@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
@@ -1,631 +0,0 @@
1
- use std::any::Any;
2
- use std::collections::{BTreeMap, BTreeSet};
3
- use std::sync::Arc;
4
-
5
- use async_trait::async_trait;
6
- use datafusion::arrow::array::{ArrayRef, BooleanArray, Int64Array, StringArray};
7
- use datafusion::arrow::datatypes::{DataType, Field, Schema, SchemaRef};
8
- use datafusion::arrow::record_batch::{RecordBatch, RecordBatchOptions};
9
- use datafusion::catalog::{Session, TableProvider};
10
- use datafusion::common::{DataFusionError, Result};
11
- use datafusion::datasource::TableType;
12
- use datafusion::execution::TaskContext;
13
- use datafusion::logical_expr::{Expr, TableProviderFilterPushDown};
14
- use datafusion::physical_expr::EquivalenceProperties;
15
- use datafusion::physical_plan::execution_plan::{Boundedness, EmissionType, PlanProperties};
16
- use datafusion::physical_plan::stream::RecordBatchStreamAdapter;
17
- use datafusion::physical_plan::{
18
- DisplayAs, DisplayFormatType, ExecutionPlan, Partitioning, SendableRecordBatchStream,
19
- };
20
- use futures_util::stream;
21
- use serde::Deserialize;
22
- use tokio::sync::Mutex;
23
-
24
- use crate::commit_graph::CommitGraphReader;
25
- use crate::serialize_row_metadata;
26
- use crate::LixError;
27
-
28
- use super::history_projection::{tombstone_identity_column_value, HistoryIdentityProjection};
29
- use super::history_route::{
30
- history_descriptor_event_matches, load_history_entries, parse_history_filter,
31
- HistoryColumnStyle, HistoryEntry, HistoryRoute, HistoryViewDescriptor, HISTORY_COL_CHANGE_ID,
32
- HISTORY_COL_COMMIT_CREATED_AT, HISTORY_COL_DEPTH, HISTORY_COL_ENTITY_ID, HISTORY_COL_FILE_ID,
33
- HISTORY_COL_METADATA, HISTORY_COL_OBSERVED_COMMIT_ID, HISTORY_COL_SCHEMA_KEY,
34
- HISTORY_COL_SNAPSHOT_CONTENT, HISTORY_COL_START_COMMIT_ID,
35
- };
36
- use super::result_metadata::json_field;
37
- use super::SqlCommitStoreQuerySource;
38
- use crate::commit_store::MaterializedChange;
39
-
40
- const DIRECTORY_DESCRIPTOR_SCHEMA_KEY: &str = "lix_directory_descriptor";
41
-
42
- pub(crate) async fn register_lix_directory_history_provider(
43
- session: &datafusion::prelude::SessionContext,
44
- commit_graph: Box<dyn CommitGraphReader>,
45
- query_source: SqlCommitStoreQuerySource,
46
- ) -> Result<(), LixError> {
47
- session
48
- .register_table(
49
- "lix_directory_history",
50
- Arc::new(LixDirectoryHistoryProvider::new(
51
- Arc::new(Mutex::new(commit_graph)),
52
- query_source,
53
- )),
54
- )
55
- .map_err(datafusion_error_to_lix_error)?;
56
- Ok(())
57
- }
58
-
59
- struct LixDirectoryHistoryProvider {
60
- schema: SchemaRef,
61
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
62
- query_source: SqlCommitStoreQuerySource,
63
- }
64
-
65
- impl std::fmt::Debug for LixDirectoryHistoryProvider {
66
- fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
67
- f.debug_struct("LixDirectoryHistoryProvider").finish()
68
- }
69
- }
70
-
71
- impl LixDirectoryHistoryProvider {
72
- fn new(
73
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
74
- query_source: SqlCommitStoreQuerySource,
75
- ) -> Self {
76
- Self {
77
- schema: lix_directory_history_schema(),
78
- commit_graph,
79
- query_source,
80
- }
81
- }
82
- }
83
-
84
- #[async_trait]
85
- impl TableProvider for LixDirectoryHistoryProvider {
86
- fn as_any(&self) -> &dyn Any {
87
- self
88
- }
89
-
90
- fn schema(&self) -> SchemaRef {
91
- Arc::clone(&self.schema)
92
- }
93
-
94
- fn table_type(&self) -> TableType {
95
- TableType::View
96
- }
97
-
98
- fn supports_filters_pushdown(
99
- &self,
100
- filters: &[&Expr],
101
- ) -> Result<Vec<TableProviderFilterPushDown>> {
102
- Ok(filters
103
- .iter()
104
- .map(|filter| {
105
- if parse_history_filter(filter, HistoryColumnStyle::Prefixed).is_some() {
106
- TableProviderFilterPushDown::Exact
107
- } else {
108
- TableProviderFilterPushDown::Unsupported
109
- }
110
- })
111
- .collect())
112
- }
113
-
114
- async fn scan(
115
- &self,
116
- _state: &dyn Session,
117
- projection: Option<&Vec<usize>>,
118
- filters: &[Expr],
119
- limit: Option<usize>,
120
- ) -> Result<Arc<dyn ExecutionPlan>> {
121
- Ok(Arc::new(LixDirectoryHistoryScanExec::new(
122
- Arc::clone(&self.commit_graph),
123
- self.query_source.clone(),
124
- projected_schema(&self.schema, projection)?,
125
- HistoryRoute::from_filters(filters, HistoryColumnStyle::Prefixed),
126
- limit,
127
- )))
128
- }
129
- }
130
-
131
- struct LixDirectoryHistoryScanExec {
132
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
133
- query_source: SqlCommitStoreQuerySource,
134
- schema: SchemaRef,
135
- route: HistoryRoute,
136
- limit: Option<usize>,
137
- properties: Arc<PlanProperties>,
138
- }
139
-
140
- impl std::fmt::Debug for LixDirectoryHistoryScanExec {
141
- fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
142
- f.debug_struct("LixDirectoryHistoryScanExec")
143
- .field("route", &self.route)
144
- .field("limit", &self.limit)
145
- .finish()
146
- }
147
- }
148
-
149
- impl LixDirectoryHistoryScanExec {
150
- fn new(
151
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
152
- query_source: SqlCommitStoreQuerySource,
153
- schema: SchemaRef,
154
- route: HistoryRoute,
155
- limit: Option<usize>,
156
- ) -> Self {
157
- let properties = PlanProperties::new(
158
- EquivalenceProperties::new(Arc::clone(&schema)),
159
- Partitioning::UnknownPartitioning(1),
160
- EmissionType::Incremental,
161
- Boundedness::Bounded,
162
- );
163
- Self {
164
- commit_graph,
165
- query_source,
166
- schema,
167
- route,
168
- limit,
169
- properties: Arc::new(properties),
170
- }
171
- }
172
- }
173
-
174
- impl DisplayAs for LixDirectoryHistoryScanExec {
175
- fn fmt_as(&self, t: DisplayFormatType, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
176
- match t {
177
- DisplayFormatType::Default | DisplayFormatType::Verbose => write!(
178
- f,
179
- "LixDirectoryHistoryScanExec(route={:?}, limit={:?})",
180
- self.route, self.limit
181
- ),
182
- DisplayFormatType::TreeRender => write!(f, "LixDirectoryHistoryScanExec"),
183
- }
184
- }
185
- }
186
-
187
- impl ExecutionPlan for LixDirectoryHistoryScanExec {
188
- fn name(&self) -> &str {
189
- "LixDirectoryHistoryScanExec"
190
- }
191
-
192
- fn as_any(&self) -> &dyn Any {
193
- self
194
- }
195
-
196
- fn properties(&self) -> &Arc<PlanProperties> {
197
- &self.properties
198
- }
199
-
200
- fn children(&self) -> Vec<&Arc<dyn ExecutionPlan>> {
201
- Vec::new()
202
- }
203
-
204
- fn with_new_children(
205
- self: Arc<Self>,
206
- children: Vec<Arc<dyn ExecutionPlan>>,
207
- ) -> Result<Arc<dyn ExecutionPlan>> {
208
- if !children.is_empty() {
209
- return Err(DataFusionError::Execution(
210
- "LixDirectoryHistoryScanExec does not accept children".to_string(),
211
- ));
212
- }
213
- Ok(self)
214
- }
215
-
216
- fn execute(
217
- &self,
218
- partition: usize,
219
- _context: Arc<TaskContext>,
220
- ) -> Result<SendableRecordBatchStream> {
221
- if partition != 0 {
222
- return Err(DataFusionError::Execution(format!(
223
- "LixDirectoryHistoryScanExec only exposes one partition, got {partition}"
224
- )));
225
- }
226
-
227
- let commit_graph = Arc::clone(&self.commit_graph);
228
- let query_source = self.query_source.clone();
229
- let schema = Arc::clone(&self.schema);
230
- let stream_schema = Arc::clone(&schema);
231
- let route = self.route.clone();
232
- let limit = self.limit;
233
- let fut = async move {
234
- let mut rows = load_directory_history_rows(commit_graph, query_source, &route)
235
- .await
236
- .map_err(lix_error_to_datafusion_error)?;
237
- if let Some(limit) = limit {
238
- rows.truncate(limit);
239
- }
240
- directory_history_record_batch(&stream_schema, &rows)
241
- .map_err(lix_error_to_datafusion_error)
242
- };
243
-
244
- Ok(Box::pin(RecordBatchStreamAdapter::new(
245
- schema,
246
- stream::once(fut),
247
- )))
248
- }
249
- }
250
-
251
- #[derive(Debug, Clone)]
252
- struct DirectoryHistoryRecord {
253
- id: String,
254
- parent_id: Option<String>,
255
- name: Option<String>,
256
- hidden: Option<bool>,
257
- entry: HistoryEntry,
258
- }
259
-
260
- #[derive(Debug, Clone)]
261
- struct DirectoryHistoryOutputRow {
262
- entity_id: String,
263
- id: String,
264
- path: Option<String>,
265
- parent_id: Option<String>,
266
- name: Option<String>,
267
- hidden: Option<bool>,
268
- descriptor_change: MaterializedChange,
269
- event: DirectoryHistoryEvent,
270
- }
271
-
272
- #[derive(Debug, Clone)]
273
- struct DirectoryHistoryEvent {
274
- directory_id: String,
275
- start_commit_id: String,
276
- depth: u32,
277
- change: MaterializedChange,
278
- observed_commit_id: String,
279
- commit_created_at: String,
280
- }
281
-
282
- #[derive(Debug, Deserialize)]
283
- struct DirectoryDescriptorSnapshot {
284
- id: String,
285
- parent_id: Option<String>,
286
- name: String,
287
- hidden: Option<bool>,
288
- }
289
-
290
- async fn load_directory_history_rows(
291
- commit_graph: Arc<Mutex<Box<dyn CommitGraphReader>>>,
292
- query_source: SqlCommitStoreQuerySource,
293
- route: &HistoryRoute,
294
- ) -> Result<Vec<DirectoryHistoryOutputRow>, LixError> {
295
- let event_route = route.traversal_only();
296
- let event_entries = load_history_entries(
297
- HistoryViewDescriptor {
298
- view_name: "lix_directory_history",
299
- start_commit_column: HISTORY_COL_START_COMMIT_ID,
300
- },
301
- Arc::clone(&commit_graph),
302
- query_source.json_reader.clone(),
303
- &event_route,
304
- vec![DIRECTORY_DESCRIPTOR_SCHEMA_KEY.to_string()],
305
- )
306
- .await?;
307
- let context_route = route.starts_only();
308
- let context_entries = load_history_entries(
309
- HistoryViewDescriptor {
310
- view_name: "lix_directory_history",
311
- start_commit_column: HISTORY_COL_START_COMMIT_ID,
312
- },
313
- commit_graph,
314
- query_source.json_reader,
315
- &context_route,
316
- vec![DIRECTORY_DESCRIPTOR_SCHEMA_KEY.to_string()],
317
- )
318
- .await?;
319
- let event_descriptors = parse_directory_history_records(&event_entries)?;
320
- let descriptors = parse_directory_history_records(&context_entries)?;
321
- let mut output = Vec::new();
322
-
323
- for descriptor in &event_descriptors {
324
- let event = directory_history_event_from_entry(&descriptor.id, &descriptor.entry);
325
- let Some(visible_descriptor) = nearest_directory_descriptor(&descriptors, &event) else {
326
- continue;
327
- };
328
- let path = if visible_descriptor.name.is_some() {
329
- resolve_directory_history_path(
330
- &visible_descriptor.id,
331
- &event.start_commit_id,
332
- event.depth,
333
- &descriptors,
334
- &mut BTreeMap::new(),
335
- &mut BTreeSet::new(),
336
- )
337
- } else {
338
- None
339
- };
340
- let id = tombstone_identity_column_value(
341
- "id",
342
- &visible_descriptor.id,
343
- HistoryIdentityProjection::SingleColumn { column: "id" },
344
- )?
345
- .and_then(|value| value.as_str().map(ToOwned::to_owned))
346
- .unwrap_or_else(|| visible_descriptor.id.clone());
347
- output.push(DirectoryHistoryOutputRow {
348
- entity_id: visible_descriptor.id.clone(),
349
- id,
350
- path,
351
- parent_id: visible_descriptor.parent_id.clone(),
352
- name: visible_descriptor.name.clone(),
353
- hidden: visible_descriptor.hidden,
354
- descriptor_change: visible_descriptor.entry.change.clone(),
355
- event,
356
- });
357
- }
358
- output.retain(|row| {
359
- let entity_id = entity_id_json_array(&row.entity_id).ok();
360
- route.matches_surface_row(
361
- DIRECTORY_DESCRIPTOR_SCHEMA_KEY,
362
- entity_id.as_deref().unwrap_or(&row.entity_id),
363
- None,
364
- row.event.depth,
365
- )
366
- });
367
-
368
- output.sort_by(|left, right| {
369
- left.entity_id
370
- .cmp(&right.entity_id)
371
- .then(left.event.start_commit_id.cmp(&right.event.start_commit_id))
372
- .then(left.event.depth.cmp(&right.event.depth))
373
- .then(
374
- left.event
375
- .observed_commit_id
376
- .cmp(&right.event.observed_commit_id),
377
- )
378
- .then(left.event.change.id.cmp(&right.event.change.id))
379
- });
380
- Ok(output)
381
- }
382
-
383
- fn parse_directory_history_records(
384
- entries: &[HistoryEntry],
385
- ) -> Result<Vec<DirectoryHistoryRecord>, LixError> {
386
- entries
387
- .iter()
388
- .filter(|entry| entry.change.schema_key == DIRECTORY_DESCRIPTOR_SCHEMA_KEY)
389
- .map(|entry| {
390
- let Some(snapshot_content) = entry.change.snapshot_content.as_deref() else {
391
- return Ok(DirectoryHistoryRecord {
392
- id: entry.change.entity_id.as_single_string_owned()?,
393
- parent_id: None,
394
- name: None,
395
- hidden: None,
396
- entry: entry.clone(),
397
- });
398
- };
399
- let snapshot: DirectoryDescriptorSnapshot = serde_json::from_str(snapshot_content)
400
- .map_err(|error| {
401
- LixError::new(
402
- "LIX_ERROR_UNKNOWN",
403
- format!("invalid lix_directory_descriptor history snapshot JSON: {error}"),
404
- )
405
- })?;
406
- Ok(DirectoryHistoryRecord {
407
- id: snapshot.id,
408
- parent_id: snapshot.parent_id,
409
- name: Some(snapshot.name),
410
- hidden: Some(snapshot.hidden.unwrap_or(false)),
411
- entry: entry.clone(),
412
- })
413
- })
414
- .collect()
415
- }
416
-
417
- fn directory_history_event_from_entry(
418
- directory_id: &str,
419
- entry: &HistoryEntry,
420
- ) -> DirectoryHistoryEvent {
421
- DirectoryHistoryEvent {
422
- directory_id: directory_id.to_string(),
423
- start_commit_id: entry.start_commit_id.clone(),
424
- depth: entry.depth,
425
- change: entry.change.clone(),
426
- observed_commit_id: entry.observed_commit_id.clone(),
427
- commit_created_at: entry.commit_created_at.clone(),
428
- }
429
- }
430
-
431
- fn nearest_directory_descriptor<'a>(
432
- descriptors: &'a [DirectoryHistoryRecord],
433
- event: &DirectoryHistoryEvent,
434
- ) -> Option<&'a DirectoryHistoryRecord> {
435
- descriptors
436
- .iter()
437
- .filter(|descriptor| {
438
- let exact_descriptor_event =
439
- history_descriptor_event_matches(&descriptor.entry, event.depth, &event.change.id);
440
- (exact_descriptor_event || descriptor.name.is_some())
441
- && descriptor.id == event.directory_id
442
- && descriptor.entry.start_commit_id == event.start_commit_id
443
- && descriptor.entry.depth >= event.depth
444
- })
445
- .min_by(|left, right| {
446
- left.entry
447
- .depth
448
- .cmp(&right.entry.depth)
449
- .then(left.entry.change.id.cmp(&right.entry.change.id))
450
- })
451
- }
452
-
453
- fn resolve_directory_history_path(
454
- directory_id: &str,
455
- start_commit_id: &str,
456
- target_depth: u32,
457
- directories: &[DirectoryHistoryRecord],
458
- cache: &mut BTreeMap<String, Option<String>>,
459
- visiting: &mut BTreeSet<String>,
460
- ) -> Option<String> {
461
- if let Some(path) = cache.get(directory_id) {
462
- return path.clone();
463
- }
464
- if !visiting.insert(directory_id.to_string()) {
465
- cache.insert(directory_id.to_string(), None);
466
- return None;
467
- }
468
- let directory = directories
469
- .iter()
470
- .filter(|directory| {
471
- directory.name.is_some()
472
- && directory.id == directory_id
473
- && directory.entry.start_commit_id == start_commit_id
474
- && directory.entry.depth >= target_depth
475
- })
476
- .min_by(|left, right| {
477
- left.entry
478
- .depth
479
- .cmp(&right.entry.depth)
480
- .then(left.entry.change.id.cmp(&right.entry.change.id))
481
- })?;
482
- let name = directory.name.as_ref()?;
483
- let path = match directory.parent_id.as_deref() {
484
- Some(parent_id) => {
485
- let parent_path = resolve_directory_history_path(
486
- parent_id,
487
- start_commit_id,
488
- target_depth,
489
- directories,
490
- cache,
491
- visiting,
492
- )?;
493
- format!("{parent_path}{name}/")
494
- }
495
- None => format!("/{name}/"),
496
- };
497
- visiting.remove(directory_id);
498
- cache.insert(directory_id.to_string(), Some(path.clone()));
499
- Some(path)
500
- }
501
-
502
- fn directory_history_record_batch(
503
- schema: &SchemaRef,
504
- rows: &[DirectoryHistoryOutputRow],
505
- ) -> Result<RecordBatch, LixError> {
506
- let columns = schema
507
- .fields()
508
- .iter()
509
- .map(|field| directory_history_column_array(field.name(), rows))
510
- .collect::<Result<Vec<_>, _>>()?;
511
- let options = RecordBatchOptions::new().with_row_count(Some(rows.len()));
512
- RecordBatch::try_new_with_options(Arc::clone(schema), columns, &options).map_err(|error| {
513
- LixError::new(
514
- "LIX_ERROR_UNKNOWN",
515
- format!("sql2 failed to build lix_directory_history record batch: {error}"),
516
- )
517
- })
518
- }
519
-
520
- fn directory_history_column_array(
521
- column_name: &str,
522
- rows: &[DirectoryHistoryOutputRow],
523
- ) -> Result<ArrayRef, LixError> {
524
- Ok(match column_name {
525
- "id" => string_array(rows.iter().map(|row| Some(row.id.as_str()))),
526
- "path" => string_array(rows.iter().map(|row| row.path.as_deref())),
527
- "parent_id" => string_array(rows.iter().map(|row| row.parent_id.as_deref())),
528
- "name" => string_array(rows.iter().map(|row| row.name.as_deref())),
529
- "hidden" => Arc::new(BooleanArray::from(
530
- rows.iter().map(|row| row.hidden).collect::<Vec<_>>(),
531
- )) as ArrayRef,
532
- HISTORY_COL_ENTITY_ID => Arc::new(StringArray::from(
533
- rows.iter()
534
- .map(|row| entity_id_json_array(&row.entity_id).map(Some))
535
- .collect::<std::result::Result<Vec<_>, _>>()?,
536
- )) as ArrayRef,
537
- HISTORY_COL_SCHEMA_KEY => {
538
- string_array(rows.iter().map(|_| Some(DIRECTORY_DESCRIPTOR_SCHEMA_KEY)))
539
- }
540
- HISTORY_COL_FILE_ID => string_array(rows.iter().map(|_| None)),
541
- HISTORY_COL_CHANGE_ID => {
542
- string_array(rows.iter().map(|row| Some(row.event.change.id.as_str())))
543
- }
544
- HISTORY_COL_SNAPSHOT_CONTENT => string_array(
545
- rows.iter()
546
- .map(|row| row.descriptor_change.snapshot_content.as_deref()),
547
- ),
548
- HISTORY_COL_METADATA => Arc::new(StringArray::from(
549
- rows.iter()
550
- .map(|row| {
551
- row.descriptor_change
552
- .metadata
553
- .as_ref()
554
- .map(serialize_row_metadata)
555
- })
556
- .collect::<Vec<_>>(),
557
- )),
558
- HISTORY_COL_OBSERVED_COMMIT_ID => string_array(
559
- rows.iter()
560
- .map(|row| Some(row.event.observed_commit_id.as_str())),
561
- ),
562
- HISTORY_COL_COMMIT_CREATED_AT => string_array(
563
- rows.iter()
564
- .map(|row| Some(row.event.commit_created_at.as_str())),
565
- ),
566
- HISTORY_COL_START_COMMIT_ID => string_array(
567
- rows.iter()
568
- .map(|row| Some(row.event.start_commit_id.as_str())),
569
- ),
570
- HISTORY_COL_DEPTH => Arc::new(Int64Array::from(
571
- rows.iter()
572
- .map(|row| i64::from(row.event.depth))
573
- .collect::<Vec<_>>(),
574
- )) as ArrayRef,
575
- other => {
576
- return Err(LixError::new(
577
- "LIX_ERROR_UNKNOWN",
578
- format!(
579
- "sql2 lix_directory_history provider does not support projected column '{other}'"
580
- ),
581
- ))
582
- }
583
- })
584
- }
585
-
586
- fn lix_directory_history_schema() -> SchemaRef {
587
- Arc::new(Schema::new(vec![
588
- Field::new("id", DataType::Utf8, false),
589
- Field::new("path", DataType::Utf8, true),
590
- Field::new("parent_id", DataType::Utf8, true),
591
- Field::new("name", DataType::Utf8, true),
592
- Field::new("hidden", DataType::Boolean, true),
593
- json_field(HISTORY_COL_ENTITY_ID, false),
594
- Field::new(HISTORY_COL_SCHEMA_KEY, DataType::Utf8, false),
595
- Field::new(HISTORY_COL_FILE_ID, DataType::Utf8, true),
596
- json_field(HISTORY_COL_SNAPSHOT_CONTENT, true),
597
- Field::new(HISTORY_COL_CHANGE_ID, DataType::Utf8, false),
598
- json_field(HISTORY_COL_METADATA, true),
599
- Field::new(HISTORY_COL_OBSERVED_COMMIT_ID, DataType::Utf8, false),
600
- Field::new(HISTORY_COL_COMMIT_CREATED_AT, DataType::Utf8, false),
601
- Field::new(HISTORY_COL_START_COMMIT_ID, DataType::Utf8, false),
602
- Field::new(HISTORY_COL_DEPTH, DataType::Int64, false),
603
- ]))
604
- }
605
-
606
- fn projected_schema(base_schema: &SchemaRef, projection: Option<&Vec<usize>>) -> Result<SchemaRef> {
607
- let Some(projection) = projection else {
608
- return Ok(Arc::clone(base_schema));
609
- };
610
- Ok(Arc::new(base_schema.project(projection)?))
611
- }
612
-
613
- fn string_array<'a>(values: impl Iterator<Item = Option<&'a str>>) -> ArrayRef {
614
- Arc::new(StringArray::from(values.collect::<Vec<_>>())) as ArrayRef
615
- }
616
-
617
- fn datafusion_error_to_lix_error(error: DataFusionError) -> LixError {
618
- super::error::datafusion_error_to_lix_error(error)
619
- }
620
-
621
- fn entity_id_json_array(entity_id: &str) -> Result<String, LixError> {
622
- serde_json::to_string(&[entity_id]).map_err(|error| {
623
- LixError::unknown(format!(
624
- "failed to encode history entity id as JSON: {error}"
625
- ))
626
- })
627
- }
628
-
629
- fn lix_error_to_datafusion_error(error: LixError) -> DataFusionError {
630
- super::error::lix_error_to_datafusion_error(error)
631
- }