Repository navigation

Issues

New issue

V3 Tracking issue
#1818 · Fokko opened on Mar 20, 2025
11
PyIceberg Production Use case survey
#1202 · kevinjqliu opened on Sep 24, 2024
16

After promoting a float column to double, > and != filters drop rows from older files
apache/iceberg-python#4203 · opened on Oct 30, 2026
A stale update_sort_order adds a second sort order with an existing id, which Java cannot load
apache/iceberg-python#4202 · opened on Oct 30, 2026
expire_snapshots on a stale table deletes a branch moved onto the expired snapshot
apache/iceberg-python#4200 · opened on Oct 30, 2026
A stale update_spec adds a partition field named like a column another writer added, which Java cannot load
apache/iceberg-python#4198 · opened on Oct 30, 2026
count() and plan_files() return nothing for an expired snapshot id
apache/iceberg-python#4197 · opened on Oct 30, 2026
dynamic_partition_overwrite by identity(s.x) deletes rows by top-level x
apache/iceberg-python#4196 · opened on Oct 30, 2026
1
Appending to a branch that does not exist creates it without main's rows
apache/iceberg-python#4195 · opened on Oct 30, 2026
A retried upsert inserts a key that a concurrent writer already inserted
apache/iceberg-python#4194 · opened on Oct 30, 2026
incremental_append_scan from an expired start fails after expire_snapshots
apache/iceberg-python#4193 · opened on Oct 30, 2026
dynamic_partition_overwrite after an in-transaction update_spec loses rows
apache/iceberg-python#4192 · opened on Oct 30, 2026
add_files puts a file holding a value and NULLs in that value's partition
apache/iceberg-python#4191 · opened on Oct 30, 2026
A stale add_column reuses a dropped column's field id and schema id
apache/iceberg-python#4190 · opened on Oct 30, 2026
A long transaction committed past a concurrent tag writes metadata Java rejects
apache/iceberg-python#4189 · opened on Oct 30, 2026
A retried append records its file under a concurrently added partition spec
apache/iceberg-python#4188 · opened on Oct 30, 2026
1
create_branch and create_tag replace an existing ref of the same name
apache/iceberg-python#4187 · opened on Oct 30, 2026
1
inspect.entries() raises TypeError when a data file has no column_sizes
apache/iceberg-python#4186 · opened on Oct 30, 2026
snapshot_as_of_timestamp returns an older snapshot after expire_snapshots
apache/iceberg-python#4185 · opened on Oct 30, 2026
Table.delete() with an OR filter deletes the wrong rows when a column is NULL
apache/iceberg-python#4184 · opened on Oct 30, 2026
A transaction writing two branches overwrites concurrent changes to the second
apache/iceberg-python#4183 · opened on Oct 30, 2026
older_than reads a naive datetime as UTC and expires snapshots it should keep
apache/iceberg-python#4182 · opened on Oct 30, 2026
Overwrite keeps deleted-only manifests, so the manifest list grows by one on every overwrite
apache/iceberg-python#4165 · opened on Oct 27, 2026
SqlCatalog: namespace_exists matches parent namespaces case-insensitively on SQLite
apache/iceberg-python#4163 · opened on Oct 27, 2026
1
Reject v3-only features on pre-v3 tables
apache/iceberg-python#4156 · opened on Oct 25, 2026
1
Local warehouse (file://): data files, manifests and metadata.json are never fsynced before the catalog commit; after a power loss the table cannot be loaded
apache/iceberg-python#4145 · opened on Oct 23, 2026
Matching Catalog Properties with Java
apache/iceberg-python#4140 · opened on Oct 22, 2026
1

Navigation Menu

Repository navigation

Issues

Issues

After promoting a `float` column to `double`, `>` and `!=` filters drop rows from older files

A stale `update_sort_order` adds a second sort order with an existing id, which Java cannot load

`expire_snapshots` on a stale table deletes a branch moved onto the expired snapshot

A stale `update_spec` adds a partition field named like a column another writer added, which Java cannot load

`count()` and `plan_files()` return nothing for an expired snapshot id

`dynamic_partition_overwrite` by `identity(s.x)` deletes rows by top-level `x`

Appending to a branch that does not exist creates it without main's rows

A retried `upsert` inserts a key that a concurrent writer already inserted

`incremental_append_scan` from an expired start fails after `expire_snapshots`

`dynamic_partition_overwrite` after an in-transaction `update_spec` loses rows

`add_files` puts a file holding a value and NULLs in that value's partition

A stale `add_column` reuses a dropped column's field id and schema id

A long transaction committed past a concurrent tag writes metadata Java rejects

A retried append records its file under a concurrently added partition spec

`create_branch` and `create_tag` replace an existing ref of the same name

`inspect.entries()` raises TypeError when a data file has no column_sizes

`snapshot_as_of_timestamp` returns an older snapshot after `expire_snapshots`

Table.delete() with an OR filter deletes the wrong rows when a column is NULL

A transaction writing two branches overwrites concurrent changes to the second

`older_than` reads a naive datetime as UTC and expires snapshots it should keep

Overwrite keeps deleted-only manifests, so the manifest list grows by one on every overwrite

SqlCatalog: namespace_exists matches parent namespaces case-insensitively on SQLite

Reject v3-only features on pre-v3 tables

Local warehouse (`file://`): data files, manifests and `metadata.json` are never fsynced before the catalog commit; after a power loss the table cannot be loaded

Matching Catalog Properties with Java