| <<<Back 1 day (to 2018/05/16) | 20180517 |
kens | tor8, sebras, FWIW that recent bug report against 1.12, the file works fine with 1.13 | 09:38.18 |
| However, the company looks like they are probably a licence infringer, so I'm sending the details to Kate. | 09:38.54 |
| Oops sorry they say 1.13. | 09:39.26 |
| My current MuPDF SHA1 is e6dfa88 and works fine on Windows. | 09:40.03 |
| Its possible of course they are using a library build. Note that the filename includes invisible non-ASCII characters. | 09:41.05 |
tor8 | kens: bug 699365? | 09:42.16 |
kens | Yes | 09:43.14 |
tor8 | I can't make heads or tails of his grammar. | 09:43.20 |
kens | and 363 but I closed that as a duplicate | 09:43.23 |
| I believe he's complaining that MuPDF won't open a file | 09:43.34 |
| WHich he has at least attached | 09:43.41 |
| The filename is not as simple as he indicates | 09:43.57 |
| It includes soe wei characters | 09:44.10 |
tor8 | so it does | 09:44.16 |
kens | GS doesn't like it | 09:44.27 |
| FOr me MuPDF is fine (on WIndows) | 09:44.36 |
| Is it possible he is using a library of MuPDF and not UTF-8 encoding the filename ? Does MuPDF need that ? | 09:45.39 |
tor8 | chrome on windows saves it with some weird '=-UTF-8-Q-=20CA... etc escaping | 09:45.43 |
| I think it's an em-dash unicode character- | 09:46.18 |
kens | GS has octal 342, 300, 224 where teh dash is | 09:46.28 |
| Could easily be an emdash | 09:47.07 |
| Just wanted to let you know I was sending an email to Kate really | 09:47.25 |
tor8 | yeah. that's the same octal I see on linux. | 09:47.29 |
avih | tor8: btw, re sort and setlength x2 vs loop of delindex, i still think setlength is preferable as it has the potential to be more efficient, even if it currently isn't. the loop is by definition O(length), while setlength could definitely be less, theoretically. | 09:47.53 |
tor8 | kens: yeah. with that level of language and "Hi Team" introduction I don't blame you for suspecting infringement | 09:48.04 |
kens | :-) | 09:48.11 |
tor8 | avih: setlength is not per spec. it will *not* do the correct thing for non-Array objects. | 09:48.33 |
avih | it's also, logically, sort of duplication | 09:48.36 |
| oh | 09:48.43 |
kens | Their website says they are a 'global business process outsourcing company' | 09:48.45 |
avih | tor8: well, at least for arrays, which i'd imagine is the main object of sort, it should | 09:49.34 |
tor8 | kens: strangely I'm getting blank pages on mupdf-gl.exe on windows but pages with text using mupdf.exe (the release 1.13.0 windows binaries) but that seems unrelated to the filename at least | 09:50.06 |
| avih: it will, but you can apply the sort function to non-array objects and it should still work | 09:50.29 |
avih | tor8: obviously, but js_isarray is just one call away :) | 09:50.50 |
| (assuming it exists, didn't check) | 09:51.04 |
kens | tor8 the pages are OK for me on Windows, but I'm not using GL | 09:51.25 |
avih | (and you can probably do that with just one comparison without function calls - with your internal knowledge :) ) | 09:52.15 |
tor8 | kens: strangely I can't reproduce it on linux... | 09:53.42 |
kens | That is indeed odd | 09:53.51 |
| I'm afraid I can't help you on that one though | 09:54.31 |
tor8 | And it seems unrelated to his issue. I'll just poke at it a bit myself, see if I discover something. | 09:55.17 |
kens | I'm sure its not related | 09:55.33 |
| oops forgot to CC tech | 10:01.25 |
| Geez, 'where's my fix ?' Seriously ? | 10:03.03 |
| Lets see what he's prepared to admit to :-) | 10:08.13 |
moolc | wow... this (presumably) Armenian bug reporters English is impeccable | 10:10.54 |
razi | Hi there. Does mupdf support extracting page labels of a PDF file (e.g.: i, ii, iii, vi,...)? | 14:06.59 |
kens | I'm certain you can extract the PageLabels dictionary from the document Catalog dictionary | 14:08.23 |
tor8 | razi: not very conveniently, I'm afraid | 14:08.35 |
| you could do something yourself, or look at what 'mutool show $PDF trailer/Root/PageLabels' shows and work from there | 14:09.40 |
razi | @tor: Thanks. I will check it. | 14:13.48 |
moolc | tor8: no luck with svg issue? | 15:33.09 |
razi | @tor8: Have you seen my update to this bug report https://bugs.ghostscript.com/show_bug.cgi?id=699317#c2 ? I have replaced it with empty string :D | 19:09.39 |
fredross-perry | tor8 - on 4/25 you introducd a change that breaks the Android build. The function load_noto_try in mudf_native.c. | 22:16.15 |
| If I fix the obvious compiler error, I then get things like error: undefined reference to '_binary_resources_fonts_urw_NimbusMonoPS_Regular_cff_start' | 22:16.41 |
| Oh wait, need to run make generate... | 22:19.22 |
| yes, that was it. But the compile issue in load_noto_try still stands. | 22:29.05 |
| Forward 1 day (to 2018/05/18)>>> | |