cfdocument does not respect page-break-inside:avoid

I found a way to do it, and although it's far from ideal, it allows for choosing any group of data to not break over a page if possible. The basic principle is to put divs around all of the nobreak data groups with a class on them, pop all of the text into a div appended to body, and .each() the class with a height counter that adds up the heights until it goes over a page, then inserts chr(30) and resets the height counter. Then submit the modified html to coldfusion, run a loop on a list with chr(30) as the delimiter, and make each page followed by cfdocumentitem type=pagebreak.

JS Code:

function paginateForPdf(report_body){   // Put a div on the page to render heights   $('body').append('<div id="pdf_render_container"></div>');   // Put the response into the div   $('#pdf_render_container').show().html(report_body);   // Summing variable for each section's height   var page_height = 0;   var page_height_max = 762;   // Container for the assembled html of the pages and page break characters   var assembled_html = "";   // Iterate over the unbreakable groups   $('#pdf_render_container .height_group').each(function(){   // Add the page height of this div to the previous total   if(page_height <= page_height_max){ // Don't page break if it's a really big one   page_height += $(this).height();   // If more than one page   if(page_height > page_height_max){   // Add a record separator before this div so page is not overfull   assembled_html += String.fromCharCode(30);   // Set the page height to this div only   page_height = $(this).height();   }   } else page_height -= page_height_max;   // Add this div's contents to the assembled html   assembled_html += $(this).html();   });   // Purpose served, remove the renderer div   $('#pdf_render_container').remove();   return assembled_html; }

CF document:

<cfdocument format="pdf" name="report_PDF"> <style> <!--- Rendering Style here ---> </style> <cfset var counter = 1> <!--- Loop over the assembled html with the record separator as the list delimiter to get whole pages ---> <cfloop list="#report_body#" delimiters="#chr(30)#" index="page"> #Trim(page)# <!--- Trim to prevent hanging spaces/newlines from tripping into the next page ---> <cfif counter LT ListLen(report_body,#chr(30)#)><!--- don't break after the last page ---> <cfdocumentitem type="pagebreak"></cfdocumentitem> <!--- page break after a page ---> </cfif> <cfset counter ++> </cfloop> </cfdocument>

All you need to do is run an ajax call to get your html (or not if it's already on the page), submit all of it to the paginate function, and submit the returned result to the coldfusion code

762 seems to be the right number for the page height for me, not sure if that's universal. Hope this helps!

We had a similar problem with table rows inside a cfloop query and tried to use, to no avail as we are on CF9 as well

This is what we did to resolve it - placing this code right before the item that is breaking in the undesired location

<span> </span><BR>

Apparently, no browser, except Opera, supports page-break-inside.

Check out Raymond Camdens pdfutils (http://pdfutils.riaforge.org/)
It allows the text from a pdf to be read page by page.

It has enabled me to do a workaround fix for the problem where cfdocument does not support
style="page-break-inside: avoid;"

The following code is not overly efficient, in that it recreates the entire report for each instance where
a page break needs to be repositioned, but the preliminary test results look good. Others may like to embellish the following.
It creates hidden fields in the pdf, formatted as
keep_together_start_001, keep_together_end_001, keep_together_start_002, keep_together_end_002 etc
If the matching tags are not on the same page, then the pdf is recreated with an appropriate page break.

In the style sheet
div.hidden1{position:absolute;
left:-3000px;
top:auto;
width:1px;
height:1px;
overflow:hidden;}

div.hidden2{position:absolute;
left:-4000px;
top:auto;
width:1px;
height:1px;
overflow:hidden;}

On initialisation of the page that creates the pdf
<cfif not isdefined("url.keep_together")>
     <cfset session.keep_together_list = "">
     <cfset keep_together_count = 0>
</cfif>

...

At the start of a section of output that I want kept together, put the following

  <cfset keep_together_count += 1>
  <cfif ListFind(session.keep_together_list,"#NumberFormat(keep_together_count,"000")#","|") gt 0>
        <cfdocumentitem type="pagebreak"></cfdocumentitem>
  <cfelse>
        <cfoutput>
              <div class="hidden1">keep_together_start_#NumberFormat(keep_together_count,"000")#</div>

        </cfoutput>
  </cfif>

...

At the end of the section to be kept together

<cfif ListFind(session.keep_together_list,"#NumberFormat(keep_together_count,"000")#","|") eq 0>
     <cfoutput>
            <div class="hidden2">keep_together_end_#NumberFormat(keep_together_count,"000")#</div>

     </cfoutput>
</cfif>

Once the PDF file has been created, use Raymond's pdfutils to check each page for a start without a matching end.
If found, indicate which section of code need to have a page break inserted before it, and then reproduce the report.

<!--- Check for keep_togethers that were split --->
<cfset pdf = createObject("component", "pdfutils")>
<cfset mypdf = expandPath("my_report.pdf")>
<cfset pdf_struct = pdf.getText(mypdf)>

<cfloop index="idx" from="1" to="#ArrayLen(pdf_struct)#">
     <cfset next_start = 1>
     <cfset start_offset = findnocase("keep_together_start_", pdf_struct[idx].text, next_start)>
     <cfloop condition="start_offset neq 0">
            <cfset next_start = start_offset + 20>
            <cfset keep_together_id = Mid(pdf_struct[idx].text,start_offset + 20,3)>
            <cfset end_offset = findnocase("keep_together_end_#keep_together_id#",pdf_struct[idx].text)>
            <cfif end_offset eq 0>
                  <cfset session.keep_together_list = session.keep_together_list & "|" & keep_together_id>
                  <cflocation addtoken="No" url="sft_report.cfm?keep_together">
            </cfif>
            <cfset start_offset = findnocase("keep_together_start_", pdf_struct[idx].text, next_start)>
     </cfloop>
</cfloop>

Message was edited by: Phil0707 - tidied up a few things - works well now

Increasing the page margin top seems to have helped me. But again it's not fail proof.

I'm curious about something.

You have nested CFLOOPS, and you give them both the same index:

  <cfloop from="1" to="10" index="i"> 

   <cfloop from="1" to="10" index="i">

...

   </cfloop>

  </cfloop>

??

^_^

I believe this is a known problem that any amount of google searching will result in the answer to. I think there is a patch that needs to be applied

to the CF server that helps with this.

I am not at my desk right now to validate, so don't quote me on that though.