table.inspect metadata tables fail (or lose manifest bounds) after dropping an old partition column

未關閉
  • 哪裡不對？
#4,355 0 則留言 0 個 reaction 已指派 0 人 在 GitHub 檢視

維護者通常 1 天內回覆

@Ankitkj1999 已經在處理了。

開始於 2026年11月12日。

  • #4356 來自 @Ankitkj1999 —— 未關閉

評估

難度
3/5
預估耗時
1-2 天
新手友好度
35/100
Issue 類型
缺陷
描述清晰度
基本清楚
活躍度
活躍
技術堆疊
python
領域
data

研究方向

Start with TableMetadata.specs_struct() and _partition_summaries_to_rows(), which call spec.partition_type(...) against the current schema, and with the InspectTable methods that use them. Run the reproduction in the issue to see the ValueError and the 'null' bounds. Done means the inspect partitions/files calls and manifest partition summary bounds work after a source column is dropped. Open PR #4356 already targets this, so check it before starting.

由索引模型根據 Issue 內容生成。

描述

kind:bug
Apache Iceberg version

main (development)

Please describe the bug 🐞
Describe the bug

If a table stops partitioning by a column (update_spec().remove_field(...)) and later drops that source column (update_schema().delete_column(...)):

  1. Calling table.inspect.partitions(), entries(), files(), data_files(), delete_files(), all_files(), all_data_files(), or all_delete_files() raises ValueError: Could not find field with id: ... inside TableMetadata.specs_struct().
  2. Calling table.inspect.manifests() / all_manifests() succeeds, but renders lower_bound and upper_bound in partition_summaries as 'null' because _partition_summaries_to_rows() evaluates spec.partition_type(self.tbl.schema()), which falls back to UnknownType().

(Note: open PR #4115 fixes delete() manifest rewrites in pyiceberg/avro/file.py and pyiceberg/manifest.py, whereas this failure occurs in TableMetadata.specs_struct() and InspectTable even without calling delete().)

To Reproduce
import tempfile
import pyarrow as pa
from pyiceberg.catalog.memory import InMemoryCatalog
from pyiceberg.partitioning import PartitionField, PartitionSpec
from pyiceberg.schema import Schema
from pyiceberg.transforms import IdentityTransform
from pyiceberg.types import LongType, NestedField, StringType

with tempfile.TemporaryDirectory() as tmp:
    cat = InMemoryCatalog("local", warehouse=f"file://{tmp}/wh")
    cat.create_namespace("ns")
    tbl = cat.create_table(
        "ns.t_v2",
        schema=Schema(NestedField(1, "id", LongType()), NestedField(2, "p", StringType())),
        partition_spec=PartitionSpec(PartitionField(2, 1000, IdentityTransform(), "p")),
        properties={"format-version": "2"},
    )
    tbl.append(pa.table({"id": [1, 2], "p": ["a", "b"]}))
    with tbl.update_spec() as spec_update:
        spec_update.remove_field("p")
    with tbl.update_schema() as schema_update:
        schema_update.delete_column("p")

    print(tbl.inspect.manifests().to_pydict()["partition_summaries"])
    # [[{'contains_null': False, 'contains_nan': False, 'lower_bound': 'null', 'upper_bound': 'null'}]]

    tbl.inspect.partitions()
    # ValueError: Could not find field with id: 2
Expected behavior

TableMetadata.specs_struct() and table.inspect.* should resolve partition source field types from historical table schemas when the source column was dropped from the current schema, keeping metadata inspection tables and manifest partition summary bounds readable.

Willingness to contribute
  • I can contribute a fix for this bug independently
  • I would be willing to contribute a fix for this bug with guidance from the Iceberg community
  • I cannot contribute a fix for this bug at this time
主要語言
Python
星號
1.2k
分支
654
平均合併
2 天 11 小時
30 天內合併 PR
122

環境準備

  • 沒有 Dockerfile 或 Docker Compose 檔案
  • 有 Pull Request 範本
  • 沒有貢獻指南

從這裡開始

  1. 先讀完整個 Issue，再讀專案的貢獻指南。
  2. 在 Issue 下留言說明你要接手 —— 這能避免兩個人做同樣的事。
  3. Fork 儲存庫，在一個分支上完成修改。
  4. 送出 Pull Request，並在描述裡引用這個 Issue 編號。

apache/iceberg-python 的其他 Issue

查看 apache/iceberg-python 的全部 Issue

相似的 Issue

更多 Python Issue

把新 issue 寄到你的電子郵件信箱

精選適合新手參與的 GitHub issue 摘要。