Re: [PATCH] New package: IPTraffic 0.8.2

[Matthias Fischer <matthias.fischer@ipfire.org>]

[-- Attachment #1: Type: text/plain, Size: 3993 bytes --]

[forgot the list, sent again]

Hi,

On 27.01.2021 12:32, Michael Tremer wrote:
> Hello Matthias,
> 
>> On 26 Jan 2021, at 16:45, Matthias Fischer <matthias.fischer(a)ipfire.org> wrote:
>> 
>> Hi,
>> 
>> On 25.01.2021 20:27, Michael Tremer wrote:
>>> Hello Matthias,
>>> 
>>> Thank you for submitting the patch.
>>> 
>>> It is great to see more people taking part in development tasks, but I am not really sure what has been done here.
>>> 
>>> The main problem is that I do not know what IPTraffic does, or how it works. The code is in a tarball and I am not aware if there is a Git repository to see what has been changed over time.

I'm working on this.

>> ...
>> I must confess I was puzzled after reading through all of it - its a pity.
> 
> I agree and I am very sorry for all the time you have invested into this with now very little result.

Yep. But thats life. I just take it as it is and try to make the best
out of it.

>> Perhaps I should have coded this for Pakfire in a different manner...
> ...
> No, I do not think that that was the thing that broke this.
> 
> As Bernhard has pointed out, the design of this add-on has some issues that would have to be ironed out and they sound to me like they are a lot of work. It might even be worth to start from scratch and get a much better design of this and only take the bits of the code that are acceptable right now.

I hope that this can be done - but I got no experience or enough
knowledge to rewrite this, so I hope we find somebody else.

>> As I see it, Bernhard has already looked through the code.
>> The only thing I can think of now: I could rewrite the build process -
>> if this makes still sense, let me know. If it doesn't fit our needs -
>> than thats it.
>> 
>>>... 
>>> You can use “git commit —-author=…” to set the correct author and you should sign-off as yourself as usual.
>> 
>> FYI:
>> This is exactly what I did in the *first* commit...
> ...
> Oh I didn’t see that. Very good :)

It was just a short search and again I learned something new about GIT.
I'm taking it positive...

>>> So to go back to the usual question: What is being proposed here and why?
>>> 
>>> Who is this add-on for? What are its features, and what are its limitations?
>>> 
>>> Why is this realised as add-on and not as part of the core system? I do not want to suggest that it should be either. It just seems that this decision has been made I would like to know based on what reasons :)
>> 
>> As I see it - it was once written as an addon and just stayed in this
>> condition. No one had the idea to integrate it. Simple.
> 
> We have a couple of those abandoned things on here, which is sad, but I suppose each of them has their own reasons.
> 
> It would be better if software is abandoned before it is being merged instead of after - because then it might cause us trouble later.

Yep. That happened in the past - I don't need it in the future.

>> Don't get me wrong - I'm not offended - just a little disappointed how
>> the whole thing has gone here at once and would definitely try to still
>> get the best out of it.
> 
> What do you suggest we should do right now?

As a start, I rewrote the whole building and installation process - I
got rid of the tarball. I would test if its ok  - one of the 'Devels' is
already working on it - and building as expected and then push it to GIT
/ Patchwork. This would make the current code readable and transparent
to everybody.

Suggestion:
Then "someone" (sorry, not me, thats far beyond my capabilities) should
be able to decide whether she/he is able to use and rewrite the existing
code to eliminate the disussed "shortcomings" (Google translate, I don't
know if this fits!) and if it can be integrated or publishd as an addon.
I can't decide if it would be better to start from scratch, see above
(Bernhards comments).

Best,
Matthias

[cut: unneeded installation code]