| | <<<Back 1 day (to 2018/07/01) | 20180702 |
| tor8 | paulgardiner: ping. | 09:43.54 |
| | I found that I have to call pdf_update_page (or loop all annotations and call pdf_update_annot) after changing a widget's field value otherwise I don't see the form changes | 10:23.41 |
| | especially when editing numbers in one of the columns, the sum total field will update with javascript but I need to make sure I call pdf_update_page to detect the changes | 10:24.08 |
| paulgardiner | tor8: pong | 11:15.46 |
| | tor8: I think that behaviour is what I expected. Certainly, in muso, I call pdf_update_page before iterating the annotations and looking for ones that have the has_new_ap flag set. | 11:17.55 |
| | Also, I ensure I do that using the fzpage object to which I sent the changing event or via which I altered an annotation. For your viewer that is trivially true I assume because you don't cycle fzpage objects. | 11:19.57 |
| | I think it was me that first introduced pdf_update_page and the intention was to mark the end of a sequence of changes to the document, ensure anything that needed recalculated was, update appearance streams and mark which annotations had changed. | 11:21.52 |
| tor8 | paulgardiner: yeah. pdf_update_page currenty only does a loop over pdf_update_annot, so it's almost largely redundant. | 12:53.24 |
| | the current forms code triggers recalculations on all widget value updates | 12:54.21 |
| paulgardiner | Actually it may have done that before, but still you don't want to regenerate the ap several times unnecessarily, so pdf_update_page is a good place to do that. | 13:05.54 |
| tor8 | paulgardiner: yeah, if doing many edits in sequence, regenerating the ap by calling pdf_update_page once at the end will do that. | 13:10.50 |
| | the field recalculation just flags the widgets as dirty; the AP update won't happen until pdf_update_page/annot is called. | 13:11.34 |
| | paulgardiner: do you still have issues with mittelabrufplan.pdf updates? | 13:11.56 |
| paulgardiner | I think so. Was that the doc I sent in my most recent error report? | 13:14.28 |
| | It was just the first field in the file that misbehaved, if I remember correctly. | 13:15.03 |
| tor8 | paulgardiner: which is the first field? "Projektleiter/Institution"? | 13:16.30 |
| | or do you mean the "Laufzeit von:" field? | 13:17.06 |
| paulgardiner | I'm not sure. That looks like the wrong file and I cannot test just at the moment because I have a very different version checked out. I'll have a look as soon as I can | 13:18.48 |
| tor8 | the only odd thing I found in that file, is the "von:" and "bis:" fields have javascript that enforces 'date' format strings (and blank out anything that doesn't match) | 13:19.38 |
| paulgardiner | Yeah, I noticed some files wont accept certain date formats and give no warning, but it may be just the way the file is constructed. I should test with AR again sometime. | 13:40.50 |
| tor8 | paulgardiner: I'd take a second look at util.js -- it seems to have issues. I get "TypeError: called object is not a function" quite a lot for those files where formatting goes wrong | 13:47.42 |
| paulgardiner | util.js may well have used some oddities that worked only in google's v4 (is it google, or am I misremembering where that came from) | 13:48.57 |
| | Didn't you already make quite a lot of changes to util.js for the sake of mujs and for the sake of good taste? :-) | 13:50.56 |
| | ... and efficiency possibly. | 13:51.30 |
| tor8 | paulgardiner: not really. I fixed a couple of oddities and bugs I ran across, and a few tweaks for mujs. | 13:51.38 |
| | it still abuses regular expressions quite heavily :) | 13:51.54 |
| paulgardiner | Possibly. I was quite pleased with some of the tricks with reg exps, giving much more concise coding of some of the functions than Adobe had, but I may well have messed up in places. | 13:54.02 |
| tor8 | paulgardiner: regexp's are fine for the most part :) | 13:56.06 |
| | I don't think fixing the JS stuff is high priority right now though. | 13:56.29 |
| paulgardiner | tor8: for those type error cases, might be good to raise a bug. I may well want to have a look at them, but probably not for a while. | 13:57.58 |
| | tor8: what should I use for testing now? Your master? | 15:38.18 |
| tor8 | paulgardiner: tor/master would be good | 15:39.20 |
| paulgardiner | tor8: those util.js problems may become important because refusing to accept input for a form and also failing to display the warning as to why is quite a nasty behaviour. | 15:44.11 |
| moolc | tor8: any approximation on when tor/nostar will make it to origin/master? | 15:46.21 |
| paulgardiner | tor8: mittelabrufplan.pdf (which I hope isn't a German swear word because I would be none the wiser) seems to work fine now. Is that definitely the attachment in the most recent problem report I emailed? | 15:50.30 |
| tor8 | paulgardiner: the date formatting code doesn't do at all the same as adobe's | 15:51.06 |
| | which either returns something or throws up an alert box refusing it | 15:51.34 |
| paulgardiner | Strange: I remember using adobe's code as a spec. | 15:51.52 |
| | ... a version of their code from who knows when admittedly | 15:52.14 |
| | I do get an alert box in most cases, so I was assuming that the problem was that some of the javascript was failing in an unrecoverable way and missing the alert | 15:53.51 |
| | Perhaps it isn't the formatting code that is supposed to perform the test | 15:54.37 |
| tor8 | paulgardiner: oops. I made a mistake cleaning some stuff in util.js | 16:02.39 |
| | there should be a fix on tor/master that fixes it onw | 16:02.51 |
| | fixes the javascript date formatting | 16:03.17 |
| avih | tor8: isn't altering Error.toString unclean? what if some code not under your control uses toString? | 16:07.06 |
| | i'd think a nicer solution would be some global catch which prints the stack | 16:07.57 |
| | for mpv i added Error.stack accessor which does what your patch did for Error.toString(), and a global catch which prints .stack if possible | 16:10.04 |
| | and my code tests if (this.stackTrace && !this.stack) ... | 16:11.35 |
| | (before setting .stack) | 16:12.13 |
| Nilsont | Hi there... is there any way to create highlights into epub files using MuPdf?? | 19:42.58 |
| | Hi there... is there any way to create highlights into epub files using MuPdf?? | 21:23.13 |
| | Forward 1 day (to 2018/07/03)>>> | |
Log of #mupdf at irc.freenode.net.