Also, in response to your other question, this bug
does not exist in 5.0. Apparently, this code was
reworked for 5.1.
-Dave
On Fri, 2007-03-02 at 08:09 -0500, David Shrewsbury wrote:
> Hello Ingo,
>
> Perhaps I may be misreading the code in the fn_format()
> function, but I don't believe always setting MY_REPLACE_EXT
> would work. It looks like that function blindly looks for
> the last occurrence of a period within in the file name.
> If it finds one, and MY_APPEND_EXT is not given but
> MY_REPLACE_EXT is, then it replaces whatever was after
> the period with the extension. If the filename given were
> `tbl.one`, then I think `tbl.one.MYI` would be renamed
> `tbl.MYI` if we always specify MY_REPLACE_EXT. Someone
> please correct me if I'm wrong as I'm sure someone knows
> the code much better than I.
>
> The code that I added was copied (and then slightly
> modified) from a few lines above the change.
>
> -Dave
>
>
> On Fri, 2007-03-02 at 09:06 +0100, Ingo Strüwing wrote:
> > Hi,
> >
> > Am Donnerstag, den 01.03.2007, 23:12 -0500 schrieb David Shrewsbury:
> > > {
> > > + char *iext= strrchr(name, '.');
> > > + int have_iext= iext && !strcmp(iext, MI_NAME_IEXT);
> > > fn_format(filename, name, "", MI_NAME_IEXT,
> > > (MY_UNPACK_FILENAME |
> > > (flags & HA_DONT_TOUCH_DATA) ? MY_RETURN_REAL_PATH : 0)
> |
> > > - MY_APPEND_EXT);
> > > + (have_iext ? MY_REPLACE_EXT : MY_APPEND_EXT));
> > > linkname_ptr=0;
> >
> > Wouldn't it be sufficient to always use MY_REPLACE_EXT? Doesn't it
> > append the extension anyway if none exists?
> >
> > And shouldn't it be fixed in 5.0 too?
> >
> > Regards
> > Ingo
> > --
> > Ingo Strüwing, Senior Software Developer
> > MySQL GmbH, Radlkoferstr. 2, D-81373 München
> > Geschäftsführer: Hans von Bell, Kaj Arnö - HRB München
> 162140
> >
> >
>
>
