Appending to a branch that does not exist creates it without main's rows #4195

Apache Iceberg version

0.12.0 (latest release)

Please describe the bug 🐞

Currently, a write to a missing branch takes no parent, so the new branch holds only the rows just written. Creating the branch with create_branch first gives [1, 2, 3].

import tempfile
import pyarrow as pa
from pyiceberg.catalog.sql import SqlCatalog

cat = SqlCatalog("c", uri="sqlite:///:memory:", warehouse=f"file://{tempfile.mkdtemp()}")
cat.create_namespace("ns")
t = cat.create_table("ns.t", schema=pa.schema([("x", pa.int64())]))
t.append(pa.table({"x": [1, 2]}))

t.append(pa.table({"x": [3]}), branch="b")
print(t.scan().use_ref("b").to_arrow()["x"].to_pylist())
# [3], should be [1, 2, 3]
# Java's SnapshotUtil.latestSnapshot takes the current snapshot as the parent of a missing branch
print(t.snapshot_by_name("b").parent_snapshot_id)
# None

overwrite does the same, and so does a retried append after a concurrent remove_branch.

tested on pyiceberg 0.12.0 and main

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