phil-s
phil-s
"Don't use `M-x org-mode`" is the answer to that, I would have thought? Changing major mode in *any* buffer will kill the buffer's header- and footer-lines (along with most other...
I just hit this testing against the staging instance of our local Matrix server, which is running version 1.85.2. ```json {"errcode":"M_UNKNOWN", "error":"'from' parameter is invalid"} ``` The logged request from...
Aha, https://spec.matrix.org/latest/client-server-api/#get_matrixclientv3roomsroomidmessages says `from` became optional in version 1.3. I would guess that related changes meant that the token values that `from` would be sourced from were also not guaranteed...
To me that sounds like ement.el could (in the absence of `prev-batch`) check to see if it had a `start` value and if so use that value; otherwise... it's not...
Oh; no I'm missing that "paginating from the first visible event" in this case is moving *backwards* in history to earlier events. So making the request without `from` seems like...
I've tested this simple change, and I believe it's working as desired in my case. The first change avoids errors in notification/mention buffers. ```diff modified ement-room.el @@ -1376,16 +1376,17 @@...
Ah, there's still an issue here. Once at the beginning of the room history, repeated calls to `ement-room-retro` are alternating between fetching zero messages and fetching the first page again....
> Purging events shouldn't be necessary. Perhaps, but it would seem useful for debugging. I'd certainly like to be able to do this and then view the room again as...
> You can "view the room again as if it was the first time I'd done it this session" by simply killing the room's buffer and using ement-view-room again. That...
I'm planning to spend some time on ement on Friday, so I'll take another look at this then and hopefully other things as well. I don't know how much I'll...