> On 19 Mar 2021, at 14:09, Bruno Voisin wrote:
>
> I'll try to investigate later today what may have triggered the problem.
Thanks to much debugging by Herb Schulz off-list the source of the problem has
now been identified, I think.
There are two dvipdfmx.cfg in tlpretest:
% kpsewhich -progn
> On 19 Mar 2021, at 13:47, Herbert Schulz wrote:
>
> If I now run
>
> xelatex -no-pdf TransparencyTest.tex
>
> and then
>
> xdvipdfmx TransparencyTest.xdv
>
> the output is still
>
> [...]
>
> and I get a correct PDF. So still no failure. Now that Bruno is also getting
> a failure I'm rea
> On Mar 19, 2021, at 2:58 AM, Bruno Voisin via XeTeX wrote:
>
>
> % xdvipdfmx TransparencyTest.xdv
>
> TransparencyTest.xdv -> TransparencyTest.pdf
> [1Error: /invalidfileaccess in --run--
> Operand stack:
> (/usr/local/texlive/2021/texmf-dist/dvips/pstricks/pstricks.pro) (r)
> Exe
> On 19 Mar 2021, at 08:39, Bruno Voisin wrote:
>
> - To be sure, I added -dSAFER to dvipdfmx.cfg. That doesn't change a thing,
Please ignore what I wrote. I had created a modified dvipdfmx.cfg in
texmf-local/dvipdfmx, but forgot to run mktexlsr afterwards, so it wasn't used.
Silly me!
Anyway
> On 19 Mar 2021, at 06:49, Akira Kakuto wrote:
>
> Sorry, I'm not so familiar with Ghostscript and I don't understand the
> difference.
> The reported security problem is for Windows, so I think it is OK.
> In the latest dvipdfmx.cfg, I added -dSAFER explicitly in order not to allow
> dangerous
Dear Bruno,
On 2021/03/19 7:40, Bruno Voisin via XeTeX wrote:
With up-to-date pretest MacTeX-2021 I get the same as Herbert, and cannot
reproduce your output:
Sorry, I'm not so familiar with Ghostscript and I don't understand the
difference.
The reported security problem is for Windows, so I
> On 18 Mar 2021, at 23:40, Bruno Voisin wrote:
>
> To be sure I tried xdvipdfmx -vvv, the output is at the end of this message:
> there's a lot of font-related info, but no indication pstricks.pro is read at
> any time.
Some complementary information to my earlier message: compiling, still wi
(Please cc possible answers to me, I'm not subscribed to the list.)
Dear Akira,
With up-to-date pretest MacTeX-2021 I get the same as Herbert, and cannot
reproduce your output:
% xdvipdfmx TransparencyTest.xdv
xdvipdfmx:warning: Invalid char in fontmap line: D
xdvipdfmx:warning: Invalid map
Dear Herbert,
On 2021/03/17 21:43, Herbert Schulz wrote:
With the pstricks transparency example I supplied xelatex alone---default
parameters for xdvipdfmx alone, -dALLOWTRANSPARENCY, no need for forcing
-dNOSAFER---works just fine.
Probably you are using an old dvipdfmx.cfg.
Or using an ol
> On Mar 16, 2021, at 5:50 PM, Akira Kakuto wrote:
>
> Dear Herbert,
>
> On 2021/03/16 22:31, Herbert Schulz wrote:
>> Xelatex seems to compile the enclosed document, which uses transparency with
>> pstricks, fine now (aside from some gs 9.53.3 warnings about opacity being
>> dpericated).
>
Dear Herbert,
On 2021/03/16 22:31, Herbert Schulz wrote:
Xelatex seems to compile the enclosed document, which uses transparency with
pstricks, fine now (aside from some gs 9.53.3 warnings about opacity being
dpericated).
But latex->dvipdmx fails. Is that expected?
Indeed. PSTricks is not
> On Mar 15, 2021, at 8:27 PM, Karl Berry wrote:
>
> FYI to all: we have changed dvipdfmx.cfg (in the pretest) so that
> Ghostscript is always invoked safely. The primary practical implication
> of this is that if you want to use xetex with pstricks,
> 1) be very sure that you trust all the inp
FYI to all: we have changed dvipdfmx.cfg (in the pretest) so that
Ghostscript is always invoked safely. The primary practical implication
of this is that if you want to use xetex with pstricks,
1) be very sure that you trust all the input files, and
2) run
xelatex -output-driver="xdvipdfmx -i
13 matches
Mail list logo
