It’s been a while since I wrote a post and I’m also going to make this one a short one.
Python Fire is a pretty cool Python library from Google that lets you turn any Python script into a command line tool, this means that you don’t have to write any logic for handling command line parameters and options, but, if you script has sophisticated use cases then you’ll be probably be better by using argparse from the standard library.
You can install Python fire from pip by running these commands:
1
| pip install ipython<br></br>pip install fire
|
Then, to turn your script into a command line tool, all you need to do is import fire and and call the Fire method by passing it a class. Fire parses the class and generates logic for handling command line arguments for each method the class has. For example:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
| class LzwCommandLineAdapter:
"""
This will be used with Google Fire for easy usage,
so we can avoid argparse/optparse.
"""
@staticmethod
def encode(filename, out="encoded.txt"):
"""
Encodes the text from the give filename.
"""
pass
@staticmethod
def decode(filename, out="decoded.txt"):
"""
Decodes the text from the given filename.
"""
pass
if __name__ == "__main__":
import fire
fire.Fire(LzwCommandLineAdapter)
|
When we run the script, we get the following result:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
| ➜ Desktop python lzw.py
Type: LzwCommandLineAdapter
File: ~/Desktop/lzw.py
Docstring: This will be used with Google Fire for easy usage,
so we can avoid argparse/optparse.
Usage: lzw.py
lzw.py decode
lzw.py encode
➜ Desktop python lzw.py decode
Fire trace:
Initial component
Instantiated class "LzwCommandLineAdapter" (lzw.py:248)
Accessed property "decode" (lzw.py:264)
('The function received no value for the required argument:', 'filename')
Type: function
String form:
File: ~/Desktop/lzw.py
Line: 264
Docstring: Decodes the text from the given filename.
Usage: lzw.py decode FILENAME [OUT]
lzw.py decode --filename FILENAME [--out OUT]
|
This is a pretty nice solution when you’re writing lots of Python scripts to experiment with things. Thank you for reading!
Further reading