Fix 6.0 source file permissions and encoding

[dmorissette]

Reported by Peter Hopfgartner on mapserver-users (MS 6.0.0-beta1)

The spec file for MapServer has the following code:

chmod -x tests/home.png
chmod -x mapscript/php/php_mapscript.c
chmod -x mappluginlayer.c
chmod -x mapscript/php/mapscript_i.c
chmod -x mapscript/php/examples/test_writeshape.phtml
chmod -x mapscript/python/examples/*.py
chmod -x mapscript/python/tests/rundoctests.dist
chmod -x mapscript/perl/examples/*.pl
chmod -x mapscript/php/README

(...)

iconv -f ISO-8859-1 -t UTF-8 < \
mapscript/java/examples/QueryByAttributeUnicode.java > \
mapscript/java/examples/QueryByAttributeUnicode-tmp.java
mv -f mapscript/java/examples/QueryByAttributeUnicode-tmp.java \
mapscript/java/examples/QueryByAttributeUnicode.java

Could these issues be fixed upstream?

[dmorissette]

I have used "svn propdel svn:executable" to remove the executable flag on the following files in r11147 and r11148:

  mappluginlayer.c
  mapkmlrenderer.cpp
  mapscript/php/Makefile.in
  mapscript/php/mapscript_i.c
  mapscript/php/Makefile.vc
  mapscript/php/php_mapscript.c
  mapscript/php/README.WIN32
  mapscript/php/php_mapscript.h
  mapscript/php/examples/test_writeshape.phtml
  tests/home.png
  mapogcfiltercommon.c
  aclocal.m4
  Makefile.in
  mapkmlrenderer.h
  mapkml.cpp

However I did not remove the executable flag on the following files from your list since I believe they were intentionally made executable. Is it a problem to leave them like this?

 mapscript/python/examples/*.py
 mapscript/python/tests/rundoctests.dist
 mapscript/perl/examples/*.pl

[dmorissette]

About the QueryByAttributeUnicode.java encoding, I'm not setup to test it and I don't understand how this may have worked before with the wrong encoding so I don't want to risk breaking it by fixing it(!).

I'll let someone more familiar with the Java wrappers take care of it.

[dmorissette]

tamas, unicoletti: What do you think of the encoding problem reported in QueryByAttributeUnicode.java ?

[unicoletti]

It used to work and it still should work if your computer had the necessary de_DE encodings. Mine hasn't and since UTF-8 is getting more and more popular it seems reasonable to modify both the source and the Makefile so that the tests pass.

[unicoletti]

Done, fixed by converting QueryByAttributeUnicode.java to utf8 and editing the Makefile so that java receives the correct environment

r11474