Re: hplip-printer-app hangs and oddities (Re: Cups automatic duplex printing problems)

Valentijn,

If this affected all PS printers we would have heard about it a long =
time ago.

I will see if I can determine what is going wrong - the whole build =
process for this printer application will make debugging challenging... =
:/


> On Jul 9, 2024, at 4:19=E2=80=AFAM, Valentijn Sessink =
<[email protected]> wrote:
>=20
> Hi Michael, hello list,
>=20
> On 09-07-2024 00:21, Michael Sweet wrote:
>> Does the PPD file in question have any JCL options (look for =
JCLOpenUI)?
>=20
> Yes it does (please see bug report, sorry it wasn't clickable due to =
the missing "h"). For your convenience:
> https://github.com/OpenPrinting/hplip-printer-app/issues/17
>=20
>> It is possible that the JCL code is missing a trailing newline which =
could cause this - in which case we should append one to make sure the =
JCL code doesn't prevent language switching...
>=20
> Nope, there actually is a <0A> at the end, I did check that.
>=20
> But please note that the %% is at the *beginning* of a PJL line, i.e. =
there are postscript comments already, where PJL is still being sent to =
the printer:
>=20
> =C2=B7%-12345X@PJL
> @PJL JOB NAME =3D "Test Page" DISPLAY =3D "4 valentyn Test Page"
> @PJL SET USERNAME =3D "valentyn"
> %%@PJL ENTER LANGUAGE =3D PostScript
> %!PS-Adobe-3.0
>=20
> I can't figure out the code path. But if I change the code at the end =
of ppdEmitJCL() to:
>=20
>  ppdEmit(ppd, fp, PPD_ORDER_JCL);
>  fputs(ppd->jcl_end, fp);
>  fputs(ppd->jcl_ps, fp);
>=20
> ... _and_ change the PPD file to say:
> *JCLEnd: "END-OF-TASK<0A>"
>=20
> ... then the output to the printer says:
>=20
> @PJL SET USERNAME =3D "valentyn"
> %%END-OF-TASK
> @PJL ENTER LANGUAGE =3D PostScript
> %!PS-Adobe-3.0
>=20
> So somewhere, something says "%%" to the printer, while there is still =
some PJL to be sent. But as said: I can't figure out the code path.
>=20
> I can't think of anything else to try. Tried to get DEBUG_printf to =
work but my grasp of the whole cups/filtering stuff is too small to =
actually be able to get that to work. Also, "debian/rules clean" for =
some reason messes up the code base and I'm sort of unable to rebuild =
cleanly without re-unpacking the debian source. But that's all in the =
category of "I'm not an actual programmer" so you shouldn't bother.
>=20
> But it is an actual bug that could potentially bug all printers with a =
ppd file that has PJL commands.
>=20
> Best regards,
>=20
> Valentijn
> --=20
> Durgerdamstraat 29, 1507 JL Zaandam; telefoon 075-7100071
>=20

________________________
Michael Sweet