How to build an openwrt image with gargoyle instead of Luci? - openwrt

First of all, I'm new to this OpenWrt, and all custom build firmwares.
How to compile the OpenWrt source with the Gargoyle or X-wrt UI instead of Luci?
I couldn't find anything in the documentation to show how to do that, and I know it's doable because of this.
In my first compilation I fallowed this steps buildroot.exigence.

Related

Link against a **custom** asan library

What is the right way of using a custom (built by me from llvm sources) asan library?
During compilation I am using -fsanitize=address.
But what should I use for linking? It is recommended everywhere to use -static-libasan or -shared-libasan, but as far as I understand, linker will use the library that comes with compiler (not mine).
Shall I drop -static-libasan / -shared-libasan and just link against the clang_rt.asan-dynamic-x86_64 built by me? Or something else?
if you use -shared-asan at build time it will link to the shared lib which is what you want, and it will default to that standard path, but you should be able to easily override it when you go to run your app to use your own build using LD_PRELOAD.
for example:
LD_PRELOAD=/home/yourname/llvm/lib/clang_rt.asan...so myapp
so...the linker won't load the standard one because yours will be loaded into memory already and myapp will link up to it.
you still might need to make sure the version you are building from source is the same or close enough to the one that it linked to myapp at build time.

Building OpenWrt with ImageBuilder offline

I'm having a Carambola2 developer Board and building a firmware with the ImageBuilder from OpenWrt. This works great. But I would like to build the firmware when I'm offline. Because by default it tries to download data from 8devices.com. When I'm offline it compiles well but it's slow because it tries to access the internet. Any sugestions?
Image builder will need internet, or otherwise you download all packages which you need and save it to "dl" folder or you can build it offline you can try building openwrt using source code. You can check this link
for building image using source code.
The only solution I have found is the following:
Step 1: Remove the rm-command for DL_DIR in the clean target of the Makefile.
Step 2: Add pkg.8devices.com to hosts.deny file.
So you will need an internet connection only the first time and every time you want to add a new package.
I think it's not a clean solution, but it works for me.

How I can run “mapbox-gl-native” project?

I’m using “Mapbox-iOS-SDK” 3.0 via cocoapods and want implement pulsing effect for annotations:
https://github.com/TransitApp/SVPulsingAnnotationView
By this link was suggested update MGLMapView.mm file:
https://stackoverflow.com/questions/33598363/implement-pulse-effect-on-annotations-in-mapbox/33768054#33768054
But for now I don’t understand how I can run “mapbox-gl-native” project?
Here are the instructions for setting up a Mapbox iOS SDK build environment. Essentially, you clone the repo and run make iproj. Once you're ready to integrate the library into your own app, you'd run make ipackage and copy the resulting binary/etc in build/ios/dynamic/ to your project.
Beware: the mapbox-gl-native project is very complicated to build, the structure still in flux, and heavy development happening every day, so any instructions that I write here are likely to not be entirely useful in the mid-future.
Adding native views to a map view is not easily accomplished (yet), but you would start here and modify/subclass/imitate MGLUserLocationAnnotationView.

Configure OpenCV on Codeblocks

I would like to use OpenCV with CodeBlocks. But I don't know to configure CodeBlocks about "Project build options". I need to precise where are the headers, and functions of openCV.
Firstly I've tried to follow this tutorial : http://kevinhughes.ca/tutorials/opencv-install-on-windows-with-codeblocks-and-mingw/ But I had some problems in the step 4 with the configuration of makefiles. In fact, I am not very sure that I need to use cmake in order to read opencv functions.
Could you help me to configure build options ? I think I need to add "C:\opencv\build\include" about compiler on Search Directories, and about linker, what should I add ?
How add the librairy on CodeBlocks ?
Thank you
In fact CMakeLists.txt is present here : ..\opencv\sources\CMakeLists.txt
I thought CMake would have searched. Then, I use correctly Cmake and he has builded OpenCV. I don't use cygwin, I use the shell of windows. So, after that, I've tried to build the exemple of opencv code in this tutorial, and it works !
Now, when I try to build a project already coded, I don't succeed. In fact, it's a code made by an other person, it use OpenCV and C++. And there is also Java code for Android. But I would like to build and execute only C++ code.
I chose the same settings for OpenCV, but how could I know what I should add in "compiler" and "linker" section ?
Thank you for your help.
Try to change your source directory to C:\opencv\sources and see if that works.

InvientCharts Vaadin not working

I am working with Vaadin 6 and want to use InvientCharts because it is free.
After compiling the widgetset, all vaadins component instead of invientCharts are working. The error which appears as text on my app is the following:
After trying everything, including setting up a complete new project, I followed this tutorial: http://www.nightswatch.de/?x=entry:entry120125-234412
But it is still not working and shows the same error.
Has anyone an idea what is going on here?
Thanks in advance!
When you receive the message that the widgetset does not contain implementation of XY,
then usually you have forgotten to compile it with the additional jar files or you did not specify the new widgetset.
The nightswatch example is incomplete, you need to compile the widgetset.
Simplest would be to use the maven build script to build all dependencies and compile stuff as needed (or use eclipse for this)

Resources