Created attachment 93488 [details]
Plase wait message
The message "Please Wait" remains forever in Navigator (with undocked window) in the follow cases:
1) Starting with navigator window docked
1.1) While editing a file, undock the Navigator window
1.2) Close Navigator
1.3) Edit/Focus the file agian
1.4) Press "ctrl+7" to reopen the navigator
(When you select another file it works again)
2) Starting with navigator window undocked
2.1) Select one file
2.2) Close the Navigator
2.3) Select another file
2.4) Press "ctrl+7" to reopen the navigator
it's not about docked navigator window only. It doesn't refresh if opened when other file is focused.
as the reporter wrote:
1. open file
2. check if the navigator shows something relevant and not please wait.
3. close navigator
4. open navigator while you have focus on editor.
you see please wait until you focus or close/open file.
if you open navigator using window -> navigating -> navigator it woks well again after you focus editor.
Product Version: NetBeans IDE Dev (Build null)
Java: 1.6.0_16; Java HotSpot(TM) 64-Bit Server VM 14.2-b01
reassigning to editor -> csl - does it belong to you guys?
Even after my last commit this seems still to cause problems. Trying to debug ...
After the last change in CSL this needs to be fixed in Parsing API. Passing the report to the parsing api category.
toRemove should better handle when someone is re-adding the same task right after its removal.
Steps to reproduce the problem: try to watch the navigator when switching between .html and .js files in the projects explorer.
Seems that task is not readded after fast sequence of add remove from multiple threads.
The reason is a dupliction of requests in toRemove queue. The problem is that the Hanz' schedulers have different semantic from the java schedulers. Hanz schedulers are run once and do not reschedule while the java schedulers do rescheduling. This causes that the request is neither among requests nor in finishedRequests. So removing it causes deferring the remove (the request is added into the toRemove queue). Fix requires non trivial change in TP :-(
fixed jet-main b5ba4cc6b65f
The committed fix introduced regression. For regression description, please see http://netbeans.org/bugzilla/show_bug.cgi?id=191289
The parsing api commit is guarded by unit tests. If the problem is in the csl part report on the proper component or add an unit test.
To pgebauer: Please do not blindly create (reopen) issues on parsing api, do it if either unit tests fail or you are sure it belongs to it. I've reverted the parsing api patch and the PHP navigator has the same problem as reported in issue #191289. Certainly nothing in common with parsing api.
I would never reopen this issue without a try with reverted fix. I have reverted the following changeset http://hg.netbeans.org/jet-main/rev/eb8ca1bb29af which has changed csl.api and it seems to be part of this issue. The issue BZ#191289 has disappeared after the reversion. Unfortunately this is not true with my todays build. It seems that the root of BZ#191289 is somewhere else.
In that case it's only problem of modules.
The issues on CSL do not belong to the parsing api they belong to CSL.
Feel free to reopen but assign to CSL.