ant-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dick, Brian E." <Brian.D...@FMR.com>
Subject RE: Cleaning a text file with tons of extra CRLF's
Date Thu, 13 Jan 2005 23:18:51 GMT
If you use OS pipes, you will appreciate filterchains.

-----Original Message-----
From: Nat Gross [mailto:natgross.rentalsystems@verizon.net] 
Sent: Thursday, January 13, 2005 4:41 PM
To: Ant Users List
Subject: Re: Cleaning a text file with tons of extra CRLF's


filterchains? I will need to catch up on that topic.
Thanks for the tip.
-nat

Matt Benson wrote:

>Also, have you tried using filterchains?  You can use
>the ignoreblank filter to remove blank lines.  And it
>might look a little prettier.
>
>-Matt
>
>--- Nat Gross <natgross.rentalsystems@verizon.net>
>wrote:
>
>  
>
>>Ok. Releasing ver 1.0RC <g>. (Using your reminder on
>>"
>>    
>>
>".)
>  
>
>>Also capitalizing on the fact that its an xml target
>>that needs the 
>>messaging, I can safely replace every single CRLF
>>with ....null, "", 
>>nada. Then replace every "/>" with "/>LF". Its real
>>simple and works, 
>>and achieves 99.9% of my goal. Here it is.
>>********
>><project name="replaceDupeLF" default="default"
>>basedir=".">
>>    <target name="default">
>>        <replace file="${basedir}/build370.xml" 
>>token="
>>    
>>
>" 
>  
>
>>value="" summary="true"></replace>
>>        <replace file="${basedir}/build370.xml" 
>>token="/&gt;" 
>>value="/&gt;
>>    
>>
>" summary="true"></replace>
>  
>
>>    </target>
>></project>
>>********
>>Thank you;
>>-nat
>>
>>
>>Julius Davies wrote:
>>
>>    
>>
>>>Hi, Nat,
>>>
>>>I never would have thought of that!  Really neat!
>>>
>>>Actually, "
>>>      
>>>
>" is a CR, not 015.
>  
>
>>>yours,
>>>
>>>Julius
>>>
>>>On Wed, 2005-01-12 at 21:25 -0500, Nat Gross wrote:
>>> 
>>>
>>>      
>>>
>>>>Thanks, but for whatever reason, I can't get 
>>>>        
>>>>
>>replacetoken to work, 
>>    
>>
>>>>regardless if using CDATA or plain strings.
>>>>BUT, you pointed me in the right direction and the
>>>>        
>>>>
>>following *partially* 
>>    
>>
>>>>works:
>>>><replace file="${basedir}/build370.xml" 
>>>>        
>>>>
>>token="
>>    
>>
>
>" 
>  
>
>>>>value="
>>>>        
>>>>
>" summary="true">
>  
>
>>>>However, it's only a partial solution, because it
>>>>        
>>>>
>>takes care of only 
>>    
>>
>>>>about half the white space, and heaven knows what
>>>>        
>>>>
>>the other white space 
>>    
>>
>>>>really is. Alas there is no valid  escape
>>>>        
>>>>
>>sequence. (That would 
>>    
>>
>>>>represent a CR).
>>>>
>>>>Also to Martin, alas, I'm not well versed in
>>>>        
>>>>
>>regex.
>>    
>>
>>>>-nat
>>>>
>>>>
>>>>Julius Davies wrote:
>>>>
>>>>   
>>>>
>>>>        
>>>>
>>>>>Try this:
>>>>>
>>>>><replace file="test.txt" value="">
>>>>><replacetoken><![CDATA[
>>>>>
>>>>>]]></replacetoken>
>>>>><replacevalue><![CDATA[
>>>>>]]></replacevalue>
>>>>></replace>
>>>>>
>>>>>Be careful to make sure the CDATA only contains
>>>>>          
>>>>>
>>the EOL's and no extra
>>    
>>
>>>>>whitespace!
>>>>>
>>>>>BEFORE:
>>>>>=======================
>>>>>this file
>>>>>
>>>>>seems to be
>>>>>
>>>>>double spaced
>>>>>
>>>>>except for the
>>>>>very last line.
>>>>>
>>>>>
>>>>>AFTER:
>>>>>=======================
>>>>>this file
>>>>>seems to be
>>>>>double spaced
>>>>>except for the
>>>>>very last line.
>>>>>
>>>>>
>>>>>yours,
>>>>>
>>>>>Julius
>>>>>
>>>>>On Wed, 2005-12-01 at 19:34 -0500, Nat Gross
>>>>>          
>>>>>
>>wrote:
>>    
>>
>>>>>     
>>>>>
>>>>>          
>>>>>
>>>>>>Hi;
>>>>>>On Windows, I get a daily text file from another
>>>>>>            
>>>>>>
>>process that has tons 
>>    
>>
>>>>>>of extra EOL's, which I manually delete with my
>>>>>>            
>>>>>>
>>editor.
>>    
>>
>>>>>>How can I use Ant's filter to copy the file and
>>>>>>            
>>>>>>
>>strip all duplicate 
>>    
>>
>>>>>>eol's? The fixeol task does not cut it. It
>>>>>>            
>>>>>>
>>should have an option to 
>>    
>>
>>>>>>remove the eol, or replace every 2 consecutive
>>>>>>            
>>>>>>
>>eol's with 1.
>>    
>>
>>>>>>If its a quick one for you, please advise.
>>>>>>Thank you.
>>>>>>-nat
>>>>>>            
>>>>>>


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@ant.apache.org
For additional commands, e-mail: user-help@ant.apache.org

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@ant.apache.org
For additional commands, e-mail: user-help@ant.apache.org


Mime
View raw message