Re: [windup-dev] Automatic upgrading (of the whole app) discussion
by Marek Novotny
First of all I think this gather to windup-dev list.
Secondly I think the safest way is to clone the current distribution and
do the upgrade on it and then before shutdown the furnace just copy
cloned and upgraded distro back to current place.
On 28.7.2015 03:43, Ondrej Zizka wrote:
> Hi all,
>
> what should be the approach to upgrading?
>
> Current implementation simply replaces the files on the fly (as a
> command), and it seems it isn't bulletproof. Changing files under
> Furnace's hands seems to break things. I get various FileNotFound and
> dependency failures. Also, we currently do not have a way to run other
> command than windup-migrate-app.
>
> Other approach could be to download the file, unzip to a temp directory,
> and right before Windup's process shuts down, after Furnace is closed,
> it would swap the existing and the new directories.
> So next time, the new version would boot.
> Would that be ok?
>
> Then we would need to address the possibility of rules changed by the
> user inside of the windup directory.
>
> Ideas?
> Ondra
>
--
Marek Novotny
--
WFK and Seam Product Lead
Red Hat Czech s.r.o.
Purkynova 99
612 45 Brno
9 years, 5 months
Fwd: Re: Tattletale future?
by Marek Novotny
Hi windup devs,
I asked Jesper what is the plan with tattletale and below is his response.
I think we probably need replace the features by merging it into our
codebase or just fork tattletale as its future is stalled and that means
our fork won't need updates from upstream/original tattletale stream.
Are you fine with that? I am asking due this related issue
https://issues.jboss.org/browse/WINDUP-699, but there will be probably
more reasons, list your own if you have any ;)
-------- Forwarded Message --------
Subject: Re: Tattletale future?
Date: Wed, 22 Jul 2015 10:39:07 -0400
From: Jesper Pedersen <jpederse(a)redhat.com>
Organization: JBoss, by Red Hat
To: mnovotny(a)redhat.com
Hi,
On 07/22/2015 10:30 AM, Marek Novotny wrote:
> I was looking in Tattletale github repository and it seems it is a long
> time from the last commit
Project has been dead for 3 years now.
> and PRs are were waiting in a queue.
>
Likely because they are incomplete or wrong - but whatever works for
people. I'm not spending time on it though.
> What's the plan for Tattletale project?
>
There are no plans to activate the project again.
Best regards,
Jesper
9 years, 5 months
Windup Meeting Minutes - 2015-07-15
by Lincoln Baxter, III
Minutes:
http://transcripts.jboss.org/meeting/irc.freenode.org/windup/2015/windup....
Minutes (text):
http://transcripts.jboss.org/meeting/irc.freenode.org/windup/2015/windup....
Log:
http://transcripts.jboss.org/meeting/irc.freenode.org/windup/2015/windup....
Meeting summary
---------------
* Agenda (lincolnthree, 14:01:49)
* Status reports (lincolnthree, 14:04:20)
* LINK:
https://plus.google.com/hangouts/_/gzeiu5f7fuxrtkbtemqlzxmmfea?hl=en
? (ozizka, 14:04:56)
* I have worked on finishing up PRs for a couple of things (including
a report that is keyed on issues rather than filenames) (jsightler,
14:05:52)
* I have also worked on a mockup of the feedback form, though I think
that may change depending upon the outcome of the meeting about the
server component (authentication, etc) (jsightler, 14:06:34)
* I have also been starting some research into openshift (setting up a
local OSE instance for some testing being the main thing there) to
start learning what is new (err, everything :)) (jsightler,
14:07:07)
* Next, I will return to the feedback form stuff and hopefully start
working on some of the tasks for it (jsightler, 14:07:34)
* I sent https://github.com/windup/windup/pull/635 for small fix in
reports (marekn, 14:07:48)
* also was discovering more sources in windup core (marekn, 14:07:48)
* and collected existing feedback for windup reports from rob and marc
(marekn, 14:07:48)
* continue on Marc's responce to my questions wrt reports (marekn,
14:09:56)
* and probably will try to implement jsf/jsp report ? (marekn,
14:10:20)
* Last week/few days I've been working on the usual PR review.
(lincolnthree, 14:11:51)
* I' also did a bunch of research on agile tools and worked on getting
us set up in Taiga (lincolnthree, 14:12:10)
* I've been working on Story 3 - reviewing the remaining non-QAd
rules/rulesets (lincolnthree, 14:12:31)
* That's done now, waiting on the discussion about which rulesets to
review/include. (lincolnthree, 14:12:46)
* I also started to engage the legal team about how we do this
feedback form and stay within the law (lincolnthree, 14:13:09)
* I plan to continue working on those things today, more PR review,
management meeting, etc. (lincolnthree, 14:13:23)
* no blockers for me (lincolnthree, 14:13:30)
* Been working on WINDUP-679, currently making the test work. See PR
633. (ozizka, 14:13:43)
* Hit and reported FURNACE-59. Then I was having difficulties with a
classloading problem - JBL LogManagager was missing in random
addons. Added to root pom as test dependency. (ozizka, 14:13:43)
* Also implemented WINDUP-665 - PR after meeting (but there's no use
of it yet). (ozizka, 14:13:44)
* Next: Finishing WINDUP-679, then I'll take one of stories. (ozizka,
14:13:44)
* I was reviewing few PRs (mbriskar, 14:19:51)
* I added File condition for the cases where file content is not
important and we want to match on the file as a whole (mbriskar,
14:19:51)
* I made rexster more convenient to work with. Now it should start
everytime you run test with -Dmaven.surefire.debug ! :) (mbriskar,
14:20:21)
* LINK: https://github.com/tinkerpop/rexster/wiki/RexPro (mbriskar,
14:23:08)
* I finished working on exporting CSV file and I am about to start
working on the server service accepting requests (mbriskar,
14:24:26)
* Meeting format (lincolnthree, 14:27:26)
* rest server service (lincolnthree, 14:34:53)
* LINK:
https://docs.google.com/a/redhat.com/drawings/d/1bMRYzEPMvyPksOhm5FBJf7g3...
(ozizka, 14:57:22)
Meeting ended at 15:04:53 UTC.
--
Lincoln Baxter, III
http://ocpsoft.org
"Simpler is better."
9 years, 5 months
Freemarker usage
by Ondrej Zizka
Hi,
I have few questions regarding Freemarker usage.
There are many places where we duplicate the functionality already
available in Freemarker.
I wonder if it's necessary, as it makes the code less readable.
1)
For example, sometimes the iterable needs to become a collection. And
then needs to be sorted.
We have this:
<#list
sortFilesByPathAscending(projectModel.fileModelsNoDirectories) as fileModel>
<@fileModelRenderer fileModel/>
</#list>
sortFilesByPathAscending() is a FM function created just for this one
single case (or there's a second one).
Wouldn't it be better to just make projectModel.fileModelsNoDirectories
a collection using some toList(projectModel.fileModelsNoDirectories) and
sort it in FM?
Did something prevent this kind of usage?
<#list
toList(projectModel.fileModelsNoDirectories)?sort_by("filePath") as
fileModel>
Or, better, having it sorted right in the database, which is probably
better place to do so as it may have indexes available for it?
2)
iterableHasContent(...) - again, wouldn't it be better, instead of
<#if iterableHasContent(projectModel.fileModelsNoDirectories)>
...
<#list
sortFilesByPathAscending(projectModel.fileModelsNoDirectories) as fileModel>
to do
<#assign myList = toList(projectModel.fileModelsNoDirectories)
<#if myList?has_content>
<#list myList?sort_by("filePath") as fileModel>
The only addition of iterableHasContent() is that it measures the time
for the function call. Is that the reason for it?
Just asking.
Ondra
9 years, 5 months
