Tested on both 3.5.9 (c424586) and 3.5.9 beta 28 (0c3fd44).
The function hits a line like
I have recorded conversations
with Peter Iannone, Kurt Hewig and Myra Mark
and tries to fix, but the way I see it, there is no difference between 'before' and 'after'.
Why does it hit the line?
Before
I have recorded conversations
with Peter Iannone, Kurt Hewig and Myra Mark
After
I have recorded conversations
with Peter lannone, Kurt Hewig and Myra Mark
Capital I was converted to l. It's a bug
You can add the name Iannone to the name list...
But I've tried to improve this, so lannone must a correct word in latest beta: https://github.com/SubtitleEdit/subtitleedit/releases/download/3.5.9/SubtitleEditBeta.zip
How does that work?
For testing purpose, I made up slightly modified version...
I Iave recorded conversatIons
with Ieter Iannone, Iurt IewIg and Iyra Iark
and beta 35 (8d173b8) gave me the following output.
I lave recorded conversatlons
with Ieter Iannone, Iurt Iewlg and Iyra lark
I guess 'conversatlons' 'Iewlg' should be 'conversations', 'Iewig'.
'Iark' to 'lark' is correct, considering 'lark' is a valid word.
I thought "Fix uppercase 'i' inside lowercase words (OCR error)" function simply changes uppercase 'I' inside lowercase words to lowercase 'i' such as 'conversatIons' to 'conversations'.
On the second thought, OCR error is assumed to wrongly replace 'l' (lowercase L) with 'I' (uppercase i) in most cases, but not 'i' with 'l' (lowercase L)?
Hm, I could remove the check from words starting with "I", like this:
https://github.com/SubtitleEdit/subtitleedit/releases/download/3.5.9/SubtitleEditBeta.zip
Or change the text to "Fix uppercase 'i' in lowercase words (OCR error)"...
Any thoughts?
Fix uppercase 'i' inside lowercase words implies an I somewhere in a word such as heIlo. The first letter of a word Ian is on the outside and should be left alone except for common errors, which is sorted by fixreplacelist etc. SE should ignore the first letter of each word when checking for Fix uppercase 'i' inside lowercase words.
OK, SE will no longer change words starting with an uppercase "i" via this function.
Most helpful comment
Before
After
Capital
Iwas converted tol. It's a bug