[jbosstools-issues] [JBoss JIRA] (TOOLSDOC-521) Review JIRA queries in Release Notes

[Michelle Murray (JIRA)]

     [ https://issues.jboss.org/browse/TOOLSDOC-521?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Michelle Murray updated TOOLSDOC-521:
-------------------------------------
    Sprint: 2014/S18 (29-Sep > 12-Oct), 2014/S19 (13-Oct > 26-Oct)  (was: 2014/S18 (29-Sep > 12-Oct))


> Review JIRA queries in Release Notes
> ------------------------------------
>
>                 Key: TOOLSDOC-521
>                 URL: https://issues.jboss.org/browse/TOOLSDOC-521
>             Project: Documentation for JBoss Tools and Developer Studio
>          Issue Type: Task
>          Components: Release Notes
>    Affects Versions: 4.2.0.CR1
>            Reporter: Martin Malina
>            Assignee: Michelle Murray
>             Fix For: 4.2.0.Final
>
>
> When reviewing RNs for JBDS 8.0.0.CR1, we came to the conclusion that the JIRA queries may need updating.
> For reference, here's is discussion between me and Michelle on the subject:
> {quote}
> 01:23 mmalina: mmurray: I have a question about RN - resolved issues: (project in (JBDS) AND affectedVersion < "8.0.0.CR1" AND fixVersion in ("8.0.0.CR1") OR project in (JBIDE) AND affectedVersion < "4.2.0.CR1" AND fixVersion in ("4.2.0.CR1")) AND type in (Bug) AND resolution in (Done)
> 01:24 mmalina: mmurray: why have the fixVersion there? any special reason?
> 01:24 mmalina: mmurray: because it's 130 issues vs. 200 without it
> 01:24 maxandersen: don't use the < > comparisons. they unforatunely dont work reliably ;/
> 01:26 mmalina: maxandersen: the problem here is there are many jiras without affectedVersion
> 01:26 mmalina: *the main problem
> 01:26 mmalina: + what you say
> 01:27 maxandersen: mmalina: ah yeah. historically affectedVersion is only rarely used.
> 01:27 maxandersen: mmalina: only when we got very scifci issues where its important.
> 01:30 mmalina: mmurray: my only guess why you have it there is that you wanted to avoid jiras found in CR1 and fixed in CR1 - the thinking being that these would be new in CR1 and not present previously, thus you shouldn't list them in "what's fixed since the previous release". but that reasoning doesn't really work - 90 % bugs found in CR1 were already there before previously ;) any other reason to have it there?
> {quote}
> Michelle wrote:
> {quote}
> The search query is a best effort.
> The affects version is as you said, to avoid picking up internally identified bugs from QE/team testing CR1respinA that were fixed in CR1respinB, say. The customer doesn't need to know or care about these.
> The fixed version is a bit more complicated. The problem is that this query goes into the doc and the search results effectively have to remain static for all time - the doc could be hosted for years to come. This CR1 RNs doc should only list the bugs fixed in the CR1 release. If I leave the fixversion out then when bugs identified in <8.0.0.CR1 get resolved for CR2 or GA or even 9.1.0 they will show up in this query and that screws up the doc. The CR1 RNs doc we get replaced with a GA version so it isn't so important for this one but it is for the GA copy and it's easier to keep basically the same query as I iterate the doc towards GA.
> Regards using < and >, they are working effectively from what I can see. To make the comparisons work it just needs the versions to be listed in chronological order on the JBIDE and JBDS jira project pages.
> {quote}



--
This message was sent by Atlassian JIRA
(v6.3.1#6329)