grep - reformat numbers

[grudgemom]

Hi all. In my work we do both English and French number formats. IE. $1,234 and 1 234 $, or 2.25% and 2,25 %. I've managed a grep f/c that changes the percents from E to F or F to E, but only if there's a decimal in the number. I cannot figure out how to change 5% (no decimal) without it finding the last digit before the percent sign (2.25%).

What would I add to my expression to find percentages that don't have decimals without also finding the last digit of a percentage with decimals? Note the number can be any number of digits long (IE 5%, 50%, 500%, 5000%, etc).

Thank you!

[winterm]

grudgemom  wrote

What would I add to my expression

What expression? You did show any.

This one will find all % with or without preceding horizontal space, and insert Nonbreaking Space between the last digit and %:

find: \d\K\h?%

change to: ~S%

For CS6 and higher.

Not sure if this is what you’re looking for.

[FRIdNGE]

English --> French

Find: \d\K%

Replace by: ~S%

French --> English

Find: \d\K\h%

Replace by: %

Best,

Michel, for FRIdNGE

[grudgemom]

I currently have 4 greps set up for this

English whole number percent to French:

Find: (\d+)% Change: $1 %

English decimal number percent to French:

Find: (\d+)\.(\d+)% Change: $1,$2 %

(I appear to have deleted my English whole number percent to French grep)

French decimal number percent to English:

Find: (\d+),(\d+) % Change: $1.$2%

-----------------------------------------

This also gets a little more difficult when I need to change dollar amounts to French. Those require moving the $ to the back end of the number, replacing commas with spaces and periods with commas. Fun!

For English to French I have:

Find: \x{0024}(\d+)\.(\d+) Change: $1,$2 \x{0024}

(I realize this doesn't take into account numbers over 999 - that's my next step)

I'm using the What the Grep script from Jongware to help breakdown all the pieces for me and it doesn't recognize \h or \K so that puts a kink in my process...