GNATS - I give up

[Joseph S. Myers]

On Sat, 5 May 2001, Neil Booth wrote:

> Incidentally, a range of PRs around 2386 only half-exist - trying to
> view them says "Error - no PR matched" which looks like database
> corruption, as they do appear on web queries.

As discussed on overseers - someone (who understands GNATS and has shell
access with whatever privileges are needed) needs to volunteer to restore
these PRs using the gcc-prs mbox archives.  Then, gen-index may need to be
rerun and the results carefully compared (see
http://gcc.gnu.org/ml/gcc/2001-04/msg00071.html ).

PRs in two places in the database: 2138 2144 2234 2354

Assuming that each PR should be in exactly one category directory:

2138: the category "c++" copy should go, the "libstdc++" one should stay.
2144: the category "pending" copy should go, the "c++" one should stay.
2234: the category "preprocessor" copy should go, the "c++" one should
stay.
2354: the category "c++" copy should go, the "target" one should stay.

PRs missing: 2386 2389 2392 2394 2397 2403 2404 2419 2420 2426

-- 
Joseph S. Myers
jsm28@cam.ac.uk