4.1 BETA

[zz]

[quote=stan]
i suggest that you wait a week or two until the compound filter is implemented and then you'll be able to add a custom filter to the ticket generator and restrict the tickets being created.
[/quote]

That will be really nice.

[stan]

[quote=zz]
It shows 50 MB which is probably OK, but I think the number of comninations is wrong. I dont know when working with file if will be used proper (all) combinations, since it shows less...For example try to make couple of milions random combinations (remembering the number), save them and then later open the source file- may show contains less ( or some different number)combinations
[/quote]

it's working fine for me.
you can verify the number of tickets in a file yourself - it's the size of file in bytes divided by the count of numbers in a ticket + 1.

e.g. if the file size is 37788993 bytes then the number of tickets in the file is:
37788993/(10+1) = 3435363 tickets (e.g. 10/80 keno lottery).

also pls note that the file size shown under the file name in the filter window is rounded to megabytes or gigabytes and is informational only.

[zz]

[quote=zz]
It shows 50 MB which is probably OK, but I think the number of comninations is wrong. I dont know when working with file if will be used proper (all) combinations, since it shows less...For example try to make couple of milions random combinations (remembering the number), save them and then later open the source file- may show contains less ( or some different number)combinations
[/quote]


However, maybe I did something wrong when saving those files, since I was actually paying attention on how many combinations the application might make. So, don't want to take the time you maybe reserved for the generator and the compound filter...or maybe for collecting your lottery wins...

[stan]

[quote=zz]
[quote=zz]
It shows 50 MB which is probably OK, but I think the number of comninations is wrong. I dont know when working with file if will be used proper (all) combinations, since it shows less...For example try to make couple of milions random combinations (remembering the number), save them and then later open the source file- may show contains less ( or some different number)combinations
[/quote]


However, maybe I did something wrong when saving those files, since I was actually paying attention on how many combinations the application might make. So, don't want to take the time you maybe reserved for the generator and the compound filter...or collecting your lottery wins:-D
[/quote]

when you click the 'cancel' button while saving the package contents only part of the package tickets gets saved to file. maybe that's why...

[zz]

[quote=stan]

when you click the 'cancel' button while saving the package contents only part of the package tickets gets saved to file. maybe that's why...
[/quote]

I checked it again and now works OK. Probably I did something wrong.
BTW, the generator and compound filter are really nice news.
Thanks Stan.

[stan]

sorry, no updates this weekend (feb 12-13). it's valentine's day and i needed to spend some time with my girlfriend.

but most of the code for compound filters is ready, i just need to update the regular filters to be compoundable. hopefully, i'll have it completed by the end of this week

[zz]

[quote=stan]
but most of the code for compound filters is ready, i just need to update the regular filters to be compoundable. hopefully, i'll have it completed by the end of this week
[/quote]

Is the generator part of it or is it separated?

Just a small hint - I was just interested why some shareware sites gave you only 4.5/5 when you deserve 5/5. Maybe because that Freetrialsoft link at the starting page is broken?

[stan]

[quote=zz]
Is the generator part of it or is it separated?
[/quote]
the filtered generator is part of the update. it's the easier task...

[quote=zz]
Just a small hint - I was just interested why some shareware sites gave you only 4.5/5 when you deserve 5/5. Maybe because that Freetrialsoft link at the starting page is broken?
[/quote]

well, the software is written in java so it may seem slow and/or memory extensive to some people.
the main reason for 'less than perfect' awards is probably that the version 4.0 doesn't look well under java runtime 1.5. but it's already fixed in version 4.1 BETA.

personally, i don't value these awards that much. imho every software that can install and uninstall itself and has a half-decent looking ui will get 4 to 5 star award.

[zz]

[quote=stan]
well, the software is written in java so it may seem slow and/or memory extensive to some people.
the main reason for 'less than perfect' awards is probably that the version 4.0 doesn't look well under java runtime 1.5. but it's already fixed in version 4.1 BETA.

personally, i don't value these awards that much. imho every software that can install and uninstall itself and has a half-decent looking ui will get 4 to 5 star award.
[/quote]


C'mon Stan I was just joking and actually wanted to tell you the link to the Freetrialsoft at your starting page is broken, so you can fix that.

We all know that your program deserves at least 5/5..

[stan]

[quote=zz]
[quote=stan]
well, the software is written in java so it may seem slow and/or memory extensive to some people.
the main reason for 'less than perfect' awards is probably that the version 4.0 doesn't look well under java runtime 1.5. but it's already fixed in version 4.1 BETA.

personally, i don't value these awards that much. imho every software that can install and uninstall itself and has a half-decent looking ui will get 4 to 5 star award.
[/quote]


C'mon Stan I was just joking and actually wanted to tell you the link to the Freetrialsoft at your starting page is broken, so you can fix that.

We all know that your program deserves at least 5/5..
[/quote]


thanx for the note, i'll fix the link.

btw, i wasn't joking about the shareware awards. but at least we have some nice looking graphics at our homepage.

[Maryland]

Stan, don't know if anyone mentioned this but when highlighting winning numbers it doesn't show up in history differences with colours. Works in original version but not Beta.

[stan]

[quote=Maryland]
Stan, don't know if anyone mentioned this but when highlighting winning numbers it doesn't show up in history differences with colours. Works in original version but not Beta.
[/quote]

thanks for the note. i think i've already fixed this (and some other coloring related issues). i didn't post an updated yet as i'm in the middle of reorganizing the filter code for the compound filter support.

[zz]

[quote=stan:1108328458]
but most of the code for compound filters is ready, i just need to update the regular filters to be compoundable. hopefully, i'll have it completed by the end of this week
[/quote]

It seems a lot of additional programming is necessary to put it into a properly working area?

[stan]

hi everybody,

beta version 4.1.0 build 421 is available for download from the usual location and it has the following updates:

bug fixes:
* fixed all panel coloring issues (pls report any bugs i forgot to fix)
* fixed the Xif5 minimizer
* package winning page (the table didn't get reset when the package contents were changed elsewhere)
* fixed some filter window sizes
* maybe some other bugs that i forgot about

new features:
* compound filters
* filtered ticket generator

usage of the compound filters should be fairly obvious as the user interface is (hopefully) self explaining. but since there's no context help yet, here's a couple of hints:
* the compound filters are shared across lotteries if possible (the lotteries must have the same parameters - min/max number, numbers drawn, schedule etc)
* compound filters use 'shortened boolean evaluation'. e.g. if a compound filter looks like A and B and C and filter A fails, filters B and C won't get evaluated because the whole compound filter will fail anyway. same way if compound filter is 'A or B or C' and filter A passed then filters B and C won't get evaluated either as the whole compound filter will pass. therefore pay attention to the order of compound filter parts as it may greatly speed the filtering process up. it's recommended to use 'fast' filters first (odd/even, sum) and then the slow filters (match winning numbers, match file)
* the AND operator takes precedence over the OR operator. it means that compound filter A and B or C and D is in fact (A and B) or (C and D).
* the compound filter names should be kept reasonable short as they are shown as items in the filters menu. use the description field for any extra info, it is then shown as a tooltip in the filters menu.
* when building a new compound filter you can doubleclick the compound filter part and change its default name to something more meaningful (e.g 'leave tickets with date of birth numbers')
* the preview button in filter windows invoked from the ticket generator will actually start the ticket generator and tell you how many tickets failed/passed for the selected filter, it's not a preview of tickets in the package.

unfortunatelly i had to add/modify much more source code than i previously thought so it's very likely to be buggy. pls watch out and report any problems.

there will be some small updates to the user interface but the basic concept should be the same (if there are no major objections ).

[Maryland]

Thanks Stan!