msooxml 1.8 KB

123456789101112131415161718192021222324252627282930313233343536
  1. #------------------------------------------------------------------------------
  2. # $File: msooxml,v 1.5 2014/08/05 07:38:45 christos Exp $
  3. # msooxml: file(1) magic for Microsoft Office XML
  4. # From: Ralf Brown <ralf.brown@gmail.com>
  5. # .docx, .pptx, and .xlsx are XML plus other files inside a ZIP
  6. # archive. The first member file is normally "[Content_Types].xml".
  7. # but some libreoffice generated files put this later. Perhaps skip
  8. # the "[Content_Types].xml" test?
  9. # Since MSOOXML doesn't have anything like the uncompressed "mimetype"
  10. # file of ePub or OpenDocument, we'll have to scan for a filename
  11. # which can distinguish between the three types
  12. # start by checking for ZIP local file header signature
  13. 0 string PK\003\004
  14. !:strength +10
  15. # make sure the first file is correct
  16. >0x1E regex \\[Content_Types\\]\\.xml|_rels/\\.rels
  17. # skip to the second local file header
  18. # since some documents include a 520-byte extra field following the file
  19. # header, we need to scan for the next header
  20. >>(18.l+49) search/2000 PK\003\004
  21. # now skip to the *third* local file header; again, we need to scan due to a
  22. # 520-byte extra field following the file header
  23. >>>&26 search/1000 PK\003\004
  24. # and check the subdirectory name to determine which type of OOXML
  25. # file we have. Correct the mimetype with the registered ones:
  26. # http://technet.microsoft.com/en-us/library/cc179224.aspx
  27. >>>>&26 string word/ Microsoft Word 2007+
  28. !:mime application/vnd.openxmlformats-officedocument.wordprocessingml.document
  29. >>>>&26 string ppt/ Microsoft PowerPoint 2007+
  30. !:mime application/vnd.openxmlformats-officedocument.presentationml.presentation
  31. >>>>&26 string xl/ Microsoft Excel 2007+
  32. !:mime application/vnd.openxmlformats-officedocument.spreadsheetml.sheet
  33. >>>>&26 default x Microsoft OOXML