[jira] Commented: (IVY-879) Using ivy in offline mode

2009-01-24 Thread Xavier Hanin (JIRA)

[ 
https://issues.apache.org/jira/browse/IVY-879?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12666896#action_12666896
 ] 

Xavier Hanin commented on IVY-879:
--

I've fixed the problem with TTL in IVY-1012, and also changed some tiny things 
to make it more usable for offline mode: I've added the option to use 'eternal' 
as ttl duration, and also changed when ttl is checked. So now if you don't have 
any specific ttl in your Ivy settings, all you have to do when offline is to 
set the property ivy.cache.ttl.default to eternal (with 
-Divy.cache.ttl.default=eternal when calling ant for instance). If you set ttls 
yourself, it's a bit more work to override the values in offline mode, but it's 
still possible.

So IMO the current trunk version is enough for offline support. WDYT?

 Using ivy in offline mode
 -

 Key: IVY-879
 URL: https://issues.apache.org/jira/browse/IVY-879
 Project: Ivy
  Issue Type: Improvement
  Components: Core
Reporter: Jean-Louis Boudart
Assignee: Xavier Hanin

 I recently tried to use ivy in offline, everything is in ivy-cache, but ivy 
 seems to check dynamique revision all the time.
 Exemple:  Suppose you have this dependency on your project
 dependency org=jasperreports name=jasperreports rev=2.0.5 
 conf=runtime-runtime/
 Looking in the POM file 
 (http://repo1.maven.org/maven2/jasperreports/jasperreports/2.0.5/jasperreports-2.0.5.pom)
  we can see that jasperreports depends on commons-logging#[1.0,)
 This revision is evaluated on every resolve/retrieve. 
 I've tried to tune my cache with a large TTL, but the problem is still here.
 My cache configuration : 
   caches default=mycache checkUpToDate=false
   cache name=mycache basedir=${ivy.home.dir}/ivy-cache 
 defaultTTL=1d
   ttl revision=latest.integration duration=1d /
   /cache
   /caches
 Maybe there is a regression on my version?
 Even if the TTL options seems to be interessant, i'm not so sure that this is 
 the best solution for working in offline mode.
 It could be a nice improvement to reuse the cache even if TTL is expired when 
 repo are not reachable.
 What do you think about it?

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Commented: (IVY-879) Using ivy in offline mode

2009-01-24 Thread Jean-Louis Boudart (JIRA)

[ 
https://issues.apache.org/jira/browse/IVY-879?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12666993#action_12666993
 ] 

Jean-Louis Boudart commented on IVY-879:


Tested on trunk Head Revision.

Works good for me i'll close the issue

 Using ivy in offline mode
 -

 Key: IVY-879
 URL: https://issues.apache.org/jira/browse/IVY-879
 Project: Ivy
  Issue Type: Improvement
  Components: Core
Reporter: Jean-Louis Boudart
Assignee: Xavier Hanin
 Fix For: trunk


 I recently tried to use ivy in offline, everything is in ivy-cache, but ivy 
 seems to check dynamique revision all the time.
 Exemple:  Suppose you have this dependency on your project
 dependency org=jasperreports name=jasperreports rev=2.0.5 
 conf=runtime-runtime/
 Looking in the POM file 
 (http://repo1.maven.org/maven2/jasperreports/jasperreports/2.0.5/jasperreports-2.0.5.pom)
  we can see that jasperreports depends on commons-logging#[1.0,)
 This revision is evaluated on every resolve/retrieve. 
 I've tried to tune my cache with a large TTL, but the problem is still here.
 My cache configuration : 
   caches default=mycache checkUpToDate=false
   cache name=mycache basedir=${ivy.home.dir}/ivy-cache 
 defaultTTL=1d
   ttl revision=latest.integration duration=1d /
   /cache
   /caches
 Maybe there is a regression on my version?
 Even if the TTL options seems to be interessant, i'm not so sure that this is 
 the best solution for working in offline mode.
 It could be a nice improvement to reuse the cache even if TTL is expired when 
 repo are not reachable.
 What do you think about it?

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.