OpenSystemsDevelopment / qpdf

This is CVE-2017-9209.

This is CVE-2017-9210.

The description string for an error message included unparsing an
object, which is too complex of a thing to try to do while throwing an
exception. There was only one example of this in the entire codebase,
so it is not a pervasive problem. Fixing this eliminated one class of
infinite loop errors.

Don't try to install HTML or PDF documentation if we're not building
docs.

The 64 Bit file functions are supported by C++-Builder as well and
need to be used, else fseek will error out on larger files than 4 GB
like used in the large file test.

Some compilers output extra newlines in some cases.

QPDFObjectHandle was used as forward declaration, but C++-Builder 10
Seattle can't use it in std::list in such cases because the type is
undefined.

qutil.cc uses strerror to print some exceptions and adds a newline
afterwards, but strerror in Windows already adds one newline at the
end of the message and the additional one from std::endl breaks the
output vs. the expected one.

Provide notes about LibTiff and such.

Working with absolute paths makes debugging easier, but some called
scripts always need / as dir separator or won't work.

/dev/null is not portable, so use File::Spec instead, which provides
portable "paths" and especially "nul" on Windows. I changed all places
with hard coded /dev/null to be sure, while I think it only is a
problem in direct system calls, because the other executed commands go
to sh.exe from MSYS which itself should port /dev/null to NUL. The
test still pass, so shouldn't have made any harm...

expr needs ARG + ARG
quote paths to support support spaces

Shebang doesn't work well on Windows.

Remove an extraneous newline to avoid useless constant warning.

Also update maintainer documentation on binary compatibility testing.

Since we have to bump soname, remove some private methods that were
just there for binary compatibility

Make sure people know that static ID should be used only for testing.

If xref table entries lack the spec-required trailing whitespace or
contain a small amount of extra space, handle them anyway.

For non-encrypted files, determinstic ID generation uses file contents
instead of timestamp and file name. At a small runtime cost, this
enables generation of the same /ID if the same inputs are converted in
the same way multiple times.

fix-qdf was previously hard-coding the number of bytes for the f2
field of the xref stream entry. This addresses issue #37. Thanks
aluebcke for reporting.

As reported in issue #40, a call to CryptAcquireContext in
SecureRandomDataProvider fails in a fresh windows install prior to any
user keys being created in AppData\Roaming\Microsoft\Crypto\RSA.

Thanks michalrames.