Todo

Todo.md

@@ -18,6 +18,8 @@
 
 # Bunch of Ideas
 
+Consider allowing curry to take expressions as baked arguments that allow to transform arguments via _.each expressions?
+
 get mybinder tutorial going so users can more easily explore fluentpy
 
     No problem. If you are curious about Spark, the high-level idea is that if you have a lot of data, you want to break your problem across multiple machines to speed up computation. I used Spark as an example as that's one of the most popular distributed computing frameworks. On a day to day basis, I actually use Apache Dask, which is basically the same thing as Spark.  Both Spark and Dask are lazy (like a generator). I looked at the source code of FluentPy, and it seems some parts of lazy (ie uses yield) and some parts are eager (ie uses return). If you are curious, take a look at this Dask syntax. It looks very similar to fluentpy: https://github.com/dask/dask-tutorial/blob/master/02_bag.ipynb In Dask, all the transformations (map/filter/reduce/etc) are lazy until you use .compute(), which triggers actual computation.  Also here's a neat tool: https://mybinder.org/    Mybinder gives you a Jupyter notebook to run in your browser for free--behind the scenes, it's just a container that clones a repo and installs the library. Hence, you can run the Dask tutorial by going to https://mybinder.org/v2/gh/dask/dask-tutorial/HEAD If you want, you can consider adding a tutorial notebook in your fluentpy repo, so then users can simply go to https://mybinder.org/v2/gh/dwt/fluent/HEAD to run the code all through the browser.   I think the fluent interface is a very cool thing that most Python programmers are not aware about, so when I show them for the first time, they are amazed! I remember when I first saw it in Java (and it was just a quick screenshot since I actually don't program in Java), I was thinking, wow this is amazing.  Hope that helps, Eugene