[gdal-dev] GDAL 1.11 release plan

Wed Mar 26 08:16:24 PDT 2014

Selon Olivier BARTHELEMY <barthelemy at geovariances.com>:

I've fixed the one in adrgdataset.cpp.
gifalloc.c, degrib18, hdf-eos are third-party code "imported" into GDAL tree. So
the best way would be to fix the issues upstream (if fix is actually needed) and
then downstream into GDAL.
I've analyze errors reported in rasterlitedataset.cpp, ogrdatasource.cpp,
ogrpgdatasource.cpp and strtod, and they are false positives.  So the rate of
false positives with cppcheck is quite significant.
I haven't looked at the other files that I'm less familiar with.

Even

> Here is what cppcheck 1.64 reports as 'errors' in gdal-1.10.1:
>
> [frmts\adrg\adrgdataset.cpp:1071]: (error) Memory leak: TILEINDEX
> [frmts\coasp\coasp_dataset.cpp:199]: (error) Undefined behavior: Variable
> 'pszItemValue' is used as parameter and destination in s[n]printf().
> [frmts\dted\dted_create.c:181]: (error) Resource leak: fp
> [frmts\envisat\EnvisatFile.c:286]: (error) Memory leak: self
> [frmts\envisat\EnvisatFile.c:609]: (error) Memory leak: dsd_text
> [frmts\envisat\EnvisatFile.c:1776]: (error) Memory leak: entry
> [frmts\georaster\georaster_dataset.cpp:188]: (error) Possible null pointer
> dereference: poGRD
> [frmts\georaster\georaster_dataset.cpp:189]: (error) Possible null pointer
> dereference: poGRD
> [frmts\georaster\georaster_dataset.cpp:190]: (error) Possible null pointer
> dereference: poGRD
> [frmts\georaster\georaster_dataset.cpp:191]: (error) Possible null pointer
> dereference: poGRD
> [frmts\georaster\oci_wrapper.cpp:1590]: (error) Pointer to local array
> variable returned.
> [frmts\gif\giflib\gifalloc.c:67]: (error) Memory leak: Object
> [frmts\grib\degrib18\degrib\clock.c:2071]: (error) Common realloc mistake:
> 'Stack' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\clock.c:2175]: (error) Common realloc mistake:
> 'Rel' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\clock.c:2197]: (error) Common realloc mistake:
> 'Rel' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\clock.c:2209]: (error) Common realloc mistake:
> 'Rel' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\grib2api.c:594]: (error) Uninitialized
> variable: igdstmpl
> [frmts\grib\degrib18\degrib\metaparse.cpp:2628]: (error) Common realloc
> mistake: 'freq' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:109]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:126]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:132]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:149]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:158]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:187]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:195]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:203]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:211]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:218]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:229]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:241]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:257]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:442]: (error) Common realloc mistake:
> 'preBuffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myerror.c:534]: (error) Common realloc mistake:
> 'buff' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myutil.c:85]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myutil.c:92]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myutil.c:99]: (error) Common realloc mistake:
> 'buffer' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\myutil.c:149]: (error) Common realloc mistake:
> 'argv' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\tdlpack.cpp:3256]: (error) Common realloc
> mistake: 'subGroup' nulled but not freed upon failure
> [frmts\grib\degrib18\degrib\tdlpack.cpp:4179]: (error) Read and write
> operations without a call to a positioning function (fseek, fsetpos or
> rewind) or fflush in between result in undefined behaviour.
> [frmts\grib\degrib18\g2clib-1.0.4\compack.c:159]: (error) Uninitialized
> variable: ival1
> [frmts\grib\degrib18\g2clib-1.0.4\comunpack.c:198]: (error) Memory leak:
> ifld
> [frmts\grib\degrib18\g2clib-1.0.4\misspack.c:203]: (error) Uninitialized
> variable: ival1
> [frmts\grib\degrib18\g2clib-1.0.4\misspack.c:118]: (error) Uninitialized
> variable: rmisss
> [frmts\grib\degrib18\g2clib-1.0.4\pngunpack.c:61]: (error) Memory leak: ifld
> [frmts\grib\degrib18\g2clib-1.0.4\pngunpack.c:61]: (error) Memory leak:
> ctemp
> [frmts\hdf4\hdf-eos\EHapi.c:2315]: (error) Common realloc mistake:
> 'metabuf' nulled but not freed upon failure
> [frmts\hdf4\hdf-eos\EHapi.c:3665]: (error) Common realloc mistake:
> 'EHXtypeTable' nulled but not freed upon failure
> [frmts\hdf4\hdf-eos\EHapi.c:3668]: (error) Common realloc mistake:
> 'EHXacsTable' nulled but not freed upon failure
> [frmts\hdf4\hdf-eos\EHapi.c:3671]: (error) Common realloc mistake:
> 'EHXfidTable' nulled but not freed upon failure
> [frmts\hdf4\hdf-eos\EHapi.c:3674]: (error) Common realloc mistake:
> 'EHXsdTable' nulled but not freed upon failure
> [frmts\hdf4\hdf-eos\GDapi.c:2411]: (error) Array 'HDFcomp[5]' accessed at
> index 19, which is out of bounds.
> [frmts\hdf4\hdf-eos\GDapi.c:2177]: (error) Array 'originNames[4]' accessed
> at index 15, which is out of bounds.
> [frmts\hdf4\hdf-eos\GDapi.c:2292]: (error) Array 'pixregNames[2]' accessed
> at index 7, which is out of bounds.
> [frmts\hdf4\hdf-eos\SWapi.c:4995]: (error) Memory leak: fillbuf
> [frmts\hdf4\hdf-eos\SWapi.c:5615]: (error) Memory leak: flag
> [frmts\hdf4\hdf-eos\SWapi.c:6239]: (error) Memory leak: flag
> [frmts\idrisi\IdrisiDataset.cpp:1886]: (error) syntax error
> [frmts\ilwis\ilwisdataset.cpp:1425]: (error) syntax error
> [frmts\msg\msgdataset.cpp:532]: (error) Object pointed by an 'auto_ptr' is
> destroyed using operator 'delete'. You should not use 'auto_ptr' for
> pointers obtained with operator 'new[]'.
> [frmts\netcdf\netcdfdataset.cpp:5997]: (error) Uninitialized variable:
> strtime
> [frmts\pcidsk\sdk\core\pcidsk_utils.cpp:644]: (error) Common realloc
> mistake: 'pszWorkBuffer' nulled but not freed upon failure
> [frmts\pcidsk\sdk\core\pcidskexception.cpp:170]: (error) Common realloc
> mistake: 'pszWorkBuffer' nulled but not freed upon failure
> [frmts\png\libpng\pngrutil.c:996]: (error) Uninitialized variable: igamma
> [frmts\rasterlite\rasterlitedataset.cpp:1179]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1185]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1186]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1187]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1188]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1190]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1191]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1192]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1193]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1194]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1195]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1196]: (error) Possible null
> pointer dereference: poDS
> [frmts\rasterlite\rasterlitedataset.cpp:1267]: (error) Possible null
> pointer dereference: poDS
> [frmts\raw\atlsci_spheroid.cpp:122]: (error) syntax error
> [frmts\raw\mffdataset.cpp:1451]: (error) syntax error
> [frmts\sgi\sgidataset.cpp:523]: (error) Using 'memcpy' on struct that
> contains a 'std::string'.
> [ogr\ogrsf_frmts\dgn\dgnhelp.cpp:1379]: (error) syntax error
> [ogr\ogrsf_frmts\dwg\ogrdwglayer.cpp:323]: (error) Invalid number of
> character ({) when these macros are defined: 'notdef'.
> [ogr\ogrsf_frmts\filegdb\FGdbLayer.cpp:1914]: (error) syntax error
> [ogr\ogrsf_frmts\generic\ogrdatasource.cpp:1425]: (error) Possible null
> pointer dereference: papoSrcLayers
> [ogr\ogrsf_frmts\mitab\mitab_coordsys.cpp:747]: (error) Buffer is accessed
> out of bounds.
> [ogr\ogrsf_frmts\mitab\mitab_geometry.cpp:223]: (error) Memory leak:
> xintersect
> [ogr\ogrsf_frmts\ntf\ntffilereader.cpp:260]: (error) Memory pointed to by
> 'poRecord' is freed twice.
> [ogr\ogrsf_frmts\pg\ogrpgdatasource.cpp:1102]: (error) Memory leak: poLayer
> [ogr\ogrsf_frmts\shape\shptree.c:241]: (error) Memory leak: psTree
> [port\cpl_minizip_zip.cpp:608]: (error) Memory leak: zi
> [port\cpl_strtod.cpp:67]: (error) Division by zero.
>
> I'd be happy to provide fix patches, but i have not much time at the
> moment, so it won't be before a 1.11 release.
>
> If you want to see the other 'warnings', and get un updated report on GDAL
> trunk, maybe you could ask to those who manage
> http://debbie.postgis.net:8080/view/GDAL/ to add the cppcheck plugin to
> their Jenkins?
>
>
> 2014-03-26 9:15 GMT+01:00 Even Rouault <even.rouault at mines-paris.org>:
>
> > Hi Olivier,
> >
> > If you could share with us the errors that have been reported (and the
> > options
> > you have used to pass cppcheck on GDAL source), that would be a good
> > contribution. Even better if you can submit a patch to fix them ;-)
> > Not saying they shouldn't be fixed, but I don't see that as a blocker point
> > however, unless they show up in nominal code paths. They have likely
> > existed for
> > years.
> > We also have registered GDAL to the Coverity scanner, and we try to fix
> > errors
> > that are reported in new code.
> >
> > Even
> >
> > > I think you should take a look at what cppcheck finds when scanning GDAL
> > > source before the next release. There are a few tens of small errors.
> > > Mostly, small leaks in particalr cases. Not very important errors, but
> > that
> > > ought to be fixed before doing an release in my opinion
> > >
> > >
> > > 2014-03-26 8:49 GMT+01:00 Paul Meems <bontepaarden at gmail.com>:
> > >
> > > > +1 for adding VS2013 binaries in GISInternals
> > > >
> > > > Thanks,
> > > >
> > > > Paul
> > > >
> > > >  *Paul Meems *
> > > > Release manager, configuration manager
> > > > and forum moderator of MapWindow GIS.
> > > > www.mapwindow.org
> > > >
> > > > Owner of MapWindow.nl - Support for
> > > > Dutch speaking users.
> > > > www.mapwindow.nl
> > > >
> > > >
> > > >
> > > > *Join us at the MapWindow GIS Conference 2014
> > > > <http://geogis.detek.unideb.hu/TKonferencia/2014/>, in Debrecen
> > Hungary*
> > > >
> > > >
> > > > 2014-03-26 6:25 GMT+01:00 xavier lhomme <lhomme.xavier at gmail.com>:
> > > >
> > > > Hi
> > > >>   Adding Visual Studio 2012 2013 in GISInternals for Gdal 1.11 should
> > be
> > > >> great too.
> > > >> Xav
> > > >> Le 25 mars 2014 22:57, "Tamas Szekeres" <szekerest at gmail.com> a
> > écrit :
> > > >>
> > > >> That sounds good to me.
> > > >>>
> > > >>> Best regards,
> > > >>>
> > > >>> Tamas
> > > >>>
> > > >>>
> > > >>>
> > > >>> 2014-03-25 22:37 GMT+01:00 Even Rouault <
> > even.rouault at mines-paris.org>:
> > > >>>
> > > >>>> Hi,
> > > >>>>
> > > >>>> GDAL 1.10 was released about one year ago, so it is time to think
> > about
> > > >>>> the next
> > > >>>> release. I think we should just call it 1.11 given that there are no
> > > >>>> major
> > > >>>> architectural changes (although there have been non trivial changes
> > in
> > > >>>> RAT
> > > >>>> management and for multiple geometry field support, but that only
> > > >>>> impacts core &
> > > >>>> drivers), and the C API and ABI is still preserved (for those who
> > have
> > > >>>> looked at
> > > >>>> the other thread, my experiments on unification are *not* in trunk).
> > > >>>> For the
> > > >>>> details, I have initiated the preliminary NEWS :
> > > >>>> http://trac.osgeo.org/gdal/export/27091/trunk/gdal/NEWS (*)
> > > >>>> There are some mentions to 2.0 in the documentation of new features,
> > > >>>> but I'll
> > > >>>> end up fixing them to reflect the 1.11 numbering.
> > > >>>>
> > > >>>> I think we are in a good shape to start the beta phase soon. If
> > people
> > > >>>> have
> > > >>>> features or bug fixes almost ready that they'd like to be in 1.11,
> > > >>>> please speak
> > > >>>> up now. Otherwise I suggest issuing a beta-1 at the end of next
> > week,
> > > >>>> give it a
> > > >>>> week for testing, and depending on the outcome, issue a beta-2, or
> > just
> > > >>>> try
> > > >>>> RC-1. Unless someone volunteers (the process is fairly well
> > documented
> > > >>>> in
> > > >>>> HOWTO-RELEASE), I can be release manager for 1.11.
> > > >>>>
> > > >>>> Any opinion about that plan ?
> > > >>>>
> > > >>>> Best regards,
> > > >>>>
> > > >>>> Even
> > > >>>>
> > > >>>> --
> > > >>>> Geospatial professional services
> > > >>>> http://even.rouault.free.fr/services.html
> > > >>>> _______________________________________________
> > > >>>> gdal-dev mailing list
> > > >>>> gdal-dev at lists.osgeo.org
> > > >>>> http://lists.osgeo.org/mailman/listinfo/gdal-dev
> > > >>>>
> > > >>>
> > > >>>
> > > >>> _______________________________________________
> > > >>> gdal-dev mailing list
> > > >>> gdal-dev at lists.osgeo.org
> > > >>> http://lists.osgeo.org/mailman/listinfo/gdal-dev
> > > >>>
> > > >>
> > > >> _______________________________________________
> > > >> gdal-dev mailing list
> > > >> gdal-dev at lists.osgeo.org
> > > >> http://lists.osgeo.org/mailman/listinfo/gdal-dev
> > > >>
> > > >
> > > >
> > > > _______________________________________________
> > > > gdal-dev mailing list
> > > > gdal-dev at lists.osgeo.org
> > > > http://lists.osgeo.org/mailman/listinfo/gdal-dev
> > > >
> > >
> > >
> > >
> > > --
> > > [image: Geovariances]
> > >
> > > Olivier BARTHELEMY *Software development engineer*
> > > Geovariances, 49bis avenue Franklin Roosevelt - 77215 AVON CEDEX - FRANCE
> > > | www.geovariances.com <http://link.geovariances.com/eml-home>
> > > Keep posted about Geovariances   <
> > http://link.geovariances.com/eml-geowidget>
> > >     <http://link.geovariances.com/eml-linkedin-gv>
> > > <http://link.geovariances.com/eml-linkedin>
> > >     <http://link.geovariances.com/eml-twitter>
> > > <http://link.geovariances.com/eml-youtube>
> > >     <http://link.geovariances.com/eml-slideshare>
> > >
> >
> >
> >
>
>
> --
> [image: Geovariances]
>
> Olivier BARTHELEMY *Software development engineer*
> Geovariances, 49bis avenue Franklin Roosevelt - 77215 AVON CEDEX - FRANCE
> | www.geovariances.com <http://link.geovariances.com/eml-home>
> Keep posted about Geovariances   <http://link.geovariances.com/eml-geowidget>
>     <http://link.geovariances.com/eml-linkedin-gv>
> <http://link.geovariances.com/eml-linkedin>
>     <http://link.geovariances.com/eml-twitter>
> <http://link.geovariances.com/eml-youtube>
>     <http://link.geovariances.com/eml-slideshare>
>