[01:10:14] *** Quits: genesis (~genesis@unaffiliated/genesis) (Remote host closed the connection) [03:04:14] *** Quits: mvglasow (~mvglasow@dslb-088-064-109-229.088.064.pools.vodafone-ip.de) (Ping timeout: 246 seconds) [03:16:57] *** Joins: mvglasow (~mvglasow@ipservice-092-211-162-207.092.211.pools.vodafone-ip.de) [04:58:41] *** Joins: genesis (~genesis@unaffiliated/genesis) [04:58:41] *** ChanServ sets mode: +v genesis [05:32:50] *** Joins: owlman (~owlman@202.92.105.238) [15:52:39] Hi thank you. Got it to work with changing the gradle version and getting saxon-b to work on macos yesterday. I have tested the trucknagigation changes on Android. There is a weird behaviour of the fgets function (documented on stackoverflow.com as well). Using mode a or a+ this returns always NULL. When changing mode to r it works. [15:53:39] So what is working so far: GUI for changing the dimensions, storing the dimensions in a separate file so they get loaded automatically after a restart of the application. [15:54:33] Routing using the given dimensions and conditional weight/speed limits. [15:54:46] The conditional speed warner. [15:56:26] CMake Error at CMakeLists.txt:6 (project): [15:56:27] No CMAKE_CXX_COMPILER could be found. [15:56:34] How can I get the FDROID build to work on CI? This always complains it can't find a suitable compiler. [15:58:42] Who could do a test on WINCE? I don't have a device to test it. [17:57:24] OLFDB about F-Droid, have you tried merging the latest trunk into your branch? [17:58:11] looking at the diff of your PR, I see that in CMakeLists.txt line 6 you reinstated `project(navit C CXX)` [17:58:51] I had to remove that in order for F-Droid CI to succeed, before doing so I got the same error [18:01:21] as far as I can tell, you're not adding any C++ code, so that shouldn't break anything [18:02:16] also, trunk may have had some changes to route.c as a result of traffic bug fixes, the sooner you merge them, the easier [18:03:18] as for WinCE, I have an old device lying around somewhere, if I get to it, I can give it a shot with truck navigation, if you tell me what to test [18:03:39] maybe you can describe the test case(s) somewhere (e.g. in the PR) [18:04:19] anyway, nice work, haven't tried it in practice but looks great so far [18:50:30] Thanks for the hint with the CMakeLists.txt. Now the error is gone but it complains about glib: [18:50:36] /usr/include/glib-2.0/glib/gtypes.h:423:3: error: '_GStaticAssertCompileTimeAssertion_0' declared as an array with a negative size [18:50:36] G_STATIC_ASSERT(sizeof (unsigned long long) == sizeof (guint64)); [18:50:36] ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [18:50:38] /usr/include/glib-2.0/glib/gmacros.h:232:103: note: expanded from macro 'G_STATIC_ASSERT' [18:51:41] I think we came across that as well when adding F-Droid CI [18:52:26] the issue is that the toolchain is trying to link against the host libraries rather than those for the target platform [18:52:43] The config.yml is the same as in trunk for the fdroid part. [18:52:44] there should be something in chat history around January-Febriary [18:53:01] Let me have a look. [18:53:41] it might not just be config.yml but some other files as well [19:07:19] I have another question. metalstrolch halped me to get low emssion zones displayed on the map. They are polygons. I would like to use them for routing based on the vehicle emission class. How can I check if a route segment is inside the polygon? [19:09:29] I guess that's pretty advanced, let me know when you find out :-> [19:09:49] If I display a map point in the GUI it lists the polygon, but I have no clue how to access the data from routing. [19:09:51] you'd probably have to analyze both geometries in detail [19:10:46] as a quick check, you could determine the bounding box of the polygon and discard everything that's outsid it [19:11:14] and do the detailed checks only on what's inside [19:12:49] every spatial DBMS has functions for that, so there should be some information on that out there [19:13:24] maybe one approach is to divide the polygon into a grid pattern, so that every point is located at a grind intersection [19:13:39] then identify rectangles which are entirely inside [19:13:51] and those cut diagonally in half by the outline [19:16:33] but that's just the first thing that came to my mind, others may have come up with better algorithms