2
Dreamer
I would be grateful, if you would contact the author of NoteXpad... Thanks

! NoteXpad got a great potential, IMHO
Here the results of two "benchmark-tests" I've made:
.WRL-File,
52,000,928 bytes, 683,131 lines
Replace 830 occurrences of ".TGA" by ".PNG"
Code: Select all
+------------+---------------+---------------+
! Editor ! Load file (s) ! Replace (s) !
+------------+---------------+---------------+
! MetaPad ! 24 ! 22 !
! NoteXpad ! 12 ! 23 !
! UltraEdit ! 3 ! > 1 !
! TextPad ! > 0 ! > 1 !
+------------+---------------+---------------+
File with SQL-"INSERT"-Statements,
1,414,571 bytes, 18,997 lines
Replace 43,264 occurrences of "G_FA_VERS" by "Version_Pegasus"
Code: Select all
+------------+---------------+---------------+
! Editor ! Load file (s) ! Replace (s) !
+------------+---------------+---------------+
! MetaPad ! > 0 ! 22 !
! NoteXpad ! > 0 ! 79 !
! UltraEdit ! > 0 ! > 0 !
! TextPad ! > 0 ! 2 !
+------------+---------------+---------------+
The performance of a "replace-call" probably depends on
* number of occurrences of the search string
* lenght of replace string vs. length of search string (faster, if both lenghts are equal, slower, if replace string is shorter/longer than search string)
I think, it costs much of time to show each replacement in realtime (as NoteXpad do)... would be faster to update the main panel not until replace process has been finished (as MetaPad, TextPad and UltraEdit do). This could be configurable.
In a single pass (of "SQL-file-test") only NoteXpad has stopped showing the replacement after a few seconds and continues its work furtive. This pass of replacement test was the fastes one (peformance comparable with MetaPad). After this, I can't reproduce this behaviour... mere chance?
Regards, van Dusen