Confusing development process

[iwahdan]

Hello @robert-hh , we will be posting about a new release candidate v1.20.0.rc0 this weekend, for now I think the problem of -lmesh is because you are using IDF. master branch , for this build we are using idf_v3.1 branch.

About the Dev process, I agree with you, that's why we are currently updating the process to make the master branch always up to date with our private Development repo, and that will take effect once we merge this v1.20.0 as an Official release and merge it to master, for Now we are keeping the branch v1.20.0.rc0 for evaluating this release candidate.

[robert-hh]

@iwahdan Hello @iwahdan . Thanks for the swift response. If you look down the posts, you sse that I checked out the idf_v3.1 branch in the pycom-esp-idf & made a git pull for updating (and submodule update). And I removed that -lmesh from the linker list. At that point I got the linker error list and stopped trying.

[iwahdan]

Hi @robert-hh , sorry didn't see that you already use idf_v3.1 , however, did you update IDF submodules git submodule update after checking out idf_v3.1 branch as libmesh.a is located in components/esp32/lib

[robert-hh]

@iwahdan I was pretty sure that I did, but doing it again changed the picture, and it build now. Thanks.

Two other question:
a) v1.20.0.rc1 (like 1.19.x) has a halfway implementation of controlling the behaviour of CTRL-F. All the lower level changes are implemented, but the API call from python is not. Is that for a specific reason?
b) I got the replaced FiPy for band 8, but trying to use it tells be, that band 8 is not supported. And downloading the most recent sequans firmware requires an authorization. I know that I downloaded a version once, but cannot recall being asked for authorization. Found the password!

[iwahdan]

@robert-hh ,

for point b) I am now updating v1.20.0.rc0 to support this band along with other bands as well, will push it today to branch.
for point a) I think Exposing this option in Micropython was not part of the pull request here https://github.com/pycom/pycom-micropython-sigfox/pull/139/files , that's why I think it was missed out.

[robert-hh]

@iwahdan about point a) This exposing was part of the pull request, if you look into the link you gave. I know it since this PR was done by me. And it is the change to modmicropython.c, which is missing. But you might skip the change to the first file, esp32/fatfs/src/drivers/sflash_diskio.c, because that is unrelated and anyhow not needed.
for point b) that's fine. Is it only the code in lte_attach(), or more?

[danielm]

I am not able to successfully build the release candidate. Am I using the right compiler?

.../xtensa-esp32-elf/bin/../lib/gcc/xtensa-esp32-elf/5.2.0/../../../../xtensa-esp32-elf/bin/ld: cannot find -lcore
collect2: error: ld returned 1 exit status

[robert-hh]

@danielm I'm using 5.2.0 too. Did you update the submodules AFTER checking out the proper branch in the IDF?

[iwahdan]

@danielm libcore.a is also part of IDF /components/esp32/lib so you might need to do git submodule update after checking out idf_v3.1 branch as well

[robert-hh]

@iwahdan With a slight modification in lte_attach() (just adding band 8) I succeeded in attaching & connecting in band 8, T-Mobile Germany, 1nce.net. Good news!

[iwahdan]

@robert-hh Good news!, I also pushed the support for bands 8,5 on the release candidate.

[kuqizz]

This post is deleted!

[Paul Thornton]

I totally agree we could do a better job documenting what each release contains. and the correct versions to be running at any given point. If people have suggestions on the best way to communicate this info. I'd be more than happy to hear it.

Drop me a forum PM or email me at paul@pycom.io I'm always happy to hear suggestions for improvement.

[robert-hh]

@iwahdan There is another branch now called stable_update. I understand that this is the trial update to 1.18.1.rx. I left a note there on one of the newer commits. Having a note somewhere about the intention of each branch would be helpful.