Dataview FLATTEN regexreplace help?

[ironhak]

What I’m trying to do

I have a dataview table that list files contained in two separate folders grouped by those folder, the result is this:

TABLE rows.file.link as "Note"
FROM "1 - Folder/1 - Journal" 
FLATTEN regexreplace(file.folder, "^(.*/)(.+)$", "$2") AS s_file_parent_name 
WHERE file.name != s_file_parent_name and file.name != "📜 Index"  and file.name != "Material to check"
and file.folder != "1 - Folder/1 - Journal/1.2 - Learnings"
sort file.ctime asc
GROUP BY  s_file_parent_name as Type
SORT Type desc

I’m trying to remove the "1.x - " part in the “Type” columns, so I’ll have the two groups called respectivelly “Working” and “Ideas”.

I’m a total noob with dataview and I built the dataview snippet by searching little pieces all over this forum. Thank’s for helping.

[holroy]

I’m out on thin ice here, as I don’t quite see the entire folder and file name structure you’re using, but maybe something like the following would work:

  regexreplace(file.folder, ".*\/\d+(?:\.\d+)* - ([^\/]+)", "$1") 

Explanations follows:

• .*\/ – Gobble up everything until and including a forward slash
• \d+ – Match digits, \d, that is one or more digits, +
• (?: ... )*– Optionally followed by zero or more, *, non-capturing group, (?: ... ) of:
  • \.\d+ – Sequences of a literal period, \., and a number, \d+
•  -  – Matching the literal sequence of  - , a dash with spaces on both sides
• ( ... ) – A capture group consisting of:
  • [^ ... ]+ – A non matching character class, [^ ... ], with at least one, +, character not matching:
    • \/ – The literal forward slash. This sequence helps anchoring the regex, and allows for the first wildcard match to be gready and get all preceding sub-folders

Since the last group was a capture group, we can use that as the replacement text, "$1". This group should strip away number sequences in front of stuff like 1 - Something, 1.12 - Another one, or 127.0.0.1 - Loopback address.

PS! I’ve not looked at the rest of query for any optimisations or such.

[ironhak]

It works properly. I’ll also study your amazing explaination of what the command is doing. I really appreciate your kind contribution, have a good day