[Nut-upsdev] Re: Another HID USB UPS

[arnaud.quette at mgeups.com]

> I have a new HID USB device, that seems to be quite similar to the
> currently supported APC HID UPS. I append the dump-tree output, and would
> much appreciate advice on how to proceed.
>
> Certain entries for the current device are not available, but maybe there
> are some replacements that could be used. This may of course require a new
> table in apc-hid.h, and I would prefer not to do this without some guidance.

FYI:
- apc-hid.h still need to be completed. John Stamp (cc'ed)
has began to work on this. Mostly, the shutdown procedure,
which varies between APC models, has to be completed...
Check with John for more info...
- this table lists all available data for all the devices of a
manufacturer,
- At launch time, newhidups check the data available
from the current device. If there are data redundancies
(ie multiple definition for a same nut variable), then the
first available is used. I put the better first (ie if there is
one RW var for high end UPS, and another RO for low end,
put the RW first, as for "battery.charge.low" in mge-hid.h)
- look at mge-hid.h for more examples and ask questions
as I've not yet made the document to extend newhidups...

> Also, it looks to me like the driver did actually start its standard loop,
> but I thought it best to first sort out the appropriate tables.

sure. what upsc gives you for the moment?

> P.S. for Arnaud, I had the option to use a different device, and so I did!
     > I will therefore unfortunately not be able to provide you with the
     > TrippLite information that you requested. I suspect we didn't loose
     > much!

no problem for me at all. although I'm the NUT project leader,
I'm paid and employed by MGE for that task, not APC nor Tripplite...
But you should have prefered an MGE to support NUT ;-)

> ...

the task is:
- to check is the below path are already in apc-hid.h
- to add it otherwise (check mge-hid for example)
and link it to the right nut var.
- the 000000 suffixed path are due to the use of the MGE
HID Parser. These are ignored, except for the dump
(I'll modify it not to pollute...)

Once more, coordinate with John to get information and
complete all that...

Arnaud Quette
---
Linux / Unix Expert - MGE UPS SYSTEMS - R&D Dpt
OpenSource Developer - http://arnaud.quette.free.fr/
Debian Developer - http://people.debian.org/~aquette/
... and much more ...