IRC Logs

Log of #ghostscript at irc.freenode.net.

Search:
 <<<Back 1 day (to 2011/10/01)2011/10/02 
sambocyn hello, i'm experiencing a bug when using preview-latex04:50.23 
  it is: "Error: /typecheck in --setfileposition--"04:50.37 
  i am running mac os x 10.7, emacs 23 with auctex, and gs9.02, if that helps04:52.45 
  so when i generate a preview in emacs, the math equations do not render04:53.32 
  i much prefer the emacs mode to lyx, as a WYSIWYG latex editor, and really want to get it working04:54.41 
  thanks!04:54.49 
marcoagpinto Hello!10:18.15 
  I have a question10:18.19 
  is anyone there who can answer it?10:18.29 
tor8 how can we know unless you actually ask it?10:18.46 
marcoagpinto ahhh10:18.51 
  sorry10:18.52 
  I have my .DOC in Times New Roman but when I print to cutewriter using ghostscript the font changes to "Times Roman" instead of "Times New Roman"10:19.27 
  is there a way of fixing it?10:19.33 
LaoLang_cool hello, is it a bug for mupdf on *nix? hit j at the end of a page will make mupdf go to the next page. Such behavior doesn't exist in windows version10:20.58 
sebras Robin_Watts: seems like ghostbot needs to be updated to trigger on another string...10:21.22 
tor8 marcoagpinto: kens is the person most qualified to answer, but he's not here at the moment. stick around, we usually answer most questions here, but not always immediately.10:21.51 
marcoagpinto tor8: it is Sunday, will he be here today?10:22.15 
tor8 well, sometimes he's here on weekends but he'll definitely be here on weekdays10:22.49 
marcoagpinto ohhhh10:23.00 
  thanks10:23.04 
  :)10:23.05 
tor8 so come back tomorrow and you'll definitely get a better answer than I can provide10:23.19 
  I'm not very familiar with the pdfwrite device10:23.26 
sebras marcoagpinto: on my locally compiled mupdf j does not change pages. nothing happens. are you compiling mupdf from some git repository over at github.com?10:23.29 
tor8 I do know that PDF tends to assume that Times Roman and Times New Roman are equivalent10:23.43 
sebras marcoagpinto: I know they have added such unix keybindings...10:23.51 
tor8 and substitutes between the two everywhere10:23.56 
marcoagpinto ahhhh10:24.04 
  tor810:24.08 
  :)10:24.09 
LaoLang_cool sebras, did you talk to me?..10:24.11 
tor8 sebras: you're answering the wrong person :)10:24.17 
sebras aww.. crap. :)10:24.30 
LaoLang_cool sebras, mupdf on FreeBSD have this weird behavior, version is 0.910:24.49 
sebras glad the correct recipient was alert though. :)10:24.53 
  LaoLang_cool: where did you get the source from? mupdf.com?10:25.34 
tor8 LaoLang_cool: the windows and unix apps should have identical key bindings. freebsd may have patched their version though.10:25.42 
sebras tor8: or some github fork...10:25.59 
LaoLang_cool sebras, get it from http://www.mupdf.com/download/10:26.46 
  http://www.freebsd.org/cgi/cvsweb.cgi/ports/graphics/mupdf/Makefile?rev=1.18;content-type=text%2Fx-cvsweb-markup10:26.57 
  This is the makefile for mupdf10:27.06 
  omg! http://www.freebsd.org/cgi/cvsweb.cgi/ports/graphics/mupdf/Makefile?rev=1.18;content-type=text%2Fx-cvsweb-markup10:27.35 
  omg OPTIONS=SCROLL"Build with scroll hacks" on \10:27.41 
  I find the issue10:27.45 
  thanks guys, sorry for noice..10:27.53 
sebras great. :)10:27.54 
  no problem. now -- breakfast! :)10:28.08 
LaoLang_cool bad hack which is the default...10:28.10 
tor8 LaoLang_cool: not a problem, we might actually want to grab their patch if its useful10:28.31 
LaoLang_cool tor8, yes, if it's an option10:28.49 
sebras tor8: currently hjkl scrolls around on the page and there are no other keybindings with overlapping functionality...10:29.50 
LaoLang_cool I don't know why mupdf hasn't bookmark support, is it a viewer's job or render's job?10:30.05 
sebras tor8: for the xcb-port however...10:30.06 
  LaoLang_cool: viewer.10:30.18 
  LaoLang_cool: and the reason is that mupdf currently links to xlib directly instead of a toolkit.10:30.40 
tor8 LaoLang_cool: viewer, we don't use a GUI toolkit so implementing bookmarks would be a sizeable amount of work10:30.47 
LaoLang_cool q feature request: please add keybinding for scroll up to <shift-space>10:30.50 
sebras LaoLang_cool: mupdf would need that to draw the bookmark menu...10:30.53 
LaoLang_cool sebras, I like xlib, small and enough, don't like bloat gtk or qt or others10:31.26 
sebras tor8: maybe we should make the keybindings configurable in ~/.mupdfrc for the xcb-port? seems like there's a lot of people that want to modify them.10:31.36 
tor8 sebras: yeah, I think that's the way to go10:31.48 
LaoLang_cool tor8, why not add a cli tool to show the bmk to std out?10:31.50 
  section ----- page <num>10:32.24 
  so I can get the section info and use <num>G to go to the page10:32.44 
sebras LaoLang_cool: there is a plan to revamp the viewer bundled with the mupdf library (this is the xcb-port I'm going on about). it's being developed albeit slowly.10:32.47 
LaoLang_cool sebras, good news, but I still want to see a cli pdf_getbmk :)10:33.21 
  No gui at all, cool!10:33.45 
tor8 LaoLang_cool, sebras: it wouldn't be all that much work to add an "outline" option to pdfshow10:34.15 
LaoLang_cool what do you guys think?10:34.22 
  tor8, cool!10:34.26 
tor8 or pdfdraw even10:34.35 
LaoLang_cool I like cli hehe10:34.35 
tor8 we all do :)10:34.43 
sebras tor8: hm. maybe I should try to link mupdf to aalib...10:35.02 
LaoLang_cool so please keep it lest dependents ;p10:35.17 
sebras tor8: hm. maybe I should try to link mupdf to aalib...10:35.26 
  agh!10:35.33 
  LaoLang_cool: tor8 intends not to revert to any standard toolkit even for the xcb-port, so I believe that you needn't worry.10:36.14 
LaoLang_cool sebras, wow, tor8 is a cool guy, I like him much!10:36.40 
  I find an issue, some fig in pdf drawn too light, but in other pdf the drawing is normal10:37.22 
  the drawing will be normal if I scale down the pdf10:37.54 
tor8 if you compare with acrobat reader, make sure to turn off the "enhance thin lines" option10:38.02 
LaoLang_cool tor8, but in other pdf viewer, say pdfvx_viwer, the drawing is like adobe too10:38.56 
  maybe they are using the same tech for pdf displaying?10:39.09 
  then why mupdf hasn't such feature?10:39.18 
tor8 in that case, open a bug and attach the file at bugs.ghostscript.com and I'll look into it10:39.27 
  afaik only acrobat has that option10:39.47 
  which is essentially turning off anti-aliasing for thin lines10:40.00 
LaoLang_cool will do it10:44.21 
  oh, website is slow here...10:50.06 
marcoagpinto ahhh... I fixed my issue: I selected in CutePDFwriter: "TrueType Font: Download as softfont" and "TrueType Font Download Option: Native TrueType"10:50.33 
  but, I have a question10:51.24 
  is the font inserted on the PDF or will the PDF download from Internet when the file is opened?10:51.51 
sebras marcoagpinto: the fonts are inserted into the PDF when it is generated.10:52.49 
marcoagpinto ahhhh10:52.57 
  thanks10:52.58 
  I was not 100% sure so I asked10:53.05 
  :)10:53.07 
  I can now create PDFs with Times New Roman10:53.31 
  :)10:53.32 
  I am glad I noticed this before finalizing my university thesis10:54.00 
sebras marcoagpinto: oh, congrats! :)10:55.02 
LaoLang_cool tor8, sorry for my english, anyway, I've describe the issue at my best...hope you can understand... bug 692564 is submitted.10:55.12 
tor8 LaoLang_cool: ah! hairline widths again...10:56.14 
LaoLang_cool tor8, what do you mean?10:56.31 
tor8 according to the PDF spec, lines with width 0 should be rendered as 1 pixel wide10:56.37 
  our detection allows a bit of fuzz. the sample file triggers that detection when you zoom out a bit.10:57.10 
  perhaps we ought to tweak that value a bit more, currently we trigger when the line width in device pixels becomes 1/10th of a pixel10:57.57 
LaoLang_cool tor8, too professional, I just hope my pdf can be rendered more readable in the next release ;p10:59.00 
tor8 the file uses a linewidth of 1.2, scaled by 0.1 so those lines are supposed to be REALLY thin10:59.18 
  if it's your document, I'd recommend that you use fatter lines10:59.47 
LaoLang_cool tor8, you mean the pdf is in a bad format?10:59.58 
  typeset in a bad format?11:00.10 
tor8 the figure uses really really narrow lines, that if printed will be barely visible11:00.26 
LaoLang_cool I got it, the doc is mine, I transformed the wmf source file to eps to include in my .tex file, I don't know how to make the line thicker...11:01.41 
tor8 those lines are 0.042mm wide11:01.54 
  where did you get the WMF file?11:02.30 
  you could edit the eps file and change the "setlinewidth" command if all else fails11:03.05 
LaoLang_cool tor8, the fig actually exported from a chemical workstation, I can't find there's an option to set the line's width, I export the fig into an .ppt file, then ues ppt's exporting fig into .wmf, then trans the .wmf into .eps :)11:03.50 
tor8 right, so editing the eps in a text editor or inkscape to change the linewidth would be the easiest option11:04.34 
LaoLang_cool tor8, I've found the setlinewidth in .ps file, but setlinewidth doesn't be followed by other value, how to specify it?11:04.57 
tor8 it's the value preceding it11:05.08 
  1.2 setlinewidth (or something similar)11:05.17 
LaoLang_cool /Lw/setlinewidth11:05.23 
  No number value...11:05.30 
tor8 ah, that's probably redefining the command11:05.33 
  so search for "Lw"11:05.41 
  I'm sure it's "/Lw/setlinewidth bind def" or something similar11:06.00 
  which makes Lw an alias for setlinewidth11:06.15 
LaoLang_cool tor8, /bn/bind ld/lw/Lw ld11:06.24 
  Too complicated to me...11:06.29 
  d}if/b{bind d}bind d/bd{bind d}bind d/xd{~ d}bd/ld{, d}bd/bn/bind ld/lw/Lw ld11:06.36 
tor8 can you send me the file?11:06.37 
LaoLang_cool yes, how to?11:06.42 
tor8 attach to the bug report11:07.05 
LaoLang_cool ok, bug website is too slow here... please wait11:08.12 
  tor8, seems the file has been uploaded ?11:10.05 
tor8 can't see anything yet11:10.41 
LaoLang_cool tor8, http://bugs.ghostscript.com/attachment.cgi?bugid=692565&action=viewall11:11.40 
  the test file application/postscript 2011-10-02 11:08 UTC37.01 KBno flags Details11:11.47 
tor8 ah, you made a new bug11:12.11 
  anyway, I have the file11:12.19 
LaoLang_cool oh! I didn't mean to do it...11:12.29 
  please delete the wrong bug...11:13.51 
tor8 will do11:13.57 
LaoLang_cool thanks11:14.02 
tor8 hm, yeah, that eps is a bit of a mess.11:18.56 
LaoLang_cool the file is transformed by metafile_to_eps_converter 1.5...11:19.40 
tor8 add the line "5 setlinewidth" before %%EndPageSetup and it looks more reasonable11:20.33 
LaoLang_cool tor8, thanks, yes, it looks good now11:22.07 
  tor8, thanks for many help, too late to time for dinner here, I must go! thanks again, mupdf is cool11:23.42 
tor8 take care11:23.56 
sebras tor8: are you testing for xps vs pdf with if (app->xref), and is the default 72dpi for PDF and 96dpi for XPS?11:43.34 
  just so I understand the code.11:43.45 
tor8 yeah, there's an adjustment for XPS which uses 96dpi measurements and PDF which uses 72dpi measures11:49.19 
LaoLang_cool mupdf, it's about pdf, so why other fomat?12:31.04 
sambocyn Hello. I have found a gs bug, "/typecheck in --setfileposition--" when invoking preview-latex from emacs. (gs v9.02, emacs23 with auctex, mac 10.7). Can anyone help?14:40.17 
alexcher sambocyn: please file a bug report at http://bugs.ghostscript.com16:11.47 
sambocyn sure16:12.10 
  alexcher, done and thanks16:24.42 
alexcher sambocyn: I think, the problem is in the file preview.ps, which has left some junk on the stack.19:21.41 
  sambocyn: The logic of the program suggest that this file should not change the stack.19:22.29 
 Forward 1 day (to 2011/10/03)>>> 
ghostscript.com
Search: