In general the code looks good but there is one major problem: if someone doesn't update their desktop app they can still run a newer version of the web app.

Rebase

Updated the diff to match the changes in previous diffs.

• Small fixes
• Keep the olm runtime objects directly in memory instead of the pickled form
• Create a new separation for olm related requests in worker request types (like in the next diffs for identity client)
• Use the globals getters introduced in the previous diff

If it returns empty object we should generate remove_all_integrity_thread_hashes operation - you could do it in a separate diff

Fixed comment. Also renamed the worker name from comm-app-database-${codeVersion} to comm-app-shared-worker-${codeVersion}.

I'm also curious how expensive it might be to change the approach at this time.

Response to the opaque URL comment:

@ashoat would something like this be more preferable?

Rebase

1. We no longer invalidate the cache. What happens if a client calls _finish method a second time?
2. Just want to make sure: we don't have any sensitive data in WorkflowInProgress that we would prefer to keep only in-memory or sth like this?

This differential removes ashoatKeyserverID from native notifs code

When using binaryen from npm I was getting this error in CI:

Try reverting binaryen to the apt install because of CI failure.

Fix script path in dockerfile

I know this was discussed earlier, but can you remind me why we're taking the rust-node-addon approach here instead of the opaque-ke-wasm approach? Is it because of issues getting the CI pipeline working?

Renamed the script to run-emscripten.sh, moved postinstall.sh to scripts. Added web/backup-client-wasm/target to .dockerignore.

Requesting review again because there were some changes:

Rebase, fix path

Rebase

Rebase, fix issues

Fix issues

Fixes for keyserver docker build.

As discussed in ENG-6873 : Add a buildkite job for checking integrity of backup client wasm file, we are changing thr approach and instead of keeping the wasm file in git repo we will be building it on cleaninstall as we do with the keyserver node addon.

is there an existing task tracking implementing backup restore for corrupt database recovery, for both native and web? I want to add a comment referencing this code in SQLiteDataHandler, to make sure it's updated as part of that task.

Resigning so I won't block landing. I don't there should be any problems with fixing the tests.

Canceling this in favour of just adding another build step to cleaninstall. See discussions in ENG-6873

Sorry, I keep forgetting that for this diff, when testing CI I have to plan changes after every new revision.

Rebase

Rebase

Rebase

Rebase

Oh sorry, I forgot this was in review, sorry for the confusion. I'm having issues with generated WASM reproducibility on CI so this diff isn't ready yet and I'm still testing.

Remap paths

Rebase

Try cleaning before building

Check that CI passes if wasm is current.

Switch to two command, as I had problems with doing both builds in one command.

Switch to one command. Check that CI fails if wasm wasn't updated.

Rebase

Rebase

Rebase

Fix rebase error

Rebase

Rebase

Rebase

Update lock file

Update lock file

Rebase

Rebase

Fix comment