SQL Server backup files → Parquet
Datatypes conversion mapping when exporting a table from SQL Server backups (.bak, full and differential) using the ArpeMSBackup ADBC driver (--connectiontype backup_mssql) to Parquet format.
backup_mssql is available on Windows x64 and Linux x64 only. It is not supported on Linux ARM64.
adbc_mssqlSince FastBCP 1.6.5, the mapping and the values are those of adbc_mssql, so a backup and a live export of the same table give the same Parquet. The exception is json and vector(n, float16) (SQL Server 2025): backup_mssql refuses them.
time(7): microsecond unit, instead of nanosecond.geography,geometry: WKB with GeoParquet metadata, instead of the SQL Server binary serialization.vector(n): list of 32-bit floats, instead of the SQL Server binary serialization.sql_variant: values are written as withadbc_mssql.
| Source | SQL Server backup file (backup_mssql) |
| Format | Parquet |
| FastBCP | 1.6.4.0; time, geography, geometry, uniqueidentifier, json and vector: 1.6.5.0 |
| Test Status | PARTIAL (32/33 Nullable) — json refused. SQL Server 2022 and 2025 backups: uncompressed, striped, MS_XPRESS and ZSTD compression, --parallelmethod Physical |
Parquet Schema: testdatatypes_nullable
| name | physical_type | converted_type | logical_type |
|---|---|---|---|
| col_bigint | INT64 | — | — |
| col_int | INT32 | — | — |
| col_smallint | INT32 | INT_16 | IntType(bitWidth=16, isSigned=1) |
| col_tinyint | INT32 | UINT_8 | IntType(bitWidth=8, isSigned=0) |
| col_bit | BOOLEAN | — | — |
| col_decimal | FIXED_LEN_BYTE_ARRAY | DECIMAL | DecimalType(scale=2, precision=18) |
| col_numeric | FIXED_LEN_BYTE_ARRAY | DECIMAL | DecimalType(scale=2, precision=18) |
| col_money | FIXED_LEN_BYTE_ARRAY | DECIMAL | DecimalType(scale=4, precision=19) |
| col_smallmoney | FIXED_LEN_BYTE_ARRAY | DECIMAL | DecimalType(scale=4, precision=10) |
| col_float | DOUBLE | — | — |
| col_real | FLOAT | — | — |
| col_date | INT32 | DATE | DateType() |
| col_time | INT64 | — | TimeType(isAdjustedToUTC=0, unit=MICROS) |
| col_datetime | INT64 | TIMESTAMP_MILLIS | TimestampType(isAdjustedToUTC=0, unit=MILLIS) |
| col_datetime2 | INT64 | TIMESTAMP_MICROS | TimestampType(isAdjustedToUTC=0, unit=MICROS) |
| col_datetimeoffset | INT64 | TIMESTAMP_MICROS | TimestampType(isAdjustedToUTC=1, unit=MICROS) |
| col_smalldatetime | INT64 | TIMESTAMP_MILLIS | TimestampType(isAdjustedToUTC=0, unit=MILLIS) |
| col_char | BYTE_ARRAY | UTF8 | StringType() |
| col_varchar | BYTE_ARRAY | UTF8 | StringType() |
| col_nchar | BYTE_ARRAY | UTF8 | StringType() |
| col_nvarchar | BYTE_ARRAY | UTF8 | StringType() |
| col_text | BYTE_ARRAY | UTF8 | StringType() |
| col_ntext | BYTE_ARRAY | UTF8 | StringType() |
| col_binary | BYTE_ARRAY | — | — |
| col_varbinary | BYTE_ARRAY | — | — |
| col_image | BYTE_ARRAY | — | — |
| col_uniqueidentifier | BYTE_ARRAY | UTF8 | StringType() |
| col_xml | BYTE_ARRAY | UTF8 | StringType() |
| col_geography | BYTE_ARRAY | — | — (WKB, GeoParquet geo metadata, spherical edges) |
| col_geometry | BYTE_ARRAY | — | — (WKB, GeoParquet geo metadata) |
| col_sql_variant | BYTE_ARRAY | UTF8 | StringType() |
| col_json | — | — | Refused |
| col_vector | FLOAT (repeated) | LIST | ListType() |
col_json (json) and col_vector (vector(3)) are SQL Server 2025 types, tested with FastBCP 1.6.5.0.
vector(n)is exported as a list ofn32-bit floats ([1.0, -2.5, 3.25]).geographyandgeometrykeep their coordinate system: the GeoParquet metadata carries the CRS from the column's SRID.uniqueidentifieris written in lowercase by default. Since FastBCP 1.6.5,FASTBCP_UID_STRING_CASINGapplies tobackup_mssqltoo.
To get geography and geometry another way, give the connection as a connection string: adbc.arpemsbackup.geospatial=wkb writes plain WKB without the metadata, and varbinary writes the SQL Server binary serialization, as before 1.6.5.
--sourceconnectstring "adbc.arpemsbackup.path=D:\Backups\MSSQL;adbc.arpemsbackup.file=mydb.bak;adbc.arpemsbackup.geospatial=varbinary"
json and vector(n, float16) columns are refusedA table with such a column is refused with a message naming the column (column j: json (SQL Server 2025) is not supported). Export the other columns by listing them in --query, or export the table with adbc_mssql.