New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
updated to v1.15.3 #52
Conversation
Started test build 42115 |
Build 42115 failed |
Started test build 42117 |
Started test build 42118 |
Build 42117 failed |
Build 42118 failed |
The generator should handle this. Let's see whether this works.
Started test build 42121 |
Build 42121 failed |
This is inspired by flathub/com.irccloud.desktop@8e764be Let's see how well this works. The commit 394786b910a1d2ee35455210d923a360a425f03d of f-b-t says, that electron/get >= 1.12.4 has a new behaviour. But we don't have such a new version. Well, let's give this a try, anyway.
Started test build 42130 |
Build 42130 failed |
dunno, really. Seems Electron is making trouble, somehow. Maybe 11.2 would help (currently, it seems to be pulling 11.1). At least, that's what seems to be working for flathub/com.irccloud.desktop@8e764be I don't understand, how |
so we should release an update that updates electron to 11.2? |
yeah, I mean, I am out of ideas what causes npm to not find the downloaded dependencies :-/ |
there you are https://github.com/deltachat/deltachat-desktop/tree/1.15.x_electron_upgrade |
Started test build 42161 |
Build 42161 failed |
This reverts commit 120f7c9.
updated electron builder in that branch |
Started test build 42170 |
Build 42170 failed |
python3 ~/vcs/flatpak-builder-tools/node/flatpak-node-generator.py npm ~/vcs/deltachat-desktop/package-lock.json --recursive -s --output generated-sources-npm.json
Started test build 42197 |
Build 42197 failed |
This reverts commit 22755ff.
Forgive me, I have no idea, what I'm doing.
Build 42203 failed |
Started test build 42205 |
Build 42204 failed |
Build 42205 failed |
Ouf, what a mess. Sorry. Looks much better now:
Looks like it has been forgotten to add That script turns out to be fragile:
|
the script tries a 3 methods of getting a git reference. Did all of them fail? maybe we need a special case for flatpak or its really unstable. looking up that error I found this stackoverflow answer https://stackoverflow.com/questions/6445148/git-fatalno-tags-can-describe-sha1-number/6445255#6445255 which suggests to do a |
yeah, you're right. But calling check types fails:
And by the time check types is run, the build hasn't yet been performed, so the shared thing hasn't been built either. |
It seems that the Electron update is not necessary. At least the npm install step seems to succeed.
This seems to be required in order for build4production to run.
Started test build 42313 |
Build 42313 failed |
This tries to find out why it can't find the zip files it's looking for. Because they sort of exist! Interestingly, it looks for both ../chromedriver-v10.0.0-linux-x64.zip and ../electron-v11.1.0-linux-x64.zip.
Started test build 42400 |
that's strange, because the file does indeed exist.
So somehow it seems to look for the file in the wrong place. If the CWD line in the log is indeed correct, i.e. Note how it seems to be looking for both chromedriver and electron. |
Build 42400 failed |
did the electron upgrade help? |
Started test build 42517 |
Build 42517 failed |
no :( Other apps use electron-builder-22, so it should work, I guess. They use yarn, though, but I don't think it should matter too much here, because the failure is past installation stage. |
log looks to me like it looks in anyways in the future we might be able to use electron-userland/electron-builder#5711, maybe you can try a build using it in a separate branch (install the branch of the electron builder pullrequest) 🤷 |
Hm. How do you arrive at that conclusion? That's the target, no?
Hm. No, I don't think this is relevant. Notice how the description also mentions that flathub is future work. |
this section in the log:
|
yeah, and I read it as if it's trying to symlink from I don't know how to proceed here. Let's hope @refi64 has seen this type of failure before and knows what to do. |
this has been superseded |
No description provided.