lilypond-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: no links in PDFs with gs 9.56.1


From: Jean Abou Samra
Subject: Re: no links in PDFs with gs 9.56.1
Date: Sun, 7 Aug 2022 11:48:38 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.12.0



Le 07/08/2022 à 11:43, Thomas Morley a écrit :
Am Sa., 30. Juli 2022 um 12:39 Uhr schrieb Werner LEMBERG <wl@gnu.org>:

In case you are forced to use 9.56.1, add `-dNEWPDF=false` to the
gs command line options (in file `Documentation/GNUmakefile`) to
make gs use the old PDF engine, which produces good results.
So this is about the post-processing step with extractpdfmark + gs,
right?
Yes – I always build the documentation with 'extractpdfmark'; maybe
this gs bug doesn't show up otherwise.


     Werner
If I build the docs I use 'extractpdfmark' as well. In the light of
this thread I tried ghostscript self-compiled from their most recent
master, giving me 9.57.0.
Though, compile time is bad:
Always doing
time make LANGS='en' doc -j5 CPU_COUNT=5
I get for

Uncompiled guile

real    111m56,272s
user    190m51,445s
sys    17m51,542s

Compiled guile

real    98m43,545s
user    162m33,293s
sys    15m33,015s

This is insane. Tbh, I did not check whether pdf links are present but
directly downgraded to gs 9.55.0, with below timing values:

Uncompiled guile

real    50m20,112s
user    129m33,921s
sys    17m42,618s

Compiled guile

real    44m32,826s
user    116m10,453s
sys    17m1,201s


Better, but still bad, regarding the fact I compiled only the english docs.

Yes, I have a weak laptop...

No idea, if we can do anything to improve the situation.


Regardless of the issue (which is problematic indeed), I would
advise building without extractpdfmark
(../configure --without-extractpdfmark). Its only advantage is to
reduce the size of PDFs. If you just want to check the doc section
you wrote, that's not really needed.


Best,
Jean


Anyway, I thought I share my observations.





reply via email to

[Prev in Thread] Current Thread [Next in Thread]