I don't know if I can describe the exact context of the problem, but I'll try...
I'll do a search for the infamous "die/the" issue (Ctrl+G), starting from the first file in the epub, and manually correcting if and as erroneous "die"s arise, until run out of hits, say on file 6/20. Then I'll load a separate search for words that erroneously insert "d" instead of "tl" (gendy, sofdy, etc.), and Count All, with 12 matches. But when I go to the next match, I'll get the message that i've already reached the end of search... because all of the matches occur on file 7/20 or later.
For the record, on relatively clean documents where the regex is only recognizing maybe a dozen matches, I can just kind of tell if the search has not fully run its course by mentally counting how many of the "Count All" matches have been reviewed. But when it's a larger match pool, it's easy to not know if I've legitimately reached the end of the matches or if I'm running into this premature termination issue.
|