commit_store/types.rs +0 -215
  60. package/dist-engine-src/src/common/error.rs +0 -313
  61. package/dist-engine-src/src/common/fingerprint.rs +0 -3
  62. package/dist-engine-src/src/common/fs_path.rs +0 -1336
  63. package/dist-engine-src/src/common/identity.rs +0 -145
  64. package/dist-engine-src/src/common/json_pointer.rs +0 -67
  65. package/dist-engine-src/src/common/metadata.rs +0 -40
  66. package/dist-engine-src/src/common/mod.rs +0 -23
  67. package/dist-engine-src/src/common/types.rs +0 -105
  68. package/dist-engine-src/src/common/wire.rs +0 -222
  69. package/dist-engine-src/src/domain.rs +0 -324
  70. package/dist-engine-src/src/engine.rs +0 -225
  71. package/dist-engine-src/src/entity_identity.rs +0 -405
  72. package/dist-engine-src/src/functions/context.rs +0 -292
  73. package/dist-engine-src/src/functions/deterministic.rs +0 -113
  74. package/dist-engine-src/src/functions/mod.rs +0 -18
  75. package/dist-engine-src/src/functions/provider.rs +0 -130
  76. package/dist-engine-src/src/functions/state.rs +0 -336
  77. package/dist-engine-src/src/functions/types.rs +0 -37
  78. package/dist-engine-src/src/init.rs +0 -558
  79. package/dist-engine-src/src/json_store/compression.rs +0 -77
  80. package/dist-engine-src/src/json_store/context.rs +0 -423
  81. package/dist-engine-src/src/json_store/encoded.rs +0 -15
  82. package/dist-engine-src/src/json_store/mod.rs +0 -12
  83. package/dist-engine-src/src/json_store/store.rs +0 -1109
  84. package/dist-engine-src/src/json_store/types.rs +0 -217
  85. package/dist-engine-src/src/lib.rs +0 -62
  86. package/dist-engine-src/src/live_state/context.rs +0 -2019
  87. package/dist-engine-src/src/live_state/mod.rs +0 -15
  88. package/dist-engine-src/src/live_state/overlay.rs +0 -75
  89. package/dist-engine-src/src/live_state/reader.rs +0 -23
  90. package/dist-engine-src/src/live_state/types.rs +0 -222
  91. package/dist-engine-src/src/live_state/visibility.rs +0 -223
  92. package/dist-engine-src/src/plugin/archive.rs +0 -438
  93. package/dist-engine-src/src/plugin/component.rs +0 -183
  94. package/dist-engine-src/src/plugin/install.rs +0 -619
  95. package/dist-engine-src/src/plugin/manifest.rs +0 -516
  96. package/dist-engine-src/src/plugin/materializer.rs +0 -477
  97. package/dist-engine-src/src/plugin/mod.rs +0 -33
  98. package/dist-engine-src/src/plugin/plugin_manifest.json +0 -118
  99. package/dist-engine-src/src/plugin/storage.rs +0 -74
  100. package/dist-engine-src/src/schema/annotations/defaults.rs +0 -275
  101. package/dist-engine-src/src/schema/annotations/mod.rs +0 -1
  102. package/dist-engine-src/src/schema/builtin/lix_account.json +0 -21
  103. package/dist-engine-src/src/schema/builtin/lix_active_account.json +0 -29
  104. package/dist-engine-src/src/schema/builtin/lix_binary_blob_ref.json +0 -29
  105. package/dist-engine-src/src/schema/builtin/lix_change.json +0 -63
  106. package/dist-engine-src/src/schema/builtin/lix_change_author.json +0 -45
  107. package/dist-engine-src/src/schema/builtin/lix_commit.json +0 -24
  108. package/dist-engine-src/src/schema/builtin/lix_commit_edge.json +0 -53
  109. package/dist-engine-src/src/schema/builtin/lix_directory_descriptor.json +0 -52
  110. package/dist-engine-src/src/schema/builtin/lix_file_descriptor.json +0 -52
  111. package/dist-engine-src/src/schema/builtin/lix_key_value.json +0 -40
  112. package/dist-engine-src/src/schema/builtin/lix_label.json +0 -29
  113. package/dist-engine-src/src/schema/builtin/lix_label_assignment.json +0 -74
  114. package/dist-engine-src/src/schema/builtin/lix_registered_schema.json +0 -25
  115. package/dist-engine-src/src/schema/builtin/lix_version_descriptor.json +0 -34
  116. package/dist-engine-src/src/schema/builtin/lix_version_ref.json +0 -48
  117. package/dist-engine-src/src/schema/builtin/mod.rs +0 -222
  118. package/dist-engine-src/src/schema/compatibility.rs +0 -787
  119. package/dist-engine-src/src/schema/definition.json +0 -187
  120. package/dist-engine-src/src/schema/definition.rs +0 -742
  121. package/dist-engine-src/src/schema/key.rs +0 -138
  122. package/dist-engine-src/src/schema/mod.rs +0 -20
  123. package/dist-engine-src/src/schema/seed.rs +0 -14
  124. package/dist-engine-src/src/schema/tests.rs +0 -780
  125. package/dist-engine-src/src/session/context.rs +0 -404
  126. package/dist-engine-src/src/session/create_version.rs +0 -88
  127. package/dist-engine-src/src/session/execute.rs +0 -541
  128. package/dist-engine-src/src/session/merge/analysis.rs +0 -102
  129. package/dist-engine-src/src/session/merge/apply.rs +0 -23
  130. package/dist-engine-src/src/session/merge/conflicts.rs +0 -63
  131. package/dist-engine-src/src/session/merge/mod.rs +0 -11
  132. package/dist-engine-src/src/session/merge/stats.rs +0 -65
  133. package/dist-engine-src/src/session/merge/version.rs +0 -427
  134. package/dist-engine-src/src/session/mod.rs +0 -27
  135. package/dist-engine-src/src/session/optimization9_sql2_bench.rs +0 -100
  136. package/dist-engine-src/src/session/switch_version.rs +0 -110
  137. package/dist-engine-src/src/session/transaction.rs +0 -76
  138. package/dist-engine-src/src/sql2/change_provider.rs +0 -331
  139. package/dist-engine-src/src/sql2/classify.rs +0 -174
  140. package/dist-engine-src/src/sql2/context.rs +0 -311
  141. package/dist-engine-src/src/sql2/directory_history_provider.rs +0 -631
  142. package/dist-engine-src/src/sql2/directory_provider.rs +0 -2453
  143. package/dist-engine-src/src/sql2/dml.rs +0 -148
  144. package/dist-engine-src/src/sql2/entity_history_provider.rs +0 -440
  145. package/dist-engine-src/src/sql2/entity_provider.rs +0 -3211
  146. package/dist-engine-src/src/sql2/error.rs +0 -215
  147. package/dist-engine-src/src/sql2/execute.rs +0 -3533
  148. package/dist-engine-src/src/sql2/file_history_provider.rs +0 -910
  149. package/dist-engine-src/src/sql2/file_provider.rs +0 -3679
  150. package/dist-engine-src/src/sql2/filesystem_planner.rs +0 -1490
  151. package/dist-engine-src/src/sql2/filesystem_predicates.rs +0 -159
  152. package/dist-engine-src/src/sql2/filesystem_visibility.rs +0 -383
  153. package/dist-engine-src/src/sql2/history_projection.rs +0 -56
  154. package/dist-engine-src/src/sql2/history_provider.rs +0 -412
  155. package/dist-engine-src/src/sql2/history_route.rs +0 -657
  156. package/dist-engine-src/src/sql2/lix_state_provider.rs +0 -2512
  157. package/dist-engine-src/src/sql2/mod.rs +0 -47
  158. package/dist-engine-src/src/sql2/predicate_typecheck.rs +0 -246
  159. package/dist-engine-src/src/sql2/public_bind/assignment.rs +0 -46
  160. package/dist-engine-src/src/sql2/public_bind/capability.rs +0 -41
  161. package/dist-engine-src/src/sql2/public_bind/dml.rs +0 -172
  162. package/dist-engine-src/src/sql2/public_bind/mod.rs +0 -26
  163. package/dist-engine-src/src/sql2/public_bind/table.rs +0 -168
  164. package/dist-engine-src/src/sql2/read_only.rs +0 -63
  165. package/dist-engine-src/src/sql2/record_batch.rs +0 -17
  166. package/dist-engine-src/src/sql2/result_metadata.rs +0 -29
  167. package/dist-engine-src/src/sql2/runtime.rs +0 -60
  168. package/dist-engine-src/src/sql2/session.rs +0 -132
  169. package/dist-engine-src/src/sql2/udfs/common.rs +0 -295
  170. package/dist-engine-src/src/sql2/udfs/lix_active_version_commit_id.rs +0 -53
  171. package/dist-engine-src/src/sql2/udfs/lix_empty_blob.rs +0 -47
  172. package/dist-engine-src/src/sql2/udfs/lix_json.rs +0 -100
  173. package/dist-engine-src/src/sql2/udfs/lix_json_get.rs +0 -99
  174. package/dist-engine-src/src/sql2/udfs/lix_json_get_text.rs +0 -99
  175. package/dist-engine-src/src/sql2/udfs/lix_text_decode.rs +0 -82
  176. package/dist-engine-src/src/sql2/udfs/lix_text_encode.rs +0 -85
  177. package/dist-engine-src/src/sql2/udfs/lix_timestamp.rs +0 -76
  178. package/dist-engine-src/src/sql2/udfs/lix_uuid_v7.rs +0 -76
  179. package/dist-engine-src/src/sql2/udfs/mod.rs +0 -89
  180. package/dist-engine-src/src/sql2/udfs/public_call.rs +0 -238
  181. package/dist-engine-src/src/sql2/version_provider.rs +0 -1202
  182. package/dist-engine-src/src/sql2/version_scope.rs +0 -394
  183. package/dist-engine-src/src/sql2/write_normalization.rs +0 -345
  184. package/dist-engine-src/src/storage/context.rs +0 -356
  185. package/dist-engine-src/src/storage/mod.rs +0 -14
  186. package/dist-engine-src/src/storage/read_scope.rs +0 -88
  187. package/dist-engine-src/src/storage/types.rs +0 -501
  188. package/dist-engine-src/src/storage_bench.rs +0 -4863
  189. package/dist-engine-src/src/test_support.rs +0 -228
  190. package/dist-engine-src/src/tracked_state/by_file_index.rs +0 -98
  191. package/dist-engine-src/src/tracked_state/codec.rs +0 -2085
  192. package/dist-engine-src/src/tracked_state/context.rs +0 -1867
  193. package/dist-engine-src/src/tracked_state/diff.rs +0 -686
  194. package/dist-engine-src/src/tracked_state/materialization.rs +0 -403
  195. package/dist-engine-src/src/tracked_state/materializer.rs +0 -488
  196. package/dist-engine-src/src/tracked_state/merge.rs +0 -492
  197. package/dist-engine-src/src/tracked_state/mod.rs +0 -32
  198. package/dist-engine-src/src/tracked_state/storage.rs +0 -375
  199. package/dist-engine-src/src/tracked_state/tree.rs +0 -3187
  200. package/dist-engine-src/src/tracked_state/types.rs +0 -231
  201. package/dist-engine-src/src/transaction/commit.rs +0 -1484
  202. package/dist-engine-src/src/transaction/context.rs +0 -1548
  203. package/dist-engine-src/src/transaction/live_state_overlay.rs +0 -35
  204. package/dist-engine-src/src/transaction/mod.rs +0 -13
  205. package/dist-engine-src/src/transaction/normalization.rs +0 -890
  206. package/dist-engine-src/src/transaction/prep.rs +0 -37
  207. package/dist-engine-src/src/transaction/schema_resolver.rs +0 -149
  208. package/dist-engine-src/src/transaction/staging.rs +0 -1731
  209. package/dist-engine-src/src/transaction/types.rs +0 -460
  210. package/dist-engine-src/src/transaction/validation.rs +0 -5830
  211. package/dist-engine-src/src/untracked_state/codec.rs +0 -307
  212. package/dist-engine-src/src/untracked_state/context.rs +0 -98
  213. package/dist-engine-src/src/untracked_state/materialization.rs +0 -63
  214. package/dist-engine-src/src/untracked_state/mod.rs +0 -15
  215. package/dist-engine-src/src/untracked_state/storage.rs +0 -396
  216. package/dist-engine-src/src/untracked_state/types.rs +0 -146
  217. package/dist-engine-src/src/version/context.rs +0 -40
  218. package/dist-engine-src/src/version/lifecycle.rs +0 -221
  219. package/dist-engine-src/src/version/mod.rs +0 -13
  220. package/dist-engine-src/src/version/refs.rs +0 -330
  221. package/dist-engine-src/src/version/stage_rows.rs +0 -67
  222. package/dist-engine-src/src/version/types.rs +0 -21
  223. package/dist-engine-src/src/wasm/mod.rs +0 -60
@@ -1,412 +0,0 @@
1
- use std::any::Any;
2
- use std::sync::Arc;
3
-
4
- use async_trait::async_trait;
5
- use datafusion::arrow::array::{ArrayRef, Int64Array, StringArray};
6
- use datafusion::arrow::datatypes::{DataType, Field, Schema, SchemaRef};
7
- use datafusion::arrow::record_batch::{RecordBatch, RecordBatchOptions};
8
- use datafusion::catalog::{Session, TableProvider};
9
- use datafusion::common::{DataFusionError, Result};
10
- use datafusion::datasource::TableType;
11
- use datafusion::execution::TaskContext;
12
- use datafusion::logical_expr::{Expr, TableProviderFilterPushDown};
13
- use datafusion::physical_expr::EquivalenceProperties;
14
- use datafusion::physical_plan::execution_plan::{Boundedness, EmissionType, PlanProperties};
15
- use datafusion::physical_plan::stream::RecordBatchStreamAdapter;
16
- use datafusion::physical_plan::{
17
- DisplayAs, DisplayFormatType, ExecutionPlan, Partitioning, SendableRecordBatchStream,
18
- };
19
- use datafusion::prelude::SessionContext;
20
- use futures_util::{stream, TryStreamExt};
21
- use tokio::sync::Mutex;
22
-
23
- use crate::commit_graph::CommitGraphReader;
24
- use crate::{serialize_row_metadata, LixError};
25
-
26
- use super::history_route::{
27
- load_history_entries, parse_history_filter, HistoryColumnStyle, HistoryRoute,
28
- HistoryViewDescriptor,
29
- };
30
- use super::result_metadata::json_field;
31
- use super::SqlCommitStoreQuerySource;
32
-
33
- pub(crate) async fn register_history_providers(
34
- session: &SessionContext,
35
- commit_graph: Box<dyn CommitGraphReader>,
36
- query_source: SqlCommitStoreQuerySource,
37
- ) -> Result<Arc<dyn TableProvider>, LixError> {
38
- let provider: Arc<dyn TableProvider> = Arc::new(LixStateHistoryProvider::new(
39
- Arc::new(Mutex::new(commit_graph)),
40
- query_source,
41
- ));
42
- session
43
- .register_table("lix_state_history", Arc::clone(&provider))
44
- .map_err(datafusion_error_to_lix_error)?;
45
- Ok(provider)
46
- }
47
-
48
- pub(crate) struct LixStateHistoryProvider {
49
- schema: SchemaRef,
50
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
51
- query_source: SqlCommitStoreQuerySource,
52
- }
53
-
54
- impl std::fmt::Debug for LixStateHistoryProvider {
55
- fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
56
- f.debug_struct("LixStateHistoryProvider").finish()
57
- }
58
- }
59
-
60
- impl LixStateHistoryProvider {
61
- pub(crate) fn new(
62
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
63
- query_source: SqlCommitStoreQuerySource,
64
- ) -> Self {
65
- Self {
66
- schema: lix_state_history_schema(),
67
- commit_graph,
68
- query_source,
69
- }
70
- }
71
- }
72
-
73
- #[async_trait]
74
- impl TableProvider for LixStateHistoryProvider {
75
- fn as_any(&self) -> &dyn Any {
76
- self
77
- }
78
-
79
- fn schema(&self) -> SchemaRef {
80
- Arc::clone(&self.schema)
81
- }
82
-
83
- fn table_type(&self) -> TableType {
84
- TableType::View
85
- }
86
-
87
- fn supports_filters_pushdown(
88
- &self,
89
- filters: &[&Expr],
90
- ) -> Result<Vec<TableProviderFilterPushDown>> {
91
- Ok(filters
92
- .iter()
93
- .map(|filter| {
94
- if parse_history_filter(filter, HistoryColumnStyle::Bare).is_some() {
95
- TableProviderFilterPushDown::Exact
96
- } else {
97
- TableProviderFilterPushDown::Unsupported
98
- }
99
- })
100
- .collect())
101
- }
102
-
103
- async fn scan(
104
- &self,
105
- _state: &dyn Session,
106
- projection: Option<&Vec<usize>>,
107
- filters: &[Expr],
108
- limit: Option<usize>,
109
- ) -> Result<Arc<dyn ExecutionPlan>> {
110
- let projected_schema = projected_schema(&self.schema, projection)?;
111
- Ok(Arc::new(LixStateHistoryScanExec::new(
112
- Arc::clone(&self.commit_graph),
113
- self.query_source.clone(),
114
- projected_schema,
115
- projection.cloned(),
116
- HistoryRoute::from_filters(filters, HistoryColumnStyle::Bare),
117
- limit,
118
- )))
119
- }
120
- }
121
-
122
- struct LixStateHistoryScanExec {
123
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
124
- query_source: SqlCommitStoreQuerySource,
125
- schema: SchemaRef,
126
- projection: Option<Vec<usize>>,
127
- route: HistoryRoute,
128
- limit: Option<usize>,
129
- properties: Arc<PlanProperties>,
130
- }
131
-
132
- impl std::fmt::Debug for LixStateHistoryScanExec {
133
- fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
134
- f.debug_struct("LixStateHistoryScanExec")
135
- .field("limit", &self.limit)
136
- .field("route", &self.route)
137
- .finish()
138
- }
139
- }
140
-
141
- impl LixStateHistoryScanExec {
142
- fn new(
143
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
144
- query_source: SqlCommitStoreQuerySource,
145
- schema: SchemaRef,
146
- projection: Option<Vec<usize>>,
147
- route: HistoryRoute,
148
- limit: Option<usize>,
149
- ) -> Self {
150
- let properties = PlanProperties::new(
151
- EquivalenceProperties::new(Arc::clone(&schema)),
152
- Partitioning::UnknownPartitioning(1),
153
- EmissionType::Incremental,
154
- Boundedness::Bounded,
155
- );
156
- Self {
157
- commit_graph,
158
- query_source,
159
- schema,
160
- projection,
161
- route,
162
- limit,
163
- properties: Arc::new(properties),
164
- }
165
- }
166
- }
167
-
168
- impl DisplayAs for LixStateHistoryScanExec {
169
- fn fmt_as(&self, t: DisplayFormatType, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
170
- match t {
171
- DisplayFormatType::Default | DisplayFormatType::Verbose => {
172
- write!(
173
- f,
174
- "LixStateHistoryScanExec(limit={:?}, route={:?})",
175
- self.limit, self.route
176
- )
177
- }
178
- DisplayFormatType::TreeRender => write!(f, "LixStateHistoryScanExec"),
179
- }
180
- }
181
- }
182
-
183
- impl ExecutionPlan for LixStateHistoryScanExec {
184
- fn name(&self) -> &str {
185
- "LixStateHistoryScanExec"
186
- }
187
-
188
- fn as_any(&self) -> &dyn Any {
189
- self
190
- }
191
-
192
- fn properties(&self) -> &Arc<PlanProperties> {
193
- &self.properties
194
- }
195
-
196
- fn children(&self) -> Vec<&Arc<dyn ExecutionPlan>> {
197
- Vec::new()
198
- }
199
-
200
- fn with_new_children(
201
- self: Arc<Self>,
202
- children: Vec<Arc<dyn ExecutionPlan>>,
203
- ) -> Result<Arc<dyn ExecutionPlan>> {
204
- if !children.is_empty() {
205
- return Err(DataFusionError::Execution(
206
- "LixStateHistoryScanExec does not accept children".to_string(),
207
- ));
208
- }
209
- Ok(self)
210
- }
211
-
212
- fn execute(
213
- &self,
214
- partition: usize,
215
- _context: Arc<TaskContext>,
216
- ) -> Result<SendableRecordBatchStream> {
217
- if partition != 0 {
218
- return Err(DataFusionError::Execution(format!(
219
- "LixStateHistoryScanExec only exposes one partition, got {partition}"
220
- )));
221
- }
222
-
223
- let commit_graph = Arc::clone(&self.commit_graph);
224
- let query_source = self.query_source.clone();
225
- let route = self.route.clone();
226
- let schema = Arc::clone(&self.schema);
227
- let stream_schema = Arc::clone(&schema);
228
- let limit = self.limit;
229
- let zero_column_projection = self
230
- .projection
231
- .as_ref()
232
- .is_some_and(|projection| projection.is_empty());
233
-
234
- let stream = stream::once(async move {
235
- let rows = if route.is_contradictory() {
236
- Vec::new()
237
- } else {
238
- load_state_history_rows(commit_graph, query_source, &route)
239
- .await
240
- .map_err(lix_error_to_datafusion_error)?
241
- };
242
- let rows = if let Some(limit) = limit {
243
- rows.into_iter().take(limit).collect::<Vec<_>>()
244
- } else {
245
- rows
246
- };
247
-
248
- let batch = if zero_column_projection {
249
- let options = RecordBatchOptions::new().with_row_count(Some(rows.len()));
250
- RecordBatch::try_new_with_options(Arc::clone(&stream_schema), vec![], &options)
251
- .map_err(|error| {
252
- DataFusionError::Execution(format!(
253
- "failed to build zero-column lix_state_history batch: {error}"
254
- ))
255
- })?
256
- } else {
257
- state_history_record_batch(Arc::clone(&stream_schema), &rows)?
258
- };
259
- Ok::<_, DataFusionError>(stream::iter(vec![Ok::<RecordBatch, DataFusionError>(
260
- batch,
261
- )]))
262
- })
263
- .try_flatten();
264
-
265
- Ok(Box::pin(RecordBatchStreamAdapter::new(schema, stream)))
266
- }
267
- }
268
-
269
- fn lix_state_history_schema() -> SchemaRef {
270
- Arc::new(Schema::new(vec![
271
- json_field("entity_id", false),
272
- Field::new("schema_key", DataType::Utf8, false),
273
- Field::new("file_id", DataType::Utf8, true),
274
- json_field("snapshot_content", true),
275
- json_field("metadata", true),
276
- Field::new("change_id", DataType::Utf8, false),
277
- Field::new("observed_commit_id", DataType::Utf8, false),
278
- Field::new("commit_created_at", DataType::Utf8, false),
279
- Field::new("start_commit_id", DataType::Utf8, false),
280
- Field::new("depth", DataType::Int64, false),
281
- ]))
282
- }
283
-
284
- fn projected_schema(base_schema: &SchemaRef, projection: Option<&Vec<usize>>) -> Result<SchemaRef> {
285
- let fields = match projection {
286
- Some(indices) => indices
287
- .iter()
288
- .map(|index| base_schema.field(*index).as_ref().clone())
289
- .collect::<Vec<_>>(),
290
- None => base_schema
291
- .fields()
292
- .iter()
293
- .map(|field| field.as_ref().clone())
294
- .collect::<Vec<_>>(),
295
- };
296
- Ok(Arc::new(Schema::new(fields)))
297
- }
298
-
299
- #[derive(Debug, Clone)]
300
- struct StateHistorySqlRow {
301
- entity_id: String,
302
- schema_key: String,
303
- file_id: Option<String>,
304
- snapshot_content: Option<String>,
305
- metadata: Option<String>,
306
- change_id: String,
307
- observed_commit_id: String,
308
- commit_created_at: String,
309
- start_commit_id: String,
310
- depth: i64,
311
- }
312
-
313
- fn state_history_record_batch(
314
- schema: SchemaRef,
315
- rows: &[StateHistorySqlRow],
316
- ) -> Result<RecordBatch> {
317
- let arrays = schema
318
- .fields()
319
- .iter()
320
- .map(|field| {
321
- Ok(match field.name().as_str() {
322
- "entity_id" => string_array(rows.iter().map(|row| Some(row.entity_id.as_str()))),
323
- "schema_key" => string_array(rows.iter().map(|row| Some(row.schema_key.as_str()))),
324
- "file_id" => string_array(rows.iter().map(|row| row.file_id.as_deref())),
325
- "snapshot_content" => {
326
- string_array(rows.iter().map(|row| row.snapshot_content.as_deref()))
327
- }
328
- "metadata" => Arc::new(StringArray::from(
329
- rows.iter()
330
- .map(|row| row.metadata.as_ref().map(serialize_row_metadata))
331
- .collect::<Vec<_>>(),
332
- )),
333
- "change_id" => string_array(rows.iter().map(|row| Some(row.change_id.as_str()))),
334
- "observed_commit_id" => {
335
- string_array(rows.iter().map(|row| Some(row.observed_commit_id.as_str())))
336
- }
337
- "commit_created_at" => {
338
- string_array(rows.iter().map(|row| Some(row.commit_created_at.as_str())))
339
- }
340
- "start_commit_id" => {
341
- string_array(rows.iter().map(|row| Some(row.start_commit_id.as_str())))
342
- }
343
- "depth" => Arc::new(Int64Array::from(
344
- rows.iter().map(|row| row.depth).collect::<Vec<_>>(),
345
- )) as ArrayRef,
346
- other => {
347
- return Err(DataFusionError::Execution(format!(
348
- "lix_state_history provider does not support projected column '{other}'"
349
- )))
350
- }
351
- })
352
- })
353
- .collect::<Result<Vec<_>>>()?;
354
- RecordBatch::try_new(schema, arrays).map_err(DataFusionError::from)
355
- }
356
-
357
- fn string_array<'a>(values: impl Iterator<Item = Option<&'a str>>) -> ArrayRef {
358
- Arc::new(StringArray::from(values.collect::<Vec<_>>())) as ArrayRef
359
- }
360
-
361
- async fn load_state_history_rows(
362
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
363
- query_source: SqlCommitStoreQuerySource,
364
- route: &HistoryRoute,
365
- ) -> Result<Vec<StateHistorySqlRow>, LixError> {
366
- let entries = load_history_entries(
367
- HistoryViewDescriptor {
368
- view_name: "lix_state_history",
369
- start_commit_column: "start_commit_id",
370
- },
371
- commit_graph,
372
- query_source.json_reader,
373
- route,
374
- Vec::new(),
375
- )
376
- .await?;
377
- let mut rows = entries
378
- .into_iter()
379
- .map(|entry| -> Result<StateHistorySqlRow, LixError> {
380
- Ok(StateHistorySqlRow {
381
- entity_id: entry.change.entity_id.as_json_array_text()?,
382
- schema_key: entry.change.schema_key,
383
- file_id: entry.change.file_id,
384
- snapshot_content: entry.change.snapshot_content,
385
- metadata: entry.change.metadata,
386
- change_id: entry.change.id,
387
- observed_commit_id: entry.observed_commit_id,
388
- commit_created_at: entry.commit_created_at,
389
- start_commit_id: entry.start_commit_id,
390
- depth: i64::from(entry.depth),
391
- })
392
- })
393
- .collect::<Result<Vec<_>, _>>()?;
394
-
395
- rows.sort_by(|left, right| {
396
- left.entity_id
397
- .cmp(&right.entity_id)
398
- .then(left.file_id.cmp(&right.file_id))
399
- .then(left.schema_key.cmp(&right.schema_key))
400
- .then(left.depth.cmp(&right.depth))
401
- .then(left.change_id.cmp(&right.change_id))
402
- });
403
- Ok(rows)
404
- }
405
-
406
- fn datafusion_error_to_lix_error(error: DataFusionError) -> LixError {
407
- super::error::datafusion_error_to_lix_error(error)
408
- }
409
-
410
- fn lix_error_to_datafusion_error(error: LixError) -> DataFusionError {
411
- super::error::lix_error_to_datafusion_error(error)
412
- }