镜像站点 · 本页由第三方 GitHub 只读镜像提供,非 GitHub 官方站点,不接受任何登录或凭据输入。前往 github.com
Skip to content

[DOC] Glossary page for core concepts #651

Closed

Description

@FrancescAlted

New contributors and users frequently confuse chunks, blocks, frames, and superchunks — terms that appear throughout the API but are never defined in one place. The getting-started section has no glossary.

Suggested work: Create doc/getting_started/glossary.rst (and link it from doc/getting_started/index.rst) defining at minimum:

  • SChunk, frame, chunk, block, subblock
  • Compression parameters (cparams): codec, clevel, filters
  • NDArray vs CTable vs SChunk — when to use each
  • Index types: SUMMARY, BUCKET, PARTIAL, FULL, OPSI — one-sentence each

Each term should link to the relevant API reference page.

Activity

  1. lanceu0128 commented on Sep 20, 2026

    @lanceu0128
    Contributor

    I'll take this issue if you don't mind @FrancescAlted (I'm in the NumFocus program)

  2. FrancescAlted commented on Sep 21, 2026

    @FrancescAlted
    MemberAuthor

    Hi @lanceu0128 ! Sure, go ahead.

  3. lanceu0128 commented on Sep 21, 2026

    @lanceu0128
    Contributor

    Actually, someone already took this issue on the separate tracker for NumFocus. I've claimed #652 there instead. @FrancescAlted

  4. FrancescAlted commented on Sep 22, 2026

    @FrancescAlted
    MemberAuthor

    Well, it is not necessary to choose different tickets for everyone. For example, this is a big one deserving quite a bit of work. Share it with another guy and collaborate!

  5. lanceu0128 commented on Sep 23, 2026

    @lanceu0128
    Contributor

    Update: we discussed this offline, I'll continue working on this issue #651 with Kumal instead of #652

  6. lanceu0128 commented on Sep 29, 2026

    @lanceu0128
    Contributor

    Opened up a draft PR. Currently has half of the terms @FrancescAlted requested + LazyExpr and LazyArray.

  7. FrancescAlted commented on Sep 30, 2026

    @FrancescAlted
    MemberAuthor

    Hey @lanceu0128 . This is looking good. Maybe I'd split it in 3 or 4 sections, like:

    • data containers (NDArray, CTable, BatchArray...)
    • compression (codecs, filters, other params...)
    • low-level data structures (SChunk, chunks, blocks...)
    • indexes (SUMMARY, BUCKET, PARTIAL, FULL, OPSI)

    I think this should improve readability.

  8. lanceu0128 commented on Oct 2, 2026

    @lanceu0128
    Contributor

    @FrancescAlted We've finished our work, including the split you suggested. Just opened the PR officially.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions