CI: Show useful data about token URLs #192
Conversation
tooomm
changed the title
|
The upstream issue is resolved and |
| # Extract domains from URLs | remove blank lines | sort | count and list | sort descending | ||
| run: | | ||
| echo "📊 **Image Hosting Statistics**" >> $GITHUB_STEP_SUMMARY | ||
| awk -F/ '{print $3}' lychee/out.md | grep . | sort | uniq -c | sort -nr >> $GITHUB_STEP_SUMMARY |
There was a problem hiding this comment.
why is grep . here? is there empty lines? why does it run sort twice?
There was a problem hiding this comment.
why is grep . here? is there empty lines?
There used to be empty lines in the output when we had token or sets added without image links. This is currently not the case.
Since we do not validate for empty/missing links that could happen again.
But I can also remove it.
Am actually not sure if empty lines would even mess with the result or not.
There was a problem hiding this comment.
I guess the comment on it could be more explicit, it takes the third column from the file, removes empty lines, sorts and then counts the duplicates, listing them with their count before the url, then sorts again but based on the number that represents the amount of duplicates descending.
There was a problem hiding this comment.
Updated the wording of the comment.
The sorting needs to be split because first it's sorting to bring same domain names next to reach other for counting to work. Only after combining the entries and adding quantity numbers it can be sorted descending.
We automatically check if imagine links are accessible/available.
This adds some more insights and helps with understanding the file. It gives us a little overview of its content and our sources/dependencies, also highlighting doubled links (which we eliminated by now).
Result:
https://github.com/Cockatrice/Magic-Token/actions/runs/10619612692/attempts/1#summary-29437582199
Screenshot: