1 2#------------------------------------------------------------------------------ 3# $File: msooxml,v 1.15 2020/09/05 22:59:14 christos Exp $ 4# msooxml: file(1) magic for Microsoft Office XML 5# From: Ralf Brown <ralf.brown@gmail.com> 6 7# .docx, .pptx, and .xlsx are XML plus other files inside a ZIP 8# archive. The first member file is normally "[Content_Types].xml". 9# but some libreoffice generated files put this later. Perhaps skip 10# the "[Content_Types].xml" test? 11# Since MSOOXML doesn't have anything like the uncompressed "mimetype" 12# file of ePub or OpenDocument, we'll have to scan for a filename 13# which can distinguish between the three types 14 150 name msooxml 16>0 string word/ Microsoft Word 2007+ 17!:mime application/vnd.openxmlformats-officedocument.wordprocessingml.document 18>0 string ppt/ Microsoft PowerPoint 2007+ 19!:mime application/vnd.openxmlformats-officedocument.presentationml.presentation 20>0 string xl/ Microsoft Excel 2007+ 21!:mime application/vnd.openxmlformats-officedocument.spreadsheetml.sheet 22>0 string visio/ Microsoft Visio 2013+ 23!:mime application/vnd.ms-visio.drawing.main+xml 24 25# start by checking for ZIP local file header signature 260 string PK\003\004 27!:strength +10 28# make sure the first file is correct 29>0x1E use msooxml 30>0x1E default x 31>>0x1E regex \\[Content_Types\\]\\.xml|_rels/\\.rels|docProps 32# skip to the second local file header 33# since some documents include a 520-byte extra field following the file 34# header, we need to scan for the next header 35>>>(18.l+49) search/6000 PK\003\004 36# now skip to the *third* local file header; again, we need to scan due to a 37# 520-byte extra field following the file header 38>>>>&26 search/6000 PK\003\004 39# and check the subdirectory name to determine which type of OOXML 40# file we have. Correct the mimetype with the registered ones: 41# https://technet.microsoft.com/en-us/library/cc179224.aspx 42>>>>>&26 use msooxml 43>>>>>&26 default x 44# OpenOffice/Libreoffice orders ZIP entry differently, so check the 4th file 45>>>>>>&26 search/6000 PK\003\004 46>>>>>>>&26 use msooxml 47>>>>>>>&26 default x Microsoft OOXML 48>>>>>&26 default x Microsoft OOXML 49