android - Parse file name from URL before downloading the file

Question

I'm downloading an ePub file from a URL.

Now I want to implement a mechanism by which if user tries to re-download the same file, he should get warning/error message and that file should not be downloaded again.

To implement this, I need to check the name of the file present in my library with the name of the file user is trying to download.

But I just have this download link, and not the file name.

How to get the name of the file before download in order to compare it with the existing file?

Answer 1

In android you can use the guessFileName() method:

URLUtil.guessFileName(url, null, null)

Alternatively, a simplistic solution in Java could be:

String fileName = url.substring(url.lastIndexOf('/') + 1);

(Assuming your url is in the format: http://xxxxxxxxxxxxx/filename.ext)

UPDATE March 23, 2018

This question is getting lots of hits and someone commented my 'simple' solution does not work with certain urls so I felt the need to improve the answer.

In case you want to handle more complex url pattern, I provided a sample solution below. It gets pretty complex quite quickly and I'm pretty sure there are some odd cases my solution still can't handle but nevertheless here it goes:

public static String getFileNameFromURL(String url) {
    if (url == null) {
        return "";
    }
    try {
        URL resource = new URL(url);
        String host = resource.getHost();
        if (host.length() > 0 && url.endsWith(host)) {
            // handle ...example.com
            return "";
        }
    }
    catch(MalformedURLException e) {
        return "";  
    }

    int startIndex = url.lastIndexOf('/') + 1;
    int length = url.length();

    // find end index for ?
    int lastQMPos = url.lastIndexOf('?');
    if (lastQMPos == -1) {
        lastQMPos = length; 
    }

    // find end index for #
    int lastHashPos = url.lastIndexOf('#');
    if (lastHashPos == -1) {
        lastHashPos = length;   
    }

    // calculate the end index
    int endIndex = Math.min(lastQMPos, lastHashPos);
    return url.substring(startIndex, endIndex);
}

This method can handle these type of input:

Input: "null" Output: ""
Input: "" Output: ""
Input: "file:///home/user/test.html" Output: "test.html"
Input: "file:///home/user/test.html?id=902" Output: "test.html"
Input: "file:///home/user/test.html#footer" Output: "test.html"
Input: "http://example.com" Output: ""
Input: "http://www.example.com" Output: ""
Input: "http://www.example.txt" Output: ""
Input: "http://example.com/" Output: ""
Input: "http://example.com/a/b/c/test.html" Output: "test.html"
Input: "http://example.com/a/b/c/test.html?param=value" Output: "test.html"
Input: "http://example.com/a/b/c/test.html#anchor" Output: "test.html"
Input: "http://example.com/a/b/c/test.html#anchor?param=value" Output: "test.html"

You can find the whole source code here: https://ideone.com/uFWxTL