revert: drop the duplicate Wi-Fi picker and /data persistence - #449
Open
tomassasovsky wants to merge 1 commit into
Open
revert: drop the duplicate Wi-Fi picker and /data persistence#449tomassasovsky wants to merge 1 commit into
tomassasovsky wants to merge 1 commit into
Conversation
Reverts #448 and #439. Both re-solved problems that feat/control-center-host-features already solved, better, in work that had been released to the console — which is why Wi-Fi worked there before tonight and stopped working after. That branch carries packages/wifi_client and packages/wifi_repository (with tests), lib/wifi/ including a tray panel, a loopy-wifi-ctl privileged helper, EAPOL/enterprise handling, association waiting and join-failure surfacing. #448 was a naive nmcli wrapper by comparison, and replacing lib/wifi/ with it was a straight regression. The persistence duplication is worse. bad09ee sets NetworkManager's own [keyfile] path=/data/NetworkManager/system-connections so NM writes profiles to /data natively, and pairs it with the settings that actually fix joining on this hardware — MAC randomization off, a real regdom, powersave off, all causes of "Secrets were required" and flaky 5 GHz on brcmfmac. #439 reimplemented the persistence half as a bind mount plus a 50-line script and a 14-assertion suite, and fixed none of the join failures. Cause: I searched lib/, packages/ and deploy/ on master, found only a coming-soon stub, and concluded no implementation existed. I never ran git log --all. Master genuinely had a stub, so every check was locally true and globally wrong. The on-screen keyboard (#447) is NOT reverted: nothing on that branch provides one, and the console still has no IME.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Reverts #448 and #439. Both re-solved problems
feat/control-center-host-featuresalready solved, better, in work that had been released to the console. That is why Wi-Fi worked on the device before tonight and reported "No networks found" after.What already existed
That branch (13 commits, unmerged, releases cut from it) has:
packages/wifi_client+packages/wifi_repository— proper packages, with testslib/wifi/— cubit, state, page, tray panel, error messages, hidden-network handlingdeploy/yocto/.../loopy-wifi-ctl— a privileged helper#448 was a naive
nmcliwrapper by comparison, and it overwrotelib/wifi/.The persistence duplication is worse
bad09eeauses NetworkManager's own supported option:NM writes profiles to
/datanatively — no bind mount. It pairs that with the settings that actually fix joining on this hardware: MAC randomisation off, a real regdom, powersave off — the documented causes of "Secrets were required" and flaky 5 GHz on brcmfmac.#439 reimplemented only the persistence half, as a bind mount plus a 50-line script and a 14-assertion suite, and fixed none of the join failures.
How this happened
I searched
lib/,packages/anddeploy/on master, found aComingSoonStub, and concluded nothing existed. I never rangit log --all. Master genuinely had a stub, so every check I ran was locally true and globally wrong. It surfaced only because the console regressed and I was told.Not reverted
#447 (on-screen keyboard) stays. Nothing on that branch provides one, and the console still has no IME — every
TextFieldthere is still dead without it, including that branch's own password field.Verification
Full suite 1345 passed / 31 author-only skips; flash-pedal 28/28;
dart analyzeclean; format clean.Next: rebase
feat/control-center-host-featuresonto master (13 ahead, 44 behind) and land it properly.