Replacing specifics spaces for tab - regex

Question

I have a big text file with addresses, and I want split the data into 3 variables. Example:
NM_LOGRADO
Street BLA BLA BLA 340
Av BLE BLI 318
Road BLI 48 Block 4
I want transform into:
NM_LOGRADO
Street(TAB)BLA BLA BLA(TAB)340
Av(TAB)BLE BLI(TAB)318
Road(TAB)BLI(TAB)48 Block 4
Basically, replace the first space and the last space before the first number space by tab.
I'm using Notepad++, and for the second replacement I tried replace ' (?=[0-9])(?<=)' by '(TAB)', but it replaced all spaces before numbers (in the third line I got Road(TAB)BLI(TAB)48 Block(TAB)4). For the first replacement I have no idea :(

Answer 1

Go to Search > Replace menu (shortcut CTRL+H) and do the following:
Find what:
(?:^.+?\K | (?=[0-9]+.+))
Replace:
\t
Select radio button "Regular Expression"
Then press Replace All
You can test it with your example at regex101.
Update1:
Based on your updated sample, try this:
Find:
^([^ ]+) ([^0-9]+) (.+)
Replace:
$1\t$2\t$3
Test it at regex101.
Update2:
Based on your updated sample, try this:
Find:
(?:^[^ ]+\K |(?<!Block|Ap) (?=[0-9]))
Replace:
\t
Test it at regex101.

Answer 2

I'm assuming that (TAB) refers to a tab character rather than a literal string.
Find what: ^(\w*) ((([A-Z]{3})( )?)+) (\d.*)$
Replace with: \1\t\2\t\6
(If my assumption was incorrect, replace \t with \(TAB\))
The key is the ungreedy space: ( )?. That leaves the leading and trailing spaces uncaptured, and therefore replaced by the tab characters.
Explanation of regular expressions:
^ Beginning of line
(\w*) Any number of alphanumeric characters, i.e. "Street", "Av", "Road"
((([A-Z]{3})( )?)+) 3 uppercase letters, followed by an ungreedy space, once or more, i.e. "BLA BLA BLA", "BLE BLI", "BLI"
(\d.*) A digit, followed by any number of any characters, i.e. "340", "318", "48 Block 4"
$ End of line
\1 First capture group, "(\w*)"
\t Tab character
\2 Second capture group, "((([A-Z]{3})( )?)+)"
\t Tab character
\6 Sixth capture group, "(\d.*)"

Answer 3

as you're using Notpad++, the easiest way is not to bother with regex but rather use a macro. simply record one and play it until the end of the line. You'll want to:
put your cursor at the first character of the file
Macros > Start Recording
find a space and convert it to tab (this will replace the first space of the row)
press END to go to the end of the line
use "find previous" command to find the last space of the line
replace that space with tab
go to the next line
Macros > Stop Recording
Run your macro till the end of the file