Marc Durdin

> @mcdurdin This seems like a good idea because it allows you to pick up where you left of before you switched to another app, but I'm not entirely sure...

On Windows, we compare the new text context from the app to the textual part of the cached context. If it is different, then we reset the cached context and...

I think it's unlikely that the keyboardid or packageid (aka keyboard and package filenames) would have mixed case -- there might be rare cases where that would happen in a...

> We have proof of at least one wild mixed-case scenario, unless my interpretation's off somehow? On that example: 1. I doubt it is an ad-hoc install -- it's a...

I note the build failed, probably because binary data no longer matches for the e2e test (haven't investigated). Additional test would be good.

@srl295 updated PR description to get automatic link to issue.

See also https://github.com/keymanapp/keyman/pull/7437/files#diff-95985cb44c34779804524113d19f84c84c0df3809790667a95bd3592e3a9e7cbR307

In terms of: > TEST_FLICKER_BANNER (PASSED) This has been observed with current master and stable builds. So we need to make sure that we are able to reproduce the issue...

Can you test these with 15.0.270 also please? We just released that version last night. On Tue, 13 Sept 2022, 6:04 pm bharanidharanj, ***@***.***> wrote: > SUITE_KNOWN_BUGS: drawn from a...

@bharanidharanj thank you -- those test results for 15.0.270 match up with what I was hoping for!