# "Don't shout" warning without capitals

I started working on an answer and first just typed three lines of arithmetic expressions with numbers into the edit box. When I later returned to the page and the content of the edit box was auto-restored, I got a warning asking me not to shout in all caps. There weren't any uppercase letters, only numbers and operators. The condition for this warning should be not only that there are no lowercase letters but also that there are at least some uppercase letters.

Here's a screenshot of how something like that happens.

Update: This just happened to me again.

• Perhaps "without consecutive uppercase letters", too.
– Asaf Karagila Mod
Aug 27 '15 at 8:04
• Perhaps the algorithm decides if the "strtoupper" or whatever provides the same functionality leaves the answer invariant. Aug 27 '15 at 8:53
• This is caused by a blacklist entry for pattern ^(?-i:\P{Ll}+){15,}$, which matches 15 or more consecutive non-lowercase characters. Too bad you can't just use ^(?-i:\p{Lu}+){15,}$, since spaces and punctuation should not be considered to break the consecutive count. Oct 2 '15 at 11:22
• "Update: This just happened to me again." Did you expect it would not happen anymore?
– quid Mod
Mar 1 '16 at 12:06
• @quid: No, I wanted to bump the thread and indicate that nothing had been done about it. Mar 1 '16 at 12:08
• Yes, I guess so and find it rude. And if not rude at least unnecessary. The "bug" being completely marginal to begin with.
– quid Mod
Mar 1 '16 at 12:08
• @quid: Why is that? 29 people upvoted the bug report, it's clearly not a desirable behaviour, and it doesn't seem like a huge effort to change, so it seems reasonable that something should be done about it. I'm not criticizing anyone for not having done anything about it, just raising a flag. Why is that rude? Mar 1 '16 at 12:10
• "and it doesn't seem like a huge effort to change" Please propose a solution. (The tone is rude, in it being egocentric.)
– quid Mod
Mar 1 '16 at 12:18
• @quid: Wow, that's pretty strong criticism. I wrote "it seems" -- that implies that I might be wrong. If so, the only harm done is that I bumped a thread that $29$ people had already found important enough to upvote -- I can't see how this minor reliance on a fallible judgement of my own with minor consequences is rude. Mar 1 '16 at 12:48
• The remark on the tone did not refer to this but to the original update. But, yes, as the report did receive some consideration by a developer, it was explained to you why it is as it is and this person thought about an alternative but had to abandon it for reasons explained, I do also find it a bit strange that you just announce it should be all so easy to fix. Further I think that bumping this thread is just inefficient; pinging the developer may have been a better option. Given that you ignored their efforts then it may not have been apparent you care so much about this.
– quid Mod
Mar 1 '16 at 13:12
• @quid: I wasn't aware that a developer had dealt with the issue. How do you identify them? Mar 1 '16 at 13:14
• User page "about me" (not all have a diamond everywhere). But I can see how one can overlook this, which helps explain our difference in perception of the situation. Anyway, sorry for my outburst over a quite minor thing.
– quid Mod
Mar 1 '16 at 13:29
• @quid: No harm done. I agree that what I did would have been suboptimal if I'd realized that a developer had responded; I should have just responded to him then. Mar 1 '16 at 13:31
• @joriki That pattern is a regular expression. In particular, the Stack Exchange codebase is written in C# and uses .NET regexes. Since I don't work for the company any more, someone else will have to implement the fix. Mar 1 '16 at 16:44
• @quid Yes, I was working for the company at the time I wrote my initial comments. I was doing bug duty at the time. Mar 1 '16 at 17:19