How to List Video Files with Specific Extensions in GNU Bash

To list video files with specific extensions like .mp4 and .avi in GNU Bash, use the find command combined with the -regex option. This approach allows you to filter files based on complex patterns rather than simple suffixes, providing a robust solution for managing large multimedia collections on Linux systems.

Understanding the Find Command Limitations

The standard find command is powerful, but its built-in options like -name or -iname only support shell-style wildcards (globbing). While *.mp4 works perfectly for a single extension, combining multiple extensions often requires chaining commands with -o (OR) operators or using grep afterwards. For example, finding both .mp4 and .avi files using only -name would look like this:

find /path/to/search -type f \( -name "*.mp4" -o -name "*.avi" \)

This method becomes unwieldy as the number of desired extensions grows. If you need to include .mkv, .wmv, and .flv, the command line gets long and difficult to maintain. Furthermore, globbing does not allow for more complex pattern matching, such as excluding hidden files or matching specific directory structures within the search path. The -regex option solves these issues by allowing full regular expression support directly within the find utility.

Using Regular Expressions with Find

GNU find supports the -regex flag, which matches the entire file path against a provided regular expression. It is crucial to understand that -regex matches the whole path, not just the filename. Therefore, your regular expression must account for the directory structure leading up to the file. To list all video files ending in .mp4 or .avi in the current directory and its subdirectories, you can use the following command:

find . -type f -regex '.*\.\(mp4\|avi\)'

Let's break down this regular expression:

• .: Matches any character (the start of the path).
• *: Matches zero or more of the preceding element (any characters in the path).
• \.: Escapes the dot to match a literal period character.
• \( ... \| ... \): Groups alternatives. In basic regular expressions (BRE), parentheses and the pipe symbol must be escaped with backslashes. This part matches either "mp4" or "avi".

If you are using GNU find version 4.3.3 or later, you can enable extended regular expressions with the -regextype option, which makes the syntax cleaner and more familiar to those used to Perl-compatible regular expressions:

find . -type f -regextype posix-extended -regex '.*\.(mp4|avi)'

Here, the parentheses and pipe symbol do not require escaping, making the pattern easier to read and write.

Handling Case Sensitivity

File extensions might appear in different cases, such as .MP4, .Mp4, or .AVI. By default, -regex is case-sensitive. To make the search case-insensitive, you can combine -regex with -iname logic, but find does not have a direct case-insensitive regex flag. Instead, you can modify the regex to explicitly match both upper and lower case letters:

find . -type f -regex '.*\.[mM][pP]4' -o -type f -regex '.*\.[aA][vV][iI]'

Alternatively, if you prefer a simpler approach and don't mind the slight performance cost, you can convert the output to lowercase or use grep with the -i flag after filtering by name, though this deviates from the pure regex approach. For most Linux distributions, ensuring your filenames follow standard casing conventions is best practice to avoid these complications.

Refining the Search Scope

When dealing with large media libraries, searching the entire filesystem is inefficient and potentially dangerous. Always specify a starting directory. Using . restricts the search to the current directory and its descendants. You can also exclude certain directories, such as hidden folders or system mounts, to speed up the process. For instance, if you want to ignore directories named cache or temp, you can add a -not condition:

find . -type d \( -name "cache" -o -name "temp" \) -prune -o -type f -regex '.*\.\(mp4\|avi\)' -print

The -prune action tells find to stop descending into matched directories, significantly improving performance on deep directory trees.

Integrating with Other Tools

Often, listing files is just the first step. You might want to count them, move them, or generate a report. The output of find with -regex can be piped to other commands. For example, to count the total number of video files:

find . -type f -regex '.*\.\(mp4\|avi\)' | wc -l

Or to copy them to a backup directory:

find . -type f -regex '.*\.\(mp4\|avi\)' -exec cp {} /backup/videos/ \;

Be cautious with -exec on large datasets, as it spawns a new process for each file. For better performance with many files, use + instead of \; to batch the arguments:

find . -type f -regex '.*\.\(mp4\|avi\)' -exec cp {} /backup/videos/ +

This technique ensures that your file management tasks remain efficient and scalable. For more information on handling edge cases in bash scripts, such as spaces in filenames, see our guide on Troubleshooting Bash Script with Suspected Issues of Spaces, Dashes, Filenames and Quoting.

Frequently Asked Questions

What is the difference between -name and -regex in find?

The -name option uses shell globbing patterns, which only match the filename component and support simple wildcards like * and ?. The -regex option matches the entire file path against a regular expression, allowing for complex logic, alternation, and precise control over the path structure.

How do I make the regex case-insensitive?

Globally, you cannot make -regex case-insensitive with a single flag. You must manually specify both uppercase and lowercase characters in the pattern, such as [mM] for 'm' or 'M'. Alternatively, you can use -iname for simple extensions, but it lacks the flexibility of regex for complex path matching.

Can I use -regex to match files in the root directory only?

No, because -regex matches the entire path, including directory prefixes. To match only files in the current directory, your regex must explicitly anchor to the start and ensure no slashes exist in the middle, which is complex. It is usually easier to use -maxdepth 1 with -name for shallow searches.

Why is my find command returning empty results?

This often happens because the regex does not account for the leading directory components. Remember that -regex matches the whole path string. Ensure your pattern starts with something like .* to consume the directory path before matching the filename extension.

Is there a performance penalty for using -regex?

Slightly, as regular expression evaluation is more computationally expensive than simple string comparison. However, the benefit of pruning directories with -prune usually outweighs the regex overhead. For massive datasets, consider using specialized tools like mlocate or fd which may offer faster indexing-based searches.