Basics

This functionality allows you to use input as a regular expression. To see how regular expressions are different from simple search click here

Some notes and implementation details:

• Regular Expressions search is only enabled for Premium+ packages. For more information click here
• Before indexing, all text is converted to lower case, to make search easier.
• Each user can run only one regex at a time.
• Regex has to match the whole filename. That basically means that you will have to add .*{regex}.* most of the time. Also ^ and $ are implicit and removed if entered.
• The domain and protocol should not be matched with the regex. A url will be returned if the regex matches the filename part of the url (bold in the following example) http://somebucket.s3-eu-west-1.amazonaws.com/images/image.jpg
• . ? + * | { } [ ] ( ) " \ are special chatacters. To search for those special characters use a backslash \ for example .*\.jpg searches for files ending with .jpg. That means it will return filename.jpg bug will exclude somefilejpg
• Full documentation here: Regular Expressions

• .*backup.*
  • Finds everything that contains backup, like thebackup, backup2 _backup_ etc.
• .*2018[\-_\. ]11.*
  
  • Find everything related to November 2018
• .*dump.*(gz|tar|zip)
  • Find all files containing keyword "dump" and end with gz, tar, or zip
• backup.*
  • Find all files that BEGIN with backup
• .*backup
  • Find all files that END with backup.
• .*\.php
  • Find all files with php extension in the site
• 19[0-9]{2}
  
  • Text that contains 19 and then exactly 2 digits from 0-9
• .*"test.txt"
  • Everything in "" is literal and its not used as in the engine.

Regex Autocorrect

Our system is build in such a way, so that the input regex must match the complete filename, start to end in order to be returned.
Most people expect that if their regex matches partially in a filename this will be returned. Because not all users are advanced, we try to autocorrect the regular expressions, so that most results are returned.
We have choose, when we autocorrect the regular expression to leave the change is visible to the user, so that advanced users can see the change and adjust. The rules are:

• When the regex does not start with .* or ^(=start of string) we add .*.
• When the regex does not end with .* or $(=end of string) we add .*.

If you do not want the system to auto-correct the regular expression, you can use ^ and $, or check the "Do not autocorrect regex" option.