| 4 Jun 2026 |
 toonn | Yes, so moving the decimal point is always a greater change than flipping a bit in the mantissa. | 12:56:59 |
 alexfmpe | It can't really ever be the case that one of these always yields a smaller change than the other | 12:58:34 |
|  @lumi:the-apothecary.club changed their profile picture. | 12:58:46 |
| alexfmpe | Otherwise you could iterate all floats after it by just iterating the mantissa | 12:59:04 |
| toonn | I'm not sure what's surprising about that. | 13:00:04 |
| toonn | Floats are "grouped" by their exponent kind of. There's no float with a different exponent in between two floats with the same exponent. | 13:01:22 |
| alexfmpe | Ah, because the mantissa is sandwhiched between 1 and 2, and the exponent is used as power of 2? | 13:12:13 |
| alexfmpe | Ok yeah that adds up. For some reason I was thinking the exponent was applied directly to the mantissa, but it's really applied to 2 with the result multiplied by the mantissa | 13:13:29 |
 Sergei Zimmerman (xokdvium) | Actually because of the hidden bit it's always between 0 and 1, though subnormals are a special case | 13:15:50 |
| Sergei Zimmerman (xokdvium) | The implicit bit is not stored anywhere for normals | 13:16:08 |
| Sergei Zimmerman (xokdvium) | That allows to have one more significand bit | 13:16:36 |
| Sergei Zimmerman (xokdvium) | And for subnormals that is derived from the exponent value | 13:17:12 |
|  bow joined the room. | 19:50:29 |
|  @logn:zirco.dev left the room. | 21:34:48 |
| 5 Jun 2026 |
|  Manuel joined the room. | 00:45:05 |
|  Shelvacu joined the room. | 09:11:39 |
| 6 Jun 2026 |
|  okobjective joined the room. | 01:43:41 |
|  @koichiio:matrix.org left the room. | 06:23:32 |
|  jutty joined the room. | 10:38:57 |
| 7 Jun 2026 |
|  Agent.69 changed their profile picture. | 06:07:34 |
|  @avdhut7009:matrix.org left the room. | 13:48:14 |
|  richarddodd joined the room. | 14:14:03 |
|  Messtron888 joined the room. | 20:27:57 |
|  new mxid @undrrdawg:kluge.cafe changed their display name from forkd [ he/him/any || UTC+3 ] to new mxid @undrrdawg:kluge.cafe. | 23:52:43 |
| 8 Jun 2026 |
|  mt joined the room. | 17:44:34 |
| @lumi:the-apothecary.club changed their profile picture. | 18:37:14 |
| @lumi:the-apothecary.club changed their profile picture. | 18:39:45 |
|  @moosecurity:matrix.org left the room. | 23:10:50 |
| 9 Jun 2026 |
 llakala | is there any way to add error context for attempting to access an attribute that doesn't exist?
… while calling 'from'
at /home/emanresu/Documents/projects/adios-wrappers/modules/mkWrapper/default.nix:11:20:
10| inputs = {
11| nixpkgs.from = { parent }: parent.nonexistent;
| ^
12| };
error: attribute 'nonexistent' missing
at /home/emanresu/Documents/projects/adios-wrappers/modules/mkWrapper/default.nix:11:39:
10| inputs = {
11| nixpkgs.from = { parent }: parent.nonexistent;
| ^
12| };
| 10:31:40 |
| llakala | i want to be able to print all the children of the parent module that do exist | 10:31:58 |
