Thread: sysSettings::Flush()
Hi,, New method sysSettings::Flush() With this patch every server setting goes immediately in persistent storage. At this moment it goes in persistent storage only when pgAdmin is normally finished. For other settings every third request will immediately put settings in persistent storage. Best regards Vladimir Kokovic Belgrade, Serbia, 6.October 2012
Attachment
Hi On Sat, Oct 6, 2012 at 6:57 AM, Vladimir Kokovic <vladimir.kokovic@gmail.com> wrote: > Hi,, > > New method sysSettings::Flush() > > With this patch every server setting goes immediately in persistent storage. > At this moment it goes in persistent storage only when pgAdmin is > normally finished. > > For other settings every third request will immediately put settings > in persistent storage. What's the rationale for that? It doesn't seem like it would be very expensive to write all settings immediately. -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise PostgreSQL Company
Hi, That is variant for somebody who don't want flush() on every setting. My preferred value is 0 (immediately). Best regards Vladimir Kokovic Belgrade, Serbia, 9.October 2012 On 10/9/12, Dave Page <dpage@pgadmin.org> wrote: > Hi > > On Sat, Oct 6, 2012 at 6:57 AM, Vladimir Kokovic > <vladimir.kokovic@gmail.com> wrote: >> Hi,, >> >> New method sysSettings::Flush() >> >> With this patch every server setting goes immediately in persistent >> storage. >> At this moment it goes in persistent storage only when pgAdmin is >> normally finished. >> >> For other settings every third request will immediately put settings >> in persistent storage. > > What's the rationale for that? It doesn't seem like it would be very > expensive to write all settings immediately. > > > -- > Dave Page > Blog: http://pgsnake.blogspot.com > Twitter: @pgsnake > > EnterpriseDB UK: http://www.enterprisedb.com > The Enterprise PostgreSQL Company >
On Tue, Oct 9, 2012 at 5:21 PM, Vladimir Kokovic <vladimir.kokovic@gmail.com> wrote: > Hi, > > That is variant for somebody who don't want flush() on every setting. > My preferred value is 0 (immediately). Doesn't seem particularly configurable though :-). I think flushing every time is fine - can you make that tweak and resend the patch please? On a related note; I'm still having trouble with your patches. This one gives the following when I try to apply it: raptor:pgAdmin3 dpage$ patch -p0 < ~/Downloads/pgadmin3-2.patch patch: **** unexpected `***' at line 12: *** a/pgadmin/include/utils/sysSettings.h How did you generate it? Was there any manual adjustment made that might have broken it? -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise PostgreSQL Company
Hi, Flushing every time version. Best regards Vladimir Kokovic Belgrade, Serbia, 10.October 2012
Attachment
On Wed, Oct 10, 2012 at 5:38 AM, Vladimir Kokovic <vladimir.kokovic@gmail.com> wrote: > Hi, > > Flushing every time version. Thanks, but it still won't apply. Using git apply (most definitely the correct option), I get: raptor:pgAdmin3 dpage$ git apply ~/Downloads/pgadmin3-2.patch fatal: unrecognized input Using patch: raptor:pgAdmin3 dpage$ patch -p1 < ~/Downloads/pgadmin3-flush.patch patching file pgadmin/include/utils/sysSettings.h patch unexpectedly ends in middle of line patching file pgadmin/utils/sysSettings.cpp patch unexpectedly ends in middle of line How are you generating it? A simple "git diff > patch.diff" should work just fine. Do you have any options in your git config file that might be causing the output to be mangled? Guillaume, can you apply the patch? -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise PostgreSQL Company
Hi, I don't know what is problem ! Maybe some blank line missing, Try to add empty line at the end of file. My git diff command: cd pgadmin3; git diff master vk-20120512-pgadmin3 | filterdiff --format=context > /tmp/pgadmin3.patch; cd ../ Best regards Vladimir Kokovic Belgrade, Serbia, 10.October 2012 On 10/10/12, Dave Page <dpage@pgadmin.org> wrote: > On Wed, Oct 10, 2012 at 5:38 AM, Vladimir Kokovic > <vladimir.kokovic@gmail.com> wrote: >> Hi, >> >> Flushing every time version. > > Thanks, but it still won't apply. Using git apply (most definitely the > correct option), I get: > > raptor:pgAdmin3 dpage$ git apply ~/Downloads/pgadmin3-2.patch > fatal: unrecognized input > > Using patch: > > raptor:pgAdmin3 dpage$ patch -p1 < ~/Downloads/pgadmin3-flush.patch > patching file pgadmin/include/utils/sysSettings.h > patch unexpectedly ends in middle of line > patching file pgadmin/utils/sysSettings.cpp > patch unexpectedly ends in middle of line > > How are you generating it? A simple "git diff > patch.diff" should > work just fine. Do you have any options in your git config file that > might be causing the output to be mangled? > > Guillaume, can you apply the patch? > > -- > Dave Page > Blog: http://pgsnake.blogspot.com > Twitter: @pgsnake > > EnterpriseDB UK: http://www.enterprisedb.com > The Enterprise PostgreSQL Company >
On Wed, Oct 10, 2012 at 4:06 PM, Vladimir Kokovic <vladimir.kokovic@gmail.com> wrote: > Hi, > > I don't know what is problem ! > Maybe some blank line missing, > Try to add empty line at the end of file. > > My git diff command: > > cd pgadmin3; git diff master vk-20120512-pgadmin3 | filterdiff > --format=context > /tmp/pgadmin3.patch; cd ../ Oh, OK - please try that again, but don't pipe it through filterdiff. I suspect that's what's causing the issue. e.g. cd pgadmin3; git diff master vk-20120512-pgadmin3 > /tmp/pgadmin3.patch; cd ../ -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise PostgreSQL Company
Hi, For me, that is correct patch, but should be applied with unix 'patch' program: cd pgadmin3; patch --context -p1 < /root/files/pgadmin/dave-good/pgadmin3.patch; cd ../ Best regards Vladimir Kokovic Belgrade, Serbia, 10.October 2012 On 10/10/12, Dave Page <dpage@pgadmin.org> wrote: > On Wed, Oct 10, 2012 at 4:06 PM, Vladimir Kokovic > <vladimir.kokovic@gmail.com> wrote: >> Hi, >> >> I don't know what is problem ! >> Maybe some blank line missing, >> Try to add empty line at the end of file. >> >> My git diff command: >> >> cd pgadmin3; git diff master vk-20120512-pgadmin3 | filterdiff >> --format=context > /tmp/pgadmin3.patch; cd ../ > > Oh, OK - please try that again, but don't pipe it through filterdiff. > I suspect that's what's causing the issue. > > e.g. > > cd pgadmin3; git diff master vk-20120512-pgadmin3 > /tmp/pgadmin3.patch; cd > ../ > > -- > Dave Page > Blog: http://pgsnake.blogspot.com > Twitter: @pgsnake > > EnterpriseDB UK: http://www.enterprisedb.com > The Enterprise PostgreSQL Company >
On Wed, Oct 10, 2012 at 4:58 PM, Vladimir Kokovic <vladimir.kokovic@gmail.com> wrote: > Hi, > For me, that is correct patch, but should be applied with unix 'patch' program: > > cd pgadmin3; patch --context -p1 < > /root/files/pgadmin/dave-good/pgadmin3.patch; cd ../ The project standard is a unified diff, the standard output from "git diff". Thanks. -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise PostgreSQL Company
On Wed, 2012-10-10 at 11:48 +0100, Dave Page wrote: > On Wed, Oct 10, 2012 at 5:38 AM, Vladimir Kokovic > <vladimir.kokovic@gmail.com> wrote: > > Hi, > > > > Flushing every time version. > > Thanks, but it still won't apply. Using git apply (most definitely the > correct option), I get: > > raptor:pgAdmin3 dpage$ git apply ~/Downloads/pgadmin3-2.patch > fatal: unrecognized input > > Using patch: > > raptor:pgAdmin3 dpage$ patch -p1 < ~/Downloads/pgadmin3-flush.patch > patching file pgadmin/include/utils/sysSettings.h > patch unexpectedly ends in middle of line > patching file pgadmin/utils/sysSettings.cpp > patch unexpectedly ends in middle of line > > How are you generating it? A simple "git diff > patch.diff" should > work just fine. Do you have any options in your git config file that > might be causing the output to be mangled? > > Guillaume, can you apply the patch? > Nope, I have the exact same messages. -- Guillaume http://blog.guillaume.lelarge.info http://www.dalibo.com