ProgressMeter doesn't work in jupyter lab

[kain88-de]

Expected behavior

A progress meter works the same in a jupyter lab environment and an ipython terminal.

Actual behavior

Currently version of MDAnalysis

• Which version are you using? (run python -c "import MDAnalysis as mda; print(mda.__version__)") 0.18.0
• Which version of Python (python -V)? 3.6
• Which operating system? Linux

[orbeckst]

Is JupyterLab different from a Jupyter notebook? It used to work in the latter, didn't it?

[kain88-de]

Jupyter lab is the *next* interface for notebooks. It's more of an IDE then the normal notebooks. It is a good bit different on the javascript side though. So things are not 100% compatible between them. The tqdm package works as expected though.

…

On Tue, Sep 18, 2018 at 7:13 PM Oliver Beckstein ***@***.***> wrote: Is JupyterLab different from a Jupyter notebook? It used to work in the latter, didn't it? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#2078 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEGnVtxW7cQp6vFwSuRsqN7hxkFOACGlks5ucSnUgaJpZM4Wt54y> .

[IAlibay]

So had a look at this as a nice excuse to wrap my head around the progress meter. It looks like Jupyter Lab doesn't like progress meter formatting assigned by the RMSD class. In particular, it breaks on the use of \r in "{step:5d}/{numsteps} [{percentage:5.1f}%]\r" (line 607).

Removing it seems to fix this issue:

This would explain why the progress meter for u.transfer_to_memory() works properly.

Within analysis, it looks this will also affect density_from_Universe (checked), helanal_trajectory, PCA, HydorgenBondAnalysis, HydrogenBondAutoCorrel, and WaterBridgeAnalysis.

I have tested removing the extra carriage return on RMSD and density_from_Universe, and (on python3) it looks like it works fine across jupyterlab, ipython and standard python calls.

[kain88-de]

That is great. Can you create a PR with the fix?

…

On Wed, Sep 19, 2018 at 7:20 PM IAlibay ***@***.***> wrote: So had a look at this as a nice excuse to wrap my head around the progress meter. It looks like Jupyter Lab doesn't like progress meter formatting assigned by the RMSD class. In particular, it breaks on the use of \r in "{step:5d}/{numsteps} [{percentage:5.1f}%]\r" (line 607). Removing it seems to fix this issue: [image: image] <https://user-images.githubusercontent.com/12460125/45765384-db61c200-bc2c-11e8-9f59-d00a1c937ab9.png> This would explain why the progress meter for u.transfer_to_memory() works properly. Within analysis, it looks this will also affect density_from_Universe (checked), helanal_trajectory, PCA, HydorgenBondAnalysis, HydrogenBondAutoCorrel, and WaterBridgeAnalysis. I have tested removing the extra carriage return on RMSD and density_from_Universe, and (on python3) it looks like it works fine across jupyterlab, ipython and standard python calls. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#2078 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEGnVkclGQO2Iszhfm3AOh2FAdNTUNgMks5ucnzCgaJpZM4Wt54y> .

[IAlibay]

@kain88-de Will do, I'm still trying to work out if anything outside analysis is affected but I'll create the PR once I'm done (and tested on p2.7).

[orbeckst]

@jbarnoud 's PR #960 is still hanging somewhere in limbo.

I also only recently discovered tqdm (which probably everyone else already knows...) and we could also think about adapting it instead of our home-grown approach. It works in notebooks and is a light-weight dependency.

[IAlibay]

Just as an update on the original fix for this, I did have most of this fixed (I think I'm missing a test for some helanal changes related to #2084), just ended up getting distracted with other work. Sorry about that. I'll try to open the PR for it over the next few days.

[orbeckst]

Update: In #928 we decided to transition to tqdm as progress meter. This might eventually solve this problem (although see #960 (comment) on tqdm in Jupyter Lab).

In the meantime. PR #2310 should at least help.

[mquevill]

This issue is fixed with #2617 (after running the extra commands in #2617 (comment)).

[IAlibay]

I've closed the issue here, but I'm going to open a new one in the user guide. We'll probably want to document the extra steps somewhere.