I also came across edits of this type and treated them as follows:
Approve
if the edit actually improved the post by triggering correct syntax highlighting while it wasn't possible to just add a tag, e.g. an answer contains a bash
script while the question isn't really about bash
.
Reject
if the syntax highlighting was already fine before the edit, e.g. an answer contains a bash
script, but the question is already (or can be → Reject and Edit) tagged bash and is unlikely to lose this tag – the edit did not produce any visible improvement.
These edits are made by the same user.
A thought about this part of your question which seems to have disappeared over the whole yes-we-really-need-syntax-highlighting discussion: The ones I encountered also were merely, if not all, from that exact user. At first glance it may seem like a loophole to quickly gain lots of reputation, but:
- you can't just go through every command-line question and blindly add
<!-- language: bash -->
to every piece of code you encounter, there is definitely work involved
- no matter how much work is needed in the end, an edit improving a post is worth a reward no matter what
--
in the title rendered as—
? That's highly disturbing in this case!