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
Update idevicerestore to 1.0.0 (+ required libraries to latest stable releases) #105846
Conversation
2ed6e94
to
f1067d2
Compare
Please do one commit per package update to easily backport or revert those commits if necessary. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Do any of those libraries tag releases?
I guess the answer is: kind of :) idevicerestore only has a single tag (1.0.0) that they recently released. My tags are all for the 1.0.0 release. It's hard to say if they will go forward and tag releases with any regularity. |
f1067d2
to
f8a35a4
Compare
@SuperSandro2000 ok, they are now 1 commit per update (1 per library + 1 for the binary) |
@GrahamcOfBorg build libplist libusbmuxd libirecovery libimobiledevice idevicerestore |
@rb2k If you use commit messagess like |
Thanks for the heads up, I'm still new to the nixpkgs review workflow :) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK, ofborg is happy with the update.
Please update the commit messages accordingly (example: b949937), then this should be merged.
f8a35a4
to
c950952
Compare
Too much mercurial at $DAYJOB, trying to revive git memory. Give me a few minutes :D |
@rb2k Yes, changing (usually with I think your last force-push wasn't as intended though, e.g. https://github.com/NixOS/nixpkgs/pull/105846/commits currently shows 12 commits, one of them being a merge commit. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ah bummer, NIX_PATH=nixpkgs=. nixpkgs-review pr --post-result --checkout commit 105846
shows that 2 packages that depend on this update newly fail (they don't fail on master
):
builder for '/nix/store/srn8ghng75qd8zla37d867g3wxxzx0rc-usbmuxd-2019-11-11.drv' failed with exit code 1; last 10 log lines:
No package 'libplist' found
builder for '/nix/store/h4gv10gfy2l6s36y5fdd82c7ik2vdbq2-libgpod-0.8.3.drv' failed with exit code 1; last 10 log lines:
No package 'libplist' found
Maybe libplist
changed its pkg-config
.pc
file somehow?
Edit:
In more detail:
libgpod
fails with:
checking for LIBGPOD... no
configure: error: Package requirements (glib-2.0 >= 2.16.0 gobject-2.0 sqlite3 libplist >= 1.0 gmodule-2.0) were not met:
No package 'libplist' found
Odd message checking for LIBGPOD... no
, given that that's the name of the library itself, but it checks for libplist
.
On nixos-unstable
it looks like this:
checking for LIBGPOD... yes
On usbmuxd
:
checking for libplist >= 1.11... no
configure: error: Package requirements (libplist >= 1.11) were not met:
No package 'libplist' found
On nixos-unstable
it looks like this:
checking for libplist >= 1.11... yes
@SuperSandro2000 just for me to clarify: why are we removing the sha256 fields? Seems like in most other places those are still present, even when using tags with fetchFromGitHub |
a99d8ab
to
20ea8b2
Compare
I am not proposing this. My proposal is about replacing the hardcoded version in fetchFromGitHub with the version already defined earlier which enables easier updates. |
20ea8b2
to
ff6800b
Compare
ok, that seemed weird. Github also redlined the 'rev' lines for me as well. Either way, replaced the version strings with the previously defined version :) |
ff6800b
to
51f54e1
Compare
Result of 16 packages marked as broken and skipped:
7 packages built:
|
I'm running a linux run as well. Currently on 82 of 133 :)
|
@rb2k Yes, though I usually use the
Ofborg only builds the packages at the beginning of your commit messages, or what is manually started by For example, if you change glibc, then ofborg will build glibc, and nixpkgs-review will build the world. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Since the remaining nixpkgs-review
failures are only marked-as-broken packages (thus also on master
):
16 packages marked as broken and skipped:
I think we're good to go here.
@SuperSandro2000 are all the changes you requested done?
which is not what we want because the could lead to immediate failures after merging. Otherwise we would need to rebase PRs before merging. Please do not run nixpkgs-review with your modified NIX_PATH or otherwise I can't really trust it.
Not really. This is nixpkgs-review on darwin which only build a 10th of the rebuilds on linux. |
My linux based nixpkgs-review run finished and I think this means 'everything ok'?:
Only thing I have to do is change my email in the commit since this is a bit of a work thing :) |
51f54e1
to
a7948ac
Compare
Result of 3 packages marked as broken and skipped:
143 packages built:
|
a7948ac
to
f813e93
Compare
@SuperSandro2000 There's certainly merit on trying to build on current
Ah sorry, I had not spotted the platform in your comment. |
Thanks for the big update, @rb2k! |
If something in master failed breaks the PR than it is broken after we merged it and we did not gain anything and maybe even broke master. |
Motivation for this change
It's been a few months and there is a new stable upstream release (1.0.0).
Since idevicerestore depends on the libraries, I wasn't sure if I should split this into multiple commits or leave it as a single one. Feedback very welcome!
Things done
sandbox
innix.conf
on non-NixOS linux)nix-shell -p nixpkgs-review --run "nixpkgs-review wip"
./result/bin/
)nix path-info -S
before and after)