Firebird into Apache Arrow through ADBC, using Firebird ODBC 3.5.0-rc1 (adbcBridge 0.1.0)

[AI Exploration]

adbcBridge is a small Apache-2.0 ADBC driver written in C11 that loads an ODBC driver and exposes it through the Arrow ADBC C ABI: block-cursor reads straight into Arrow record batches, bulk ingest, metadata, partitioned parallel reads. I ran Firebird through it as one of 46 databases in a single compatibility workload and wanted to share the entry here, since the quirks it records are Firebird-specific.

What was verified (Firebird 5; driver: Firebird ODBC 3.5.0-rc1):
   Linux:        PASS
   macOS arm64:  driver unavailable: firebird-odbc-driver v3-0-1 ships Linux and Windows assets only
   Windows x64:  PASS (Firebird ODBC 3.0.1.18; the driver answers SQL_DRIVER_NAME "FirebirdODBC" there and "OdbcFb" on Linux)

What the compatibility entry records:
   SQL_C_WCHAR sized in 4-byte wchar_t; no usable parameter arrays. Firebird's dialect has neither multi-row VALUES (-104 Token unknown at the second row-group's comma) nor Oracle's INSERT ALL, so bulk ingest batches through the third form it does take: INSERT INTO t (cols) SELECT CAST(? AS <type>), ... FROM RDB$DATABASE UNION ALL SELECT ...
   Full entry: https://github.com/singhpratech/adbcbridge/blob/main/ docs/COMPATIBILITY.md

As far as I know there is no ADBC driver for Firebird of any kind, so this is currently the way to get Firebird result sets into Arrow through the ADBC API — from Python, Rust, Go, Java or C#, one shared library. While running it I hit three bugs in the ODBC driver itself and wrote them up with pure-ODBC C reproductions (no adbcBridge involved) on the driver tracker: parameter arrays step fixed-length C types by BufferLength so every row receives row 1's values (https://github.com/ FirebirdSQL/firebird-odbc-driver/issues/299), a SQL_BIGINT parameter bound once as SQL_C_DEFAULT keeps writing NULL after later SQL_C_SBIGINT rebinds (https://github.com/FirebirdSQL/firebird-odbc-driver/ issues/300), and SQLPrepare discards SQL_ATTR_ROWS_FETCHED_PTR set before it (https://github.com/FirebirdSQL/firebird-odbc-driver/ issues/301). The first two are silent — SQL_SUCCESS all the way and wrong rows in the table — so if you drive Firebird through ODBC from anything that uses parameter arrays, they are worth a look. adbcBridge works around all three.

Trying it (Python; Rust, Go, Java and C# are on the docs site):

   pip install adbcbridge

   import adbcbridge
   with adbcbridge.connect(uri="Driver=/opt/firebird-odbc/ libOdbcFb.so;DBNAME=inet://localhost:3050//data/ mydb.fdb;UID=SYSDBA;PWD=...;CHARSET=UTF8;") as conn:
       with conn.cursor() as cur:
           cur.execute("SELECT * FROM RDB$RELATIONS")
           table = cur.fetch_arrow_table()   # a pyarrow.Table

Links:
   Repository:     https://github.com/singhpratech/adbcbridge
   Docs:           https://adbcbridge.org/docs/
   Compatibility:  https://github.com/singhpratech/adbcbridge/blob/main/ docs/COMPATIBILITY.md
   Upstream notes: https://github.com/singhpratech/adbcbridge/blob/main/ docs/UPSTREAM.md
   PyPI:           https://pypi.org/project/adbcbridge/

It is a 0.1.0. If the entry says something wrong about Firebird, or you run a version or driver I didn't, an issue on the repository with the details is the most useful thing you could send.