Extract text after a pattern in a text string

[xxx59712]

I have the following text sample

Code: Select all

01. since the 1500s, when an
02. opposed to using 'Content here
03. and a search for 'lorem ipsum'
20 - randomised words which don't look even slightly believable
19 - going through the cites of the word
01. Finibus Bonorum et Malorum 02. essentially unchanged it was popularised in 03. The first line of Lorem Ipsum 04. predefined chunks as necessary

I need to extract the text excluding numbers ("02." or "20 -") and store each sentence in the part of an array. So the array would be Variable(0) = "since the 1500s, when an",
Variable(1) = "opposed to using 'Content here" and so forth.

I managed to write what I think would be the regex to find the separating pattern

Code: Select all

 \d\d(\.| -)

but I do not manage to come up with the function to use and the loop
to achieve that. Can you somebody please help me?

Thank you.

[mikeyww]

AHK simple arrays start with a default key of 1. This typically makes working with simple arrays straightforward in AHK.

Code: Select all

str =
(
01. since the 1500s, when an
02. opposed to using 'Content here
03. and a search for 'lorem ipsum'
20 - randomised words which don't look even slightly believable
19 - going through the cites of the word
01. Finibus Bonorum et Malorum 02. essentially unchanged it was popularised in 03. The first line of Lorem Ipsum 04. predefined chunks as necessary
)
text := []
For each, line in StrSplit(str, "`n", "`r")
 text.Push(RegExReplace(line, "^.*?[.-]\h*"))
For lineNum, line in text
 MsgBox, 0, Line #%lineNum%, %line%

Explained: Simple arrays

[xxx59712]

Thank you very much, we are almost there.

What would be the function to use to :
1. go through the text
2. Find a number pattern (e.g. "01.")
3. Take the sentence following that number pattern
4. Store it in as item in the array

Thanks.

[mikeyww]

The current script does not do it? If so, you can provide the specific input and output strings that you need, as comprehensive examples.

[xxx59712]

My apologies, I did not notice there was more lines than visible in your post.

I agree it does that.

The last challenge I am facing is the last line of the example :

Code: Select all

01. Finibus Bonorum et Malorum 02. essentially unchanged it was popularised in 03. The first line of Lorem Ipsum 04. predefined chunks as necessary

Sometimes the text provided does not have some line break to separate 01. from 02. from 03. etc, the text would be a block without any line break.

I would need to extract data the same way as the code you wrote, is that possible

Final result would be :

Code: Select all

[1]since the 1500s, when an
[2]opposed to using 'Content here
[3]and a search for 'lorem ipsum'
[4]randomised words which don't look even slightly believable
[5]going through the cites of the word
[6]Finibus Bonorum et Malorum
[7]essentially unchanged it was popularised in
[8]The first line of Lorem Ipsum
[9]predefined chunks as necessary

Thanks.

[mikeyww]

Use:

Code: Select all

For each, line in StrSplit(RegExReplace(str, " (\d\d(\.| -) )", "`n$1"), "`n", "`r")

[xxx59712]

Thank you so much.

I admit this is well above what I can do in term of scripting.

[mikeyww]

Not by much; my regex is similar to yours. My idea was to "normalize" the text by prepending line feeds to those strings. That puts each desired part on its own line. The subsequent line then removes the leading text, and adds the remaining text to an array. There might be simpler approaches overall.