1) When searching and/or replacing from editor, empty lines are not matched with regular expression ^$ in a single-line mode. I have to use a multi-line regexp and search for \n\n instead. Why?
2) RESearch supports named subpatterns: (?P<name>pattern) in searches. Can I use them in replace strings, and if so, what is the proper syntax?
3) Is it possible to set the starting value of $R?
4) When replacing from panels, $R is set once at the beginning of the process. Is it possible to force a reset for each file?
5) Is it possible to redefine character classes? For example, I need a slightly reduced \w which would not include numbers. At the moment the only solution I know of requires me to type all characters by hand, e.g. [a-záčď...ž]
6) If some text is selected when I start RESearch from the editor, option "In selection only" is checked. Is it possible to force this option to unchecked by default, regardless of the existence or nonexistence of a selected block?
7) This is more of a feature request: The ability to count the number of found strings (with either search or search/replace) would be nice.
8) I suppose this is a natural consequence of how RESearch reads text from the editor, but maybe it is possible to fix it? The plugin treats the last line rather strangely in respect to a newline character. For example, input
Code: Select all
(Five lines, each [including the last] ending with a newline character) is, for the purpose of searching, considered to end with e\n\n while in reality it ends with e\n. on the other hand, input
Code: Select all
(same as above, except that the last line does NOT end with a newline) will only match e but not e\n. But if I replace newlines with hashes (search = \n, replace = #, multiline), I don't get the expected result of a#b#c#d#e but a#b#c#d#e\n (a newline character is added).