These are chat archives for spyder-ide/public
Error: Cask 'spyder' is unavailable: No Cask with this name exists.
That says it all: there's no longer a cask available for Spyder because we stopped distributing DMGs a long time ago
View --> Window Layoutsand then select either one of the preset options, or
Reset to defaulton the current one. Finally, you can try switching to the pane via the keyboard with the shortcuts next to each pane, to see if that reveals them.
conda create -n name-your-new-env python=3.6 source activate name-your-new-env conda install spyder spyder
Preferences --> Python Interpereter); you can keep one copy of Spyder and use it with any Anaconda environments, other Python installs on your machine (including system Python), or even kernels started on remote machines over SSH, with full functionality, all at the same time in different Consoles if you wish. All you need to install is the new
spyder-kernelspackage in any environment you want to work in (which has much more minimal dependencies), and you're set.
this would happen when I update spyder
Thanks. I see you're mixing
conda-forge packages with
defaults ones, which is a recipe for the sorts of problems you're reporting since they aren't 100% binary compatible. That's another great reason to keep separate environment, one using only
defaults packages, and one for the
conda-forge ones; if you're going to use one or the other, use only one or the other in a given environment or you're likely to get problems. If you have to mix them for some reason (generally, specially on Linux its best to go all-conda-forge), installing via
conda install -c defaults -c conda-forge will only pull the specific packages that
defaults doesn't have, rather than the whole stack.
for the the one I want to use currently in the console
thus have to change it every time I want to work in a different anaconda environment?
Yes, but it remembers all the interpreters/environments you've previously used with it so you can just select your desired environment from a list, thus it should only take a few seconds to switch once you've used any given environment once. Also, you can always keep multiple consoles open from different environments and switch between them at will just like files in the Editor (or tabs in a web browser), and whichever console is the active tab will be the one that files/lines, etc are run in.
I'm quite impassionate for one or the other. What should I go for when I have the choice?
Go with all
defaults if all your packages are available in defaults; otherwise go with all
conda-forge; of course, you can always keep environments for each and switch between them at will. whichever one you choose, I recommend you "start fresh" with a new environment at this point with your desired packages since your current one is "contaiminated" with lots of one or the other. Often if a package is pure Python, and there's only one specific package on conda-forge that you need (with no non-defaults dependencies), you can install just that one in a defaults environment, but make sure to do
conda install -c defaults -c conda-forge package-name-here so it doesn't grab the conda-forge version of the deps and potentially "dirty" your current env.
I was more thinking of having backlogs for possibly reading-back when I hang around afk.
The chat on the Gitter site itself shows a full backlog (maybe not the embed version), and you can also search or filter it via the options in the left sidebar.
Its actually what's hosting this chat you're reading; I'm accessing it from their web UI rather than the site version actually.
Goodnight and good luck with Spyder!