See loadStrings(), saveStrings() and search the forum for stopwords or HashMap.

Hi PhiLho

Thank you for the pointers and I've checked all the references including the referenced code.  I've modified the RiTa Markov example to load multiple files and save a single text output but, since I'm not a programmer, I'm waaay over my head here when it comes to altering the code above to meet my simple four step process.

The main Wordnet processing occurs from Lines 32 to the end and I'm hoping the addition of  saveStrings("WordnetOut.txt", lines);  will save the processed file the same way it does in the Markov example.

That's great because it means more than 50% of the code can be used with only a small change!

In fact, I tried using some of the Markov code and changing it. For example,  loadStrings(Sample.txt)  at line 8 (after deleting the Stringtest reference and text.

Unfortunately, I can't move beyond this error:  It looks like you're mixing "active" and "static" modes.

Since I want the loaded text processed from the beginning in sequence I removed the word "random" from

1. / loop from a random spot
2.   int count =  (int)random(0, words.length);
3.   for (int i = count; i < words.length; i++)

but it produced another error.

So what about lines 1 to 31? If I knew what I was doing or had similar code examples I could adapt then I would enjoy experimenting like I did with Markov, but with the additions of stopswords, loading the file, removing the frames,  and processing each word in sequence from the beginning, I'm stumped.

Looks like your a busy person on this forum and I'm sure that comes from your knowledge and expertise. If you have a little more you can share to solve my problem I would appreciate it.


Cheers,

You should load the text in setup() (assigning it to a global variable).
The It looks like you're mixing "active" and "static" modes. message means you have put some code outside any function.

Hello:

After searching java-related forums, other processing code, and asking questions I've managed to cobble this together with some help. It loads a text file, uses stopwords.txt, and saves a file but I doesn't process every word in sequence through Wordnet.

The "random" function should only apply to Wordnet processing and the returning of a synonym (if available). 

Line 53 applies to the input text.  Is there a way to change this so every word in the input file is processed and not just random words?

thanks,

1. import rita.*;
2. import rita.wordnet.*;
3. import javax.swing.*;
4.  
5. RiWordnet wordnet;
6. String textFileData = "";
7. void setup()
8. {
9.   setVisible(false);
10.   wordnet = new RiWordnet(this);
11.  
12.   JFileChooser fc = new JFileChooser();
13.   int fileOpen = fc.showOpenDialog(this);
14.  
15.   switch (fileOpen) {
16.     case JFileChooser.APPROVE_OPTION:
17.       File file = fc.getSelectedFile();
18.       if (file.getName().endsWith("txt")) {
19.         String[] datas = loadStrings(file.getAbsolutePath());
20.         for (String data : datas) {
21.           textFileData += (data + "\n");
22.         }
23.       } else {
24.         println("You have not selected a .txt file.");
25.       }
26.       break;
27.     default:
28.       print("Please select a .txt file.");
29.       break;
30.   }
31.   doSubstitution();
32.   int fileSave = fc.showSaveDialog(this);
33.  
34.   switch (fileSave) {
35.     case JFileChooser.APPROVE_OPTION:
36.       File file = fc.getSelectedFile();
37.       if (file.getName().endsWith("txt")) {
38.         saveStrings(file.getAbsolutePath(), split(textFileData, "\n"));
39.       } else {
40.         println("Please save the file as a .txt document.");
41.       }
42.       break;
43.     default:
44.       print("Please save the file.");
45.       break;
46.   }
47.   exit();
48. }
49. void doSubstitution()
50. {
51.   String[] blocked = loadStrings("stopwords.txt");
52.   String[] words = textFileData.split(" ");
53.   for (int x = 0; x < (int)random(words.length / 2, words.length); x++) {
54.     int count =  (int)random(0, words.length);
55.     for (int i = count; i < words.length; i++) {
56.       for (String wordBlock : blocked) {
57.         if (words[i].equals(wordBlock))
58.           continue;
59.       }
60.       if (words[i].length() < 3) continue;
61.      
62.       String pos = wordnet.getBestPos(words[i].toLowerCase());       
63.       if (pos != null) {
64.         String[] syns = wordnet.getSynset(words[i], pos);
65.         if (syns == null) continue;
66.         String newStr = syns[(int)random(0, syns.length)];
67.         if (Character.isUpperCase(words[i].charAt(0)))             
68.           newStr = RiTa.upperCaseFirst(newStr);
69.         textFileData = RiText.regexReplace("\\b"+words[i]+"\\b", textFileData, newStr);
70.         RiText.deleteAll();    
71.         break;
72.       }
73.       if (count == words.length) count = 0;
74.     }
75.   } 
76. }