boost::split – warning C4996: ‘std::copy::_Unchecked_iterators::_Deprecate’: Call to ‘std::copy’ with parameters that may be unsafe

If you’re using the very handy boost::split() on Visual Studio, then you may run into the following annoying warning:

warning C4996: ‘std::copy::_Unchecked_iterators::_Deprecate’: Call to ‘std::copy’ with parameters that may be unsafe – this call relies on the caller to check that the passed values are correct. To disable this warning, use -D_SCL_SECURE_NO_WARNINGS. See documentation on how to use Visual C++ ‘Checked Iterators’

It’s a big warning that rightly litters the compiler’s output!!

Everything’s OK, it’s just VS whinging about std::copy() being called with pointers. As the warning mentions, you can disable the warnings by defining _SCL_SECURE_NO_WARNINGS, its probably best to define it only for the affected files and not your whole project however!

Another way is to explicitly instruct that warning 4996 be ignored in the code like this:

// Suppress annoying warning for boost:split()
#if defined(_MSC_VER) && _MSC_VER >= 1400 
#pragma warning(push) 
#pragma warning(disable:4996) 


#if defined(_MSC_VER) && _MSC_VER >= 1400 
#pragma warning(pop) 

More Info

Remote Desktop Connection Closes / Disappears silently soon after Connecting

I had a strange problem with my Windows 10 Remote Desktop connections over the last few days, soon after connecting the connection would simply disappear without any message or warning, this would happen typically under a minute after connecting to the remote machine – amusing, but quite frustrating if you wanted to get anything done!

I searched around and couldn’t find any mention of similar things happening to other folk, but there was some mention of cleaning out stale connection data, so I managed to make the problem go away by going to the Remote Desktop connection manager and deleting the ‘Saved Credentials’ from the connection. When I reconnected everything was fine and has remained so ever since!

Connections from multiple Browsers to Mosquitto MQTT via Web Sockets not Working

If you are using the paho javascript MQTT client library to connect to MQTT via web sockets, you might notice that you can’t connect from multiple browser sessions at the same time. This may be because you are using the same client ID for each connection – you should use a different client ID for each connection.

client = new Paho.MQTT.Client('', 9001, 'some-client-id');  // Doesn't work

To get around this you could generate a random client Id like this:

client = new Paho.MQTT.Client('', 9001, Math.random().toString());

This generates a thoroughly meaningless/ugly client Id, but it works…

There is some worry that using an approach like this can lead to a buildup of stale sessions on the server side as the browser will pass a completely different client Id on each connection/refresh but I am not sure, I think this may only affect persistent sessions, will keep an eye on it…

A recipe for building a Web UI using MQTT for your IoT / Embedded App

Cook up a simple & tasty web UI for your embedded project with just a few simple ingredients!  Served on a bed of wholesome Linux goodness on an SBC, (my favourites are: Raspbery Pi or Odroid N2 but substitute alternatives for your own taste) – this is sure to delight family and friends alike!


For the back-end you will need:

Your embedded App  – this is the centrepiece of your creation, it will do some magic and communicate with the UI via MQTT.

python3 – always handy for thickening the sauce!

Flask – tiny & easy to set-up python system for serving your UI’s web pages and static files, perhaps a little lighter on the palette than the ubiquitous Node.js?

Mosquito – for sending MQTT messages between your app and the web UI for updating status and sending commands, MQTT lends a more contemporary taste than the sometimes overused AJAX!

For the front-end you will need:

bootstrap – presentation is soo important!

paho-js – for communicating with MQTT.

HTML & Javascript – the prefect seasoning, no need for some of the heavier spices like Angular.

How To Cook

(details coming soon, hopefully!)

Format std::time_point in ISO 8601 format with fractional seconds / microseconds in C++

A C++ function to format an std::time_point as an ISO 8601 string.

The C++ std chrono stuff is very useful but a bit of a head-wreck!  One of the things I had problems with was how to take an std::time_point value and format it as a string with the fractional seconds / microseconds included.  This sort of time resolution is often required for accurately time stamping machine vision images, especially when acquiring at  high rates from multiple cameras – accurate timestamps allow you to compare images from different cameras that were taken at the ‘same time’.

Anyway if you’re happy to wait for C++20 then you will have access to a format() function; but if you’re more eager to format your time strings now, then here is a function which may fit the bill, it (the function) has to jump through some hoops, but gets there in the end.

Note: This function uses the std::chrono::system_clock but it could be converted (or templated) for other std clocks..

//  Licensed under the Apache License, Version 2.0 (the "License");
//   you may not use this file except in compliance with the License.
//   You may obtain a copy of the License at
//   Unless required by applicable law or agreed to in writing, software
//   distributed under the License is distributed on an "AS IS" BASIS,
//   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
//   See the License for the specific language governing permissions and
//   limitations under the License.

// Format an std::time_point as an ISO 8601 string with fractional seconds to 6
// decimal places, e.g. 2014-08-30T08:18:51.867479
// Warning will not work for any date/times before the start of the UNIX epoch.


inline std::string to_iso_8601(std::chrono::time_point t) {

	// convert to time_t which will represent the number of
	// seconds since the UNIX epoch, UTC 00:00:00 Thursday, 1st. January 1970
	auto epoch_seconds = std::chrono::system_clock::to_time_t(t);

	// Format this as date time to seconds resolution
	// e.g. 2016-08-30T08:18:51
	std::stringstream stream;
	stream << std::put_time(gmtime(&epoch_seconds), "%FT%T");

	// If we now convert back to a time_point we will get the time truncated
	// to whole seconds 
	auto truncated = std::chrono::system_clock::from_time_t(epoch_seconds);

	// Now we subtract this seconds count from the original time to
	// get the number of extra microseconds..
	auto delta_us = std::chrono::duration_cast(t - truncated).count();

	// And append this to the output stream as fractional seconds
	// e.g. 2016-08-30T08:18:51.867479
	stream << "." << std::fixed << std::setw(6) << std::setfill('0') << delta_us;

	return stream.str();

Microsoft Teams stuck in a loop saying it needs sign in after a password change

An amusing MS Teams related problem hit me this morning after I changed my Domain Password – The MS Teams App piped up and said it need Sign In, when I hit OK it seemd to restart and then repeated the message, we were in an infinite loop of fun!

Now as much as I can happily live without Teams bugging me, I unfortunately needed it for a conference call so I had to coax it back to life.  What worked for me in the end was to wait until Teams came up with its message and then right-clicked on the teams icon in the task bar and chose the ‘Sign Out’ option.  I then shutdown and restarted Teams and everything was OK again.


Launching Flask on Port 80 without using sudo

Flask is a great python based HTTP server that’s really small and easy/fast to setup, it is really useful for deploying small Web based User Interfaces for IoT type devices.  By default flask will attach itself to port 5000.  To get my flask script to attach to port 80 I use:

if __name__ == '__main__':'', port=80, debug=True)

This works fine, the only problem is that in order to successfully attach to port 80 the script must be run as root, so instead of just running:


I have to run:

sudo ./

Which isn’t great.

To get around this problem I used the tool: authbind

To Install:

sudo apt-get install authbind

And configure it for access to port 80:

sudo touch /etc/authbind/byport/80

sudo chmod 777 /etc/authbind/byport/807

(Not sure if the very loose 777 permission is required, must experiment)

Now I just have to launch my flask UI script using authbind and it will take care of the script’s permissions to bind to port 80 and I can connect from a browser:

authbind -deep python3 ./

More info on authbind here

Octave – Can’t scroll Window, Workaround

Strange problem with Octave (my version is 4.4.0), in the GUI I can’t scroll the ‘Command Window’, so if some code outputs lots of info I can’t scroll up to see it all, the window keeps jumping to the bottom as I try to scroll!

I am not sure why its happening, but a workaround is to enter the ‘pause’ command within the window, once pause executes I can scroll to my heart’s content.  I  then hit ctrl-c to exit pause.

It is possible that installing a newer version of Octave would fix this but I am not bothered to upgrade at the moment as all of the Maths bits seem to work very well!


Paho Javascript Client – Figure out received message’s MQTT topic

When using the Paho Javascript client from MQTT; when a message arrives via client.onMessageArrived(), how can we figure out the message’s topic.

This threw me for a bit as the documentation for the message object doesn’t mention ‘topic’ at all – but it turns out that the topic name is stored in the message.destinationName field! (panic over)

KDevelop hangs during C++ build on ARM Odroid N2

I have been using Hardkernel’s new Odroid N2 to develop a computer vision system with multiple Basler USB3 cameras. I have been using KDevelop on the Odroid to engineer the C++ code and in general all has been going very well (the Odroid N2 is a fantastic device). I did however hit an occasional problem where KDevelop would cause the Odroid to ‘hang’ during a build – especially when re-building a lot of files.

On investigation it seemed like KDevelop was just using up too much memory and putting the Odroid into a very bad place, sometimes the system would free up after a (long) period of time but most often it wouldn’t, it just ground to a halt thrashing memory (I presume). I am using the boost libraries and I think that a lot of IDEs and build systems have problems with boost as it’s very big! (Clion does especially!)

Anyway the solution was to limit how many parallel build instances (or jobs as it calls them) that KDevelop can run, to do this open the ‘Project / Open Configuration..’ menu and click on the ‘Make’ tab in the left-hand column, now check the ‘Override number of simultaneous jobs’ checkbox, and enter a number in the ‘Number of Simultaneous jobs’ box. A value of 3 works well for me, the builds are a little slower but no longer hang the system! I might try increasing it to see if I can get away with quicker builds…