Hi,
On 2014-07-04 01:39, Norbert Preining wrote:
As I'm currently writing, the patch has been considered a good
starting point but it is not complete yet, nor it has been ACKed.
Agreed. But the whole point is that it is a *regression* because
in older versions of poppler it did work. So there should be code
or a way to find what caused the problem.
Sure, just like any other regression.
I will wait when there's at least a complete fix upstream, otherwise
it is piling up hacks on hacks.
But I am not sure whether it will be included. The devs stated in
https://bugs.freedesktop.org/show_bug.cgi?id=73291
that the *font* should be fixed, which means they are not
acknowledging that the problem is within the proper interpretation
for glyph names in poppler.
The only poppler developer talking in that bug is Adrian Johnson
(comments #5, #7, #8, #10), which only notes the issue is about the
different glyph naming used by texgyretermes-bold.ttx.
I don't see how that implies "the font is buggy" nor "we won't change
poppler", nor how you can draw such conclusions from it, especially
when Adrian is quite skilled.
The rest of the comments there are by simple users.
OTOH, the font developers have a clear stand and try to be according
to the guidelines put forth by Adobe.
>Poppler upstream maintainers seem to be ignorant about this, and
>even worse, seem to be completely incooperative.
Pointers to this? In both the freedesktop bugs (#73291 and #80093),
See the abov bug report where in comment 19 Christopher Yeleighton
states that it is a bug in the fonts.
It seems you missed what I said about him. Simply ignore what he
says, really.
As a side note: If you follow
https://bugs.freedesktop.org/show_bug.cgi?id=25240
then you will see what the attitude of (at least one) poppler
developer
is concerning bugs including patches.
This regards just one specific feature, I don't see how that should
represent the "official view" on what poppler developers say or do,
and having contributed to poppler for many years I can tell you quite
a lot of features and bugfixes have been added over the years.
Please don't generalize.
Moreover, few months ago I contacted you about testing poppler the
0.25.x pre-releases of 0.26 and other details, and never got
anything
I know, I was busy packaging 2014, and maybe you don't know through
which library hell I had to go due to libpng maintainers keeping
libpng >= 1.5 out of Debian.
OK, but if you don't reply, even in a short way, telling me so, how
can I know? I can imagine you could be busy, yet a bit of testing
from a core component like TeX with newer poppler series would be
nice to have.
I have spent weeks trying to fix these problems - I honestly cannot
invest more time just to test compatibilities.
I do think it would be a very useful thing to do, especially to not
be later in a situation where a new poppler version causes some issue
(regression or not), and then I read bad remarks about poppler...
And with the above email I suggested to include
a fix since my feeling from the two freedesktop bugs is that
poppler will not fix this regression.
Can you please try to not be that pessimist like that, for a moment?
I told you above that the first review of the patch in fdo#80093
considered it a good starting point, and that I will handle the issue
in Debian once there is an ACKed patch upstream.
There is no "secret plot" or whatever meant to make the interaction
with TeX worse; just like you said above you were busy, also poppler
people are.
Thanks for the patience,
--
Pino Toscano
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
