"sb" <stormfi***@yahoo.com> wrote in message
news:uBeWGCWxGHA.4576@TK2MSFTNGP03.phx.gbl...
> I'm using the following regex expression where "searchword" is a string:
> string regrexstr = @"( |^)\b" + searchword + @"+( |\b)"
>
> My goal is to use this string to ultimately replace all occurances of
> searchword as long as:
>    a) it has a preceeding space or it's the beginning of a line AND
>    b) it is suffixed with a space or a newline character
>
> The problem right now is that using the above string and calling
> Regex.Replace also replaces the leading/trailing spaces.  I want the
> spaces to count during the match, but be ignored during the replace.  Is
> that easily do-able?
>
> I'm probably going about this the wrong way.  In short, I want to do a
> word replacement throughout a large file.
>
> TIA!
> -sb
>

"Kevin Spencer" <u**@ftc.gov> wrote in message
news:OS55tOWxGHA.4960@TK2MSFTNGP04.phx.gbl...
> Hi sb,
>
> This should be solvable by using LookArounds. A LookAround is a
> non-capturing condition. That is, it defines a match without consuming the
> characters being "looked-ahead" or "looked-behind."
>
> (?=expression)    Positive LookAhead. Match must be followed by the
> expression.
> (?<=expression) Positive LookBehind. Match must be preceded by the
> expression.
> (?!expression)    Negative LookAhead. Match must NOT be followed by the
> expression.
> (?<!expression) Negative LookBehind. Match must NOT be followed by the
> expression.
>
> Example:
>
> (?<= )Some words(?!\.)
>
> This means that the phrase "Some words" must be preceded by a space, and
> must NOT be followed by a period:
>
> Some words.
> Some words?    *Match*
> Some words.
> Some words
>
> --
> HTH,
>
> Kevin Spencer
> Microsoft MVP
> Chicken Salad Surgery
>
> Orange you bland I stopped splaying bananas?
>
>
> "sb" <stormfi***@yahoo.com> wrote in message
> news:uBeWGCWxGHA.4576@TK2MSFTNGP03.phx.gbl...
>> I'm using the following regex expression where "searchword" is a string:
>> string regrexstr = @"( |^)\b" + searchword + @"+( |\b)"
>>
>> My goal is to use this string to ultimately replace all occurances of
>> searchword as long as:
>>    a) it has a preceeding space or it's the beginning of a line AND
>>    b) it is suffixed with a space or a newline character
>>
>> The problem right now is that using the above string and calling
>> Regex.Replace also replaces the leading/trailing spaces.  I want the
>> spaces to count during the match, but be ignored during the replace.  Is
>> that easily do-able?
>>
>> I'm probably going about this the wrong way.  In short, I want to do a
>> word replacement throughout a large file.
>>
>> TIA!
>> -sb
>>
>
>