A personal project I’ve wanted to build for many years now is a system to get people thinking about the value of a proposed spending, by showing examples of what else they could get for that same amount of money.

Originally I wanted to be able to enter what you wanted to buy and the price and then show alternatives for less and the amounts you could get for that same cost. But figured that it would be a whole task in itself to gather the data so I put the project on hold. For years….

Eventually I realised that I could invert it and instead ask for the amount you are thinking of spending on that particular category and then show a simple list of pre-selected items. This way I can just make up examples of purchases based on real products or services and include them in the tool completely statically. Or host them in a web API or something.

So now I had a feasible idea finally, and an existing React project I could put this into. Following are a few iterations of the design.

First iteration

I started with generic minimum and maximum costs for each item/service with an idea of the amount left over for snacks or the appropriate “side purchase”. I included a scenario where you could take 20% out and keep extra for side purchases. I realised that this is kind of boring and hard to follow.

Iteration 2: Tabular design

So I moved the 20% scenario into a column, adding emojis and alternating row background colours. See how it’s just more visually interesting and slightly easier to follow. But it’s not very useful actually.

Iteration 3: itemised

I realised that if I explicitly included examples for each kind of cost, like “adult ticket”, “child ticket”, “Netflix with ads”, “Netflix Standard”, “4k Netflix”, it’s clearer, more useful, and actually easier to code because the grammar is easier to write inside the string interpolation.

Iteration 4: legibility and styling

By now I’ve just made some wording and styling improvements. I also corrected the CSS imports in another area of the app so now I have to fix the form layout here but I’m having issues with the imports so I’ve put the project away for this coding session!

A week or so later, I finally fixed the form layout a bit. It’s not visible but my next improvement has been to make the categories and data dynamic so I only need to update some JSON files to make a new category. Since it’s the same React component now the form keeps your entered cost amount.

If you would you like play with the current product, it’s available at Money-Choices.info.

I’ve found that when I update my website, it’s tedious to manually copy the files up to my CPanel hosting platform. And my website source code is already on Github version control, so I wanted to be able to automatically update the static files when I push my HTML up to Github.

The only way to achieve this is to utilise Git’s post-receive hook (and Github’s webhooks) and trigger some code on my site which could a perform git pull.

Git hooks are actions that can run your own scripts when events happen to your repository – commit or pull for example – then do whatever task is handy for you. Webhooks are HTTP payloads sent from one server to another, when some event occurs, to trigger off another action in turn on the receiving server.

Some research led me to oxguy3’s deploy.php Github Gist which receives Github’s post-receive webhook and initiates a git pull on on the web host:

Oxguy3’s deploy.php script.

Set up a webhook on your repository in the settings of the Github repository and give it a password. Really I just followed oxguy3’s directions, it’s well documented.

Their PHP script is quite good, it even uses Github’s authorisation headers to prevent randoms from triggering the deploy script.

But oxguy3’s script had to ssh to their server (inside the script) since their host had blocked PHP shell_exec. But my web host hasn’t blocked shell_exec so I could take out those pieces of code. So my forked gist becomes:

My deploy.php script

The script also emails me when it’s complete and the output. If you started from an empty public_html folder you’d be right to run the script now. (I wasn’t though, read on if interested.)

Continue reading →

Finishing semester 2 2018 and passing both COMS7003 – Information Security and COMP7506 – Algorithms and Data Structures this semester now gives me a Graduate Diploma in Information Technology

So this whole degree I have done 9 subjects:

1. CSSE7030 – Introduction to Software Engineering
2. INFS7900 – Introduction to Information Systems
3. INFS7903 – Relational Database Systems
4. INFS7204 – Services Oriented Architecture
5. CSSE7202 – Advanced Software Engineering
6. INFS7202 – Web Information Systems
7. COMS7201 – Computer Networks 1
8. COMS7003 – Information Security
9. COMP7505 – Algorithms and Data Structures

I’ve been wondering this semester which subjects have I found the most useful or enjoyable or interesting. It’s really hard to say if I’m going to be objective about it. The first 2 certainly gave me the biggest jump in knowledge – from almost nothing to actually being a programmer and building databases!

INFS7204, COMS7201, and CSSE7202 probably gave me the next biggest jump in knowledge and practical skills. While the SOA side of INFS7204 may not be so relevant these days, the overall idea of spreading functionality over simple services is very much relevant still, and the experience with C# was good. CSSE7202 taught some programming concepts which explained some of the workings we experienced in C#. Whilst I may not need the in-depth understanding of TCP, the overall networking experience and theory from COMS7201 was quite interesting and I think it’s quite useful in this day and age. That subject gave me experience in network programming in Python at the byte and binary level, as well as my first introduction to base 2 and hexadecimal notation.

Algorithms and Data Structures is a hard one to discuss – while I hated the course, I learnt a lot of things, got more experience in object oriented programming and recursion, and well graduated earlier by doing 2 subjects this semester.

Information Security was quite an interesting course and well taught, I think it covered a range of topics that are useful in the modern IT field. The assignments were satisfying at least.

Web Information Systems and Relational Database Systems were probably the least useful subjects. Unfortunately I probably could have learnt much the same content in WIS for free online – though it was a good chance to build a real web application from scratch. Relational Database Systems was interesting, but honestly I find it hard to relate the knowledge back to writing better SQL – if I was going to work for Oracle or work on PostgreSQL maybe it would be useful. But still learning is and these 2 were part of the overall degree anyway.

Trying to remain objective, I think my Grad Dip IT was a good choice and has been good for me overall. I think a shorter course like a Graduate Certificate or Graduate Diploma is a good way to break into IT. Maybe you could to a bootcamp and combine with some university study for theory and background too. I won’t get into that further here though.

Emotionally, I suppose these 3 years of part-time study has been good for my resilience with some rather stressful (for me at least) times, which I won’t get into. Suffice to say, what “they” say about pushing yourself is true, you do grow when you move out of your comfort zone.

At work I have an ASP.net Core 2.1 web app I’m developing. At work we use Windows and Visual Studio 2017 to develop on with a local MS SQL Server as the database backend. But, I want to be able to work on the project away from the office and I only have a Mac laptop. So I’m trying to get it running on Mac, which should be possible since it’s .Net Core after all. There were a few things I needed to change, which turned out to be one line!

Continue reading →

In my earlier post, I was installing Django on Mac OS X by following their own tutorial partly to get a feel for the database site of the MVC framework. This was to evaluate whether I should use an off-the-shelf framework to just build my own. My requirements are:

• Easy to code.
• Preferably easy to manage the database and existing data.
• Preferably users, security etc for free.
• Reusable View components for menus etc.
• Quality framework-created database table structure.

I don’t want to write forms to manage the data or all the user and authentication and session stuff. I can, but I don’t want to.

After I followed the tutorial and inspected the database generated I could critically evaluate the framework.

Django’s model-generated-data

• Logically structured OK
• Table names are the same as the entity in the model, but by default prepended with the Django app name and lower-cased. (You can change the table name in the model.)
• Foreign-keys are generated and handled sensibly.
• There is also a Shell – which gives you programmatic access to the data model. It’s basically just a standard Python REPL with some extra enhancements for Django stuff.

Django’s Admin Site

• supports multiple users
• Seems to keep a history of the changes (the meta-data not the actual changes) as made in the Admin site.
• Intelligently builds the forms based on the data type in the model – dates have date pickers, strings have text inputs etc.

Views/URL Mappings

• Can be as simple as a return statement or return a render() method for an HTML template or any other Python library as long as it can fit into the HTTP response – JSON/PDF/HTML/media/whatever… From the website:

“Your view can read records from a database, or not. It can use a template system such as Django’s – or a third-party Python template system – or not. It can generate a PDF file, output XML, create a ZIP file on the fly, anything you want, using whatever Python libraries you want.

All Django wants is that HttpResponse. Or an exception.”

• URLs and parameters are mapped in a separate file which ‘captures’ parts of the GET request and converts parts of it to Python variables to pass along to the appropriate view method. It seems like you can have heaps of view methods all contained in different Python files, it’s just standard Python syntax to call things and pass data around. Possibly similar to other frameworks like ASP.net which also has URL mappings.

Things I’m not sure about yet

• Can you password protect an entire site? Like for an internal app?
• I read somewhere about reverse engineering an existing database or even generating an ER diagram based on Django’s model. These sound pretty cool if possible.

Conclusion

• The Django Project’s tutorial is one of the best getting-started tutorials I’ve used, props to them.  They must have tested it well, on various audiences. Only thing is some parts seem like suggestions, but it turns out later steps refer to code given so I had to go back. But if you did everything shown, you wouldn’t have this problem.
• Programmatic testing is built right into Django, I’m surprised that you can even test Views. Not necessarily the HTML output, but at least whether or not a view gives you a particular HTTP response code – they give the example of permissions or hidding polls without questions.

I’ll finish the testing tutorial and static files tutorial later, but so far Django seems a pretty capable MVC framework and I’ll probably use it for at least some of the app ideas I have.

I’m writing this partially for myself so I have a reference for what’s wrong while trying to install Django in a Python virtual environment on OS X. I’m just going to write about errors I encountered I think.

Create the virtualenv

In the directory where I wanted to put the virtual environment type:

virtualenv django-ve

where django-ve will be the virtual environment. It should exist first.

Activate it:

cd django-ve
source bin/activate

Install Django

Following Django’s own guide still use pip to install it:

pip install Django

But when I did this I got an error with the SSL connection to PyPy exactly as in, this Stack Overflow post. The solution was as they describe in the accepted answer though you also have to update pip both inside the new virtual environment and outside. (I first tried updating it, then recreating the virtual environment, but this didn’t help, which kind of makes sense if it fetches a new pip each time.)

Install the database drivers

Since I want to use a full MySQL database I needed to install the Python connectors. I stuck with the Django recommended way. I tried to follow their instructions.

When I tried to install mysqlclient I got this same error. Updating Wheels as per one of the suggestions didn’t help, but running

xcode-select --install

did fix the problem by installing the Apple provided developer command line tools which install a C compiler.

Verify?

I verified that the Python virtualenv could see Django as instructed and it could.

At this point, the setup guide ended so I’m just proceeding along with the tutorial.

Tutorial

I continued through the tutorial pretty well OK up until the part about configuring timezones. Silly me, I wrongly assumed I could modify it from ‘UTC’ to ‘AEST’, but even as is linked in their own guide, it’s based on the TZ Database so I should use ‘Australia/Queensland’

Then proceeding along and running

python manage.py migrate

runs successfully and going back to MySQL Workbench I can see all the tables that Django created. Looking good so far.

By this stage I’ve set it up and worked connect it to a real database, so that’s enough for now. I may come back later and critically evaluate Django for my purposes.

The end of this semester I finished COMS7201 – Computer Networks 1.  For some reason, it was one of the most satisfying courses I’ve taken.

I’ve written about all about it right here – Computer Networks 1, describing the assessment and several of the main concepts taught. Since writing that page, I still haven’t gone back and fixed the issues in the assignments, particularly Assignment 3. I want to implement the 1s Complement checksum and totally native IP header generation.  I’ve been doing other things with my spare time instead, like adding a feature to my Chat Log Parser/Viewer to group chats by day, writing for Weekend Notes on Mt Gambier plus other life skills updating and of course general life.

One approach I’ve started a few weeks ago is to create a daily goal to practice coding (and I’ve added writing) every day, but only for 15 minutes each day. This forces me to break up the task into discrete units of work – one (or two) unit test/s, one (or 2) method/s or classes, one paragraph or just the outline. I’ve found this method to work well so far, I don’t get put off with a long session when I want to relax but I still make progress. If I find I’m going really well and get into the groove (Flow), I can certainly keep going, but if not I can stop.

The above approach doesn’t work so well with university study because the assignments have due dates. This is good for personal projects and hobbies though.