No announcement yet.

Metadata based file formats

  • Filter
  • Time
  • Show
Clear All
new posts

  • Metadata based file formats

    We have some legacy stuff going on and have to deal with many files which in fact have different types from a semantic point of view but have no extension or share a common extension.

    As far as i can tell it is currently not possible to tell Beyond Compare to investigate the file itself and derive the format from internal metadata / magic numbers but the only possibility is to use distinct file extensions.
    Or is there a way?

  • #2
    Sorry, it isn't possible to select file format based on file contents. File formats are selected based on file mask, unless no mask matches. If no file format mask matches a file then Beyond Compare will look at file contents to determine if a file is binary or text and open it in the appropriate viewer.

    Masks don't have to match just on file extension, they can match on the full filename, so if there's a common part of the name for each file type, you can use that in the mask. Example: Define a format named format1 with mask "format1*.*" to open any file with a name that starts with format1.

    After a file is opened you can also select the format dropdown below the left and right path edits in the Text Compare to force a specific format.
    Chris K Scooter Software


    • #3
      Hi Chris
      the filenames are almost random and there are thousands of them :/ Manually selecting the format is unfortunately not an option.
      Is it a feature which might be implemented in future versions of Beyond Compare?


      • #4

        Detecting on other information (such as metadata and first line) is something on our wishlist, but not a feature we'll likely be able to tackle short-term. If you have specific example files and how they should be detected, email to with a link back to this forum thread for our reference, and I can add those notes to our wishlist entry.
        Aaron P Scooter Software


        • #5
          Hi Aaron
          thanks for the offer. These formats might not be worthy to be hardcoded into your software as they're our own fileformats which unfortunately also change or grow infrequently. The only option to solve this i could see is to provide some sort of script/regex based header matching specification and/or hook additionally to the file extensions.