Active Topics
-
Is there a section to talk about Java ME? Apps, etc. (5)
to General by teroyk - 6 days, 3 hrs ago -
Extra softwares in Sailfish using CLI, repositories, etc (118)
to SailfishOS by nieldk - 6 days, 9 hrs ago - more...
|
2008-04-18
, 13:41
|
|
Posts: 27 |
Thanked: 4 times |
Joined on Nov 2007
|
#22
|
terrencegf is my new hero. THANK YOU!
 |
|
2008-04-18
, 16:28
|
|
Posts: 221 |
Thanked: 182 times |
Joined on Jul 2007
@ Central Illinois
|
#23
|
Updated .deb to 1.0.3. I just can't help tinkering! 
Since there was some concern about xkbd, I decided to put the xkbd stuff within the python code rather than launching it separately via the xword.sh shell script. There is now a Preferences Menu option (with a check mark) to "Show xkbd". This option gets saved so that xkbd will show up (or not) when you start xword.
It still checks for the presence of /usr/bin/xkbd, and does not give you the Preference menu option if the binary isn't available.
So this really isn't a necessary update, but it makes it a bit nicer since xkbd is launched by xword. Also, this way the xkbd gets closed when the xword application exits.
Since there was some concern about xkbd, I decided to put the xkbd stuff within the python code rather than launching it separately via the xword.sh shell script. There is now a Preferences Menu option (with a check mark) to "Show xkbd". This option gets saved so that xkbd will show up (or not) when you start xword.
It still checks for the presence of /usr/bin/xkbd, and does not give you the Preference menu option if the binary isn't available.
So this really isn't a necessary update, but it makes it a bit nicer since xkbd is launched by xword. Also, this way the xkbd gets closed when the xword application exits.
 |
|
2008-04-18
, 16:32
|
|
Posts: 5,478 |
Thanked: 5,222 times |
Joined on Jan 2006
@ St. Petersburg, FL
|
#24
|
Be cool if you could use some sort of Graffiti like text entry to write the letters into the cells directly. 
 |
|
2008-04-18
, 16:46
|
|
Posts: 566 |
Thanked: 145 times |
Joined on Feb 2008
@ Tallahassee, FL
|
#25
|
Originally Posted by terrencegf
Thanks for the update!! It works well. I just can't help tinkering!
Now, there is a problem with how the application handles "Today's Crossword" from the NYT (but not the archives)... the current puzzle has the solution scrambled (it's part of the .puz file format specification), and there's no provision in xword to provide the 4-digit unscramble key (and, thus, check your solution) which is provided the following day, when the puzzle is released (in unscrambled form) to the archives.
Obviously, this is beyond the scope of what you're doing, and more along the lines of making changes to the actual xword code so that it handles these files properly. I'll have to dig through my archives and see if I can find the C routine I wrote years ago to determine the scramble code from a given *.puz file... if I can only remember which computer was my "main workstation" at that time....
|
|
2008-04-18
, 18:04
|
|
Posts: 221 |
Thanked: 182 times |
Joined on Jul 2007
@ Central Illinois
|
#26
|
Originally Posted by briand
I don't do the NYT puzzles, so I've never seen a scrambled solution. However, if you go to the main xword page, you will find this note near the bottom of the page.
Now, there is a problem with how the application handles "Today's Crossword" from the NYT (but not the archives)... the current puzzle has the solution scrambled (it's part of the .puz file format specification)
Using a locked puzzle
Sometimes a puzzle will be locked so that the answers are unavailable. Unfortunately, Xword's support for locked puzzles is somewhat flaky. When using a locked puzzle, you should not click on the "Check" or "Solve" buttons, since they will give incorrect information. However, you can still use Xword to enter answers for a locked puzzle. Later, when an unlocked version of the puzzle is released (usually the next day), open it using Xword. When asked, choose to continue where you left off. Now you can use the "Check" and "Solve" buttons to see how well you did. (Sometimes crosswords will come with a four-digit code to unlock a locked puzzle file. This code is not needed by Xword.)
|
|
2008-04-18
, 18:50
|
|
Posts: 566 |
Thanked: 145 times |
Joined on Feb 2008
@ Tallahassee, FL
|
#27
|
yeah, I saw that. :/
If you crack open a completed .puz file in a hex editor, you can see the layout with the user's answers and the "key" answers, laid out as the grid is laid out. If the .puz isn't scrambled (or "locked"), then it's a plain english text read. If it's scrambled, then the second ('answer') grid will be scrambled. In the actual AcrossLite application, entering the 4-digit (1-9, no 0's!) code will de-scramble the "answer key" copy in memory, and allow direct comparisons (check word, check grid, etc). After applying the 4-digit key to the puzzle, subsequent saves to disk will save an 'unlocked' version.
Anyway, several years ago, during a time I should have (no doubt) been working on something else, I fiddled around and found where (and how) the 4-digit key is stored in the .puz file, and (subsequently) how to use that key number to translate the 'scrambled' grid in the .puz file into an 'unscrambled' grid. Again, I'll snoop around and see if I can find that code snippet, and apply it to the project.
More properly, "This code is not needed by Xword, as long as the site in question offers an unscrambled version of the same puzzle for download at a later date/time."
If all the site offers is the 4-digit code and no "unlocked" puzzle, then there is no way to check your answers.
If you crack open a completed .puz file in a hex editor, you can see the layout with the user's answers and the "key" answers, laid out as the grid is laid out. If the .puz isn't scrambled (or "locked"), then it's a plain english text read. If it's scrambled, then the second ('answer') grid will be scrambled. In the actual AcrossLite application, entering the 4-digit (1-9, no 0's!) code will de-scramble the "answer key" copy in memory, and allow direct comparisons (check word, check grid, etc). After applying the 4-digit key to the puzzle, subsequent saves to disk will save an 'unlocked' version.
Anyway, several years ago, during a time I should have (no doubt) been working on something else, I fiddled around and found where (and how) the 4-digit key is stored in the .puz file, and (subsequently) how to use that key number to translate the 'scrambled' grid in the .puz file into an 'unscrambled' grid. Again, I'll snoop around and see if I can find that code snippet, and apply it to the project.
This code is not needed by Xword.
If all the site offers is the 4-digit code and no "unlocked" puzzle, then there is no way to check your answers.
|
|
2008-04-18
, 19:05
|
|
Posts: 364 |
Thanked: 54 times |
Joined on Feb 2008
|
#28
|
Originally Posted by GeneralAntilles
Absolutely!! I use graffiti on my Clie when running BeCrossword and it's so easy. I never thought I had become so dependent on that Graffiti for input but it really is/was very useful.
Be cool if you could use some sort of Graffiti like text entry to write the letters into the cells directly.
|
|
2008-04-18
, 19:10
|
|
Posts: 364 |
Thanked: 54 times |
Joined on Feb 2008
|
#29
|
Originally Posted by terrencegf
I love you and wanna have your baby....hehehehe...seriously though thanks a bunch for getting this on the road to ported over.
Updated .deb to 1.0.3. I just can't help tinkering!
Since there was some concern about xkbd, I decided to put the xkbd stuff within the python code rather than launching it separately via the xword.sh shell script. There is now a Preferences Menu option (with a check mark) to "Show xkbd". This option gets saved so that xkbd will show up (or not) when you start xword.
It still checks for the presence of /usr/bin/xkbd, and does not give you the Preference menu option if the binary isn't available.
So this really isn't a necessary update, but it makes it a bit nicer since xkbd is launched by xword. Also, this way the xkbd gets closed when the xword application exits.
I had to uninstall everything last night because something buggered screen taps in microB. And it was after xkbd just quit on me a few times. But have not had time to make sure I did the xkbd install correctly.
I think the opriginal author must have gotten bizzy with other stuff and/or nobody actually showed interest. But I the app seems to run OK for the most part on the tablet.
Can't wait to give the new setup a try tonight. thanks a bunch for all the work on it!
|
|
2008-04-18
, 19:33
|
|
Guest |
Posts: n/a |
Thanked: 0 times |
Joined on
|
#30
|
A couple other bugs I've noticed:
1. The touch pad is a little TOO sensitive: more often than not a letter gets printed twice when I tap it with the stylus, which results in a large amount of backspacing.
2. When you are in the lower half of a 21x21 puzzle, you can no longer go to a particular square by tapping the square -- it goes to the square that had been in the place tapped when the top half of the puzzle was showing on the screen. I do not like having to maneuver to the place I want to go with the arrow keys.
1. The touch pad is a little TOO sensitive: more often than not a letter gets printed twice when I tap it with the stylus, which results in a large amount of backspacing.
2. When you are in the lower half of a 21x21 puzzle, you can no longer go to a particular square by tapping the square -- it goes to the square that had been in the place tapped when the top half of the puzzle was showing on the screen. I do not like having to maneuver to the place I want to go with the arrow keys.
i had to use a crowbar (xterm as root) to get libxpm4 and your xword to install without complaint, but I have managed to get xkbd and xword both installed on my n800 (mind you, all without direct connectivity [there's no accessible wireless here], to boot!), so I'll play around with it (...some more. qwerty12 had, as he mentioned, previously compiled up the package and kindly sent it to me, but I only gave it a very rough once-over.)
anyway... on to the NYT puzzle page for a couple .puz files. thanks, again.