| | <<<Back 1 day (to 2015/11/17) | 20151118 |
| sebras | fredross-perry: testing 789 | 00:15.38 |
| fredross-perry | well played | 00:15.47 |
| Robin_Watts | testing .0# ? | 00:16.38 |
| fredross-perry | Indeed. | 00:17.27 |
| marcosw | Robin_Watts: speaking of working late... | 00:21.12 |
| HenryStiles | mvrhel_laptop: hey sorry I was away before what's up? | 04:03.55 |
| kens | and again henrysi7 produces a seg fault on a cluster test | 11:11.27 |
| chrisl | Same file or different file each time? | 11:15.45 |
| kens | different every time as far as I can recall | 11:15.58 |
| chrisl | So, it could be a fault with the machine, or a compiler specific issue..... | 11:16.57 |
| kens | Yes, could be either, I mention it because I'm now mostly seeing 1 seg fault per cluster run, always on henrysi7 | 11:17.30 |
| | I'm inclined to remove the node since its causing problems. | 11:17.55 |
| chrisl | We could disable that machine, pending investigation | 11:17.59 |
| kens | tor8 our emails crossed :-) | 11:26.51 |
| tor8 | kens: and we both checked different things :) | 11:31.00 |
| kens | Well I thought it was worth actually checking the MS viewer | 11:33.58 |
| | I bet the customer didn't..... | 11:34.11 |
| HenryStiles | kens: I'll look at the logs here. I am doing work on the machine now and don't see anything odd. | 14:28.30 |
| kens | HenryStiles : I don't know what might be up, at the moment its merely an observation | 14:28.56 |
| | I'ver seen 3 inexplicable seg faults over the last few days, and 2 of them have been on that node. The other one might have been, it was the first I noticed and I didn't note which node it was | 14:29.55 |
| Robin_Watts | kens: Same file, or random files? | 15:11.38 |
| kens | Robin_Watts : the file seems to differ each time | 15:13.26 |
| | It seems, at least generally, to be be low resolution | 15:13.54 |
| HenryStiles | kens: the one I have here is pcl -> pdf -> ppm crashing going to ppm. But it happens with single transformations as well I assume. | 15:16.27 |
| kens | HenryStiles : I'm not certain | 15:17.24 |
| | I haven't checked ths in any detail, its relatively new (as is the node) and doesn't go wrong often. I only recently noticed that it was ths particular node | 15:18.02 |
| faLUCE | hello. I have onepage.pdf .How can I remove 10 pts left, 20 pts right, 30 pts top and 40 pts bottom from it? | 16:56.14 |
| kens | Set a /CropBox | 16:56.29 |
| | http://ghostscript.com/regression/cgi-bin/clustermonitor.cgi?log=log&machine=henrysi7&report=d368650e9f3e7e6e4af2d840ffa0dd1e161a9694&project=ghostpdlOr modify any existing one | 16:56.38 |
| | Grrr | 16:56.43 |
| | Or modify an existing CropBox if there is one present | 16:56.53 |
| faLUCE | kens: can you write again the link? | 16:57.25 |
| kens | Ignore the link, its an accident | 16:57.36 |
| | You need to add a /CropBox to the page, or modify any exsiting CropBox | 16:58.00 |
| | Alternatively, go back to the source document, modify it appropriately, and remake the PDF | 16:59.15 |
| faLUCE | kens: how can I modify the source doc? | 16:59.32 |
| kens | Open it in the relevant application, and do whatever you have to in order to make the changes, how should I know precisely ? l have no idea what application wa used to create hte original. | 17:00.39 |
| faLUCE | kens: I have to remove spaces of many pdfs from command line | 17:02.34 |
| kens | Well, I'd start by berating however created them incorrectly | 17:02.57 |
| | Like I said, add a /CropBox to the Pages tree. If the files have multiple pages and each page has a CropBox you will have to modify each Page object | 17:04.03 |
| Robin_Watts | faLUCE: Didn't you ask about this last week? | 17:07.29 |
| | kens: If I understand faLUCEs request properly, a cropbox will not do it. | 17:08.04 |
| | He needs to modify the Mediabox. | 17:08.11 |
| kens | Robin_Watts : Possibly, its not clear at all from what he is saying | 17:08.23 |
| | And simply modifying hte MediaBox won't work, as the 0,0 will still be at bottom left, so the content won't shift. | 17:08.50 |
| Robin_Watts | faLuce: Do you want to keep the page the same size and just restrict the area on it that is marked (so you get blank borders) or do you want to shrink the page ? | 17:08.55 |
| | kens: We've discussed that before, and I fear you're wrong. | 17:09.09 |
| kens | If he wantsa to modify the content (and I can see no reason to do ths otherwise) then he'll need to set a fixed page size, set a BeginPage or Install procedure to modify the CTM by translating before rendering. Then run the whole thorugh pdfwrite to create a new PDF file | 17:09.48 |
| | Robin_Watts : possibly you're correct :-) | 17:10.07 |
| Robin_Watts | There has to be a first time for everything. | 17:10.19 |
| kens | With no example file, no clear explanation, and no GS command line, I'm fumbling in the dark | 17:10.29 |
| faLUCE | Robin_Watts: I can have blank borders (then I will remove them with pdfcrop) | 17:10.31 |
| Robin_Watts | I am unfamiliar with pdfcrop. | 17:11.03 |
| kens | does caguely remember altering the MediaBox from 0 0 and being surprised by the result | 17:11.11 |
| | http://ghostscript.com/regression/cgi-bin/clustermonitor.cgi?log=log&machine=henrysi7&report=d368650e9f3e7e6e4af2d840ffa0dd1e161a9694&project=ghostpdlBut like I said, no example file, and no explanation of what is required to be altered. | 17:11.40 |
| faLUCE | Robin_Watts: anyway, I can keep the same size and just restrict the area on it that is marked | 17:11.47 |
| kens | Bahm stupid insert key | 17:11.47 |
| faLUCE | how can I do? | 17:11.49 |
| kens | Set a CropBox | 17:11.57 |
| Robin_Watts | So how does pdfcrop work? | 17:12.37 |
| kens | suspects it sets a CropBox :-) | 17:12.51 |
| Robin_Watts | kens: That wouldn't remove the borders. | 17:13.06 |
| | I could believe it set the mediabox to the cropbox. | 17:13.17 |
| kens | faLUCE : To be perfectly frank, if you want help, post an example file somewhere we can see it, then write a clear description of what you want it to look like afterwards. Then maybe we will understand what you are trying to do, and be able to help | 17:13.31 |
| Robin_Watts | Or that it set the mediabox to the bbox of the marked content. | 17:13.37 |
| chrisl | "-First, PDFCrop calls ghostscript to find the borders of the PDF's bounding box. Next, it determines the page orientation of each page of the PDF file. Finally, it calls PDFedit to crop and resize the pages." | 17:13.55 |
| faLUCE | kens: Robin_Watts http://s000.tinyupload.com/index.php?file_id=21678900709178525082 <---- this is the file. I want to 1) split into two equals horizontal pages and 2) remove these black borders | 17:17.12 |
| kens | I posted some code to Stack Overflow some time back for splitting pages up. In that case it was a 4x4 powerpoint presentation that the user wanted split into 4 single pages | 17:18.52 |
| | I imagine that you oculd reuse that code | 17:19.03 |
| Robin_Watts | faLUCE: mutool poster will do the splitting for you. | 17:21.11 |
| kens | is having trouble finding my answer..... | 17:21.25 |
| Robin_Watts | mutool poster -x 2 in.pdf out.pdf | 17:22.03 |
| faLUCE | Robin_Watts: it is not exactly the center and, in addition I need also to remove black spaces | 17:23.07 |
| Robin_Watts | faLUCE: Well, mutool poster works by just setting mediaboxes. You could fiddle with the source of that :) | 17:24.07 |
| faLUCE | Robin_Watts: do you mean that I can specify these args through cli? | 17:24.55 |
| Robin_Watts | faLUCE: No. I mean you can fiddle with the C. | 17:27.01 |
| faLUCE | Robin_Watts: this is weird... there should be an easier and faster solution | 17:28.13 |
| kens | OK I found my PostScript program | 17:28.32 |
| | faLUCE : The fact that you *want* an easier solution doesn't mean there has to be one. | 17:28.51 |
| faLUCE | kens: with imagemagick there should be a way | 17:29.06 |
| kens | Imagemagick is an image manipulatoin program. | 17:29.37 |
| | Not a PDF manipulation tool | 17:29.43 |
| | Of course, what you actually have here is clearly an image wrapped up as a PDF file | 17:29.58 |
| | You could extract the image, then use a script-driven image manipulatoin package to modify the image. You could evne then probably re-export back to PDF | 17:30.31 |
| faLUCE | kens: I could leave the wrapped image, no matter if the size of the file is big. But how can I modify the borderbox? | 17:33.29 |
| | kens: I mean, should I modify the header with a hex editor? | 17:33.55 |
| kens | If you are extracting the image, you do it in an image manipulatoin program | 17:33.55 |
| faLUCE | kens: let's try a solution without extracting the image | 17:34.15 |
| | how can I modify the borderbox? | 17:34.23 |
| kens | There is no borderbox | 17:34.36 |
| | There are a series of boxes, including CropBox | 17:34.50 |
| | Ths won't help you if you want a PDF file with 2 pages, each of which is one of the pages in the origibal single page file | 17:35.10 |
| | Nothign you can readliy to to that file will turn it into a 2 page file. | 17:35.29 |
| faLUCE | kens: no; i don't need a pdf with two pages | 17:35.42 |
| | kens: I can have two pdfs as well | 17:35.50 |
| kens | Well what *do* you need ? | 17:35.55 |
| faLUCE | kens: I need to generate, from the orig pdf, two separate pdfs with the borders that I said | 17:36.24 |
| kens | Start by copying the PDF file | 17:36.43 |
| faLUCE | kens: then? | 17:36.51 |
| kens | Then modify the MediaBox of each | 17:36.51 |
| faLUCE | kens: ok, and in order to modify the mediabox, what should I modify? the hex file? | 17:37.08 |
| | through a hex editor? | 17:37.13 |
| kens | Any editor that does not mangle the binary | 17:37.26 |
| faLUCE | but: is there a cli for that, instead the hex editor? | 17:37.49 |
| kens | Warning, your first problem will be that you cannot add bytes | 17:37.51 |
| | A CLI for what ? | 17:38.07 |
| | Since you cannot add bytes, modifying the MediaBox is likely to be tricky | 17:38.21 |
| faLUCE | http://linux.die.net/man/1/pdfcmb | 17:38.25 |
| kens | Right, well use that then | 17:38.47 |
| | None of ths is relevant to Ghostscript or MuPDF mind you | 17:39.00 |
| | We are not a general purpose PDF support forum | 17:39.12 |
| faLUCE | kens: I know, sorry, but I really did not know where to start and ask | 17:39.40 |
| kens | TIme for me to go eat, goodnight all | 17:39.46 |
| faLUCE | kens: there's not a #pdf chan | 17:39.53 |
| | thanks kens, for all | 17:39.57 |
| kens | faLUCE : Try Stack Overflow | 17:39.58 |
| | faLUCE : One thing that occurs to me is that you could write somethign to read a PDF file, search for MediaBox, remove the lower left x,y and upper right x,y co-ordinates and replace them with ones you calculate. At the same time you could either modify or remove the Crop Box. You could write such a thing in C or probably a decent scripting language. | 20:30.42 |
| | Because PDF is a binsrty format, ths would break the PDF file. However, your files seem very simple, so you could use MuPDF to fix them afterwards. | 20:31.13 |
| | "mutool --clean broken.pdf fixed.pdf " I thnk is the correct incantation | 20:31.46 |
| Robin_Watts | mutool clean broken.pdf fixed.pdf | 20:43.42 |
| kens | Close.... | 20:44.15 |
| marcosw_ | mvrhel_laptop: you around? | 21:11.41 |
| mvrhel_laptop | marcosw: yes | 21:12.16 |
| marcosw_ | one of the icc testing command is: | 21:12.52 |
| | bin/gs -sDEVICE=ppmraw -sSourceObjectICC="./toolbin/color/src_color/objsrc_profiles_example.txt" -o test.ppm -f ./examples/text_graph_image_cmyk_rgb.pdf | 21:12.53 |
| | it produces this error: | 21:13.03 |
| | ./base/gsicc_manage.c:1098: gsicc_open_search(): Could not find cmyk_src_cyan.icc | 21:13.03 |
| | + ./base/gsicc_manage.c:660: gsicc_set_srcgtag_struct(): setting of ./toolbin/color/src_color/objsrc_profiles_example.txt src obj color info failed | 21:13.04 |
| | | ./base/gsicc_manage.c:2777: gs_setsrcgtagicc(): cannot find srctag file | 21:13.05 |
| | I tried adding â-I./toolbin/color/src_colorâ but that didnât help. | 21:15.16 |
| | never mind, I just saw your comment about copying the icc profiles. | 21:17.29 |
| mvrhel_laptop_ | ok | 21:17.34 |
| marcosw_ | mvrhel_laptop: no wait, that doesnât help. they arenât found in iccprofiles either. The have to be in â.â to be found. | 21:20.27 |
| mvrhel_laptop | marcosw_ : I had copied them into iccprofiles and then built, which puts them in the romfs | 21:21.11 |
| | if you want to put them in . and that works then that is fine | 21:21.29 |
| marcosw_ | iccprofiles is searched when using the romfs but not otherwise? | 21:22.22 |
| | in any case, putting them in . works for me. | 21:22.40 |
| mvrhel_laptop | good question. There probably needs to be a bit of work on the search paths for this sort of thing | 21:22.57 |
| | marcosw_: ok , well that will work for this testing | 21:23.26 |
| | bbiaw. need to go eat | 21:26.12 |
| marcosw_ | mvrhel_laptop: also, all the slashes go the wrong way :-) | 21:29.18 |
| | Forward 1 day (to 2015/11/19)>>> | |
IRC Logs