[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: [MoM] FAST progress (was Re: [MoM] fast: Add further dependencies to enable chroot / cowbuilder to build)



Hello Andreas,

Just a reminder whenever you get the chance to check the commits and the
test suite as I know you're busy.

Also, is it worth updating the other packages (upstream code for
mmseqs2) now or is it better to wait until the package has finished
pending upload and is in unstable. Additionally, how long does it
usually take before a package is put into unstable from being NEW? I've
seen some packages as NEW for over a year although maybe these have been
rejected?

Kind regards,
Shayan Doust

On 02/09/2019 00:57, Shayan Doust wrote:
> Hello Andreas,
> 
>> Both sounds sensible.  I try to check your commits soon.
> 
> Thanks.
> 
> I also see no reason why the test suite will not work now. As far as the
> binary executes for me (until the opencl limitation), I see no asset-
> related assertion or error. Curious to see if we get a success with testing.
> 
> Kind regards,
> Shayan Doust
> 
> On 01/09/2019 22:12, Andreas Tille wrote:
>> On Sat, Aug 31, 2019 at 10:30:33PM +0100, Shayan Doust wrote:
>>> I have used /usr/share/fast instead, I didn't think of this.
>>>
>>>> Those binaries that have a purpose as user utilities should probably
>>>> go to /usr/bin and adding a manpage (perhaps by using
>>>> createmanpages[1]).
>>>
>>> A bit more investigation. The issue with this is that it is generally
>>> not user friendly (maybe proof of concept). These binaries also expect
>>> that you have the FAST_Test_Data.xz extracted in the same directory as
>>> them or using your own data set. In this case, it probably won't be wise
>>> to have it in /usr/bin as there is no interactivity. Maybe your initial
>>> suggestion of having these in /usr/lib/fast and having a documentation
>>> explaining these under /usr/share/doc/libfast-examples/examples would be
>>> better. Sorry if I have not been clear to begin with, making more and
>>> more discoveries and understanding with a fairly big project.
>>
>> Both sounds sensible.  I try to check your commits soon.
>>
>> Kind regards
>>
>>       Andreas.
>>
> 

Attachment: signature.asc
Description: OpenPGP digital signature


Reply to: