One liner for those of you who don't have machine with hudreds of gigs of RAM and want to process `page_views.csv.zip` in a mini-batch fashion:
```
mkdir page_views_splits
unzip -p page_views.csv.zip | split --line-bytes=300m --filter='gzip --fast > ./page_views_splits/$FILE.gz'
```

In a pipe we 

1. uncompress zip stream
2. with argument `--line-bytes=300m` we create batches of 300 mb (commabd doesn't break lines!)
3. compress to `gzip` file (fast comression, minimal compression rate)
4. save it to `page_views_splits` directory which we created at first line


Enjoy!

# Edit for OS X users

First install gnu command line utils `coreutils` - built in tools are heavily outdated:
```
brew install coreutils
```
Then use `gsplit` instead of `split`
```
mkdir page_views_splits
unzip -p page_views.csv.zip | gsplit --line-bytes=300m --filter='gzip --fast > ./page_views_splits/$FILE.gz'
```