On Tue, 28 Oct 2014 17:24:08 +0000 Brian Potkin wrote: > On Sat 25 Oct 2014 at 12:46:48 +0200, Francesco Poli wrote: > > > Hello, > > Hello Francesco, thank you for the report. You're welcome! > > > I experience exactly the same bug: I often find PDF files which cannot > > be correctly printed with lpr from cups-bsd (some images and/or parts > > of text are missing in the hardcopy). > > To be clear here - we are dealing only with some PDFs which behave like > this and not every PDF? Definitely not every PDF file behaves like this. But, in recent times, it seems to have happened more often than it used to in the past. I cannot remember having experienced the issue with PDF files produced by myself (e.g.: with pdflatex): it usually happens when dealing with PDF files generated by third parties (e.g.: technical-scientific articles written by other people). Unfortunately, I haven't identified any clear pattern... :-( > Is it possible to quantify "often" in some way? It is not easy, since I tend to limit my use of printers (in an attempt to save some trees...). But I have recently noticed a worrying number of cases where the printed hard copy misses some images and/or parts of text. > > > Printing from evince and zathura mysteriously works... > > These applications use cairo; a PDF submitted to cups is not the same > PDF submitted by lpr and lp. It seems to me that cairo uses more robust ways to submit print jobs to cups: would it be possible to modify lpr so that it may use cairo to do the "rendering" before submitting the print job to cups? > In fact, lp and lpr send identical PDFs so > a poor output for lpr should also be a poor output for lp. Having a PDF > which produces an unsatisfactory print with both would be useful. I haven't ever used lp to submit print jobs: I usually use lpr, as I am used to its command-line options. Anyway, I tested the two files originally sent by jaakov [1], and I get blank pages when printing HensgenFinkelManber-TwoAlgorithmsForBarrierSynchronization.pdf with lpr, as reported by jaakov at the time. [1] the two attachments of https://bugs.debian.org/613987#5 > > > I experience the bug with a number of different printers, hence I don't > > think it is a hardware-specific issue. > > We could do with knowing what printers are used, their PPDs and how they > are connected. Two examples. HP LaserJet 1320n ----------------- connected via Ethernet LAN and configured with the following commands: # lpadmin -p lj -E -v lpd://x.y.z.w/lp0 \ -m foomatic:HP-LaserJet_1320-pxlmono.ppd \ -o pdftops-renderer=pdftops \ -D "HP LaserJet 1320" # lpoptions -p lj -o media=A4 -o sides=two-sided-long-edge Xerox Phaser 6280 DN -------------------- connected via Ethernet LAN and configured with the following commands: # lpadmin -p ph -E -v lpd://x.y.z.w/lp0 \ -m foomatic:Xerox-Phaser_6250N-Postscript.ppd \ -D "Xerox Phaser 6280 DN" # lpoptions -p ph -o media=A4 -o sides=two-sided-long-edge I hope that this may help in pinpointing the issue. Please let me know, if you need further information. Thanks for your time! -- http://www.inventati.org/frx/ fsck is a four letter word... ..................................................... Francesco Poli . GnuPG key fpr == CA01 1147 9CD2 EFDF FB82 3925 3E1C 27E1 1F69 BFFE
Attachment:
pgp4grnaWwQ1f.pgp
Description: PGP signature