Output of selected pages of viewed PDF to vector PDF file - such as through Virtual PDF Printers #4864
Replies: 2 comments 2 replies
-
The Windows Printing System changes of more than 10 years ago were highly problematic and variable (and has never got better, unless you had full control of your System components). So with the code alterations to 64 bit in 2015 the poor printout problems were avoided by not have to port or write any fresh printer code for every printer on the planet. There are two very good historic commercial grade PDF printing systems that SumatraPDF can share files to print. One is the highly versatile Artifex GhostScript command line printing highly suited to high quality vector printer often found in clones like PDF24 bullzip etc etc etc. So 10 years ago the Printing System was handed entirely over to the WinPrint system which is perfect for cheaper Inkjets. Thermal printers and all those others that support using PCLm and other PDF CUPS raster formats which are PDFs as printing images. ONLY the commercial printers like Apple Laser printer need the Original PDF Out as PostScript PDF. |
Beta Was this translation helpful? Give feedback.
-
There is by the way a workaround, using the nice idea of GitHubRulesOK of using the external viewers functionality to run cmds and other exe files (that are not external viewers). See #4853 (comment). |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
(Edited for clarity)
As I have learned the hard way, printing a viewed PDF from SumatraPDF to any available virtual PDF printer produces a bitmap PDF.
Virtual printers such as PDFCreator allow are installed as system printers, and do produce PDF files. If they receive a vector file, they do output a vector PDF (unless requested otherwise). In my humble understanding, SumatraPDF works in a way that one gets only bitmap PDFs. I tried MS, Bullzip, PDFCreator, .... to no avail. For me this is a defect in SumatraPDF, being unable to send vector to printout devices that support it.
Why users might need to produce a PDF for a PDF they are viewing? Different use cases:
a) wish for only a range of pages
b) change from color to b&w
c) change of page size
d) you name your favorite....
Consequences of the bitmap content of the PDF is that files become unnecessary bulky and search/copy of text is lost. Besides, user has no control of resolution output of any image. Edition capability is lost.
I was told that this happens with SumatraPDF is due to the use of muPDF (the use of muPDF seems to be an architectural and programming decision) and WinPrint. Note that I am not knowledgeable of how front/back end works in this environment, nor how muPDF is used or how SumatraPDF interfaces with WinPrint.
I am unaware of how this could be solved. I believe Virtual Printers do handle receiving vector PDFs, otherwise I cannot see any value on them (nowadays that MS PDF is there). PDFCreator used to allow define resolution, b&w, sign PDF, and many more.
A way could be to "stream" the current PDF to "another tool" (there are many libraries to handle pdfs) so that the virtual PDF printers receive an exact copy of the original PDF and not a bitmaped ghost of it... A workaround would be to use another library to split and merge and output the outcome PDF file. Perhaps this last is the easiest way to
While one could find a way to carry out the task by the use of the external viewers and command use of printing tools, it is quite cumbersome and not user friendly.
From my point of view, SumatraPDF future could be jeopardized due to this issue (not being able to output true vector PDFs, either due to defective printing or by lack of turnaournd. Users will find out other viewers able to print "as is" and not as bitmap - I have been unable but iI would expect someone would be able to do (Okular does in Linux but not in Windows...). Users abandoning SumatraPDF would be a real pity ...
As user myself I see this issue as a bug of the program and not a new feature. How come one would expect that the outcome is not same as the input .... Vector elements of PDF are of the essence of PDF, creating just bitmaps is really an entropy damage, loss of information.... Which ever way you could solve it, either solving the defect or providing an enhancement, please do it!!! (besides should not be terribly difficult; should I be able to code I would be proposing a patch ...)
Just my 2cents
CT
Beta Was this translation helpful? Give feedback.
All reactions