Ramen Junkie

100DaysOfCode

100 Days of Python, Projects 66-70 #100DaysofCode

Ramen Junkie No Comments

Whew, I didn’t really think I’d get to 9 parts in this series, and I am only around 2/3rds of the way through even.  I actually may change up the format later with the last 20 projects that are listed as “Professional”.  Maybe one post each.

The bulk of this round is wrapping up the Flask projects and building a simple blog that runs on Python.  It’s been fun.  I’ve been a bit busier than normal slow my pace has slowed, but that’s ok too.  Day 66 in particular felt like it took longer than it really should have, given how little it felt like it was doing.

Day 66 – Build RESTful APIs

Kind of a different sort of project.  It’s building something, but not really anything, with any sort of interface.  All of the interaction is done through Postman (or the URL if one wants), and the responses are all JSON of some sort.  

We built 6 API interfaces to work with a database of Coffee shops.

  • One returned all of the shops.
  • One returned all of them in a particular city.
  • One returned a specific shop only.
  • One updated the price of coffee.
  • One added a new Coffee Shop.
  • One deleted a Coffee Shop.

Including built in error handling for if a Coffee Shop didn’t exist etc.  It all feels like it could be useful later in the Blog Project.  Not too useful on it’s own accord.

Day 67 – Blog Capstone Project Part 3

Back to the Blog Project, and after this round it’s a LOT more Blog Like, though incredibly insecure.  The Security Part looks to be the subject of the next couple of lessons though, so it’s all good.  We basically started with the last Blog Project from Day 59, and added the ability to add, edit, and delete posts.  Also with a Database back end.

One tricky moment here.  The idea was to use the same form page for New Post and Edit Post.  Originally I had nested these together, and had some if/else cases to check if it was Editing or Adding, and kind of got stuck on how to handle the date, since the date doesn’t get changed on updates.

But then I had a bit of an epiphany moment, the kind you get often with coding, where you band against a problem, convinced it’s the way to solve things, only to realize there is an obvious, easy method.

I split them up.  No more if/else, and no more worrying about what the DB is doing at the end.  Because two separate @app.route and related functions, can route to the same HTML template (the Edit Form).  The only thing needed was to pass a header variable along to change the title.  

It was a total “Duuuuuuh” moment.

Anyway, no security though, because there are no users or security keys and anyone can just come in and post anything and delete anything and it would be total chaos as an actual production Blog platform.

Day 68 – Authentication with Flask

Oops, I spoiled what this topic was going to be in the last project write up.  This whole section is working with Authentication using a Database for persistence.  It’s a simple website, a home page, a user creation page, a log in page, and a secrets page.  Users can create an account, and once registered, they can access the Secrets page and download a PDF.  

Part of this exercise is also restricting access for not logged in users and another part is proper security and handling passwords with hashing and salting.

Of all of the projects so far, I feel like this may be one of the most important ones, despite it’s basic simplicity.  I’ve been planning to work out the best way to share some of the projects built so far, and using Flask was a good “first step”, but being security conscious, I would rather not throw a bunch of web based Flask pages up with no restrictions on access.  Like last session, there was the Library App or the Top Ten Movies page.  Sure, I could put them up on the web-server, map some sub domain to them, but anyone can edit them, without proper, persistent security and restricted access.  

Now, I can make that happen.  Plus, with the modulator of routing and such, I can easily slip them into parts of the overall Blog Project, in time.

Day 69 – Blog Capstone Project Part 4

And here it is, the culmination of the Python based Flask Blog.  It’s neat.  I like it, well, I like the basic functionality of it.  The layout is a bit plain but that’s fixed.  I doubt it ever replaces WordPress, I love WordPress, but I could definitely see uses for this finished product.  I may actually modify it to work as a sort of “Twitter Replacement” since Twitter is currently burning down.  Using what I have learned, I could easily set this up to take a post, then “Syndicate” it out via API calls to Twitter, Facebook, Mastodon, or anywhere.  While keeping my own archive of posts.

My current, next up To Do Items:

  • Create RSS Feed Page  
  • Create Admin Page  
  • Enable/Disable Comments for Posts  
  • Allow Admin to delete Comments  
  • Allow Users to delete their own comments  
  • Add Mailer.class and add email notifications  
  • Add Pagination to Home Page  
  • Add Tags and Category Options to Posts

Also, on a side note, this actually isn’t the first time I’ve built a “Blog Platform”. I built a basic one a few years ago in HTML and PHP. Heck, the system I used int he early 2000s with SHTML Pages was sort of a “Blog Platform”.

I’d recommend, for anyone going through this course, check the notes on the Blog courses.  There are a lot of good suggestions for improvements, especially in stopping things like Java Script injection in the comments.  When you go to Angela’s (The Instructor) Blog link, it immediately throws out a JavaScript pop up that someone dropped in the comments.

Day 70 – Hosting with Heroku

So, there isn’t really a Day 70 project.  I did go through the section, and the most useful part for me was the more robust SQL solution mentioned in the last section.  The first few parts were about GIT and Github, which I am already using.  The middle part was about using Heroku, which I have heard of and used a bit before but for the long term, I don’t need to use a freemium hosting service in a jankey way.  I have a VPS, and later I will figure out how to get Apache to play nice with Flask.

For now, I’ve split the Blog Project into it’s own repository and worked up my initial planned ToDo List.  But I need to keep my focus on the course for now.

AI Art and Chat

A Progressive Journey Through Working With AI Art – Part 1 – On the Web

Ramen Junkie No Comments

This was originally one huge post, but I decided it would be better to be split up across a few posts.

Is it Dall*E? Dall-E? DALLE? I’m honestly not sure, and I’ll probably jump around a bit on which I use. Whatever it is, it’s AI Generated Art from text based prompts. And it’s pretty neat, and progressively more and more neat.

Initially, it was very limited to even do anywhere, but there was this great little service called Dall*E Mini, which let you make simple images with a sort of simplified version of the Dall*E 2 Engine. Or something like that. It eventually evolved into crAIyon.com. I don’t really use it much anymore and I’ve misplaced the images I generated with it… somewhere. Aside from that “Ramen Junkie” image above.

I have tried a few other similar systems with mixed results.

This is long and there are lots of images, so feel free to click through to read more…

—– Read More —-: A Progressive Journey Through Working With AI Art – Part 1 – On the Web Read More
100DaysOfCode

100 Days of Python, Projects 58-65 #100DaysofCode

Ramen Junkie No Comments

Judging by the comments on the lessons, The lessons are getting harder, though frankly, I am finding they are a bit easier.  I feel like my experience with Web Design is a lot of this.  I also have been looking a bit into how to properly host some of these apps to share here, on my Code Projects Page.  I believe I could set them up to run on a production Flask environment, with different ports, then just map sub directories n the domain to different ports.

But I am doing my best not to get distracted.  Yes, I have been building a few smaller projects here and there lately, but I don’t want to get TOO distracted.

Day 58 – Bootstrap Overview

Not a lot of really say here, like the Web Foundational section, this was a chunk of the Web Dev course offered by the same Instructor.  I may look into picking it up if it’s on sale for cheap, since at this point I’ve done like half of it.  My next plan for a major learning push is going to be Javascript.  I really need to get familiar with Javascript for some work projects.  

The project was essentially using different Bootstrap bits to format a silly Tinder knockoff website, a site for Dogs called Tindog.  Bootstrap is definitely useful, but I generally try to avoid it because it makes things look very samey.  I suppose familiarity in design is useful at times though.

Day 59 and 59 – Blog Capstone Project Part 2

The last part of the Intermediate+ section was the bones of a Clean Blog running in Python Flask.  These two days expanded on the Blog concept a bit, and a few later lessons bring it back around again.  I am almost interested in trying to use the blog once finished, except that I already have plenty of proper outlets for Blogging.  Wordpress works fine.  

The core concept was using Bootstrap to format the blog up nicely, as well as setting it up to reuse Header and Footer files.  It’s a common method for webdesign, and it’s one I have been using since Geocities when I discovered something called SHTML which would let me write Blog Texts in HTML, then encapsulate them into a common top and bottom.

Day 61 – Flask Forms

An intro to easier to use Flask based forms and libraries.  The project itself was to construct a simple log in page, which lead to a Rick Roll when unlocked.  I could see this piece being useful later to add in on the more complete Blog project.

Which brings to a bit of an aside topic.  A lot fo these projects feel disconnected, but really, they are demonstrating a proper way of handling larger Code Projects.

Break it into parts.

We built a simple blog page that pulled posts from a CSV file.

We formatted that blog page.

We created a log in form, that could easily be slipped into a blog page.

Later lessons work with more dynamic forms and data and SQL Lite databases.

As the pieces get laid out, the end goal becomes more and more clear.  My prediction is that the later Blog Capstone project will amount to, “Take everything from the last ten weapons and mash it into a functioning Blog app.

Day 62 – Coffee Shop Wifi Tracker

Like I said, building on the last.  For this lesson, we built a little table that would list Coffee Shops in  pretty Bootstrap table with ratings for Coffee Quality, Wifi Quality, and Power Outlet availability.  Instead of just a form that takes an input and verifies it’s good, it actually inserts new data into the CSV that holds all the coffee shops.

Day 63 – Book Collection App

I may revisit this one later to build an app for my other collections.  And to make it more robust.  The core though is once again, building on the Coffee App, essentially.  Instead of just a tracked list with an Add Form, we added persistence by learning about SQL Lite databases.  So the data persists even if the server is shut down.  

I took a few extra non required steps on this app and reused the Coffee Shop code to make this app look much nicer.  The core app and instructions just had a very basic page with an unordered list.  I turned it into a table with a bit more formatting and color.

Day 64 – Top Ten Movies App

This whole project was essentially the same as the Book Collection App, except it was intnded from the start to look prettier.  The main difference is that the add process had a few extra steps to pull from the API of The Movie DataBase website to get data about each film, and have the user select from a list of results.  In the instructions for the class, the idea was to make multiple API calls for the list, then specific movie data, but I had already worked ahead on the project and instead it pulls the data, the user selects the movie, then it’s just done.  It was a bit complicated to get the data to pass around between pages since it was a dictionary and not just a single variable, but I got it working using Global variables.

I mostly try to avoid Global variables, but it felt like this was a good use case and made sense to use given the way Flask works.

I am thinking I may integrate the log in app created previously with this website and use it to test out mapping some of these projects to a domain to make it a for real, live website. Also, you can’t tell it in the snapshot but the posters do this cool flip over effect when you mouse over them to show more information. Not really a Python thing, but it’s still a neat effect.

Day 65 – Web Design Principles

No project for this day, it’s another slice of the Web Developer course from the same instructor.  I’m definitely going to keep an eye out for a deep discount, especially with Black Friday coming, since I’ve basically completed half of that course now.

I was hoping to fit all of this more advanced Flask based projects into one page but it feels like this post is getting pretty long so I’m going to go ahead and break it up here.

Music

Amazon Music Now Sucks Completely

Ramen Junkie No Comments

Amazon Echo, probably the easiest and simplest way to listen to music I have, had just completely shat itself and it’s so incredibly frustrating.  I don’t understand why companies are so dead set on ruining every product they make (Oh wait, I do, something something endless growth bull shit, but that’s a rant for elsewhere).  I don’t care for Ala-carte subscription based music at all.  Artists general don’t get paid enough cut, and I like being able to listen to my music years to decades later, without paying monthly, endlessly.  I still own every CD I have ever bought.  Granted, I do buy some of my CDs used so the artist still gets nothing from that so I guess that’s mildly hypocritical.   Here’s an article about the change.

https://variety.com/2022/digital/news/amazon-music-prime-100-million-songs-shuffle-mode-podcasts-ad-free-1235416844/

My point is, I buy music digitally.  Hell, I bought a couple of albums just last Friday for #BandCampFriday.  I have literally hundreds of albums and thousands of trcks PURCHASED through Amazon Music.  I’ve been buying music through Amazon for over a decade.  More recently I shifted away a bit because I wanted to get FLAC options but when I can’t find FLAC I still buy from Amazon because the quality offered is still a pretty good MP3.  This made the Echo and Alexa super great because I could just tell it, “Alexa, play Aurora from my library”, and it would play my tracks, or whatever artist or album I wanted.

I even have an echo hooked to my sound system down in the basement so it sounds better.

But this recent change.  This isn’t even an option for music I have purchased.  Like now, as I type this, I played my album, “A Different Kind of Human”.  And it’s just randomly inserted some other random track in.  It’s also playing the tracks out of order.  I even tried playing it from the Alexa App on my phone.  Maybe it works better in the Amazon Music app but I had to dump that because it kept asking EVERY SINGLE TIME I opened it to subscript to Music Unlimited.

Which is the goal with this new Shuffle Play idiocy.  If you subscribe up, you can get all the old features back.  Which I have no interest in.  I own my music, and I already pay too much for Amazon Prime, which is going up AGAIN next year.  I don’t even have it on Auto Renew anymore because it’s like $150/year now and twice as much as when I started subscribing.  I really just want to pay $60/year or whatever for easier shipping.  I don’t care about all this extra nonsense.

I wouldn’t even mind this annoying change as much if it just worked with music I’ve purchased, but it doesn’t.

Looking for some answers or suggestions on fixing this around the web says that a lot of people are upset about this. Not just because of situations like mine where you can’t listen to your own music, but because of things like custom routines and playlists being completely broken as well. A couple that stood out to me were people who had routines set up for their kids to go to sleep listening to particular songs using playlists and timers. Another was how it completely breaks soundtracks to musicals, and I imagine it breaks other similar content with a need for linear tracks that run together like Stand Up Comedy and Dance mixes.

It all feels incredibly poorly thought out and frankly, it doesn’t drive me to upgrade my subscription (which is the point) but makes me want to find a different smart speaker solution that works better with my local network shares. If I were going to subscribe to music, it would be Spotify or Tidal over Amazon.

Programming Projects

FreshRSS and RSS Feed Posts

Ramen Junkie No Comments

Keen observers (ha ha ha no one reads this), might have noticed that a few posts of links showed up in the feed.  These are basically, stories I read in my RSS reader that I found interesting, and wanted to share, or at least, keep track of.  The posts as of now are a little ugly, and I’ll probably clean up the formatting over time, but I wanted to go ahead and write a bit about the process.  I’ll have the Code on Github at some point.

As for the factors, firstly, this is something I’ve wanted to have on my blog for a while.  Like a long while.  I might even try to see if there are ways to better slit up the links by topic later.  A fair number of blogs I subscribe to have these sort of link digest posts, and I’ve always just liked the idea.  It’s also good for personal reference to when I may have read something.  It is limited as it only comes from y RSS Reader.

Speaking of my RSS Reader.  I’ve moved on from TinyTinyRSS, for a few reasons.  One, the interface is a little meh, honestly.  Maybe the newer version is better but it’s only available in Docker, and Docker is such a PItA to use.  Also, while looking for alternatives, it sounds like the folks who make TTRSS are kind of a bunch of gatekeeping jerk types, and I’d rather not support that.  I also find the need to keep the update daemon running with Screen to be a pain.  So I’ve moved over to FreshRSS, which I just run locally on a Raspberry Pi.  I may move it to a publicly accessibly machine at some point, but I am not entirely convinced that TT-RSS wasn’t the entry point for my previous server malware woes.

So, like TT-RSS, Fresh RSS has a way to get an RSS feed out of your Favorited posts.  In the past I’ve used tools like IFTTT to automate posting these links around, but I don’t use IFTTT anymore for reasons I’m not going into.  Fortunately, I’ve been working to become a pretty good Python coder for the last month or so.  So instead I wrote a script.  

It’s not even a particularly complicated script.  There are only two things it really needs to do, get new articles, and then post them to WordPress. Since the script runs locally, on the same Raspberry Pi even, it easily can reach and pull the RSS feed.  One nice thing I noticed with Fresh RSS, the feed included a time interval, so just getting new posts was super simple, because the interval is just “24” for “24 hours”.  The script eventually will run on a cronjob at the exact same time daily.  Anyway, after pulling the RSS, the entries are already in an easily usable Dictionary.  which gets fed into the construction of the WordPress Post.

def get_feed(feed_url):
    NewsFeed = feedparser.parse(feed_url)
    return NewsFeed

The posting part was pretty easy as well, WordPress has an API, and Python also has a library that can use that API.  It just needs some log in information and a post payload to send.  

def make_post(NewsFeed):
    wp = Client(f'https://{wp_url}/xmlrpc.php', wp_user, wp_pass)
    post = WordPressPost()
    post.title = f"{cur_date} - Link List"
    post.terms_names = {'category': ['Link List'], 'post_tag': ['links', 'FreshRSS']}
    post.content = f"<p>Blogging Intensifies Link List for {cur_date}</p>"
    for each in NewsFeed.entries:
        post.content += f'{each.published[5:-15].replace(" ", "-")} - <a href="{each.links[0].href}">{each.title}</a></p>'

The trickiest part was formatting the date a bit prettier.  I mentioned cleaning up the formatting a bit, I’m thinking maybe a simple invisible table, so the date and the links don’t wrap oddly like they do now.   i also added a check that if there are no new favorited posts, it will skip making a post.  Otherwise I’ll end up with empty posts on days I forget to check my feed reader

While writing the script, at first I was just outputting a text copy of the post to the console until satisfied.  Eventually, I pushed out a real post, then verified that things worked.  The next day, was just a straight test by opening the project, then running it again.  The third day, I copied the files and installed the lobraries needed, then posted from the Pi.  Phase 4 of this will be to set up Cron to run it automatically.  If that works then it will certainly, “just run” for the foreseeable future.

« Previous Page — Next Page »