py-cheetah: Update to version 3

[mf2k (Frank Schima)]

py-cheetah appears to be forked to ​py-cheetah3. Should the existing py-cheetah port simply be updated or should a new py-cheetah3 port be created?

Here are the dependents:

$ port echo depends:py27-cheetah
gnuradio                        
gnuradio-devel                  
gnuradio-next                   
py27-tahchee                    
py27-turbocheetah               
volk                         

[mf2k (Frank Schima)]

See also ​https://github.com/macports/macports-ports/pull/1110

[michaelld (Michael Dickens)]

If they can live side by side, I'd vote for keeping cheetah as is & adding in cheetah3.

[mf2k (Frank Schima)]

@michaelld: Can gnuradio work with cheetah3?

[michaelld (Michael Dickens)]

No idea. Do you know how backward compatible 3 -> 2 is? Maybe I'll give it a go & see what comes of it.

GR devs are working on moving away from cheetah, but that's still a bit down the road since cheetah is used extensively.

[mf2k (Frank Schima)]

Replying to michaelld:

Do you know how backward compatible 3 -> 2 is?

I don't know.

[mf2k (Frank Schima)]

While I have not tested it, it appears they would conflict because they both use the same package directory "Cheetah".

[emcrisostomo (Enrico Maria Crisostomo)]

I asked upstream: ​https://github.com/CheetahTemplate3/cheetah3/issues/2 There's not a lot of activity on that repository but perhaps we are lucky.

[emcrisostomo (Enrico Maria Crisostomo)]

The author of Cheetah3 says he reluctantly forked because of Cheetah's author inaction. The forks should be a drop-in replacement:

The worst deficiency for Cheetah 2 is that it only supports Python 2. I needed to update it for Python 3. The old repository has a number of pull requests — I applied them to Cheetah 3. Also I received a few pull requests for Cheetah 3. And I implemented a number of small features, modernized build and test infrastructure, updated docs to be build with Sphinx. I continue working on it but I don't plan any revolution. Not even major features.

Given this, the fact that Cheetah2 hasn't been updated in 7 years and the fact that Cheetah2 is being used by just a handful of ports, both options are feasible. Perhaps the simplest one would be to update py-cheetah to use Cheetah3 and test the ports which currently use it.

[mf2k (Frank Schima)]

@michaelld: Would you be willing to update the py-cheetah on your computer to version 3 and test your ports against it?

• py-turbocheetah looks dead because upstream has not been updated since 2007 and only officially supports python 2.5.
• py-tahchee: appears to be abandoned.

If michaelld gives the go ahead, let's just update the existing py-cheetah port to version 3.

[michaelld (Michael Dickens)]

yes it's on my queue. will try it hopefully tomorrow or Wednesday & report back here.

[michaelld (Michael Dickens)]

As far as I can tell, Cheetah3 plays nicely with GNU Radio and Volk, or at least their specific use of Cheetah. So I'm going to go ahead and update the py*-cheetah port to 3.

[michaelld (Michael Dickens)]

In a3cf774715fda60e83c4d90b31fa388fd8b0ecb1/macports-ports:

py*-cheetah: update to 3.0.0

Closes: #55512