PraNaMa

7/29/2019 PraNaMa

1/28

PraNaMaScripting Language in Kannada

Dept. Of CS & E, GEC, Hassan Page 1

Chapter 1

Introduction

In computer science, an interpreter [10] normally means a computer program that

executes instructions written in a programming language.

An interpretermay be a program that either

1. Executes the source code directly

2. Translates source code into some efficient intermediate representation (code) and immediately

executes this

3. Explicitly executes stored precompiled code made by a compiler which is part of the

interpreter system

A scripting language [8],[9],[10] is a command set for controlling some specific piece of

hardware, software, or operating system, often with rudimentary and in some cases more

advanced programming-like control flow constructs, and is almost always usable from a stored

format such as a simple text file, a section of read-only persistent storage in an embedded device,

a deck of punched cards, or other mechanism. Stored sets of commands, especially those stored

in files, can be generically called scripts, but usually have more specific names based upon some

more conventional programming language, although a key differentiator is that even these

scripting languages almost never use the normal tool chain characteristic of the language they

resemble Kannada Scripting Language PRANAMA is one such application specific language

which can be used to program in Kannada. The language provides a new wave in computer

programming by facilitating the users who want to learn and implement computer programs in

Kannada.

1.1 Kannada scripting language has some of the following features:

Support for programming in a regional language Kannada and there by allow userswho know the language edit and interpret programming in the same.

7/29/2019 PraNaMa

2/28



Well defined keywords, identifiers, functions implemented to support programming anduse the same.

Well defined set of rules to govern the syntax and semantics of programming. This makesprogramming much easier and stick to one particular pattern.

Efficient GUI based editor to edit and also interpret the programs written. The editor alsoprovides the features like file open, save as, save etc.

A separate output window to display the executed results in Kannada itself in a userreadable format.

Efficient memory management techniques used to store and retrieve the values ofvariables used in the programs.

A parser which handles the written program to be passed to computer understandableformat and thereby process it to give out the results.

A separate file extension for the files to be saved called .ks which can be recognized bythe editor even when file is loaded.

This paper aims mainly at providing its users a programming experience in a regional language

Kannada. The users can edit, interpret and thus observe the results also in Kannada.

7/29/2019 PraNaMa

3/28



Chapter 2

System Design

The system architecture affects the performance, robustness, distributability and

maintainability of a system. The particular style and structure chosen for an application may

therefore depend on the non-functional system requirements:

1. Performance

2. Security

3. Safety

4. Availability

5. Maintainability

2.1 Design Considerations

The main aim is to provide a platform for those who want to program using a language

that is typed not in English but in Kannada. This would aid those who are not well versed in

English, who know Kannada and want to learn programming. Thus we have kept the features

supported by PraNaMa as few as possible only retaining those features that would help the

students of PraNaMa in learning higher level languages such as C, C++ in the future.

2.2 Designing PraNaMa:

Figure 1Design/Flow of the PraNaMa

7/29/2019 PraNaMa

4/28



The high-level design shows three major components:

1. First Pass

2. Second Pass

3. Meta Data

The first pass reads the input file and extracts information that will be used during the

second pass and stores it as Meta Data. This information could be things like name, type and

scope of variables, name and return type of functions and information about any files that are to

be linked with the existing file. The second pass handles the major part of interpreting the

program written.

It handles various programming constructs like the conditional statements and looping

constructs. Second pass also handles the evaluation of expressions, function calls and linking of

files.

A detailed explanation of each of the above components is given below.

2.3 Pass 1:

Figure 2 shows the flow/design of the first pass of the PraNaMa interpreter

The first pass scans each line of the program and identifies all lines that contain variable

declarations and function calls. Important information such as the variable name, type and scope

is stored as meta-data. This meta-data is crucial for in the second pass where the values of

variables forming a part of function calls and expressions need to be retrieved in order to

successfully call the functions and evaluate the expressions.

7/29/2019 PraNaMa

5/28



2.4 Metadata:Metadata is used in the second pass to retrieve values of variables, to retrieve function return

types and to retrieve variable types and scope of variables. Two separate tables are used to store

metadata about functions and variables. The details about the entries in each table are shown in

table 1 and 2.

Table 1 Memory Table

2.5 Pass 2:

Each line from the file is read and analyzed. The analysis helps in identifying the next

action that has to be performed. The action to be performed is based on the analysis of the

current line. The line may be a function call, start of a loop, a conditional statement or an

expression to be evaluated. Based on the type of the line an appropriate function is called to carry

out the correct action. The function call may display a result or alter the memory table entries.

Once the action appropriate for the current line is carried out , Pass 2 checks if there is another

line that is to be read. If there is then it loops back to the step where the line is analyzed and thiscontinues till an end of file is encountered.

7/29/2019 PraNaMa

6/28



2.6 Identifying Keywords and operators

An important step in designing any computer language is to identify the keywords[6] and

determine the rules to be followed while writing the program in the new language. The sections

that follow will give a brief description of the keywords.

2.7 Keywords

The following are the keywords that are to be used when writing a program.

2.8 Operators

Operators [6] form an important part of a programming language allowing us to perform

Mathematical and logical operations.

The following operators were identified as important and operators that ought to be

supported by PraNaMa.

7/29/2019 PraNaMa

7/28



7/29/2019 PraNaMa

8/28



Chapter 3

System Working

3.1 Basic Logic of Parsing:

Each .ks file will be parsed [5] twice. The idea of two parses is to separate file including,

function declaration from actual running of the program.

3.2 First Pass:

First parse handles four things.

later). This is analogous to the #include statement of a C program. This is done by appending

each line of the included file to the present file.

appending of new files).

3.3 Second Pass:

In Intrepreter.java the second pass is handled by: private int second pass (int scopeNum,

int new_scope) function. Here new scope is used for handling user defined functions (as

explained later) and scope Num is the scope number of the block which one wants to execute.

The first call to this function is done as: second pass (0,-1) ;

Here 0 is for global scope. The advantage of scopeNum lies in the fact that handling of

if-else and while constructs becomes easy to implement. The function returns 1 on successful

execution of function body or -1 when an error occurs.

7/29/2019 PraNaMa

9/28



In second_pass function each line in the file is examined and the type of the line is

determined. Depending on the type of the line appropriate functions are called to handle that

type. Following things are handled in second pass:

-else and while.

3.4 Declaration of Statements:

The rule in Ganlipi is that every variable must be declared before it is used., every

declaration statement can be of two

types one for each different data type supported by ganaka i.e. Akshara ( string) and Sankhye (

float ).

The type of variable declared is identified by first word in statement and appropriate

variable declaration functions are called.

User can declare a single variable or an array. A single variable can be initialized in

declaration but an array cannot be initialized.

The functions take care of allocation of memory for variables in memory table and

initialization of values.

3.5 Handling of Expressions:

An expression statement is one of the following :

a = 10 ;

b = a+10 ;

c = a*b ;

( Operators can be logical, relational or arithmetic )

7/29/2019 PraNaMa

10/28



a = "hello"+"how"+"are"+"you";

b = a+ nice weather ;

a = b = c = 0 ;

a ;

When a line in the file is judged to be an expression statement the Exrpession.Evaluate

function is called with the current line , memory object and current line number as

parameters.Expression.java and ExpressionVal.java are generated using Byacc/J[2] with calc.y

as input grammar and ExpressionLex.java is generated using Jflex[1] with calc.flex as input file.

The Evaluate function basically tries to reduce one of the many grammar rules. Once a reductionis done the corresponding action associated with the grammar rule is executed.

Besides these, other tokens include:

NUM = All real numbers.

KANNUM = Real numbers in Kannada.

CONSTANT = All characters within double quotes ( ).

7/29/2019 PraNaMa

11/28



NUMVAR = A variable which contains floating point value

STRVAR = A variable which contains string value.

Operator Precedence[6]:

Handling of implicit function: WRITE

The function is to print a statement given by the user. The statement may be a collection of String literals, literal Numbers, Sankhye and

Akshara variables.

Write functions analyses the statement access memory table to get values of variables andconverts everything into a string to be displayed on Output window.

On invoking write, the Output window GUI is initialized if it is not yet and the output isappended to display buffer.

7/29/2019 PraNaMa

12/28



Handling of implicit function: READ

This function is used to take input from user. Currently it only supports input to a single variable. This function analysis the type of variable to which the value is to be read and the value

given by users is stored into proper variables.

On invoking a READ, the Output window GUI is initialized if it is not yet and a

text-field is appended to output screen buffer so that user can enter data into text field. On

pressing enter-key on text field, the text field is set to non-editable mode and the program

continues. The Program does a blocked read for a READ command, i.e. on execution of

READ command, the interpreter halts execution until user gives input.

3.6 Handling of User defined functions

Ganaka also supports function's as C, but the slight variation is every function declaration

is followed by a keyword KAARYA (function) and every function call is to be called by an

implicit CALL api KARE.

Function declaration:

Every function declaration must start with keyword FUNCTION followed by return type

SANKHYE, AKSHARA or KHAALI ( void ) followed by a function name followed by

parameter list in parenthesis.

On parsing such a line, the function definition handler is called which puts all the details

of function in function table (explained later).

The logic of function must be enclosed in '{' '}' and the immediate block following the

function declaration is skipped.

3.7 Function call and return value handling:

The function call is executed by

1) CALL (function_name (parameter list ) );

or

2) CALL (function_name (parameter list ) , RVariable );

7/29/2019 PraNaMa

13/28



In method 1, the return value if any is ignored, where as in method 2 the return value is

expected and it is stored in RVariable.

Such statements are handled by CallHandler function, which take care of parsing all the

parameters to be sent, access their values, initialize the function block to execute the block.

The return values from function are stored in appropriate variables.

It is a mentionable fact that ganaka can handle recursion, and this is made possible by

storing the current state of function and initializing the function with new values for execution,

after the execution of function the old state of function is restored.

The states of functions are handled with scope value.

3.8 Handling of Constructs

The constructs if, else-if, else and while have a condition stipulated to them that the body

of the construct mustbe enclosed in flower brackets ({and}). This condition ensures that the

body of the construct lies in a new scope. Hence by identifying the scope the body can be

executed by calling the second_pass function with that scope as the parameter.

int ifHandler(String line) During second pass if the current line contains an if statement then it

invokes the ifHandler function passing the contents of the line in string format. The ifHandler

will handle if, else-if and else constructs. This function returns the line number from where theexecution must continue in second_pass function, in case of error it returns -1.

Algorithm used in if Handler function:

Step 1: From the string line extract conditional part of the if statement (withinparentheses).

Step 2: Call Expression.Evaluate function for the conditional expression and store theresult.

Step 3:If result is true do the following:o Find the scope of the next line (since blank lines are removed, the next line

contains { ) which is done using paranPos() function.

7/29/2019 PraNaMa

14/28



o Call second_pass function passing the scope number found in the previous step asparameter.

oSkip all the corresponding else-if and else constructs associated with this ifconstruct and return the line number from which execution should continue.

Step 4: If result is false do the following:o Skip the if (or else-if ) block of statements.o Check if next statement contains else-if or else keyword.o If it does not then return that statement.o If it contains else-if statement do the following:

Extract the expression part of else-if statement. Evaluate the expression. If it is true execute the corresponding block and skip the rest of

else-if ladder and else block and return the appropriate line number

from which execution will resume.

If it is false go back to step 4. If it contain else statement do the following: Find the scope of the else block. Execute this block using second_pass function. Return the next line after the else block.

int whileHandler(String line)

This function is used to handle the while construct. Like the ifHandler this function returns

the line number from where the execution must resume or -1 in case of an error. Likewise it also

takes the contents of the statement as a parameter.

Algorithm for whileHandler function:

Step 1: From the string line extract conditional part of the while statement (withinparentheses).

7/29/2019 PraNaMa

15/28



Step 2: Call Expression.Evaluate function for the conditional expression and store theresult.

Step 3: If result is true:

Step 4: If result is false return the line number immediately after the while loop.

3.9 The Memory Handling[6]:

All variables are stored in hash table for easy access, the hash table stores the variable

name along with score number as key and the value part is a set of variable name, value in string

and num format, scope num and other associated information.

A noteworthy feature is both the string and numvalue of a variable are stored which

makes coercing easier. On inserting a value into table, the number is coerced to string and it is

stored as string value and a string is parsed if it has a number and its values is stored

correspondingly , if it does not have a number, 0 is stored.

The memory handling also provides function to store, access and alter variables. Any

variable can be accessed by its name and the line number where it is used. Such an interface was

developed as an interpreter executes line by line and this interface would make variable access

easier.

Another reason for creation of separate memory handling functions is that, Variables

forms the soul of program, a variable is used in almost every line of program. Hence it was

necessary to build a flawless and efficient handling of variables.

The auto coercing feature in memory table adds an advantage in various functions of parser.

Example: WRITE handler.

7/29/2019 PraNaMa

16/28



3.10 The Function handling:

The function handling[4] mainly concentrates on handling data of functions. The function

table (a hash table) stores all detail of function. The function handler provides interface to insert

data regarding a function and also to access specific details regarding function whenever needed.

The access queries may be to check if a function exists, which line the function is defined, what

is the return type, what are the type of arguments the function expects and so on.

The function handler makes the task of call handler easy and structured.

7/29/2019 PraNaMa

17/28



Chapter 4

Usage guidelines

This section gives a description of various types of constants, data types, operators and control

statements available in PraNaMa. It also gives a brief description of how the language includes

function handling, array handling features in it.

The following series of steps are to be followed before learning any computer programming

language:

1. Learn the alphabet of the language.

2. Grouping characters from the alphabet set to form meaningful tokens.

3. Grouping meaningful tokens to form meaningful instructions.

4. Clustering the instructions in a meaningful and logical order to result in efficient computer

programs.

PraNaMa is a programming language concentrating mainly on providing a programming feature

in Kannada. Forming meaningful tokens here can lead to write efficient programs. The character

set of Kannada PraNaMa is as follows.

).

to of Kannada and 0 to 9 of English.

efer table 8 for symbols used.

\t), new line (\n), carriage return (\r) etc.

7/29/2019 PraNaMa

18/28



4.1 Tokens :

A token is a smallest basic unit of a program. The tokens can be classified into the following

groups :

might include _ underscore also. Example: , _etc

change when during the execution of the

program. PraNaMa has the following constants.

alled the floating

point constant.

constants.

stored, accessed or manipulated. The identifier which is used to store a value is called as

variable. The variable is always different from that of keywords and no variable name can be

same as that of a keyword.

eration. Example : + , - , * , /

etc

7/29/2019 PraNaMa

19/28



program. They are used along with well defined keywords or identifiers to give out special

functions to the program. Example : {} , [ ] , etc

4.2 Data types:

The data type[13] is a type of a variable that is stored in the memory while a variable is declared

in the program.

The general syntax for declaring variables of any datatype is as follows:

Datatype variable1 = value1 , variable2 = value2 ,, variable = valuen ;

PraNaMa is interpreted language and to give user a more generalized programming experience it

includes only 3 basic datatypes as follows :

) : this datatype is a keyword that is used to define numbers. The numbers can

be whole numbers, integers, floating point values. The data type is used to store the numeric

values in the memory. Example : , = , ;

() : this datatype is a keyword that is used to define strings. The variables

declared of this datatype store string values in them and can be used for string manipulationpurposes. PraNaMa considers even a single character as a string and there by provides wide

string handling capability to the language. Example : =" " ;

) : this datatype is a keyword that is used to declare array of mixed datatype

variables ( from above 2 types ). The array can include any values and can efficiently be used in

reference type passing etc.

4.3 Operators:

Operators[12] are the symbols that are used to specify the particular operation on a set of

variables. The operators can be classified into the following types:

7/29/2019 PraNaMa

20/28



Arithmetic Operators : The operators that are used to perform fundamental mathematical

operations like addition, subtraction, multiplication, division. PraNaMa accepts two operands for

a particular arithmetic operator and thus gives out the result. There is a precedencerule to be

followed by the operators which is BODMAS rule and can be explained as follows.

o Brackets have the first precedence.

o Of operators will have the second precedence.

o Division operator will have third precedence.

o Multiplication operator will have forth precedence.

o Addition operator has the fifth precedence.

o Subtraction operator has the sixth precedence.

Relational Operators : The relational operators are used to compare two operands and thus

return the result. The two operands can be variables or constants.

: As logical gates such as AND, OR and NOT whose ouput is 1 or 0, we

also have logical operators. After evaluation, expression consisting of logical operators results in

either true or false and hence such are called logical expressions.

into a memory location is called and assignment operator. Copying and storing into a memory

location is called assigning and hence the name. the assignment operator is denoted by = sign.

7/29/2019 PraNaMa

21/28



4.4 Control statements :

The order in which the statements are executed are called the control flow. The statements that

are used to control the flow of execution of the program are called control statements. These

statements alter the sequence of execution of the program. Based on the order in which the

statements are executed, the various control statements are classified as follows.

4.5 Sequential statements :

The programmer writes a sequence of statements to do a specific activity. All these statements in

a program are executed in order in which they appear in the program. These programming

statements that are executed sequentially, that is one after the other, are called sequential control

statements.

The following program skeleton shows the sequence of flow of statements.

4.6 Branching statements :

In sequential execution of statements the computer surpasses every line and interprets it.

Branching allows the skipping option of certain statements when its execution is not needed.

There are five types of branching statements in PraNaMa and can be implemented according to

the need.

1. The simple if statement( ) : It is a simple selection statement which allows for

skipping/executing of statements following it based on the condition mentioned to be true or

false.

Example :

7/29/2019 PraNaMa

22/28



2. The ifelse statement ( - - ) ): The ifelse statement is a selection statement.If the condition mentioned is true the if scope is executed and if the condition is false the else

scope is executed.

Example :

3. The return statement ( ): The return statement returns the datatype mentioned in the

function definition. Here the return statement can return the data types that are discussed already.

Example:

7/29/2019 PraNaMa

23/28



4.7 Loop statements :

The statements that allow the programmer to repeatedly execute a particular set of statements

until a condition is true is called as loop statements. PraNaMa includes a loop statement called

which is analogous to while loop in C.

( ( ): A set of statements might have to be repeated until a certain

condition is reached and in such cases where a definite value is not known in prior the while

statement is used. It allows the programmer to repeatedly execute a particular set of statements

until a condition is true.

Example

4.8 Arrays :

An array is a collection of data of same datatype. PraNaMa allows the user to add arrays in

his/her programs which help in storing similar data items in a set format. The arrays here can be

categorized into the following types :

1. Single dimensional arrays : The array which is a linear list of data items with only single array

index to refer to them is called single dimension arrays. PraNaMa allows for the declaration of

arrays with the following general syntax. Syntax : datatype variable[ index ] ;

Example

7/29/2019 PraNaMa

24/28



4.9 Functions :

The language PraNaMa provides the feature of splitting up the program into well defined

modules called functions (). The functions have a definite scope and canaccept parameters.

They can even return results which can be stored on specific memory locations. There are two

important elements for the functions in PraNaMa. They are as follows :

and a specific task is achieved. The function definition can be clearly understood with the

following example.

is made in PraNaMa with a suffix of a keyword called and then the function name with

the appropriate parameters. The function is thus called referring to the function table and

executed. The function call looks something like this.

7/29/2019 PraNaMa

25/28



Chapter 5

Screen Shots

1. Simple welcome program in PraNaMa.

2. Calculator program.

7/29/2019 PraNaMa

26/28



7/29/2019 PraNaMa

27/28



Chapter 6

Conclusion

PraNaMa is a scripting language interpreter which accepts input from a user and gives

the output in Kannada. The application is implemented on Java platform and parses the scripts

written in Kannada and generates results in the same language. The application allows the user to

enjoy the scripting comfort in his/her language.

PraNaMa is primarily meant for those who can read and write only in their native

language and those who are not well versed in English but are interested in learning computerprogramming. PraNaMa can be a stepping stone for those who do not know or know little

English and plan to learn programming, by introducing them to the basic concepts of

programming in their native language Kannada. Those who are used to conventional

programming can also write programs using PraNaMa provided they know Kannada.

The interpreter scans every line written in the program and executes. The language

has its own syntax which has to be followed when writing the script and is closely related to C

language syntax. All important features that are expected of a scripting language are provided by

the language, thus the user can use these small but powerful set of features to write programs and

make himself familiar with programming before diving into much more complex languages like

C, C++ and Java.

7/29/2019 PraNaMa

28/28



References

[1] www.sourceforge.jflex.net-scanner generator for Java

[2] http://byaccj.sourceforge.net/ - Yacc compatible parser

[3] http://www.google.com/transliterate/Kannada - Transliteration APIs

[4] www.sun.java.com- http://www.oracle.com/technetwork/java/api-141528.html

[5] The Principles of Compiler Design, 4th edition by Jeffery Ulmann et.al.

[6] The Complete Reference, Java by Herbert Schildt.

[7]http://www.health.qld.gov.au/multicultural/interpreters/guidelines_int.pdf - Working with

Interpreter Guidelines

[8]http://www.stanford.edu/class/cs242/slides/2005/scripting.pdf - Scripting Languages

[9] http://www.sensi.org/~ak/impit/studies/report.pdf - An overview of scripting languages

[10] http://en.wikipedia.org/wiki/Interpreter_(computing) Advantages and disadvantages of

using an Interpreter

[11] http://en.wikipedia.org/wiki/Scripting_language - Working Guidelines for Scripting

Languages

[12]http://en.wikipedia.org/wiki/Operator- Precedence and associativity of operators

[13]http://en.wikipedia.org/wiki/Data_type - Primitive and user-defined data types.

PraNaMa

Documents