[Debian-science-sagemath] status page: legend: Waiting in NEW: a more distinctive colour ?

[Ximin Luo]

Jerome BENOIT:
> On 20/08/16 22:42, Ximin Luo wrote:
>> Jerome BENOIT:
>>> On 20/08/16 20:23, Ximin Luo wrote:
>>>> if we add ITP/ITA/RFS later, then we'll need even more colours.
>>>
>>> No more colours, but jut in place of PTS
>>>
> 
>> Oh so I have an idea, put NEW into the PTS column and only have one shade of yellow. We can reuse the existing colours like this:
> 
>> ITP/ITA light red
>> RFS orange
>> NEW yellow
>> PTS no colour
> 
>> the downside is the version would be hidden whilst it's in NEW. But I can set it as a title="" tooltip.
> 
>> I can also move the PTS column after next to the Debian column since they're more related.
> 
>> What do you think?
> 
> Keep things simple and readable for the rest of the world.
> 
> Let keep the PTS which points to a full charts of useful data when it exists;
> when the PTS webpage does not yet exist, put as substitute ITP xor NEW when possible.
> ITA information is given in PTS, and it will not hurt to discard RFS (which is rather
> a maintainer issue (when [s]he is not DM or DD) that a package one).
> Within this scheme, a light red for the state ITP/NEW instead of grey.
> 

I tried to implement this, but it was harder than I expected, and I really wanted to do other things today so I'm going to leave this for now. 

We didn't think about the following case:

1. A package is already in Debian in unstable/experimental at version X, so PTS already exists.
2. The package is also in NEW at version Y (because the new version brings in new binary packages).

The "obvious behaviour" to display this is to show (1) in the Debian column and (2) in the PTS column.

However, the existing logic in dss.py makes it very hard to do this - if there is a version in NEW that is higher than unstable/experimental, it will discard the info for the latter. So to implement "the obvious behaviour" we'd need to:

(a) change get_status_and_debian_version() to not discard this data
(b) refactor a lot of logic in print_tr() to handle this data

and I just can not be bothered to do that right now, given that this is just a status page and doesn't directly improve sagemath packaging. Sorry :(

X

-- 
GPG: ed25519/56034877E1F87C35
GPG: rsa4096/1318EFAC5FBBDBCE
https://github.com/infinity0/pubkeys.git