[Lldb-commits] [PATCH] D14790: cmake: add -DLLDB_ALLOW_STATIC_BINDINGS=1, defaults off

[Todd Fiala via lldb-commits]

Moot point now, but thanks.

The reason for the flag name is that the prepare_bindings.py script as it
stands will always generate the bindings with swig if a swig is either
specified or allowed to be found in some common places.  So the
"USE_STATIC_BINDINGS" would be misleading --- they wouldn't necessarily be
used - only when a swig could not be found.  So that flag literally
controlled whether static bindings were even permitted (i.e. allowed) to be
used, not that they would be.

On Thu, Nov 19, 2015 at 2:25 AM, Pavel Labath <labath at google.com> wrote:

> labath added a comment.
>
> A bit late to the party, but anyway... I would change the detection logic
> here:
> Instead of `ALLOW_STATIC_BINDINGS`, have `USE_STATIC_BINDINGS`. If it's
> set, don't even bother checking for swig's presence and use the static
> bindings. If it's not set, then detect swig, and make it an error if it's
> not found.
> I like this more because then the build will not do something completely
> different depending on whether it finds some binary on your system or not.
> I don't feel so strongly about that as in the case of curses (this is the
> dynamic dependency kamil was referring to) because this only kicks in if
> some option is specified, but I do think this provides more visibility into
> what the build system is doing.
>
>
> http://reviews.llvm.org/D14790
>
>
>
>


-- 
-Todd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/lldb-commits/attachments/20151119/3a91a5f0/attachment-0001.html>