Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 308096

Summary: "Using from Orbit CQ" function is not limiting use from Orbit
Product: Community Reporter: Barb <barb.cochrane>
Component: IPZillaAssignee: Eclipse Foundation IPZilla inbox <foundation.ipzilla-inbox>
Status: RESOLVED INVALID QA Contact:
Severity: normal    
Priority: P3 CC: barb.cochrane, gabe.obrien, janet.campbell, sharon.corbett, wayne.beaton
Version: unspecified   
Target Milestone: ---   
Hardware: PC   
OS: Windows XP   
Whiteboard: stalebug
Attachments:
Description Flags
screenshot of the choices I got when I asked to use CQ3507 as a third-party library from Orbit none

Description Barb CLA 2010-04-05 14:56:39 EDT
When committers try to piggyback on Orbit CQs, it would be nice if their choice of CQs upon which they can piggyback is actually limited to Orbit CQs.  

Currently, committers can select "using from Orbit" function, yet select a CQ from any project.  The summary line in the resulting CQ erroneously contains the description "using from Orbit CQxxxx".  

This is important for the release train:  the only CQs that releasing projects are *supposed to* piggyback on are Orbit CQs.

For avoidance of doubt, "piggybacked" CQs could be one of three scenarios:

1.  Piggybacking on any project CQ *except* Orbit:  "PB CQxxxx"
2.  Contributing a previously approved CQ to Orbit:  "ATO CQxxxx" 
3.  Piggybacking on an Orbit CQ:  "using Orbit CQxxxx"

It's the third one that doesn't seem to be working properly.  

Recent examples:  
https://dev.eclipse.org/ipzilla/show_bug.cgi?id=3904
https://dev.eclipse.org/ipzilla/show_bug.cgi?id=3905

Note:  for your bug-resolution-planning purposes, I know you are working on https://bugs.eclipse.org/bugs/show_bug.cgi?id=249135; perhaps work on this bug may be somehow related to that one.
Comment 1 Barb CLA 2010-04-30 09:31:39 EDT
Adding another example, and the user described the steps he took that resulted in a CQ erroneously being tagged as "using from Orbit"

https://dev.eclipse.org/ipzilla/show_bug.cgi?id=4045#c3
Comment 2 Gabe O'Brien CLA 2010-04-30 15:55:20 EDT
I took sometime today to really look into this bug.  What I am trying to figure out is what the code should be looking for when picking a CQ. There is no keyword 'inorbit' (or even something like this).  So the only place that seems to make sense to look is the subject?
 
The CQ the code picked (CQ3466) has a Subject of:
   h2 Database Version: 1.1.117 (2009-08-09) (PB CQ3508) 

While the CQ Barb picked (CQ3507) has a subject of:
   h2 Database Version: 1.1.117 (2009-08-09) (ATO CQ3508) 

So, would it be correct to say that the CQs the code should pick would have a (ATO CQXXXX) in the subject?  Or is this too simplistic?
Comment 4 Barb CLA 2010-05-03 16:55:21 EDT
Hi Gabe, I'm not sure if we have a common understanding of the problem yet, so let me try from a different angle.  

Project A enters CQ1234 for a brand new package called Joe's Software v1.  For various reasons, other projects may have entered distinct CQs for Joe's Software v1 as well.  Let's say Project D entered CQ9999 all on its own.  

Project A then wants to contribute that package to Orbit.  Project A would have to create *another* CQ for Orbit.  Let's say that Orbit CQ number is CQ3456.  CQ3456 would essentially "piggybacking" on CQ1234, but it would have a special identifier that distinguishes it from any other kind of piggyback.  The Orbit CQ3456 would normally be identified as follows:  "Joe's Software v1 (ATO CQ1234)"  This signifies that CQ1234 contains the code that is Added To Orbit (hence the "ATO").

Then, let's say Project C knows that package is in Orbit and wants to piggyback off of it.  Let's say their new CQ number ends up being CQ5678.  There is a unique identifier for this type of "piggybacking" as well:  "using Orbit" rather than "PB".  So, Project B goes into the portal and select the option that prompts them to use a package from Orbit.  From this point, we think their choice should be limited to Orbit CQs, but I don't think they are:  

If all worked according to plan, Project B's CQ5678 should read "Joe's Software v1 (ATO CQ1234)(using Orbit CQ3456)".  

BUT....for some reason, what's happening is that Project B's choice when they create CQ5678 is not limited to that one bundle that was contributed to Orbit.  From the place in the portal where they are apparently choosing to use a bundle from Orbit, they can actually choose CQ9999, and their resulting CQ5678 erroneously says "Joe's Software v1 (using Orbit CQ9999)".  

CQ9999 is not an Orbit CQ, so the question is...how is that happening?  I'd be happy to have a call with you if you think that would help Gabe :-).  

Cheers,  Barb
Comment 5 Gabe O'Brien CLA 2010-05-03 18:17:25 EDT
Created attachment 166878 [details]
screenshot of the choices I got when I asked to use CQ3507 as a third-party library from Orbit

Barb,

   Thanks for the recap, it helped allot.  

   The issue with the Portal is not only with the choices the Portal gives to users, but with how it automatically tries to figure out which CQ is really in Orbit after the user selects a CQ (this is the error you are fixing when you update the title of the bugs).

   I followed the steps from CQ comment 2.  The only options the Portal gave me was the same CQ I entered (see the screen shot).  I tried a few different CQs and all just gave me back the same CQ I entered.  The Portal appear to be doing not validation on the CQ that user is picking.

   So I can see two thing the Portal could do when a CQ entered doesn't have the ATO XXXX in the subject:

1) the Portal could display an error message and allow you to enter a new CQ
2) the Portal could try and find the right CQ and show this to the users

#2 seems to be what you are asking for and I will look into how hard this will be to do.  I can see this working where the tittles are 100% the same minus the ATO part.

Let me know if this makes senses, if not then we should schedule a time to have a phone conversation so we can get on the same page and get this bug wrapped up.
Comment 6 Eclipse Genie CLA 2014-11-17 11:36:30 EST
This bug hasn't had any activity in quite some time. Maybe the problem got resolved, was a duplicate of something else, or became less pressing for some reason - or maybe it's still relevant but just hasn't been looked at yet.

If you have further information on the current state of the bug, please add it. The information can be, for example, that the problem still occurs, that you still want the feature, that more information is needed, or that the bug is (for whatever reason) no longer relevant.

--
The automated Eclipse Genie.
Comment 7 Eclipse Genie CLA 2016-11-07 19:12:03 EST
This bug hasn't had any activity in quite some time. Maybe the problem got resolved, was a duplicate of something else, or became less pressing for some reason - or maybe it's still relevant but just hasn't been looked at yet.

If you have further information on the current state of the bug, please add it. The information can be, for example, that the problem still occurs, that you still want the feature, that more information is needed, or that the bug is (for whatever reason) no longer relevant.

--
The automated Eclipse Genie.
Comment 8 Wayne Beaton CLA 2017-07-27 14:36:18 EDT
I believe that the PMI-based implementation invalidates this request. There is now only one entry point into the CQ creation process and a "from Orbit" designation is implied from the source of a piggyback CQ.