I'm going to take that back, somewhat embarrassed 😳. My code contains the
following comment:
# Version: 2025-02-25 - allow for future format of RFC index (no leading zeros)
so it seems I handled this a long time ago. But anyone else who has code
that expects leading zeros, or depends on 4-digit RFC/BCP/STD numbers, needs
to fix it now.
Regards/Ngā mihi
Brian Carpenter
On 30-Oct-25 09:20, Brian E Carpenter wrote:
On 30-Oct-25 03:10, Robert Sparks wrote:
The current plan is to remove leading zeros in places like this.
Which places *exactly*?
There's a potential problem for code that "knows" that rfc-index.xml
(or even rfc-index.txt) has the leading zeros. Changing those files
to remove the leading zeros would break some of my code. I'm fairly
sure I was already assured that the file format will not change, and
I assumed that meant the leading zeros would remain within the metadata.
(Of course, I can program round it, but only if I know...)
The DOI system allows for registering aliases and specifying a canonical
DOI and we're planning to register that for the set that currently have
leading zeros.
That's a good idea and will simplify life.
This has been in the works for several years, and the leading-zero-less
variants of the names have been deployed for quite some time at places
like https://datatracker.ietf.org/doc/rfc20/
Absolutely, and that is the right thing to do.
Brian
RjS
On 10/29/25 3:10 AM, Mark Nottingham wrote:
To be clear - I’m not asking for any particular change, I’m asking for clarity.
Cheers,
Sent from my iPhone
On 29 Oct 2025, at 6:58 pm, Carsten Bormann <[email protected]> wrote:
On Oct 29, 2025, at 08:11, Brian E Carpenter <[email protected]>
wrote:
I don't see why they would change. It would be a nuisance, and it would be a bug
to change the DOIs. Code I've written has inserted/removed leading zeros for RFCs
<1000, but should work seamlessly when RFC10000 appears.
+1
I think the implementation would be: "RFC%04d”
So we do RFCnnnn for N < 10k, and the necessary number of digits for N ≥ 10k.
This should work throughout the system for the stable identifiers.
We are referencing older RFCs in the text with anchors like “RFC20” inside
newer I-Ds and RFCs, but that stays internal to that document; the DOI remains
10.17487/RFC0020 .
Grüße, Carsten
_______________________________________________
rfc-interest mailing list -- [email protected]
To unsubscribe send an email to [email protected]
_______________________________________________
rfc-interest mailing list -- [email protected]
To unsubscribe send an email to [email protected]
_______________________________________________
rfc-interest mailing list -- [email protected]
To unsubscribe send an email to [email protected]
