1 |
On Mon, Jan 23, 2006 at 11:16:03AM +0100, Marius Mauch wrote: |
2 |
> On Wed, 11 Jan 2006 12:39:03 -0800 |
3 |
> Brian Harring <ferringb@g.o> wrote: |
4 |
> |
5 |
> > Regex you've got there allows for pulling the wrong text- recall, ebd |
6 |
> > originally was doing grep based filtering (regex). Had to rewrite |
7 |
> > that in a major hurry since bash syntax (specifically here ops) |
8 |
> > forces you to track state/constructs rather then just a regex... |
9 |
> |
10 |
> Not really an issue in this case. First the code bails out if more than |
11 |
> one match is found, so unless the metadata assignment is NOT found by |
12 |
> it we don't get the wrong info. |
13 |
> Also a mismatch in this special is so |
14 |
> extremely unlikely that honestly I don't really care about it, |
15 |
> especially as this is a one time conversion (might be different if I'd |
16 |
> have added the on the fly extraction). |
17 |
|
18 |
Re-read that statement. It's a one time conversion- meaning we better |
19 |
get it right the first time, else the user's data is effectively |
20 |
corrupted. Forcing a full regen from the saved environment is not a |
21 |
solution for fixing past corruptions either. |
22 |
|
23 |
If it were on the fly extraction, I wouldn't care quite as much- but |
24 |
the fact this is an untracked change to the users data means we *do* |
25 |
need to cover corner cases. |
26 |
|
27 |
I know you want this in, but it has to be done *right* covering all |
28 |
known corner cases for it- I already wrote the tool that handles the |
29 |
corner cases properly, use it, don't adhoc a solution that |
30 |
re-introduces the potential gaps. |
31 |
|
32 |
If we're not going to do it right, we really shouldn't do it when it |
33 |
comes to upgrades of the vdb. |
34 |
|
35 |
Aside from that, if the code is in debate (as this is), I really |
36 |
don't think it should get slid into svn 2 weeks later effectively |
37 |
unchanged- didn't write that original email just for the hell of it :) |
38 |
|
39 |
~harring |