Coding Standards and Best Programming Practices
1. Introduction
Anybody can write code. With a few months of programming experience, you can write 'working
applications'. Making it work is easy, but doing it the right way requires more work, than just
making it work.
Believe it, majority of the programmers write 'working code', but not ‘good code'. Writing 'good
code' is an art and you must learn and practice it.
Everyone may have different definitions for the term ‘good code’. In my definition, the following
are the characteristics of good code.
o Reliable
o Maintainable
o Efficient
Most of the developers are inclined towards writing code for higher performance, compromising
reliability and maintainability. But considering the long term ROI (Return On Investment),
efficiency and performance comes below reliability and maintainability. If your code is not reliable
and maintainable, you (and your company) will be spending lot of time to identify issues, trying to
understand code etc throughout the life of your application.
2. Purpose of coding standards and best practices
To develop reliable and maintainable applications, you must follow coding standards and best practices.
The naming conventions, coding standards and best practices described in this document are compiled from
our own experience and by referring to various guidelines.
There are several standards exists in the programming industry. None of them are wrong or bad and you
may follow any of them. What is more important is, selecting one standard approach and ensuring that
everyone is following it.
3. Naming Conventions and Standards
Note :
The terms Pascal Casing and Camel Casing are used throughout this document.
Pascal Casing - First character of all words are Upper Case and other characters are lower case.
Example: BackColor
Camel Casing - First character of all words, except the first word are Upper Case and other characters are
lower case.
Example: backColor
Use Pascal casing for Class names
public class HelloWorld
{
...
}
Use Pascal casing for Method names
void SayHello(string name)
{
...
}
Use Camel casing for variables and method parameters
int totalCount = 0;
void SayHello(string name)
{
string fullMessage = "Hello " + name;
...
}
Use the prefix “I” with Camel Casing for interfaces ( Example: IEntity )
Do not use Hungarian notation to name variables.
In earlier days most of the programmers liked it - having the data type as a prefix for the variable
name and using m_ as prefix for member variables. Eg:
string m_sName;
int nAge;
However coding standards, this is not recommended. Usage of data type and m_ to represent member variables
should not be used. All variables should use camel casing.
Some programmers still prefer to use the prefix m_ to represent member variables, since there is no other
easy way to identify a member variable.
Use Meaningful, descriptive words to name variables. Do not use abbreviations.
Good:
string address
int salary
Not Good:
string nam
string addr
int sal
Do not use single character variable names like i, n, s etc. Use names like index, temp
One exception in this case would be variables used for iterations in loops:
for ( int i = 0; i < count; i++ )
{
...
}
If the variable is used only as a counter for iteration and is not used anywhere else in the loop, many people
still like to use a single char variable (i) instead of inventing a different suitable name.
Do not use underscores (_) for local variable names.
All member variables must be prefixed with underscore (_) so that they can be identified from
other local variables.
Do not use variable names that resemble keywords.
Prefix boolean variables, properties and methods with “is” or similar prefixes.
Ex: private bool _isFinished
Namespace names should follow the standard pattern
<company name>.<product name>.<top level module>.<bottom level module>
Use appropriate prefix for the UI elements so that you can identify them from the rest of the variables.
There are 2 different approaches recommended here.
a. Use a common prefix ( ui_ ) for all UI elements. This will help you group all of the UI elements together
and easy to access all of them from the intellisense.
b. Use appropriate prefix for each of the ui element. A brief list is given below. Since .NET has given
several controls, you may have to arrive at a complete list of standard prefixes for each of the controls
(including third party controls) you are using.
Control Prefix
Label lbl
TextBox txt
DataGrid dtg
Button btn
ImageButton imb
Hyperlink hlk
DropDownList ddl
ListBox lst
DataList dtl
Repeater rep
Checkbox chk
CheckBoxList cbl
RadioButton rdo
RadioButtonList rbl
Image img
Panel pnl
PlaceHolder phd
Table tbl
Validators val
File name should match with class name.
For example, for the class HelloWorld, the file name should be [Link] (or, [Link] or, [Link],
etc)
Use Pascal Case for file names.
4. Indentation and Spacing
Use TAB for indentation. Do not use SPACES. Define the Tab size as 4.
Comments should be in the same level as the code (use the same level of indentation).
Good:
// Format a message and display
string fullMessage = "Hello " + name;
DateTime currentTime = [Link];
string message = fullMessage + ", the time is : " +
[Link]();
[Link] ( message );
Not Good:
// Format a message and display
string fullMessage = "Hello " + name;
DateTime currentTime = [Link];
string message = fullMessage + ", the time is : " +
[Link]();
[Link] ( message );
Curly braces ( {} ) should be in the same level as the code outside the braces.
Use one blank line to separate logical groups of code.
Good:
bool SayHello ( string name )
{
string fullMessage = "Hello " + name;
DateTime currentTime = [Link];
string message = fullMessage + ", the time is : " +
[Link]();
[Link] ( message );
if ( ... )
{
// Do something
// ...
return false;
}
return true;
}
Not Good:
bool SayHello (string name)
{
string fullMessage = "Hello " + name;
DateTime currentTime = [Link];
string message = fullMessage + ", the time is : " +
[Link]();
[Link] ( message );
if ( ... )
{
// Do something
// ...
return false;
}
return true;
}
There should be one and only one single blank line between each method inside the class.
The curly braces should be on a separate line and not in the same line as if, for etc.
Good:
if ( ... )
{
// Do something
}
Not Good:
if ( ... ) {
// Do something
}
Use a single space before and after each operator and brackets.
Good:
if ( showResult == true )
{
for ( int i = 0; i < 10; i++ )
{
//
}
}
Not Good:
if(showResult==true)
{
for(int i= 0;i<10;i++)
{
//
}
}
Keep private member variables, properties and methods in the top of the file and public members
in the bottom
5. Good Programming practices
Avoid writing very long methods. A method should typically have 1~25 lines of code. If a method
has more than 25 lines of code, you must consider re factoring into separate methods.
Method name should tell what it does. Do not use mis-leading names. If the method name is
obvious, there is no need of documentation explaining what the method does.
Good:
void SavePhoneNumber ( string phoneNumber )
{
// Save the phone number.
}
Not Good:
// This method will save the phone number.
void SaveDetails ( string phoneNumber )
{
// Save the phone number.
}
A method should do only 'one job'. Do not combine more than one job in a single method, even if
those jobs are very small.
Good:
// Save the address.
SaveAddress ( address );
// Send an email to the supervisor to inform that the address is
updated.
SendEmail ( address, email );
void SaveAddress ( string address )
{
// Save the address.
// ...
}
void SendEmail ( string address, string email )
{
// Send an email to inform the supervisor that the address is changed.
// ...
}
Not Good:
// Save address and send an email to the supervisor to inform
that
// the address is updated.
SaveAddress ( address, email );
void SaveAddress ( string address, string email )
{
// Job 1.
// Save the address.
// ...
// Job 2.
// Send an email to inform the supervisor that the address
is changed.
// ...
}
Always watch for unexpected values. For example, if you are using a parameter with 2 possible values,
never assume that if one is not matching then the only possibility is the other value.
Good:
If ( memberType == [Link] )
{
// Registered user… do something…
}
else if ( memberType == [Link] )
{
// Guest user... do something…
}
else
{
// Un expected user type. Throw an exception
throw new Exception (“Un expected value “ +
[Link]() + “’.”)
// If we introduce a new user type in future, we can easily
find
// the problem here.
}
Not Good:
If ( memberType == [Link] )
{
// Registered user… do something…
}
else
{
// Guest user... do something…
// If we introduce another user type in future, this code will
// fail and will not be noticed.
}
Do not hardcode numbers. Use constants instead. Declare constant in the top of the file and use it in
your code.
However, using constants are also not recommended. You should use the constants in the config file or
database so that you can change it later. Declare them as constants only if you are sure this value will never
need to be changed.
Do not hardcode strings. Use resource files.
Convert strings to lowercase or upper case before comparing. This will ensure the string will
match even if the string being compared has a different case.
if ( [Link]() == “john” )
{
//…
}
Use [Link] instead of “”
Good:
If ( name == [Link] )
{
// do something
}
Not Good:
If ( name == “” )
{
// do something
}
Avoid using member variables. Declare local variables wherever necessary and pass it to other
methods instead of sharing a member variable between methods. If you share a member variable
between methods, it will be difficult to track which method changed the value and when.
Use enum wherever required. Do not use numbers or strings to indicate discrete values.
Good:
enum MailType
{
Html,
PlainText,
Attachment
}
void SendMail (string message, MailType mailType)
{
switch ( mailType )
{
case [Link]:
// Do something
break;
case [Link]:
// Do something
break;
case [Link]:
// Do something
break;
default:
// Do something
break;
}
}
Not Good:
void SendMail (string message, string mailType)
{
switch ( mailType )
{
case "Html":
// Do something
break;
case "PlainText":
// Do something
break;
case "Attachment":
// Do something
break;
default:
// Do something
break;
}
}
Do not make the member variables public or protected. Keep them private and expose
public/protected Properties.
The event handler should not contain the code to perform the required action. Rather call another
method from the event handler.
Do not programmatically click a button to execute the same action you have written in the button
click event. Rather, call the same method which is called by the button click event handler.
Never hardcode a path or drive name in code. Get the application path programmatically and use
relative path.
Never assume that your code will run from drive "C:". You may never know, some users may run
it from network or from a "Z:".
In the application start up, do some kind of "self-check" and ensure all required files and
dependencies are available in the expected locations. Check for database connection in startup,
if required. Give a friendly message to the user in case of any problems.
If the required configuration file is not found, application should be able to create one with default
values.
If a wrong value found in the configuration file, application should throw an error or give a
message and also should tell the user what are the correct values.
Error messages should help the user to solve the problem. Never give error messages like "Error
in Application", "There is an error" etc. Instead give specific messages like "Failed to update
database. Please make sure the login id and password are correct."
When displaying error messages, in addition to telling what is wrong, the message should also
tell what the user should do to solve the problem. Instead of message like "Failed to update
database.” suggest what should the user do: "Failed to update database. Please make sure the
login id and password are correct."
Show short and friendly message to the user. But log the actual error with all possible
information. This will help a lot in diagnosing problems.
Do not have more than one class in a single file.
Avoid having very large files. If a single file has more than 1000 lines of code, it is a good
candidate for refactoring. Split them logically into two or more classes.
Avoid passing too many parameters to a method. If you have more than 4~5 parameters, it is a
good candidate to define a class or structure.
If you have a method returning a collection, return an empty collection instead of null, if you have
no data to return. For example, if you have a method returning an ArrayList, always return a valid
ArrayList. If you have no items to return, then return a valid ArrayList with 0 items. This will make
it easy for the calling application to just check for the “count” rather than doing an additional check
for “null”.
Logically organize all your files within appropriate folders. Use 2 level folder hierarchies. You can
have up to 10 folders in the root folder and each folder can have up to 5 sub folders. If you have
too many folders than cannot be accommodated with the above mentioned 2 level hierarchy, you
may need re factoring into multiple assemblies.
Make sure you have a good logging class which can be configured to log errors, warning or
traces. If you configure to log errors, it should only log errors. But if you configure to log traces, it
should record all (errors, warnings and trace). Your log class should be written such a way that in
future you can change it easily to log to Windows Event Log, SQL Server, or Email to
administrator or to a File etc without any change in any other part of the application. Use the log
class extensively throughout the code to record errors, warning and even trace messages that
can help you trouble shoot a problem.
If you are opening database connections, sockets, file stream etc, always close them in the
finally block. This will ensure that even if an exception occurs after opening the connection, it
will be safely closed in the finally block.
Declare variables as close as possible to where it is first used. Use one variable declaration per
line.
Use StringBuilder class instead of String when you have to manipulate string objects in a loop.
The String object works in weird way. Each time you append a string, it is actually discarding the
old string object and recreating a new object, which is a relatively expensive operations.
Consider the following example:
public string ComposeMessage (string[] lines)
{
string message = [Link];
for (int i = 0; i < [Link]; i++)
{
message += lines [i];
}
return message;
}
In the above example, it may look like we are just appending to the string object ‘message’. But
what is happening in reality is, the string object is discarded in each iteration and recreated and
appending the line to it.
If your loop has several iterations, then it is a good idea to use StringBuilder class instead of
String object.
See the example where the String object is replaced with StringBuilder.
public string ComposeMessage (string[] lines)
{
StringBuilder message = new StringBuilder();
for (int i = 0; i < [Link]; i++)
{
[Link]( lines[i] );
}
return [Link]();
}
6. Comments
Good and meaningful comments make code more maintainable. However,
Do not write comments for every line of code and every variable declared.
Use // or /// for comments. Avoid using /* … */
Write comments wherever required. But good readable code will require very less comments. If
all variables and method names are meaningful, that would make the code very readable and will
not need many comments.
7. Exception Handling
Never do a 'catch exception and do nothing'. If you hide an exception, you will never know if the
exception happened or not. Lot of developers uses this handy method to ignore non significant
errors. You should always try to avoid exceptions by checking all the error conditions
programmatically. In any case, catching an exception and doing nothing is not allowed. In the
worst case, you should log the exception and proceed.
In case of exceptions, give a friendly message to the user, but log the actual error with all
possible details about the error, including the time it occurred, method and class name etc.
Always catch only the specific exception, not generic exception.
Good:
void ReadFromFile ( string fileName )
{
try
{
// read from file.
}
catch (FileIOException ex)
{
// log error.
// re-throw exception depending on your case.
throw;
}
}
Not Good:
void ReadFromFile ( string fileName )
{
try
{
// read from file.
}
catch (Exception ex)
{
// Catching general exception is bad... we will never know
whether
// it was a file error or some other error.
// Here you are hiding an exception.
// In this case no one will ever know that an exception
happened.
return "";
}
}
No need to catch the general exception in all your methods. Leave it open and let the application crash. This
will help you find most of the errors during development cycle. You can have an application level (thread
level) error handler where you can handle all general exceptions. In case of an 'unexpected general error',
this error handler should catch the exception and should log the error in addition to giving a friendly
message to the user before closing the application, or allowing the user to 'ignore and proceed'.
When you re throw an exception, use the throw statement without specifying the original exception. This
way, the original call stack is preserved.
Good:
catch
{
// do whatever you want to handle the exception
throw;
}
Not Good:
catch (Exception ex)
{
// do whatever you want to handle the exception
throw ex;
}
Do not write try-catch in all your methods. Use it only if there is a possibility that a specific exception may
occur and it cannot be prevented by any other means. For example, if you want to insert a record if it does
not already exists in database, you should try to select record using the key. Some developers try to insert a
record without checking if it already exists. If an exception occurs, they will assume that the record already
exists. This is strictly not allowed. You should always explicitly check for errors rather than waiting for
exceptions to occur. On the other hand, you should always use exception handlers while you communicate
with external systems like network, hardware devices etc. Such systems are subject to failure anytime and
error checking is not usually reliable. In those cases, you should use exception handlers and try to recover
from error.
Do not write very large try-catch blocks. If required, write separate try-catch for each task you perform and
enclose only the specific piece of code inside the try-catch. This will help you find which piece of code
generated the exception and you can give specific error message to the user.
Write your own custom exception classes if required in your application. Do not derive your custom
exceptions from the base class SystemException. Instead, inherit from ApplicationException.
