#pyramid

Show me an OS/distribution that does that by default and I will show you some poor person somewhere that made a HUGE mistake.

not with forwarding, but by routing

Nope.

Won't happen.

it does

Prove it.

Show me the steps I need to take to test this on CentOS 7.0 or Ubuntu 15.04.

Pick one, I don't care.

yeah, not for 127/8

That's what we are talking about here ...

but possible for other ifs if you do not enable the reverse filtering

I've done this before back in times of 2.4 kernel..

Zart: That's a bug, not a feature.

osx :P

it's not a bug

it's an implementation detail of tcp/ip stack

The whole idea behind binding to 127.0.0.1 is that it is not externally reachable.

127/8 isnt magic unicorn

x58 aha! i found an explanation of the relationship between the <1.x and >=2.0.0 versions of deform: https://pypi.python.org/pypi/deform/2.0a1

Zart: yes it is not a magic unicorn in that it still is an ip address and it is possible to have a packet with that IP

but there is no legal use of sending 127/8 over any link

Zart: Alright. I'm going to spin up a service on 127.0.0.1 on a Linux machine that does not have forwarding enabled, 127.0.0.1 is assigned to a loopback, as it should be. And you can try and get the web page.

iptables disabled.

stevepiercy: Glad you foudn it.

so i guess that means the 1.x-branch is not released on pypi yet, although there was a stalled attempt at doing so

Zart: Actually, I am not going to. That would break so many assumptions.

That I am going to claim no go.

stevepiercy: and pip does not install alpha by default :P

ztane: I'd argue that is a bug in OS X, and I hope that is fixed =)

x58: definitely

stevepiercy: Hmm

deform 2 will be out using bs3 the week after bs4 is deprecated.

osx is bsd-based, which has strict address binding

ztane but easy_install will :)

easy_install quietly defaults to ==dev

nasty packagezzzz managerzezzzzz

we don't likes them!

Good morning.

maybe we should write yet another one :P

yappi

damn that is already taken

no more pixie dust for you ztane

morning Charlie_X

i wonder why there just aren't two separate projects, deform and deform2

I seem to remember that that was once the case.

or better, deform and deform-bootstrap

Charlie_X i don't think so

And then they got merged. It seemed like a good idea at the time. I think it was because iElectric and others were working hard on a project and decided that Bootstrap solved lots of problems for them.

sure, so that become deform v2.0.0

but still deform

Well, more bootstrap than deform.

But understandable. HTML forms are such a fucking mess.

hm, made my vm to send traffic to 127.0.0.1 externally, but seems like my openswitch doesnt forward somehow 8\

stevepiercy: Bootstrap is popular for developers building mobile phone apps.

I think it makes less sense for websites.

and tablets

i use bootstrap a lot

I've yet to find a use for it.

i'm essentially trying to sort out the organization of deform's tags/versions/branches, and i'm slowly grokking it

But maybe it will be Twitter's legacy?

it looks like the 1.x-branch in the deform repo is a continuation after the release of deform 0.9.9

That is the case.

and i assume that the 1.x-branch is intended to be released some day to pypi

but...

there is a 2.0.0a2 release on pypi now

If anyone can ever find the energy and time to do it.

so i don't know if it is even possible to do a deform 1.0 release on pypi now

i don't know enough about pypi

Interesting question.

i've put my project up on github https://github.com/cdunklau/paildocket

it's pretty rough still. you can basically ignore the user and checklist views, and most of the templates. but i'd like feedback on the auth policy in paildocket.security if anyone has a chance

stevepiercy: Charlie_X: the idea was that 1.0 could still live and have bugfixes if people wanted support

iElectric ya, that much is clear from what is published https://pypi.python.org/pypi/deform/2.0a1

but it is not clear whether another release of < 2.0 on pypi will happen

i.e., deform 1.0

stevepiercy: if someone does/wants the release, it will happen

stevepiercy: is there something you're interested in there?

iElectric nothing in particular, other than is it possible to do a release of deform 1.0 on pypi, now that deform 2.0a2 is released?

sure it's possible

ok

pip will ignore alpha releases anyway

right

i've looked through some PRs and the 2.0 release blockers in regards to documentation, and it's not too bad

so i figured i could improve the docs

btw, I'm so sad to see posts like this https://impythonist.wordpress.com/2015/09/12/bu...

and author claims that pyramid would be the choice; "But due to its documentation it is a nightmare for beginners"

hahaha

now we know that's not really true, but something does draw such people away if they feel this way.

i mean... i would say pyramid isn't well-suited for beginners to web applications

translation: "I'm too lazy to read documentation"

but saying it's because the documentation is bad is just not true

the troubles I've faced with pyramid docs is that they they're really good once you know what you're looking for

hyriand: don't tell me we need video tutorials :-P

yes, terminology is a tough nut to crack

it's a matter of building up vocabulary

that's one of the things in the pyramid docs that shines

the extensive glossar

y

even in getting started we have http://pyramid.readthedocs.org/en/latest/quick_...

maybe we have *too much* documentation

mostly my own fault though, if I'd started with just reading it all instead of trying to look for something specific, that would've helped a lot

I'd count that as a ~good reason

hyriand: i've done that so many times

hyriand: "i don't need to read the quick start, i know what i'm doing"

hours later... "oh hey the quick start is really freaking useful"

in my case it was more like "this project that uses pyramid was dumped into my lap, I have no use for a quick start"

even worse :D

ok that answers my question :)

sooo i'm considering something bad. writing my own forms libr

i mean... i like colander. like a lot.

but deform is... strange

what do you find strange about deform? it makes a lot of sense to me

cdunklau: I don't want to be hipster and all, but react.js is very similar to form libraries, but just separated from the backend technology

well, honestly the thing that makes me consider writing my own is that deform's current stable doesn't have an easy way to do csrf