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.
Created attachment 91393 [details] The NPS-File while Scaning-Projects Today I tried to format some Code (about 10 lines) while Scanning projects was running. I had to wait more than 50 Minutes, then the Exception from bug 176666 occured and the code wasn´t formated. Why Code-Formating is blocked by Scanning-Projects? Editor-Actions shouldn´t be blocked by Scanning-Projects.
Created attachment 91394 [details] The Logfile.
Yes, this is a known problem, but currently not resolvable under current infrastructure. However, there should be a dialog displayed, allowing to cancel the action, has it been displayed? Thanks.
The dialog was displayed.
Unfortunately, the situation is a bit more difficult. The formatting does not wait for the scan to finish (pure Java formatting can run concurrently with scan - formatting of Java scriptlets in JSP cannot), but is waiting in a queue for a canceled Go to Declaration to finish, which is waiting for the scan to finish. When the scan finishes, GtD is woken up and cancels itself, but that is useless in this case. As a separate concern, 50 minutes for scan seem quite a lot to me - is there anything that you could tell us about your project and setting? E.g. how many files it contains, what kinds of files, how many megabytes of sources and jars it contains, if the project/homedir/${userdir} are on slow media (e.g. network disk), how much physical memory do you have, custom memory settings for the IDE, if any, etc.? Thanks.
I think I already interrogated Stefan about his setup in issue #168328. One thing that you should keep in mind is that he's got a significant portion of scanned files on a network drive. But even with that 50 minutes is quite a long time. I'll have a look in the log file, if I can see some obvious problem why scanning took so long. Also, Stefan, I can see from the log file that you are you using that Scan-On-Demand module. Please note that this is not an officially supported plugin and it may have (and actually it most likely has) adverse/unknown effect on scanning and the editor/java features. That said, we are very very reluctant to even analyze defects with this plug installed and turned on. Sorry, but with this plugin anything is possible.
-) E:/Entwicklung65: Standard-Java-Projekt *) src: 3500 Java-Files *) build/classes: 5500 Classes -) Library *) H:/src: 25.000 Java-Files *) H:/build/classes: 45.000 Class-Files *) Additional Jar-File: 3500 Classes No Network-Drive! I think the scanning was until one week ago faster than now. > but is waiting in a queue for a canceled Go to Declaration to finish, Very strange constellation.
(In reply to comment #5) > I think I already interrogated Stefan about his setup in issue #168328. One > thing that you should keep in mind is that he's got a significant portion of > scanned files on a network drive. But even with that 50 minutes is quite a long > time. I'll have a look in the log file, if I can see some obvious problem why > scanning took so long. I changed my configuration and make every night a sync to my local-drive. > Also, Stefan, I can see from the log file that you are you using that > Scan-On-Demand module. Please note that this is not an officially supported > plugin and it may have (and actually it most likely has) adverse/unknown effect > on scanning and the editor/java features. That said, we are very very reluctant > to even analyze defects with this plug installed and turned on. Sorry, but with > this plugin anything is possible. I hopped this module can removes the scanning-bugs, but I think I wait, until this works better.
Integrated into 'main-golden', will be available in build *200912010200* on http://bits.netbeans.org/dev/nightly/ (upload may still be in progress) Changeset: http://hg.netbeans.org/main/rev/dea778a478f0 User: Jan Lahoda <jlahoda@netbeans.org> Log: #177261: no need to compute classes to recompile for binary roots.
Similar bug 166668 "First process collapsing code folds" should be valued as DEFECT too.
Fixed jet-main 41985eb45913