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 249917 - GCC 4.9.2 integration issues
Summary: GCC 4.9.2 integration issues
Status: RESOLVED WORKSFORME
Alias: None
Product: cnd
Classification: Unclassified
Component: -- Other -- (show other bugs)
Version: 8.1
Hardware: PC Windows 7
: P2 normal (vote)
Assignee: issues@cnd
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-01-20 21:32 UTC by aschwarz1309
Modified: 2015-01-23 15:29 UTC (History)
0 users

See Also:
Issue Type: DEFECT
Exception Reporter:


Attachments
IDE log (737.30 KB, text/plain)
2015-01-20 21:32 UTC, aschwarz1309
Details
SLIP Project source code (264.27 KB, application/octet-stream)
2015-01-22 15:41 UTC, aschwarz1309
Details

Note You need to log in before you can comment on or make changes to this bug.
Description aschwarz1309 2015-01-20 21:32:05 UTC
Product Version = NetBeans IDE 8.0.2 (Build 201411181905)
Operating System = Windows 7 version 6.1 running on amd64
Java; VM; Vendor = 1.7.0_40
Runtime = Java HotSpot(TM) 64-Bit Server VM 24.0-b56
cygwin
gcc 4.9.2

I just downloaded gcc 4.9.2 and I believe there seems to be some integration issues:
1: Bogus "Unable to resolve identifier" messages for scoped variables and C++ include files. This is an longstanding issue.
2: Include file gray's out. After initial editing, exit, and return the include file is grayed out - making it very difficult to read.
3: Build of one project but identification with another. Once, and only once, after I downloaded gcc 4.9.2 I started Netbeans and switched projects. When I started a build the previous project was identified as the build root, but the build was to the switched project. This has been non-repeatable and as such, I consider it very low priority and not a critical issue.

The bogus diagnostics with C++ include files leads to a cascade of errors - the same a for scoped variables but more extensive. Please tell me if this is of no concern and I will stop identifying it as an error. It seems to never be repaired (although sometimes it is better than others) and I don't really know if you feel that this is "operator error" and ignorable, hard to identify and fix, or just not a pressing concern. Please identify you treatment to me.

thanks
art
Comment 1 aschwarz1309 2015-01-20 21:32:12 UTC
Created attachment 151593 [details]
IDE log
Comment 2 aschwarz1309 2015-01-20 22:04:50 UTC
One additional comment:
4: Right clicking on a file in the Projects Window has the "Compile File F9" option grayed out (not operational). However the pulldown option (Run -> Compile File) works fine (on the same file). When the Project -> Compile File option does not work for the current file, it apparently doesn't work for any file. Build works fine.
Comment 3 Alexander Simon 2015-01-22 08:01:30 UTC
Hi art,

What are your projects?
Are its managed or unmanaged (created with existing sources or from binary)?
Could you provide test case to reproduce bugs?

Alexander
Comment 4 aschwarz1309 2015-01-22 15:41:14 UTC
Created attachment 151621 [details]
SLIP Project source code

No external libraries required. All code is C++. Install it into any directory (src/ slip/) as a C++ project and compile with 

Properties > Build > C++ Compiler > Additional Options > -std=gnu++11 -Wall -Wno-reorder -Wno-unused-value -Wno-address -Wno-sequence-point

I have had issues with filenames not having the correct case (Windows being case insensitive) and I have tried to correct that.

After parsing in Netbeans, look at any file.

Now, I just built my source and found no errors but, now, I can compile a file using the project window (right click on file name) but can not compile a file using Run > Compile File. Go figure.

So it looks like all of my errors have changed. Now there is one Netbeans anomaly and none of my files have errors. I don't know what to say and I don't know what to do. I suspect the situation will change by this afternoon.

Feel free to critique the code. 

art
Comment 5 aschwarz1309 2015-01-22 23:37:14 UTC
I've worked all day and haven't seen any errors. I don't think that there's been an update so I can only say, "Pfft, all gone".

This has happened before. There is a period of welcome relief and then bogus errors start appearing. But for now, the issue (of bogus errors) appears to be a non-issue. Mark it up to operator error.

Same result (almost) for compiles. This morning it appeared that I could compile using the Navigator but not Run > Compile File. Then "Pfft, all gone" and everything worked.

Thanks
art
Comment 6 Vladimir Voskresensky 2015-01-23 12:26:45 UTC
So, for now it is works-for-me, right?
Comment 7 aschwarz1309 2015-01-23 15:29:17 UTC
Yep.

Thanks for all.
art