[gdal-dev] building with PDF driver on Windows + ?

[Joaquim Luis]

Brent,

 From memory, as I don't find the CMake cash with exact details. I first 
built freetype, which I think is mandatory, and than because I already 
had them built but don't know if they are a must-be, I added also zlib, 
jpeg to the CMake list.
My first error was to not select 'splash' because I didn't know what it 
was but it's part of poppler and needs to be selected. I did not include 
any Qt stuff and probably some other.
Once it is built (I used only VC2010) there is the mess of the includes. 
You will need to include the root directory where the whole poppler tree 
lives, plus that ../poppler  plus still one other dir that holds only 
one .h file. I moved that .h to one of the other places and now use this 
(my names) in nmake.opt

POPPLER_CFLAGS = -IC:/programs/compa_libs -IC:/programs/compa_libs/poppler

Good luck

Joaquim


>  I'm about to embark on compiling Poppler on Windows to get Geospatial 
> PDF support in GDAL.  Any recommendations with respect to compiler 
> version, dependencies, CMake options, etc?
>
> Thanks!
> Brent Fraser
>
> On 10/14/2010 3:23 PM, Joaquim Luis wrote:
>> Did I mention before that the propeller (sorry, poppler) doesn't have 
>> any building instructions for Windows?
>>
>> Well, the CMakeLists.txt is incomplete and does not add the contents 
>> of the "splash" directory to the project. After adding all *.cc from 
>> 'splash' to project, GDAL is happy with the poppler.lib and now 
>> gdalinfo says
>>
>> ...
>>   KMLSUPEROVERLAY (rw): Kml Super Overlay
>>   XYZ (rwv): ASCII Gridded XYZ
>>   HF2 (rwv): HF2/HFZ heightfield raster
>>   PDF (rov): Geospatial PDF
>>
>> but I'm now confused with the presence of the KMLSUPEROVERLAY because 
>> my nmake.opt has
>>
>> # Uncomment out the following to enable KML Super-Overlay support.
>> #KMLSUPEROVERLAY_SUPPORTED = YES
>> #MINIZIP_INCLUDE = -I$(OSSIM_HOME)\minizip\src
>> #MINIZIP_LIBRARY = $(OSSIM_HOME)\minizip\release\minizip.lib
>>
>>
>> Not complaining. Just reporting.
>>
>> Joaquim
>>
>>
>>
>>
>>> On 14-10-2010 18:32, Even Rouault wrote:
>>>> Le jeudi 14 octobre 2010 17:01:06, Joaquim Luis a écrit :
>>>>> Hi,
>>>>>
>>>>> I tried to build GDAL on Win with PDF support and CV2010
>>>>>
>>>>> Well that's an adventure.
>>>> I trust you and didn't even try this way.
>>>>
>>>> Instead I just downloaded the kde-win32 installer, used the 
>>>> "packager mode"
>>>> (or whatever it is called. I'm just quoting from my memory of doing 
>>>> this a few
>>>> weeks ago), and selected the poppler, freetype and lcms packages 
>>>> and their
>>>> developement packages. The only requirement for GDAL is poppler, 
>>>> freetype and
>>>> lcms appears to be poppler requirements on this KDE build. But 
>>>> those lib only
>>>> work with MSVC 2008 if I remember.
>>>
>>> Even,
>>>
>>> Hmm, on a further check those symbols are from poppler so it's not a 
>>> lcms fault.
>>> I get another error building poppler referring iconv.h that I ignore 
>>> either but selected as a no dependency in CMake, but this affect the 
>>> creation of a poppler-cpp.lib, not poppler.lib so I'm not sure it 
>>> relates to the GDAL error.
>>> I would like to build these dependencies myself because:
>>>
>>> 1. I don't want to use anything that dares to create manifests 
>>> dependencies
>>> 2. I want to be able to build 64 bits versions too.
>>>
>>> Anyway, I found these warnings too that are unrelated to this PDF 
>>> driver issue
>>>
>>> json_util.c
>>> json_util.c(62) : warning C4013: 'open' undefined; assuming extern 
>>> returning int
>>> json_util.c(71) : warning C4013: 'read' undefined; assuming extern 
>>> returning int
>>> json_util.c(74) : warning C4013: 'close' undefined; assuming extern 
>>> returning int
>>> json_util.c(109) : warning C4013: 'write' undefined; assuming extern 
>>> returning int
>>>
>>> ogrsf_frmts.lib(resolvexlinks.obj) : warning LNK4221: This object 
>>> file does not define any previously undefined public symbols, so it 
>>> will not be used by any link operation that consumes this library
>>>
>>> and many "blabla ... possible loss of data" warnings.
>>>
>>>
>>> _______________________________________________
>>> 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
>>
>
>
>