insightdatascience / parsing-workshop Goto Github PK

Chapter1:Both find and uniq has the same description

In chapter 2, I have a couple of notes:

• The argparse module is a bit heavy handed for most purposes. In most cases, it is faster and easier to use sys.argv. Surprisingly, it was a little hard to find good, simple examples of sys.argv on a site that isn't riddled with popups, but I think this site should do. sys.argv is simply a list of arguments given to a python program, with the name of the program at index 0.
• This comprehensive python cheatsheet should be included in the resources. Between this cheatsheet, google, and practice practice practice, you have the recipe for a good time learning python. For a less terse, more example-driven reference, I like learnXinYminutes

Notes for chapter 3:

The awk tutorial in the references is not very learner friendly for several reasons. It would be good to establish the usual awk 'condition {action}; another condition {another action}' file structure and then talk about possible conditions and actions in increasing complexity using examples. There are some ok tutorials out there, like this one, although this one makes a huge oversight in not mentioning regular expressions and piping into other commands like sort or uniq. Basically, I think it's worth fleshing out awk a bit.

• First with regex, like printing out field 3 of lines matching a regex:

awk '/^A[bc]+/ {print "The third field of this matching line is ",$3}' file.txt

• Then get into examples of the built in variables NR,FNR,NF,FS,OFS to do things like averages.
• Then maybe get into BEGIN{} END{}, variables and associative arrays (using BEGIN rather than -v).
• Then some basic flow control.
• Then show the possibilities of piping into sort -rn, uniq -c, etc.
• All of this, of course, driven by examples. I think that would give a strong enough foundation to tackle to parsing exercises.

Chapter1:The copy command goes by cp, in the syntax and example, "copy" is used.