[unterrainer_ale]

Well it really depends what you actually want to do. they keyword parse (parse | Reference | kdb+ and q documentation – Kdb+ and q documentation (kx.com)) simply parses an expression into a parse tree, in this case it parses a q-sql statement into a functional select. The output you got, is what your qsql looks like in form of a functional select.

q)parse "select time,sym,bid,ask from t"

?
t
()
0b
time
sym
bid
ask!
time
sym
bid
ask

q)t:([] time:5?.z.p; sym:5?
3; bid:5?10.0; ask:5?10.0)

q)t
time sym bid ask
---------------------------------------------------
2014.11.16D17:25:17.686664320 enf 3.017723 3.927524
2012.10.31D16:48:20.456221120 plh 7.85033 5.170911
2016.10.06D02:12:56.529433024 nni 5.347096 5.159796
2005.07.26D06:48:01.359027808 glc 7.111716 4.066642
2016.10.09D03:08:41.154352640 gkp 4.11597 1.780839

q)eval parse "select time,sym,bid,ask from t"

time sym bid ask
---------------------------------------------------
2014.11.16D17:25:17.686664320 enf 3.017723 3.927524
2012.10.31D16:48:20.456221120 plh 7.85033 5.170911
2016.10.06D02:12:56.529433024 nni 5.347096 5.159796
2005.07.26D06:48:01.359027808 glc 7.111716 4.066642
2016.10.09D03:08:41.154352640 gkp 4.11597 1.780839

However, you don’t actually need to parse/transform your qsql to a functional select inside your function if you don’t really “want” (I recommend using functional selects only if you really need to, i.e. if you need to parameterise your queries)

Here you have a function that takes a table as input and then selects from it using qsql

q){select time,sym,bid,ask from x}[t]

time sym bid ask
---------------------------------------------------
2014.11.16D17:25:17.686664320 enf 3.017723 3.927524
2012.10.31D16:48:20.456221120 plh 7.85033 5.170911
2016.10.06D02:12:56.529433024 nni 5.347096 5.159796
2005.07.26D06:48:01.359027808 glc 7.111716 4.066642
2016.10.09D03:08:41.154352640 gkp 4.11597 1.780839

A very helpful read about functional selects is the Whitepaper about parse trees.

Parse trees and functional forms | White Paper | kdb+ and q documentation – Kdb+ and q documentation (kx.com)

Hope this helps

Alex

code.kx.com

parse | Reference | kdb+ and q documentation - Kdb+ and q documentation

parse is a q keyword that returns a parse tree for a string expression.

[unterrainer_ale]

I use kdb studio to test, debug and develop code and IntelliJ as an editor. I like IntelliJ because you can easily search for function usage, jump to definitions etc. It’s great when reading code. It also has a kdb analyst and q plugin so you can actually run code in IntelliJ (similar to studio you connect to a q process) and display results but I never really tried it.

[unterrainer_ale]

I would avoid using ChatGPT for KDB/Q. It’s often very far off. Read Q for Mortals, it’s a good start to learn Q

[unterrainer_ale]

Hi Simon,

is there any chance you could provide an example of bigNestedThing? or an extract? In the meantime, try to have a look at cross-section and how it works with apply.

[unterrainer_ale]

Something worth adding to the previous answers: If you can transform your filter into a table then the select is much easier. To illustrate

q)t:flip `date`data!(2023.05.20 2023.05.20 2023.05.20 2023.05.19 2023.05.19 2023.05.19;`a`c`b`b`b`c) 
q)t 
date      data 
--------------- 
2023.05.20 a 
2023.05.20 c 
2023.05.20 b 
2023.05.19 b 
2023.05.19 b 
2023.05.19 c 

q)select from t where ([] date;data) in ([] date:2023.05.20 2023.05.20 2023.05.19;data:`a`b`b) 
date      data 
--------------- 
2023.05.20 a 
2023.05.20 b 
2023.05.19 b 
2023.05.19 b 

q)filter:([] date:2023.05.20 2023.05.19 2023.05.18 2023.05.17; data:(`a`b;enlist `b;`c`d`a;`d`a)) 
q)filter 
date        data 
----------------- 
2023.05.20 `a`b 
2023.05.19 ,`b 
2023.05.18 `c`d`a 
2023.05.17 `d`a 

q)ungroup filter 
date      data 
--------------- 
2023.05.20 a 
2023.05.20 b 
2023.05.19 b 
2023.05.18 c 
2023.05.18 d 
2023.05.18 a 
2023.05.17 d 
2023.05.17 a 

q)t1:ungroup filter 
q)select from t where ([] date;data) in t1 
date      data 
--------------- 
2023.05.20 a 
2023.05.20 b 
2023.05.19 b 
2023.05.19 b