Using regular expression to remove the parent paragraph of a placeholder in an RTF template

Question

I am working on contract module for my project. The contract templates are stored as RTF template with many placeholders with syntax as @placeholder_name@. Each campaign entry is associated with a particular contract template at any given time. When the contract for the campaign is requested/downloaded:

1. the RTF template is read as a variable.
2. placeholders within the file variable are replaced with the values from the query objects for the campaign.
3. the variable is then send to the browser for download using cfcontent.

Problem

I need to remove the whole section in the rtf file, if the value for a particular placeholder is empty. For example: the Additional Information section here:

I was able to find out the following rtf block inside the file, which is the whole section of Additional Information above including the rtf table styling.

\par \ltrrow}\trowd \irow0\irowband0\lastrow \ltrrow\ts78\trgaph108\trleft-
810\trbrdrt\brdrdot\brdrw10 \trbrdrl\brdrdot\brdrw10 \trbrdrb\brdrdot\brdrw10 \trbrdrr\brdrdot\brdrw10 \trbrdrh\brdrdot\brdrw10 \trbrdrv\brdrdot\brdrw10     \trftsWidth3\trwWidth11520\trftsWidthB3\trftsWidthA3\trautofit1\trpaddl108\trpaddr108\trpaddfl3\trpaddft3\trpaddfb3\trpaddfr3\tblrsid12942116\tbllkhdrrows\tbllkhdrcols\tbllknocolband\tblind-702\tblindtype3 \clvertalt\clbrdrt\brdrdot\brdrw10 \clbrdrl \brdrdot\brdrw10 \clbrdrb\brdrdot\brdrw10 \clbrdrr\brdrdot\brdrw10 \cltxlrtb\clftsWidth3\clwWidth3510\clshdrawnil \cellx2700\clvertalt\clbrdrt\brdrdot\brdrw10 \clbrdrl\brdrdot\brdrw10 \clbrdrb\brdrdot\brdrw10 \clbrdrr\brdrdot\brdrw10 
\cltxlrtb\clftsWidth3\clwWidth8010\clshdrawnil \cellx10710\pard \ltrpar\ql \li0\ri0\sa200\widctlpar\intbl\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\pararsid4544034 {\rtlch\fcs1 \af1 \ltrch\fcs0 \insrsid3568873 Additional Information}{
\rtlch\fcs1 \af1 \ltrch\fcs0 \insrsid4544034 \cell }\pard \ltrpar\ql \li0\ri0\sa200\widctlpar\intbl\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0 {\rtlch\fcs1 \af1 \ltrch\fcs0 \insrsid3568873\charrsid4544034 @additional_contract_info@}{
\rtlch\fcs1 \af1 \ltrch\fcs0 \insrsid4544034 \cell }\pard \ltrpar\ql \li0\ri0\widctlpar\intbl\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0 {\rtlch\fcs1 \af1 \ltrch\fcs0 \insrsid4544034 \trowd \irow0\irowband0\lastrow \ltrrow
\ts78\trgaph108\trleft-810\trbrdrt\brdrdot\brdrw10 \trbrdrl\brdrdot\brdrw10 \trbrdrb\brdrdot\brdrw10 \trbrdrr\brdrdot\brdrw10 \trbrdrh\brdrdot\brdrw10 \trbrdrv\brdrdot\brdrw10 
\trftsWidth3\trwWidth11520\trftsWidthB3\trftsWidthA3\trautofit1\trpaddl108\trpaddr108\trpaddfl3\trpaddft3\trpaddfb3\trpaddfr3\tblrsid12942116\tbllkhdrrows\tbllkhdrcols\tbllknocolband\tblind-702\tblindtype3 \clvertalt\clbrdrt\brdrdot\brdrw10 \clbrdrl
\brdrdot\brdrw10 \clbrdrb\brdrdot\brdrw10 \clbrdrr\brdrdot\brdrw10 \cltxlrtb\clftsWidth3\clwWidth3510\clshdrawnil \cellx2700\clvertalt\clbrdrt\brdrdot\brdrw10 \clbrdrl\brdrdot\brdrw10 \clbrdrb\brdrdot\brdrw10 \clbrdrr\brdrdot\brdrw10 
\cltxlrtb\clftsWidth3\clwWidth8010\clshdrawnil \cellx10710\row }\pard \ltrpar\ql \li0\ri0\sa200\widctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0 {\rtlch\fcs1 \af1 \ltrch\fcs0 \insrsid4544034 
\par }

I have been working on it to find a solution for days now. What I need is a regular expression statement in ColdFusion to find the block of \par control word wrapped around the placeholder @additional_contract_info@ i.e Only the parent para for the placeholder:

the portion: "\par ...@additional_contract_info@ ...." until the ending \par

Assuming the paras are not nested.

I am not very proficient with Regular expression and I tried googling and searching SO for all types of related questions but couldn't solve it. I need help with it!

Answer 1

Try:

\\par\b((?!\\par\b).)*@additional_contract_info@.*?\\par\b

• \b matches a word boundary so you don't match \pard.
• (?!\\par\b). will first do a negative look-ahead to ensure that there are no other instances of \par between the start of the match and the @ then will consume a single character. Repeating this will match the entire string between the most recent \par and the @.
• After the final @ you can just use a non-greedy wildcard match .*? (so it will only match the minimum number of characters) to find the ending paragraph code.

Example:

<cfscript>
  str = '\par \par \pard text \par \pard text @additional_contract_info@ text \pard \par text \pard \par } \par }';
  output = REReplace( str, '\\par\b((?!\\par\b).)*@additional_contract_info@.*?\\par\b', '' );
  WriteOutput( output );
</cfscript>

Should output:

\par \par \pard text  text \pard \par } \par }

Update:

You can also try doing it without regular expressions:

<cfscript>
  str      = '\par \par \pard text \par \pard text @additional_contract_info@ text \pard \par text \pard \par } \par }';
  pos      = find( '@additional_contract_info@', str );
  endPos   = find( '\par ', str, pos ) + 4;
  startPos = left( str, pos ).lastIndexOf( '\par ' );
  output   = left( str, startPos ) & right( str, len( str ) - endPos + 1 );
  WriteOutput( output );
</cfscript>

(Note: this assumes that you will always find \par with a trailing space, whereas the Regular Expression looks for word boundaries, if this isn't the case then you might need to try other ways of finding the boundaries of the text to remove).