Re: [gdal-dev] Trying to debug a GDAL crash in "GDALClose(hDataset); "

2015-11-13 Thread Kurt Schwehr
nLockCount should be set by https://trac.osgeo.org/gdal/browser/trunk/gdal/gcore/gdal_priv.h#L598 https://trac.osgeo.org/gdal/browser/trunk/gdal/port/cpl_atomic_ops.cpp#L56 which uses InterlockedCompareExchange to change it. Can you give the whole stack trace? On Fri, Nov 13, 2015 at 6:47 PM,

Re: [gdal-dev] Current SVN doesn't build with VS

2015-11-13 Thread Kurt Schwehr
woohoo! Thank! I'll see if I can address any of those other warnings: https://trac.osgeo.org/gdal/ticket/6219 I will give https://trac.osgeo.org/gdal/ticket/6212 a couple days and close it out if I don't get any more failure reports On Fri, Nov 13, 2015 at 4:10 PM, Joaquim Luis wrote: > Kurt

[gdal-dev] Trying to debug a GDAL crash in "GDALClose(hDataset); "

2015-11-13 Thread Joaquim Luis
Hi, I'm having strange crashes in GDAL (SVN) when called via GMT. The crashes occur at a call to GDALClose(hDataset); Now, this is only occurs when I try to read a sub-region of a grid but those are GMT details. It doesn't happen to colleagues on OSX but they are using gdal 1.11. Can't s

Re: [gdal-dev] Current SVN doesn't build with VS

2015-11-13 Thread Joaquim Luis
Kurt, Yes, that fixed it thanks. I see now only this warning v:\gdal\frmts\hdf4\hdf4imagedataset.cpp(2550) : warning C4701: potentially uninitialized local variable 'iWrkNumType' used but it's probably a false positive. As are ... v:\gdal\frmts\hdf4\hdf-eos\ehapi.c(449) : warning C4701: pot

Re: [gdal-dev] Current SVN doesn't build with VS

2015-11-13 Thread Kurt Schwehr
Hi Joaquim, Yes, MAX should work - that was what was originally in the code. I'm trying hard to get away from macros and such and get to the most robust code we can. I really want to find a way that we can use std::max which is the C++ way to g

Re: [gdal-dev] OGR_Dr_Open deprecated

2015-11-13 Thread Ari Jolma
I'm afraid I've missed this change quite much since it was so nicely backwards compatible. It doesn't make sense to have two classes in the Perl bindings when there is really only one in the library - it's too confusing. That means I'll need to remove the OGR Driver and DataSource classes comp

Re: [gdal-dev] Current SVN doesn't build with VS

2015-11-13 Thread Joaquim Luis
Hi Kurt, I'll update but later. Now I have to go. Meanwhile I had replaced std::max by MAX and it passed that point. Hi Joaquim, I broke the build with https://trac.osgeo.org/gdal/changeset/31403/. Even created a ticket: https://trac.osgeo.org/gdal/ticket/6212 and I added your >erro

Re: [gdal-dev] Current SVN doesn't build with VS

2015-11-13 Thread Kurt Schwehr
Hi Joaquim, I broke the build with https://trac.osgeo.org/gdal/changeset/31403/. Even created a ticket: https://trac.osgeo.org/gdal/ticket/6212 and I added your error message to the ticket. http://build.gisinternals.com/sdk/Default.aspx has been down, so it's been tough to see what is happening.

[gdal-dev] Current SVN doesn't build with VS

2015-11-13 Thread Joaquim Luis
The error is v:\gdal\frmts\hdf4\hdf4imagedataset.cpp(2386) : error C2784: '_Ty std::max(std::initializer_list<_Elem>,_Pr)' : could not deduce template argument for 'std::initializer_list<_Elem>' from 'int' c:\program files (x86)\microsoft visual studio 12.0\vc\include\algorithm(4108

Re: [gdal-dev] OGR_Dr_Open deprecated

2015-11-13 Thread Ari Jolma
On 13.11.2015 16:37, Even Rouault wrote: Not sure if it is a "swig" limitation, or something in the Perl bindings. But the Python and Java bindings for sure can use the vector part of GDALDatasetShadow. Ok, I've missed all that. Sigh. Until now there's been a strict difference between raste

Re: [gdal-dev] OGR_Dr_Open deprecated

2015-11-13 Thread Even Rouault
Le vendredi 13 novembre 2015 15:17:40, Ari Jolma a écrit : > On 13.11.2015 15:58, Even Rouault wrote: > > Le vendredi 13 novembre 2015 14:48:02, Ari Jolma a écrit : > >> OGR_Dr_Open became deprecated in 2.0 but it is still the main method to > >> open vector datasets in the bindings. > >> > >> I j

Re: [gdal-dev] OGR_Dr_Open deprecated

2015-11-13 Thread Ari Jolma
On 13.11.2015 15:58, Even Rouault wrote: Le vendredi 13 novembre 2015 14:48:02, Ari Jolma a écrit : OGR_Dr_Open became deprecated in 2.0 but it is still the main method to open vector datasets in the bindings. I just realized that I would need one method to open all data sets. OpenEx seems to

Re: [gdal-dev] OGR_Dr_Open deprecated

2015-11-13 Thread Even Rouault
Le vendredi 13 novembre 2015 14:48:02, Ari Jolma a écrit : > OGR_Dr_Open became deprecated in 2.0 but it is still the main method to > open vector datasets in the bindings. > > I just realized that I would need one method to open all data sets. > OpenEx seems to be it but I would still need to cas

[gdal-dev] OGR_Dr_Open deprecated

2015-11-13 Thread Ari Jolma
OGR_Dr_Open became deprecated in 2.0 but it is still the main method to open vector datasets in the bindings. I just realized that I would need one method to open all data sets. OpenEx seems to be it but I would still need to cast the object into a Dataset or a DataSource. Maybe OpenEx should

[gdal-dev] Error building Gdal on Mac OS X with Filegeodb

2015-11-13 Thread Gane R
I was trying to build gdal-2.0.1 with Filegeodb (File Geodatabase API 1.4 version for Mac 64-bit (Gcc) for Mac running on intel OS X Yosemite 10.10.5 I get the error undefined symbols for architecture x86_64 at Filegeodb modules. I think all the compiler and OS requirements are met. I tried to b