No announcement yet.

Not recognizing Word docx zipped contents

  • Filter
  • Time
  • Show
Clear All
new posts

  • Not recognizing Word docx zipped contents


    I am writing an application that reads a Word dotx template file, modifies the content, and then outputs a Word docx file. I am using MiniZip (version 1.1, Feb 14, 2010) to uncompress the dotx file and later create the docx file using the deflate method. Word can open these docx files without any problems, however BC4 cannot compare them (it is as though they are not being uncompressed before comparison).

    I've attached a generated file (zipped because I can't upload a docx file). Any ideas?

    Attached Files

  • #2

    Thanks for the sample file. I haven't narrowed down exactly what is injecting this issue with the conversion process, but there is something about how this file was generated that is causing a problem with the conversion process. If I open the file in Word, perform a very small edit and re-save, the re-saved version opens in BC4 without issue.

    I did also test our Word Extended format, which uses your install of Word to perform the conversion. You can download the alternate Word version here:
    Aaron P Scooter Software


    • #3
      Hi Aaron,

      Using the extended format has fixed the issue. Not only that, but the extended format does a better job of comparing table content. Colour me happy. Thank you!

      I'd be even happier if you could figure out why the extended format is necessary. As a test, I've used 7zip to extract the contents of the docx and then re-archived it using the deflate method. BC4 is then able to read the re-archived file, but as mentioned above the the extended format does a better job of comparing the file content. So it comes down to BC4 not liking the Deflate method used by MiniZip (and yes, I tried various compression strengths including no-compression). I'll look around for a different source of compression software.

      Thanks again,


      • #4
        Hi Brian,

        BC's built-in docx support is failing because it expects the folder structure for the filenames within the zip to use '/' as the path delimiter and your files are using '\'. The zip file format spec does require '/' delimiters, but most archive utilities, including BC's folder compare, will handle either direction, and obviously MS Word does too.

        We have a fix in a future version of BC, but for now you should see if you can change the way MiniZip is storing the embedded filenames.
        Zoë P Scooter Software


        • #5
          Hi Zoe,

          MiniZip was not the culprit; I was the one constructing the paths in the archive. After a simple edit, all is now well.

          A tip of the hat to you, Aaron and the rest of the Scooter Software team. Great product. Great support.

          Thank you.


          • #6
            Beyond Compare 4.1.7 added support for .docx files that use '\' as a path delimiter.
            Chris K Scooter Software