On Thu, 2017-07-20 at 19:46 +0200, Joerg Sonnenberger wrote:
> Whether the second or third is choosen still depends on the
> conversion rules, which is tricky at best.
Hi,
yes, that's what I meant. I made the change in revision 1856:
http://sourceforge.net/p/podofo/code/1856
> Don't get m
On Thu, Jul 20, 2017 at 05:17:01PM +0200, zyx wrote:
> Hi,
>
> On Thu, 2017-07-20 at 15:30 +0200, Joerg Sonnenberger wrote:
> > Passing anything to Load() which was not explicit a long, i.e.
> > std::string::size() would result in the former method be choosen...
>
> Ah, I see, that's unfort
Hi,
On Thu, 2017-07-20 at 15:30 +0200, Joerg Sonnenberger wrote:
> Passing anything to Load() which was not explicit a long, i.e.
> std::string::size() would result in the former method be choosen...
Ah, I see, that's unfortunate and unintended. The idea behind the
change was to not break
On Thu, Jul 20, 2017 at 02:17:36PM +0200, zyx wrote:
> Well, every release has bumped soname version, thus I do not know what
> silence you are talking about.
Old:
PDFMemDocument::Load(const char *);
PdfMemDocument::Load(const char *, long);
New:
PDFMemDocument::Load(const char *, bool = false);
On Thu, 2017-07-20 at 12:22 +0200, Joerg Sonnenberger wrote:
> please include the attached patch to fix the ABI nightmare that r1810
> creates.
Hi,
this had been fixed long time ago, shortly after the release. I guess
you are not subscribed to the list that long, otherwise you'd save you
p
On Thu, Jul 20, 2017 at 01:33:38PM +0200, Mattia Rizzolo wrote:
> On Thu, Jul 20, 2017 at 12:22:13PM +0200, Joerg Sonnenberger wrote:
> > While I love PoDoFo, it certainly needs a lot of love when it
> > comes to API and ABI stability.
>
> podofo never had any ABI stability, indeed the soname is
On Thu, Jul 20, 2017 at 12:22:13PM +0200, Joerg Sonnenberger wrote:
> While I love PoDoFo, it certainly needs a lot of love when it
> comes to API and ABI stability.
podofo never had any ABI stability, indeed the soname is versioned with
all the version string (including the "patch number" in sem
Hello,
please include the attached patch to fix the ABI nightmare that r1810
creates. Mixing C++ standard versions is still very popular, especially
since the GCC versions in LTS Linux distributions default to C++03
still. The patch avoids the difference by making it explicit in the one
place where
