Combining noun-class with aspect.

I don't understand why we're still talking about this, but since we apparently are . . .

When {{t}} has both a noun-class and an aspect, e.g. {{t|ru|foo|c1|impf}}, you get a big red "Script error" message:

That's what I was referring to.

Find me a Russian translation which has both. If there is a single one, it was an error in the first place. Russian has no noun classes. Only gender and animacy. And especially verbs have no noun classes.

I just finished scanning the latest dump and found 0 (in words: zero) translations which would be affected by this.

Which is why I wrote:

> I think it's unlikely that we have any instances of anything like {{t|ru|foo|c1}} {{pf}}: it's probably not worth worrying about.

Now, having seen your further comments at Wiktionary:Grease pit . . . I said "It looks good overall" and then focused on the few things that seemed like they might need improvements. I take it that you were hoping for bigger accolades? :-/

Not really. That discussion summary was my poor attempt at being humorous. I try to suppress my vanity most of the time. Though it does sort of annoy me that you seemingly tend to focus on hypothetical situations and portray them as rather disastrous without doing research to determine whether they would actually arise. But maybe this is just my perception.

I'm not sure what you mean. Can you offer any examples?

Some discussions about xte long ago, back when I announced the first version (though in your defence, that was actually rather reasonable, and I actually implemented changes to address your objections), or later with linkifying bare text in the same tool. Or the {{it-noun}} issue, where you were all like "But someone could have objected!". Or this very discussion about mixing gender with aspect markers, which is rather easy to read this way.

But maybe I am overgeneralising. Not quite sure about that.

Sorry, I don't remember the details of the xte or linkifying-bare-text discussions. :-/

With {{it-noun}} the problem isn't that someone could have objected, but that CodeCat didn't bother to ask.

With this very discussion about mixing gender with aspect markers, the situation did in fact exist — not "hypothetical" — so that actually seems to justify my approach. :-P   (Though admittedly, since I misremembered the case that the template threw an error for, the situation wasn't the problem I thought it was.)

(Actually, I guess I didn't really read your comment critically enough. You're saying that it's not a bot-owner's job to research and make sure (s)he's not breaking things, but rather, that it's everyone else's job to prove that (s)he is — or else they should just keep their mouth shut. That's a pretty fundamental disagreement we have, so I guess I just have to accept that I will annoy you!)

Well, no. Of course the bot operator should do their own research.

My own research was done by running the script manually thorough Wiktionary, examining potential problems, and coming up with a common case that could be uncontroversially converted mechanically. And I knew that combining aspects with genders in this way did not cause problems. The point was — maybe I already checked how things work. Test how and if things break under some specific conditions instead of vaguely shouting "but it will break!". The only response you will get to "what if it would cause Problems™" is "it would not", and that is certainly not very productive.

And to be fair, CodeCat could have handled that {{it-noun}} issue better too. Still, it is the end result that should have more weight.

And here is the xte thread: User talk:Kephir/gadgets/xte#Unwikified links always become ttbc.

O.K., fair enough. :-)