This is an automated email from the ASF dual-hosted git repository.
stevenwu pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/iceberg.git
The following commit(s) were added to refs/heads/main by this push:
new cab0decbb0 REST Spec: Add row lineage fields (#13010)
cab0decbb0 is described below
commit cab0decbb0e32bf314039e30807eb033c50665d5
Author: Prashant Singh <35593236+singhpk...@users.noreply.github.com>
AuthorDate: Tue May 27 14:50:22 2025 -0700
REST Spec: Add row lineage fields (#13010)
Co-authored-by: Prashant Singh <prashant.si...@snowflake.com>
---
api/src/main/java/org/apache/iceberg/DataFile.java | 6 +++++-
open-api/rest-catalog-open-api.py | 15 +++++++++++++++
open-api/rest-catalog-open-api.yaml | 12 ++++++++++++
3 files changed, 32 insertions(+), 1 deletion(-)
diff --git a/api/src/main/java/org/apache/iceberg/DataFile.java
b/api/src/main/java/org/apache/iceberg/DataFile.java
index 58742e18d7..6433822834 100644
--- a/api/src/main/java/org/apache/iceberg/DataFile.java
+++ b/api/src/main/java/org/apache/iceberg/DataFile.java
@@ -99,7 +99,11 @@ public interface DataFile extends ContentFile<DataFile> {
optional(140, "sort_order_id", IntegerType.get(), "Sort order ID");
Types.NestedField SPEC_ID = optional(141, "spec_id", IntegerType.get(),
"Partition spec ID");
Types.NestedField FIRST_ROW_ID =
- optional(142, "first_row_id", LongType.get(), "Starting row ID to assign
to new rows");
+ optional(
+ 142,
+ "first_row_id",
+ LongType.get(),
+ "The first row ID assigned to the first row in the data file");
Types.NestedField REFERENCED_DATA_FILE =
optional(
143,
diff --git a/open-api/rest-catalog-open-api.py
b/open-api/rest-catalog-open-api.py
index 20674f4564..c01ce71275 100644
--- a/open-api/rest-catalog-open-api.py
+++ b/open-api/rest-catalog-open-api.py
@@ -215,6 +215,11 @@ class Snapshot(BaseModel):
alias='manifest-list',
description="Location of the snapshot's manifest list file",
)
+ first_row_id: Optional[int] = Field(
+ None,
+ alias='first-row-id',
+ description='The first _row_id assigned to the first row in the first
data file in the first manifest',
+ )
summary: Summary
schema_id: Optional[int] = Field(None, alias='schema-id')
@@ -1004,6 +1009,11 @@ class ValueMap(BaseModel):
class DataFile(ContentFile):
content: str = Field(..., const=True)
+ first_row_id: Optional[int] = Field(
+ None,
+ alias='first-row-id',
+ description='The first row ID assigned to the first row in the data
file',
+ )
column_sizes: Optional[CountMap] = Field(
None,
alias='column-sizes',
@@ -1141,6 +1151,11 @@ class TableMetadata(BaseModel):
table_uuid: str = Field(..., alias='table-uuid')
location: Optional[str] = None
last_updated_ms: Optional[int] = Field(None, alias='last-updated-ms')
+ next_row_id: Optional[int] = Field(
+ None,
+ alias='next-row-id',
+ description="A long higher than all assigned row IDs; the next
snapshot's first-row-id.",
+ )
properties: Optional[Dict[str, str]] = None
schemas: Optional[List[Schema]] = None
current_schema_id: Optional[int] = Field(None, alias='current-schema-id')
diff --git a/open-api/rest-catalog-open-api.yaml
b/open-api/rest-catalog-open-api.yaml
index ebabf56718..e9d5ab9a10 100644
--- a/open-api/rest-catalog-open-api.yaml
+++ b/open-api/rest-catalog-open-api.yaml
@@ -2430,6 +2430,10 @@ components:
manifest-list:
type: string
description: Location of the snapshot's manifest list file
+ first-row-id:
+ type: integer
+ format: int64
+ description: The first _row_id assigned to the first row in the
first data file in the first manifest
summary:
type: object
required:
@@ -2515,6 +2519,10 @@ components:
last-updated-ms:
type: integer
format: int64
+ next-row-id:
+ type: integer
+ format: int64
+ description: A long higher than all assigned row IDs; the next
snapshot's first-row-id.
properties:
type: object
additionalProperties:
@@ -4299,6 +4307,10 @@ components:
content:
type: string
const: "data"
+ first-row-id:
+ type: integer
+ format: int64
+ description: "The first row ID assigned to the first row in the data
file"
column-sizes:
allOf:
- $ref: '#/components/schemas/CountMap'
