Prasad Saya

Closing a topic is not an end to anything really. George G. can always solve the way he perceives it. He can try to understand whats posted by the ranchers and figure the solution. May be there is another way. Sometimes pausing little bit can clear things up.

Perhaps there are many thoughts and ideas to focus on. May be its little overwhelming (it can be sometimes when there is pressure to figure a solution on time, etc.). Can't say what are the factors affecting. Over years I have come to know that the most deterring factor in solving an issue in software field is human and his intervention; its never a technology or hardware or software.

Good to know you are finishing your code without any issues   . While posting the code make sure its properly indented (and do remove the commented code). Also, if code comments (saying what it does) are placed strategically helps others to read the code and respond with a comment or a suggestion.

RandomAccessFile has mechanism called file pointer which allows access to data from a known particular position in a file. But, I don't know if it can be used with a file with serialized objects.

In case you choose to use a List collection to store and then retrieve objects from it, one can navigate thru a list using a ListIterator - an iterator for lists that allows traversal of the list in either direction.

But what i want in my project is to get first object , last object, previous etc. ...

Prasad Saya wrote:I think, first you may have to write the code without streams API (functional-style programming) using pre-Java 8 constructs and find the solution. And, then translate or refactor it into using the streams API.

Enthusiasm to do functional-style programming is not enough. It requires study and effort. Switching from OO style programming to functional-style requires understanding of why functional style, what are the differences, advantages and disadvantages, use cases, etc. Learning or browsing thru the API's is not sufficient, one needs to know how the provided features can be used and in the right situations.

A way to learn (other than books, tutorials, tests, etc) is to actually take some non-functional code and convert it. One can immediately see the differences - like the compactness of the code, the self-documentation, the new features and the efficiency.

I think the GO command need to be studied and list out the use cases. Then, work how these cases can be handled within this application.

Stephan van Hulst wrote:

Prasad Saya wrote:I think, first you may have to write the code without streams API (functional-style programming) using pre-Java 8 constructs and find the solution. And, then translate or refactor it into using the streams API.

I don't agree with this advice. People don't learn functional programming by translating imperative programs. It's good to get into a functional mindset straight away.

Then one should learn functional-style Java programming. One should put some effort, in a professional manner.

I think, first you may have to write the code without streams API (functional-style programming) using pre-Java 8 constructs and find the solution. And, then translate or refactor it into using the streams API.

I just used most of your code and refactored it like this (as I had explained earlier) and you may have to figure some details:

public class StringCounter {     private static HashMap<String, Integer> dictionary = new HashMap<String, Integer>(); private static List<Vector<Integer>> vectors = new ArrayList<>(); public static void main(String[] args) { List<String> textFileList = Arrays.asList("Test.txt", "Test2.txt");         try { Dictionary reader = new Dictionary(dictionary);             for (String text : textFileList) { reader.fileScanner(text); } List<String> inputFiles = Arrays.asList("A.txt", "B.txt", "C.txt"); // more files             for (String inFile : inputFiles) { processDocument(inFile); }   } catch(IOException | FileNotFoundException e){ e.printStackTrace(); }                                // use the vectors list further here, etc...    } private void processDocument(String inFileName)    throws IOException { // initialize dictionary (the hash map) values to zero ... // read the new document file to an array list Scanner textFile = new Scanner(new File(inFileName)); ArrayList<String> file = new ArrayList<String>(); while(textFile.hasNext()) { file.add(textFile.next().trim().toLowerCase()); } // iterate the arraylist of words -and- update dictionary with count for(String word : file) { Integer dict = dictionary.get(word); if (!dictionary.containsKey(word)) { dictionary.put(word, 1); } else { dictionary.put(word, dict + 1); } }             textFile.close(); // create the vector with the dictionary values and print Vector<Integer> vec1 = new Vector<>(dictionary.values());                         for (Integer count : vec1) { System.out.println(count); } // add vector to the vector list vectors.add(vec1); } }

The main method does this:

- from the original code: creates the Dictionary reader = ... using the two files Text.txt and Text2.txt (or more files)
- create a list of your 10-12 input files (A, B, C,...)
- process each file in a loop: processDocument

The GO command of sql server and its usage: http://www.sqlines.com/sql-server/sql_batch_go

You need to figure how to create batches of queries and how the GO affects the way the batched sql statements will be executed.

Would the dictionaries values be reset to 0 after each call?

Yes. The value part of the map.

But would that not overwrite the pre-existing vector(s)? If so, would this guarantee each vector would be different?

Yes, it overwrites.

You need to create another static variable List<Vector> myVectors = ....
-and-
for each call store the vectors in that. So, at the end of processing all the 10-12 files you will be left with myVectors list with the output of those 10-12 input files/documents.

You can try this.

Define "dictionary" as a static variable. Store the 10-12 file names to be processed in an ArrayList and process each file in a for-loop in a new function called processDocument or processInputFile:

private static void processDocument(String fileName) { // initialize dictionary's values to zero // read the new document file to an array list of words // iterate the arraylist of words -and- update dictionary with count // create the vector with the dictionary values and print }

I am not familiar with SQL Server; but it looks like you have to execute sql statements as a batch. Look for something like JDBC execute batch on Google search.

What is he SQL script you are trying to run?

File tempFile = new File(inFile.getAbsolutePath() + ".tmp");

This above statement doesn't create a file; its a File object, but not an actual physical file in a directory. The file gets created when you do this:

tempFile.createNewFile();

Then this tempFile can be renamed to the inFile, as you intended to.