Hi Kim,
From your reply, it happens that one SPACE should be present, in the search regex, between the opening round bracket and the block [+-] !
Well, I just realized that your numbers could be separated by more than one space characters ! Moreover, your file may, also, contain some tabulation characters. In that case, prefer the search regex below ( The replacement regex is unchanged ):
(\h+[+-]?\d+(\.(\d+)?)?[Ee][+-]\d\d){3}
Notes :
The \h regex syntax matches one horizontal blank character, that is to say :
The normal space character ( \x20 )
The tabulation character ( \x09 )
The No-Break space ( abbreviated NBSP ) ( \xA0 )
As usual, the + quantifier means one or more times the previous character
So given the test example below, where I put, between numbers, several spaces, in the first line, several tabulations in the second line and a mix of these two, in the third line :
ospht421 4.52E-02 -1.20E-02 2.98E-03 3.19E-02 -1.93E-02 -6.87E-02 4.16E-02 1.71E-02 -2082.95E-02
ospht000 +9.9999998E+03 +1.23e-07 -5.4625e+04 -6.258E-00 5.789E+02 1.008E-06 +9.E+03 +1.E-07 -5.e+04
ospht999 -6.E-00 5.E+02 1.e-06 +9E+03 +1e-07 -5E+04 -6e-00 5E+02 1E-06
After clicking on the Replace All button, we get the following text :
ospht421
4.52E-02 -1.20E-02 2.98E-03
3.19E-02 -1.93E-02 -6.87E-02
4.16E-02 1.71E-02 -2082.95E-02
ospht000
+9.9999998E+03 +1.23e-07 -5.4625e+04
-6.258E-00 5.789E+02 1.008E-06
+9.E+03 +1.E-07 -5.e+04
ospht999
-6.E-00 5.E+02 1.e-06
+9E+03 +1e-07 -5E+04
-6e-00 5E+02 1E-06
Now, we have to get rid of all the possible blanks characters, at the beginning of the lines and just keep one space character between the different numbers. The following S/R will achieve it, nicely :
Find what : ^\h+|(\h+)
Replace with : (?1 ) , with a space after the digit 1
Notes :
Due to the alternative regex symbol |, this search regex looks :
for any non null range of blank characters, that begins the line ( the ^ symbol represents the zero length assertion beginning of line )
for any subsequent range of these blank characters, between the numbers
The second part (\h+), ( the subsequent blanks ), is enclosed in round brackets, and represents the group 1
In replacement, we use the conditional form (?n....) :
which replace the searched regex with the string located after digit n , till the ending round bracket, if the group n exists
which does NO replacement if the group n does NOT exist
So :
When the first part, of the search regex, matches possible blank characters, at beginning of a line, group 1, does not exist. Therefore, there’s NO replacement
When the second part, of the search regex, matches blank characters, before the numbers, group 1 exists. Then they are replaced with the single space, located between the digit 1 and the ending round bracket, of the replacement regex
This time, after performing this second S/R, we obtain the final text below :
ospht421
4.52E-02 -1.20E-02 2.98E-03
3.19E-02 -1.93E-02 -6.87E-02
4.16E-02 1.71E-02 -2082.95E-02
ospht000
+9.9999998E+03 +1.23e-07 -5.4625e+04
-6.258E-00 5.789E+02 1.008E-06
+9.E+03 +1.E-07 -5.e+04
ospht999
-6.E-00 5.E+02 1.e-06
+9E+03 +1e-07 -5E+04
-6e-00 5E+02 1E-06
Cheers,
guy038
