| | <<<Back 1 day (to 2019/09/25) | Fwd 1 day (to 2019/09/27) >>> | 20190926 |
| kens | fluffypony: ShowAnnots does indeed override ShowAnnotTypes. If you disable all annots then we don't bother checking the types (they are all disabled) | 08:22.31 |
| chrisl | fluffypony: one thing that occurs to me: Are you sure they links that aren't being dropped are actually Link annotations? Acrobat and various other viewers have an annoying habit of spotting URLs in text, and making it appear like a Link annotation | 08:22.40 |
| kens | chrisl its because he's using pdfwrite (i was bout to write this) | 08:22.53 |
| | The pdfwrite device uses PreserveAnnots, and that does not honour ShowAnnotTypes | 08:23.19 |
| | ShowAnnotTypes is limited to rendering | 08:23.33 |
| chrisl | Oh, okay... that's a bit surprising | 08:23.53 |
| kens | Two different code paths I'm afraid | 08:24.05 |
| | One has to render the annot, the other has to create a pdfmark from it | 08:24.15 |
| | I do believe it would be useful to have the smae filtering available when preserving Annotations though | 08:24.49 |
| | I'll try and remember to create an enhancement bug report (unless fluffypony would like to do so, and supply an example, that would be useful) | 08:25.24 |
| chrisl | I just would have expected ShowAnnotTypes to he handled at a higher level that deciding between rendering and pdfmark - but then, that's applying logic to the PDF interpreter.... silly of me! | 08:26.02 |
| kens | I guess if I'd been writing it from scratch, that's exctly what I would have done, but the pdfwrite stuff was retro-fitted to the existing annotations code | 08:26.41 |
| | I think by the time I came to it the split had already been made. I added the ShowAnnotTypes IIRC, but it just didn't occur to me at the time to duplicate it for the pdfwrite preserve annots | 08:27.25 |
| | I suspect I can reuse much of the same logic | 08:27.38 |
| | chrisl bug #701313, did you manage to do magic for shared libtiff ? It would be really nice to get that in the release if we could. | 08:29.40 |
| | Mind you, that latest comment looks like spam | 08:30.37 |
| chrisl | kens: No, I never looked at it further, because no one had replied | 08:30.39 |
| kens | Yeah it was the reply that motivated me to think again, but it looks like the reply is spam, so I will go disable the user | 08:31.18 |
| | Done. | 08:33.20 |
| chrisl | kens: http://git.ghostscript.com/?p=user/chrisl/ghostpdl.git;a=commitdiff;h=8992f00edfd1c39154c013489de2a01d2e9a92ee | 08:45.54 |
| kens | That seems entirely reasonable to me | 08:46.22 |
| chrisl | Except, it's the wrong way around.... oops | 08:47.22 |
| kens | Is it ? I thought if SHARE_LIBTIFF was 0 then we were using a static linkage, so having the error handler was coprrect | 08:47.54 |
| chrisl | Oh, god, no it isn't - I'm not coffeed up enough yet :-( | 08:48.06 |
| kens | is releived | 08:48.15 |
| chrisl | Okay, coffee... | 08:50.49 |
| kens | Good plan, I think I'll have some too while I wait for my test run | 08:51.05 |
| fluffypony | apologies for the delayed reply | 15:11.24 |
| | chrisl: what's the best way for me to go about checking that? | 15:11.34 |
| kens | fluffypony: see my comments | 15:11.49 |
| fluffypony | ah ok got it | 15:12.16 |
| kens | You can't do what you want with the pdfwrite device because it uses PreserveAnnots not ShowAnnots and ShowAnnotTypes doesn't work with PreserveAnnots | 15:12.18 |
| fluffypony | ok so then the best path is to nuke all annotations | 15:12.41 |
| kens | I'm open to making it do so (or something similar) but I'm wrapped up with other problems right now | 15:12.43 |
| fluffypony | preserving annotations is pretty low-priority for us, crippling phishing PDFs is the critical path and ShowAnnots=false will do that just fine | 15:13.15 |
| | thanks so much for coming back to me | 15:13.21 |
| kens | Which is why I suggest opening an enhancement bug report, my memory is notoriously porr and I will likely forget all about it without a prompt | 15:13.26 |
| fluffypony | 100% will do so | 15:14.30 |
| kens | That's great, thanks! | 15:14.43 |
| wchannell | Hello | 18:40.10 |
| ghostbot | Welcome to #ghostscript. If you have a question, please ask it, don't ask to ask it. Do be prepared to wait for a reply as devs will check the logs and reply when they come on line. If you are looking for help or infomation about MuPDF, try the new #mupdf channel. | 18:40.10 |
| wchannell | I am using fedora linux with ghostscript. Ghostscript does not work with fedora 29 and 30. See Bug 1658013 - Ghostscript fails to print PDF file to screen in F29 with terminal mode gs command. Do any of the experts on this site have any insight as to a fix? Thanks. | 18:43.08 |
| | https://bugzilla.redhat.com/show_bug.cgi?id=1658013 | 18:44.36 |
| | Ghostscript has been inoperable on fedora (29 and 30) for at least a year now. Fedora 31 is on the verge of being released. Are there any ideas out there regarding how to fix this problem? Is it due to the os software or is it due to gs software? | 18:48.20 |
| | Ghostscript used to work great in fedora 28 and before. | 18:50.08 |
| kens | wchannell (for our logs) Nobody has reported a bug to us, and so no, we have no clues. If nobody reports bugs then they don't get fixed. | 19:25.29 |
| | <<<Back 1 day (to 2019/09/25) | Forward 1 day (to 2019/09/27)>>> | |
Log of #ghostscript at irc.freenode.net.