2.1.0-follow-up master drop #2055
Conversation
The following functions were removed: - `wifi.sta.eventMonReg()` - `wifi.sta.eventMonStart()` - `wifi.sta.eventMonStop()` The corresponding documentation has also been removed This PR addresses issue #1893
#1903) * Add event callbacks to wifi.sta.config() and wifi.ap.config() and more Added option to register event callbacks during configuration of both station and AP. Added option to register event callbacks to `wifi.sta.connect()` and `wifi.sta.disconnect()` * Add note about event registration to wifi module documentation Other minor changes to wifi documentation are also included * Add more detail to documentation for wifi.sta.config()
* Only handle errors if we have not parsed a complete object * Fix typo which means that \n was not an ending character
* backport fix for espressif/esp-idf#631 * remove code from intermediate fix
* fix 2007 Call callback in all cases, call callback with errorcode -1 if no connection could be establioshed * change logging from ERR to DEBUG * make debug output more clear (hopefully) * add handling of errors to docs, note error handling on every call instead of only in the main documentation
From what I can tell they *should* be harmless, but I'd rather we keep a neat house in the first place.
- Move jsonsl build to regular library build rather than #include the .c file - Provide wrappers for malloc/calloc/free to fix undefined symbol warnings.
Sjson undefined symbol fix
* Modified wifi.sta.get*config() to return AP's MAC even if bssid_set==0 * Improved documentation for wifi.sta.getapinfo, fixes #2025
Fix the SPIFFS blocksize issues
* Improve RTC timekeeping -- includes clock rate tracking * Improved division by 1M * Fix crash in sntp * Disable RTC debug * Get the offset correct * Add comments on where the mysterious numbers came from * Fix a crash with auto repeat mode and errors on repeat
Unaligned loads are a no-no on many architectures, the ESP8266 included.
|
@marcelstoer Marcel, I notice that no one has commented on #2048. OK, let's do this drop, but I really feel that we should discuss and resolve this before any more master drops. We shouldn't make this sort of change to master as we have done without assessing the implications. |
|
@TerryE My apologies I have been busy and I haven't been able to get on here and comment, but I have been working on a solution to #2048. I'll push a branch later with the changes and post back. On a side note: @marcelstoer I have found that SDK 2.1.0 broke the current implementation of the timed portion of the light sleep functionality and I think it would be a good idea to disable it before the master drop. |
|
@TerryE That sounds reasonable, sure. No one has commented on that issue and no one has come forward volunteering to do these master drops aka releases for the past year. That's why I just keep sticking to our self-imposed 2 months interval - unless there's good reason not to. If we postpone the master drop due to #2048 and #2056 then we should postpone it for at least two weeks so the changes from #2056 can settle. @dnc40085 is usually extremely diligent but we all know we can't prove the absence of bugs. |
|
@dnc40085 has opened #2056, but this is something that we should do for a dev release first, IMO. The only change that I would propose for incorporating into this push from dev to master would be to make the |
|
@marcelstoer Marcel, why close this? why not just a a push for the two liiner I suggest? |
|
@marcelstoer PR #2064 covers the modifications @TerryE is suggesting. |
…suspend Comment out pmsleep and timer_suspend options in user_config.h
|
@ fellow maintainers |
|
Marcel, that's a bit of a catch-22, isn't it? Shoot me later. |
|
I finished up as per https://github.com/nodemcu/nodemcu-firmware/blob/master/CONTRIBUTING.md#handling-releases and created a new release: https://github.com/nodemcu/nodemcu-firmware/releases/tag/2.1.0-master_20170811 Feel free to revise the release notes. |
|
Very impressive. Thanks Marcel! 😄 |
|
What, the release notes? Since I keep tagging merged PRs with a milestone writing release notes is a 1min job, ok 2min. |
|
No, the process and that you drive it. 👏 |
Contains all issues referenced in the "2.1 follow-up" milestone.