-
Notifications
You must be signed in to change notification settings - Fork 31
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
Document a few dependencies more clearly in the README, and also work on fixing the build #555
Conversation
The build will still fail to dependency issues (documented in biocore#555), but when those are fixed it should at least pass.
@fedarko, while the issue in IOW is sorted out, would we be able to pin the version requirement? |
@ElDeveloper Just pinned the iow version requirement to 0.1.3 for now. This fixes the build (McHelper is broken which causes the "Main CI + McHelper" build to fail, but all of the installations / tests pass). I think 0.1.3 is the only iow version that will work with EMPress (see https://pypi.org/project/iow/#history): earlier versions of iow will not have the updates Erfan added in 2020, and later versions of iow will have the problematic linear-tree check. Thanks! 🌲 🌳 🎄 |
Thanks so much @fedarko |
Thanks @ElDeveloper!!!
…On Mon, May 23, 2022, 7:39 AM Yoshiki Vázquez Baeza < ***@***.***> wrote:
Merged #555 <#555> into master.
—
Reply to this email directly, view it on GitHub
<#555 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AA736P2XCPARQAMPTYB34V3VLOKAHANCNFSM5T2RMXXQ>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
This PR adds an "installation troubleshooting" section to the README; this addresses #550 and #553. It also adds two short places indicating where users can get support (opening an issue / via the Q2 forum).
Update: Multiple tests in the build are failing due to changes in the dependencies' behavior:
One of the failures is due to
biom-format
now explicitly raising errors whenTable.update_ids()
is called with a dict containing non-string values (as done in this test). This PR fixes this test. I've also submitted a PR tobiom-format
which clarifies this behavior.Another issue (causing two failing tests) is due to our use of "linear" trees causing problems with iow. This time, the problem is not on our side but on iow's, and I contend that we should keep the linear trees in the tests (if only to make sure that these sort of trees remain explicitly supported). I have submitted a PR to iow which fixes the relevant problem.
Even though this build is currently failing, I imagine that it's probably better to merge this PR in sooner rather than later (I'm aware this is me being a hypocrite considering the backlog of things I need to review, sorry...) -- having the installation documentation up-to-date will prevent user confusion.
Thanks!