#pyramid

+1

how do you I can approach the problem asynchronously?

multiple requests from the client right... one to start the job and then later some way to tell the client the job is done

so you mean that i can't compile more than on code for different users at a time ?

one#

[redacted]

[redacted]

dowwie howdy

[redacted]

dowwie yeah... too much work on my plate atm

[redacted]

dowwie problem is how to make a relevant demo

Michael, r u still there :) ?

amr_: sure you can... the stuff I described is "best practice" regardless of which framework you use

how do u think i can implement an intrpreter like that : codepad.org?

well anything that's executing random code on my servers is a dangerous/complicated game

the least of your concerns there should be the web framework

sadly it looks like codepad is executing in containers with no network access :(

I can't run my bitcoin miner on it :(

soryy my connectio disconnected

how do you think i benefit from performance of pyramid in this project? why it's said that it's the fastest among all py frameworks

[redacted]

dowwie: not actually :-)

I read that reddit faces scalability issues when they weren on django, and they had to go with pyramid to solve that

[redacted]

[redacted]

sorry for typos

yes they go with it

[redacted]

reddit was on webpy and then pylons, when were they on django?

reddit went from lisp to pylons

[redacted]

err lisp to webpy to pylons I guess

right, I meant just the python part

[redacted]

what does 'python setup.py build' do? Can I use this to deploy my pyramid app?

dowwie: reddit doesn't use pyramid

[redacted]

they say: Used the Pylons (Django was too slow)

they use an old version of pylons

they never use in prod django to my knowledge

completely tweaked... hardly pylons anymore

yeah right

pylons was used as a base

[redacted]

[redacted]

last commit 16 minutes ago

but the commit says 2 hours ago :P

so it looks active

still uses the pylons controller approach with @validate decorator

funny to look at this, remember pre-pyramid time

so guys If i have to take the decision whther it's pyramid or django, what shall i do?

the one you know more

and feel more confortable

the right answer is to hire blaflamme

no, hire raydeo

I only do what he's saying anyways :P

[redacted]

[redacted]

[redacted]

kazoo is a python binding to zookeeper

written by benbangert

flask is micro-framework!!!!

[redacted]

[redacted]

[redacted]

[redacted]

they are probably using it for leader selection

amr_: first thing to do is to get much more clearer objectives than "need great performance". From what you say Dajngo is not a good option, because you'd end up fighting the parts of the framework that you don't use. Next, is that compiling code idea even workable? Write a proof of concept of what you need to do, use Pyramid, or Flask. See which one you like best

its not so much that we don't want to do your homework for you. It more that we can't. Only you know your skills and objectives. And we are lousy salesmen.

You know, there are intermediate steps between 1. I need a web app, 2. For StackOverflow level audiences with great performance.

@cguardia how much time do I need to be into pyramid, for someone who comes from django background ?

amr_: what's important is Python background. A good Python dev can work well with both

great

what are these intermediate steps you mentioned, sorry for the inconvinience

how comfortable are you with the python ecosystem? Pyramid (and flask) are much more about integration with other stuff that is already available externally to the projects.

But, look at the tutorials and the narrative doc. The docs are pretty good.

yes i found it really good, i'm just a little afraid to not going in depth as quick as i think, as I have to make progress with my grad. project

thank u all for help and support, i really admire this community

In depth questions will usually get answers here (usually of the 3 guys, 5 answer variety, but ...). It is the overview questions that are hard.

[redacted]

[redacted]

#dowwie you're doing it right :D

I doubt he'll be back.

He just wants to install shit and have it work: Django/Plone with high performance and easy to customise…

Aw, he's gone, I was working out the intermediate steps on my head

It is unusual to hear that the docs are not sufficiently in depth, or that they take too long to get there.

It's understandable: wanting to have something working ASAP but being forced to think about things like storage.

jpenny: not from one who wants to get to "how to handle SO level visitors" on page 2, after the installation

We all know that the best modern websites and apps don't need storage. They have magic pixie dust!

[redacted]

[redacted]

dstufft: ping

uralbash: ping

dowwie: asset light strategies only get you so far but they are indeed currently very à la mode.

cguardia, maybe. But, no one person can handle SO levels of traffic. No one can even plan for SO traffic. You have to do it when you start seeing the problems.

at least he's optimistic about his project success :)

jpenny: i think you are being too pessimistic about handling part

jpenny: yes, that was my point when I mentioned "intermediate steps". It's counterproductive to worry about that so early. Lots of wasted time looking for the "best", "fastest" framework

blaflamme: they all are

Ergo: hi

dstufft: did you find the time test paginate?

Ergo: nope, let me do it now

Ergo: master branch riht?

yup

Ergo: so it raises an error, but I can work around it. Basically https://github.com/Pylons/paginate/blob/master/... the boolean on self.collection calls len(). It might be worthwhile to check for page < 1 first, but I can also just override the __bool__ method

after overiding __bool__ in my wrapper class, it seems to be working fine

dstufft: well i think the boolean is not necessary

but original author for some weird reason tests if page is 0 is list is emty

ok maybe i should do len() then