record sequence fields

David Wilson davidwwilson at attbi.com
Wed Jun 18 19:12:35 CEST 2003


>From: "Frank Ellermann" Frank.Ellermann at t-online.de
>To: "David Wilson" davidwwilson at attbi.com
>Cc: "EIS fans" seqfan at ext.jussieu.fr
>Subject: Re: record sequence fields
>Date: Wednesday, June 18, 2003 11:05 AM
>
>David Wilson wrote:
>
>> I am attempting to convert the OEIS to a MySQL database.
>
>That's an interesting idea. Changing records is probably
>simple, a reference to sequence X is only allowed if X exists.
>
>But additions and deletions can be difficult, because you need
>transactions for several records at once (e.g. cross-referenced
>pre-numbered sequences).

Right now, I am only concerned with getting a working database
incarnation of the existing sequences, and not with inserts or
crossrefs, I will deal with those if and when I decide to extend
the application.

>> Longest single element: A019439 (196 chars)
>How do you define "element" ? I tried "concatenation of %S,
>%T, and %U" for A19439, but that resulted in 212 characters.

Sorry, that should have been A019437

>> Longest name: A076740 (641 chars)

>Hm... :-( I'm still unhappy with too long %N lines.

If you could give me an idea of what is "too long" for various
fields, I could provide you with data about where the problems
are.

>> Longest comment line: A075886 (2148 chars)
>
>Oops, thanks for info. My default line length limit is 2048,
>because the DOS-version of my editor is restricted. But maybe
>I should change this to 4000 (the hard DOS-limit is 4095).

Hey, at least the database is no longer on punch cards.

> Bye, Frank
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.seqfan.eu/pipermail/seqfan/attachments/20030618/f373f601/attachment-0001.htm>


More information about the SeqFan mailing list