This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.

Bug 25400 - Remount filesystem action doesn't close opened documents.
Summary: Remount filesystem action doesn't close opened documents.
Status: CLOSED FIXED
Alias: None
Product: platform
Classification: Unclassified
Component: Text (show other bugs)
Version: 3.x
Hardware: PC Linux
: P2 blocker (vote)
Assignee: Peter Zavadsky
URL:
Keywords: ARCH
Depends on: 25707
Blocks:
  Show dependency tree
 
Reported: 2002-07-08 14:58 UTC by Martin Schovanek
Modified: 2008-12-23 00:11 UTC (History)
5 users (show)

See Also:
Issue Type: DEFECT
Exception Reporter:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Martin Schovanek 2002-07-08 14:58:40 UTC 
[Nb Build 200207080100, jdk1.4.0_01]

Steps to reproduce:
===================
1) open .java in Source Editor and modify it
2) remount filesystem (change Root Directory in
Property Editor),
   the Source Editor remains opened
3) click the Source Editor,
   question dialog "File is modified. Save it?"
appears
4) click "Yes"
5) remount the filesystem back and open the .java

The changes wasn't seved (potential data lost).

In case of XML documents dialog "File was modified
externaly. Reload it?" appears. In this manner I
can e.g. open XML document twice.
Comment 1 Peter Zavadsky 2002-07-08 17:19:54 UTC 
Problem is when changing the root dir name, i.e. system name of the
filesystem, the FileObject are made invalid that way.

Problem is the support is not informed about the change neither via
FileObject nor via DataObject level. 

Radek suggested to fix it the way listen in support also on the
filesystem property change and handle it. I'll make it that probably,
but don't like the solution.
I think the support should be informed from data system about the
upcoming change (the FO's will be invalid, thus the DO too), to be
able take measures about that.
Comment 2 Jesse Glick 2002-07-08 17:32:17 UTC 
Agreed, shouldn't DataObject fire PROP_VALID from this? If so,
openide/datasystems. Also would P1 be more appropriate given that
there is potential user data loss?
Comment 3 Peter Zavadsky 2002-07-09 12:13:07 UTC 
Fixed in [trunk]

Fix:
openide/../loaders/OpenSupport.java [1.34]
          /text/EditorSupport.java [1.169]

Note:
It is the solution which Radek suggested.

Data system has an intention to fire the PROP_VALID change after the
root system name has chaned (it is not working due to some bug).

But in that case is late for support, beacuse when needed to save the
document, then the FO's are invalid already, thus no save is possible.
FileSystem fires also vetoable change event before going to change the
system name, and that would be probably nice if somehow propagated via
data system to support as vetoable PROP_VALID change. That's a
question. Anyway the support is needed to be informed at the time the
FO's are still valid, i.e. before the actual system name change, thus
allow the possible save.
Comment 4 Marian Mirilovic 2002-07-26 13:38:36 UTC 
verified in [nb_release34_rc1](200207232340)
Comment 5 Quality Engineering 2003-07-01 16:22:53 UTC 
Resolved for 3.4.x or earlier, no new info since then -> closing.