iceberg/spec/
snapshot.rs

1// Licensed to the Apache Software Foundation (ASF) under one
2// or more contributor license agreements.  See the NOTICE file
3// distributed with this work for additional information
4// regarding copyright ownership.  The ASF licenses this file
5// to you under the Apache License, Version 2.0 (the
6// "License"); you may not use this file except in compliance
7// with the License.  You may obtain a copy of the License at
8//
9//   http://www.apache.org/licenses/LICENSE-2.0
10//
11// Unless required by applicable law or agreed to in writing,
12// software distributed under the License is distributed on an
13// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
14// KIND, either express or implied.  See the License for the
15// specific language governing permissions and limitations
16// under the License.
17
18/*!
19 * Snapshots
20 */
21use std::collections::HashMap;
22use std::sync::Arc;
23
24use chrono::{DateTime, Utc};
25use serde::{Deserialize, Serialize};
26use typed_builder::TypedBuilder;
27
28use super::table_metadata::SnapshotLog;
29use crate::error::{Result, timestamp_ms_to_utc};
30use crate::io::FileIO;
31use crate::spec::{ManifestList, SchemaId, SchemaRef, TableMetadata};
32use crate::{Error, ErrorKind};
33
34/// The ref name of the main branch of the table.
35pub const MAIN_BRANCH: &str = "main";
36
37/// Reference to [`Snapshot`].
38pub type SnapshotRef = Arc<Snapshot>;
39#[derive(Debug, Default, Serialize, Deserialize, PartialEq, Eq, Clone)]
40#[serde(rename_all = "lowercase")]
41/// The operation field is used by some operations, like snapshot expiration, to skip processing certain snapshots.
42pub enum Operation {
43    /// Only data files were added and no files were removed.
44    #[default]
45    Append,
46    /// Data and delete files were added and removed without changing table data;
47    /// i.e., compaction, changing the data file format, or relocating data files.
48    Replace,
49    /// Data and delete files were added and removed in a logical overwrite operation.
50    Overwrite,
51    /// Data files were removed and their contents logically deleted and/or delete files were added to delete rows.
52    Delete,
53}
54
55impl Operation {
56    /// Returns the string representation (lowercase) of the operation.
57    pub fn as_str(&self) -> &str {
58        match self {
59            Operation::Append => "append",
60            Operation::Replace => "replace",
61            Operation::Overwrite => "overwrite",
62            Operation::Delete => "delete",
63        }
64    }
65}
66
67#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, Clone)]
68/// Summarises the changes in the snapshot.
69pub struct Summary {
70    /// The type of operation in the snapshot
71    pub operation: Operation,
72    /// Other summary data.
73    #[serde(flatten)]
74    pub additional_properties: HashMap<String, String>,
75}
76
77#[derive(Debug, PartialEq, Eq, Clone)]
78/// Row range of a snapshot, contains first_row_id and added_rows_count.
79pub struct SnapshotRowRange {
80    /// The first _row_id assigned to the first row in the first data file in the first manifest.
81    pub first_row_id: u64,
82    /// The upper bound of the number of rows with assigned row IDs
83    pub added_rows: u64,
84}
85
86#[derive(Debug, PartialEq, Eq, Clone, TypedBuilder)]
87#[builder(field_defaults(setter(prefix = "with_")))]
88/// A snapshot represents the state of a table at some time and is used to access the complete set of data files in the table.
89pub struct Snapshot {
90    /// A unique long ID
91    pub(crate) snapshot_id: i64,
92    /// The snapshot ID of the snapshot’s parent.
93    /// Omitted for any snapshot with no parent
94    #[builder(default = None)]
95    pub(crate) parent_snapshot_id: Option<i64>,
96    /// A monotonically increasing long that tracks the order of
97    /// changes to a table.
98    pub(crate) sequence_number: i64,
99    /// A timestamp when the snapshot was created, used for garbage
100    /// collection and table inspection
101    pub(crate) timestamp_ms: i64,
102    /// The location of a manifest list for this snapshot that
103    /// tracks manifest files with additional metadata.
104    /// Currently we only support manifest list file, and manifest files are not supported.
105    #[builder(setter(into))]
106    pub(crate) manifest_list: String,
107    /// A string map that summarizes the snapshot changes, including operation.
108    pub(crate) summary: Summary,
109    /// ID of the table’s current schema when the snapshot was created.
110    #[builder(setter(strip_option(fallback = schema_id_opt)), default = None)]
111    pub(crate) schema_id: Option<SchemaId>,
112    /// Encryption Key ID
113    #[builder(default)]
114    pub(crate) encryption_key_id: Option<String>,
115    /// Row range of this snapshot, required when the table version supports row lineage.
116    /// Specify as a tuple of (first_row_id, added_rows_count)
117    #[builder(default, setter(!strip_option, transform = |first_row_id: u64, added_rows: u64| Some(SnapshotRowRange { first_row_id, added_rows })))]
118    // This is specified as a struct instead of two separate fields to ensure that both fields are either set or not set.
119    // The java implementations uses two separate fields, then sets `added_row_counts` to Null if `first_row_id` is set to Null.
120    // It throws an error if `added_row_counts` is set but `first_row_id` is not set, or if either of the two is negative.
121    // We handle all cases infallible using the rust type system.
122    pub(crate) row_range: Option<SnapshotRowRange>,
123}
124
125impl Snapshot {
126    /// Get the id of the snapshot
127    #[inline]
128    pub fn snapshot_id(&self) -> i64 {
129        self.snapshot_id
130    }
131
132    /// Get parent snapshot id.
133    #[inline]
134    pub fn parent_snapshot_id(&self) -> Option<i64> {
135        self.parent_snapshot_id
136    }
137
138    /// Get sequence_number of the snapshot. Is 0 for Iceberg V1 tables.
139    #[inline]
140    pub fn sequence_number(&self) -> i64 {
141        self.sequence_number
142    }
143    /// Get location of manifest_list file
144    #[inline]
145    pub fn manifest_list(&self) -> &str {
146        &self.manifest_list
147    }
148
149    /// Get summary of the snapshot
150    #[inline]
151    pub fn summary(&self) -> &Summary {
152        &self.summary
153    }
154    /// Get the timestamp of when the snapshot was created
155    #[inline]
156    pub fn timestamp(&self) -> Result<DateTime<Utc>> {
157        timestamp_ms_to_utc(self.timestamp_ms)
158    }
159
160    /// Get the timestamp of when the snapshot was created in milliseconds
161    #[inline]
162    pub fn timestamp_ms(&self) -> i64 {
163        self.timestamp_ms
164    }
165
166    /// Get the schema id of this snapshot.
167    #[inline]
168    pub fn schema_id(&self) -> Option<SchemaId> {
169        self.schema_id
170    }
171
172    /// Get the schema of this snapshot.
173    pub fn schema(&self, table_metadata: &TableMetadata) -> Result<SchemaRef> {
174        Ok(match self.schema_id() {
175            Some(schema_id) => table_metadata
176                .schema_by_id(schema_id)
177                .ok_or_else(|| {
178                    Error::new(
179                        ErrorKind::DataInvalid,
180                        format!("Schema with id {schema_id} not found"),
181                    )
182                })?
183                .clone(),
184            None => table_metadata.current_schema().clone(),
185        })
186    }
187
188    /// Get parent snapshot.
189    #[cfg(test)]
190    pub(crate) fn parent_snapshot(&self, table_metadata: &TableMetadata) -> Option<SnapshotRef> {
191        match self.parent_snapshot_id {
192            Some(id) => table_metadata.snapshot_by_id(id).cloned(),
193            None => None,
194        }
195    }
196
197    /// Load manifest list.
198    pub async fn load_manifest_list(
199        &self,
200        file_io: &FileIO,
201        table_metadata: &TableMetadata,
202    ) -> Result<ManifestList> {
203        let manifest_list_content = file_io.new_input(&self.manifest_list)?.read().await?;
204        ManifestList::parse_with_version(
205            &manifest_list_content,
206            // TODO: You don't really need the version since you could just project any Avro in
207            // the version that you'd like to get (probably always the latest)
208            table_metadata.format_version(),
209        )
210    }
211
212    #[allow(dead_code)]
213    pub(crate) fn log(&self) -> SnapshotLog {
214        SnapshotLog {
215            timestamp_ms: self.timestamp_ms,
216            snapshot_id: self.snapshot_id,
217        }
218    }
219
220    /// The row-id of the first newly added row in this snapshot. All rows added in this snapshot will
221    /// have a row-id assigned to them greater than this value. All rows with a row-id less than this
222    /// value were created in a snapshot that was added to the table (but not necessarily committed to
223    /// this branch) in the past.
224    ///
225    /// This field is optional but is required when the table version supports row lineage.
226    pub fn first_row_id(&self) -> Option<u64> {
227        self.row_range.as_ref().map(|r| r.first_row_id)
228    }
229
230    /// The total number of newly added rows in this snapshot. It should be the summation of {@link
231    /// ManifestFile#ADDED_ROWS_COUNT} for every manifest added in this snapshot.
232    ///
233    /// This field is optional but is required when the table version supports row lineage.
234    pub fn added_rows_count(&self) -> Option<u64> {
235        self.row_range.as_ref().map(|r| r.added_rows)
236    }
237
238    /// Returns the row range of this snapshot, if available.
239    /// This is a tuple containing (first_row_id, added_rows_count).
240    pub fn row_range(&self) -> Option<(u64, u64)> {
241        self.row_range
242            .as_ref()
243            .map(|r| (r.first_row_id, r.added_rows))
244    }
245
246    /// Get encryption key id, if available.
247    pub fn encryption_key_id(&self) -> Option<&str> {
248        self.encryption_key_id.as_deref()
249    }
250}
251
252pub(super) mod _serde {
253    /// This is a helper module that defines types to help with serialization/deserialization.
254    /// For deserialization the input first gets read into either the [SnapshotV1] or [SnapshotV2] struct
255    /// and then converted into the [Snapshot] struct. Serialization works the other way around.
256    /// [SnapshotV1] and [SnapshotV2] are internal struct that are only used for serialization and deserialization.
257    use std::collections::HashMap;
258
259    use serde::{Deserialize, Serialize};
260
261    use super::{Operation, Snapshot, Summary};
262    use crate::spec::SchemaId;
263    use crate::spec::snapshot::SnapshotRowRange;
264    use crate::{Error, ErrorKind};
265
266    #[derive(Debug, Serialize, Deserialize, PartialEq, Eq)]
267    #[serde(rename_all = "kebab-case")]
268    /// Defines the structure of a v3 snapshot for serialization/deserialization
269    pub(crate) struct SnapshotV3 {
270        pub snapshot_id: i64,
271        #[serde(skip_serializing_if = "Option::is_none")]
272        pub parent_snapshot_id: Option<i64>,
273        pub sequence_number: i64,
274        pub timestamp_ms: i64,
275        pub manifest_list: String,
276        pub summary: Summary,
277        #[serde(skip_serializing_if = "Option::is_none")]
278        pub schema_id: Option<SchemaId>,
279        #[serde(skip_serializing_if = "Option::is_none")]
280        pub first_row_id: Option<u64>,
281        #[serde(skip_serializing_if = "Option::is_none")]
282        pub added_rows: Option<u64>,
283        #[serde(skip_serializing_if = "Option::is_none")]
284        pub key_id: Option<String>,
285    }
286
287    #[derive(Debug, Serialize, Deserialize, PartialEq, Eq)]
288    #[serde(rename_all = "kebab-case")]
289    /// Defines the structure of a v2 snapshot for serialization/deserialization
290    pub(crate) struct SnapshotV2 {
291        pub snapshot_id: i64,
292        #[serde(skip_serializing_if = "Option::is_none")]
293        pub parent_snapshot_id: Option<i64>,
294        pub sequence_number: i64,
295        pub timestamp_ms: i64,
296        pub manifest_list: String,
297        pub summary: Summary,
298        #[serde(skip_serializing_if = "Option::is_none")]
299        pub schema_id: Option<SchemaId>,
300    }
301
302    #[derive(Debug, Serialize, Deserialize, PartialEq, Eq)]
303    #[serde(rename_all = "kebab-case")]
304    /// Defines the structure of a v1 snapshot for serialization/deserialization
305    pub(crate) struct SnapshotV1 {
306        pub snapshot_id: i64,
307        #[serde(skip_serializing_if = "Option::is_none")]
308        pub parent_snapshot_id: Option<i64>,
309        pub timestamp_ms: i64,
310        #[serde(skip_serializing_if = "Option::is_none")]
311        pub manifest_list: Option<String>,
312        #[serde(skip_serializing_if = "Option::is_none")]
313        pub manifests: Option<Vec<String>>,
314        #[serde(skip_serializing_if = "Option::is_none")]
315        pub summary: Option<Summary>,
316        #[serde(skip_serializing_if = "Option::is_none")]
317        pub schema_id: Option<SchemaId>,
318    }
319
320    impl From<SnapshotV3> for Snapshot {
321        fn from(s: SnapshotV3) -> Self {
322            Snapshot {
323                snapshot_id: s.snapshot_id,
324                parent_snapshot_id: s.parent_snapshot_id,
325                sequence_number: s.sequence_number,
326                timestamp_ms: s.timestamp_ms,
327                manifest_list: s.manifest_list,
328                summary: s.summary,
329                schema_id: s.schema_id,
330                encryption_key_id: s.key_id,
331                row_range: match (s.first_row_id, s.added_rows) {
332                    (Some(first_row_id), Some(added_rows)) => Some(SnapshotRowRange {
333                        first_row_id,
334                        added_rows,
335                    }),
336                    _ => None,
337                },
338            }
339        }
340    }
341
342    impl TryFrom<Snapshot> for SnapshotV3 {
343        type Error = Error;
344
345        fn try_from(s: Snapshot) -> Result<Self, Self::Error> {
346            let (first_row_id, added_rows) = match s.row_range {
347                Some(row_range) => (Some(row_range.first_row_id), Some(row_range.added_rows)),
348                None => (None, None),
349            };
350
351            Ok(SnapshotV3 {
352                snapshot_id: s.snapshot_id,
353                parent_snapshot_id: s.parent_snapshot_id,
354                sequence_number: s.sequence_number,
355                timestamp_ms: s.timestamp_ms,
356                manifest_list: s.manifest_list,
357                summary: s.summary,
358                schema_id: s.schema_id,
359                first_row_id,
360                added_rows,
361                key_id: s.encryption_key_id,
362            })
363        }
364    }
365
366    impl From<SnapshotV2> for Snapshot {
367        fn from(v2: SnapshotV2) -> Self {
368            Snapshot {
369                snapshot_id: v2.snapshot_id,
370                parent_snapshot_id: v2.parent_snapshot_id,
371                sequence_number: v2.sequence_number,
372                timestamp_ms: v2.timestamp_ms,
373                manifest_list: v2.manifest_list,
374                summary: v2.summary,
375                schema_id: v2.schema_id,
376                encryption_key_id: None,
377                row_range: None,
378            }
379        }
380    }
381
382    impl From<Snapshot> for SnapshotV2 {
383        fn from(v2: Snapshot) -> Self {
384            SnapshotV2 {
385                snapshot_id: v2.snapshot_id,
386                parent_snapshot_id: v2.parent_snapshot_id,
387                sequence_number: v2.sequence_number,
388                timestamp_ms: v2.timestamp_ms,
389                manifest_list: v2.manifest_list,
390                summary: v2.summary,
391                schema_id: v2.schema_id,
392            }
393        }
394    }
395
396    impl TryFrom<SnapshotV1> for Snapshot {
397        type Error = Error;
398
399        fn try_from(v1: SnapshotV1) -> Result<Self, Self::Error> {
400            Ok(Snapshot {
401                snapshot_id: v1.snapshot_id,
402                parent_snapshot_id: v1.parent_snapshot_id,
403                sequence_number: 0,
404                timestamp_ms: v1.timestamp_ms,
405                manifest_list: match (v1.manifest_list, v1.manifests) {
406                    (Some(file), None) => file,
407                    (Some(_), Some(_)) => {
408                        return Err(Error::new(
409                            ErrorKind::DataInvalid,
410                            "Invalid v1 snapshot, when manifest list provided, manifest files should be omitted",
411                        ));
412                    }
413                    (None, _) => {
414                        return Err(Error::new(
415                            ErrorKind::DataInvalid,
416                            "Unsupported v1 snapshot, only manifest list is supported",
417                        ));
418                    }
419                },
420                summary: v1.summary.unwrap_or(Summary {
421                    operation: Operation::default(),
422                    additional_properties: HashMap::new(),
423                }),
424                schema_id: v1.schema_id,
425                encryption_key_id: None,
426                row_range: None,
427            })
428        }
429    }
430
431    impl From<Snapshot> for SnapshotV1 {
432        fn from(v2: Snapshot) -> Self {
433            SnapshotV1 {
434                snapshot_id: v2.snapshot_id,
435                parent_snapshot_id: v2.parent_snapshot_id,
436                timestamp_ms: v2.timestamp_ms,
437                manifest_list: Some(v2.manifest_list),
438                summary: Some(v2.summary),
439                schema_id: v2.schema_id,
440                manifests: None,
441            }
442        }
443    }
444}
445
446#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, Clone)]
447#[serde(rename_all = "kebab-case")]
448/// Iceberg tables keep track of branches and tags using snapshot references.
449pub struct SnapshotReference {
450    /// A reference’s snapshot ID. The tagged snapshot or latest snapshot of a branch.
451    pub snapshot_id: i64,
452    #[serde(flatten)]
453    /// Snapshot retention policy
454    pub retention: SnapshotRetention,
455}
456
457impl SnapshotReference {
458    /// Returns true if the snapshot reference is a branch.
459    pub fn is_branch(&self) -> bool {
460        matches!(self.retention, SnapshotRetention::Branch { .. })
461    }
462}
463
464impl SnapshotReference {
465    /// Create new snapshot reference
466    pub fn new(snapshot_id: i64, retention: SnapshotRetention) -> Self {
467        SnapshotReference {
468            snapshot_id,
469            retention,
470        }
471    }
472}
473
474#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, Clone)]
475#[serde(rename_all = "lowercase", tag = "type")]
476/// The snapshot expiration procedure removes snapshots from table metadata and applies the table’s retention policy.
477pub enum SnapshotRetention {
478    #[serde(rename_all = "kebab-case")]
479    /// Branches are mutable named references that can be updated by committing a new snapshot as
480    /// the branch’s referenced snapshot using the Commit Conflict Resolution and Retry procedures.
481    Branch {
482        /// A positive number for the minimum number of snapshots to keep in a branch while expiring snapshots.
483        /// Defaults to table property history.expire.min-snapshots-to-keep.
484        #[serde(skip_serializing_if = "Option::is_none")]
485        min_snapshots_to_keep: Option<i32>,
486        /// A positive number for the max age of snapshots to keep when expiring, including the latest snapshot.
487        /// Defaults to table property history.expire.max-snapshot-age-ms.
488        #[serde(skip_serializing_if = "Option::is_none")]
489        max_snapshot_age_ms: Option<i64>,
490        /// For snapshot references except the main branch, a positive number for the max age of the snapshot reference to keep while expiring snapshots.
491        /// Defaults to table property history.expire.max-ref-age-ms. The main branch never expires.
492        #[serde(skip_serializing_if = "Option::is_none")]
493        max_ref_age_ms: Option<i64>,
494    },
495    #[serde(rename_all = "kebab-case")]
496    /// Tags are labels for individual snapshots.
497    Tag {
498        /// For snapshot references except the main branch, a positive number for the max age of the snapshot reference to keep while expiring snapshots.
499        /// Defaults to table property history.expire.max-ref-age-ms. The main branch never expires.
500        #[serde(skip_serializing_if = "Option::is_none")]
501        max_ref_age_ms: Option<i64>,
502    },
503}
504
505impl SnapshotRetention {
506    /// Create a new branch retention policy
507    pub fn branch(
508        min_snapshots_to_keep: Option<i32>,
509        max_snapshot_age_ms: Option<i64>,
510        max_ref_age_ms: Option<i64>,
511    ) -> Self {
512        SnapshotRetention::Branch {
513            min_snapshots_to_keep,
514            max_snapshot_age_ms,
515            max_ref_age_ms,
516        }
517    }
518}
519
520#[cfg(test)]
521mod tests {
522    use std::collections::HashMap;
523
524    use chrono::{TimeZone, Utc};
525
526    use crate::spec::TableMetadata;
527    use crate::spec::snapshot::_serde::SnapshotV1;
528    use crate::spec::snapshot::{Operation, Snapshot, Summary};
529
530    #[test]
531    fn schema() {
532        let record = r#"
533        {
534            "snapshot-id": 3051729675574597004,
535            "timestamp-ms": 1515100955770,
536            "summary": {
537                "operation": "append"
538            },
539            "manifest-list": "s3://b/wh/.../s1.avro",
540            "schema-id": 0
541        }
542        "#;
543
544        let result: Snapshot = serde_json::from_str::<SnapshotV1>(record)
545            .unwrap()
546            .try_into()
547            .unwrap();
548        assert_eq!(3051729675574597004, result.snapshot_id());
549        assert_eq!(
550            Utc.timestamp_millis_opt(1515100955770).unwrap(),
551            result.timestamp().unwrap()
552        );
553        assert_eq!(1515100955770, result.timestamp_ms());
554        assert_eq!(
555            Summary {
556                operation: Operation::Append,
557                additional_properties: HashMap::new()
558            },
559            *result.summary()
560        );
561        assert_eq!("s3://b/wh/.../s1.avro".to_string(), *result.manifest_list());
562    }
563
564    #[test]
565    fn test_snapshot_v1_to_v2_projection() {
566        use crate::spec::snapshot::_serde::SnapshotV1;
567
568        // Create a V1 snapshot (without sequence-number field)
569        let v1_snapshot = SnapshotV1 {
570            snapshot_id: 1234567890,
571            parent_snapshot_id: Some(987654321),
572            timestamp_ms: 1515100955770,
573            manifest_list: Some("s3://bucket/manifest-list.avro".to_string()),
574            manifests: None, // V1 can have either manifest_list or manifests, but not both
575            summary: Some(Summary {
576                operation: Operation::Append,
577                additional_properties: HashMap::from([
578                    ("added-files".to_string(), "5".to_string()),
579                    ("added-records".to_string(), "100".to_string()),
580                ]),
581            }),
582            schema_id: Some(1),
583        };
584
585        // Convert V1 to V2 - this should apply defaults for missing V2 fields
586        let v2_snapshot: Snapshot = v1_snapshot.try_into().unwrap();
587
588        // Verify V1→V2 projection defaults are applied correctly
589        assert_eq!(
590            v2_snapshot.sequence_number(),
591            0,
592            "V1 snapshot sequence_number should default to 0"
593        );
594
595        // Verify other fields are preserved correctly during conversion
596        assert_eq!(v2_snapshot.snapshot_id(), 1234567890);
597        assert_eq!(v2_snapshot.parent_snapshot_id(), Some(987654321));
598        assert_eq!(v2_snapshot.timestamp_ms(), 1515100955770);
599        assert_eq!(
600            v2_snapshot.manifest_list(),
601            "s3://bucket/manifest-list.avro"
602        );
603        assert_eq!(v2_snapshot.schema_id(), Some(1));
604        assert_eq!(v2_snapshot.summary().operation, Operation::Append);
605        assert_eq!(
606            v2_snapshot
607                .summary()
608                .additional_properties
609                .get("added-files"),
610            Some(&"5".to_string())
611        );
612    }
613
614    #[test]
615    fn test_v1_snapshot_with_manifest_list_and_manifests() {
616        {
617            let metadata = r#"
618    {
619        "format-version": 1,
620        "table-uuid": "d20125c8-7284-442c-9aea-15fee620737c",
621        "location": "s3://bucket/test/location",
622        "last-updated-ms": 1700000000000,
623        "last-column-id": 1,
624        "schema": {
625            "type": "struct",
626            "fields": [
627                {"id": 1, "name": "x", "required": true, "type": "long"}
628            ]
629        },
630        "partition-spec": [],
631        "properties": {},
632        "current-snapshot-id": 111111111,
633        "snapshots": [
634            {
635                "snapshot-id": 111111111,
636                "timestamp-ms": 1600000000000,
637                "summary": {"operation": "append"},
638                "manifest-list": "s3://bucket/metadata/snap-123.avro",
639                "manifests": ["s3://bucket/metadata/manifest-1.avro"]
640            }
641        ]
642    }
643    "#;
644
645            let result_both_manifest_list_and_manifest_set =
646                serde_json::from_str::<TableMetadata>(metadata);
647            assert!(result_both_manifest_list_and_manifest_set.is_err());
648            assert_eq!(
649                result_both_manifest_list_and_manifest_set
650                    .unwrap_err()
651                    .to_string(),
652                "DataInvalid => Invalid v1 snapshot, when manifest list provided, manifest files should be omitted"
653            )
654        }
655
656        {
657            let metadata = r#"
658    {
659        "format-version": 1,
660        "table-uuid": "d20125c8-7284-442c-9aea-15fee620737c",
661        "location": "s3://bucket/test/location",
662        "last-updated-ms": 1700000000000,
663        "last-column-id": 1,
664        "schema": {
665            "type": "struct",
666            "fields": [
667                {"id": 1, "name": "x", "required": true, "type": "long"}
668            ]
669        },
670        "partition-spec": [],
671        "properties": {},
672        "current-snapshot-id": 111111111,
673        "snapshots": [
674            {
675                "snapshot-id": 111111111,
676                "timestamp-ms": 1600000000000,
677                "summary": {"operation": "append"},
678                "manifests": ["s3://bucket/metadata/manifest-1.avro"]
679            }
680        ]
681    }
682    "#;
683            let result_missing_manifest_list = serde_json::from_str::<TableMetadata>(metadata);
684            assert!(result_missing_manifest_list.is_err());
685            assert_eq!(
686                result_missing_manifest_list.unwrap_err().to_string(),
687                "DataInvalid => Unsupported v1 snapshot, only manifest list is supported"
688            )
689        }
690    }
691
692    #[test]
693    fn test_snapshot_v1_to_v2_with_missing_summary() {
694        use crate::spec::snapshot::_serde::SnapshotV1;
695
696        // Create a V1 snapshot without summary (should get default)
697        let v1_snapshot = SnapshotV1 {
698            snapshot_id: 1111111111,
699            parent_snapshot_id: None,
700            timestamp_ms: 1515100955770,
701            manifest_list: Some("s3://bucket/manifest-list.avro".to_string()),
702            manifests: None,
703            summary: None, // V1 summary is optional
704            schema_id: None,
705        };
706
707        // Convert V1 to V2 - this should apply default summary
708        let v2_snapshot: Snapshot = v1_snapshot.try_into().unwrap();
709
710        // Verify defaults are applied correctly
711        assert_eq!(
712            v2_snapshot.sequence_number(),
713            0,
714            "V1 snapshot sequence_number should default to 0"
715        );
716        assert_eq!(
717            v2_snapshot.summary().operation,
718            Operation::Append,
719            "Missing V1 summary should default to Append operation"
720        );
721        assert!(
722            v2_snapshot.summary().additional_properties.is_empty(),
723            "Default summary should have empty additional_properties"
724        );
725
726        // Verify other fields
727        assert_eq!(v2_snapshot.snapshot_id(), 1111111111);
728        assert_eq!(v2_snapshot.parent_snapshot_id(), None);
729        assert_eq!(v2_snapshot.schema_id(), None);
730    }
731}
iceberg/spec/snapshot.rs

iceberg/spec/
snapshot.rs
