Home

Previous 20

Jul. 2nd, 2008

July 2nd Entry

I spend most of day finding bugs and fixing them.
I also changed the open file option in JFLAP.
It used to be the case, when you open file, you always get the root directory.
Now, when you click open file, you will get the current directory.
I've also finished implemeting tutorial page for "creating trap state" feature.
It can be founded under JFLAP08/tutorial/fa/trapsate.

-Jason

July 1st Entry

I've finish implementing multi-run file input feature in JFLAP.
Now, user can load input file to test FA/PDA/Turing/Grammar
Old multi-run pane has been modified to implement this new feature.
Also, I've made some changes in detecting Grammar feature.
It can now detect context-sensitive  grammar.

-Jason

Jun. 28th, 2008

June 28th Entry

 I worked at my aunt's house in New Jersey.
And I finished implementing grammar detection feature in JFLAP.
I forgot to bring Linz's book, so "context-sensitive" grammar is not detected yet.
Also, this feature is not fully tested yet.
So, I will test this out tomorrow and make .jar file available in JFLAP08 folder.
Also, I've changed the logo and verison information of JFLAP to 6.4 (learned from last mistake).

-Jason

Jun. 27th, 2008

June 27th Entry

 I've finished implementing GUI.
I've selected the GUI and I believe that is the best design to ask for input.
I finished implementing this feature on all FA/PDA, Turing machine (even multitape).
I performed bug testing to see, if our new feature is reliable.

-Jason

Jun. 26th, 2008

June 26th Entry

I am still working on new input GUI.
GUI was working well with FA and PDA.
However, with multiple-turing machine GUI appears to be very unfriendly to users.
I have few other possible GUI design to implement.
Also, I've started working on creating tutorial page for adding trap state.
My goal for this weekend (making up for my missed work on Mon, Tues) is finish implementing GUI and add identifying feature on JFLAP.

-Jason

Jun. 25th, 2008

June 25th Entry

I've struggled with creating GUI for "Input with file feature".
I ended up creating two separate GUIs.
I like my second GUI, which utilizes JOptionPane, much better than my original GUI.
So, I am going to use that.
I have implemented this feature to FA and PDA.
I am going to further test and implement this feature on TM and Grammar.
(Not sure, if I have to do this for Mealy and Moore machine).

-Jason

Jun. 20th, 2008

June 20th Entry

I did additional bug testing on the new feature that I created yesterday.
Also, I finished implementing basic GUI design for the next feature that I am going to work on.
I decided to use the standard txt file with space or new line delimiter to read in the input.

-Jason

Jun. 19th, 2008

June 19th Entry

I've finished implementing "adding trap state" feature in JFLAP.
The problem I mentioned yesterday is finally fixed.
Now, I changed my focus on implementing new feature of loading inputs through file.
Also, I set up the basic DTD format for JFLAP XML to conform. 

Jun. 18th, 2008

June 18th Entry

I finished the algorithm of creating trap state and making transitions.
However, I've stumbled upon the problem when I represent this algorithm on GUI.
I keep get null Transition, even though I create transition object in GUI.
It appears to be GUI viewer and controller is running on separate thread.
I worked on fixing this problem.
There are several solution to the problem, but I am trying to find a way to fix it without changing current JFLAP classes.
Hopefully, I can fix the problem tomorrow.

Jun. 17th, 2008

June 17th Entry

I've fixed saving problems in JFLAP.
Now, user can save Regular expression.
I've started working on adding trap state to DFA feature.
I've finished most of interface and will be able to finish implementing this feature by tomorrow.
I also looked at JFLAP XML forms and started working on creating XML DTD.

-Jason

Jun. 16th, 2008

June 16th Entry

I had severe stomachache. (So, I stayed home and worked here)
I've looked over XML format in JFLAP and it appears to be that we have XML DTD.
I've worked on the bug report about XML, but I cannot regenerate that error.
Perhaps, that bug reporter was using older version of JFLAP.
I am going to contact him to find out more detail tomorrow.

-Jason

Jun. 13th, 2008

June 13th Entry

I've started looking at changing XML for newer version of JFLAP.
It appears to be that DFA/NFA has no problem saving notes and state names.
I am still trying to find out what exactly the bug reporter was doing.
Also, I've added some additional comments in my code.

-Jason

Jun. 12th, 2008

June 12th Entry

I've mostly worked on trying to find more bugs in JFLAP.
I also worked on commenting on the codes that I wrote.
In addition, I tested my building blocks, so it can be used in the future.

-Jason

Jun. 11th, 2008

June 11th Entry

I've started looking at the bug reports that was under jflaperror6.3 folder.
I think the first error was caused by the java running out of heap memory and as we discussed before, nothing could be done to fix that.
I've looked at the new CYK parser bug. I contacted the person who sent the bug report and got that .jff file to work on it.
I noticed that the error was caused by the difference in CNF grammar in the parser and the tracer.
Now, that bug is fixed.
Also, instead of changing the logo picture in "About" menu, I just created the new picture that is in similar style.

-Jason

May. 23rd, 2008

May 23rd Entry

Today, we are going to release new version of JFLAP.
Additional bug check was done and web-pages and .jar file's version information is correctly updated.
I've also added two additional files into the zip file.
After finishing final preparation for JFLAP release, I've read two journal articles that utitlizes JFLAP.

-Jason

May 22nd Entry

To fix the problem posed in my previous entry, I decided to sort the productions before I export them to grammar window, so user can get exactly same grammar with the same order when they perform conversion.
I've decided to give higher priority to variables that derives the terminals.
I believe that could help the parsing to be done more efficiently.
I have also updated the tutorial page and conversion.html file is ready to go.
It appears to be that everything is now set up for the JFLAP release.

-Jason

May. 22nd, 2008

May 21st Entry

When I was creating tutorial page, I discovered that productions in the  grammar converted from TM are not necessarily in the same order.
And sometimes, it caused the delay in the parsing.
I spend most of the day in attempt to solve this problem.
I've also updated web tutorial page and created changes.html

-Jason

May. 20th, 2008

May 20th Entry

CYK Parser bug is finally fixed.
The bug was caused when I perfrom backtracking from converted CNF to regulat CFG.
After fixing the bug, I've tested several grammars and they all worked fine.
I am still trying to find the second example for TM to URG.
However, it appears to be that the grammar is just too big.
I will go over the algorithms one more time tomorrow, just to make sure there are no programming bug.

-Jason

May. 19th, 2008

May 19th Entry

I've fixed some bugs founded in building blocks that I built last week.
There was a "println" statement in some portion of JFLAP code and it took me almost 3 hours to find that println statement and get rid off it.
CYK bug that was reported in jflaperror35 appears to be pretty complicated.
I am still thinking about the way to solve the problem.
I have few ideas, but before I start coding I would like to talk with Prof. Rodger about it.
Also there was bug in special brute force parser for converted grammar from TM (parser never stops).
Now that bug is fixed.

-Jason

May. 16th, 2008

May 16th Entry

I have fixed additional bugs in building blocks.
I've also created 2 examples that utilizes these building blocks.
They are located under jflap08/needTMBB/examples folder.
JFLAP bug that occurs while saving TM has been fixed.

-Jason

Previous 20