> -----Original Message-----
> From: Joerg Bruehe [mailto:joerg.bruehe@stripped]
> Sent: Freitag, 28. Januar 2011 16:25
> To: Vladislav Vaintroub
> Cc: commits@stripped
> Subject: Re: bzr push into mysql-5.5-bugteam branch (joerg:3108 to 3109)
>
> Hi Wlad,
>
>
> thanks for the hint:
>
> Vladislav Vaintroub wrote:
> > Hi Joerg,
> >
> >> -----Original Message-----
> >> From: Joerg Bruehe [mailto:joerg@stripped]
> >> Sent: Freitag, 28. Januar 2011 15:28
> >> To: commits@stripped
> >> Subject: bzr push into mysql-5.5-bugteam branch (joerg:3108 to 3109)
> >
> >
> >> FILE(READ ${CMAKE_SOURCE_DIR}/INFO_SRC SOURCE_INFO)
> >> FILE(WRITE ${INFO_SRC} "${SOURCE_INFO}\n")
> >
> > if you just want to copy a file,
> > configure_file(<input> <output> COPYONLY)
> >
> > is your friend, as documented in the inline-help : cmake --help-command
> configure_file
>
> I'm still in the test/development phase, therefore the many debug lines.
> But yes, "configure_file(... copyonly)" looks quite clean.
>
> However, I just stumbled over the fact that using a relative "path" in
> IF(EXISTS path)
> FILE(READ path ...)
> I get a read error, because the "exists" and the "read" interpret that
> name relative to different start points:
> - "exists" uses the current directory,
> - "read" uses the source tree.
I found it simpler to always use full paths (except in ADD_EXECUTABLE or ADD_LIBRARY)
Variables
${CMAKE_CURRENT_BINARY_DIR}
${CMAKE_BINARY_DIR}
${CMAKE_CURRENT_SOURCE _DIR}
${CMAKE_SOURCE _DIR}
make it easy (also for the reader) and there is no ambiguity.
> So I realize any change needs lots of testing.
>
>
> Regards,
> Jörg
>
> --
> Joerg Bruehe, MySQL Build Team, joerg.bruehe@stripped
> (+49 30) 417 01 487
> ORACLE Deutschland B.V. & Co. KG, Komturstrasse 18a, D-12099 Berlin
> Geschaeftsfuehrer: Juergen Kunz, Marcel v.d. Molen, Alexander v.d. Ven
> Amtsgericht Muenchen: HRA 95603
