Is there a limit to the size of PDF files?
Could it be that PDFTeX (as in TeXlive2007 for win32) aborts with an obscure error message when the size of the generated PDF file exceeds 2GB?
The prototype of fseek() in my old MSVC++ reads int fseek(FILE *stream, long offset, int origin) and long (the same as int) is signed 32 bit integer. Therefore I think 2GB is the limit in the TL win32 pdfTeX.
If yes, is there a chance to work around this in some way?
I cannot find a workaround. Best regards, Akira
Akira Kakuto
Could it be that PDFTeX (as in TeXlive2007 for win32) aborts with an obscure error message when the size of the generated PDF file exceeds 2GB?
The prototype of fseek() in my old MSVC++ reads int fseek(FILE *stream, long offset, int origin) and long (the same as int) is signed 32 bit integer. Therefore I think 2GB is the limit in the TL win32 pdfTeX.
If yes, is there a chance to work around this in some way?
I cannot find a workaround.
In Linux/glibc, I find The fseeko() and ftello() functions are identical to fseek() and ftell() (see fseek(3)), respectively, except that the offset argument of fseeko() and the return value of ftello() is of type off_t instead of long. On many architectures both off_t and long are 32-bit types, but compilation with #define _FILE_OFFSET_BITS 64 will turn off_t into a 64-bit type. Is there something like this to be had in MSVC++? Is MSVC++ obligatory for compiling PDFTeX in win32? -- David Kastrup
2007/10/17, David Kastrup
Is there something like this to be had in MSVC++? Is MSVC++ obligatory for compiling PDFTeX in win32?
I don't know about win32, but the size of PDFs is limited to 10GiB. If we can't easily fix the 2G limit on win32, we should at least fix the error message. Best Martin
Taco Hoekwater
David Kastrup wrote:
Is there something like this to be had in MSVC++? Is MSVC++ obligatory for compiling PDFTeX in win32?
Have you tried already whether a pdf > 2GB can be opened by your target audience?
Presumably not a problem. But I have to search for other venues, anyway, since binary replacement is not feasible right now in this project. And frankly, 2GB for a 150 page document is ridiculous.
If not, there is not much point in investigating further.
Maybe. -- David Kastrup
2007/10/17, David Kastrup
Taco Hoekwater
writes: Have you tried already whether a pdf > 2GB can be opened by your target audience?
Presumably not a problem. But I have to search for other venues, anyway, since binary replacement is not feasible right now in this project. And frankly, 2GB for a 150 page document is ridiculous.
AR8@Linux can not open files >2GB (I just tested). Best Martin
"Martin Schröder"
2007/10/17, David Kastrup
: Taco Hoekwater
writes: Have you tried already whether a pdf > 2GB can be opened by your target audience?
Presumably not a problem. But I have to search for other venues, anyway, since binary replacement is not feasible right now in this project. And frankly, 2GB for a 150 page document is ridiculous.
AR8@Linux can not open files >2GB (I just tested).
Actually, I just discovered \pdfobjcompresslevel. Maybe we will get this circumnavigated. -- David Kastrup
David Kastrup writes:
Actually, I just discovered \pdfobjcompresslevel. Maybe we will get this circumnavigated.
Don't expect too much if there are many graphics. They are compressed already and take most of the space. Regards, Reinhard -- ---------------------------------------------------------------------------- Reinhard Kotucha Phone: +49-511-4592165 Marschnerstr. 25 D-30167 Hannover mailto:reinhard.kotucha@web.de ---------------------------------------------------------------------------- Microsoft isn't the answer. Microsoft is the question, and the answer is NO. ----------------------------------------------------------------------------
Reinhard Kotucha
David Kastrup writes:
Actually, I just discovered \pdfobjcompresslevel. Maybe we will get this circumnavigated.
Don't expect too much if there are many graphics. They are compressed already and take most of the space.
Given their size, they can't be compressed all too much (5 to 10MB PDF per piece, presumably generated by Imagemagick). I already wrote a ticket for replacing Imagemagick with sam2p (I guess I trust Peter more about squeezing size), but I don't know whether my guesses about which does what in image preprocessing are correct. -- David Kastrup
On Wed, 17 Oct 2007, Reinhard Kotucha wrote:
David Kastrup writes:
Actually, I just discovered \pdfobjcompresslevel. Maybe we will get this circumnavigated.
Don't expect too much if there are many graphics. They are compressed already and take most of the space.
... and any kind of stream objects are not touched at all by object compression. The \pdfobjcompresslevel affects only objects that are pure dictionaries. Regards, Hartmut
participants (6)
-
Akira Kakuto
-
David Kastrup
-
Hartmut Henkel
-
Martin Schröder
-
Reinhard Kotucha
-
Taco Hoekwater