[RFC] hgweb: ease download of binary files

[Christian Ebert]

* Jesse Glick on Wednesday, April 09, 2008 at 14:48:15 -0400
> Christian Ebert wrote:
>> Actually I do this in a similar way. Only, as this a longterm 
>> project, I want to conserve interim pdf files as well.
> 
> I would suggest you use some kind of continuous builder to generate PDFs 
> from sources after every change, daily, on demand, etc. Then you do not 
> need to pollute your sources with build products.

I won't be allowed to install even a slimmed down LaTeX system on
the server. And even if I would, as I don't work with standard
classes, I'd have to version control all the class and style
files that are not created by me too. -- Even on CTAN you can't
get older versions of packages, they don't have version control
(yet); LaTeX is a dinosaur.

If I tried to make sure that I got all those dependencies tracked
I'd never find time to work on actual content -- which is the
main purpose. So I decided for a dirty compromise: track all
changes in the source that I am responsible for, and the main
results.

> As an example, if you 
> use Hudson you can ask for the PDF from a build to be tracked as a build 
> artifact, so it will be archived and easily available from the web 
> interface (both the latest version and historical versions), and the 
> changelog will be visible too.

In an ideal world, yes ;) Again I probably would not be able to
set up something else on the server that is not owned by me.
Already with Mercurial it was quite tricky, and I think I'll be
stuck with 0.9.4 for quite some time on the server -- on the flip
side, I've got branchview working, hehe.

Mind you, if I find time I will look into Hudson, never heard of
it before, thanks for the pointer.

Anyway, I'll just tell people to click on the "raw" button for
now.

Thanks.

c
-- 
_B A U S T E L L E N_ lesen! --->> <http://www.blacktrash.org/baustellen.html>