Nice idea for my git log visualiser

I could use this excellent idea in my Git log visualiser. Never thought I could do this! ?

Currently I am parsing plain text output from git log –numstat command, looking at certain order of lines and line beginnings to determine what is on the line.

Exporting from git log to JSON using jq like Simon does in his post would make parsing extremely simple using Swift Codable.

Posting this here to remind myself.

Summertime blues

Last night we turned the time towards summer again.

So if we program all the clocks to go too fast (~20 secs a day) for half of the year and, correspondingly, go too slow (again ~20 secs a day) for the other half of the year, we would not need to do this clock manipulation thing twice a year at all.

Sadness

Update March 26th: they seem to be up and running again!

I’ve been using weather.willab.fi weather service located here in Oulu, at the VTT research center, for years in many of my courses.

It has been a good demo for teaching many basic computing and networking topics, for example:

  • how to use curl to execute network requests and look at the different HTTP headers the server provides in the response;
  • how HTTP works on top of lower level protocols (TCP), using Wireshark for traffic analysis;
  • localization of simple Android and iOS apps;
  • what is JSON and XML and how they look like with simple weather data;
  • what is a server and what is a client on the Internet;
  • and maybe something else I do not remember now.

But anyways, the service has been down for some time, returning weather data from around December 2022. Now it is even worse; it just displays the nginx server welcome page, and HTTP 404 when you try to get the current weather contents using http://weather.willab.fi/weather.json.

Hopefully they will kick the service up and running soon! Otherwise I need to rewrite many exercises and demo apps to use some other simple service.

It’s both

I stumbled to a YouTube video bashing those “quote unquote Clean Code” people for creating badly time performing (fast) code. While he was going through Clean Code principles, he showed, one by one, how not using them produced more performant code.

Every thing he did to make code more time performant – I agree, he is right.

Otherwise, he either is deeply ignorant or was deliberately trolling in a very nasty way. That is why I will not share the link to the video. He had even disabled commenting on the video – a sign he is not after genuine discussion on the topic.

Of course there are conflicting requirements in software development. To satisfy some (many) of those requirements, principles of Clean Code are very useful.

For some other requirements, not using Clean Code principles in some areas of the code is a good thing, to achieve those requirements. Like time performance.

Many times you have to compromise and then pay the price for that compromise. This should be obvious also to the people who made that video.