Re: regular expression problem

Hi,

It works fine.
Thank you for your help.

Eric.

Le 22/04/2010 13:43, Eddy De Greef a écrit :
> The problem is that the '.+' matches all characters greedily,
> including the comma. To make it non-greedy, use '.+?':
>    replace_in_selection("^(.+?),*$", "input\t\\1,", "regex")
>
> Alternatively, you can use:
>    replace_in_selection("^([^,]+),*$", "input\t\\1,", "regex")
> This will prevent any commas from being consumed by the part between
> brackets.
>
> Eddy
>
> On 04/22/2010 01:27 PM, Eric Bouyoux wrote:
>> Hi,
>>
>> Thank you for your answer. What I do is the following, which seems to be
>> more or less the same as your last example :
>> replace_in_selection("^(.+),*$", "input\t\\1,", "regex")
>>
>> The idea is : I take what is in (), I add "input\t" before and ","
>> after.
>> But it seems to be a greedy expression which goes to the end of line and
>> does not stop to the first ",".
>>
>> Regards.
>>
>> Eric.
>>
>> Le 22/04/2010 10:24, Eddy De Greef a écrit :
>>> I guess that you used something like
>>>     replace_all("toto|toto,", "input toto,", "regex")
>>> in which case the first match takes precedence.
>>>
>>> You could reverse the order:
>>>     replace_all("toto,|toto", "input toto,", "regex")
>>> or better yet, use something like
>>>     replace_all("toto,?", "input toto,", "regex")
>>> where the comma is quantified (quantified matching is greedy by
>>> default). That should give you the desired behaviour.
>>>
>>> Eddy
>>>
>>> On 04/22/2010 09:43 AM, Eric Bouyoux wrote:
>>>> Hi,
>>>>
>>>> I have created a small macro to help me to modify some Verilog
>>>> files to
>>>> replace a line that contains some characters followed or not by a
>>>> comma
>>>> by the same line starting with 'input' and ending by a comma.
>>>> Ex :
>>>> the line contains :
>>>> toto
>>>> or the line contains :
>>>> toto,
>>>>
>>>> In both case, I want to replace it by :
>>>> input toto,
>>>>
>>>> The result I get is :
>>>> if the line does not contain any comma, then it works.
>>>> If the line already contains a comma, the resulting line contains 2
>>>> commas.
>>>>
>>>> Could you help me ?
>>>>
>>>> Regards.
>>>>
>>>> Eric.
>>
>
>

-- 
NEdit Discuss mailing list - [email protected]
http://www.nedit.org/mailman/listinfo/discuss