This needed a little sprucing up as it has grown quite organically over the
life of this script. Make things a bit more pythonic through the use of
iterators rather than collection indexing, and try to generalize the special
cases of things a bit.
Also catch encoding problems early and fail gracefully rather than blow up
the entire package parser. A failed decode of a file should cause us to just
skip it rather than stop the entire parser. Worst case, this leaves that
package out of the web interface.
Signed-off-by: Dan McGee <dan@archlinux.org>
We need to do a little dropping into SQL to accomplish this, but it isn't
all that bad to actually do and we can do the whole thing in one query.
Signed-off-by: Dan McGee <dan@archlinux.org>
When I have caught reporead behaving badly on the production box, I haven't
been able to successfully get a traceback without killing the process.
Hopefully using a different signal will allow me to actually capture some
data.
Signed-off-by: Dan McGee <dan@archlinux.org>
Add total package count and incomplete package count columns. Also reduce
the number of total queries by killing the query per row that was happening
before.
Signed-off-by: Dan McGee <dan@archlinux.org>
The one thing that people forget to mention when moving models from one
Django app to another. I was fooled here.
Signed-off-by: Dan McGee <dan@archlinux.org>
Add a column flagging whether this architecture is agnostic (e.g. 'any') or
not. This will remove the hardcoded name checks we have all over the place
and replace it with a boolean.
Signed-off-by: Dan McGee <dan@archlinux.org>
With package pooling we don't put things in the '/any/' folder anymore under
each repository; they only show up in the actual architectures. Use a 'real'
architecture as part of the download link to rectify this. Fixes FS#20779.
Signed-off-by: Dan McGee <dan@archlinux.org>
As reported by Thomas, the dashboard JS dies when we try to apply table
sorting to a table with no entries (those in the top section). Use some
selector magic to only apply tablesorting if the table actually has rows
worth sorting.
Also move the package todo lists table down in the dashboard as the other
two tables are more relevant for the individual developer.
Signed-off-by: Dan McGee <dan@archlinux.org>
This view was getting huge with ~500 items on it, and most people are not
really interested in seeing every single news item. Use the drop in
pagination and add some controls that still allow browsing to any page of
the list.
Signed-off-by: Dan McGee <dan@archlinux.org>
And also add the new style class we will need so the search form retains the
prior look, now that we are using this style elsewhere.
Signed-off-by: Dan McGee <dan@archlinux.org>
This is done as client-side JS which makes the page nice and fast. Minor
versions can be excluded, as can packages in [multilib]. In addition,
architecture filtering is in place so you can limit the subset of shown
packages to those in any, both, one or the other.
Signed-off-by: Dan McGee <dan@archlinux.org>
Add a second file for use in production environments where we know we
use memcached and MySQL, and for those that want to replicate the
production environment as close as possible. Slim down the primary file
to those requirements the app directly needs rather than how it is
installed. Also bump the version of South to the newer point release.
Signed-off-by: Dan McGee <dan@archlinux.org>
South actually makes this relatively painless if you get everything right,
so might as well start getting these out of the legacy main application to
eventually eliminate models being separate from their views.
Signed-off-by: Dan McGee <dan@archlinux.org>
Rather than the need to include RequestContext() calls directly, we can
just use direct_to_template to do all the work for us.
Signed-off-by: Dan McGee <dan@archlinux.org>
We can't cache most of the package actions links stuff because we look at
the user there, but we can cache everything on the left and the content of
the bottom boxes.
Signed-off-by: Dan McGee <dan@archlinux.org>
This will allow clicking through to a few more associated packages from a
given package, including those for the other architecture and potentially in
other repositories. The box floats to the right with the package actions,
which have also been restyled slightly so the whole area is a bit more
styled and set off from the rest of the page.
This should allow us to close the current oldest bug open in the Arch bug
tracker, FS#7787. I remember the days when every bug was a four-digit
number...
Signed-off-by: Dan McGee <dan@archlinux.org>
E.g. if we are looking at one version for a particular architecture, I want
to see all versions for all architectures in all repositories.
Signed-off-by: Dan McGee <dan@archlinux.org>
Add another custom parser to the mix here so we can sort by package count,
which just involves extracting the digits from the table cell contents and
ignoring the text.
Signed-off-by: Dan McGee <dan@archlinux.org>
Because it was a link, the sort didn't properly sort on the
complete/incomplete status when viewing a todo list. Add a custom parser to
fix it.
Signed-off-by: Dan McGee <dan@archlinux.org>
We had a bunch of extra imports, non-conventional variable names, spacing
issues, etc. that were relatively low-hanging fruit to clean up. Fix them
and make the code a bit cleaner in the process.
Signed-off-by: Dan McGee <dan@archlinux.org>
Make it more visible to the public, and cache the results of the group
information function since it was designed and made separate for that
purpose and the results can safely be reused without needing to worry too
much about it being stale.
Signed-off-by: Dan McGee <dan@archlinux.org>
Just another annotation to the queryset to get this data, and a little more
manipulation in the group data function. This will help when adding a
sitemap in a subsequent commit.
Signed-off-by: Dan McGee <dan@archlinux.org>
This makes the support asked for in FS#19801 a lot more useful since we
actually have an overview page for the entire group.
Signed-off-by: Dan McGee <dan@archlinux.org>
