Re: CVS update: subversion/subversion/libsvn_fs dag.h

[Greg Stein <gs...@lyra.org>]

On Thu, Feb 08, 2001 at 05:26:43PM -0000, jimb@tigris.org wrote:
>...
>   +/* Return a new dag_node_t object referring to the same node as NODE,
>   +   allocated in TRAIL->pool.  */
>   +dag_node_t *svn_fs__dag_dup (dag_node_t *node,
>   +			     trail_t *trail);
>...
>   +/* Return the filesystem containing NODE.  */
>   +const svn_fs_t *svn_fs__dag_get_fs (dag_node_t *node);

The parameters in the above two functions can/should be "const". But that
svn_fs_t return value probably shouldn't. You wouldn't be able to pass the
FS anywhere if you do that.

>...
>   +/* Return the node revision ID of NODE.  The value returned is shared
>   +   with NODE, and will be deallocated when NODE is.  */
>   +const svn_fs_id_t *svn_fs__dag_get_id (dag_node_t *node);
>...
>   +/* Return true iff NODE is mutable.  */
>   +int svn_fs__dag_is_mutable (dag_node_t *node);
>...
>   +/* Return true iff NODE is a file/directory.  */
>   +int svn_fs__dag_is_file (dag_node_t *node);
>   +int svn_fs__dag_is_directory (dag_node_t *node);

These could be const parameters, too.

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/

Re: CVS update: subversion/subversion/libsvn_fs dag.h

[Jim Blandy <ji...@zwingli.cygnus.com>]

Greg Stein <gs...@lyra.org> writes:

> 
> On Thu, Feb 08, 2001 at 05:26:43PM -0000, jimb@tigris.org wrote:
> >...
> >   +/* Return a new dag_node_t object referring to the same node as NODE,
> >   +   allocated in TRAIL->pool.  */
> >   +dag_node_t *svn_fs__dag_dup (dag_node_t *node,
> >   +			     trail_t *trail);
> >...
> >   +/* Return the filesystem containing NODE.  */
> >   +const svn_fs_t *svn_fs__dag_get_fs (dag_node_t *node);
> 
> The parameters in the above two functions can/should be "const". But that
> svn_fs_t return value probably shouldn't. You wouldn't be able to pass the
> FS anywhere if you do that.

That `const svn_fs_t' was a cut-and-paste-o.  Thanks.

I don't want the node arguments to be const, though.  There's no
reason those functions shouldn't be permitted to change the contents
of the node for some internal reason if they please.