Hi!
Checkpatch regulary complains about bad whitespace:
pavel@amd:/data/l/linux-msm/drivers/input/touchscreen$
../../../scripts/checkpatch.pl --file synaptics_i2c_rmi.c
ERROR: trailing whitespace
#245: FILE: synaptics_i2c_rmi.c:245:
+^Ipr_info("synaptics: version %x, product property %x\n", $
ERROR: trailing whitespace
#250: FILE: synaptics_i2c_rmi.c:250:
+static void compute_areas(struct synaptics_ts_data *ts,^I$
ERROR: trailing whitespace
#326: FILE: synaptics_i2c_rmi.c:326:
+^I^I^I -inactive_area_left, max_x + inactive_area_right, $
ERROR: trailing whitespace
#468: FILE: synaptics_i2c_rmi.c:468:
+^I^Ipr_err("synaptics: Unable to register %s input device\n", $
ERROR: trailing whitespace
#497: FILE: synaptics_i2c_rmi.c:497:
+^Ipr_info("synaptics: Start touchscreen %s in %s mode\n", $
total: 5 errors, 0 warnings, 635 lines checked
I don't think that a bit of trailing whitespace hurts that much (it is
very easy to create when fixing overly long lines, another checkpatch
complain), and at least akpm autofixes the whitespace, anyway, but...
Would it be possible to add autofixing to checkpatch? Fixing
whitespace issues by hand is useless & boring task...
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
On Tuesday 21 July 2009 12:01:14 Pavel Machek wrote:
> Would it be possible to add autofixing to checkpatch? Fixing
> whitespace issues by hand is useless & boring task...
That's why most people use scripts for it.. ;)
scripts/cleanfile
scripts/cleanpatch
Maybe we should make checkpatch inform people about them
(if any whitespace errors were detected)..
On Wed, Jul 22, 2009 at 01:05:27PM +0200, Bartlomiej Zolnierkiewicz wrote:
> On Tuesday 21 July 2009 12:01:14 Pavel Machek wrote:
>
> > Would it be possible to add autofixing to checkpatch? Fixing
> > whitespace issues by hand is useless & boring task...
>
> That's why most people use scripts for it.. ;)
>
> scripts/cleanfile
> scripts/cleanpatch
>
> Maybe we should make checkpatch inform people about them
> (if any whitespace errors were detected)..
That sounds like a good idea. Will have a look at doing that.
-apw
On Tue, 21 Jul 2009, Pavel Machek wrote:
> I don't think that a bit of trailing whitespace hurts that much (it is
> very easy to create when fixing overly long lines, another checkpatch
> complain), and at least akpm autofixes the whitespace, anyway, but...
The problem with trailing whitespace in files is that future developers
have to keep it the same while making changes in the area, or it'll add
noise to their commits and patches. Fixing up a patch to stop removing
trailing whitespace is even more of a pain than fixing a patch to stop
adding it.
> Would it be possible to add autofixing to checkpatch? Fixing
> whitespace issues by hand is useless & boring task...
Until checkpatch does it, "git apply" can be configured to auto-correct.
-Daniel
*This .sig left intentionally blank*
On Tue 2009-07-21 14:06:18, Andy Whitcroft wrote:
> On Wed, Jul 22, 2009 at 01:05:27PM +0200, Bartlomiej Zolnierkiewicz wrote:
> > On Tuesday 21 July 2009 12:01:14 Pavel Machek wrote:
> >
> > > Would it be possible to add autofixing to checkpatch? Fixing
> > > whitespace issues by hand is useless & boring task...
> >
> > That's why most people use scripts for it.. ;)
> >
> > scripts/cleanfile
> > scripts/cleanpatch
> >
> > Maybe we should make checkpatch inform people about them
> > (if any whitespace errors were detected)..
>
> That sounds like a good idea. Will have a look at doing that.
Yes, that would be great. Thanks!
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html