State of Ramaze

[Yorick Peterse]

Dear Rubyists,

I'd like to talk a bit about Ramaze and its current state in particular.
As most of you have probably noticed by now we recently pushed out a new
version that was supposed to introduce a few non backwards compatible
changes that in turn would make Ramaze/Innate a lot faster. While this
was the case it sadly has a few side effects that make the current
release pretty much worthless in my opinion. Currently there are the
following issues:

* `before_all` blocks are not executed properly, it seems they are only
executed the first time (Innate issue #7).
* when calling `respond!` with a status such as 404 and 403 the request
will be executed multiple times due to how Rack::Cascade works (issue
#57).

Both these issues are pretty big show stoppers for the current release.
Although things like these happen I have to admit that I'm not very
happy with how the results eventually turned out.

Traditionally the solution would be easy: find the problem and fix it.
Sadly both me and Michael have been quite busy lately and as such
haven't been able to contribute a whole lot of code to Ramaze/Innate.
On top of that I'm also not an expert on the Innate internals. While the
code is well documented there is a lot going on and as such it can be
quite a pain to debug issues like the ones mentioned above.

Now to be honest I'm not entirely sure how to approach the issue. On one
hand I'd say revert the changes but on the other hand the performance
gain would be rather nice to have. Looking at my schedule for the coming
weeks I highly doubt I have time to sit down for at least an entire day
to figure out what the hell is going on. Michael knows this the best but
I don't want to take away too much of his spare time.

Now what I'm going to ask may be a bit odd but it's quite simple: I'd
like for those that are interested to spend some time debugging the
following two issues:

* https://github.com/Ramaze/innate/issues/7
* https://github.com/Ramaze/ramaze/issues/57

Even something as simple as "The issue is located in file X on line Y"
would save me a lot of time. I've already spent quite a bit of time on
debugging the issue but I just can't seem to find it. I realize this
more or less comes down to "Go fix the problem yourself" but I'm afraid
that due to lack of time and people there isn't really a different
option available, at least not for the coming month or two (if not
more).

I will see if I can spend some time with the Ramaze/Innate code this
week but as mentioned I can't really guarantee anything. If there's
anybody willing to help out feel free to contact me in #ramaze, I keep
an eye on IRC pretty much most of the day with the exception of between
16:30 and 18:30 UTC.

Yorick

[Yorick Peterse]

Just a small heads up, after debugging some of the Innate code together
with Michael and Lasso we've fixed the problems regarding action
caching. Looking at the problems it caused compared to the performance
gain I decided to remove action caching all together and instead cache
two small parts that take up most of the time of a request. These
changes were made in the following commits:

*
https://github.com/Ramaze/innate/commit/d2aeb7f9107da739f155e626fce02807b888d112
*
https://github.com/Ramaze/innate/commit/06dd4c52c18c572dc2cd56b630b2b9b509b6d4ad

Once the last issue for Innate (issue #7) has been confirmed to work
we'll push out a new release of Innate.

Yorick