Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
RE: [orbit-dev] Check out these beautiful new download pages!

Hi Dave,
 
thanks for these great enhancements!
 
I'd think that a contact info is in order, if it's shown in a format that's readable
by humans only but not by spambots.
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 


From: orbit-dev-bounces@xxxxxxxxxxx [mailto:orbit-dev-bounces@xxxxxxxxxxx] On Behalf Of David M Williams
Sent: Dienstag, 11. März 2008 02:44
To: Orbit Developer discussion
Subject: RE: [orbit-dev] Check out these beautiful new download pages!


>  Is that a happy coincidence or is it the case that we are only shipping ones that have Orbit CQs?

The later, all bundles have Orbit CQs ... with 2 qualifications: 1. I didn't actually check all the CQ numbers -- that they exist, that they are in tools.orbit -- the page just finds and lists the CQ numbers in the "ip_logs" files that committers have provided, and 2) since I was making the webpage to report missing ip_logs, before I announced it, I made sure there was an ip_log for everything  :)    There was a CQ for all the missing ones I added, but in some cases was not correctly entered in the tools.orbit project, so I entered ones for those cases.

The page basically lists all the bundles we build (not necessarily all the ones in cvs .. as others have pointed out (DJ?) ... I think there is some in cvs that are not built, since not in the feature.xml file ... not sure if that's intentional, or an oversight.

And, then once the page has the list of all the bundles we build, the page searches through all the ip_log files looking for matching bundle IDs and versions, and then "merges" the data together. Besides listing a found CQ number, two other outcomes are "No IP Log file exists" or "No CQ for this version exists".

It should make it easier to "manually" check that all bundles are covered, and provide an ongoing quick check as new bundles are added. Long term, I suppose, someone could write SQL queries to look up the CQ number and verify the data ... but, not sure it's worth the effort and/or we could have JUnit tests to flag an error if data was missing ... but, I like the visual-and-hyperlinking display of information better as a starting point. The other question I hear a lot is "who should I contact about bundle so-and-so" ... so, if no objections, one other piece of data I'll display (eventually) is a column of "contacts" for each bundle.








From: "Jeff McAffer" <jeff@xxxxxxxxx>
To: "'Orbit Developer discussion'" <orbit-dev@xxxxxxxxxxx>
Date: 03/10/2008 07:17 PM
Subject: RE: [orbit-dev] Check out these beautiful new download pages!





This is great!  I especially like that all bundles on the page have corresponding Orbit CQs!!  Is that a happy coincidence or is it the case that we are only shipping ones gthat have Orbit CQs?
 
Thanks David for all your work on this.  Very useful.
 
Jeff
 
From: orbit-dev-bounces@xxxxxxxxxxx [mailto:orbit-dev-bounces@xxxxxxxxxxx] On Behalf Of David M Williams
Sent:
Monday, March 10, 2008 4:34 AM
To:
Orbit Developer discussion
Subject:
[orbit-dev] Check out these beautiful new download pages!

 

Beginning this week, I've tried to convey more information in the download pages, so it will be easier to tell when
a bundle may or may not have the proper book-keeping files and CQs in place.


For example, see

http://download.eclipse.org/tools/orbit/committers/drops/I20080310070939/

Eventually I may add more info to this summary page, to help others (committers and legal staff) to check for accuracy, etc. .... and, who knows,
maybe it'll even be pretty one day. :)


So ... please click around and make sure _your_ links are accurate and ... harder to spot ... that there are none omitted.


Thanks,
_______________________________________________
orbit-dev mailing list
orbit-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/orbit-dev


Back to the top