| Time |
Nick |
Message |
| 02:29 |
|
YuGiOhJCJ joined #luanti-dev |
| 02:39 |
|
user2_ joined #luanti-dev |
| 02:45 |
|
aliasreadytaken joined #luanti-dev |
| 02:47 |
|
SpaceManiac joined #luanti-dev |
| 04:00 |
|
MTDiscord joined #luanti-dev |
| 07:08 |
|
user2_ joined #luanti-dev |
| 11:59 |
|
imi joined #luanti-dev |
| 12:12 |
|
alias joined #luanti-dev |
| 14:45 |
[MatrxMT] |
<birdlover32767> <90 pull requests 🎉 |
| 14:51 |
repetitivestrain |
Krock: regarding the documentation change in 16397: not only is it unnecessary (very few boolean parameters are expressly stated to be booleans in lua_api.md), it is also incorrect, inasmuch as the default value is nil rather than false, and POS1 and POS2 must _match_ the extents of the chunk undergoing generation rather than simply remain within these extents |
| 14:53 |
repetitivestrain |
and for what it's worth i don't really believe the patches render the function any more or less comprehensible |
| 14:59 |
repetitivestrain |
"in respect to" is also not terribly contemporary English, and certainly is not being used in an appropriate context, it expressing relation rather than adherence |
| 15:40 |
|
user2_ joined #luanti-dev |
| 16:31 |
Krock |
repetitivestrain: the type and default value was not documented at all, which raises questions for mod writers. Just because the types or value ranges are not defined in the other API does not mean that it's helpful.  As for the pos1/pos2 topic  - yes, that's bound to the earlier question |
| 16:31 |
Krock |
.. and moreso incorrect, as I've learnt. |
| 16:37 |
Krock |
as for the patches - I suppose that's somewhat subjective. For example, it was not clear to me why the mapgen context error is raised based on the vmanip object. More precisely, it should error because the mapgen object does not have the biomemap present (which is exactly what this code depends on) |
| 16:41 |
Krock |
In addition, I like to keep variables in the smallest possible scope such that it is easier to see (often without scrolling) where it is used. For example, "oldvm" and "mg" would now be less spread out, and even have less branches that need testing ("vm_old" would act as a backup in any case) |
| 16:43 |
Krock |
I thought about this PR a few times - in long-term it might be saner to create a separate Mapgen instance dedicated to out-of-mapgen-scope map manipulations, such that it is no longer necessary to tinker with members that should be private |
| 16:56 |
|
fluxionary_ joined #luanti-dev |
| 18:58 |
|
fluxionary_ joined #luanti-dev |
| 19:45 |
|
YuGiOhJCJ joined #luanti-dev |
| 20:58 |
|
fluxionary_ joined #luanti-dev |
| 22:32 |
|
panwolfram joined #luanti-dev |
| 23:05 |
|
Eragon joined #luanti-dev |
