[CWB] A formal specification of CQL?

[Hardie, Andrew]

Re: CWB 2 and 3 – Yes, that’s exactly what it means. CWB 3.9 will only be able to work with CWB 3.9 indexes.

My hope is to add a conversion tool to 3.9, however, if it is not too complex to do so. (Shouldn’t be – all it will involve is piping cwb-decode3.5 into cwb-encode3.9).

As for NoSkE – this is the first I’ve heard that you even could use corpora compiled by Manatee with CWB! So I really have no clue there.

In general, everyone should expect 3.9 to break more or less everything.

There will be a stable branch 3.5 which we will maintain for some years, if you don’t want to upgrade to 3.9. I expect it will be quite a while until the bugs are shaken out of 3.9 enough that we become confident enough to release it as 4.0!

This will be a massive pain. However, the v2/3 binary format is now one of the main obstacles to us improving CWB in the ways that Stefan & I have identified as priorities. So the pain is necessary in the long run….

Andrew.

From: cwb-bounces at sslmit.unibo.it [mailto:cwb-bounces at sslmit.unibo.it] On Behalf Of Yannick Versley
Sent: 30 January 2015 20:00
To: Open source development of the Corpus WorkBench
Subject: Re: [CWB] A formal specification of CQL?

Does this mean that the binary corpus files made with CWB 2.x through 3.2 (and possibly
any corpora compiled with the NoSke tools) will stop working with CQP 3.9/4?

Best wishes,
Yannick

On Fri, Jan 30, 2015 at 6:34 PM, Hardie, Andrew <a.hardie at lancaster.ac.uk<mailto:a.hardie at lancaster.ac.uk>> wrote:
I see!  That ticket is now here (due to sourceforge changing the tracker software):

http://sourceforge.net/p/cwb/feature-requests/46/

Actually, I'm going to close it as WONTFIX: there seems little point in spending a lot of time writing about file formats that we are going to chuck  out in versions 3.9 and 4, and I will document the new file formats as I'm in the process of inventing them...

Andrew.

PS thanks for that parser state diagram. It confirms what I always suspected, to be frank.


-----Original Message-----
From: cwb-bounces at sslmit.unibo.it<mailto:cwb-bounces at sslmit.unibo.it> [mailto:cwb-bounces at sslmit.unibo.it<mailto:cwb-bounces at sslmit.unibo.it>] On Behalf Of Serge Heiden
Sent: 30 January 2015 15:25
To: cwb at sslmit.unibo.it<mailto:cwb at sslmit.unibo.it>
Subject: Re: [CWB] A formal specification of CQL?

Thank you Andrew, I found it: it was the content of a ticket:
http://devel.sslmit.unibo.it/pipermail/cwb/2012-July/001042.html

Best,
Serge

Le 30/01/2015 16:05, Hardie, Andrew a écrit :
> We generally call it "CQP Syntax" rather than "Corpus Query Language", although lots of people working on other software have gravitated to the latter.
>
> The code of the parser is written as a Bison grammar, which *is* thus in effect a formal specification!
>
> It's in cqp/parser.y (see also parser.l).
>
> See:
>
> http://sourceforge.net/p/cwb/code/HEAD/tree/cwb/trunk/cqp/parser.y
>
> And to search the list, use
>
> site:devel.sslmit.unibo.it/pipermail/cwb<http://devel.sslmit.unibo.it/pipermail/cwb>
>
> in a Google search.
>
> best
>
> Andrew.
>
>
> -----Original Message-----
> From: cwb-bounces at sslmit.unibo.it<mailto:cwb-bounces at sslmit.unibo.it> [mailto:cwb-bounces at sslmit.unibo.it<mailto:cwb-bounces at sslmit.unibo.it>]
> On Behalf Of Serge Heiden
> Sent: 30 January 2015 14:46
> To: cwb at sslmit.unibo.it<mailto:cwb at sslmit.unibo.it>
> Subject: Re: [CWB] A formal specification of CQL?
>
> Not to my knowledge.
>
> There's been a thread on a similar topic before.
> Is there any search engine available to search in all CWB mailing list
> archive at once? (I can't remember when it was).
>
> For fun, you can have a look at the network of the CQP parser states
> here:
> https://groupes.renater.fr/wiki/txm-info/_media/cqpsyntax.dot.jpeg
>
> Serge H.
>
> Le 30/01/2015 14:36, Jörg Knappen a écrit :
>> Is there a formal specification of the Corpus Query Language, e.g.,
>> in EBNF format, available?
>>
>> --Jörg Knappen
>>
>> _______________________________________________
>> CWB mailing list
>> CWB at sslmit.unibo.it<mailto:CWB at sslmit.unibo.it>
>> http://devel.sslmit.unibo.it/mailman/listinfo/cwb

--
Dr. Serge Heiden, slh at ens-lyon.fr<mailto:slh at ens-lyon.fr>, http://textometrie.ens-lyon.fr ENS de Lyon/CNRS - ICAR UMR5191, Institut de Linguistique Française 15, parvis René Descartes 69342 Lyon BP7000 Cedex, tél. +33(0)622003883<tel:%2B33%280%29622003883>

_______________________________________________
CWB mailing list
CWB at sslmit.unibo.it<mailto:CWB at sslmit.unibo.it>
http://devel.sslmit.unibo.it/mailman/listinfo/cwb
_______________________________________________
CWB mailing list
CWB at sslmit.unibo.it<mailto:CWB at sslmit.unibo.it>
http://devel.sslmit.unibo.it/mailman/listinfo/cwb

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://devel.sslmit.unibo.it/pipermail/cwb/attachments/20150130/776af9bb/attachment-0001.html>