The java. URL class in Java is a built-in library that offers multiple methods to access and manipulate data on the internet. In this case, we will be using the openStream function of the URL class. The method signature for the openStream function is:.
The openStream function works on an object of the URL class. The URL class opens up a connection to the given URL and the openStream method returns an input stream which is used to read data from the connection.
These classes are used for reading from a file and writing to it, respectively. The contents are read as bytes and copied to a file in the local directory using the FileOutputStream.
To lower the number of lines of code we can use the Files class available from Java 7. The Files class contains methods that read all the bytes at once and then copies it into another file.
Here is how you can use it:. Java NIO is an alternative package to handle networking and input-output operations in Java. The main advantage that the Java NIO package offers is that it's non-blocking, and has channeling and buffering capabilities. When we use the Java IO library we work with streams that read data byte by byte. However, the Java NIO package uses channels and buffers.
The buffering and channeling capabilities allow the system to copy contents from a URL directly into the intended file without needing to save the bytes in application memory, which would be an intermediary step. The ability to work with channels boosts performance. The downloaded contents will be transferred to a file on the local system via the corresponding file channel.
After defining the file channel we will use the transferFrom method to copy the contents read from the readChannel object to the file destination using the writeChannel object. The transferFrom and transferTo methods are much more efficient than working with streams using a buffer. The transfer methods enable us to directly copy the contents of the file system cache to the file on the system.
Thus direct channeling restricts the number of context switches required and enhances the overall code performance. Now, in the following sections, we will be looking at ways to download files from a URL using third-party libraries instead of core Java functionality components.
Now you may be thinking why would we use this when Java has its own set of libraries to handle IO operations. However, Apache Commons IO overcomes the problem of code rewriting and helps avoid writing boilerplate code. In order to start using the Apache Commons IO library, you will need to download the jar files from the official website. When you are done downloading the jar files, you need to add them to use them. If you are using an Integrated Development Environment IDE such as Eclipse , you will need to add the files to the build path of your project.
There is only a single line of code required to download a file, which looks like:. The connection and read timeouts convey the permissible time for which either the connection may stay idle or reading from the URL may stop.
We will use the copy inputStream, fileOS method to download a file into the local system. Check out our hands-on, practical guide to learning Git, with best-practices, industry-accepted standards, and included cheat sheet. Stop Googling Git commands and actually learn it! The function returns the number of bytes copied.
If the value of the variable i is -1, then it indicates that the contents of the file are over 2GB. I'm not familiar with ImageIO, but I guess it has to make some additional work. Did you try increasing buffer size in solution 2? Have you tried the accepted anwser here: stackoverflow.
I'm surprised that the first version would be faster, considering it's decoding and reencoding the image. For a standard library solution, I'd go with Files. It is impossible that 1 is faster. How did you test it? Did you use the same image for 2 and 2? Did you eliminate the effects of caching? Add a comment. Active Oldest Votes. Improve this answer. There's no speed difference between this and your original 2 , except whatever is caused by the buffer size, which I would increase to at least bytes.
EJP Yes but I think 3 is far more elegant than the 2. Less code and more compact! What do you think? Sign up or log in Sign up using Google. Sign up using Facebook. Sign up using Email and Password. Post as a guest Name. Email Required, but never shown. The Overflow Blog. Podcast what if you could invest in your favorite developer?
Who owns this outage? Building intelligent escalation chains for modern SRE. Featured on Meta. Now live: A fully responsive profile. Reducing the weight of our footer. Linked 2.
0コメント