UniVersal Grep (uvgrep)
The uvgrep tool lets you grep text, PDF, Microsoft Office (docx, xlsx, pptx) and LibreOffice/OpenOffice (ODF) files simultaneously. It requires that bash (for getopts), grep, pdfgrep, sed, xmllint, and unzip (for unpacking docx, xlsx, pptx, odt, odp and ods files) are installed.
Get the file uvgrep, make sure it is executable (chmod a+x uvgrep), and move it to a folder which is in the PATH variable, e. g. /usr/local/bin. Check that the required tools (see above) are installed.
uvgrep [options] [files]
-i: ignore case
-n: output line numbers (text) or page numbers (PDF)
The current and initial version of uvgrep relies solely on filename extensions and should be modified to use the output of the file program.
uvgrep cannot detect on which pages, tables or slides, of a LibreOffice document a search term was found; it cannot detect on which pages of a Microsoft Word document it was found. (However it will display information about slides or tables in Microsoft documents.)
uvgrep uses German error messages for the two problems that can occur (file not found, unsupported file type). In order to localize it, change the three message variables defined in the first lines.
[esser@quad:~]$ uvgrep -in libreoffice *.sh *.pdf *.odt *.???x
uvgrep.sh:5:# uvgrep: grep txt, PDF and LibreOffice files
uvgrep.pdf:1: 5 # uvgrep: grep txt, PDF and LibreOffice files
test.odt:<text:p text:style-name="Standard">This test file contains the word "LibreOffice".</text:p>
test2.pptx[/slide9.xml]:<a:t>This is not a LibreOffice but a Microsoft Office file.</a:t>
uvgrep was meant to be named "ugrep" (universal grep), but a different project already uses that name, so I picked "uvgrep".
Copyright (c) 2016 Hans-Georg Eßer, licensed under the GPL version 3.