Hi Sebastian,
* Sebastian Pipping <webmaster@hartwork.org> [2008-01-06 01:03]:
> Don Armstrong wrote:
> >There's no SQL to talk to. If you want actual accurate data, the SOAP
> >service is the thing to talk to. [Anything else has various delay
> >lengths in it before you actually see the data reported.]
>
> Where does SOAP get the data from?

http://bugs.debian.org:80/cgi-bin/soap.cgi

Have a look at: http://wiki.debian.org/DebbugsSoapInterface

Kind regards
Nico
--
Nico Golde - http://www.ngolde.de - nion@jabber.ccc.de - GPG: 0x73647CFF
For security reasons, all text in this mail is double-rot13 encrypted.

Nico Golde wrote:

http://bugs.debian.org:80/cgi-bin/soap.cgi

Have a look at: http://wiki.debian.org/DebbugsSoapInterface


I guess you misunderstood my question. Again:
Where does SOAP (itself) get the data from?



Sebastian


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Jan 6, 2008 9:49 AM, Sebastian Pipping <webmaster@hartwork.org> wrote:
> Nico Golde wrote:
> > http://bugs.debian.org:80/cgi-bin/soap.cgi
> >
> > Have a look at: http://wiki.debian.org/DebbugsSoapInterface
>
> I guess you misunderstood my question. Again:
> Where does SOAP (itself) get the data from?

The debbugs "database" is a bunch of on-disk files, presumably it gets
the data from there like the rest of debbugs does.

--
bye,
pabs

http://wiki.debian.org/PaulWise


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On 11256 March 1977, Sebastian Pipping wrote:

>> http://bugs.debian.org:80/cgi-bin/soap.cgi>
>> Have a look at: http://wiki.debian.org/DebbugsSoapInterface
> I guess you misunderstood my question. Again:
> Where does SOAP (itself) get the data from?

Its the BTS itself, so from its data storage. And no, thats not SQL or
something.

--
bye Joerg
<liw> HE, we had sex in Debian for many years, yes, before I put a stop
to it


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Sun, 06 Jan 2008, Sebastian Pipping wrote:
> Don Armstrong wrote:
>> There's no SQL to talk to. If you want actual accurate data, the SOAP
>> service is the thing to talk to. [Anything else has various delay
>> lengths in it before you actually see the data reported.]
>
> Where does SOAP get the data from?

From the exact same place the BTS does; the on-disk files in various
different formats, none of which are SQL databases.

SOAP exposes the actual functions that the BTS itself uses to display
information; if there's something that it doesn't currently do that
you want to, feel free to file wishlist bugs or e-mail me or owner@
(or talk to me on IRC.)


Don Armstrong

--
Frankly, if ignoring inane opinions and noisy people and not flaming
them to crisp is bad behaviour, I have not yet achieved a state of
nirvana.
-- Manoj Srivastava in 87n04pzhmh.fsf@glaurung.internal.golden-gryphon.com

http://www.donarmstrong.com http://rzlab.ucr.edu

Don Armstrong wrote:

There's no SQL to talk to. If you want actual accurate data, the SOAP
service is the thing to talk to. [Anything else has various delay
lengths in it before you actually see the data reported.]


i started working on this project and i have reached the point
where it makes sense to get more eyes into play:

http://debian.binera.de/wnpp/

it's a start, not a full solution yet.


-- status quo details
my wnpp page is run by two PHP 5 scripts: one is syncing a MySQL database with
BTS, the other one allows to browse the current content of the database. syncing
currently means adding new bugs and removing old bugs, but not yet updating the
others, e.g. currently bug retitles will not show up there.


-- todos
* decide how to best setup the cron job
* update existing database entries
* make it pretty
* make it bullet-proof
* make it more usable


help with the todos, as well as any feedback is very welcome.



sebastian



--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Mon, 2008-01-14 at 23:43 +0100, Sebastian Pipping wrote:
> Don Armstrong wrote:
> > There's no SQL to talk to. If you want actual accurate data, the SOAP
> > service is the thing to talk to. [Anything else has various delay
> > lengths in it before you actually see the data reported.]


> my wnpp page is run by two PHP 5 scripts: one is syncing a MySQL
> database with
> BTS, the other one allows to browse the current content of the
> database. syncing
> currently means adding new bugs and removing old bugs, but not yet
> updating the
> others, e.g. currently bug retitles will not show up there.

Have you thought about not relying on a cron job but doing the whole
thing via the PHP SOAP interface? I'm using PHP and SOAP to query the
BTS for Emdebian.
http://www.emdebian.org/bugs.php
(code available in the Emdebian SVN repo)
http://buildd.emdebian.org/svn/browser/current/website/trunk/english

The actual SOAP interface in PHP is almost trivial - it only takes 5
lines of PHP to replace all the MySQL interface code (or sqlite in my
case).

> i started working on this project and i have reached the point
> where it makes sense to get more eyes into play:
>
> http://debian.binera.de/wnpp/
>
> it's a start, not a full solution yet.

It could do with some sub-divisions, separating ITP from RFP etc.

Also some form of re-ordering using the PHP query-string? To allow
sorting by age, title, type and even description. (Think how Apache does
index pages for directories or the various RCS web frontends allow
sorting by age etc.)

One very useful addition would be an internal search - help people find
out if an RFP or ITP already exists before filing a new ITP etc. Sorting
by description or project could help with that but a full text search
would be v.useful. The hardest thing to do right now is work out if
someone has already filed an ITP or RFP when preparing an ITP myself.
There are so many that my eyes just glaze over when browsing the WNPP
list and in-browser text search isn't always friendly.

--


Neil Williams
=============
http://www.data-freedom.org/
http://www.nosoftwarepatents.com/
http://www.linux.codehelp.co.uk/

Neil Williams wrote:

Have you thought about not relying on a cron job but doing the whole
thing via the PHP SOAP interface? I'm using PHP and SOAP to query the
BTS for Emdebian.
http://www.emdebian.org/bugs.php
(code available in the Emdebian SVN repo)
http://buildd.emdebian.org/svn/browser/current/website/trunk/english


yes, i have. i need some sort of cache here because i cannot
send traffic for 1500 HTTP request per page view and also
not wait for them to return. actually about 200 requests is
where PHP's runtime limitation comes into play on the machine
running binera.de . so using SOAP alone was no option to me.





It could do with some sub-divisions, separating ITP from RFP etc.


i plan to add a checkbox for each kind so you can have any
combination.




Also some form of re-ordering using the PHP query-string? To allow
sorting by age, title, type and even description. (Think how Apache does
index pages for directories or the various RCS web frontends allow
sorting by age etc.)


have you tried clicking the links in the table header?
currently sorting by age, type and project is possible.




One very useful addition would be an internal search - help people find
out if an RFP or ITP already exists before filing a new ITP etc. Sorting
by description or project could help with that but a full text search
would be v.useful. The hardest thing to do right now is work out if
someone has already filed an ITP or RFP when preparing an ITP myself.
There are so many that my eyes just glaze over when browsing the WNPP
list and in-browser text search isn't always friendly.


adding a search is a good idea.

btw thanks for your PHP code on the wiki page. without that i wouldn't
have anything now. i think the first PHP example is missing some
"$bugnumber = X"; line.



sebastian


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Neil Williams wrote:

It could do with some sub-divisions, separating ITP from RFP etc.


turning on/off each type is now possible.




One very useful addition would be an internal search [..]


Filtering by project name and/or description is now possible.
An example filter would be

"unicode -font"

All positive words have to match, all negative words must not.
Words starting with '+' or '-' are used this way:

++a -+a +-b --b e.g. "perl --perl"


also a sync with BTS can be manually invoked now, link at the top.
check out the new version here:

http://debian.binera.de/wnpp/



sebastian


PS: sorting is not cool yet. i plan to address this next.


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

>>>>> "Sebastian" == Sebastian Pipping <webmaster@hartwork.org> writes:
Sebastian> also a sync with BTS can be manually invoked now, link
Sebastian> at the top. check out the new version here:

Sebastian> http://debian.binera.de/wnpp/

Looking better all the time. I would really like some (approximate)
popcon data myself, especially for orphaned packages. But I'm like,
just this guy you know?

david


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org