Scanner vs. BufferedReader

Question

As far I know, the two most common methods of reading character-based data from a file in Java is using Scanner or BufferedReader. I also know that the BufferedReader reads files efficiently by using a buffer to avoid physical disk operations.

My questions are:

• Does Scanner perform as well as BufferedReader?
• Why would you choose Scanner over BufferedReader or vice versa?

Personalty

I generally also use Scanner for reading from standard in ('Scanner in = new Scanner(System.in)' feels much cleaner). Not sure if that's actually less efficient, but since reading from std in is blocking, I can't imagine the efficiency of Scanner would be the issue. – Roman

Answer 1

Scanner is used for parsing tokens from the contents of the stream while BufferedReader just reads the stream and does not do any special parsing.

In fact you can pass a BufferedReader to a scanner as the source of characters to parse.

Song

BufferedReader is synchronized and Scanner is not, so its up to you to decide. – Tantalic

I know this topic is old, but I have had mixed results among operating systems using BufferedReader when trying to slurp up content from the streams provided by Process (i.e. capturing output of an external command). Once I changed my code to use Scanner instead, as noted in a separate answer, things started behaving consistently and as expected. – Haver

@Tantalic But Scanner depends ultimately on something else for its input, which may well be synchronized. – Shuffleboard

If the token is a line, then BufferedReader can read tokens. – Eipper

Answer 2

In currently latest JDK 18 release/build (b37), the Scanner has a smaller buffer (1024 chars) as opposed to the BufferedReader (8192 chars), but it's more than sufficient.

As to the choice, use the Scanner if you want to parse the file, use the BufferedReader if you want to read the file line by line. Also see the introductory text of their aforelinked API documentations.

• Parsing = interpreting the given input as tokens (parts). It's able to give back you specific parts directly as int, string, decimal, etc. See also all those nextXxx() methods in Scanner class.
• Reading = dumb streaming. It keeps giving back you all characters, which you in turn have to manually inspect if you'd like to match or compose something useful. But if you don't need to do that anyway, then reading is sufficient.

Elana

Nice one. Thanks for the buffer tip. Was looking for it all along as native reads are extremely expensive. – Forta

@Elana Ok I already used, readInt(); readFloat(); etc. Now got what is the mean of parsing. and BalusC can you give little time to me just 10minutes in chat room, I want to ask little about buffered, how it works. – Pastille

What is I wrap BufferedReader in Scanner's constructor? is this a good idea? – Pol

Scanner’s buffer will be expanded as needed for the pattern matching. So if you want a larger buffer, you only need to invoke, e.g. findWithinHorizon("\\z", 8192), on it and afterwards, it will use a buffer with a capacity of 8192 chars (or the entire file if it is smaller than that). – Nimiety

Answer 3

See this link, following is quoted from there:

A BufferedReader is a simple class meant to efficiently read from the underling stream. Generally, each read request made of a Reader like a FileReader causes a corresponding read request to be made to underlying stream. Each invocation of read() or readLine() could cause bytes to be read from the file, converted into characters, and then returned, which can be very inefficient. Efficiency is improved appreciably if a Reader is warped in a BufferedReader.

BufferedReader is synchronized, so read operations on a BufferedReader can safely be done from multiple threads.

A scanner on the other hand has a lot more cheese built into it; it can do all that a BufferedReader can do and at the same level of efficiency as well. However, in addition a Scanner can parse the underlying stream for primitive types and strings using regular expressions. It can also tokenize the underlying stream with the delimiter of your choice. It can also do forward scanning of the underlying stream disregarding the delimiter!

A scanner however is not thread safe, it has to be externally synchronized.

The choice of using a BufferedReader or a Scanner depends on the code you are writing, if you are writing a simple log reader Buffered reader is adequate. However if you are writing an XML parser Scanner is the more natural choice.

Even while reading the input, if want to accept user input line by line and say just add it to a file, a BufferedReader is good enough. On the other hand if you want to accept user input as a command with multiple options, and then intend to perform different operations based on the command and options specified, a Scanner will suit better.

Kathlyn

"A scanner on the other hand has a lot more cheese built into it; it can do all that a BufferedReader can do and at the same level of efficiency as well." Don't agree, BufferedReader is a bit faster as compared to Scanner because Scanner does parsing of input data and BufferedReader simply reads sequence of characters. – Humor

Answer 4

1. BufferedReader has significantly larger buffer memory than Scanner. Use BufferedReader if you want to get long strings from a stream, and use Scanner if you want to parse specific type of token from a stream.

2. Scanner can use tokenize using custom delimiter and parse the stream into primitive types of data, while BufferedReader can only read and store String.

3. BufferedReader is synchronous while Scanner is not. Use BufferedReader if you're working with multiple threads.

4. Scanner hides IOException while BufferedReader throws it immediately.

Thanatopsis

Answer 5

I suggest to use BufferedReader for reading text. Scanner hides IOException while BufferedReader throws it immediately.

Squirm

Answer 6

The difference between BufferedReader and Scanner are following:

1. BufferedReader is synchronized but Scanner is not synchronized.
2. BufferedReader is thread-safe but Scanner is not thread-safe.
3. BufferedReader has larger buffer memory but Scanner has smaller buffer memory.
4. BufferedReader is faster but Scanner is slower in execution.
5. Code to read a line from the console:

BufferedReader:

InputStreamReader isr = new InputStreamReader(System.in);
BufferedReader br = new BufferedReader(isr);
String st = br.readLine();
       
// You can make the object InputStreamReader object inside the BufferReader method.
BufferReader br = new BufferedReader(InputStreamReader(System.in));
String st = br.readLine();

// You can even inspect the type of the input stream manually by using Parse method which accepts string parameter.
int x = Integer.parseInt(br.readLine());

// Or you can pass the object directly.
int x = Integer.parseInt(st);

Scanner:

Scanner sc = new Scanner(System.in);
String st = sc.nextLine();

Cafeteria

Answer 7

The differences between BufferedReader and Scanner are:

1. BufferedReader reads data, but Scanner parses data.
2. You can only read String using BufferedReader, using Scanner you can read to different data types like int.
3. BufferedReader is older than Scanner, it was added on JDK 1.1, while Scanner was added on JDK 5 release.
4. The buffer size of BufferedReader is larger (8KB) as compared to Scanner's 1KB.
5. BufferedReader is more suitable for reading files with long String, while Scanner is more suitable for reading small user input from command prompt.
6. BufferedReader is synchronized, while Scanner is not, which means you cannot share Scanner among multiple threads.
7. BufferedReader is faster than Scanner because it doesn't spend time on parsing.
8. BufferedReader is a bit faster as compared to Scanner.
9. BufferedReader is from java.io package, while Scanner is from java.util package.

On basis of the points we can select our choice.

Thanks for reading!

Cotquean

Answer 8

The Main Differences:

1. Scanner

---

• Simple text scanner which can parse primitive types and strings using regular expressions.
• Scanner breaks its input into tokens using a delimiter pattern, which by default matches whitespace. The resulting tokens may then be converted into values of different types using the various next methods.

Example:

 String input = "1 fish 2 fish red fish blue fish";
 Scanner s = new Scanner(input).useDelimiter("\\s*fish\\s*");
 System.out.println(s.nextInt());
 System.out.println(s.nextInt());
 System.out.println(s.next());
 System.out.println(s.next());
 s.close(); 

prints the following output:

1
2
red
blue

The same output can be generated with this code, which uses a regular expression to parse all four tokens at once:

 String input = "1 fish 2 fish red fish blue fish";

 Scanner s = new Scanner(input);
 s.findInLine("(\\d+) fish (\\d+) fish (\\w+) fish (\\w+)");
 MatchResult result = s.match();
 for (int i = 1; i <= result.groupCount(); i++) {
     System.out.println(result.group(i));
 }
 s.close();

---

2. BufferedReader:

• Reads text from a character-input stream, buffering characters so as to provide for the efficient reading of characters, arrays, and lines.

• The buffer size may be specified, or the default size may be used. The default is large enough for most purposes.

In general, each read request made of a Reader causes a corresponding read request to be made of the underlying character or byte stream. It is therefore advisable to wrap a BufferedReader around any Reader whose read() operations may be costly, such as FileReaders and InputStreamReaders. For example,

BufferedReader in = new BufferedReader(new FileReader("foo.in"));
 

will buffer the input from the specified file. Without buffering, each invocation of read() or readLine() could cause bytes to be read from the file, converted into characters, and then returned, which can be very inefficient. Programs that use DataInputStreams for textual input can be localized by replacing each DataInputStream with an appropriate BufferedReader.

Source used: https://docs.oracle.com

Cassiecassil

Answer 9

There are different ways of taking input in java like:

1) BufferedReader 2) Scanner 3) Command Line Arguments

BufferedReader Read text from a character-input stream, buffering characters so as to provide for the efficient reading of characters, arrays, and lines.

Where Scanner is a simple text scanner which can parse primitive types and strings using regular expressions.

if you are writing a simple log reader Buffered reader is adequate. if you are writing an XML parser Scanner is the more natural choice.

For more information please refer:

http://java.meritcampus.com/t/240/Bufferedreader?tc=mm69

Glenine

Why are you bringing up command line? Then you forgot env variables, config files, the preferences api, network resourcs, ... – Eipper

Answer 10

The answer below is taken from Reading from Console: JAVA Scanner vs BufferedReader

When read an input from console, there are two options exists to achieve that. First using Scanner, another using BufferedReader. Both of them have different characteristics. It means differences how to use it.

Scanner treated given input as token. BufferedReader just read line by line given input as string. Scanner itself provides parsing capabilities just like nextInt(), nextFloat().

But, what is others differences between?

• Scanner treated given input as token. BufferedReader as stream line/String.
• Scanner tokenized given input using regex. Using BufferedReader must write extra code.
• BufferedReader faster than Scanner *point no. 2
• Scanner isn’t synchronized, BufferedReader synchronized

Scanner came with since JDK 1.5 and higher.

When should use Scanner, or Buffered Reader?

Look at the main differences between both of them, one using tokenized, others using stream line. When you need parsing capabilities, use Scanner instead. But, I am more comfortable with BufferedReader. When you need to read data from a File, use BufferedReader, because it uses buffer memory when it reads a file, and that reduces physical drive usage. Or you can use BufferedReader as input to Scanner.

Harilda

Answer 11

1. BufferedReader will probably give you better performance (because Scanner is based on InputStreamReader, look sources). oops, for reading data from files it uses nio. When I tested nio performance against BufferedReader performance for big files nio shows a bit better performance.
2. For reading data from a file try Apache Commons IO.

Per

Answer 12

I prefer Scanner because it doesn't throw checked exceptions and therefore it's usage results in a more streamlined code.

Guenzi