Re: [hyde-dev] Current state of Hyde

[Ben]

As an outside observer, this is what I can say:

The source code works to some extend, but there is a severe lack of documentation for anyone trying to do something with it. Looks like the project maintainer got other priorities, which could also explain the lack of reactivity on the pull requests.

However, if you dare asking you question on this list, chances are big that they will get answered.

Regards,

Ben.

On Sunday, August 12, 2012, Jens Rantil wrote:

Hi,

I am looking into Hyde right now and wondered; What's the current state of Hyde? I am asking for the following reasons:

 * http://hyde.github.com contains broken images on its front page.
 * There are 24/31 pull requests (to [1]) are lacking any type of comments.
 * Last commit to [1] was more than three months ago.

I've seen references to [1] saying that it is under active development. But really, it doesn't look that way. Or have Hyde development moved elsewhere (again)?

[1] https://github.com/hyde/hyde/

I'd love to hear you input,
Jens

--
You received this message because you are subscribed to the Google Groups "Hyde" group.
To post to this group, send email to hyde...@googlegroups.com.
To unsubscribe from this group, send email to hyde-dev+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msg/hyde-dev/-/rUKx2T2XYCEJ.
For more options, visit https://groups.google.com/groups/opt_out.

[Lakshmi Vyas]

Ben - thanks for sharing a very good observation :) Thats pretty accurate I would say.

----

To explain:

1. I consider hyde's codebase to be in a stable state. I manage several hyde websites (one very large and others smaller) and I find that it works 

     consistently although always not as desired.

2. I have identified some core issues with hyde(see below) that I want to address before merging the pull requests. Simply merging the pull requests:

a. Might make the code base unstable

b. Might make it harder to make the changes

3. I have some scheduled time coming up where I can completely focus on hyde and I am planning to:

a. Make the changes/fixes below

b. Merge pull requests

c. Add additional layouts that cover more scenarios

d. Complete documentation

----

These are some of the things I want to fix properly (and to the extent possible in a backward compatible manner) before I merge pull requests:

1. The core model object with expando attributes is not very friendly considering hyde enhances the model with user provided text.

2. The auto generation is not robust. It ignores some changes while processes others. I have noticed that this is essential for being productive 

     with larger sites. I am also considering using watchdog instead of just in time generation in the server module.

3. I am not happy with the chaining / composability of plugins

4. I consider the django support pull request to be very important and probably pushes hyde to 1.0 state. I want a lot of tests / refactoring of the 

     template module to make sure its very easy to add support for other tempting engines.

----

I know the lack of documentation is probably the worst issue with hyde. Until the time its complete, I will be more than happy to address any 

questions you have on this group or on a personal email. The only requirement from me is that, if the problem is not obvious to diagnose, I 

would like the source (or the minimal subset of it) to be shared in some manner.

Thanks

Lakshmi

[Jens Rantil]

Late reply - sorry for that.

Anyway, thanks you for your observation, Ben! ;) Also, thanks for taking the time to answer, Lakshmi!

I'll respond inline. See below.

Den måndagen den 13:e augusti 2012 kl. 04:45:28 UTC+2 skrev Lakshmi:

Ben - thanks for sharing a very good observation :) Thats pretty accurate I would say.

----

To explain:

1. I consider hyde's codebase to be in a stable state. I manage several hyde websites (one very large and others smaller) and I find that it works 

     consistently although always not as desired.

2. I have identified some core issues with hyde(see below) that I want to address before merging the pull requests. Simply merging the pull requests:

a. Might make the code base unstable

b. Might make it harder to make the changes

Sure, merging pull requests without code review is not a wise thing to do. However, I see there are some tests, how well do you trust them? Also, how about adding Travis CI to Hyde to automatically test all the pull requests? At least that would minimize the effort a bit when going through them eventually. I'll be willing to create a pull request for it, if you are interested in accepting it.

 

3. I have some scheduled time coming up where I can completely focus on hyde and I am planning to:

a. Make the changes/fixes below

b. Merge pull requests

c. Add additional layouts that cover more scenarios

d. Complete documentation

I'm just curious, when do you have scheduled time coming up for Hyde? Since it's been standing still for the past ~7 months... If not in the next two months, "under active development" should probably be removed from the README. :)

 

These are some of the things I want to fix properly (and to the extent possible in a backward compatible manner) before I merge pull requests:

1. The core model object with expando attributes is not very friendly considering hyde enhances the model with user provided text.

2. The auto generation is not robust. It ignores some changes while processes others. I have noticed that this is essential for being productive 

     with larger sites. I am also considering using watchdog instead of just in time generation in the server module.

3. I am not happy with the chaining / composability of plugins

4. I consider the django support pull request to be very important and probably pushes hyde to 1.0 state. I want a lot of tests / refactoring of the 

     template module to make sure its very easy to add support for other tempting engines.

Do all of the above bullets have Github issues created for them? Just based on the fact that there are 66 open issues and 31 open pull requests would make me think someone could be interested in helping out with this if that's all it takes to get the project going again. Or, is this a one-man show?

 

I know the lack of documentation is probably the worst issue with hyde. Until the time its complete, I will be more than happy to address any 

questions you have on this group or on a personal email. The only requirement from me is that, if the problem is not obvious to diagnose, I 

would like the source (or the minimal subset of it) to be shared in some manner.

That's good to know!

Cheers,

Jens