On Sat, 20 July 2013 13:35:08 +0200, Stefan Gehn wrote:
>
> - VARIANTS could be replaced by an enum for the four variants. That way
> setup of the arrays in setup_sysfs_battery would be a bit more
> descriptive. The hardcoded indices inside the code and the VARIANTS
> define could get out of sync easily if somebody has to touch that code
> again. I thought about something along the lines of:
>
> typedef enum SysChargeVariants_ { SCVEnergy, SCVEnergyDesign, SCVCharge,
> SCVChargeDesign, SCVCapacity, SCVCount } SysChargeVariants;
Would be fine with me, with one caveat: I really don't believe anyone
will touch that code again. Gkrellm is effectively abandonware. How
else could a patch like mine that - code issues aside - fixes an
easily reproducable bug linger for 15 months without comment? ;)
> - The patch drops all debug output in setup_sysfs_battery(). I think the
> output should be ported to at least log the stored paths for the
> multiple variants.
That is actually a good thing. Debug output is random noise in the
source code and makes it harder to read/understand/change. That
downside always exists. In some cases debug output also has upsides
that sufficiently counter the downsides. I didn't need the debug
output at all and the gkrellm maintainer... *cough*
Btw, how much inconvenience does it involve to become the gkrellm
maintainer for Debian? All I can promise is to fix any bugs that
happen to annoy me personally and occasionally merge patches from
others, provided the code is reasonable. If you can do a better job
than that, go Stefan go! This is just a fallback offer.
Jörn
--
Linux [...] existed just for discussion between people who wanted
to show off how geeky they were.
-- Rob Enderle
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
