Gregor Sturm

So far, I didn't run into any situation where I'd have had to exclude certain files from autoreloading. However, one situation where I believe this could be useful is if...

Thanks for your elaborate response! I was mostly thinking along the lines of a file watcher. A general solution would ensure that it extends to other use-cases as well and...

> Unfortunately the busy polling technique used by ‘testthat’ can’t be used here because it’s synchronous. I see! > But I’m realising that we don’t actually need this: If (and...

How about [TrieSearch](https://www.npmjs.com/package/trie-search)? I racall it to work quite well... But actually, I think a server-side solution would be the way to go, as already the transfer-time is non-negligible and...

Just wondering... is there an alternative by now? This issue is basically the only thing preventing me to switch from Shiny to dash...

Nice, @chriddyp! With the dynamic option available, feel free to close the issue from my side.

To pass through columns with constant value per group like this `old_col=_.old_col.iloc[0])`, couldn't you just include them in `group_by` as additional grouping variable?

It seems this can already be achieved using a `lambda` function: ```python ( fmri >> filter(_.region == "parietal") >> pipe( lambda _: sns.lineplot( x="timepoint", y="signal", hue="region", style="event", data=_ ) )...

The underscore access to the last values is a feature of the IPython shell: https://ipython.org/ipython-doc/3/interactive/reference.html#output-caching-system You could try setting `InteractiveShell.cache_size = 0` in your [IPython config file](https://ipython.org/ipython-doc/3/config/intro.html).

What about not overloading an operator at all, but using a pipe function? There's already one in the [functoolz](https://toolz.readthedocs.io/en/latest/_modules/toolz/functoolz.html) package: [`pipe`](https://toolz.readthedocs.io/en/latest/api.html#toolz.functoolz.pipe) For instance: ```python from siuba.data import mtcars from siuba...