Closed lumberbot-app[bot] closed 2 years ago
@bollwyvl @jtpio
This seems to be msa / Firefox related: https://github.com/wilzbach/msa/issues/257
Xref: jupyterlite/jupyterlite#284
same error on chromium:
jupyterlab: 3.1.2 jupyterlab-fasta: 3.1.1 chromium: 92.0.4515.131
Replace if(this.seqs=[],!t instanceof Array||"at"in t))
by if(this.seqs=[],!t instanceof Array||"on"in t)
.
I guess there was a typo "at" --> "on".
Thanks @guignonv! Would you like to open a PR?
So the upstream library (https://github.com/wilzbach/msa) looks abandoned (last PR in 2018 and the NPM package was pushed 5 years ago).
The root issue isn't in that library, however, but is in https://github.com/wilzbach/stat.seqs, which also looks abandoned. Specifically, it's in https://github.com/wilzbach/stat.seqs/blob/master/lib/index.js#L52.
The issue can't easily be fixed here (unless you do a hacky find and replace at the end of the build stage). It can't be fixed upstream given they seem abandoned.
Thankfully, there's a fork with fixes - https://github.com/niaid/msa. The specific fix is https://github.com/niaid/msa/commit/e82512e8512e46d77e15ff04368596d9504b277a.
Unfortunately, this isn't published to NPM and is a bit of a pain to get working.
"msa"
entry of package.json
in the fasta-extension
folder to point to "https://github.com/niaid/msa"
.bio.io
package not being built.bio.io
in node_modules
and manually 'build' it by running babel -d lib src
.Hopefully, this helps a little...
Thanks @zacharyrs for investigating this.
Thankfully, there's a fork with fixes - https://github.com/niaid/msa. The specific fix is niaid/msa@e82512e.
It would have been interesting to see if they planned to open a PR to merge the fixes upstream. Although as you mentioned it doesn't seem to be actively maintained unfortunately.
Another option would also be to maintain a fork of the msa
library, for example in https://github.com/jupyterlab-contrib (and move the jupyterlab-renderers
there too). But this adds quite a bit of maintenance cost.
Personally I suspect we'd be best to move to their fork.
For reference, NIAID is the National Institute of Allergy and Infectious Diseases, so I suspect their fork will be well maintained.
I've contacted them to find out if they'll be publishing it to NPM, else I suppose we'll need a fork...
Hey @jtpio! Unfortunately I don't thing NIAID will be publishing their fork to NPM. I think it might be worth forking it to https://github.com/jupyterlab-contrib and publishing that to NPM ourselves.
Thanks @zacharyrs for following up here.
Would you like to help with this and releasing to npm? Probably we could then fork to https://github.com/jupyterlab-contrib and add you as maintainer of the repo?
The package could be published under the @jlab-contrib/msa
name for example.
Hey @jtpio, I'm keen to give it a go but I'll be honest that I'm not sure how much time I have to dedicate long-term. I can at least get the NPM builds working though so we can use it!
@zacharyrs I fork the NIAID msa repository in jupyterlab-contrib and you should have receive an invitation to become admin: https://github.com/jupyterlab-contrib/msa
Would you mind giving me your npm user id so I can invite you to the NPM organization?
@zacharyrs I fork the NIAID msa repository in jupyterlab-contrib and you should have receive an invitation to become admin: https://github.com/jupyterlab-contrib/msa
Awesome! I've just accepted that.
Would you mind giving me your npm user id so I can invite you to the NPM organization?
Should be zacharyrs
.
Thanks!
Thanks I invited you to the jlab-contrib npm organization. Let us know if you have trouble publishing the package.
Released in https://pypi.org/project/jupyterlab-fasta/3.2.0
Description
The
jupyterlab-fasta
extension does not work with3.1.0rc2
Reproduce
Can be reproduced with this gist:
Expected behavior
The
jupyterlab-fasta
extensions should work out of the box with JupyterLab 3.1.Context
Originally reported by @bollwyvl in https://github.com/jupyterlite/jupyterlite/issues/284
Originally opened as jupyterlab/jupyterlab#10675 by @jtpio, migration requested by @blink1073