Unicode PDF Output Looks Wrong -- How to correct?

rbuser · May 2010

Hello,

http://www.ipsc-tech.org/distribution/ReportBuilder.zip

The URL above leads to two reports that were generated by the very same
ReportBuilder report generating code, except that was was saved to a PDF
file and the other was sent to the doPDF.com PDF print driver to be saved
as a PDF file (with EmbedFontOptions being AllFonts and FontEncoding
being Unicode).

The doPDF.com PDF file looks great!!!

The ReportBuilder output saved as a PDF file looks terrible!!!

What am I doing wrong (what do I need to do to get ReportBuilder to generate
doPDF.com -like output, without the use of doPDF)?

I am using the following:

* CodeGearT Delphi® 2009 Professional
* ReportBuilder 12 Standard Edition
* Windows XP Professional SP1 (32-bit)

With regards from,
Christopher

nicocizik · May 2010

Hi Christopher,

For future reference, please either post a question to the newsgroups or
send it to support@, not both.

Unicode for PDF is a very complex feature and we are currently working on a
number of enhancements and fixes that will improve its output. Current
there are a few steps you can take to produce a valid PDF.

1. Try using a different font for the text objects that doe not display
correctly. The reason the characters are not displaying is that the font
you are currently using does not actually include them. We are working on a
solution to this issue however currently you can try changing the font to a
major Unicode font such as Arial Unicode MS and see if that helps.

2. Use the font subset feature by enabling the efUseSubset option of the
Report.PDFSettings.EmbedFontOptions property. This should drastically
decrease the size of the file being generated.

3. There is a patch available that fixes some of the text spacing issues.
I sent it to your email address.

--
Regards,

Nico Cizik
Digital Metaphors
http://www.digital-metaphors.com

rbuser · June 2010

Nico Cizik (Digital Metaphors) wrote:

I am also waiting for the patch (I have sent an e-mail to support).
Does it also address the problem that efUseSubset seems to have no
effect when the font encoding is Unicode? My PDFs with Unicode
encoding are always 20MB or so, as well as having completely
unacceptable spacing.

My principal reasons for spending the money to upgrade to 12 were to
get Unicode PDF support and built-in TIF support. The former is
unusable and the latter is still somewhat primitive. However we have
been able to update our home-brew TIF support from version 11 and build
a useful replacement for ppTIFFDevice; this is a worthwhile improvement
over the hacks we had made to ExtraDevices for earlier RB releases.

nicocizik · June 2010

Hi Tim,

I'm very sorry you are having issues with Unicode and PDF.

1. I sent a copy of the patch for the spacing issues on Friday and I just
resent another copy. Please check your junk email folder to be sure it is
not being placed in there for some reason.

2. The current release of ReportBuilder (12.0) currently has issues
creating subsets of TTC (TrueType Collection) fonts. These are generally
Asian fonts that have the file extension .ttc. We are working to fix this
issue and hopefully have it ready for the next maintenance release.
Currently however Unicode TTF fonts should function correctly. Try using a
.ttf font such as Arial Unicode MS as a temporary workaround.

3. Please let us know what enhancements you would like added to the TIFF
device. The next release will include single file TIFF export meaning
multiple report pages can be exported to the same TIFF file. I would be
happy to research and possibly add any other features you would like to see.

--
Regards,

Nico Cizik
Digital Metaphors
http://www.digital-metaphors.com

rbuser · June 2010

Nico Cizik (Digital Metaphors) wrote:

Thanks for the quick response. I got one copy of the patch, anyway. The
spacing of text in the report is now fixed and looks good.

I am using Arial Unicode MS and still creating 25MB PDF files. One
detail-band field on our 3-page test-data report can contain non-ASCII
characters and is in this font, the remainder are Arial and some
headings are Times Roman. The Acrobat PDF properties page reports all
three fonts as having embedded subsets, but there are a lot of subset
entries for each font on the Fonts page (several hundred entries in
all). An ANSI-encoded PDF contains only three subset entries on the
properties/fonts page (Arial, Times New Roman, and Times New Roman
Bold) when I view the PDF in Acrobat.

Thanks, I will take this to e-mail. We are creating directly-faxable
reports in TIF files.

Unicode PDF Output Looks Wrong -- How to correct?

Comments