Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
  • #21001

    My PC: Win7 64bit + 16GB RAM + Intel 3.1GHZ CPU

    Recently, I choose Emeditor to handle big text files(from 20MB ~ 800MB html+css style data),
    Most of features was satisfied for me, excepted for the following problem.
    In Emeditor, if want to delete blank lines, I will use regular expression to find “^\n”, and replace all with “”(nothing).
    (Macro “delBlankLines.jsee” will do the similar thing.)

    It spent a long time on this, for example, one of the smaller file, size around 42MB, more than 970000 lines.
    After one minute past, just 4% finished (15250 found), even checked the “Accelerate by Discarding Undo Info.”
    Refer to

    Therefore, I will change to use notepad++(a free source code editor) as workaround solution.
    It is just about 50 seconds, all 320801 occurrence was replaced.

    Another solution: use Emeditor’s Extract All with “^[^\n]” to get the same result in the new tab.
    It will take around 65 seconds only, a little slower than notepad++, much faster than Emeditor standard method.
    So suppose Emeditor Core have chance to optimize the replacement algorithm.

    But if I want to find all “xxxx \n yyyy” and replace all with “xxxx yyyy” for combining data, it will be unable to do so.
    In this situation, could you kindly advise how to increase the speed for removing “\n” ?

    Yutaka Emura

    To delete blank lines, I would bookmark all regular expressions:


    (use “Bookmark All” button in the Find dialog box)

    and then use the Delete Bookmarked Lines command.

    If you have enough physical memory in your PC, and if you need to replace many multi-line text (combine lines or split lines), you might want to try clearing the Use Temporary File to Reduce Memory Usage check box in the Advanced tab of the Customize dialog box.



    Thanks for the tip about using Bookmarks.


    To delete blank lines, “Delete Bookmarked Lines” is very good, less than 1 second !!

    But even I have enough physical memory(16GB ram vs. 42MB data), combine lines or split lines is still slow.

    Anyway, thanks for your tip for the first question.


    If I change to Binary (Hex View), and find “hh hh 0D 0A hh hh” replace with “hh hh 40 40 hh hh”
    Then change back to UTF-8, and find “@@” replace with “”(nothing).

    It will be very fast for combine some lines, but can not combine all lines.
    Due to there are some strings will have one more “space” between hex code.
    Ex. “hh hh 0D 0A hh hh” or “hh hh 0D 0A hh hh”

    Therefore, if Emeditor can provide pure Hex View without any additional “space”.
    Or suggest Emeditor can provide pure command line mode without Windows interface.
    I believe it can be done.

Viewing 5 posts - 1 through 5 (of 5 total)
  • You must be logged in to reply to this topic.