Stewart Stremler wrote:
begin quoting Ralph Shumaker as of Thu, Feb 01, 2007 at 09:59:22PM -0800:
[snip]
How would I search for every numerical sequence *not* followed by a
space, and append a space after each such instance?
In vi(m):
:%s/\([0-9][0-9]*\)\([^ 0-9]\)/\1 \2/g
: -> command mode
% -> for every line in the file
s -> substitute
/ -> use / for delimiter
\( -> start first group
[0-9] -> any digit
[0-9] -> any digit
* -> 0 or more of previous character
(This gives the effect of [0-9]+)
\) -> end first group
\( -> start second group
[^ 0-9] -> every character except space and digits (and newlines)
\) -> end second group
/ -> end "find" part, start "replace" part
\1 -> first group
\2 -> second group
/ -> end pattern
g -> global, that is, replace all matches, not just the first on
each line
The key is the [^ ] construct. [0-9] says "match any characters
zero through nine", while [^0-9] says "match any characters NOT
0-9 or newline" (since vi(m) is basically line-oriented).
I wish I had gone online and downloaded my email to see this before I
found my own solution. I did man regex, and then did /\. and looked for
any occurence of "." that was not end of sentence. I just had a feeling
that it would yield what I wanted. I quickly found (and relearned) the
utility of "^" as the NOT.
I was able to follow along with the formula you gave above (even without
the explanation this time, although the bit about newlines was
informative). I think I need to start using regex a lot more often so
as to become proficient, learn more, and not forget what I already know
(i.e. ^==NOT and that "0-9" can be combined with " ").
Although I would change (of your formula) the final g to c.
Thanks again.
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
