[gdal-dev] Removing one of the two svn:keyword Id lines in all source files

[Kurt Schwehr]

See: https://trac.osgeo.org/gdal/ticket/6547

I propose to remove the $id line from the comments at the top of the file
for files that already have CPL_CVSID.  Having two lines means we have a
redundant copy of all this info, which is changing for every commit.

In addition, I propose to set the svn:keywords for these files to just be
Id.  Some have more and some have none.  e.g.

svn propset svn:keywords Id gdal_rpc.cpp
svn propset svn:keywords Id gdaltransformgeolocs.cpp
svn commit

https://trac.osgeo.org/gdal/changeset/34506

grep Id gdal_rpc.cpp gdaltransformgeolocs.cpp

gdal_rpc.cpp: * $Id: gdal_rpc.cpp 34506 2016-07-01 17:08:58Z goatbar $
gdal_rpc.cpp:CPL_CVSID("$Id: gdal_rpc.cpp 34506 2016-07-01 17:08:58Z
goatbar $");
gdaltransformgeolocs.cpp: * $Id: gdaltransformgeolocs.cpp 34506 2016-07-01
17:08:58Z goatbar $
gdaltransformgeolocs.cpp:CPL_CVSID("$Id: gdaltransformgeolocs.cpp 34506
2016-07-01 17:08:58Z goatbar $");

Then removing the redundant line, it becomes:

grep Id gdal_rpc.cpp gdaltransformgeolocs.cpp

gdal_rpc.cpp:CPL_CVSID("$Id: gdal_rpc.cpp 34506 2016-07-01 17:08:58Z
goatbar $");
gdaltransformgeolocs.cpp:CPL_CVSID("$Id: gdaltransformgeolocs.cpp 34506
2016-07-01 17:08:58Z goatbar $");

Why?  It's extra churn in diffs/patches and adds yet another line to check
for those of use using merge tools like meld a lot (which might just be me).

Anyone against this?  I'm proposing this just for head on trunk.

-kurt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/gdal-dev/attachments/20160701/bf816a0a/attachment.html>