Downgrade to previous version of Spyder

Question

I'm currently using the Anaconda package with Python 2.7.9-1

Yesterday I upgraded the Spyder-app via the Anaconda Launcher and all of my scripts now crash. (Some bugs are already reported in the issues page https://github.com/spyder-ide/spyder/issues ). How can I downgrade to the previous Spyder version meanwhile those bugs are fixed?

Answer 1

You can install a specific version of spyder with conda. For instance, to install 2.3.5.2, run

conda install spyder-app=2.3.5.2

in the terminal.

Answer 2

In Anaconda Navigator, on Spyder tab, you have a settings symbol on top right corner. Press that and select 'Install specific version' option

Answer 3

Thanks for the answers. What worked in the end was the following (I have windows 7 and the Anaconda Package):

1) In the anaconda prompt, type:

conda list -r

this is going to give you a list of all the packages updates you've done in the past.

2) Check the number of the update where you had the previous version of spyder (for example in my case was 2015-06-25 10:43:34 rev 4) and type:

conda install --revision=4

It would ask you if you want to downgrade, etc and thats it! Now all scripts run again.

Answer 4

The previous answers didn't work for me, but going here: ( https://docs.spyder-ide.org/installation.html ). I found the following command works.

conda install -c conda-forge/label/beta spyder=4.1.3

or

conda install spyder=4.1.3*

Note: I had to do this today (7/17/20) where the most recent version is spyder=4.1.4 and I'm getting the following error in the console: "QSocketNotifier: Multiple socket notifiers for same socket 2068 and type Read"

Answer 5

As it is hosted on GitHub, you can find the desired version from the commit history and download that version easily.

First browse the commits:

https://github.com/spyder-ide/spyder/commits/master

After you find the version that doesn't break things. Then you can easily download and install it.

Answer 6

As mentioned before there is an option in the GUI now to select the preferred version.

But before that make sure to run conda update conda . At least in my case that helped.