Well, Jonas, here's another item that ghostscript needs to attend to. Earlier in this problem record, I noted that the second control sequence sent to the printer was
ESC 3 0 (hex 1B 33 30), which sets the line spacing to 48/216 of an inch. That is double what it should be for 8-pin graphics. But then I said that it was moot because all the line spacing is done by sending ESC J ^X (hex 1B 4A 18), which is an explicit request to advance the paper by 24/216 of an inch, which is the correct spacing for 8-pin graphics. Well, it turns out not to be moot after all. It affects the printing of raw ASCII text files. Here's the scenario. I set DIP switch #6 (automatic carriage return) to the on position (while the printer is off of course). I then turn the printer on and send a Unix-style plain ASCII text file to the printer via lpr -Ppp2 -l plain.txt (My printer name in CUPS is pp2, and plain.txt is the name of the text file I want to print. The -l option means that raw data is being sent and it should go directly to the printer without passing through any filters.) The text file prints normally, single spaced. I then print the printer test page via the CUPS administration web interface. It prints correctly. I then print my plain ASCII text file again as above. This time, the spacing is almost double spaced. If I turn the printer off and back on again, and then print the file again, it prints correctly. So that ESC 3 0 sequence is being remembered and is not harmless after all. It messes up the printing of plain text. This escape sequence should not be sent at all! Perhaps we should split this into two problem records: one for the things that ghostscript needs to fix and another for the things that foomatic needs to fix. How should we do this? -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
