Specification status report for TPAC 2020

[inexorabletash]

The specification status report should include:

• What progress has your spec made in the last 12 months?
• Is anything blocking your spec from moving to CR?
• If yes, what is your plan to unblock it and do you need any help?

[inexorabletash]

Progress in the last 12 months

• New functionality
  • added transaction durability option, per discussion at TPAC 2019 Define flush-to-disk guarantees and control #50
    • Supported in Chrome and Edge
• Other normative changes
  • remove File lastModifiedDate references, per discussion at TPAC 2019 lastModifiedDate on File is not a thing #215
  • specify that read-only transactions can run in parallel but block a later read/write transaction from starting, and the read/write transactions similarly block later read/write and read-only transactions, matching implementations Disallow starting readwrite transactions while readonly transactions are running? #253
  • disallow proxies as keys, matching implementations. Array proxies as keys? #309
  • make transactions inactive during the value clone process, to prevent side effects (e.g. getters) from re-entering algorithms.
• Spec quality
  • Alignment with Infra and WebIDL syntax, term and convention updates.
  • Added Accessibility Concerns section Add accessibility section #327

Pending changes:

• New functionality
  • (has PR) IDBObjectStore putAll() per TPAC 2019 discussion - Chrome has implementation experience, but no data ready to share yet.
  • (has PR) Manual commit per TPAC 2019 discussion - no implementation plans I'm aware of.
  • (has PR) IDBObjectStore / IDBIndex getAllEntries() per TPAC 2019 discussion -
    no implementation plans I'm aware of.
  • (has PR) IDBCursor close() per TPAC 2019 discussion - no implementation plans I'm aware of.
• Spec quality
  • (has PR) Define Indexed DB as a storage endpoint, per Storage, and integrate with "bottles" terms, etc.

Moving to CR

Seems premature:

• Goals for V3 in charter state "The third edition adds new capabilities and improves developer ergonomics by using promises." KV-Storage as a Promise-based layer on top of Indexed DB has stalled. An alternative is to push on low-level support in Indexed DB (example) and encourage libraries to implement Promise support.
• Overall, new functionality since V2 is limited: see https://w3c.github.io/IndexedDB/#revision-history
  • databases() method API to enumerate databases #31, commit() method Proposal: Add explicit IDBTransaction.commit() (was: add "writeonly" mode) #234, request attribute cursor.request #255, durability option Define flush-to-disk guarantees and control #50

How to unblock:

• IMHO, at least one more "big" feature is needed (e.g. one of the above PRs) to justify the effort
• Need further implementation experience
• Any other interested editors? Both current editors are from Chromium-based browsers.
• Need a WD published at least.

See also

• TPAC 2019 report

[inexorabletash]

FYI @LJWatson @marcoscaceres

[marcoscaceres]

Moved this to the wiki https://github.com/w3c/IndexedDB/wiki/Specification-status-report-for-TPAC-2020