GCP Cloud Architect Exam Experience

Last October it was announced that Desert Oasis Healthcare (the company I work for) signed on to pilot Google's Care Studio. DOHC is the first ambulatory clinic to sign on.

I had been in some of the discovery meetings before the announcement and was really excited about the opportunity. So far our use of any Cloud platforms at work has been extremely limited (that is to say, we don't use ANY of the big three cloud solutions for our tech) so this seemed to provide a really good opportunity.

As we worked through the project scoping there were conversations about the handoff to DOHC and it occurred to me that I didn't have any knowledge of what GCP offered, what any of it did, or how any of it could work.

I've had on my 'To Do' list to learn one of the Big Three Cloud services (AWS, Azure, or GCP) but because we didn't use ANY of them at work I was (a) worried about picking the 'wrong' one and (b) worried that even if I picked one I'd NEVER be able to use it!

The partnership with Google changed that. Suddenly which cloud service to learn was apparent AND I'd be able to use whatever I learned for work!

Great, now I know which cloud service to start to learn about ... the next question is, "What do I try to learn?". In speaking with some of the folks at Google they recommended one of three Certification options:

  1. Digital Cloud Leader
  2. Cloud Engineer
  3. Cloud Architect

After reviewing each of them and having a good idea of what I need to know for work, I opted for the Cloud Architect path.

Knowing which certification I was going to work towards, I started to see what learning options were available for me. It just so happens that Coursera partnered with the California State Library to offer free training which is great because Coursera has a learning path for the Cloud Architect Exam! So I signed up for the first course of that path right before Thanksgiving and started to work my way through the courses.

I spent most of the holidays working through these courses, going pretty fast through them. The labs offered up are so helpful. They actually allow you to work with GCP for FREE during your labs which is amazing.

After I made my way through the Coursera learning Path I bought the book Google Cloud Certified Professional Cloud Architect Study Guide which was really helpful. It came with 100 electronic flash cards and 2 practice exams, and each chapter had questions at the end.

I will say that the practice exams and chapter questions from the book weren't really like the ACTUAL exam questions BUT it did help me in my learning, especially regarding the case studies used in the exams.

I read through the book several times, and used the practice questions in the chapters to drive what parts of the documentation I'd read to shore up my understand of the topics.

Finally, after about 3 months of pretty constant studying I took the test. I opted for the remote proctoring option and I'd say that I really liked this option. I was able to take the test in the same place I had done most of my studying. I did have to remove essentially EVERYTHING from my home office, but not having to drive anywhere, and not having to worry about unfamiliar surroundings really helped me out (I think).

I had 2 hours in which to answer 60 questions. My general strategy for taking tests is to go through the test, mark questions that I'm unsure of and eliminate answers that I know to not be true on those questions. Once I've gone through the test I revisit all of the unsure questions and work through those.

My final pass is to go through ALL of the questions and make sure I didn't do something silly.

Using this strategy I used 1 hour and 50 minutes of the 2 hours ... and I passed!

The unfortunate part of the test is that you only get a Pass or Fail so you don't have any opportunity to know what parts of the exam you missed. Now, if you fail this could be a huge help in working to pass it next time, but even if you pass it I think it would be helpful to know what areas you might struggle in.

All in all this was a pretty great experience and it's already helping with the GCP implementation at work. I'm able to ask better questions because I'm at least aware of the various services and what they do.

Contributing to Django or how I learned to stop worrying and just try to fix an ORM Bug

I went to DjangoCon US a few weeks ago and hung around for the sprints. I was particularly interested in working on open tickets related to the ORM. It so happened that Simon Charette was at Django Con and was able to meet with several of us to talk through the inner working of the ORM.

With Simon helping to guide us, I took a stab at an open ticket and settled on 10070. After reviewing it on my own, and then with Simon, it looked like it wasn't really a bug anymore, and so we agreed that I could mark it as done.

Kind of anticlimactic given what I was hoping to achieve, but a closed ticket is a closed ticket! And so I tweeted out my accomplishment for all the world to see.

A few weeks later though, a comment was added that it actually was still a bug and it was reopened.

I was disappointed ... but I now had a chance to actually fix a real bug! I started in earnest.

A suggestion / pattern for working through learning new things that Simon Willison had mentioned was having a public-notes repo on GitHub. He's had some great stuff that he's worked through that you can see here.

Using this as a starting point, I decided to walk through what I learned while working on this open ticket.

Over the course of 10 days I had a 38 comment 'conversation with myself' and it was super helpful!

A couple of key takeaways from working on this issue:

  • Carlton Gibson said essentially once you start working a ticket from Trac, you are the world's foremost export on that ticket ... and he's right!
  • ... But, you're not working the ticket alone! During the course of my work on the issue I had help from Simon Charette, Mariusz Felisiak, Nick Pope, and Shai Berger
  • The ORM can seem big and scary ... but remember, it's just Python

I think that each of these lesson learned is important for anyone thinking of contributing to Django (or other open source projects).

That being said, the last point is one that I think can't be emphasized enough.

The ORM has a reputation for being this big black box that only 'really smart people' can understand and contribute to. But, it really is just Python.

If you're using Django, you know (more likely than not) a little bit of Python. Also, if you're using Django, and have written any models, you have a conceptual understanding of what SQL is trying to do (well enough I would argue) that you can get in there AND make sense of what is happening.

And if you know a little bit of Python a great way to learn more is to get into a project like Django and try to fix a bug.

My initial solution isn't the final one that got merged ... it was a collaboration with 4 people, 2 of whom I've never met in real life, and the other 2 I only just met at DjangoCon US a few weeks before.

While working through this I learned just as much from the feedback on my code as I did from trying to solve the problem with my own code.

All of this is to say, contributing to open source can be hard, it can be scary, but honestly, I can't think of a better place to start than Django, and there are lots of places to start.

And for those of you feeling a bit adventurous, there are plenty of ORM tickets just waiting for you to try and fix them!

Upgrading to PostgreSQL 14

Django 4.1 was released on August 3, 2022 and I was excited to upgrade to it. I did the testing locally and then pushed my changes up to GitHub to deploy. The deployment was succesful, but when I went to visit my sites ... womp womp. I got a Server Error 5XX.

What happened? Well, it turns out that Django 4.1 dropped support for Postgres 10 and that just so happens to be the version I was running on my production server (but not on my local dev machine ... I was running Postgres 14).

OK, so I am going to need to upgrade in order to get the features of anything above Django 4.0 ... and honestly, I've needed to upgrade past Postgres 10 for a while.

I found this StackOverflow question and answer and it helped me a ton! It was to upgrade from Psotgres 10 to 12, but the ideas were the same (but replace 12 with 14). There is also a step that indicates you need to run ./analyze_new_cluster.sh but that seems to be only for version 12(maybe 13) and lower.

Everything was fine until I visited my site and got a Server Error 5XX AGAIN!

What gives?

My first assumption was that maybe the postgres server didn't start back up properly after the upgrade. I checked the service to verify that it was running, and it was

ps -aux | grep postgres

which returned

postgres   988  0.0  1.3 321668 27588 ?        Ss   16:55   0:01 /usr/lib/postgresql/14/bin/postgres -D /var/lib/postgresql/14/main -c config_file=/etc/postgresql/14/main/postgresql.conf
postgres  1034  0.0  0.2 321788  6112 ?        Ss   16:55   0:00 postgres: 14/main: checkpointer
postgres  1035  0.0  0.2 321800  5996 ?        Ss   16:55   0:00 postgres: 14/main: background writer
postgres  1036  0.0  0.4 321668  9388 ?        Ss   16:55   0:00 postgres: 14/main: walwriter
postgres  1039  0.0  0.3 322356  8080 ?        Ss   16:55   0:00 postgres: 14/main: autovacuum launcher
postgres  1040  0.0  0.2 176828  5108 ?        Ss   16:55   0:00 postgres: 14/main: stats collector
postgres  1041  0.0  0.3 322224  6628 ?        Ss   16:55   0:00 postgres: 14/main: logical replication launcher
root      4868  0.0  0.0  14860  1072 pts/0    S+   18:47   0:00 grep --color=auto postgres

I also checked

systemctl status postgresql

which returned as expected

 postgresql.service - PostgreSQL RDBMS
   Loaded: loaded (/lib/systemd/system/postgresql.service; enabled; vendor preset: enabled)
   Active: active (exited) since Sun 2022-08-28 16:55:32 UTC; 1h 54min ago
  Process: 1169 ExecStart=/bin/true (code=exited, status=0/SUCCESS)
 Main PID: 1169 (code=exited, status=0/SUCCESS)

Aug 28 16:55:32 server-name systemd[1]: Starting PostgreSQL RDBMS...
Aug 28 16:55:32 server-name systemd[1]: Started PostgreSQL RDBMS.

One last thing to try

python manage.py makemigrations

This gave me a hint as to what the issue was:

RuntimeWarning: Got an error checking a consistent migration history performed for database connection 'default': connection to server at "", port 5432 failed: FATAL:  password authentication failed for user "user" connection to server at "", port 5432 failed: FATAL:

Hmmm ... a quick google search doesn't specifically answer it, but it helps me to get the to answer.

The 'user' isn't able to connect to the database. Maybe the upgrade process resets the password of users in the database or it just doesn't keep the users.

A quick look at the users on the database showed me that the users were still there, so the only thing left to do at this point was to set the user passwords to be what my settings are expecting.

To do that I ran


I did this for the databases that were associated with my websites that were returning 5XX errors and voila! That fixed the issue.

I'm sure that there is a way to keep the passwords for the users after the upgrade, but I haven't been able to find it.

The next time I need to upgrade PostgreSQL I am going to refer back to this post to remind myself what I did last time 😀

A Goodbye to Vin

One of the earliest memories of my grandmother is visiting her in 29 Palms 1 2 in her permanent mobile home. I remember sitting on the davenport watching the Dodgers on a small 13" COLOR CRT TV. I remember that the game was broadcast on KTLA5. But what I remember the most is the voice of Vin Scully.

I don't know what who the Dodgers were playing, but I remember how much my grandmother LOVED to listen to Vin call the game. And it stuck with me. I was probably about 7 or 8 and I thought baseball was "boring". To be fair, I thought most sports were boring, but especially baseball. Nothing ever happens! But, I loved my grandmother, and I loved hanging out with her 3 and so I watched the game with her.

Years later I discovered that yes, I did like baseball, and no, it was not boring. And since my grandmother was a Dodgers fan, then I would be too. It was something that connected us. it didn't matter where I lived, or how old I was, we both loved baseball. We both loved the Dodgers. We both loved to hear Vin call the game.

My grandmother died in 2007, but something that helped to connect me to her in the years since was watching the Dodgers. Listening to Vin.

As Vin got older, he still called the home games, but he handed most of the road games to a new crew. I still loved to Watch Dodgers games, but I loved watching the games he called a little bit more. At the start of each season I always kind of wondered, "is this the last year for Vin?". And in 2016 the answer was yes.

I still remember the last game he called in Dodgers Stadium. I remember the back and forth. I remember the Rockies going up 1 run in the top of the 9th. And the Dodgers tying it back up in the bottom of the 9th. And I remember when Charlie Culberson hit the game winning home run in the bottom of the 10th.

I remember the last game Vin called in San Francisco. I remember the Dodgers lost ... but it was Vin's last game, so I still loved getting the chance to watch it. And to listen to him call the game.

Vin passed at the age of 94 on Aug 2, 2022. Just as I knew that there would be a day when Vin retired from calling games, I knew there would be a day when he wouldn't be with us anymore.

I've been trying process this and figure out why this is hitting me as hard as it is.

It all comes back to my grandmother. They never met each other (at least I don't think they did), but in my head they were inextricably connected. Vin was a connection to my grandmother that I didn't fully realize I had, and with his passing that connection isn't there anymore. He hasn't called a game in more than 5 years, but still, knowing that he NEVER will again is hitting a bit hard for me. And I think it's because it reminds me that my grandma isn't here to watch the games with me anymore, and that bums me out. She was a cool lady who always loved the Dodgers ... and Vin.


  1. Yes that 29 Palms, right next to the LARGEST Marine Corp Base in the WORLD
  2. also the 29 Palms that is right next to Joshua Tree home to the National Park that is the current catnip of Hipsters
  3. she always had the butter scotch hard candies that were my favorite

Django and Legacy Databases

I work at a place that is heavily investing in the Microsoft Tech Stack. Windows Servers, c#.Net, Angular, VB.net, Windows Work Stations, Microsoft SQL Server ... etc

When not at work, I really like working with Python and Django. I've never really thought I'd be able to combine the two until I discovered the package mssql-django which was released Feb 18, 2021 in alpha and as a full-fledged version 1 in late July of that same year.

Ever since then I've been trying to figure out how to incorporate Django into my work life.

I'm going to use this series as an outline of how I'm working through the process of getting Django to be useful at work. The issues I run into, and the solutions I'm (hopefully) able to achieve.

I'm also going to use this as a more in depth analysis of an accompanying talk I'm hoping to give at Django Con 2022 later this year.

I'm going to break this down into a several part series that will roughly align with the talk I'm hoping to give. The parts will be:

  1. Introduction/Background
  2. Overview of the Project
  3. Wiring up the Project Models
  4. Database Routers
  5. Django Admin Customization
  6. Admin Documentation
  7. Review & Resources

My intention is to publish one part every week or so. Sometimes the posts will come fast, and other times not. This will mostly be due to how well I'm doing with writing up my findings and/or getting screenshots that will work.

The tool set I'll be using is:

  • docker
  • docker-compose
  • Django
  • MS SQL
  • SQLite

Inserting a URL in Markdown in VS Code

Since I switched my blog to pelican last summer I've been using VS Code as my writing app. And it's really good for writing, note just code but prose as well.

The one problem I've had is there's no keyboard shortcut for links when writing in markdown ... at least not a default / native keyboard shortcut.

In other (macOS) writing apps you just select the text and press ⌘+k and boop! There's a markdown link set up for you. But not so much in VS Code.

I finally got to the point where that was one thing that may have been keeping me from writing because of how much 'friction' it caused!

So, I decided to figure out how to fix that.

I did have to do a bit of googling and eventually found this StackOverflow answer

Essentially the answer is

  1. Open the Preferences Page: ⌘+Shift+P
  2. Select Preferences: Open Keyboard Shortcuts (JSON)
  3. Update the keybindings.json file to include a new key

The new key looks like this:

    "key": "cmd+k",
    "command": "editor.action.insertSnippet",
    "args": {
        "snippet": "[${TM_SELECTED_TEXT}]($0)"
    "when": "editorHasSelection && editorLangId == markdown "

Honestly, it's little things like this that can make life so much easier and more fun. Now I just need to remember to do this on my work computer 😀

Logging Part 2

In my previous post I wrote about inline logging, that is, using logging in the code without a configuration file of some kind.

In this post I'm going to go over setting up a configuration file to support the various different needs you may have for logging.

Previously I mentioned this scenario:

Perhaps the DevOps team wants robust logging messages on anything ERROR and above, but the application team wants to have INFO and above in a rotating file name schema, while the QA team needs to have the DEBUG and up output to standard out.

Before we get into how we may implement something like what's above, let's review the parts of the Logger which are:


In a logging configuration file you can have multiple formatters specified. The above example doesn't state WHAT each team need, so let's define it here:

  • DevOps: They need to know when the error occurred, what the level was, and what module the error came from
  • Application Team: They need to know when the error occurred, the level, what module and line
  • The QA Team: They need to know when the error occurred, the level, what module and line, and they need a stack trace

For the Devops Team we can define a formatter as such1:

'%(asctime)s - %(levelname)s - %(module)s'

The Application team would have a formatter like this:

'%(asctime)s - %(levelname)s - %(module)s - %(lineno)s'

while the QA team would have one like this:

'%(asctime)s - %(levelname)s - %(module)s - %(lineno)s'


The Handler controls where the data from the log is going to be sent. There are several kinds of handlers, but based on our requirements above, we'll only be looking at three of them (see the documentation for more types of handlers)

From the example above we know that the DevOps team wants to save the output to a file, while the Application Team wants to have the log data saved in a way that allows the log files to not get too big. Finally, we know that the QA team wants the output to go directly to stdout

We can handle all of these requirements via the handlers. In this case, we'd use

Configuration File

Above we defined the formatter and handler. Now we start to put them together. The basic format of a logging configuration has 3 parts (as described above). The example I use below is YAML, but a dictionary or a conf file would also work.

Below we see five keys in our YAML file:

version: 1

The version key is to allow for future versions in case any are introduced. As of this writing, there is only 1 version ... and it's version: 1


We defined the formatters above so let's add them here and give them names that map to the teams

version: 1
    format: '%(asctime)s - %(levelname)s - %(module)s'
    format: '%(asctime)s - %(levelname)s - %(module)s - %(lineno)s'
    format: '%(asctime)s - %(levelname)s - %(module)s - %(lineno)s'

Right off the bat we can see that the formatters for application and qa are the same, so we can either keep them separate to help allow for easier updates in the future (and to be more explicit) OR we can merge them into a single formatter to adhere to DRY principals.

I'm choosing to go with option 1 and keep them separate.


Next, we add our handlers. Again, we give them names to map to the team. There are several keys for the handlers that are specific to the type of handler that is used. For each handler we set a level (which will map to the level from the specs above).

Additionally, each handler has keys associated based on the type of handler selected. For example, logging.FileHandler needs to have the filename specified, while logging.StreamHandler needs to specify where to output to.

When using logging.handlers.RotatingFileHandler we have to specify a few more items in addition to a filename so the logger knows how and when to rotate the log writing.

version: 1
    format: '%(asctime)s - %(levelname)s - %(module)s'
    format: '%(asctime)s - %(levelname)s - %(module)s - %(lineno)s'
    format: '%(asctime)s - %(levelname)s - %(module)s - %(lineno)s'
    class: logging.FileHandler
    level: ERROR
    filename: 'devops.log'
    class: logging.handlers.RotatingFileHandler
    level: INFO
    filename: 'application.log'
    mode: 'a'
    maxBytes: 10000
    backupCount: 3
    class: logging.StreamHandler
    level: DEBUG
    stream: ext://sys.stdout

What the setup above does for the devops handler is to output the log data to a file called devops.log, while the application handler outputs to a rotating set of files called application.log. For the application.log it will hold a maximum of 10,000 bytes. Once the file is 'full' it will create a new file called application.log.1, copy the contents of application.log and then clear out the contents of application.log to start over. It will do this 3 times, giving the application team the following files:

  • application.log
  • application.log.1
  • application.log.2

Finally, the handler for QA will output directly to stdout.


Now we can take all of the work we did above to create the formatters and handlers and use them in the loggers!

Below we see how the loggers are set up in configuration file. It seems a bit redundant because I've named my formatters, handlers, and loggers all matching terms, but 🤷‍♂️

The only new thing we see in the configuration below is the new propagate: no for each of the loggers. If there were parent loggers (we don't have any) then this would prevent the logging information from being sent 'up' the chain to parent loggers.

The documentation has a good diagram showing the workflow for how the propagate works.

Below we can see what the final, fully formed logging configuration looks like.

version: 1
    format: '%(asctime)s - %(levelname)s - %(module)s'
    format: '%(asctime)s - %(levelname)s - %(module)s - %(lineno)s'
    format: '%(asctime)s - %(levelname)s - %(module)s - %(lineno)s'
    class: logging.FileHandler
    level: ERROR
    filename: 'devops.log'
    class: logging.handlers.RotatingFileHandler
    level: INFO
    filename: 'application.log'
    mode: 'a'
    maxBytes: 10000
    backupCount: 3
    class: logging.StreamHandler
    level: DEBUG
    stream: ext://sys.stdout
    level: ERROR
    formatter: devops
    handlers: [devops]
    propagate: no
    level: INFO
    formatter: application
    handlers: [application]
    propagate: no
    level: DEBUG
    formatter: qa
    handlers: [qa]
    propagate: no
  level: ERROR
  handlers: [devops, application, qa]

In my next post I'll write about how to use the above configuration file to allow the various teams to get the log output they need.

  1. full documentation on what is available for the formatters can be found here: https://docs.python.org/3/library/logging.html#logrecord-attributes

Logging Part 1


Last year I worked on an update to the package tryceratops with Gui Latrova to include a verbose flag for logging.

Honestly, Gui was a huge help and I wrote about my experience here but I didn't really understand why what I did worked.

Recently I decided that I wanted to better understand logging so I dove into some posts from Gui, and sat down and read the documentation on the logging from the standard library.

My goal with this was to (1) be able to use logging in my projects, and (2) write something that may be able to help others.

Full disclosure, Gui has a really good article explaining logging and I think everyone should read it. My notes below are a synthesis of his article, my understanding of the documentation from the standard library, and the Python HowTo written in a way to answer the Five W questions I was taught in grade school.

The Five W's

Who are the generated logs for?

Anyone trying to troubleshoot an issue, or monitor the history of actions that have been logged in an application.

What is written to the log?

The formatter determines what to display or store.

When is data written to the log?

The logging level determines when to log the issue.

Where is the log data sent to?

The handler determines where to send the log data whether that's a file, or stdout.

Why would I want to use logging?

To keep a history of actions taken during your code.

How is the data sent to the log?

The loggers determine how to bundle all of it together through calls to various methods.


Let's say I want a logger called my_app_errors that captures all ERROR level incidents and higher to a file and to tell me the date time, level, message, logger name, and give a trace back of the error, I could do the following:

import logging

message='oh no! an error occurred'
formatter = logging.Formatter('%(asctime)s - %(levelname)s - %(message)s - %(name)s')
logger = logging.getLogger('my_app_errors')
fh = logging.FileHandler('errors.log')
logger.error(message, stack_info=True)

The code above would generate something like this to a file called errors.log

2022-03-28 19:45:49,188 - ERROR - oh no! an error occurred - my_app_errors
Stack (most recent call last):
  File "/Users/ryan/Documents/github/logging/test.py", line 9, in <module>
    logger.error(message, stack_info=True)

If I want a logger that will do all of the above AND output debug information to the console I could:

import logging

message='oh no! an error occurred'

logger = logging.getLogger('my_app_errors')

ch = logging.StreamHandler()
fh = logging.FileHandler('errors.log')

formatter = logging.Formatter('%(asctime)s - %(levelname)s - %(message)s - %(name)s')



logger.error(message, stack_info=True)
logger.debug(message, stack_info=True)

Again, the code above would generate something like this to a file called errors.log

2022-03-28 19:45:09,406 - ERROR - oh no! an error occurred - my_app_errors
Stack (most recent call last):
  File "/Users/ryan/Documents/github/logging/test.py", line 18, in <module>
    logger.error(message, stack_info=True)

but it would also output to stderr in the terminal something like this:

2022-03-27 13:18:45,367 - ERROR - oh no! an error occurred - my_app_errors
Stack (most recent call last):
  File "<stdin>", line 1, in <module>

The above it a bit hard to scale though. What happens when we want to have multiple formatters, for different levels that get output to different places? We can incorporate all of that into something like what we see above, OR, we can stat to leverage the use of logging configuration files.

Why would we want to have multiple formatters? Perhaps the DevOps team wants robust logging messages on anything ERROR and above, but the application team wants to have INFO and above in a rotating file name schema, while the QA team needs to have the DEBUG and up output to standard out.

You CAN do all of this inline with the code above, but would you really want to? Probably not.

Enter configuration files to allow easier management of log files (and a potential way to make everyone happy) which I'll cover in the next post.

New Theme, who dis?

Because I have a couple of posts that I need/want to work on, and I have the time to work on them, I have of course decided to instead to update the theme on my blog because that was a way better use of my time 😂

Also, because the day is just too nice to not be sitting outside watching baseball (even if it's on TV ... and even if it's the ping of the bat and not the crack of the bat1)

  1. Since the MLB Lockout is still going on and there's no end in sight, I've resorted to watching NCAA Baseball. I have to say, it's really entertaining AND it seems like there's 100 games on each day!

I made a Slackbot!

Building my first Slack Bot

I had added a project to my OmniFocus database in November of 2021 which was, "Build a Slackbot" after watching a Video by Mason Egger. I had hoped that I would be able to spend some time on it over the holidays, but I was never able to really find the time.

A few weeks ago, Bob Belderbos tweeted:

And I responded

I didn't really have anymore time now than I did over the holiday, but Bob asking and me answering pushed me to actually write the darned thing.

I think one of the problems I encountered was what backend / tech stack to use. I'm familiar with Django, but going from 0 to something in production has a few steps and although I know how to do them ... I just felt ~overwhelmed~ by the prospect.

I felt equally ~overwhelmed~ by the prospect of trying FastAPI to create the API or Flask, because I am not as familiar with their deployment story.

Another thing that was different now than before was that I had worked on a Django Cookie Cutter to use and that was 'good enough' to try it out. So I did.

I ran into a few problems while working with my Django Cookie Cutter but I fixed them and then dove head first into writing the Slack Bot

The model

The initial implementation of the model was very simple ... just 2 fields:

class Acronym(models.Model):
    acronym = models.CharField(max_length=8)
    definition = models.TextField()

    def save(self, *args, **kwargs):
        self.acronym = self.acronym.lower()
        super(Acronym, self).save(*args, **kwargs)

    class Meta:
        unique_together = ("acronym", "definition")
        ordering = ["acronym"]

    def __str__(self) -> str:
        return self.acronym

Next I created the API using Django Rest Framework using a single serializer

class AcronymSerializer(serializers.ModelSerializer):
    class Meta:
        model = Acronym
        fields = [

which is used by a single view

class AcronymViewSet(viewsets.ReadOnlyModelViewSet):
    serializer_class = AcronymSerializer
    queryset = Acronym.objects.all()

    def get_object(self):
        queryset = self.filter_queryset(self.get_queryset())
        acronym = self.kwargs["acronym"]
        obj = get_object_or_404(queryset, acronym__iexact=acronym)

        return obj

and exposed on 2 end points:

from django.urls import include, path

from .views import AcronymViewSet, AddAcronym, CountAcronyms, Events

app_name = "api"

user_list = AcronymViewSet.as_view({"get": "list"})
user_detail = AcronymViewSet.as_view({"get": "retrieve"})

urlpatterns = [
    path("", AcronymViewSet.as_view({"get": "list"}), name="acronym-list"),
    path("<acronym>/", AcronymViewSet.as_view({"get": "retrieve"}), name="acronym-detail"),
    path("api-auth/", include("rest_framework.urls", namespace="rest_framework")),

Getting the data

At my joby-job we use Jira and Confluence. In one of our Confluence spaces we have a Glossary page which includes nearly 200 acronyms. I had two choices:

  1. Copy and Paste the acronym and definition for each item
  2. Use Python to get the data

I used Python to get the data, via a Jupyter Notebook, but I didn't seem to save the code anywhere (🤦🏻), so I can't include it here. But trust me, it was 💯.

Setting up the Slack Bot

Although I had watched Mason's video, since I was building this with Django I used this article as a guide in the development of the code below.

The code from my views.py is below:

ssl_context = ssl.create_default_context()
ssl_context.check_hostname = False
ssl_context.verify_mode = ssl.CERT_NONE

SLACK_BOT_USER_TOKEN = getattr(settings, "SLACK_BOT_USER_TOKEN", None)
CONFLUENCE_LINK = getattr(settings, "CONFLUENCE_LINK", None)
client = slack.WebClient(SLACK_BOT_USER_TOKEN, ssl=ssl_context)

class Events(APIView):
    def post(self, request, *args, **kwargs):

        slack_message = request.data

        if slack_message.get("token") != SLACK_VERIFICATION_TOKEN:
            return Response(status=status.HTTP_403_FORBIDDEN)

        # verification challenge
        if slack_message.get("type") == "url_verification":
            return Response(data=slack_message, status=status.HTTP_200_OK)
        # greet bot
        if "event" in slack_message:
            event_message = slack_message.get("event")

            # ignore bot's own message
            if event_message.get("subtype"):
                return Response(status=status.HTTP_200_OK)

            # process user's message
            user = event_message.get("user")
            text = event_message.get("text")
            channel = event_message.get("channel")
            url = f"https://slackbot.ryancheley.com/api/{text}/"
            response = requests.get(url).json()
            definition = response.get("definition")
            if definition:
                message = f"The acronym '{text.upper()}' means: {definition}"
                confluence = CONFLUENCE_LINK + f'/dosearchsite.action?cql=siteSearch+~+"{text}"'
                confluence_link = f"<{confluence}|Confluence>"
                message = f"I'm sorry <@{user}> I don't know what *{text.upper()}* is :shrug:. Try checking {confluence_link}."

            if user != "U031T0UHLH1":
                    blocks=[{"type": "section", "text": {"type": "mrkdwn", "text": message}}], channel=channel
                return Response(status=status.HTTP_200_OK)
        return Response(status=status.HTTP_200_OK)

Essentially what the Slack Bot does is takes in the request.data['text'] and checks it against the DRF API end point to see if there is a matching Acronym.

If there is, then it returns the acronym and it's definition.

If it's not, you get a message that it's not sure what you're looking for, but that maybe Confluence1 can help, and gives a link to our Confluence Search page.

The last thing you'll notice is that if the User has a specific ID it won't respond with a message. That's because in my initial testing I just had the Slack Bot replying to the user saying 'Hi' with a 'Hi' back to the user.

I had a missing bit of logic though, so once you said hi to the Slack Bot, it would reply back 'Hi' and then keep replying 'Hi' because it was talking to itself. It was comical to see in real time 😂.

Using ngrok to test it locally

ngrok is a great tool for taking a local url, like localhost:8000/api/entpoint, and exposing it on the internet with a url like https://a123-45-678-901-234.ngrok.io/api/entpoint. This allows you to test your local code and see any issues that might arise when pushed to production.

As I mentioned above the Slack Bot continually said "Hi" to itself in my initial testing. Since I was running ngrok to serve up my local Server I was able to stop the infinite loop by stopping my local web server. This would have been a little more challenging if I had to push my code to an actual web server first and then tested.


This was such a fun project to work on, and I'm really glad that Bob tweeted asking what Slack Bot we would build.

That gave me the final push to actually build it.

  1. You'll notice that I'm using an environment variable to define the Confluence Link and may wonder why. It's mostly to keep the actual Confluence Link used at work non-public and not for any other reason 🤷🏻

Page 1 / 17