Wireshark-dev: Re: [Wireshark-dev] range_string checking

From: Jaap Keuter <jaap.keuter@xxxxxxxxx>
Date: Sat, 4 Apr 2020 15:53:08 +0200


On 2 Apr 2020, at 23:08, Martin Mathieson via Wireshark-dev <wireshark-dev@xxxxxxxxxxxxx> wrote:

It is common to have a 'catch-all' case for parts or all of the range, which is Ok if it comes after more specific entries.  I'm wondering if its worth complaining if *part* of an entry is hidden by an earlier one?  Current output from master is as below.  I will try to fix them up where I can access the relevant specs, but wanted to check my understanding of how they work and how fussy we should be?  I will most likely update README.dissector to make sure it is clear how it is evaluated in order.

Cool stuff. 
I can definitely see use for catch-all-in-certain-range, opposite of filling every gap with their specifics, which is maintenance heavy. This matches the val_to_string() default string used when no match is found, but then in a higher dimension. I would say let the ranges decide, if their union is the same as the catch-all then it’s okay, otherwise mark it.

just my €0.02
Jaap