chore(deps): bump hotdata-framework from 0.4.0 to 0.9.0

[dependabot]

Bumps hotdata-framework from 0.4.0 to 0.9.0.

Release notes

Sourced from hotdata-framework's releases.

hotdata-framework 0.9.0

Added

• list_managed_tables, load_managed_table, add_managed_table, delete_managed_table, delete_managed_database, and execute_sql accept an already-resolved ManagedDatabase (as returned by create_managed_database) in place of a name/id. When passed one, they skip the get_database / list_databases read probe. This lets an API key scoped to create + load but not read /databases bootstrap a managed database and load into it within a single run: the caller holds the ManagedDatabase from create and drives the load/add/query ops with zero reads. The name/id string path is unchanged.

hotdata-framework 0.8.0

Changed

• load_managed_table accepts a key argument — the merge key columns for delete/update/upsert loads, matched per-load instead of requiring a key declared at table creation. Omit it to use the table's declared key; ignored for replace/append. Requires hotdata>=0.8.0.

hotdata-framework 0.7.3

Changed

• upload_parquet() now delegates to the SDK's hotdata.uploads.UploadsApi.upload_file() instead of hand-rolling the session → PUT → finalize flow. Uploads gain concurrent part PUTs under a peak-memory budget, per-part retries, and ETag/size validation, making large uploads substantially faster. Errors still surface as RuntimeError with the underlying ApiException as the direct cause.

Fixed

• classify_sdk_error now classifies HTTP 501 (Not Implemented) as terminal instead of transient — a permanent capability gap must not burn retries.

hotdata-framework 0.7.2

Removed

• The POST /v1/files fallback in upload_parquet(). Presigned upload sessions (POST /v1/uploads) are now required; a server that responds 501 raises a clear RuntimeError instead of silently falling back to the full-file-in-memory upload path.

hotdata-framework 0.7.1

Changed

• upload_parquet() now uses the presigned upload session API (POST /v1/uploads) instead of reading the entire file into memory before uploading. For multipart mode the file is streamed one part_size chunk at a time, eliminating the memory spike that caused OOM on large Parquet files. Falls back to POST /v1/files when the server returns 501.

hotdata-framework 0.7.0

Added

• load_managed_table(..., mode=...) selects the load mode (replace (default), append, delete, update, upsert) instead of always replacing the table. replace/append apply the upload directly; delete/update/upsert match rows by the table's declared key. Backward compatible — omitting mode still replaces.
• create_managed_database(..., keys={table: [cols]}) and add_managed_table(..., key=[cols]) declare a table's row-identity key, enabling the key-based load modes on it. Requires a hotdata client whose managed-table decl models carry key (see the dependency floor bump); tables declared without a key stay replace/append-only.

Fixed

• load_managed_table(..., mode="append") is no longer retried on transient errors. Every other mode is idempotent, but retrying an append whose commit succeeded before the response was received would duplicate the uploaded rows; append now runs at most once. mode is also now typed as a literal of the accepted values.

hotdata-framework 0.6.3

... (truncated)

Changelog

Sourced from hotdata-framework's changelog.

[0.9.0] - 2026-07-23

Added

• list_managed_tables, load_managed_table, add_managed_table, delete_managed_table, delete_managed_database, and execute_sql accept an already-resolved ManagedDatabase (as returned by create_managed_database) in place of a name/id. When passed one, they skip the get_database / list_databases read probe. This lets an API key scoped to create + load but not read /databases bootstrap a managed database and load into it within a single run: the caller holds the ManagedDatabase from create and drives the load/add/query ops with zero reads. The name/id string path is unchanged.

[0.8.0] - 2026-07-20

Changed

• load_managed_table accepts a key argument — the merge key columns for delete/update/upsert loads, matched per-load instead of requiring a key declared at table creation. Omit it to use the table's declared key; ignored for replace/append. Requires hotdata>=0.8.0.

[0.7.3] - 2026-07-16

Changed

• upload_parquet() now delegates to the SDK's hotdata.uploads.UploadsApi.upload_file() instead of hand-rolling the session → PUT → finalize flow. Uploads gain concurrent part PUTs under a peak-memory budget, per-part retries, and ETag/size validation, making large uploads substantially faster. Errors still surface as RuntimeError with the underlying ApiException as the direct cause.

Fixed

• classify_sdk_error now classifies HTTP 501 (Not Implemented) as terminal instead of transient — a permanent capability gap must not burn retries.

[0.7.2] - 2026-07-15

Removed

• The POST /v1/files fallback in upload_parquet(). Presigned upload sessions (POST /v1/uploads) are now required; a server that responds 501 raises a clear RuntimeError instead of silently falling back to the full-file-in-memory upload path.

[0.7.1] - 2026-07-15

Changed

• upload_parquet() now uses the presigned upload session API (POST /v1/uploads) instead of reading the entire file into memory before uploading. For multipart mode the file is streamed one part_size chunk at a time, eliminating the memory spike that caused OOM on large Parquet files. Falls back to POST /v1/files when the server returns 501.

[0.7.0] - 2026-07-14

Added

• load_managed_table(..., mode=...) selects the load mode (replace (default), append, delete, update, upsert) instead of always replacing the table. replace/append apply the upload directly; delete/update/upsert match rows by the table's declared key. Backward compatible — omitting mode still replaces.
• create_managed_database(..., keys={table: [cols]}) and add_managed_table(..., key=[cols]) declare a table's row-identity key, enabling the key-based load modes on it. Requires a hotdata client whose managed-table decl models carry key (see the dependency floor bump); tables declared without a key stay replace/append-only.

... (truncated)

Commits

• 0fb624d feat: accept resolved ManagedDatabase in managed-table ops to skip read probe...
• 2ec35c5 Merge pull request #51 from hotdata-dev/chore/add-codeowners
• 3cc9ae4 chore: add default CODEOWNERS
• c846523 Merge pull request #50 from hotdata-dev/release/v0.8.0
• cd1bc61 chore: release v0.8.0
• 4f3769d Merge pull request #49 from hotdata-dev/feat/per-load-key
• c266493 chore: lock hotdata 0.8.0
• 5e21767 feat: add per-load key to load_managed_table
• 268c9e2 feat: delegate upload_parquet to SDK upload_file; classify 501 as terminal (0...
• 7d12553 feat!: remove POST /v1/files fallback — presigned uploads required (0.7.2) (#47)
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)