-
Notifications
You must be signed in to change notification settings - Fork 0
/
TODO
28 lines (28 loc) · 1.47 KB
/
TODO
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
- do by using regex to split on last by <field> and assume none if missing and just run as map over groups
- _2 in dump by _1 in other
- _2 in dump by _1 in other
+ _1 if /www/
+ _1 if _2 =~ x
+ .parselyrc
- split by: if a X or not, or if _1 in other_file (subsides comm)
+ in "foo 1\nbar 2" "sum(_2) if _2 < 3" to see how much I can delete
- _1._1 for submatch (e.g. user agent)
- index of <foo> (e.g. when something is processing ines and you want to check where it is)
- skip line e.g. 123\na\nb\nc -> parsely _2 unless idx = 0
- normalize e.g. uniq on
/?email=digest%40digg.com&name=&callback=jsonp1310305312089
/?email=digest%40digg.com&name=&callback=jsonp1310305312088
/?email=digest%40digg.com&name=&callback=jsonp1310305312085
/?email=digest%40digg.com&name=&callback=jsonp1310305312086
skip jsonp bit
summaryse is a cool extension from which to steal stuff. Maybe provide basic functionalities through that
- given [/m/09_pbpl /type/object/key /soft/isbn/9780491035453,..] and [/m/077601h /book/isbn/book_editions /m/09_pbpl] get [/m/09_pbpl /soft/isbn/9780491035453]
- use merge sorted and merge unsorted probably
- find a simpler way to do aggregates
- top k
selext top/max N or N% ditto for minus
select outliers
parsely '_1.each_cons(3).to_a if rand(400)==1' /usr/share/dict/words |pbcopy #make PerlVar enumerable
document GEOIP, and everything else
list fields, eg pr -l => _1 _2(_1 _2) _3 _4
on -k also perform file guessing (e.g. headers for csv, common log format)