Index

1. Problem Statement
2. Motivation
3. Proposal and Implemented Features
4. Time taken for implementation
5. Solution Design(Modules)
   • ITRANS to UNICODE conversion
     • Mapping
     • Parsing
6. Implementation Details
7. Testing
8. Challenges
9. Future Work

Problem Statement

To construct a tool for ITRANS to unicode conversion(Devanagari script) having word meanings associated with each word.

Input : A String of english characters
Output: The devanagari unicode value of the input and word meaning of the most recently written word.

Sample Input: "bhaarat"
Sample Output: "भारत"

Motivation

The main motivation is the people whose native language is other than English

Often people do not know English,people knowing the itrans notation can use this program

To get output in my mother tongue

The motivation was also get something in my mother tongue

Proposal and Implemented features

• Initial Proposal

Initilally i planned to take input as a text file containing itrans code to convert.
Initially no GUI feature is considered to implement.
Output can be saved in a file.
Meaning of the words as much as possible can also be displayed.

• Current status

Logic has been implemented
Now input can be given to program in a text file.
Output can be shown in a text file.
70% work has been completed. My project will be completed till October 20.

• What is remaining

GUI feature is not implemented till now

Solution Design

  Modules in the Project

• ITRANS to UNICODE conversion: This module converts english text into its Devanagari equivalent by using its two submodules:
  • Mapping: This module contains the mapping between english character(s) and its corresponding unicode value. Input to this module will be will be a set of english characters(say 'gh') and the corresponding unicode('घ ') will be returned if exists otherwise 'null' will be returned.
  • Parsing: This module parses the given input text and finds the corresponding unicode for the given english string. Input to this module will be a word(say 'meraa') and corresponding devanagari unicode('मेरा ') will be returned.src
• Testing: The tool will be tested for the many different test cases(including two songs)which are created using open source website then the outputs are compared using using automated program.src

• Architecture of the application

Implementation Details

• Programming Language: Python
• Packages used: Tkinter package for GUI
• Lines of Code: All part of code is written by me only
  appx. 200 lines will be written for GUI
  210 lines of code in converting itranse to unicode (convert.py) file.
• Resources Used:
  • Python 2.6.5 on linux
  • Standard Unicode 5.1 code chart, by Unicode Consortium
  
  
  

Testing

Testing: The tool will be tested for the 14 different test cases(including two songs)which are created using open source website. The work will be done for tesing is as follows:

1. Creat 14 test cases using the software inhttp://www.giitaayan.com/x.htm by making input files and corresponding output file.
2. Exhaustive testing of the tools will be done by constructing such test cases which prints all possible devanagari words of length 1.
3. Two test cases having National Anthem and National Song of India are also will be created.
4. For each test case, we will have two files "in" and "out". "in" file is the input to the tool constructed and the output obtained by the tool is compared to the "out" file. If both match, the output of the testing will be successful otherwise it will print the line which is not matched.

Challenges

• Getting input from user

For this problem i plan to use Tkinter package available in python for GUI.

• Logic for conversion

This was solved by implementing small modules step by step and testing each module for various inputs
and subsequent funcationality was added to accept various inputs and checking whether the output is correctly displayed or not.

• Getting output from programm

It will be displayed in a file which open as user hit "Translate" button
and file will be opened automatically and output will be saved.

Future Work

Basically there could be two important extensions to this project:

1. One of the most useful extension for ITRANS to unicode converter can be indic text predictive typing Which will be useful in typing frequently used words.
2. As i will creata desktop application using Python, so to use the tool everyone will be required to download the tool and then execute it. To make it easily accessible to the users, this application can be converted into a web application so that the overhead of downloading and then executing the code can be removed. I will also try to do that in my application.