launchpad-dev team mailing list archive
-
launchpad-dev team
-
Mailing list archive
-
Message #09761
Re: getting the source
On Mon, May 25, 2015 at 12:28 PM, William Grant
<william.grant@xxxxxxxxxxxxx> wrote:
> On 25/05/15 19:02, anatoly techtonik wrote:
>> https://dev.launchpad.net/Getting
>>
>> 1. I'd place tl;dr section at the top with:
>> bzr branch lp:launchpad
>
> The "Getting it" section already says that, in a small section that's
> visible without scrolling. The page's introduction includes important
> information about supported environments and copyright licences, so a
> TL;DR wouldn't be a good idea.
Then it is better to make the keywords bold. Including all this stuff
doesn't mean page becomes more attractive and people like me will
actually read it. Some facts:
http://uxmyths.com/post/647473628/myth-people-read-on-the-web
>> 2. The download process is long (15+ minutes already)
>> Is there a way for bzr to download only last 1000 revisions?
>> Something like `git clone --depth 1000 ...`?
>> It would be worthy to mention there.
>
> Bazaar doesn't implement shallow branches. You need to wait for the
> entire branch to download.
>
> Now that Launchpad has beta Git hosting support
> (http://blog.launchpad.net/general/git-code-hosting-beta), we intend to
> move the code for Launchpad itself to Git in the coming weeks. There's a
> test import of our Bazaar branches into Git at the top of
> https://code.launchpad.net/launchpad, from which you can do a shallow
> clone if you desire. But it's not always up to date, and not supported
> by the Launchpad team yet.
That would make my life easier. Is it possible to update the page with this
info? And also update Running page with info how to run LP if you already
have the copy of source code?
>> 3. The next page to be read is advised to be
>> https://dev.launchpad.net/Running
>> which uses another method. Why? Who needs to
>> get the source without running it for development?
>
> Some people just want to look at the code, for example to work out why a
> particular piece of Launchpad behaves the way it does. It's much quicker
> to download it that way than using the full rocketfuel-setup script.
If you're going to download unusable code, it should be mention at the top of
https://dev.launchpad.net/Getting as a first thing.
Now after 20min+ I have a copy of sources, read them and interested
to see how it really works. I need to repeat the download process now?
https://dev.launchpad.net/Running
> But if you want to run it, you want to use rocketfuel-setup. That's why
> there's no TL;DR -- you don't want to bzr branch lp:launchpad if you
> want to do more than look at it.
This is also more complicated than necessary:
bzr --no-plugins cat
http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/utilities/rocketfuel-setup
> rocketfuel-setup
why not
wget http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/download/head:/rocketfuelsetup-20070604080655-qvg2xintg0bed61s-1/rocketfuel-setup
? Is it because bazaar web doesn't support pointers to latest file version
and all downloads require specific hash?
P.S. Another bug - the list address is not added to CC automatically.
--
anatoly t.
Follow ups
References