I have a PDF file that I need to change the orientation from landscape to portrait, using this command GS. The Orientation switch it is not changing the orientation. Is there a extra setting I need to add?
::Path to your Ghostscript EXE
set GSC="C:\Program Files (x86)\gs\gs9.16\bin\gswin32.exe"
%GSC% -sDEVICE=pdfwrite -sOutputFile="D:\TRANS\PDF\2_ZNOR_00000000000000000070.pdf" -dNOPAUSE -dEPSCrop -c "<</Orientation 3>> setpagedevice" -f "D:\TRANS\PDF\1_ZNOR_00000000000000000070.pdf"
It worked for me
List<string> switches = new List<string>
{
"-empty",
"-dQUIET",
"-dSAFER",
"-dBATCH",
"-dNOPAUSE",
"-dNOPROMPT",
"-dDEVICEHEIGHTPOINTS=612",
"-dDEVICEWIDTHPOINTS=792",
"-dFIXEDMEDIA",
"-dFitPage",
"-sDEVICE=pdfwrite",
"-dAutoRotatePages=/None",
"-sOutputFile=" + resultFilePath,
"-c",
"<</Orientation 1>> setpagedevice",
"-f"
};
switches.Add(this.FilePath);
Most of the switches you are using there will have no effect at all, and you haven't done anything to rotate the PDF file either.
The first thing to note is that Ghostscript is really the wrong tool for this job, it doesn't modify PDF files, it generates completely new ones. The process is described here
The -dEPSCrop switch only has an effect if the input is an EPS file, otherwise it does nothing.
The operands to setpagedevice are incorrect, you are passing an empty hex string <> when setpagedevice expects a dictionary <<>>. Even then an empty dictionary will do nothing.
I expect you intended to put <</Orientation 1>> setpagedevice but actually that's only a media matching request. In order to get Ghostscript to create a rotated PDF file you would have to alter the media size, set it to fixed, and set -dFitPage. You would also have to set the AutoRotatePages switch to None in order to prevent the automatic reorientation of the output so that the text is horizontal.
Since you haven't supplied the PDF file to look at, nor stated its media size, I can only guess at the sizes. However, something like
gswin32c.exe -sDEVICE=pdfwrite -sOutputFile=new.pdf -dDEVICEHEIGHTPOINTS=612 -dDEVICEWIDTHPOINTS=792 -dFitPage -dAutoRotatePages=/None input.pdf
Related
i use GS with this command line
gswin32c.exe -sDEVICE=tiffg4 -dSAFER -dBATCH -dNOPROMPT -sPAPERSIZE=a4 -dFIXEDMEDIA -dPDFFitPage -dNOPAUSE -r600 -g6120x7920 -dPrinted=false -sOutputFile="C:\test\test_%d.tif c:\test\xyz.pdf
That works fine, but it always start with the "1" value, so the output is "test_1.tif".
How do i need to modify the comanndline, to force GS to start with another value, e.g. 7.
so GS convert the multipage PDF "xyz.pdf" to singlepage tiff like "test_7.tif, test_8.tif" and so on.
background:
i have several pdf and the naming of the pdf is allways different. but i want one ongoing name for the tif file which allways begin with "test" in this case and the incremental value for each pdf page.
also my first pdf extract have to be the first tiff files, then the 2nd PDF file extract have to append and so on.
hopefully its a litte bit clear what i need ;)
BR Ralf
The %d convention is simply a C printout format and as mentioned in KenS comment cannot be "offset"
see https://www.ghostscript.com/doc/current/Use.htm#One_page_per_file
but there are workarounds mentioned in
How to set printf %d offset for Ghostscript?
However they are perhaps not the most efficient, but may help in some cases, just beware the comments below them.
The simplest way I will alter the range is to apply a preset such as
1%d to produce 10 11 12..., or
2%02d = 201 202 ...
Again you are limited to the first page will always be ###1 and if you need it to start at ###5 then you need to rename all files post production. That can often be easily scripted depending how well you prepare the desired names.
I'm trying to write a number of gs commands for server-side use. The user-provided PDF/JPEG files, I have to work with cannot be assumed to be safe (broken or even malicious files could be provided). Therefore, I'm trying to write all of my Ghostscript commands with -dSAFER, to guarantee at least a basic level of security.
Unfortunately, -dSAFER appears to be incompatible with certain gs commands. Take for example the following command:
# count number of pages in PDF
gs -dQUIET -dBATCH -dNOPAUSE -dNOSAFER -dNODISPLAY \
-c "(input.pdf) (r) file runpdfbegin pdfpagecount = quit"
How would you re-write this command with -dSAFER? The command fails if I just add -dSAFER, because gs can't read the file input.pdf (which is what I expect). How do I tell gs that is permitted to read input.pdf, but nothing else? Maybe there's a way to permit reading of files only from certain directories?
Here's a second example command:
# convert JPEG to single-page PDF
gs -dQUIET -dBATCH -dNOPAUSE -dNOSAFER \
-sDEVICE=pdfwrite -dCompatibilityLevel=1.5 -dPDFSettings=/printer \
-sColorConversionStrategy=Gray -dProcessColorModel=/DeviceGray \
-sOutputFile=output.pdf \
viewjpeg.ps \
-c "(input.jpg) << /PageSize 2 index viewJPEGgetsize 2 array astore >> setpagedevice viewJPEG"
This command has exactly the same problem as the first one. How do I re-write this with -dSAFER?
Please include a link to the relevant documentation in your answer if you think that -dSAFER is really not needed for my commands.
You can add directories to the search list using -I, IIRC any such directory is permitted to be read. For the current directory you can also use -P- and -P.
See Use.htm in the ghostpdl/doc directory.
However, even if adding the current directory allows you to read the file, this will soon stop working with -dSAFER anyway. There's an ongoing programme to 'harden' the interpreter when -dSAFER is set by hiding/remobing any non-standard operators especially if there's any possibility they could be misused.
Your proposed usage is, simply, incompatible with -dSAFER. The commands you are using will almost certainly be specifically prohibited because they are inherently unsafe. Essentially by messing about inside PDF files like that, your PostScript program is unsafe.
Of course, you know that your PostScript program is safe, and since you are not executing any part of the PDF, the PDF is also safe. The PDF would only be unsafe if you attempted to actually execute the contents, which your program doesn't do. It simply opens the file reads the Pages tree, and tells you the value associated with the Count key.
So you don't need to set -dSAFER in this particular example anyway. However any such program which did execute the PDF content (eg by rendering any part of it, or sending it to the pdfwrite device) would not be safe.
Your second example also doesn't need SAFER, since a JPEG can't contain code to access the disk.
I am using ghost script to convert my postscript file to PDF/A.
Below is the command to generate the PDF/A file
gs -q -dPDFA=2 -dBATCH -dNOPAUSE -dNOOUTERSAV -sProcessColorModel=DeviceGray \
-sDEVICE=pdfwrite -dFastWebView=true -dPSFitPage -dDOPDFMARKS \
-sOutputFile=NewPDF.pdf PDFA_def.ps abc.ps pdfmarks
The PDFMarks file looks like:
[ /Title (Document title)
/DOCINFO pdfmark
[ /Page 1 /Title (COVER PAGE) /OUT pdfmark
[ /PageMode /UseOutlines /View [/FitPage]/Page 1 /DOCVIEW pdfmark
[ {Catalog} <</ViewerPreferences<</DisplayDocTitle true>>/PageLayout/OneColumn>> /PUT pdfmark
Still when the PDF/A generates, the Magnification Property is set to Default Only. I need to set it to FitPage.
I tried using /Fit,/FitB,/FitH etc but none of them seems to work.
There is no 'FitPage', using this option causes Acrobat to raise a warning when opening the resulting PDF file (Ghostscript does not attempt to fully validate pdfmark operands)
Fit uses the CropBox to determine the scaling, and you haven't specified a CropBox so unless your other PostScript does so, then there won't be one. Similar problems may exist with the other options you have mentioned.
The PUT pdfmark is also incorrect, the '<>' should be '<<'.
You do not need (and should not use) -dNOOUTERSAVE.
I wouldn't use -dFastWSebView unless you have a very good reason, it simply takes longer to process and produces a larger file for no really good reason.
PSFitPage won't do anything useful if you don't specify a fixed media size (which you haven't).
You don't need -dDOPDFMARKS, that's for PDF input only (and in any event defaults to true).
If you want the PDF/A file to be Gray you should set -sColorConversionStrategy=DeviceGray rather than meddling with the ProcessColorModel.
I used a simple empty PostScript program and added:
[ /PageMode /UseOutlines /View [/FitV] /Page 1 /DOCVIEW pdfmark
The resulting PDF file displayed as expected.
I have created a small script to run my PS file through the windows printer driver for printout.
I use this code:
echo mark /NoCancel true /BitsPerPixel 4 /OutputFile (%printer%%PNAME%) /UserSettings ^<^</DocumentName (%MYDOCNAME%) ^>^> (mswinpr2) finddevice putdeviceprops setdevice>setup.ps
gswin32c -dNOPAUSE -dBATCH -r84 setup.ps %1
The script is creating a setup.ps file as recommended by the gs documentation for the mswinpr2 driver.
It looks like this:
mark
/NoCancel true
/BitsPerPixel 4
/OutputFile (Windows printer name)
/UserSettings
<<
/DocumentName (the text for the job in the spooler queue)
>>
(mswinpr2)
finddevice
putdeviceprops
setdevice
%printer% resolves to the empty string on my machine.
%PNAME% is the name of the windows printer, "C364PS" for me.
The %printer%%PNAME% is taken from the documentation to give the desired target printer. To stay with the example printer it will resolve to C364PS
gs is then called as
gswin32c -dNOPAUSE -dBATCH -r84 setup.ps <name of desired ps file here>
This works as expected, regarding the printed result. But I want a silent printout, since I have chosen the printer before. Whatever I choose for %PNAME%, it will always pop up the "select printer" dialog.
I am running Windows 8.1 and tried gs 8.70 and 9.19.
If I add /QueryUser 3 to the setup.ps, the printout starts without further confirmation of the printer, but then it will only print on the default printer, not on the one selected in the /OutputFile.
What am I missing?
Update: I have checked this question's result, but it does not work for me, dialog keeps popping up: https://superuser.com/questions/807027/how-to-print-with-ghostscript-in-silent-mode
Update 2: I have now (as requested) tried a commandline, with the same result. I also tried without the %printer%, no difference here:
gswin32c -dNOPAUSE -dBATCH -dNOPROMPT -dNOQUERY -sOutputFile="%printer%C364SeriesPCL" -r84 setup.ps 151008172940#123#000001#000001#PR_CCCC_Vertragspruef_4#CCCC_CAAA.358636.ps
setup.ps is this, no linebreaks have been added:
mark /NoCancel true /BitsPerPixel 4 /OutputFile (C364SeriesPCL) /UserSettings <</DocumentName (151008172940#123#000001#000001#PR_CCCC_Vertragspruef_4#CCCC_CAAA.358636) >> (mswinpr2) finddevice putdeviceprops setdevice
Solution:
The string %printer% is actually a literal in the setup.ps /OutputFile () or on the commandline. So changing the setup.ps worked:
mark /NoCancel true
/BitsPerPixel 4
/OutputFile (%printer%C364SeriesPCL)
/UserSettings <<
/DocumentName (151008172940TBS000001000001PR_CONT_Vertragspruef_4CONT_CONT.358636)
>>
(mswinpr2)
finddevice
putdeviceprops
setdevice
This way, the printer is found. I tried to copy the behavior with a commandline switch -sOutputFile but was not able to make the dialog go away without the setup.ps - I tried %%printer%%, but to no avail. As I am fine with the setup.ps, I am not following that further.
The root of my problem was, that the ghostscript documentation is correct, nevertheless, easily misunderstood here. Especially in the context of windows, it would not hurt to state explicitely, that %printer% is actually no env-variable, but really a string literal, that must be present in the output filename.
-sOutputFile="%printer%printer_name"
Specifies which printer should be used. The printer_name should be typed exactly as it appears in the Printers control panel, including
spaces.
To really answer my initial question. The commandline script must be changed, to escape the % signs in the %printer% literal:
echo mark /NoCancel true /BitsPerPixel 4 /OutputFile (%%printer%%%PNAME%) /UserSettings ^<^</DocumentName (%MYDOCNAME%) ^>^> (mswinpr2) finddevice putdeviceprops setdevice>setup.ps
gswin32c -dNOPAUSE -dBATCH -dNOPROMPT -r84 setup.ps %1
This works as initially planned.
I've moved this to an answer, because there is insufficient room in the comments.
So, what I would normally expect to see as a (simple) command line for this would be:
gswin32c -dNOPAUSE -dBATCH -dNOPROMPT -dNOQUERY -sOutputFile="%printer%C364SeriesPCL" 151008172940#123#000001#000001#PR_CCCC_Vertragspruef_4#CCCC_CAAA.358636.ps
Your command line includes -r84 which would normally set the resolution to 84 dpi, if that actually gets honoured you are going to print a horribly low resolution image of the PostScript, I very much doubt you really want that.
On top of that, you also are still sending setup.ps before your actual PostScript program. The problem with that is that setup.ps overrides the values you set on the command line. So your -sOutputFile="%printer%C364SeriesPCL" is overridden by the contents of setup.ps where you define /OutputFile (C364SeriesPCL)
Now, as I was suggesting earlier, your problem here is that the correct syntax for a printer is "%printer%printer name" and what you have in setup.ps is lacking the %printer%. I would imagine the reason that %printer% is missing is because you put it in a batch file and %printer% is being treated as an environment variable. Since it isn't set, it gets replaced by nothing.
So, I would suggest that you start off by using the command line you gave, but NOT including setup.ps and see what that does. If that works, then you know what the problem is. In order to use %printer% in a batch file you need to 'escape' the '%' which means you will need (I think) %%printer%%
Given that you have a PostScript printer (C364SeriesPS) why don't you just send the original PostScript file ? What you are doing here is having Ghostscript render the PostScript to an image (a Windows Bitmap in fact) which it then passes to the printing system, which will send the bitmap to the printer. This is slow, and involves shuffling a lot of data around, when you could just send the PostScript straight to the printer.
I m converting a pdf into image using ghostscript
It almost produces good images but for one pdf the created image is having a white strip
however this strip is not present in the original pdf.
script which i hv used is below
gs -dNOPAUSE -sDEVICE=jpeg -dFirstPage=1 -dLastPage=2 -dTextAlphaBits=4 -dGraphicsAlphaBits=4 -sOutputFile=fileName.gif -dJPEGQ=100 -r300 -q pdfName.pdf -c quit
Here the created image http://www.pictureshack.us/images/65486_1.gif
i am using Ghostscript 8.70 and the issue is fixed for the latest update i.e 9.06, but i cant upgrade to this...moreover found the issue its due -r300, so when i specify resolution of 100 then it works fine for this pdf but not for many other pdf...
You don't say what version of Ghostscript you are using, and you haven't provided the original PDF file, so I can't test it on current code. However, this 'looks like' a transparency bug, and may well be fixed in recent versions of Ghostscript.
I would suggest you upgrade to 9.06 if you haven't already and try that. If the problem persists please open a bug report at http://bugs.ghostscript.com Please attach the original PDF file and the command line you are using.