Find

[gcotterl]

How can I find rows with a decimal-point at the 58th position?

[bveldkamp]

Code: Select all

Find what ^.{57}\.
[X] Regular Expression

(POSIX regular expressions assumed)

[gcotterl]

What would your example be if I'm looking for a comma in position 23?

[talleyrand]

The general form would be

Code: Select all

Find what ^.{N-1}\.

where N is the position.

Note, you must do the math and not put 58-1 or 23-1

[gcotterl]

Your example contains TWO periods.

^.{N-1}\.

Which one is the "Find" character?

[gcotterl]

I re-read the postings and noticed the reference to "POSIX regular expressions assumed".

I clicked on CONFIGURE/PREFERENCES/EDITOR and clicked on "Use POSIX regular expressions syntax".

It's now working!!

Thanks loads! This will save me many hours of work!!

[ben_josephs]

That was a slip.

It should be

Find what: ^.{N},

[X] Regular expression

where N is one less than the column position.

In your case:

Find what: ^.{22},

[X] Regular expression

In the search text, when you are searching using a regular expression, a simple dot ( . ) represents any single character (except newline); a quoted dot ( \. or [.] ) represents a literal dot.

This assumes you are using Posix regular expression syntax:

Configuration | Preferences | Editor

[X] Use POSIX regular expression syntax

[ben_josephs]

"Use POSIX regular expressions syntax"

In this example, it's just the braces ( {...} ) that are affected. The meanings of {...} and \{...\} are swapped.

[gcotterl]

What would be the "Find What" entry for finding a dash (i.e., "-") in position 77?

[ben_josephs]

Have a look through the preceding messages and try to find a pattern.

[gcotterl]

Hi Ben,

It looks like the character that I want to find follows the ^

So:

To find a dash, the entry would be: ^-{N},

To find an "BC", the entry would be: ^BC{N},

To find any single character (except newline), the entry would be: ^.{N},

To find a literal dot, the entry would be" ^\.{N}, or ^[.]{N},


correct?



I don't know what you mean by "In this example, it's just the braces
( {...} ) that are affected. The meanings of {...} and \{...\} are swapped).

[ben_josephs]

No. To find lines with a hyphen in position 77, you need: ^.{76}-

Code: Select all

^        beginning of a line
.        any character
.{76}    any character, 76 times; i.e., any 76 characters
^.{76}   any 76 characters at the beginning of a line
-        a hyphen
^.{76}-  any 76 characters at the beginning of a line, followed by a hyphen

To find lines with a dot in position 77, you need: ^.{76}\.

By "The meanings of {...} and \{...\} are swapped" I meant that they have opposite meanings in Posix and non-Posix syntax.

Edit: Added missing hyphen in last line of explanation.

[gcotterl]

So the character I want to find follows the last brace (instead of the ^). Right?

[ben_josephs]

I think that a more effective way for you to learn how to use regular expressions might be for you to try to understand how they are constructed, rather than repeatedly relying on others to give you a new expression each time there is a small change in your requirements.

If there is anything in the explanation I gave earlier that is not clear, please let me know, and I will try to express it more carefully.

[gcotterl]

(I'm sorry for the questions but I just want to understand the syntax of the expression so I can use it correctly).

Your previous posting says:
To find lines with a hyphen in position 77, you need: ^.{76}-
To find lines with a dot in position 77, you need: ^.{76}\.

So, I was correct in saying that "the character I want to find follows the last brace (instead of the ^)".

In your second example above:
}\. looks for a single DOT
}. looks for ANY character (except newline).