Blog | ~dbms37

Missing Blog Data & Programming

Turns out I was missing a chunk of information from the last blog post I wrote. I'll rectify that here in this update but I do wish to detail some additional matters too.

RSS

An RSS feed is a bit overkill for this website I believe but I've found it helps keep track of what folks (or myself) are doing. Prior to using more open source software, I was quite fond of Google Reader. Simpler times, one could argue.

Regarding how the RSS feed is generated for this site, I whipped together a small script to slap a new entry ontop of the item list.

### Variable declaration for certain items
link_format="https://tilde-dbms37.w10.site/blogosphere/?post="

# Find item insertion line
ln=`awk '/<!--items-->/{print NR+1}' $rss_feed`
if [ $? -ne 0 ]; then
	echo "RSS feed syntax error - missing item comment"
	exit 1
fi

if [ $# -lt 1 ]; then
	latest=`ls $blog_dir | sort -rn | head -n 1`
	if [ $? -ne 0 ]; then
		echo "Issue finding latest blog entry, aborting!"
		exit 1
	fi
else
	latest=$1
	if [ ! -f "$blog_dir/$latest" ]; then
		echo "File not found in $blog_dir/$latest"
		exit 2
	fi
fi

### Details for finding the item info from post and writing using "ed"

As for actually creating the RSS feed itself, it was a matter of making a template file from the RSS specification and altering details as necessary. If I need to erase the existing RSS feed due to consistency errors, then there is a secondary script to make it from scratch and run the updater in loop.

Uploading Scripts

Everything is handled via lftp, which wasn't the first choice of application; however, it works remarkably well with here-documents. Exactly as done with the updaterss script, the uploadscript is used to convert the Markdown files I write into basic HTML. Note, I don't use the staticgen.py file I originally wrote to handle this anymore. Doesn't make any sense to not use cmark as it handles most if not every edge case. Only exception is that for the generated files to work with the blog engine, the headers must be converted upward from h3 to h5, etc.

Source Code

I've started publishing my programs, if any, to my Codeberg page. I don't see a need to be using something like "GitHub" as they are owned by the selatious Micro$lop. The continued usage of AI in everything is disappointing to incredulous levels.

AI tangent aside, I'll see about uploading my blogging/rss scripts to Codeberg in the near future as to provide folks with something to base off of if interested. I don't anticipate updating the scripts much if ever, as they seem to be sufficient. It's not like Web1.0, smolweb, etc. ever need to update with new "fancy" features. Defeats the purpose, no?

Biggest hurdle prior to publishing too much on Codeberg is ensuring I don't have inappropriate or mistranslated contributing guides. Being explicitly clear is the intention. No need for being brash with anything via OSS.

Upcoming Projects

I need to find a good mirror for Codeberg projects and see what I can do to establish myself on the YGG network. Beyond that, I'll keep working on some personal applications and jot my thoughts when I can.

>>