XP Roadmap
v1.1
----
* Integrate a proper grammar
* Support 3rd party kernels
v1.2
----
* Implement fx commands: import, export, info, and dry_run
* Add support for code block execution code checking
* block_type.check()
* fx command: check_block_support
* Add parallel task execution
v1.3
----
* Add task generators. Something like this::
# there are two general approaches we might take here...
# Approach 1: internal task generators
gen_${year}_data: download_${year}_data
generator.py:
for x in range(2011,2015):
yield {'year':x}
# Approach 2: combined task generators
download_${year}_data (dd):
code.sh:
curl http://some.web.server/myfile_${year}.tgz > $PLN(myfile_${year}.tgz)
gen_${year}_data (gd): download_${year}_data
code.sh:
blah blah
task generator:
Distant Future
--------------
* Add distributed task execution (flex cluster)
* Implement notebook interface (ala ipython & jupyter)
* Add support for comment blocks that have markdown format or something. This would allow pipelines to be workbooks in themselves.