coreutils
[Top][All Lists]
Advanced

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

Re: How to implement the V comparsion used by sort in python?


From: Assaf Gordon
Subject: Re: How to implement the V comparsion used by sort in python?
Date: Mon, 28 Oct 2019 13:04:35 -0600
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.9.0

Hello,

On 2019-10-28 3:00 a.m., Florian Weimer wrote:
* Assaf Gordon:
On Oct 26, 2019, at 5:05 PM, Peng Yu <address@hidden> wrote:

Are you sure they are 100% compatible with V? I don’t want to use
them just later find they are not 100% compatible.

There are no such guarantees, especially not with free software.

I don't know why you say that.

Perhaps my writing wasn't clear enough.

What I meant was: *I* can not provide any such guarantees
(since the question was "are *you* sure").

I can't speak for other coreutils maintainers (or the people who
wrote the gnulib version-compare module), but I highly suspect
that they will also not be willing to guarantee such %100 compatibility.

As for the "free software" part - (almost?) every free software license
explicitly mentions that the software comes with no warranty what so
ever. Typically the license include the phrase "[no] FITNESS FOR A PARTICULAR PURPOSE" - meaning that even there is some implied purpose
(such as sorting 'naturally' for "sort -V"), there is no guarantee
it is even fit for that purpose.

In practice, it means that even if I (or others) took a cursory look
at both "sort -V" and the mentioned python package and deemed them "compatible", there is still *no* guarantees they are actually 100%
compatible. There could always be a bug or an unexpected result.

It seemed to me the OP wanted some very strong guarantees regarding
that code that would save him time and effort, without investing time
or other resources to do the testing themselves.
To that, my answer was "no such guarantees".

If my previous answer was too brief, I hope this clarifies it.

But someone certainly has to do this work.

I completely agree.

If the OP wants reasonable assurance they are compatibly,
they can read the details about "sort -V" and invest the time
and effort in comparing it to the python package algorithm.

Or for stronger guarantees - perhaps they can consider hiring someone
to do a very thorough investigation and provide them with some
concrete guarantees.

regards,
 - assaf



reply via email to

[Prev in Thread] Current Thread [Next in Thread]