Re: [flashrom] [PATCH 1/7] Introduce and use enum test_state

[Stefan Tauner]

On Sat, 03 Mar 2012 23:32:23 +0100
Carl-Daniel Hailfinger  wrote:
> Am 03.03.2012 21:11 schrieb Stefan Tauner:
> > Previously boards in the wiki were tagged either as working or as known
> > bad. But we added support to various boards via board enables that were
> > then never tested because the owners have not reported back. This can
> > now be tagged with NT and is shown appropriately.
> >
> > Also, the underlying data structure indicating state was converted from
> > macros to an enum while preserving original integer values.
> >
> > Because all lines specifying supported boards and laptops were touched
> > anyway, this patch also re-indents them.
> >
> > TODO: change other occurrences to use it.  wanted to get feedack first.
> >
> > Signed-off-by: Stefan Tauner
> >
> > diff --git a/print_wiki.c b/print_wiki.c
> > index 377154d..9a9cd83 100644
> > --- a/print_wiki.c
> > +++ b/print_wiki.c
> > @@ -136,9 +136,9 @@ static void wiki_helper(const char *devicetype, int cols,
> >    const struct board_match *b = board_matches;
> >
> > for (i = 0; boards[i].vendor != NULL; i++) {
> > if (boards[i].working)
> > if (boards[i].working == OK)
> > boardcount_good++;
> > else
> > if (boards[i].working == BAD)
> > boardcount_bad++;
>
> You could replace that construct with a switch(), and it might make
> sense to count untested boards as well.

i could, but it would increase the line count by 100%, untested board
count was added.
> > }
> >
> > @@ -171,7 +171,8 @@ static void wiki_helper(const char *devicetype, int cols,
> >    	       b[k].lb_vendor ? b[k].lb_vendor : "",
> >    	       b[k].lb_vendor ? ":" : "",
> >    	       b[k].lb_vendor ? b[k].lb_part : "",
> >        (boards[i].working) ? "OK" : "No");
> >        (boards[i].working == OK) ? "OK" :
> >        (boards[i].working == NT) ? "?3" : "No");
>
> The ?3 looks odd, but I assume you tested it.

it is a mediawiki template:
http://flashrom.org/index.php?title=Template:%3F3&action=edit
> […]
> Is it possible that this patch made the board table one or two tabs
> wider? The patch looks like that, and while I agree that some
> files/sections should not have line length limits, adding another 16
> columns of whitespace is something I'd like to avoid.

reverted that part...

i have committed a refined version now in r1555, thanks for your
suggestions!