| <<<Back 1 day (to 2014/11/18) | 20141119 |
zlnk | Hi, I have a question about mupdf on android, if on the C part instead of use one simple struct "global", I use an array of it, when the function get globals is called, it is instanciated the complete array or should I do an iteration?, sorry for the silly question | 14:33.31 |
kens | I can't see any reason why it would set any other than the first element of the array. | 14:35.21 |
zlnk | Oh, you are right, I will try that | 14:37.29 |
| thanks ;) | 14:37.40 |
kens | NP, Be aware I'm not a MuPDF developer though, so I can often be mistaken | 14:37.58 |
jogux | zlnk: I'm not sure I'm clear what you're trying to do? Why would you want multiple sets of globals? | 14:38.30 |
kens | wondered that too, but didn't like to ask..... | 14:38.46 |
zlnk | because the original pdf it`s splitted into many pdf`s of one page, and should be treated as if was one book | 14:40.25 |
| basically for that | 14:44.10 |
jogux | right. hm. I don't remember enough about jni to sensibly answer. Paul or Robin might. | 14:44.38 |
zlnk | Well jogux , I will try to return the first array and see what happen, and comment the result if is useful for someone later | 14:47.44 |
paulgardiner | I don't think you want an array of global structs. More likely you'd want an array of fz_document objects, or more likely still, you'd want to move "doc" out of globals and into the page_cache, so that the information for each cached page would include the fz_document | 14:53.50 |
Robin_Watts | Yes, what paulgardiner said. | 14:55.05 |
| I can tell you now that randomly poking at the code without really understanding it, is a great way to get nowhere fast. | 14:55.40 |
zlnk | thanks for the point, I understand how mupdf.c works, I have been studying the code and I wasnt sure if it was a good way, to manage global arrays out of the structure, as it was done on previous versions of mupdf, like 1.2 | 15:08.21 |
Robin_Watts | zlnk: Mupdf on android runs in response to intent calls. | 15:09.07 |
| As such, you can launch one version from say a file explorer, and another version looking at another file from mupdf itself. | 15:09.33 |
| Any globals you hold in the library are shared between the two instances. | 15:09.56 |
| hence using statics is bad, as it gets you into all sorts of problems (as we saw with 1.2) | 15:10.21 |
zlnk | ohh I see | 15:10.43 |
Robin_Watts | Therefore we now have a "globals" structure that is held one per user of the library, by storing it in the thread local storage. | 15:11.03 |
| So, for what you are doing you still want a single globals. | 15:11.24 |
| but you want an array of fz_documents. | 15:11.37 |
zlnk | that's true | 15:11.58 |
| it will be more comfortable to work on that way | 15:12.19 |
| thanks to all for the help :) | 15:26.07 |
kens | Well, that's another 17,000 diffs resolved. | 15:45.34 |
| Only 10,000 left..... | 15:46.09 |
Robin_Watts | Hmm. Latest email from Virgin... it will now cost at least 25 quid to choose your seats more than 24 hours in advance of flying. | 16:15.08 |
kens | O.O !! | 16:15.18 |
| I haven't received that meail yet.... | 16:15.41 |
Robin_Watts | http://www.virgin-atlantic.com/gb/en/the-virgin-experience/economy/seating.html | 16:15.43 |
jogux | BA charge 30 I think (though our travel agent managed to do it for free) | 16:15.48 |
Robin_Watts | It's wrapped up as a "look how we've improved things!" mail. | 16:16.06 |
kens | Well knickers to paying £25 for the privelge of booking early | 16:16.19 |
Robin_Watts | Red members can choose from 3 days in advance. | 16:16.42 |
kens | Whoppie do :-( | 16:16.50 |
Robin_Watts | Time to stop flying virgin. | 16:17.26 |
kens | And I bet you still can't select which extra legroom seat you get in advance :-( | 16:17.33 |
kens | is inclined to agree | 16:17.43 |
Robin_Watts | United give seatback power, and you can choose in advance, I think. | 16:18.04 |
kens | Yeah but the inflight entertainment is crap | 16:18.16 |
Robin_Watts | I rarely watch the inflight entertainment, especially if I have seatback power. | 16:18.36 |
kens | I do though | 16:18.44 |
| THere's not enough room in economy for me to work comfortably the whole flight | 16:19.08 |
Robin_Watts | It's a pain, cos I have loads of virgin miles. Have just swapped over to using my virgin credit card to generate more. | 16:19.11 |
kens | Oh.... | 16:19.18 |
| I'm not sure how many miles I have. A fair number I guess. | 16:19.33 |
| But if I can't guarantee a decent seat, I'm not flying with them any more, and I'm not coughing up £25 for it either | 16:20.02 |
| Not as long as someone else does it for free | 16:20.09 |
| Hmm seems I have 53k miles at the moment | 16:21.44 |
chrisl | Unfortunately, it seems several airlines are charging for it now :-( | 16:26.27 |
kens | Bunch of gougers :-( | 16:26.42 |
chrisl | Well, I get it for short/cheap flights, but for long haul flights, I feel it's a swizz | 16:27.25 |
kens | I see the hotel that charged for bad reviews is to stop doing that, now its hit the national press. Streisand effect anyone :-) | 16:27.44 |
chrisl | I kind of thought that would be illegal | 16:30.09 |
kens | SO did a lot of opther people, but they signed as part of teh T&Cs when they booked it | 16:30.31 |
| Trading Stanards are (or were) looking into it though | 16:30.48 |
Robin_Watts | kens: Nonetheless, I suspect it's unreasonable behaviour. I bet Trading Standards would kick them into touch. | 16:31.04 |
chrisl | I'd have thought that was an unenforcible clause | 16:31.35 |
Robin_Watts | And I bet a complaint to their credit card company would have resulted in the threat of their ability to take cards being removed. | 16:32.00 |
kens | Yes, all probably true, but a number of people used debit cards. | 16:32.22 |
chrisl | Pretty scummy thing to do. | 16:33.33 |
Robin_Watts | "This is a positive review of Hotel X. I had a positively horrible time when I visited. I thought the service was positively dreadful. I thought their "no negative reviews or we'll charge you 100 pounds" policy was positively inspired. I'm positive I'll never be going back!" | 16:33.47 |
kens | Given they were charging £36 for B&B for two, its probably the only way they made any money ;-) | 16:34.02 |
Robin_Watts | had to find B&B for Scott and his sisters for December. | 16:34.23 |
| One place wanted 150quid per room per night. | 16:34.39 |
| Another wanted 35 quid per room per night. | 16:34.48 |
| Guess which we went with? :) | 16:34.56 |
kens | That's quite a difference | 16:35.01 |
Robin_Watts | (Both have been recommended to us) | 16:35.10 |
kens | OK I'm off. I'm out tomorrow, bacxk Friday, night all. | 17:05.33 |
Robin_Watts | Night Kens. | 17:05.43 |
rayjj | yuck. Pattern scaling problems trying to match Adobe :-( This is an area where there are RJW footprints as well | 17:50.38 |
Robin_Watts | rayjj: The spec essentially says "do what you want". | 17:51.14 |
| So we'll go mad trying to match adobe in all cases. | 17:51.28 |
rayjj | Robin_Watts: yes, I know. But cust 532 is important enough for me to do a little more work | 17:51.58 |
mvrhel_laptop | bbiaw | 17:54.37 |
Robin_Watts | marcosw: ping | 18:07.22 |
ddaanet | Hello. | 18:56.52 |
| I was looking for an alternative to poppler's pdftohtml | 18:57.20 |
| And on the ghostscript docs about an xml output device I read about MuPDF | 18:57.49 |
Robin_Watts | ddaanet: mudraw with -t or -tt or -ttt. I forget which will get you something. | 18:57.55 |
ddaanet | Robin_Watts: Cool. I find binary tools for Windows downloadable on the mupdf website. Do they also build on headless linux? | 18:58.49 |
| I found no mention on that on the website, and I just want to know if I need to spend time investigating it. | 18:59.16 |
Robin_Watts | mupdf builds on everything. | 18:59.25 |
ddaanet | I got the idea, that the library builds on anything. But I found very little info about command line tools. | 19:00.04 |
| I'll have look tomorrow. Thanks. | 19:01.14 |
Robin_Watts | np. | 19:07.15 |
lilltiger | Hi, have there been any changes regaring the handling of colorspaces between 9.04 and 9.14/9.15 because I get the wrong colors when using 9.14/9.15 but the right ones with 9.04? the colorspace is PMS from indesign, I think it uses LAB colors. | 19:55.48 |
Robin_Watts | lilltiger: Colour Management has been a theme of 9.x development, but I don't think you should have seen radical changes between 9.04 and 9.14 | 19:58.22 |
| lilltiger: Are you in a position to narrow the change down any more? Can you git bisect? | 19:58.41 |
lilltiger | Robin_Watts: I am not :/ | 19:59.51 |
Robin_Watts | Then you'll have to open a bug on bugs.ghostscript.com, add the file that shows the problem, and hope for a developer to get bored. | 20:00.26 |
lilltiger | Could it be some change in defaults flags? | 20:00.35 |
Robin_Watts | I doubt it. | 20:01.00 |
lilltiger | Unfortunally I cant provide a test file now, using a logo from a customer, will try to get another test image | 20:10.25 |
| Robin_Watts: i tried with 9.10 and it get's the colors right, but I cant find the sources for 9.11-9.13 | 20:18.36 |
Robin_Watts | lilltiger: If you open a bug, and let us know, we can accept the file and keep it private. | 20:19.06 |
lilltiger | ohh ok | 20:19.34 |
| Robin_Watts: should i file it as "Color" or as "Color Management"? | 20:22.55 |
Robin_Watts | Either. | 20:23.05 |
lilltiger | Robin_Watts: Can you check this: http://www.colorguides.net/pantone_spot_color.html this seems to explain exactly my issue (even the same color) I get the CMYK color with 9.14/9.15 but the spot color with 9.10 (and i want the spot color). | 20:42.32 |
| Not sure if I should submit it as a bug, because I might be doing it wrong, and one now should specify that one wants the spot colors insted of cmky | 20:44.08 |
Robin_Watts | Submit it as a bug. If it's something you're doing wrong, I'm sure someone will point that out to you. | 20:47.18 |
lilltiger | ok :) | 20:47.28 |
Robin_Watts | (Remember, you are dealing with Software Engineers here, and there is nothing a Software Engineer does better than tell people that they are doing it wrong :) ) | 20:47.44 |
lilltiger | emm | 20:50.02 |
| Robin_Watts: I get "ERR_ACCESS_DENIED" when i try to submit a bug | 20:55.33 |
| Robin_Watts: Also using "-dColorConversionStrategy=/RGB" generates the colors I want. So I suppose the default print colors changed from RGB to CMYK. | 21:02.50 |
| But someone should realy check if there is something wrong with the bugzilla, as it's not all too good when submiting a bug gives access error ;D | 21:04.41 |
Robin_Watts | lilltiger: That doesn't sound good. | 21:31.19 |
| Works for me. Have you made an account? | 21:32.08 |
lilltiger | Robin_Watts: yes, made account and tried to reset password and relogg in etc | 21:41.04 |
| several times | 21:41.08 |
Robin_Watts | lilltiger. I am confused then. My apologies. | 21:41.30 |
lilltiger | not your fault | 21:41.42 |
Robin_Watts | Can you email details of everything to robin.watts at artifex.com please and I'll chase it up? | 21:41.48 |
lilltiger | Robin_Watts: about not being able to submit a bug? | 21:44.36 |
Robin_Watts | yes. And the bug details themselves - I'll open it. | 21:45.24 |
lilltiger | Robin_Watts: well, I dont think it is a bug | 21:45.37 |
Robin_Watts | If you can send me the file, I'll upload it to the bug but mark it as private. | 21:45.46 |
| OK, up to you. | 21:46.00 |
lilltiger | Robin_Watts: it might be the other way around, as adding "-dColorConversionStrategy=/RGB" gives the result i want, insted of the CMYK result | 21:46.27 |
| while it is very resonable that the CMYK result is the default | 21:46.49 |
| bug dubmit worked now in firefox | 21:47.50 |
| Robin_Watts: I now can submit in chrome as well, hmm how do i remove the tests? | 21:50.06 |
Robin_Watts | what tests? | 21:50.25 |
ghostbot | testing | 21:50.25 |
lilltiger | Robin_Watts: I tested submiting bugs agains, after you said it workt for you, and they both workt, but now there is two useless bug reports from me | 21:51.09 |
Robin_Watts | lilltiger. Oh, I see. I'll remove 'em. | 21:51.30 |
lilltiger | ok, thanks =) | 21:51.42 |
Robin_Watts | np. | 21:52.31 |
| Forward 1 day (to 2014/11/20)>>> | |