[proteowizard-developer] more LGPL code in ProteoWizard?

[Brian P. <bri...@in...>]

Hey All,

 

I've been adding gzip output to various TPP converters (makes it possible,
for example, to directly generate a converted file as .mzML.gz to save on
disk and network bandwidth) and I want to do the same for msconvert.

 

I looked at doing this in a properly boost-y fashion, but the bummer there
is that you have to replace std::ostream with
boost::iostreams::filtering_ostream, which implements an ostream-like
interface but is not actually an ostream.  Thus a lot of code that uses
ostream has to be torn up to use the boost stuff (this seems like a screwup
on boost's part, but I digress).
 
Happily the TPP codebase has a lovely std::ostream derived class that does
gzip output, based on the fairly ubiquitous gzstream implementation by
Deepak Bandyopadhyay and Lutz Kettner.  I'd be happy to move this into pwiz
as we did with my random access gzip reader, but the code is based on LGPL
work so it gives me pause.  I see that this wouldn't be the only LGPL code
in ProteoWizard (there's also
pwiz\utility\vendor_api\thermo\ScanFilter.cpp/h) but it would be darned near
the only.
 
So, I thought I should ask first.
 
Brian