Jeremy Conlin schrieb:
> On Fri, Sep 25, 2009 at 12:07 AM, Christian Brabandt
> <[email protected]>wrote:
>
>> On Fri, September 25, 2009 7:43 am, Jeremy Conlin wrote:
>>> I know there are some regex wizards out there, I am in need of your help.
>>> I have lines in my file that look like one of the following two
>>>
>>> Some text & more text & (some white space) & ...
>>> Some text & more text & (some numbers, :, or -) & ...
>>>
>>> For the life of me I can't get my regular expression to differentiate
>>> between the two. This is what I tried (but it found both lines):
>>>
>>> \_^\(.\{-}\s&.\{-}&\s*\)\(\s\{-}[\d:-]\{1,}\)
>> You are at least missing the paranthesis in your pattern and I believe
>> the \d form is not valid inside [].
/[[:digit:]:-] or /[0-9:-] will do.
>> Using the following pattern
>>
>> ^\(.\{-}\s&.\{-}&\s*\)\((\s\{-}[0-9:-]\{1,})\)
>>
>> works, assuming you want to match the following line:
>> Some text & more text & (12349:-) & ...
It's confusing to require a match for the surrounding parens ...
>> regards,
>> Christian
>>
>
> That didn't quite work and it's my fault for not giving better examples.
> Here are two lines:
>
> The genealogies & & 1:2--17 & & 3:23--38 & & \\
> Elisabeth's seclusion & Judaea: Bethlehem & & & 1:24--25 & & \\
>
> I want my regular expression to find the first line and not the second
> because there is non-whitespace between the second and third ampersand.
> What I'm trying to do is replace "1:2--17" with "\index{1:2--17}"
>
> Is that example more clear?
>
> Thanks,
> Jeremy
this way?
pattern:
/^\%([^&]*&\)\{2}\s*[0-9:-]\+
substitute command (with \zs added in the pattern):
:%s/^\%([^&]*&\)\{2}\s*\zs[0-9:-]\+/\\index{&}/
will not work if "&" somehow occurs escaped as part of an entry.
--
Andy
--~--~---------~--~----~------------~-------~--~----~
You received this message from the "vim_use" maillist.
For more information, visit http://www.vim.org/maillist.php
-~----------~----~----~----~------~----~------~--~---
