| | <<<Back 1 day (to 2018/11/14) | 20181115 |
| sebras | tor8: ok, I give up. | 03:44.21 |
| | tor8: I have a few changes on sebras/master for mupdf-android-viewer-mini and mupdf-android-fitz | 03:44.58 |
| | with these I can compile in git-repos that I have checked out a long time ago. | 03:45.16 |
| | however checking out mupdf-android-fitz anew in a clean directory ends with "Could not resolve all files for configuration ':lintClassPath'." for me. | 03:46.27 |
| | next it gives me a list of directories in android/sdk where it has been looking for lint-gradle-26.1.10, and I'm not sure where this comes from. | 03:47.30 |
| | rerunning with ./gradlew --info assembleDebug has the first failure "Taks :extractDebugAnnotations FAILED" beacuse it cannot find mupdf-android-fitz/src/main/java even though we on purpose do not include this directory in srcDirs. | 03:48.40 |
| | whyyyyy?! :) | 03:50.24 |
| | also I hate the fact that developing for android _always_ ends in an endless search for discussions on forums on groovy, gradle, maven, android on stackoverflow and elsewhere. I feel that this never happens when developing in other languages.... | 03:53.49 |
| | I think I might have cracked it just now actually. | 03:55.11 |
| | we had jcenter() and google() only in buildscripts.repositories, but not int allprojects.repositories. I'm too tired to read up on the difference though. | 03:55.49 |
| | oh and we did have it at both locations in mupdf-android-viewer-mini, but not mupdf-android-fitz... | 03:56.14 |
| | tor8: ok... so with what is currently on sebras/master in those two repos I can now successfully compiler and build for android | 04:13.03 |
| | the Quad stuff is still not fixed proper. | 04:13.13 |
| | if we just want to compile 1.14.0 we can probably remove my ugly Quad-hack since 1.14.0 doesn't appear to have Quads yet. | 04:14.08 |
| | tor8: oh and mupdf:sebras/master has had its commits updated too. I opted for the $(sort $(wildcard)) pattern used elsewhere, even for this javah replacement. I hope that would iron out all the sorting issues. | 04:19.46 |
| | time to sleep. | 04:19.51 |
| | tor8: ugh.... https://developer.android.com/distribute/best-practices/develop/target-sdk | 22:55.07 |
| | apparently we must target api level 26 starting from the 1st of this month. that means we must adapt before we do the next upload. | 22:55.42 |
| | I guess this means finally sorting out the permissions model. | 22:55.57 |
| tor8 | sebras: I think we've got that mostly sorted already | 23:14.31 |
| | or at least I remember working on it at one point | 23:14.40 |
| | but it says "new apps" -- our apps are not new. does this also affect updates? | 23:15.24 |
| | yeah, we already do targetSdkVersion 25 | 23:15.57 |
| | moving to 26 shouldn't be a problem | 23:16.09 |
| sebras | yes, it applis for app updates after november 1st. | 23:22.14 |
| | new apps from august 1st. | 23:22.20 |
| | tor8: oh... it seems to me like it is only the old viewer that targets level 16. | 23:24.56 |
| | tor8: I was under the impression that it was all of them. | 23:25.06 |
| | tor8: I'm surprised that there are no commits changing targetSdkVersion. it appears to have been 25 since the start of viewer-mini and viewer. | 23:26.34 |
| | Forward 1 day (to 2018/11/16)>>> | |
Log of #mupdf at irc.freenode.net.