[02:31:46] <wm-bb>	 <u99of9> Good question. I don't know. I suppose the other precedent to match is the URL data that comes in from Wikidata, which is not separated into components. (re @wmtelegram_bot: <[[smlckz]]> "should be more complicated" I wonder.. I'd obviously would like it as you say, but look at Z20838.. what is the pr...)
[11:25:36] <wm-bb>	 <vrandecic> for example Z28803 or a function call that creates a sentence fragment such as "Battle of Trondheim" (re @u99of9: I don't understand this solution. Which ZID would be the value of that triple for the subject Q7819319 ?)
[11:26:55] <wm-bb>	 <vrandecic> My suggestion would be to for structureless URLs, and then have function that do things like "protocol from URL" or "TLD from HTTP URL" etc. (re @u99of9: Good question. I don't know. I suppose the other precedent to match is the URL data that comes in from Wikidata, which is not se...)
[23:17:13] <wm-bb>	 <Al> 🥳 A coherent diff 🙏 : https://tools-static.wmflabs.org/bridgebot/5d875217/file_83557.jpg