mirror of https://github.com/huggingface/xet-core.git synced 2026-06-04 13:30:29 +08:00

Files

Hoyt Koepke 9caf7fcc44 V2 reconstruction with client-side optional single range splitting (#703 )

This PR introduces V2 multirange URL fetching for xorbs, but optionally
splits the multirange requests into multiple single-range requests that
can be executed in parallel. This allows the reconstruction process to
generate full multirange presigned URLs, but the client effectively
performs the retrieval stage as a sequence of parallel single-range
queries.

The config variable `client.enable_multirange_fetching` controls this
behavior; by default it is set to false due to the current observed
slowness of fetching multiranged URLs.

---------

Co-authored-by: Adrien <adrien@huggingface.co>

2026-03-16 14:10:50 -07:00

README.md

Code reorganization towards release of xet cargo package (#693 )

2026-03-11 12:02:38 -07:00

update_260304_cas_xorb_name_clarification.md

Code reorganization towards release of xet cargo package (#693 )

2026-03-11 12:02:38 -07:00

update_260309_package_restructure.md

Code reorganization towards release of xet cargo package (#693 )

2026-03-11 12:02:38 -07:00

update_260316_unify_upload_download_sync_async.md

Unify sync and async download/upload groups in session interface. (#719 )

2026-03-16 13:33:46 -07:00

update_260316_v2_reconstruction_multirange.md

V2 reconstruction with client-side optional single range splitting (#703 )

2026-03-16 14:10:50 -07:00

update_260316_xet_config_get_set_python.md

Optimize config struct for direct access in python (#706 )

2026-03-16 12:23:43 -07:00

README.md

API changes

This folder contains a record of API changes in main. It's indended for AI agents to read in order to correctly apply merges or update dependencies and PRs.

The updates are listed by date in the form: update_<yymmdd>_<description>.md

When applying a merge, rebase, or downstream update, all AI agents should first scan this folder to understand what relevant information may need to be applied.

When creating a PR that involves an API change potentially requiring downstream updates, an AI agent should create such a file. This file should be humanly readable but contain enough information to correctly apply the needed changes without scanning the code.