This makes these lists directly useable.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>

ISTM that these are what most people will be looking for most of the
time.  It's probably worth making a fast path for them.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>

We only support, here, a subset of the possible purposes.

This serves both to avoid html escaping and also to limit the size of
the drop down (which otherwise contains a lot of singleton purposes).

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>

This is compared (literally) with the purpose values in LDAP.

Currently there is nothing in the form to cause this to be sent, but
URLs like
   machines.cgi?purpose_search=porterbox
will now work.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>

We are going to want to put something else under this same condition
in a moment.

No functional change.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>

If this is set, read that as our config, rather than the hardcoded
file in /etc.  This will make it possible to test this code from a
clean git tree.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>

An example of how to use this will come in a moment.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>

Thanks to weasel for spotting

doc-mail: drop single quotes around 'foo IN TXT plain text here' in the example as these are probably wrong.

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

This brings the functionality more in line with the previous
incarnation.

The configuration for the paging functionality is left in place to
make it simpler to enable / disable as desired, and potentially
allow users the option at some future point.

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Based on an example from datatables.net

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

From https://datatables.net/



Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

This avoids needing the Content-Security-Policy to allow "unsafe-inline"

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>