Artifact 18ba960d3df7ebbdc4102678a54ebffe1578a905aac808c2924ed336db702c62:
Instance of technote
[
a88bbe840b]
- More on (daemonp) vs server-process.
by
jaccarmac
2018-11-12 01:59:53.
C More\son\s(daemonp)\svs\sserver-process.
D 2018-11-12T02:21:04.972
E 2018-11-12T01:59:53 a88bbe840b3eca64159146dacf711c91bd598007
P 90ce235b67eab5f0c3c5aa66ee2e881c375447058cf0cf796b00121dbd69a604
N text/x-fossil-wiki
U jaccarmac
W 1099
For background, see [7395c48a0179cdde063afe54f3a9f7493c510a8c5dd0e4d92b7c51c1936ed092] and [ba7732f136ba1704abec5fd43f6819c67d4e1367f3abb6762773beb6554f66cd]
Pretty sure I've solved the puzzle of why (daemonp) works to create a loading hook for themes on startup while the supposedly more robust (and (boundp 'server-process) server-process) does not. It seems that the latter works for determining whether or not a server is running, but emacs --daemon or equivalent does not start the server until it is finished executing Elisp. Which makes sense in a way.
So, in conclusion, if there's some way to start a headless Emacs without running emacs --daemon I may need the even more robust (or (daemonp) (and (boundp 'server-process) server-process)) but for now I'll stick to the old way.
Another related consideration is starting Emacs in a terminal where Nord's colors are not available all at once. I don't do this, but I can see that being a reason to use condition-based handling of this case rather than trying to figure out every possible permutation of factors impeding Nord-loading.
Z 7416ea99682c3fc584b929bcb20db20d