Exit Python loop when replacement count is 0

[M Andre Z Eckenrode]

When performing manual find/replace, the bottom of the dialog states how many occurrences were replaced. Is it possible to use some form of equivalent information in a Python script utilizing editor.rereplace in a loop, to exit when the last execution resulted in 0 replacements? If so, could someone provide an example? I’m not seeing any useful clues in the Python Script help.

Example of code I want to have repeating in a loop:

editor.rereplace(r'^(.+?) (\(tracks: .+?\)), (.+?) — (.+?)$', ur'\1 — \4 \2\r\n\3 — \4')

[Alan Kilborn]

@M-Andre-Z-Eckenrode

One technique you could use is to get the text of the document into a variable before the replacement (e.g. temp = editor.getText()), then get the text after the replacement, and see if it is the same as the saved variable, e.g. if editor.getText() == temp. If so, you know that the replacement replaced nothing.

[Alan Kilborn]

@M-Andre-Z-Eckenrode

This may also be of interest:

Feature request: editor.rereplace() could return number of replacements made.

[Alan Kilborn]

Maybe it goes without saying that you could get the match count before doing the replacement – and of course if the match count is zero then you can break out of your loop before doing a replacement that wouldn’t replace anything:

while True:
    matches = []
    editor.research(your_regex, lambda m: matches.append(m.span(0)))
    if len(matches) == 0: break
    editor.rereplace(your_regex, ...)

[M Andre Z Eckenrode]

@Alan-Kilborn

All great ideas that I didn’t think of, and I like that last one the most, so I think I’ll go with it. Thanks much, Alan!

[Alan Kilborn]

@M-Andre-Z-Eckenrode said in Exit Python loop when replacement count is 0:

I like that last one the most

A modification to that one could be to limit the matches that research() will look for to 1, e.g.:

editor.research(your_regex, lambda m: matches.append(m.span(0)), 0, pos1, pos2, 1)

The last argument of 1 will limit the possible matches to 1, which is sufficient to determine that there are matches remaining. This could speed up a slow running regex by not looking for things you don’t need.

[M Andre Z Eckenrode]

@Alan-Kilborn said in Exit Python loop when replacement count is 0:

A modification to that one could be to limit the matches that research() will look for to 1, e.g.:

Noted, and I like it. Thanks again.