Alex Alabuzhev

@MKadaner I am fine with either behavior, as long as: 1. it is consistent between the regular search and the regex search (looks like currently it is not) 2. next/previous...

> What do you mean "stuck?" The search does not progress on repeated pressing of `Shift+F7`/`Alt+F7`? Yes, and `F7 Enter`.

Thank you for the explanation, I agree that it's a PITA from a dev perspective. The `ReadConsoleOutputCharacter` API doesn't sound like a viable alternative for our workflow: we need not...

Yes, it's the same idea as those F2/F4/F7/F9 popups. Also, `TakeShapshot` / `PutSnapshot` / `DeleteSnapshot` relatively nicely fit into the existing API: ```C // Take a snapshot of the current...

@zadjii-msft Thanks for getting back to this. To summarise, this is what we do now: ``` ┌─────┐ │Start│ └──┬──┘ ┌───────────────────────────────────►┤ │ ▼ │ ┌─────────┴─────────┐ │ │ Take a snapshot │...

Thanks. Related question: after aa18839 error messages are less detailed - the xml content isn't displayed anymore [here](https://github.com/leethomason/tinyxml2/commit/aa1883928bf20605db7ac1193c016364b580c85c#diff-ecca82e15b27e99711e9960d29378f69R1858) and in some other places. Is it intentional?

Closing in favor of 050e3b4a869e7b351e6fe68c48ae034d93139ac1.

> I suppose that the most compatible behavior can be achieved if update item's timestamp both on opening and closing. Ok, as requested. 6051.

> I wonder if @alabuzhev knows about any specific cases that are at risk for which we can add explicit synthetic testing. Could you please briefly summarize what is this...

Thanks for the explanation. > If you make a snapshot with `ReadConsoleOutputW` and then restore it with `WriteConsoleOutputW`, we'll have to be careful that any _partially_ captured leading/trailing characters are...