Bug#811846: #811846: status?
Hi Pino! On 09/25/2016 02:07 PM, Pino Toscano wrote: > Note that, if nothing happens within few days (say, friday 30th), then > I'll NMU/2 the patch I sent earlier; it has been already 3 months since > the switch to GCC 6, and this bug was reported months earlier. > xbase64 is not a leaf package, so leaving it unfixed means blocking > other packages (calligra, at least in my case). I uploaded xbase64 earlier today but I'm still waiting for the ACCEPT mail. If that doesn't happen within the next two hours, I will try to reupload the package. Please do therefore not NMU this package. Thanks, Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaub...@debian.org `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 signature.asc Description: OpenPGP digital signature
Bug#811846: #811846: status?
In data lunedì 19 settembre 2016 11:32:18 CEST, John Paul Adrian Glaubitz ha scritto: > On 09/19/2016 09:08 AM, Pino Toscano wrote: > > Looking further at this very issue (#811846), I don't think the > > solution implemented is actually the correct one: > > > > https://anonscm.debian.org/cgit/collab-maint/xbase64.git/commit/?id=39d6239543df5dff1707e14e129574ac393325af > > > > This just forces the C++ dialect to C++98, but that's only a workaround; > > looking at the build log in this bug (which can be produced also by > > rebuilding xbase64 in an up-to-date unstable/testing environment), > > there is a real bug in xbase64/xblock.cpp: > > > > std::cout << "xbLock constructor" << std::cout; > > > > std::cout is clearly wrong at the end, where std::endl is most probably > > what should be used; there are 5 occurrences of this in xbase64, all of > > them in xblock.cpp. > > Ugh, you're right. I somehow missed the debian/rules file completely, I > automatically interpreted the first changelog entry as "New upstream > release" and assumed those were upstream changes. > > Your fix is the correct one, of course. Interesting that the compiler > let's this slip with a warning instead of bailing out completely when > specifying -std=c++98. > > Anyway, I'll help Joerg incorporate the changes and the updated package > uploaded. Friendly ping. Note that, if nothing happens within few days (say, friday 30th), then I'll NMU/2 the patch I sent earlier; it has been already 3 months since the switch to GCC 6, and this bug was reported months earlier. xbase64 is not a leaf package, so leaving it unfixed means blocking other packages (calligra, at least in my case). Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#811846: #811846: status?
On 09/19/2016 09:08 AM, Pino Toscano wrote: > Looking further at this very issue (#811846), I don't think the > solution implemented is actually the correct one: > > https://anonscm.debian.org/cgit/collab-maint/xbase64.git/commit/?id=39d6239543df5dff1707e14e129574ac393325af > > This just forces the C++ dialect to C++98, but that's only a workaround; > looking at the build log in this bug (which can be produced also by > rebuilding xbase64 in an up-to-date unstable/testing environment), > there is a real bug in xbase64/xblock.cpp: > > std::cout << "xbLock constructor" << std::cout; > > std::cout is clearly wrong at the end, where std::endl is most probably > what should be used; there are 5 occurrences of this in xbase64, all of > them in xblock.cpp. Ugh, you're right. I somehow missed the debian/rules file completely, I automatically interpreted the first changelog entry as "New upstream release" and assumed those were upstream changes. Your fix is the correct one, of course. Interesting that the compiler let's this slip with a warning instead of bailing out completely when specifying -std=c++98. Anyway, I'll help Joerg incorporate the changes and the updated package uploaded. Thanks, Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaub...@debian.org `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 signature.asc Description: OpenPGP digital signature
Bug#811846: #811846: status?
tag 811846 + patch
thanks
In data domenica 18 settembre 2016 19:15:50 CEST, John Paul Adrian Glaubitz ha
scritto:
> On 09/18/2016 01:07 PM, John Paul Adrian Glaubitz wrote:
> > Sorry guys, was busy this weekend. Will look at it in a few hours :).
>
> There are two issues with the package on mentors that I see:
Looking further at this very issue (#811846), I don't think the
solution implemented is actually the correct one:
https://anonscm.debian.org/cgit/collab-maint/xbase64.git/commit/?id=39d6239543df5dff1707e14e129574ac393325af
This just forces the C++ dialect to C++98, but that's only a workaround;
looking at the build log in this bug (which can be produced also by
rebuilding xbase64 in an up-to-date unstable/testing environment),
there is a real bug in xbase64/xblock.cpp:
std::cout << "xbLock constructor" << std::cout;
std::cout is clearly wrong at the end, where std::endl is most probably
what should be used; there are 5 occurrences of this in xbase64, all of
them in xblock.cpp.
Attached there is an actual patch fixing the issue: please integrate it,
and remove the -std=c++98 workaround.
Thanks,
--
Pino Toscano--- a/xbase64/xblock.cpp
+++ b/xbase64/xblock.cpp
@@ -84,7 +84,7 @@ xbLock::xbLock(xbDbf * pdbf)
TableLockCnt = 0;
MemoLockCnt = 0;
IndexLockCnt = 0;
- std::cout << "xbLock constructor" << std::cout;
+ std::cout << "xbLock constructor" << std::endl;
}
/*/
xbLock::~xbLock()
@@ -169,7 +169,7 @@ else if( LockType == XB_LOCK || LockType
/*/
xbaseLock::xbaseLock( xbDbf * pdbf ) : xbLock( pdbf )
{
- std::cout << "xbaseLock constructor" << std::cout;
+ std::cout << "xbaseLock constructor" << std::endl;
}
/*/
xbShort xbaseLock::LockTableHeader( xbShort LockType )
@@ -336,7 +336,7 @@ xbShort xbaseLock::LockInit()
/*/
dbaseLock::dbaseLock( xbDbf * pdbf ) : xbLock( pdbf )
{
- std::cout << "dbaseLock constructor" << std::cout;
+ std::cout << "dbaseLock constructor" << std::endl;
}
/*/
xbShort dbaseLock::LockTableHeader( xbShort LockType )
@@ -460,7 +460,7 @@ xbShort dbaseLock::UnlockAll()
/*/
clipperLock::clipperLock( xbDbf * pdbf ) : xbLock( pdbf )
{
- std::cout << "clipperLock constructor" << std::cout;
+ std::cout << "clipperLock constructor" << std::endl;
}
/*/
xbShort clipperLock::LockTableHeader( xbShort LockType )
@@ -523,7 +523,7 @@ xbShort clipperLock::UnlockAll()
/*/
foxproLock::foxproLock( xbDbf * pdbf ) : xbLock( pdbf )
{
- std::cout << "foxproLock constructor" << std::cout;
+ std::cout << "foxproLock constructor" << std::endl;
}
/*/
xbShort foxproLock::LockTableHeader( xbShort LockType )
signature.asc
Description: This is a digitally signed message part.
Bug#811846: #811846: status?
On 09/18/2016 01:07 PM, John Paul Adrian Glaubitz wrote: > Sorry guys, was busy this weekend. Will look at it in a few hours :). There are two issues with the package on mentors that I see: * It has symbols files for some architectures but not all architectures despite the debian/changelog file claiming otherwise. * The postinst scripts for the -dev and -bin packages call dpkg-maintscript-helper to install symlinks to the documentation folder of the -doc package. However, neither the -bin nor the -dev package depend on any way on the -doc package. So, in the worst case, the postinst script will fail (I didn't test how dpkg-mainscript-helper behaves), the best cases, it creates dangling symlinks. Neither is an optimal situation. The rest of the changes look fine. Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaub...@debian.org `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 signature.asc Description: OpenPGP digital signature
Bug#811846: #811846: status?
On 09/18/2016 12:41 PM, Jörg Frings-Fürst wrote: >> what's the status of this bug (xbase64 that FTBFS with GCC 6)? >> Because of it, xbase64 was removed from testing a couple of months >> ago, > > The package is uploaded to mentors and Adrian is looking for it. > >> and now it will prevent calligra (which ships a sybase access driver >> for kexi) to migrate to testing. >> > > Ok. I have Adrian add as CC. Sorry guys, was busy this weekend. Will look at it in a few hours :). Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaub...@debian.org `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 signature.asc Description: OpenPGP digital signature
Bug#811846: #811846: status?
Hello Pino, Am Sonntag, den 18.09.2016, 07:06 +0200 schrieb Pino Toscano: > Hi, > > what's the status of this bug (xbase64 that FTBFS with GCC 6)? > Because of it, xbase64 was removed from testing a couple of months > ago, The package is uploaded to mentors and Adrian is looking for it. > and now it will prevent calligra (which ships a sybase access driver > for kexi) to migrate to testing. > Ok. I have Adrian add as CC. > Any way I can help with this? > Many thanks. Please ask Adrian for it. > Thanks, Cu Jörg -- New: GPG Fingerprint: 63E0 075F C8D4 3ABB 35AB 30EE 09F8 9F3C 8CA1 D25D GPG key (long) : 09F89F3C8CA1D25D GPG Key: 8CA1D25D CAcert Key S/N : 0E:D4:56 Old pgp Key: BE581B6E (revoked since 2014-12-31). Jörg Frings-Fürst D-54470 Lieser Threema: SYR8SJXB IRC: j_...@freenode.net j_...@oftc.net My wish list: - Please send me a picture from the nature at your home. signature.asc Description: This is a digitally signed message part
Bug#811846: #811846: status?
Hi, what's the status of this bug (xbase64 that FTBFS with GCC 6)? Because of it, xbase64 was removed from testing a couple of months ago, and now it will prevent calligra (which ships a sybase access driver for kexi) to migrate to testing. Any way I can help with this? Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
