Skip navigation
Currently Being Moderated

CFDocument Screwing Up Bulleted Lists from Word Documents

Mar 25, 2013 2:13 PM

Given a simple Word (2010) document (.docx) that is a buleted list of three lines of text using completely default formatting, CFDocument will generate a PDF that has bizzarre icons for the bullets.


We have 3 separate servers that behave in 3 different ways.

All 3 servers are running the same versions of Coldfusion (9) and OpenOffice (3.4.1).  They're all Windows Server 2008 R2 (64-bit) servers.


One server generates a PDF with clock symbols for the bullets.  0xB7 in Wingdings.

One server generates a PDF with diamond symbols for the bullets.  0xB7 in Wingdings 2.

One server generates a PDF with 35/17 symbols for the bullets.  0xB7 in MS Reference Specialty.


Word is using the Symbol font for the bullet character, and Calibri for the body.

All servers have the Symbol font listed in the Font Management of the CFIDE Administrator:

symbol Symbol SymbolMT TRUETYPE PDF/FlashPaper c:\windows\fonts\symbol.ttf


We're calling CFDcoument like so:








Changing FONTEMBED to NO doesn't change anything.  The generated PDFs still include an embedded subset of the respective fonts, and still display the incorrect symbol.


Should I be looking on the OpenOffice end?  Documents open up and display fine there, and all 3 servers have just a basic installation we havent changed any settings for.  We only use it for document conversion to PDF via Coldfusion.


Should I try installing Office 2010 on the web servers just in case it futzes with the fonts in some way?


If no one has any suggestions, my current plan is to reboot the servers at some point and try to kill off the Wingings and MS Reference Specialty fonts, in the hopes that Coldfusion will use the diamond symbols it uses on one of the servers (which is more acceptible looking than the other two options).  I have no idea why Coldfusion decides to pick whatever font it does when it reads the .docx file (which only uses Symbol and Calibri).

  • Currently Being Moderated
    Mar 25, 2013 2:36 PM   in reply to bw_bloodletter

    Just wondering, but does this process have the same failures if you try the conversion on an HTML file?  I'm wondering if all the bloat proprietary code that Word documents has is causing this issue.

    Mark as:

More Like This

  • Retrieving data ...

Bookmarked By (0)

Answers + Points = Status

  • 10 points awarded for Correct Answers
  • 5 points awarded for Helpful Answers
  • 10,000+ points
  • 1,001-10,000 points
  • 501-1,000 points
  • 5-500 points