Search

This project is a simple and fast command line tool that allows the user to Search for file names, file types, and text strings within the files. Speed has always been a high priority for Search, currently it takes about 8.4 seconds to trawl ~350GB of files (on a SSD) but we're always trying to make it faster.

What Do I Need?

If you already have Python 3.5 (or higher) installed on your computer then you can just add search.py to PATH and use Search from the command line whenever you want. Otherwise, if you are a Windows user we have a compiled executable search.exe that you can download and add to your PATH environment variable and Search that way.

How To Use Search

Search is based as a command line tool that accepts numerous arguments e.g. if you want to Search for HelloWord.txt.

$ search.py -d=/home/andy -n=HelloWorld.txt

Or if you want to Search for some text in every Python file on your computer, try this.

$ search.py -e=.py -t="import numpy"

Note that if you don't specify the directory then Search will start from C:\ for Windows users or / for Linux users.

Alternatively, if you're just curious about the number of lines of Python that you've written, Search can do that.

$ search.py -d=/home/andy/my_project -e=.py -c

If your Search is likely to yield a lot of results e.g. Searching for every text file on your computer, then you can output the results to a file like so.

$ search.py -d=/home/andy -e=.txt -o=My_Text_Files.txt

That's basically it for your day to day use, but there are a couple of other options that can also be useful.

• -i Indicates a case-insensitive Search, for either text Searches or file name Searches.
• -h Prints out the help message, showing all your options and what each of them does.
• -v Indicates a verbatim Search for file names. Ordinarily a file name Search will return any file with the Search term in it's name, e.g. a Search with -n=hello would match with files hello_there, why_hello_there, and oh_hello_there, while a verbatim Search will only match with a perfect name match.

Things to do

• Implement text find and replace option.
• Add benchmarking tests to more accurately quantify Search speed.