Announcing PyCascading, a Python wrapper for Cascading
Gabor Szabo
Cascading flows from Python, and user-defined functions, aggregators,
and filters are also written natively in Python. The Python
interpreter is Jython, so all the libraries available to both Python
and Java are reusable for putting together Cascading jobs quickly.
The usual word counting example would look like this:
@map(produces=['word'])
def split_lines(tuple):
for word in tuple.get('line').split():
yield [word]
def main():
...
input | split_lines | GroupBy('word') | Count() | output
....
Decorators are used to decorate functions to be used in flows, pipes
are chained together with the '|' operator, and merged with '&' in
joins.
PyCascading is open source, and hosted at https://github.com/twitter/pycascading
. Please see the examples folder for usage patterns, and submit any
questions or reports to the cascading-user list.
Happy hacking,
Gabor
Chris K Wensel
chris
> --
> You received this message because you are subscribed to the Google Groups "cascading-user" group.
> To post to this group, send email to cascadi...@googlegroups.com.
> To unsubscribe from this group, send email to cascading-use...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/cascading-user?hl=en.
>
--
Chris K Wensel
ch...@concurrentinc.com
http://concurrentinc.com
