Python shell pipeline builder
Python
Switch branches/tags
Nothing to show
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
bin
pypit
tests
.gitignore
AUTHORS
LICENSE
MANIFEST.in
README.rst
ez_setup.py
setup.py

README.rst

Pypit

Pypit (pronounce "Pipe it!") is a python library for building shell pipeline easily via configuration.

The principe is simple: create a configuration which describes a programmes list with all options and your done.

Exemple

Let's say we have a file.txt with the following content:

HellO
Foo
BAR
Bla

The following configuration will sort the file.txt and lower all capitalize chars:

>>> config = [
...    {
...        "cmd": "sort -r {{input}}",
...    },
...    {
...        "cmd": "tr A-Z a-z",
...        "use_stdin": True
...    }
...]

To process this configuration:

>>> from pypit import Pypit
>>> result = Pypit(config).run(file_name="file.txt")
>>> print result
hello
foo
bla
bar

Explanation

Only cmd is required. This is the command line to use.

If use_sdtin is True, the input will be taken from the standard input.

{{input}} will describes the input file given to the run function. You can use a the {{output}} description to specify output file. Be carrefull to specify the output_ext then:

>>> config = [
...    {
...        'cmd': 'sort -r {{input}} {{output}}',
...        'output_ext': sorted # we need to specify the output extension. Here, the file will be "file.txt.sorted"
...    },
...    {
...        'cmd': 'wc -c {{input}}',
...    }
...]
>>> Pypit(config).run(file_name='file.txt')
'6\n'

Note that if the file is in another directory, you can specify if with cwd argument:

>>> Pypit(config).run(file_name='file.txt', 'cwd'='/tmp')
'6\n'

Finally, it is possible to take a look at the generated command line via the cmdline attribute:

>>> pypit = Pypit(config)
>>> res = pypit.run(file_name='file.txt')
>>> pypit.cmdline
'sort -r file.txt file.txt.sorted && wc -c file.txt.sorted

Errors may occure in your pipeline. To debug it and display errors use the errors attribute which act like a file:

>>> pypit.errors.read()

Shell script usage

Pypit package provides a shell commands to execute pypit configuration in yaml format. To do the following example, you have to create a config.yaml file:

-
    path: /bin
    shell: true
    name: echo "12345"

-
    path: /usr/bin
    name: wc
    input: STDIN
    options: -c

then call the pypit programme with the config file as argument:

$ pypit config.yaml
6

If you build your config to handle dynamic file, you can pass those file in arguments:

-
    path: /usr/bin
    name: sort
    input: STDIN
    options: -r
-
    path: /usr/bin
    name: wc
    input: STDIN
    options: -l

$ pypit config.yaml file.txt
6

Version

v0.2.2

  • add errors handle

v0.2.1

  • fix issue when with utf-8 options

v0.2

  • add dynamic file input support