scidb-users Mailing List for Scidb

Hi,

 

Using:

clang version 5.0.1-svn316607-1~exp1 (branches/release_50)

 

Full build log here: https://pastebin.com/8MgQvZnP

 

Relevant part:

Compiling db_move_list.cpp                [-Wall -g -I/usr/include/tcl8.6
-DSCI_NAMEBASE_FIX=1]

db_move_list.cpp:277:16: error: explicit instantiation of 'db::MoveBuffer'
must occur in namespace 'db'

template class MoveBuffer<position::Maximum_Moves>;

               ^

./db_move_list.h:38:7: note: explicit instantiation refers here

class MoveBuffer

      ^

db_move_list.cpp:278:16: error: explicit instantiation of 'db::MoveBuffer'
must occur in namespace 'db'

template class MoveBuffer<opening::Max_Line_Length>;

               ^

./db_move_list.h:38:7: note: explicit instantiation refers here

class MoveBuffer

      ^

2 errors generated.

Makefile:161: recipe for target 'db_move_list.o' failed

make[2]: *** [db_move_list.o] Error 1

Makefile:205: recipe for target 'recursive' failed

make[1]: *** [recursive] Error 1

Makefile:18: recipe for target 'all' failed

make: *** [all] Error 2

 

Of course, clang is probably not supported, but I thought it interesting to
report it here anyway.

 

Best,

Bogdan

> CQL++ ;-)

I agree, this sounds like a very good alternative, so it will be named CQL++.

> If eCQL will be a functional superset of CQL, how much work would be 
> involved in placing a CQL compatibility layer on top of eCQl to 
> translate the syntax?

I'm a specialist in compiler development and formal languages, and I don't think that the effort is high. But this will not be done before CQL++ is finished.

> I'm anticipating that some users will have a rather large library of CQL 
> syntax which they will not care to translate by hand.

I'm always kept in mind that CQL scripts (v3.1 and v.5.1) should be runnable under Scidb.

> I'm wondering how much of a gap there will be between the two syntaxes.  I 
> believe that you mentioned earlier that the current documentation is not 
> representative of the final syntax.

Although it's a preliminary draft, the current documentation is quite representative. The difference between the two syntaxes is enormous, but it is convertible in both directions - considering that CQL++ is a superset.

Gregor

Dear Scidb mailing list,
 
I was going to ask if Scidb could be run on ChromeOS, but now I remember that even though it's based on Linux it's still kind of like a special type of OS. So, that leaves me with a question. Will Scidb ever be release as an app to be downloaded from the Play store. I'm thinking kind of like a limited version of Scidb where it's main features, look-n-feel, and general appearance would be included, but not the entire full blown application.  So far the best thing I've been able to find on the Play store seems to be Perfect Chess Database so making Scidb available on the Play would be a really nice thing.
 
Thanks.

Mit TravelXtra profitieren Sie von 5% Rückvergütung auf den Reisepreis  – bekannte dt. Reiseanbieter und ein umfangreiches Reiseangebot wie im Reisebüro!
https://email.freenet.de/reisen/index.html

On 11/01/2017 04:00 AM, Gregor Cramer wrote:
>
> BTW: Also CQL v5.1 is a subset of eCQL (ChessQuery).
>

If eCQL will be a functional superset of CQL, how much work would be 
involved in placing a CQL compatibility layer on top of eCQl to 
translate the syntax?

I'm anticipating that some users will have a rather large library of CQL 
syntax which they will not care to translate by hand.

To be clear, I'm not suggesting that *you* implement that layer.  I'm 
wondering how much of a gap there will be between the two syntaxes.  I 
believe that you mentioned earlier that the current documentation is not 
representative of the final syntax.


-- Lionel

>> Thanks for your suggestion, I agree that SCQL sounds good, but I'm
>> not sure that I want a name which contains the project name Scidb,
>> because the design of eCQL (ChessQuery) is independent from Scidb's
>> internals. Anyone else who has suggestions/proposals for the naming
>> of the query language?
>>
>
> CQL++  ;-)


Yes to CQL++.


-- Lionel

Am Wed, 01 Nov 2017 11:00:28 +0100
schrieb Gregor Cramer <re...@gm...>:

> > >> CQL is a nice play of words on the well known  SQL (Structured
> > >> Query Language) and i like keeping it. But maybe renaming it
> > >> will help differentiate Gregor's implementation from the
> > >> original ?  
> > > 
> > > The point is that Scidb's design will be quite different from
> > > current CQL 5.0, so I think that a different name like ChessQuery
> > > is required. It's a bit  unluckily that the term CQL is already
> > > taken. And the term eCQL - used in my current documentation of
> > > ChessQuery - is not very satisfactory for me.  
> > 
> > I like "SCQL'.... Scidb CQL.  
> 
> Thanks for your suggestion, I agree that SCQL sounds good, but I'm
> not sure that I want a name which contains the project name Scidb,
> because the design of eCQL (ChessQuery) is independent from Scidb's
> internals. Anyone else who has suggestions/proposals for the naming
> of the query language?
> 

CQL++  ;-)

> >> CQL is a nice play of words on the well known  SQL (Structured Query
> >> Language) and i like keeping it. But maybe renaming it will help
> >> differentiate Gregor's implementation from the original ?
> > 
> > The point is that Scidb's design will be quite different from current CQL
> > 5.0, so I think that a different name like ChessQuery is required. It's a
> > bit  unluckily that the term CQL is already taken. And the term eCQL -
> > used in my current documentation of ChessQuery - is not very satisfactory
> > for me.
> 
> I like "SCQL'.... Scidb CQL.

Thanks for your suggestion, I agree that SCQL sounds good, but I'm not sure 
that I want a name which contains the project name Scidb, because the design 
of eCQL (ChessQuery) is independent from Scidb's internals. Anyone else who 
has suggestions/proposals for the naming of the query language?

BTW: Also CQL v5.1 is a subset of eCQL (ChessQuery).

Gregor

On 10/27/2017 05:43 AM, Thomas Porschberg wrote:
>
> thank you. Where do I come from?
> How I told you, currently I use chess assistant with CQL.
> Now a friend with OSX asked me if I know a chess database with CQL capability for OSX.
> I don't know...
> To speak frankly: I only know this CQL-version from chess assistant and this standalone version
> for windows. I think chessbase did not come with a CQL-interface.
>
> Do you know another chess database which has this feature (may be even running on OSX)?

Hello Thomas,

CQL v5.1 is now up and running as an integrated search engine with Scid 
vs PC.  It is fully functional for all CQL syntax which is not 
non-sensical in an Scid context, and looks to be quite solid.  You can 
tell your friend that a patch for the svn source tree should be 
available toward the end of November (there are still a couple dozen 
minor porting issues which need attention... call it finishing polish).

One caveat is that, for the foreseeable future, your friend will have to 
compile and build his own executable (or maybe you could build it for him).

Best regards,

-- Lionel