chessdb-users Mailing List for ChessDB - a Free Chess Database (Page 13)

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

The earlier submission was a patch, not the actual file.
So is this; apply with:
  patch -p0 < cvsignore.fix.diff

Cheers,

Jeremy

>> Why not specify a directory and a start issue number, then 
>> periodically determine the latest issue in the directory and download 
>> the "next" one, repeating until you get a 404 error? If you want to 
>> clean out the directory, then you can set the start issue apporpriately.
> 
> One possibility, although I'm not sure if its a good idea to keep 
> accessing the web to find them.
> 
> First it would cause an error message if an Internet connection is not 
> available. Secondly TWIC might get rather fed up with it. If there are 
> lots of copies, all hitting their server all day, they might get 
> annoyed.  .

The level of automation should surely be left up to the user. Users will 
have a sense of whether they want to avoid errors by doing it manually 
or let the application do it for them. Having the games automatically 
arrive is definitely a convenience that a lot of people would take 
advantage of.

TWIC will care about total bandwidth used. More on this below...

> (Incedentely, the web server logs at TWIC will show something like:
> 
> 
> 192.168.0.10 - - [05/Jan/2007:19:08:26 +0000] "GET 
> /twic/zips/twic633g.zip HTTP/1.0" 200 169814 "-" "ChessDB 
> http://chessdb.sourceforge.net/"
> 
> Some people writing things like this that download from web sites in a 
> non-interact way, often fake the 'user agent' to make it appear to be 
> Internet Explorer or similar. I decided to set the user agent to be the 
> program's name. It does mean TWIC could stop the downloads if they so 
> chose by filtering on this, although if its not abused, they would have 
> no reason to do it. (Of course, one could easily change the user agent 
> in the source code, but I'd prefer not to do that).

I definitely agree with your upfront policy here. We have to be kind to 
twic. A possibility we should consider is to contact twic and offer to 
set up a mirror. I work at a hosting company, so I could help with this 
if necessary. You could use sourceforge for this also.

>> Once you download them are you importing to a chessdb database? 
> 
> Yes. it creates  a database called 'newgames'. The basic code does this:
> 
> 1) Download compressed zip file of pgn format games.
> 2) extract the pgn file from the zip file.
> 3) remove zip file
> 4) append the pgn onto a larger pgn.
> 5) Remove the pgn just downloaded.
> 
> that lot goes in a loop.

Two thoughts come to mind. First, we should make chessdb read zip files 
directly. I think it can already read .gz files, IIRC. Doing this would 
simplify the procedure and save disk space.

Second, why append them? It seems like the most efficient process would 
download a file and import it into the database directly. This comes 
down to reading zips as above.

> Once its all done, a database is created.
> Finally the large pgn file is removed, leaving only the database.
> 
> I guess it may need something on the options menu to let the user 
> configure what happens.
> 
> 
>> It would be very cool to be able to apply a filter criteria to this 
>> import, and to have such criteria be allowed to be permanently 
>> associated with the database, so that the game feed flows into 
>> multiple databases with different purposes. This is one reason to keep 
>> the downloaded files in raw form.
> 
> They are large - if you download all of them it goes into 100's of MB. 
> And once you have them in a database, is there a lot of point keeping 
> the individual pgn files?

To me, the cost of drive space is cheap enough that I don't worry about 
things that are smaller than 1G. Perhaps others worry about a few 100MB, 
I can't speak for them. I have a directory with all the twic zipped pgn 
files currently available, they add up to about 600MB uncompressed. The 
scid database put them in is about 75MB total.. The latest file, 
twic634g.zip is 390K, which is one datapoint to say what the weekly 
download volume would be. My PC has an 80GB drive that's about 25% 
utilized. To me, I'd rather just save the files so that I never have to 
redownload the same file twice. I use more tools than just chessdb, as I 
suspect many do.

What I really want is a way to have lots of databases, each with a 
"theme" and have IM/GM games matching the theme flow in with minimal 
work for me. Perhaps another approach would be to automate the process 
of applying a filter to an external database and periodically running it 
and adding the new hits into the first database.

This might be better anyway, because it would be more general than a 
twic specific data feed.

>> It would also be cool to do duplicate detection during an import. 
> 
> Don't you think that is better for someone to do manually? That often 
> needs human interaction.

It's not an either/or choice: I want both. I want the easy dup's to be 
caught automatically. Harder ones can be done manually. When I'm doing 
an import, the context gives me addition information that helps me. For 
example, if I'm adding games from the same source (e.g. TWIC), then 
duplicate games happen when either A) twic failed to spot the dup or B) 
I reload the same source game twice. If I can have automation prevent B 
and catch really blatent cases of A, I'm happy.

Somebody mentioned wanting to do deletion in the best games window. I 
definitely think this is a good idea. That's where I usually catch dups 
that snuck through. I'd also like to be able to very quickly correct the 
ELO of a player. There's nothing more annoying that getting some weird 
name you've never heard of there with a 4000 rating.

>> Which reminds me: the duplicate detection using game moves doesn't 
>> seem to work well: it sometimes fails to detect that two games have 
>> the same moves, maybe because of move evaluation symbols? I can dig up 
>> an example of this if needed.
> 
> That would be interesting to see an example, although I imagine tracing 
> the bug might be hard.

Next time I see one, I'll confirm that this happens in chessdb (I've 
only observed it in scid) and try to simplify it to a minimal test case.

Bryan Taylor,  sobota, 6 stycznia 2007 02:11:
>Dr. David Kirkby wrote:
>> The default is to download the 3 most recent TWIC games collection.
>> a) Don't ask for user intervention - just download the latest collection
>> to a file. But I'm not too keen on so much automation.
>>
>> b) Let the user chose which issues, but store the number of those
>> already download (to options file?), so they don't repeat them.
>>
>> c) Make periodic checks to see if later collections are online, and if
>> so do WHAT ??
What about putting last downloaded TWIC in config file? You can ask user when 
he uses the feature for the first time.
-- 
Michal Rudolf

Bryan Taylor wrote:
> Dr. David Kirkby wrote:
>
> Why not specify a directory and a start issue number, then periodically 
> determine the latest issue in the directory and download the "next" one, 
> repeating until you get a 404 error? If you want to clean out the 
> directory, then you can set the start issue apporpriately.

One possibility, although I'm not sure if its a good idea to keep 
accessing the web to find them.

First it would cause an error message if an Internet connection is not 
available. Secondly TWIC might get rather fed up with it. If there are 
lots of copies, all hitting their server all day, they might get 
annoyed.  .

(Incedentely, the web server logs at TWIC will show something like:

192.168.0.10 - - [05/Jan/2007:19:08:26 +0000] "GET 
/twic/zips/twic633g.zip HTTP/1.0" 200 169814 "-" "ChessDB 
http://chessdb.sourceforge.net/"

Some people writing things like this that download from web sites in a 
non-interact way, often fake the 'user agent' to make it appear to be 
Internet Explorer or similar. I decided to set the user agent to be the 
program's name. It does mean TWIC could stop the downloads if they so 
chose by filtering on this, although if its not abused, they would have 
no reason to do it. (Of course, one could easily change the user agent 
in the source code, but I'd prefer not to do that).

> Once you download them are you importing to a chessdb database? 

Yes. it creates  a database called 'newgames'. The basic code does this:

1) Download compressed zip file of pgn format games.
2) extract the pgn file from the zip file.
3) remove zip file
4) append the pgn onto a larger pgn.
5) Remove the pgn just downloaded.

that lot goes in a loop.

Once its all done, a database is created.
Finally the large pgn file is removed, leaving only the database.

I guess it may need something on the options menu to let the user 
configure what happens.

> It would 
> be very cool to be able to apply a filter criteria to this import, and 
> to have such criteria be allowed to be permanently associated with the 
> database, so that the game feed flows into multiple databases with 
> different purposes. This is one reason to keep the downloaded files in 
> raw form.

They are large - if you download all of them it goes into 100's of MB. 
And once you have them in a database, is there a lot of point keeping 
the individual pgn files?

> It would also be cool to do duplicate detection during an import. 

Don't you think that is better for someone to do manually? That often 
needs human interaction.

> Which 
> reminds me: the duplicate detection using game moves doesn't seem to 
> work well: it sometimes fails to detect that two games have the same 
> moves, maybe because of move evaluation symbols? I can dig up an example 
> of this if needed.

That would be interesting to see an example, although I imagine tracing 
the bug might be hard.

Denis J Navas wrote:
> 
> ----- Original Message ----- From: "Dr. David Kirkby" 
> <dav...@on...>
> To: "Jeremy White" <jw...@co...>
> Cc: "ChessDB Users" <che...@li...>
> Sent: Tuesday, January 02, 2007 10:46 PM
> Subject: Re: [Chessdb-users] [Patch] Refinements to the blunder analysis 
> mode
> 
> 
> 
>> I was going to look at the code at one point with a view to putting the
>> name of the analysis engine, since that information might be useful if
>> you review the game at a later date. I know that information is
>> displayed in scid/chessdb, but I don't know at what bits of the code it
>> is available.
>>
>> -- 
>> Dr. David Kirkby
>>
> 
> 
> Dr. Kirkby:
> 
> Months ago I searched the same problem and reached the following 
> solution (I am not a programmer and do not have the software that 
> produces the diff file):
> 

Thanks for that - I will investigate where the engine name is placed and 
how best to get it into the annotations. It should not be too hard.

-- 
Dr. David Kirkby

Dr. David Kirkby wrote:
> I've made a new release (ChessDB-3.6.11-alpha) which has two significant 
> changes from the previous release.

IMPORTANT - I forgot to say, the latest release is in a new 'package' 
called 'chessdb-unstable-for-testing'

(the following direct URL might work - depends if this is a dynamic URL 
created only for me by sourceforge, or it its a permanent one)

http://sourceforge.net/project/showfiles.php?group_id=184864&package_id=217075

The new code ( 	ChessDB-3.6.11-alpha.tar.gz) is *NOT* in the main 
chessdb "package" (Sourceoforge's term). There are in fact now 3 
packages on the Sourceforge site:

1) chessdb - where the main files are.
2) extras - currently only tablebases, but I might add some games in 
databases.
3) chessdb-unstable-for-testing - pretty obvious that one.

I've put the new code in an 'unstable' package as I am aware there are 
bugs in this. There is nothing that should corrupt any databases, as 
none of that code is touched, but the new code connects to the Internet 
for the first time, so there are quite a few major changes.

I would appreciate feedback on this. Currently there is only the source 
files - no setup.exe.

Dr. David Kirkby wrote:

> The default is to download the 3 most recent TWIC games collection.
> 
> Other possibilities include
> 
> a) Don't ask for user intervention - just download the latest collection 
> to a file. But I'm not too keen on so much automation.
> 
> b) Let the user chose which issues, but store the number of those 
> already download (to options file?), so they don't repeat them.
> 
> c) Make periodic checks to see if later collections are online, and if 
> so do WHAT ??

Why not specify a directory and a start issue number, then periodically 
determine the latest issue in the directory and download the "next" one, 
repeating until you get a 404 error? If you want to clean out the 
directory, then you can set the start issue apporpriately.

Once you download them are you importing to a chessdb database? It would 
be very cool to be able to apply a filter criteria to this import, and 
to have such criteria be allowed to be permanently associated with the 
database, so that the game feed flows into multiple databases with 
different purposes. This is one reason to keep the downloaded files in 
raw form.

It would also be cool to do duplicate detection during an import. Which 
reminds me: the duplicate detection using game moves doesn't seem to 
work well: it sometimes fails to detect that two games have the same 
moves, maybe because of move evaluation symbols? I can dig up an example 
of this if needed.

----- Original Message ----- 
From: "Dr. David Kirkby" <dav...@on...>
To: "Jeremy White" <jw...@co...>
Cc: "ChessDB Users" <che...@li...>
Sent: Tuesday, January 02, 2007 10:46 PM
Subject: Re: [Chessdb-users] [Patch] Refinements to the blunder analysis 
mode

> I was going to look at the code at one point with a view to putting the
> name of the analysis engine, since that information might be useful if
> you review the game at a later date. I know that information is
> displayed in scid/chessdb, but I don't know at what bits of the code it
> is available.
>
> -- 
> Dr. David Kirkby
>

Dr. Kirkby:

Months ago I searched the same problem and reached the following solution (I 
am not a programmer and do not have the software that produces the diff 
file):

====================================================================
Changint the way SCID annotate the games from "depth:value" to "value/depth 
EngineName"

The changes must be done at various places in the code.  This description is 
on the code inside the file
named "scid.gui", Scid version 3.6.1.   To insert the engine name, it is 
necesary a new variable:  $analysisName

But, when I inserted this in Scid 3.6.2, discovered that the new variable 
must be declared in a different instruction,
but my machine crashed after, and don't have those changes.

Original code:

----at line 22970
proc addAnalysisVariation {{n 1}} {
  global analysis annotateMoves annotateMode analysisName

follows the following code (original from Scid 3.6.1):

---A)  line 22943

  set text [format "%d:%+.2f" $analysis(depth$n) $analysis(score$n)]

---B)  line 22986:
  # If line is true, add the whole line, else just add the score:
  if {$line} {
    set scoretext [format "%+.2f: %s" $score $analysis(moves$n)]
  } else {
    set scoretext [format "%+.2f" $score]
  }

---C)  line 22439:  Engine priority

  set analysis(priority$n) normal     ;# CPU priority: idle/normal

---D)  Línea 22433:  Tiempo de análisis en comentario de movimientos

  set analysis(automoveTime$n) 4000

-------------------------------------------------------------------------
Changed to:
-------------------------------------------------------------------------

----22970
proc addAnalysisVariation {{n 1}} {
  global analysis annotateMoves annotateMode analysisName

A) --- 22943 New text format, with the new variable incorporated.

  set text [format "%+.2f/%d %s" $analysis(score$n) $analysis(depth$n) 
$analysisName]

B) ---22986 Add the lines with the changed format

  # If line is true, add the whole line, else just add the score:
  if {$line} {
    set scoretext [format "%+.2f/%d: %s %s" $score $analysis(depth$n) 
$analysisName $analysis(moves$n)]
  } else {
    set scoretext [format "%+.2f/%d %s" $score $analysis(depth$n) 
$analysisName]
  }

C) --- Change the engine priority

  set analysis(priority$n) idle       ;# CPU priority: idle/normal

D) --- Increase default analysis time

  set analysis(automoveTime$n) 15000

====================================================================

Hope, this could give you some ideas about where to search.

Sincerily

Denis J Navas 

I've made a new release (ChessDB-3.6.11-alpha) which has two significant 
changes from the previous release.

1) You can check if your release is current with an option on the help menu.

1) You can download games from the TWIC web site and have them converted 
to a ChessDB/Scid format database all from within the GUI. The program 
works out what should be the latest version, based on the date. It does 
not 'scrape' the TWIC web site.

There will be a problem if the program thinks week 635 should be online, 
but it it not. I will need to add something to detect this possibility. 
At the minute it thinks you can download up 634, which is true. The 
changeover occurs in the software on a Sunday.

I'd be interested in thoughts on how best to use this. The interface 
looks a bit awful, as I know little Tcl, but I hope to improve that, so 
will tidy it up later. There needs to be more error checking, as there 
is a lot that can go wrong when downloading files. Hence I have marked 
it as alpha release and have made a prominent note to this effect when 
the software starts.

The default is to download the 3 most recent TWIC games collection.

Other possibilities include

a) Don't ask for user intervention - just download the latest collection 
to a file. But I'm not too keen on so much automation.

b) Let the user chose which issues, but store the number of those 
already download (to options file?), so they don't repeat them.

c) Make periodic checks to see if later collections are online, and if 
so do WHAT ??

Both need networking support. For this I have used http::geturl in Tcl 
(I originally thought to used wget, but decided it was not a very 
elegant solution). There are likely to be a few bugs.

-- 
Dr. David Kirkby

Denis J Navas wrote:
>>I've copied what I found on the repetoire editor to the ChessDB web site:
>>
>>http://chessdb.sourceforge.net/tutorial/t_adv_reper.php
>>
>>but it is very spartan
>>
>>Is there any chance you would be willing to expand on that so I can add
>>it to the web site, along with some of your examples files? It might
>>help if ohers had some examples. Unfortunatly, I don't play the same
>>openings as you, so I will probably not gain as much as from some otther.
>>
>>Like Pascal, I have some difficulty in understanging how best to use the
>>Repertoire editor.
>>
>>
>>-- 
>>Dr. David Kirkby
>>
> 
> 
> 
> I will check the document you refers and write something about repertoire 
> usage.
> 
> 
> Denis J Navas

Thnaks a lot. Just email me a document and I'll convert it to html and 
put it on there (don't bother sending me html though, as most HTML 
generated by computers is pretty alful, with lots of unneeded rubbish in 
it).

-- 
Dr. David Kirkby

>>>Both archives are available now:
>>>http://sourceforge.net/mailarchive/forum.php?forum_id=51158
>>
>>Good, so they are not lost.

Nice!

Cheers,

Jeremy

Dr. David Kirkby,  =C5=9Broda, 3 stycznia 2007 11:49:
>Michal Rudolf wrote:
>>>Sourceforge can be a bit lax some times - I've know them fail to fix
>>>problems that have gone on for years.
>>
>> Both archives are available now:
>> http://sourceforge.net/mailarchive/forum.php?forum_id=3D51158
>
>Good, so they are not lost.
>
>> http://sourceforge.net/mailarchive/forum.php?forum_id=3D1830
>
>Em, I can't access that one.
>
>Permission Denied
>
>Access to this page is restricted (either to project members or to
>project administrators) and you do not meet the requirements to access
>this page. Please contact the administrator of this project for further
>assistance
Martin Skj=C3=B6ldebrand is listed as one of Project Admins. He used to pos=
t to =20
this list, so maybe he should be asked to make archives public?

=2D-=20
Michal Rudolf

Michal Rudolf wrote:

>>Sourceforge can be a bit lax some times - I've know them fail to fix
>>problems that have gone on for years.
> 
> 
> Both archives are available now:
> http://sourceforge.net/mailarchive/forum.php?forum_id=51158

Good, so they are not lost.

> http://sourceforge.net/mailarchive/forum.php?forum_id=1830

Em, I can't access that one.

Permission Denied

Access to this page is restricted (either to project members or to 
project administrators) and you do not meet the requirements to access 
this page. Please contact the administrator of this project for further 
assistance

Perhaps they are in the process of making changes to the site.
-- 
Dr. David Kirkby

Dr. David Kirkby,  =C5=9Broda, 3 stycznia 2007 05:27:
>Hi,
>I noticed that there was only one message archived. I'm not sure why
>this would be at all. I will ask Sourceforge what is going on. It may be
>there are only archived once per month or something, but that would seem
>very odd.
>
>I can forward you the messages. I could also stick them on a web page as
>just one long page.  I can't be bothered to sort them in any way, so
>they would be a lot harder to follow than if Sourceforge made the
>archives available.
>
>I've just checked the scid web site and can't even find a reference to
>scid-users, although there is a scid-developers, although nobody has
>posted there since 2004

>Sourceforge can be a bit lax some times - I've know them fail to fix
>problems that have gone on for years.

Both archives are available now:
http://sourceforge.net/mailarchive/forum.php?forum_id=3D51158
http://sourceforge.net/mailarchive/forum.php?forum_id=3D1830
=2D-=20
Michal Rudolf

I thought I'd forward this to the list I've reported this issue Jeremy 
reported so hopefully the archives can restored, or at least they start 
archiving soon.

-------- Original Message --------
Subject: [ alexandria-Support Requests-1626626 ] Messages to mailing 
list are not being archived
Date: Tue, 02 Jan 2007 21:09:36 -0800
From: SourceForge.net <no...@so...>
To: no...@so...

Support Requests item #1626626, was opened at 2007-01-03 05:09
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=200001&aid=1626626&group_id=1

Please note that this message will contain a full copy of the comment 
thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Project Mailing Lists/Archives/Services
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: Dr. David Kirkby (drkirkby)
Assigned to: Nobody/Anonymous (nobody)
Summary: Messages to mailing list are not being archived

Initial Comment:
I'm the administrator for ChessDB  - a chess database
http://chessdb.sourceforge.net/

I have a mailing list set up,

che...@li...

which has 49 subscribers and I'm guessing there have been 50 - 100 posts 
on the list. But only one is archived at 
http://sourceforge.net/mailarchive/forum.php?forum_id=51158

It is possible for you to add archives back? Someone has reported he 
deleted his posts, thinking they would be on sourceforge, only to find 
they are not.

I've looked at the administrative page for the list and see no option to 
either enable or disable the archival process.

----------------------------------------------------------------------

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=200001&aid=1626626&group_id=1

-- 
Dr. David Kirkby

> I've not looked at the code, but does the blunder analysis stop
> permanently if score gets very large?
> 
> The reason I ask is that among weaker players, it is not unknown for
> someone to draw a won position (like king + pawn vs king) by loosing
> opposition. So it is quite possible for the game to get from a forced
> mate to a draw with one move. If the analysis mode actually stops
> permanently when after the forced mate is seen, important information
> may be lost if the situation changes.

No (or at least no modulo any bugs <grin>);
the relevant code is in tcl/tools/analysis.tcl.
Search for 'meaningless', and you'll find the code.

Basically the idea is that if the score was above the
threshold last move, and the new move leaves it above that
same threshold, then we don't care.  Unless it's a sudden
reversal in fortune.  (That is, score was -15 and is now suddenly
+327.64).  The one hole is that if the score is, say, +20, and
it suddenly drops to +10, that would not be flagged either,
even though it is likely an interesting move.

I've got a set of kids games I'm using as benchmarks, and
it is indeed useful to track wild swings (my current favorite example
is where the kids each miss a set of forced checkmates; swings the score
nicely <grin>).

> 
> I was going to look at the code at one point with a view to putting the
> name of the analysis engine, since that information might be useful if
> you review the game at a later date. I know that information is
> displayed in scid/chessdb, but I don't know at what bits of the code it
> is available.
> 

Ah, good idea, I hadn't thought of that.  I took a quick look and couldn't
find where the engine name is stashed, so perhaps I'll leave that to you.

Cheers,

Jeremy

Jeremy White wrote:
> I have incorporated most of Pascal's suggestions about
> the blunder analysis mode, including some code he sent to me,
> into a newer version of the 'Blunders' annotation mode.
> 
> The specific changes are as follows:
>   * Add a 'mistake' (?) threshold, in addition to the
>     blunder ('??') threshold
>   * Add a filter so we stop processing after the score
>     reaches a certain point.
>   * Handle engines that do not treat black advantage as a negative score
>   * Improve the dialog box a bit so that when blunders
>     mode isn't selected, the connected fields are greyed out
> 
> I did not update anything other than English, nor did I update
> the screen shot for the tutorial (my WM is very different than
> that in the tutorials already, and I thought it would be jarring).
> 
> Cheers,
> 
> Jeremy

Hi,
thank you for that. I'll patch the source later.

I've not looked at the code, but does the blunder analysis stop 
permanently if score gets very large?

The reason I ask is that among weaker players, it is not unknown for 
someone to draw a won position (like king + pawn vs king) by loosing 
opposition. So it is quite possible for the game to get from a forced 
mate to a draw with one move. If the analysis mode actually stops 
permanently when after the forced mate is seen, important information 
may be lost if the situation changes.

I was going to look at the code at one point with a view to putting the 
name of the analysis engine, since that information might be useful if 
you review the game at a later date. I know that information is 
displayed in scid/chessdb, but I don't know at what bits of the code it 
is available.

-- 
Dr. David Kirkby

Jeremy White wrote:
> So...I deleted a bunch of emails from this list, thinking I could
> just review them in the archives, as it's a shiny new list and all.
> 
> Only when I went to scan the archives, I found exactly one email.
> 
> Does anyone know why that is?  Is this list archived somewhere
> other than on the sourceforge page that comes at the bottom of each email?
> 
> Thanks,
> 
> Jeremy

Hi,
I noticed that there was only one message archived. I'm not sure why 
this would be at all. I will ask Sourceforge what is going on. It may be 
there are only archived once per month or something, but that would seem 
very odd.

I can forward you the messages. I could also stick them on a web page as 
just one long page.  I can't be bothered to sort them in any way, so 
they would be a lot harder to follow than if Sourceforge made the 
archives available.

I've just checked the scid web site and can't even find a reference to 
scid-users, although there is a scid-developers, although nobody has 
posted there since 2004

Sourceforge can be a bit lax some times - I've know them fail to fix 
problems that have gone on for years.

-- 
Dr. David Kirkby

So...I deleted a bunch of emails from this list, thinking I could
just review them in the archives, as it's a shiny new list and all.

Only when I went to scan the archives, I found exactly one email.

Does anyone know why that is?  Is this list archived somewhere
other than on the sourceforge page that comes at the bottom of each email?

Thanks,

Jeremy

I have incorporated most of Pascal's suggestions about
the blunder analysis mode, including some code he sent to me,
into a newer version of the 'Blunders' annotation mode.

The specific changes are as follows:
  * Add a 'mistake' (?) threshold, in addition to the
    blunder ('??') threshold
  * Add a filter so we stop processing after the score
    reaches a certain point.
  * Handle engines that do not treat black advantage as a negative score
  * Improve the dialog box a bit so that when blunders
    mode isn't selected, the connected fields are greyed out

I did not update anything other than English, nor did I update
the screen shot for the tutorial (my WM is very different than
that in the tutorials already, and I thought it would be jarring).

Cheers,

Jeremy

General Newcomb wrote:
> Hi Dr. Kirkby
> 
> I installed the chessdb 3.6.9, unfortunately, the "download from twic" 
> feature was not working for me. I'm not sure what the problem is. I'll 
> try to investigate it some more tomorrow.
> 
> thanks,
> 
> Alex

Hi, that does not surprised me, since I have been updating the CVS when 
the feature was not working.

Try this, which will check out a version I have tagged as 
'Snapshot-3_6_10_alpha_TWIC_working', which I believe works.

---

cvs -z3 
-d:pserver:ano...@ch...:/cvsroot/chessdb co -P 
-r Snapshot-3_6_10_alpha_TWIC_working chessdb

---

That will not be the latest source files, but should be in a working state.

It is now downloading for a minimum issue you specify (210 or more) to 
the current issue, where the current one is calculated based on the date 
I. I'm assuming one is out every week, so in a years time it will try to 
download one that is 52 versions higher than today

-- 
Dr. David Kirkby

>
>
> Date: Mon, 01 Jan 2007 11:50:48 +0000
> From: "Dr. David Kirkby" <dav...@on...>
> Subject: [Chessdb-users] Downloading from TWIC
> To: ChessDB Users <che...@li...>
> Message-ID: <459...@on...>
> Content-Type: text/plain; charset=us-ascii; format=flowed
>
> I've seen various comments about it being useful if one could download
> games from TWIC. There is a feature request in the Scid site:
>
>
> http://sourceforge.net/tracker/?group_id=26963&atid=389083
>
> Someone posted a bash script the other day and I note there is a python
> one in scripts directory in the source code.
>
> http://chessdb.sourceforge.net/tutorial/scripts/twic2chessdb.py
>
> so I guess this is what at least a few people want
>
> I assume TWIC has changed the web site a bit, since they python script
> does not work Seems a bit excessively long, to do a trivial thing too.
>
> As a UNIX guy, I agree with someone the other day about the merits of
> the command line for that sort of thing - a script can be written in no
> time at all, making use of curl, wget or similar.
>
> But anyway, I decided to have a go at adding a menu item under tools.
>
> Tools -> Download from TWIC
>
>
> My thoughts were
>
> 1) Download to the current database - makes it easier.
>
> 2) Store in the options (or perhaps other) file, the last one
> downloaded, so you don't download the same issue more than once - not
> that finding duplicates is that hard, but it is a bit pointless.
>
> Comments, suggestions ...
> --
> Dr. David Kirkby

This sounds great. Thanks for working on this Dr. Kirkby.

Alex

Hello,

You can download a new spelling.ssp and ratings.ssp from my homepage:=20
http://members.aon.at/schachverein.steyr/

Like FIDE now I only added Birthyear because of protection of privacy.
The format is now:

Nagl, Franz =A0 =A0 =A0 =A0 =A0 =A0 #- =A0 AUT [2220] 1960.??.??

I can also remove the .??.?? but I don't know, how ChessDB use the
datas internal.

Only for historic players I added full Born/Died date.

What do you think about this?

Any feedback are welcome!

=46ranz

Thanks, Dr. Kirkby. I downloaded and installed 3.6.9 and this appears  
to have fixed the problem.

Thanks again,
Cory Helfrich
cor...@ya...

On Jan 1, 2007, at 4:59 PM, Dr. David Kirkby wrote:

> Cory Helfrich wrote:
>> Hello Dr.Kirkby,
>> First, I would l like to express my gratitude for continuing the  
>> development of ChessDB and also giving me a great deal of help  
>> with the installation. I regret that I am not a programmer so I  
>> cannot participate more actively in the development.
>> I have found a little problem with the sc_spell utility. When I  
>> run it, I get the following error (note that I have never had scid  
>> installed on this machine):
>> Begin Error
>> /usr/local/bin/sc_spell: line 10: exec: tcscid: not found
>> End Error
>> Thanks again for your help,
>> Cory Helfrich
>> cor...@ya... <mailto:cor...@ya...>
>
> Hi,
> I've looked at the source and realise that the script still called  
> tccsid until release 3.6.8, where it was changed.
>
> I suggest you download the latest at least 3.6.8 (the latest is  
> 3.6.) and rebuild it.
>
> You could try editing the file /usr/local/bin/sc_spell and changing  
> line 10 from
>
> exec tcscid "$0" "$@"
> to
> exec tcchessdb "$0" "$@"
>
> But you might run into other similar issues, so I suggest you just  
> download the latest source.
> -- 
> Dr. David Kirkby

Denis J Navas wrote:
> I can't avoid to gave some opinion about the repertoire editor.  I used 
> it extensively about two years ago.  I include a pretty small example 
> when I was thinking on a repertoire as black.

Hi,
the scid tutorial has nothing on the repetoire editor.

http://scid.sourceforge.net/tutorial/t_adv_reper.html

But there are some pages in the actual program - they will be on the 
sicd site somewhere, but I can't recall where

I've copied what I found on the repetoire editor to the ChessDB web site:

http://chessdb.sourceforge.net/tutorial/t_adv_reper.php

but it is very spartan

Is there any chance you would be willing to expand on that so I can add 
it to the web site, along with some of your examples files? It might 
help if ohers had some examples. Unfortunatly, I don't play the same 
openings as you, so I will probably not gain as much as from some otther.

Like Pascal, I have some difficulty in understanging how best to use the 
Repertoire editor.

-- 
Dr. David Kirkby

2006	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec (36)
2007	Jan (101)	Feb (82)	Mar (24)	Apr (11)	May (1)	Jun (6)	Jul (9)	Aug (3)	Sep (13)	Oct (1)	Nov (1)	Dec (3)
2008	Jan (3)	Feb	Mar (1)	Apr	May	Jun	Jul	Aug	Sep	Oct (1)	Nov (1)	Dec
2009	Jan	Feb (3)	Mar (3)	Apr (4)	May	Jun (1)	Jul (1)	Aug (2)	Sep (4)	Oct (1)	Nov	Dec
2010	Jan (2)	Feb (6)	Mar (3)	Apr (6)	May	Jun	Jul (1)	Aug (1)	Sep (4)	Oct (1)	Nov (4)	Dec
2011	Jan (1)	Feb (1)	Mar (1)	Apr	May	Jun	Jul	Aug (1)	Sep	Oct (2)	Nov	Dec
2012	Jan	Feb (1)	Mar	Apr	May	Jun (1)	Jul	Aug	Sep	Oct	Nov (1)	Dec (7)
2013	Jan (3)	Feb	Mar (3)	Apr (1)	May	Jun (1)	Jul	Aug	Sep	Oct	Nov	Dec
2014	Jan	Feb (1)	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
2015	Jan (1)	Feb	Mar	Apr (1)	May	Jun	Jul	Aug	Sep	Oct	Nov (1)	Dec
2016	Jan	Feb (1)	Mar	Apr (1)	May (1)	Jun	Jul	Aug (1)	Sep	Oct	Nov	Dec

chessdb-users Mailing List for ChessDB - a Free Chess Database (Page 13)

chessdb-users — User mailing list - used by developers too.