|
Regarding the regular expressions issue:
I think in normal mode, it should have treated the \1 as pure text and not a capture pattern as it has nicely done for \2 later on.
In regex mode it should treat them as invalid chars since no captures were done.
I will look into this but as DiapDealer said the lack of actual capture groups in the find regex does cause many issues.
I think in this case we pad the caputure_group_offsets list and that creates an off by 1 situation, so it tries to replace \1 when it should have treated it as pure text. That said the \1 also fits the escape for an octal character in C strings so who knows!
|