<br><br><div class="gmail_quote">2008/10/28 Howard Butler <span dir="ltr"><<a href="mailto:hobu.inc@gmail.com">hobu.inc@gmail.com</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="Ih2E3d"><br>
On Oct 23, 2008, at 9:47 AM, Martin Vales wrote:<br>
<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi:<br>
<br>
today Roberto Antolin and me are working with liblas for first time with c++ and .NET. We found several problems with c++ and C APIs:<br>
<br>
- CAPI not exposed in windows. I am using the mateus compilation and mi mingw compilation.<br>
I tried use this file but donīt have the CAPI or i can not see it:<br>
<a href="http://liblas.org/raw-attachment/wiki/1.0.0/liblas-1.0.0-win32.zip" target="_blank">http://liblas.org/raw-attachment/wiki/1.0.0/liblas-1.0.0-win32.zip</a><br>
<br>
<br>
</blockquote>
<br></div>
When I look at this file with depends.exe, I can see the C API entry points. The liblas_i contains the implib for liblas. Is this what you are looking for?</blockquote><div><br>humm. I donīt know much about visual c++ compiler but in the mateus compilation i can see the entry points, but in your compilation i can not. But you have the C utilities working (based in the CAPI). Mateus are you compiling at the same way than hobu? <br>
By default in mingw like in gcc the entries point are exported by default. In visual c++ i think itīs a little different but i donīt know that compiler. In mingw you can use the <b>__declspec(dllexport) to export but is not mandatory.</b><br>
<a href="http://www.geocities.com/yongweiwu/stdcall.htm">http://www.geocities.com/yongweiwu/stdcall.htm</a><br><br>Itīs very extrange but i am totally sure i can not see the CAPI in .NET with your compilation or not using the same number of functions.<br>
<br>the other day a visual c++ friend show me his bindings to .NET and he had the entries point based in numbers (@1,@2, etc...) I instead try to find the real name in the API.<br><br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="Ih2E3d"><br>
<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
- CAPI GetPointAt always the same pointer:<br>
</blockquote>
<br></div>
Mateusz,<br>
<br>
This makes sense right? &(reader->GetPoint()) is going to get the address of the reader's current point? The Python API uses GetPointAt (C API) and it seems to work ok.<br>
</blockquote><div><br>the problem is i can not have more than 2 points at the same time. Last week we need read the point i,i-1 and i+1, but like in the CAPI you use the same point i have the same point always. I think is the same in python because use the same API. I think the reader need create a new instante for every point.<br>
<br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>
<br>
<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div class="Ih2E3d">
LAS_DLL const LASPointH LASReader_GetPointAt(const LASReaderH hReader, liblas::uint32_t position)<br>
<br></div><div class="Ih2E3d">
291 LASReader *reader = ((LASReader*) hReader);<br>
292 if (reader->ReadPointAt((std::size_t) position))<br>
<br></div><div class="Ih2E3d">
294 return (LASPointH) &(reader->GetPoint());<br>
<br>
</div></blockquote>
<br>
</blockquote></div><br>