Mailing-List: contact user-help@ant.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Ant Users List" <user@ant.apache.org>
Received-SPF: neutral (nike.apache.org: local policy)
Date: Thu, 06 Sep 2007 09:31:21 -0700
From: Bruce Atherton <bruce@callenish.com>
Subject: Re: extract properties from file and corresponding template
In-reply-to: <EF39D77A486C234B9D9933091CE6D21E020F83CA@z011034.bk.fin.local>
To: Ant Users List <user@ant.apache.org>
Message-id: <46E02B59.7060900@callenish.com>
MIME-version: 1.0
Content-type: text/plain; charset=ISO-8859-1; format=flowed
Content-transfer-encoding: 8BIT
References: <EF39D77A486C234B9D9933091CE6D21E020F83CA@z011034.bk.fin.local>
User-Agent: Thunderbird 2.0.0.6 (X11/20070728)

Actually, if you wanted to do it programatically in code and you could 
guarantee that each line had no more than one token, you could probably 
make it work by creating regular expressions. Here is some pseudocode:

    for each line in the template file
        pull out the token name from the line
        create a regular expression from the line that anchors both ends 
and replaces the @token@ with a "(.*)"
        read equivalent line from the resulting file
        run regular expression on line
        if no match
            error out, this file was not generated from the template
        value of token is group 1
        register token name, token value in a Properties object (or 
equivalent dict or hash construct)
    write Properties object to disk

So in your example, the regular expression would be "^this (.*) file$" 
and the group 1 match would give you "is a sample".

Translated into Java, this could be a custom task. Or it could be done 
with some script in an ant build file.

You could even make it work with multiple tokens on a line so long as 
enough characters showed up between the tokens that would not also show 
up in the substitution values.

Jan.Materne@rzf.fin-nrw.de wrote:
> No there isnt.
>
> And I think this cant be done:
> If you combine the template with the property values you doesnt have any markers where a property value starts and where it ends (if your template doesnt provide these). Therefore you cant parse the resulting file.
>
> Example:
>
> template
>   this @value@ file
> property
>   value=is a sample
> result
>   this is a sample file
>
> How you would parse the "this is a sample file" string?
>
>
> Jan
>
>   
>> -----Urspr�ngliche Nachricht-----
>> Von: Patrick Martin [mailto:antuser@gmail.com] 
>> Gesendet: Donnerstag, 6. September 2007 10:18
>> An: Ant User
>> Betreff: extract properties from file and corresponding template
>>
>> Hello,
>>
>> I have a file that was generated from a template and a set of
>> properties using ant filters.
>> Basically, the template contained tokens like @my.property@ which
>> where replaced by the corresponding value of ${my.property}.
>> template + properties >>> file
>>
>> Now, I'd like to be able to get the set of properties back from this
>> file and the corresponding template.
>> file + template >>> properties
>>
>> Is there any task out there that already exists for that?
>>
>> Thank you,
>>
>> Patrick
>>
>> ---------------------------------------------------------------------
>> 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
>
>   

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