Though the Fonts are not right, as when I try to copy and paste from the PDF
to Notepad, all I get is garbage characters.
When running the Accessibility on the Document it reports:
"None of the text in this document maps to Unicode reliably"
The encoding on the Type 1 Fonts is Custom and it says they are an embedded
subset.
We are using an Arial font throughout the Document.
The Ghostscripts that are setup do not seem to handle TrueType Fonts be
default. I'm not sure were in the Scripts to add the commands to process the
TTF font files. I can see where it can handle them...
Printing the Same Document using Adobe PDF Writer produces a document
without the embedded subset fonts and I can copy and paste just fine.
Any tips would be appreciated.
Scott<-
> The Ghostscripts that are setup do not seem to handle TrueType Fonts be
> default. I'm not sure were in the Scripts to add the commands to process the
> TTF font files.
> Any tips would be appreciated.
Don't embed True Type fonts with Ghostscript.
Sounds like you have GS set to create a Screen PDF. For size reduction
GS automatically substitutes for fonts which come with AR. Helvetica
and Arial are the same font and since Helvetica comes with AR, Arial is
replaced. Use either prepress or printer to have all fonts embedded.
First make sure the change is not a setting in your printer driver.
The default for the Distiller driver to replace Arial with Helvetica.
Also, you can edit the lib files for GS to stop the substitution if you
have turned off the driver substitution.
Phil
fabrizio
> Helvetica and Arial are the same font
Hmm, it may be the case that 99.9% of people cant tell the difference or
dont care, but they are different.
http://www.ms-studio.com/articlesarialsid.html
> > Helvetica and Arial are the same font
>
> Hmm, it may be the case that 99.9% of people cant tell the difference or
> dont care, but they are different.
I should have phrased that better. Years ago I went through an
exercise on another list of explaining how similar fonts were not the
same font--in that case it was differences between TTF and Type1 TNR
fonts. I used an OS/2 font tool to provide the differences.
> http://www.ms-studio.com/articlesarialsid.html
Thanks for the article, it was a good and informative read.
Phil
This is supposed to say "Thank You" "7KDQN \RX"
It looks fine on the Screen. Here is a link to the file:
http://www.enm.com/quote.pdf
Thanks,
Scott<-
"PGAGA" <grif...@glinx.com> wrote in message
news:1141096531.5...@j33g2000cwa.googlegroups.com...
That's what's going to happen when you use Ghostscript to embed fonts.
I wish people would stop doing it. The world doesn't need any more bad PDFs.
(And if the big white rectangle with the red border was intended to
"redact" the doc, that's a bad idea too.)
On Tue, 28 Feb 2006 19:08:47 -0500 PDFrank <pdf...@some.net> wrote:
> That's what's going to happen when you use Ghostscript to embed fonts.
>
> I wish people would stop doing it. The world doesn't need any more
> bad PDFs.
Sorry, but there's nothing wrong with the fonts embedded. But you are
continuing to spread FUD on ghostscript. The fonts are, in fact,
perfectly embedded. The only thing that stops copy & paste is a known
charset which is set to "user defined". It's still a perfectly valid
PDF.
Why are you continuously badmouthing ghostscript? Don't tell me things
that apply to versions that are outdated today...
-hwh
... and, the fonts seem to be converted from TrueType to Type1. If you look into the PDF, you will see
font names like Arial_Black0233, and in the encoding you see glyph names like G055 etc., which obviously
are generated by some font conversion process or a (windows-based?) printer driver.
PDFrank, Please tell us the history of the file's creation (application, platform, etc.).
Helge
--
Helge Blischke
Softwareentwicklung
SRZ Berlin | Firmengruppe besscom
http://www.srz.de
We are using a Programming Language Called Powerbuilder from Sybase. We are
printing directly to a Locally defined printer, "Sybase DataWindow PS". that
printer is using the Adobe Distiller as its Driver.
I cannot seem to print directly to the defined printer and generate a PDF.
The Code in the Application requires that there be a GS folder in the
Application Directory and in that folder there needs to be a folder with the
name of "gs7.06" though the version of ghostscript that is in there does
not seem to matter.
When I started with version 8.15 the Fonts that showed up in the Document
Properties were Type3 and had names like T3Font_6. I upgraded to 8.53 and
that is when they Switched to Type1
I don't really care if they are embedded or not. I don't care if I need to
change the Font to Helvetica or something similar. When our sales people
create the Quote, they want to be able to copy and paste information from
the PDF to another document.
I know that somehow the Powerbuilder code or the "Sybase DataWindow PS"
printer is calling/using the scripts in the lib folder as I can change
things in them and get different results in the PDF.
What do I need to change to let it know that I want to use a normal encoding
for the fonts so they can be copy and pasted?
Thank you,
Scott<-
"Helge Blischke" <h.bli...@srz.de> wrote in message
news:44059C29...@srz.de...
I know it was probably lame to most of you who know how to manipulate a PDF,
but the big white rectangle with the red border was to cover the customer
info on the quote.
If I knew what fonts to use to not embed them and what options/changes I
need to make to allow them to be normally encoded so I can copy and paste
text from it, I'd be more then happy to make the changes.
Thank you,
Scott<-
"PDFrank" <pdf...@some.net> wrote in message
news:14qdnTVfU_m...@rcn.net...
On Wed, 01 Mar 2006 15:48:36 GMT "Scott Townsend"
<scott-i@.-N0-SPAMplease.enm.com> wrote:
> As I mentioned in my other reply, it does not matter to me if they
> are embedded or not.
Well, it doesn't matter as long the RIP does The Right Thing (TM)...
> The Smaller the file the better actually. I'm
> not familiar with Ghostscript. Our Programming environment just says,
> put this folder here and print to ""Sybase DataWindow PS" and it will
> work. Other then that there is no other info on how to adjust the
> creation of the PDF.
Googling a bit I found this:
http://manuals.sybase.com/onlinebooks/group-dwnet/dnetp0150e/dwninst/@Generic__BookTextView/1159;pt=168
I guess this is what you're speaking about?
Did you install the mentioned adist5.inf (Adobe Distiller profile, I
guess) printer driver or a custom (other) one?
> If I knew what fonts to use to not embed them and what
> options/changes I need to make to allow them to be normally encoded
> so I can copy and paste text from it, I'd be more then happy to make
> the changes.
Start with trying to print with another app to the "Sybase DataWindow
PS" driver. Maybe you can put the results for a few fonts online, too,
so we have a chance to inspect them.
BTW, whereas pdftotext (from xpdf suite) doesn't manage to extract the
text, Adobe Acrobat doesn't show any problems for me with the sample
doc you've linked.
-hwh
> Sorry, but there's nothing wrong with the fonts embedded. But you are
> continuing to spread FUD on ghostscript. The fonts are, in fact,
> perfectly embedded. The only thing that stops copy & paste is a known
> charset which is set to "user defined". It's still a perfectly valid
> PDF.
>
> Why are you continuously badmouthing ghostscript? Don't tell me things
> that apply to versions that are outdated today...
I like Ghostscript. I use Ghostscript.
I don't like it when people use Ghostscript to make "perfectly valid"
PDFs (with "perfectly embedded" fonts) that are all screwed up.
Because then I have to go through the trouble of fixing them to make
them searchable.
On Wed, 01 Mar 2006 23:06:07 -0500 PDFrank <pdf...@some.net> wrote:
> > Why are you continuously badmouthing ghostscript? Don't tell me
> > things that apply to versions that are outdated today...
>
> I like Ghostscript. I use Ghostscript.
>
> I don't like it when people use Ghostscript to make "perfectly valid"
> PDFs (with "perfectly embedded" fonts) that are all screwed up.
>
> Because then I have to go through the trouble of fixing them to make
> them searchable.
:-) OK, so it was just an accidental rant... I can pretty well
understand that. And in fact, you're quite right if I understand the
documentation page "ps2pdf.htm" from CVS version correctly:
--snip--
Ghostscript has been writing incorrect ToUnicode CMap without CMapName
into the PDF since version 8.10 (rev. 3611) . This bug is fixed in
version 8.54 (rev. 6201). We recommend to re-generate PDF files created
by the affected Ghostscript versions. Since version 8.54 (rev. 6590)
Ghostscript can read the incorrect PDF files.
--snip--
So I let you continue to rant ;-)
-hwh
"Hans-Werner Hilse" <hw.h...@stud.ugoe.de> wrote in message
news:20060301185910.a...@stud.ugoe.de...
> Googling a bit I found this:
> http://manuals.sybase.com/onlinebooks/group-dwnet/dnetp0150e/dwninst/@Generic__BookTextView/1159;pt=168
>
> I guess this is what you're speaking about?
Yes, that is what We have done.
>
> Did you install the mentioned adist5.inf (Adobe Distiller profile, I
> guess) printer driver or a custom (other) one?
The printer "Sybase DataWindow PS" is created automaticaly when you print if
its not already there. Its Driver is Acrobat Distiller.
I've also created a printer using the Supplied Adist5.inf file.
>
> Start with trying to print with another app to the "Sybase DataWindow
> PS" driver. Maybe you can put the results for a few fonts online, too,
> so we have a chance to inspect them.
I've printed test pages with both and both are not readable by Acrobat.
They do have PS Contents in the file, but something is not formatted right.
>
> BTW, whereas pdftotext (from xpdf suite) doesn't manage to extract the
> text, Adobe Acrobat doesn't show any problems for me with the sample
> doc you've linked.
Yeah, the Documents look great... The customers do not seem to complain
when they get them, though the sales people want to be able to copy and
paste info from them, and that's the real issue.
I've Posted my Docs here: http://www.enm.com/pdfissues/
I've added a few new Quotes, one from GS 8.15 and one from GS 8.53. The
funny thing was when I tried to Downgrade my 8.53 install back to 8.15, the
fonts when printed with the 8.15 are embedded like the 8.53. I had to go to
another machine to get the real 8.15 output.
So should I install 8.54?
Thank you for your assistance in this. It is appreciated!
Scott<-
On Thu, 02 Mar 2006 15:42:38 GMT "Scott Townsend"
<scott-i@.-N0-SPAMplease.enm.com> wrote:
> I've linked all the files I mention here:
> http://www.enm.com/pdfissues/
OK, luckily, those aren't PDFs but the originating Postscript. Thus I
can now tell that the "garbage" is produced by the printer driver
(adist5.ppd in combination with pscript.dll) which seems to be very
outdated (v.4.00, according to the test page).
Postscript created with Adobe's newest postscript driver and adist5.ppd
available from
http://www.adobe.com/support/downloads/product.jsp?product=44&platform=Windows
contains a more proper encoding for me.
So my suggestion would be to visit that side and install the newest
driver. The adist5.ppd is in the "PPD Files" section a bit down the
mentioned page (under "PPD Files: Adobe"). You need to download it
before starting to install the postscript driver (top of the page) as
you'll be asked for that PPD then.
Most probably that will do the trick.
> I've added a few new Quotes, one from GS 8.15 and one from GS 8.53.
> The funny thing was when I tried to Downgrade my 8.53 install back to
> 8.15, the fonts when printed with the 8.15 are embedded like the
> 8.53. I had to go to another machine to get the real 8.15 output.
>
> So should I install 8.54?
Errr, if you want ;-) But that's only the name for the CVS version
which is still in development before the definate 8.54 gets published.
So you'll probably have to compile it yourself which I have no
experience with -- on Windows. I think upgrading your Postscript driver
should do.
-hwh
Thanks!
"Hans-Werner Hilse" <hw.h...@stud.ugoe.de> wrote in message
news:20060302174054.7...@stud.ugoe.de...
Okay now that is cleared up... (-;
I installed AdobePS 1.0.6 and the "PPD Files: Adobe" files and all is good
in the World.
It installed a "Adobe PS Acrobat Distiller" as the print driver, the one it
was using was just "Acrobat Distiller"
So now I need to figure out if I really need the AdobePS 1.0.6 installed or
just the "PPD Files: Adobe" PPD.
The new file is here: http://www.enm.com/pdfissues/q23101.pdf
Though the Fonts are still funny names and are embedded and custom encoded.
I'll see if I can switch it all over to Helvetica and see if that deals with
that issue.
Thanks again for your help!
Scott<-
"Hans-Werner Hilse" <hw.h...@stud.ugoe.de> wrote in message
news:20060302174054.7...@stud.ugoe.de...
According to the DataWindows installation guide (found on google), the default printer driver
does not permit creation of searchable PDF (as it installs the NT4 kernel mode PostScript driver,
which converts TrueType fonts to Type1 and, of couse, does font subsetting and in any case
uses a custom encoding for the generated Type1 font subsets which depend on the sequence
the requested glyphs appear in the input sream).
If you install the "official" PS driver for W2K and upwards, you must patch your DataWindows
afterwards as described in the installation guide (see http://download.sybase.com/pdfdocs/dnetp0150e/dwninst.pdf)
Hi Scott!
Glad things are starting to work.
Besides the printer driver, you have also changed the version of the
PDF being created from 1.5. to 1.3. Also the increased file size
suggests that you have also changed from screen to default (when I saw
a similar problem using JAWS, the cause was my use of the screen/Web
setting).
> Though the Fonts are still funny names and are embedded and custom encoded.
> I'll see if I can switch it all over to Helvetica and see if that deals with
> that issue.
The names are created when the TTF is changed to either a Type1 or
Type42 font depending on your setting.
Phil
> Ghostscript has been writing incorrect ToUnicode CMap without CMapName
> into the PDF since version 8.10 (rev. 3611) . This bug is fixed in
> version 8.54 (rev. 6201).
I didn't know that. I'll get version 8.54 right away.
Before (or while) doing this, read its licence agreement CAREFULLY.
>
> It installed a "Adobe PS Acrobat Distiller" as the print driver, the one it
> was using was just "Acrobat Distiller"
>
> So now I need to figure out if I really need the AdobePS 1.0.6 installed or
> just the "PPD Files: Adobe" PPD.
Dont' use the Distiller PPD, see dicussion below.
>
> The new file is here: http://www.enm.com/pdfissues/q23101.pdf
Adobe Reader 7.0.5/ Windows copies the text OK.
>
> Though the Fonts are still funny names and are embedded and custom encoded.
> I'll see if I can switch it all over to Helvetica and see if that deals with
> that issue.
>
> Thanks again for your help!
> ...
- Font subsetting DURING PS->PDF conversion is not a problem.
- Reencoding done by old PS drivers in order to subset fonts
("incremental download" of fonts) usually results in
non-searchable PDFs.
- You do not need the Distiller PPD. Ghostscript does come with
a PPD and an INF for installing a printer on Windows 2000/XP,
though I'm not sure in which version were these added. Use
the Windows "Add Printer" wizard, choose "Have disk", and point
it to Ghostscript's LIB\ directory. Also note there are pending
patches for the PPD and for extending support to Win9X, see
http://bugs.ghostscript.com/show_bug.cgi?id=687981
Now back to the "searchable PDF" problem, for PDFs created
on Windows:
- Use a PS driver for a printer that includes the base-14
(or base-35) fonts. GS does, but IIRC Distiller has only Courier
[new drivers work OK anyway, but not the old ones];
- Let the driver do the standard substitutions of system TrueType
fonts with printer-resident ones (Arial->Helvetica, etc);
- Use only the fonts that are substituted. This can be limiting for
some applications, but it seems you use only Arial.
This takes care of the reencoding issue, even with old drivers,
because there won't be any "incremental download of fonts" (in
fact, no font download al all) and the diver will have to use the
standard encoding of the printer-resident fonts. There won't be
any strange font names. And you get smaller PDFs, because the
base-14 fonts don't need to be embeded, any PDF viewer will
provide them.
> Monday 27 February 2006
>
> Sounds like you have GS set to create a Screen PDF. For size reduction
> GS automatically substitutes for fonts which come with AR. Helvetica
> and Arial are the same font and since Helvetica comes with AR, Arial is
> replaced. Use either prepress or printer to have all fonts embedded.
The fonts that "come with AR" are the Adobe base 13 fonts that every pdf
viewer is assumed to provide. Helvetica is one of the original base 13
fonts, but now Adobe uses Arial. Ghostscript uses the URW fonts, but if
you don't embed, you get whatever the viewer chooses to use, which may be
Arial, Helvetica, or the corresponding URW font. There are gross
differences in certain glyphs, particularly the Helvetica "italic", which
was originally a synthetic font (e.g., Helvetica with a slant applied)
and had slanted parentheses, brackets, etc. Arial Italic has upright
glyphs. People often use italic for math variables so |a| becomes /a/
depending on the font used to display the document.
Just to complicate matters, some printers with non-Adobe rasterizers
likely use Helvetica "clone" fonts.
One way to be sure a consitent font is used is to embed the URW fonts
by using them instead of the Adobe base fonts in the original document.
The original URW fonts were high quality, but be aware that some recent
versions with additional glyphs had bugs.
--
George N. White III <aa...@chebucto.ns.ca>
Hi George!
George N. White III wrote:
> The fonts that "come with AR" are the Adobe base 13 fonts that every pdf
> viewer is assumed to provide. Helvetica is one of the original base 13
> fonts, but now Adobe uses Arial. Ghostscript uses the URW fonts, but if
> you don't embed, you get whatever the viewer chooses to use, which may be
> Arial, Helvetica, or the corresponding URW font. There are gross
> differences in certain glyphs, particularly the Helvetica "italic", which
> was originally a synthetic font (e.g., Helvetica with a slant applied)
> and had slanted parentheses, brackets, etc. Arial Italic has upright
> glyphs. People often use italic for math variables so |a| becomes /a/
> depending on the font used to display the document.
With AR6 the switch was made to outline fonts. I am not sure how AR6
and AR7 builds them for display, but it does. Even with the switch the
full IBM Character set is still not in the the base 13 fonts. This
creates a problem for DOS users who want to turn files designed for PCL
printers into PDF files.
> One way to be sure a consitent font is used is to embed the URW fonts
> by using them instead of the Adobe base fonts in the original document.
> The original URW fonts were high quality, but be aware that some recent
> versions with additional glyphs had bugs.
As an OS/2 user and the maintainer of the Postscript Printer driver for
WordStar Dos, creating a custom font map for GS and embedding fonts is
something with which I am familiar. My driver comes with a custom
fontmap for GS which makes use of the IBM fonts (either the one which
comes with OS/2 or the free Courier released for Linux).
Phil