Opalvoip
Author: s | 2025-04-24
OpalVOIP การดาวน์โหลดฟรีและปลอดภัย OpalVOIP เวอร์ชันล่าสุด OpalVOIP เป็นการสื่อสารฟรี a ซอฟต์แวร์ที่ช่วยให้คุณเพลิดเพลินกับ Voice over Internet Prot
Thread: [Opalvoip-devel] opalvoip vs h323plus
In the mean time I'll>> just>>>> let it simmer and see if anyone else has the issue. It may be some>>>> compiler flag or setting.>>> And of course, it might mysteriously disappear with a random compiler>>> upgrade ...>> It disappeared by removing enable-pic from the compiler flags... just in>> case anyone runs across this in the future.>>>> -Dennis>>>>>> ------------------------------------------------------------------------->> ----->> All of the data generated in your IT infrastructure is seriously>> valuable.>> Why? It contains a definitive record of application performance, security>> threats, fraudulent activity, and more. Splunk takes this data and makes>> sense of it. IT sense. And common sense.>> _______________________________________________>> Opalvoip-devel mailing list>> Opa...@li...>> ------------------------------------------------------------------------------> All the data continuously generated in your IT infrastructure contains a> definitive record of customers, application performance, security> threats, fraudulent activity and more. Splunk takes this data and makes> sense of it. Business sense. IT sense. Common sense.> _______________________________________________> Opalvoip-devel mailing list> Opa...@li...> From: Robert J. - 2011-10-11 22:10:58 Weird.We have had a very stable system for a number of years. Then, recent GNUcompiler versions (last 6 months or so) have broken all sorts of stuff.Mostly in that "static object" constructor/destructor area.It seems the llvm/clang has more changes from the previous stable compilerthan most. And more problems.Very frustrating, but that's Open Source for you!Robert JongbloedOPAL/OpenH323/PTLib Architect and Co-founder.> -----Original Message-----> From: Dennis Muhlestein [mailto:djm...@gm...]> Sent: Wednesday, 12 October 2011 2:30 AM> To: opa...@li...> Subject: Re: [Opalvoip-devel] iOS issues> > Yeah no kidding. Here is another one for you:> > I have to pass
opalvoip-devel Mailing List for OpalVOIP - SourceForge
The> behavior of the ios_base io manipulator classes. I get an> EXC_BAD_INSTRUCTION after a little bit when the anything like this is> done:> > PTRACE(n, "xxx" or cout > Any of those ios_base manipulators cause the program to crash.> > I've searched through differences in ptlib's object.h and through the> code for other issues that might cause this but nothing is turning up.> > I got around it temporarily by just lowering the ptrace level and> commenting out a couple spots that used those functions. I can call the> underlying functions that those classes call on the streams without a> crash. Something is causing it to not work though. Haven't had much> luck today in finding the cause.> > Anyone else using iOS tried the trunk code? The ptrace stuff seems to> work fine on the 3.8 branch so I don't think it has to do with using the> 5.0 sdk instead of the 4.x sdk to compile/run. It does seem to be> related to something with the compiler though.> > > > -------------------------------------------------------------------------> -----> All the data continuously generated in your IT infrastructure contains a> definitive record of customers, application performance, security> threats, fraudulent activity and more. Splunk takes this data and makes> sense of it. Business sense. IT sense. Common sense.> _______________________________________________> Opalvoip-devel mailing list> Opa...@li...> From: Robert J. - 2011-09-20 00:19:24 Integrators? I meant iterators!Robert JongbloedOPAL/OpenH323/PTLib Architect and Co-founder.> -----Original Message-----> From: Robert Jongbloed [mailto:ro...@vo...]> Sent: Tuesday, 20 September 2011 9:59 AM> To: Opa...@li...> Subject: Re: [Opalvoip-devel]opalvoip-user Mailing List for OpalVOIP - SourceForge
> > On Sep 21, 2011, at 5:38 PM, Robert Jongbloed wrote:> > >> -----Original Message-----> >> From: Dennis Muhlestein [mailto:djm...@gm...]> >> Sent: Thursday, 22 September 2011 4:07 AM> > ...> >> On 9/19/11 8:15 PM, Robert Jongbloed wrote:> >>> Random thought, can you try making it explicitly std::hex, rather> than> >> just> >>> hex ? I remember having some bizarre behaviour once with that, though> >> not a> >>> crash.> >>>> >> I actually tried that before. It's really weird. To get around it I> >> made my own hex/right/left etc classes and my own operators. They do> >> the same thing the ios_base.h defined ones do. It works fine. I'll> >> have to dig into this more when I have time. In the mean time I'll> just> >> let it simmer and see if anyone else has the issue. It may be some> >> compiler flag or setting.> >> > And of course, it might mysteriously disappear with a random compiler> > upgrade ...> > It disappeared by removing enable-pic from the compiler flags... just in> case anyone runs across this in the future.> > -Dennis> > > -------------------------------------------------------------------------> -----> All of the data generated in your IT infrastructure is seriously> valuable.> Why? It contains a definitive record of application performance, security> threats, fraudulent activity, and more. Splunk takes this data and makes> sense of it. IT sense. And common sense.> _______________________________________________> Opalvoip-devel mailing list> Opa...@li...> From: Dennis M. - 2011-10-11 15:30:18 Yeah no kidding. Here is another. OpalVOIP การดาวน์โหลดฟรีและปลอดภัย OpalVOIP เวอร์ชันล่าสุด OpalVOIP เป็นการสื่อสารฟรี a ซอฟต์แวร์ที่ช่วยให้คุณเพลิดเพลินกับ Voice over Internet Prot OpalVOIP การดาวน์โหลดฟรีและปลอดภัย OpalVOIP เวอร์ชันล่าสุด OpalVOIP เป็นการสื่อสารฟรี a ซอฟต์แวร์ที่ช่วยให้คุณเพลิดเพลินกับ Voice over Internet ProtThread: [Opalvoip-devel] Opalvoip-opal 3.18.8 release
Temporarily by just lowering the ptrace level and> > commenting out a couple spots that used those functions. I can call> the> > underlying functions that those classes call on the streams without a> > crash. Something is causing it to not work though. Haven't had much> > luck today in finding the cause.> >> > Anyone else using iOS tried the trunk code? The ptrace stuff seems to> > work fine on the 3.8 branch so I don't think it has to do with using> the> > 5.0 sdk instead of the 4.x sdk to compile/run. It does seem to be> > related to something with the compiler though.> >> >> >> > -----------------------------------------------------------------------> --> > -----> > All the data continuously generated in your IT infrastructure contains> a> > definitive record of customers, application performance, security> > threats, fraudulent activity and more. Splunk takes this data and makes> > sense of it. Business sense. IT sense. Common sense.> > > _______________________________________________> > Opalvoip-devel mailing list> > Opa...@li...> > > > -------------------------------------------------------------------------> -----> All the data continuously generated in your IT infrastructure contains a> definitive record of customers, application performance, security> threats, fraudulent activity and more. Splunk takes this data and makes> sense of it. Business sense. IT sense. Common sense.> _______________________________________________> Opalvoip-devel mailing list> Opa...@li...> From: Dennis M. - 2011-09-20 00:48:35 On Sep 19, 2011, at 5:58 PM, Robert Jongbloed wrote:> Wow! I cannot imagine what could have changed between 3.8 and 3.11 to cause> that! There wereThread: [Opalvoip-svn] SF.net SVN: opalvoip:[ ] opal
My own operators. They> do> >>>> the same thing the ios_base.h defined ones do. It works fine. I'll> >>>> have to dig into this more when I have time. In the mean time I'll> >> just> >>>> let it simmer and see if anyone else has the issue. It may be some> >>>> compiler flag or setting.> >>> And of course, it might mysteriously disappear with a random compiler> >>> upgrade ...> >> It disappeared by removing enable-pic from the compiler flags... just> in> >> case anyone runs across this in the future.> >>> >> -Dennis> >>> >>> >> ----------------------------------------------------------------------> ---> >> -----> >> All of the data generated in your IT infrastructure is seriously> >> valuable.> >> Why? It contains a definitive record of application performance,> security> >> threats, fraudulent activity, and more. Splunk takes this data and> makes> >> sense of it. IT sense. And common sense.> >> >> _______________________________________________> >> Opalvoip-devel mailing list> >> Opa...@li...> >> >> > -----------------------------------------------------------------------> -------> > All the data continuously generated in your IT infrastructure contains> a> > definitive record of customers, application performance, security> > threats, fraudulent activity and more. Splunk takes this data and makes> > sense of it. Business sense. IT sense. Common sense.> > > _______________________________________________> > Opalvoip-devel mailing list> > Opa...@li...> > > > -------------------------------------------------------------------------> -----> All the data continuously generated in your IT infrastructure contains a> definitive record of customers, application performance, security> threats, fraudulent activity and more. Splunk takes this data and makes> sense ofThread: [Opalvoip-announce] Plan to move OpalVoip source
Into this more when I have time. In the mean time I'll just> let it simmer and see if anyone else has the issue. It may be some> compiler flag or setting.And of course, it might mysteriously disappear with a random compilerupgrade ...Robert JongbloedOPAL/OpenH323/PTLib Architect and Co-founder. From: Dennis M. - 2011-10-07 15:08:18 On Sep 21, 2011, at 5:38 PM, Robert Jongbloed wrote:>> -----Original Message----->> From: Dennis Muhlestein [mailto:djm...@gm...]>> Sent: Thursday, 22 September 2011 4:07 AM> ...>> On 9/19/11 8:15 PM, Robert Jongbloed wrote:>>> Random thought, can you try making it explicitly std::hex, rather than>> just>>> hex ? I remember having some bizarre behaviour once with that, though>> not a>>> crash.>>> >> I actually tried that before. It's really weird. To get around it I>> made my own hex/right/left etc classes and my own operators. They do>> the same thing the ios_base.h defined ones do. It works fine. I'll>> have to dig into this more when I have time. In the mean time I'll just>> let it simmer and see if anyone else has the issue. It may be some>> compiler flag or setting.> > And of course, it might mysteriously disappear with a random compiler> upgrade ...It disappeared by removing enable-pic from the compiler flags... just in case anyone runs across this in the future.-Dennis From: Robert J. - 2011-10-10 00:17:33 Mystery upon mystery ...Robert JongbloedOPAL/OpenH323/PTLib Architect and Co-founder.> -----Original Message-----> From: Dennis Muhlestein [mailto:djm...@gm...]> Sent: Saturday, 8 October 2011 2:08 AM> To: opa...@li...> Subject: Re: [Opalvoip-devel] iOS issues>Thread: [Opalvoip-devel] opalvoip vs h323plus - SourceForge
Summary Files Reviews Support Wiki Mailing Lists Tickets ▾ Bugs Support Requests Patches Feature Requests Git ▾ ptlib opal Menu ▾ ▴ From: Dennis M. - 2011-09-19 23:36:12 I've actually got my H.460 patch close to ready to submit. It seems to be working properly on the latest trunk code. I'm having a difficult time testing it though because I've run into another issue.Something between the 3.8 branch and current trunk has affected the behavior of the ios_base io manipulator classes. I get an EXC_BAD_INSTRUCTION after a little bit when the anything like this is done:PTRACE(n, "xxx" From: Robert J. - 2011-09-19 23:58:52 Wow! I cannot imagine what could have changed between 3.8 and 3.11 to causethat! There were no fundamental changes there. There hasn't been in years!The largest change is the addition of integrators to PDictionary, but howcould that affect io manipulators? Bizarre.You say "after a little bit", does that imply that sometimes std::hex worksand sometimes it doesn't?Are your builds for 3.8 and 3.11 use identical compiler and environment,correct? What you say below implies it might not be.Robert JongbloedOPAL/OpenH323/PTLib Architect and Co-founder.> -----Original Message-----> From: Dennis Muhlestein [mailto:djm...@gm...]> Sent: Tuesday, 20 September 2011 9:36 AM> To: Opa...@li...> Subject: [Opalvoip-devel] iOS issues> > I've actually got my H.460 patch close to ready to submit. It seems to> be working properly on the latest trunk code. I'm having a difficult> time testing it though because I've run into another issue.> > Something between the 3.8 branch and current trunk has affected. OpalVOIP การดาวน์โหลดฟรีและปลอดภัย OpalVOIP เวอร์ชันล่าสุด OpalVOIP เป็นการสื่อสารฟรี a ซอฟต์แวร์ที่ช่วยให้คุณเพลิดเพลินกับ Voice over Internet Prot OpalVOIP การดาวน์โหลดฟรีและปลอดภัย OpalVOIP เวอร์ชันล่าสุด OpalVOIP เป็นการสื่อสารฟรี a ซอฟต์แวร์ที่ช่วยให้คุณเพลิดเพลินกับ Voice over Internet Prot
[Opalvoip-devel] Opalvoip-opal 3.18.8 release compilation error
IOS issues> > Wow! I cannot imagine what could have changed between 3.8 and 3.11 to> cause> that! There were no fundamental changes there. There hasn't been in> years!> The largest change is the addition of integrators to PDictionary, but how> could that affect io manipulators? Bizarre.> > You say "after a little bit", does that imply that sometimes std::hex> works> and sometimes it doesn't?> > Are your builds for 3.8 and 3.11 use identical compiler and environment,> correct? What you say below implies it might not be.> > Robert Jongbloed> OPAL/OpenH323/PTLib Architect and Co-founder.> > > > -----Original Message-----> > From: Dennis Muhlestein [mailto:djm...@gm...]> > Sent: Tuesday, 20 September 2011 9:36 AM> > To: Opa...@li...> > Subject: [Opalvoip-devel] iOS issues> >> > I've actually got my H.460 patch close to ready to submit. It seems to> > be working properly on the latest trunk code. I'm having a difficult> > time testing it though because I've run into another issue.> >> > Something between the 3.8 branch and current trunk has affected the> > behavior of the ios_base io manipulator classes. I get an> > EXC_BAD_INSTRUCTION after a little bit when the anything like this is> > done:> >> > PTRACE(n, "xxx" > or cout >> > Any of those ios_base manipulators cause the program to crash.> >> > I've searched through differences in ptlib's object.h and through the> > code for other issues that might cause this but nothing is turning up.> >> > I got around itOpalVOIP for Windows - download.seask.net
One for you:I have to pass special compiler flags for ptlib/src/ptlib/common/contain.cxx for the release build.-O0 instead of -O3. with -O3, the regular expression code doesn't match when it is supposed to match. It ends up causing calls not to be routed. It also causes some calls with getting the endpoint addresses to not return all the addresses. I've tried internal regex build as well as the OS regular expressions. I haven't tracked down why the optimization breaks things but for now I just don't optimize that file in the release build.This is with the llvm/clang compiler still of course. Not sure if gcc behaves the same way or not.On 10/9/11 6:17 PM, Robert Jongbloed wrote:> Mystery upon mystery ...>> Robert Jongbloed> OPAL/OpenH323/PTLib Architect and Co-founder.>>> -----Original Message----->> From: Dennis Muhlestein [mailto:djm...@gm...]>> Sent: Saturday, 8 October 2011 2:08 AM>> To: opa...@li...>> Subject: Re: [Opalvoip-devel] iOS issues>>>>>> On Sep 21, 2011, at 5:38 PM, Robert Jongbloed wrote:>>>>>> -----Original Message----->>>> From: Dennis Muhlestein [mailto:djm...@gm...]>>>> Sent: Thursday, 22 September 2011 4:07 AM>>> ...>>>> On 9/19/11 8:15 PM, Robert Jongbloed wrote:>>>>> Random thought, can you try making it explicitly std::hex, rather>> than>>>> just>>>>> hex ? I remember having some bizarre behaviour once with that, though>>>> not a>>>>> crash.>>>>>>>>> I actually tried that before. It's really weird. To get around it I>>>> made my own hex/right/left etc classes and my own operators. They do>>>> the same thing the ios_base.h defined ones do. It works fine. I'll>>>> have to dig into this more when I have time.. OpalVOIP การดาวน์โหลดฟรีและปลอดภัย OpalVOIP เวอร์ชันล่าสุด OpalVOIP เป็นการสื่อสารฟรี a ซอฟต์แวร์ที่ช่วยให้คุณเพลิดเพลินกับ Voice over Internet ProtOpalVOIP - Browse Files at SourceForge.net
Special compiler flags for> ptlib/src/ptlib/common/contain.cxx for the release build.> -O0 instead of -O3. with -O3, the regular expression code doesn't match> when it is supposed to match. It ends up causing calls not to be> routed. It also causes some calls with getting the endpoint addresses> to not return all the addresses. I've tried internal regex build as> well as the OS regular expressions. I haven't tracked down why the> optimization breaks things but for now I just don't optimize that file> in the release build.> > This is with the llvm/clang compiler still of course. Not sure if gcc> behaves the same way or not.> > On 10/9/11 6:17 PM, Robert Jongbloed wrote:> > Mystery upon mystery ...> >> > Robert Jongbloed> > OPAL/OpenH323/PTLib Architect and Co-founder.> >> >> -----Original Message-----> >> From: Dennis Muhlestein [mailto:djm...@gm...]> >> Sent: Saturday, 8 October 2011 2:08 AM> >> To: opa...@li...> >> Subject: Re: [Opalvoip-devel] iOS issues> >>> >>> >> On Sep 21, 2011, at 5:38 PM, Robert Jongbloed wrote:> >>> >>>> -----Original Message-----> >>>> From: Dennis Muhlestein [mailto:djm...@gm...]> >>>> Sent: Thursday, 22 September 2011 4:07 AM> >>> ...> >>>> On 9/19/11 8:15 PM, Robert Jongbloed wrote:> >>>>> Random thought, can you try making it explicitly std::hex, rather> >> than> >>>> just> >>>>> hex ? I remember having some bizarre behaviour once with that,> though> >>>> not a> >>>>> crash.> >>>>>> >>>> I actually tried that before. It's really weird. To get around it> I> >>>> made my own hex/right/left etc classes andComments
In the mean time I'll>> just>>>> let it simmer and see if anyone else has the issue. It may be some>>>> compiler flag or setting.>>> And of course, it might mysteriously disappear with a random compiler>>> upgrade ...>> It disappeared by removing enable-pic from the compiler flags... just in>> case anyone runs across this in the future.>>>> -Dennis>>>>>> ------------------------------------------------------------------------->> ----->> All of the data generated in your IT infrastructure is seriously>> valuable.>> Why? It contains a definitive record of application performance, security>> threats, fraudulent activity, and more. Splunk takes this data and makes>> sense of it. IT sense. And common sense.>> _______________________________________________>> Opalvoip-devel mailing list>> Opa...@li...>> ------------------------------------------------------------------------------> All the data continuously generated in your IT infrastructure contains a> definitive record of customers, application performance, security> threats, fraudulent activity and more. Splunk takes this data and makes> sense of it. Business sense. IT sense. Common sense.> _______________________________________________> Opalvoip-devel mailing list> Opa...@li...> From: Robert J. - 2011-10-11 22:10:58 Weird.We have had a very stable system for a number of years. Then, recent GNUcompiler versions (last 6 months or so) have broken all sorts of stuff.Mostly in that "static object" constructor/destructor area.It seems the llvm/clang has more changes from the previous stable compilerthan most. And more problems.Very frustrating, but that's Open Source for you!Robert JongbloedOPAL/OpenH323/PTLib Architect and Co-founder.> -----Original Message-----> From: Dennis Muhlestein [mailto:djm...@gm...]> Sent: Wednesday, 12 October 2011 2:30 AM> To: opa...@li...> Subject: Re: [Opalvoip-devel] iOS issues> > Yeah no kidding. Here is another one for you:> > I have to pass
2025-04-14The> behavior of the ios_base io manipulator classes. I get an> EXC_BAD_INSTRUCTION after a little bit when the anything like this is> done:> > PTRACE(n, "xxx" or cout > Any of those ios_base manipulators cause the program to crash.> > I've searched through differences in ptlib's object.h and through the> code for other issues that might cause this but nothing is turning up.> > I got around it temporarily by just lowering the ptrace level and> commenting out a couple spots that used those functions. I can call the> underlying functions that those classes call on the streams without a> crash. Something is causing it to not work though. Haven't had much> luck today in finding the cause.> > Anyone else using iOS tried the trunk code? The ptrace stuff seems to> work fine on the 3.8 branch so I don't think it has to do with using the> 5.0 sdk instead of the 4.x sdk to compile/run. It does seem to be> related to something with the compiler though.> > > > -------------------------------------------------------------------------> -----> All the data continuously generated in your IT infrastructure contains a> definitive record of customers, application performance, security> threats, fraudulent activity and more. Splunk takes this data and makes> sense of it. Business sense. IT sense. Common sense.> _______________________________________________> Opalvoip-devel mailing list> Opa...@li...> From: Robert J. - 2011-09-20 00:19:24 Integrators? I meant iterators!Robert JongbloedOPAL/OpenH323/PTLib Architect and Co-founder.> -----Original Message-----> From: Robert Jongbloed [mailto:ro...@vo...]> Sent: Tuesday, 20 September 2011 9:59 AM> To: Opa...@li...> Subject: Re: [Opalvoip-devel]
2025-04-13Temporarily by just lowering the ptrace level and> > commenting out a couple spots that used those functions. I can call> the> > underlying functions that those classes call on the streams without a> > crash. Something is causing it to not work though. Haven't had much> > luck today in finding the cause.> >> > Anyone else using iOS tried the trunk code? The ptrace stuff seems to> > work fine on the 3.8 branch so I don't think it has to do with using> the> > 5.0 sdk instead of the 4.x sdk to compile/run. It does seem to be> > related to something with the compiler though.> >> >> >> > -----------------------------------------------------------------------> --> > -----> > All the data continuously generated in your IT infrastructure contains> a> > definitive record of customers, application performance, security> > threats, fraudulent activity and more. Splunk takes this data and makes> > sense of it. Business sense. IT sense. Common sense.> > > _______________________________________________> > Opalvoip-devel mailing list> > Opa...@li...> > > > -------------------------------------------------------------------------> -----> All the data continuously generated in your IT infrastructure contains a> definitive record of customers, application performance, security> threats, fraudulent activity and more. Splunk takes this data and makes> sense of it. Business sense. IT sense. Common sense.> _______________________________________________> Opalvoip-devel mailing list> Opa...@li...> From: Dennis M. - 2011-09-20 00:48:35 On Sep 19, 2011, at 5:58 PM, Robert Jongbloed wrote:> Wow! I cannot imagine what could have changed between 3.8 and 3.11 to cause> that! There were
2025-04-05My own operators. They> do> >>>> the same thing the ios_base.h defined ones do. It works fine. I'll> >>>> have to dig into this more when I have time. In the mean time I'll> >> just> >>>> let it simmer and see if anyone else has the issue. It may be some> >>>> compiler flag or setting.> >>> And of course, it might mysteriously disappear with a random compiler> >>> upgrade ...> >> It disappeared by removing enable-pic from the compiler flags... just> in> >> case anyone runs across this in the future.> >>> >> -Dennis> >>> >>> >> ----------------------------------------------------------------------> ---> >> -----> >> All of the data generated in your IT infrastructure is seriously> >> valuable.> >> Why? It contains a definitive record of application performance,> security> >> threats, fraudulent activity, and more. Splunk takes this data and> makes> >> sense of it. IT sense. And common sense.> >> >> _______________________________________________> >> Opalvoip-devel mailing list> >> Opa...@li...> >> >> > -----------------------------------------------------------------------> -------> > All the data continuously generated in your IT infrastructure contains> a> > definitive record of customers, application performance, security> > threats, fraudulent activity and more. Splunk takes this data and makes> > sense of it. Business sense. IT sense. Common sense.> > > _______________________________________________> > Opalvoip-devel mailing list> > Opa...@li...> > > > -------------------------------------------------------------------------> -----> All the data continuously generated in your IT infrastructure contains a> definitive record of customers, application performance, security> threats, fraudulent activity and more. Splunk takes this data and makes> sense of
2025-04-08Summary Files Reviews Support Wiki Mailing Lists Tickets ▾ Bugs Support Requests Patches Feature Requests Git ▾ ptlib opal Menu ▾ ▴ From: Dennis M. - 2011-09-19 23:36:12 I've actually got my H.460 patch close to ready to submit. It seems to be working properly on the latest trunk code. I'm having a difficult time testing it though because I've run into another issue.Something between the 3.8 branch and current trunk has affected the behavior of the ios_base io manipulator classes. I get an EXC_BAD_INSTRUCTION after a little bit when the anything like this is done:PTRACE(n, "xxx" From: Robert J. - 2011-09-19 23:58:52 Wow! I cannot imagine what could have changed between 3.8 and 3.11 to causethat! There were no fundamental changes there. There hasn't been in years!The largest change is the addition of integrators to PDictionary, but howcould that affect io manipulators? Bizarre.You say "after a little bit", does that imply that sometimes std::hex worksand sometimes it doesn't?Are your builds for 3.8 and 3.11 use identical compiler and environment,correct? What you say below implies it might not be.Robert JongbloedOPAL/OpenH323/PTLib Architect and Co-founder.> -----Original Message-----> From: Dennis Muhlestein [mailto:djm...@gm...]> Sent: Tuesday, 20 September 2011 9:36 AM> To: Opa...@li...> Subject: [Opalvoip-devel] iOS issues> > I've actually got my H.460 patch close to ready to submit. It seems to> be working properly on the latest trunk code. I'm having a difficult> time testing it though because I've run into another issue.> > Something between the 3.8 branch and current trunk has affected
2025-03-28IOS issues> > Wow! I cannot imagine what could have changed between 3.8 and 3.11 to> cause> that! There were no fundamental changes there. There hasn't been in> years!> The largest change is the addition of integrators to PDictionary, but how> could that affect io manipulators? Bizarre.> > You say "after a little bit", does that imply that sometimes std::hex> works> and sometimes it doesn't?> > Are your builds for 3.8 and 3.11 use identical compiler and environment,> correct? What you say below implies it might not be.> > Robert Jongbloed> OPAL/OpenH323/PTLib Architect and Co-founder.> > > > -----Original Message-----> > From: Dennis Muhlestein [mailto:djm...@gm...]> > Sent: Tuesday, 20 September 2011 9:36 AM> > To: Opa...@li...> > Subject: [Opalvoip-devel] iOS issues> >> > I've actually got my H.460 patch close to ready to submit. It seems to> > be working properly on the latest trunk code. I'm having a difficult> > time testing it though because I've run into another issue.> >> > Something between the 3.8 branch and current trunk has affected the> > behavior of the ios_base io manipulator classes. I get an> > EXC_BAD_INSTRUCTION after a little bit when the anything like this is> > done:> >> > PTRACE(n, "xxx" > or cout >> > Any of those ios_base manipulators cause the program to crash.> >> > I've searched through differences in ptlib's object.h and through the> > code for other issues that might cause this but nothing is turning up.> >> > I got around it
2025-04-02