Support for LIWC 2022's .dicx Format

[jonkush]

Requested feature

Thank you for all you work on quanteda. LIWC released a new version of their software in May that uses a new format for their dictionary files with a new extension (.dicx). The format is similar so I hope it won't be too difficult to support it with the dictionary() command. I've copied an example of the code format below [first Github post so apologies if anything is off with this post].

#Original format (.dic)
%
1	Correct
2	Incorrect
%
rope*	1
candlestick*	2
candle stick	2
pistol*	2
knife	2
knives	2
pipe*	2

#New format (.dicx)
Entry,Correct,Incorrect
rope*,1,
candle stick*,,1
candlestick*,,1
knife,,1
knives,,1
pipe*,,1
pistol*,,1

Use case

LIWC 2022 offers a very handy dictionary building tool but it only exports into .dicx.

[kbenoit]

Thanks - I will get on this asap. I hadn't realised they just issued a new release. Looks like I will need to purchase a license for the new version.

In the dictionary format, it looks like they went in their format from really bad to slightly bad. Why are we the only ones suggesting something sensible like a YAML format?

[lluaces1]

Would love to know if you or someone else has a fix for reading "dicx" in R (terrible format, worse name...)

[kbenoit]

Not yet but it's on the short list. Was hoping someone might take this issue on, but if not, I'll tackle it soon.

Yep, pretty terrible name!