Louis Maddox
@permutans.bsky.social
Combinatorially curious https://spin.systems
interesting, so Python encourages thread reuse in ThreadPoolExecutor because threads are more expensive because of the GIL [Python interpreter state], Rust threads are thin wrappers around OS threads
Rust thread size is 2 MiB doc.rust-lang.org/std/thread/#...
Python’s is 8 MiB (i.e. the OS default)
Rust thread size is 2 MiB doc.rust-lang.org/std/thread/#...
Python’s is 8 MiB (i.e. the OS default)
November 20, 2025 at 4:17 PM
interesting, so Python encourages thread reuse in ThreadPoolExecutor because threads are more expensive because of the GIL [Python interpreter state], Rust threads are thin wrappers around OS threads
Rust thread size is 2 MiB doc.rust-lang.org/std/thread/#...
Python’s is 8 MiB (i.e. the OS default)
Rust thread size is 2 MiB doc.rust-lang.org/std/thread/#...
Python’s is 8 MiB (i.e. the OS default)
rustfmt concurrency headache thread github.com/rust-lang/ru...
November 20, 2025 at 2:20 PM
rustfmt concurrency headache thread github.com/rust-lang/ru...
rusqlite ! It has a query parsing/metadata exposing Statement type docs.rs/rusqlite/lat...
Makes things possible you can’t do in other dialects collected in the connectorx lib, seemed “ahead” when I reviewed it
I used it this year in a Polars bridge library called polite github.com/lmmx/polite/...
Makes things possible you can’t do in other dialects collected in the connectorx lib, seemed “ahead” when I reviewed it
I used it this year in a Polars bridge library called polite github.com/lmmx/polite/...
November 20, 2025 at 1:29 PM
rusqlite ! It has a query parsing/metadata exposing Statement type docs.rs/rusqlite/lat...
Makes things possible you can’t do in other dialects collected in the connectorx lib, seemed “ahead” when I reviewed it
I used it this year in a Polars bridge library called polite github.com/lmmx/polite/...
Makes things possible you can’t do in other dialects collected in the connectorx lib, seemed “ahead” when I reviewed it
I used it this year in a Polars bridge library called polite github.com/lmmx/polite/...
💫🌎 v0.5 crates.io/crates/syncdoc
`syncdoc --migrate; syncdoc --restore` now ‘round trips’ code through omnidoc macro rewrites & back perfectly
No net effect except docstring ‘fixes’ (moving above attrib macros, trimming) like a linter in effect. Long tail no doubt but I see it work on my codebases
`syncdoc --migrate; syncdoc --restore` now ‘round trips’ code through omnidoc macro rewrites & back perfectly
No net effect except docstring ‘fixes’ (moving above attrib macros, trimming) like a linter in effect. Long tail no doubt but I see it work on my codebases
November 20, 2025 at 12:54 PM
💫🌎 v0.5 crates.io/crates/syncdoc
`syncdoc --migrate; syncdoc --restore` now ‘round trips’ code through omnidoc macro rewrites & back perfectly
No net effect except docstring ‘fixes’ (moving above attrib macros, trimming) like a linter in effect. Long tail no doubt but I see it work on my codebases
`syncdoc --migrate; syncdoc --restore` now ‘round trips’ code through omnidoc macro rewrites & back perfectly
No net effect except docstring ‘fixes’ (moving above attrib macros, trimming) like a linter in effect. Long tail no doubt but I see it work on my codebases
HuggingFace xet-core using ctor to initialise constants 👀 github.com/huggingface/... setting the env var like coreutils nooo
November 20, 2025 at 2:08 AM
HuggingFace xet-core using ctor to initialise constants 👀 github.com/huggingface/... setting the env var like coreutils nooo
Figured out how to cfg(test)-gate debug logging with env var control such that you get ergonomic defaults but can always override github.com/lmmx/syncdoc...
November 19, 2025 at 7:27 PM
Figured out how to cfg(test)-gate debug logging with env var control such that you get ergonomic defaults but can always override github.com/lmmx/syncdoc...
iOS version 18+ feature I think, I don't see any option for it on Android nope. My home settings has "themed icons" (Motorola UI settings) but it doesn't look like the above
November 19, 2025 at 6:15 PM
iOS version 18+ feature I think, I don't see any option for it on Android nope. My home settings has "themed icons" (Motorola UI settings) but it doesn't look like the above
Hold an icon (any icon) down, click edit home screen, then press edit in the corner of the screen, click customise, change the icon mode to dark
November 19, 2025 at 5:51 PM
Hold an icon (any icon) down, click edit home screen, then press edit in the corner of the screen, click customise, change the icon mode to dark
Not on iPad you don’t
November 19, 2025 at 5:47 PM
Not on iPad you don’t
My hunks runneth over
November 19, 2025 at 2:57 AM
My hunks runneth over
It was great to see this come along, I just wish I could go back and unsee the drivel that came the day after from the peanut gallerists. In lieu of dwelling on that please accept my pre-PEP Ferrisnek concept art
November 19, 2025 at 2:15 AM
It was great to see this come along, I just wish I could go back and unsee the drivel that came the day after from the peanut gallerists. In lieu of dwelling on that please accept my pre-PEP Ferrisnek concept art
The round tripping itself is still in finalisation but first glimpse at a kind of docstring linting it achieves as a side effect: this was all automated from parsing the outer signatures of the module items as tokens github.com/lmmx/asteris...
(syncdoc puts docstrings before other inner attributes)
(syncdoc puts docstrings before other inner attributes)
November 18, 2025 at 4:53 PM
The round tripping itself is still in finalisation but first glimpse at a kind of docstring linting it achieves as a side effect: this was all automated from parsing the outer signatures of the module items as tokens github.com/lmmx/asteris...
(syncdoc puts docstrings before other inner attributes)
(syncdoc puts docstrings before other inner attributes)
graphic::design.is(&my_passion);
November 18, 2025 at 1:21 PM
graphic::design.is(&my_passion);
Some mighty fine looking path creation expectation tests using insta inline snapshots and braces now in the syncdoc CLI crate github.com/lmmx/syncdoc...
November 17, 2025 at 3:02 PM
Some mighty fine looking path creation expectation tests using insta inline snapshots and braces now in the syncdoc CLI crate github.com/lmmx/syncdoc...
∧＿∧
(｡･ω･｡)つ━☆・*。
⊂/ / ・゜
しーＪ
(｡･ω･｡)つ━☆・*。
⊂/ / ・゜
しーＪ
November 17, 2025 at 12:12 PM
∧＿∧
(｡･ω･｡)つ━☆・*。
⊂/ / ・゜
しーＪ
(｡･ω･｡)つ━☆・*。
⊂/ / ・゜
しーＪ
Now preserves trailing slashes and generally way better
November 17, 2025 at 2:06 AM
Now preserves trailing slashes and generally way better
Needed something like this today so I made it :)
New {library,CLI} for brace {compaction,expansion,reprocessing} with customisable {separator,sorting,deduplication} and ability to set max {depth,brace size}
📂 github.com/lmmx/braces
📦 docs.rs/braces
New {library,CLI} for brace {compaction,expansion,reprocessing} with customisable {separator,sorting,deduplication} and ability to set max {depth,brace size}
📂 github.com/lmmx/braces
📦 docs.rs/braces
November 16, 2025 at 7:47 PM
Needed something like this today so I made it :)
New {library,CLI} for brace {compaction,expansion,reprocessing} with customisable {separator,sorting,deduplication} and ability to set max {depth,brace size}
📂 github.com/lmmx/braces
📦 docs.rs/braces
New {library,CLI} for brace {compaction,expansion,reprocessing} with customisable {separator,sorting,deduplication} and ability to set max {depth,brace size}
📂 github.com/lmmx/braces
📦 docs.rs/braces
Struct-valued enum variants now supported in syncdoc 0.3.4 which means they can get formatted onto a single line 😍
Lines 56-75 become lines 32-38 after migrating with `syncdoc -m`
Docs path is now added to Cargo.toml automatically so isn’t needed in the omnidoc macro calls themselves
Lines 56-75 become lines 32-38 after migrating with `syncdoc -m`
Docs path is now added to Cargo.toml automatically so isn’t needed in the omnidoc macro calls themselves
November 15, 2025 at 11:00 PM
Struct-valued enum variants now supported in syncdoc 0.3.4 which means they can get formatted onto a single line 😍
Lines 56-75 become lines 32-38 after migrating with `syncdoc -m`
Docs path is now added to Cargo.toml automatically so isn’t needed in the omnidoc macro calls themselves
Lines 56-75 become lines 32-38 after migrating with `syncdoc -m`
Docs path is now added to Cargo.toml automatically so isn’t needed in the omnidoc macro calls themselves
File under things I can’t believe I didn’t already set up: open a file (using fd to match once from partial filename) at a given substring by passing a -c command (like python) to search for it at start up gist.github.com/lmmx/6112346...
November 15, 2025 at 9:05 PM
File under things I can’t believe I didn’t already set up: open a file (using fd to match once from partial filename) at a given substring by passing a -c command (like python) to search for it at start up gist.github.com/lmmx/6112346...
admin reveal: mlem edition
November 15, 2025 at 5:26 PM
admin reveal: mlem edition
migrating an entire crate to syndoc + checks with missing_docs lint ✅
November 15, 2025 at 11:23 AM
migrating an entire crate to syndoc + checks with missing_docs lint ✅
First hint of the LOC changes to source files in a real world code when migrating to syncdoc (+/-)
(the section_move test one is spurious: still some stray formatting diffs being applied)
(the section_move test one is spurious: still some stray formatting diffs being applied)
November 15, 2025 at 1:33 AM
First hint of the LOC changes to source files in a real world code when migrating to syncdoc (+/-)
(the section_move test one is spurious: still some stray formatting diffs being applied)
(the section_move test one is spurious: still some stray formatting diffs being applied)
Claude on hunk splitting algorithms:
November 14, 2025 at 11:46 PM
Claude on hunk splitting algorithms:
syncdoc-migrate 📤 (WIP) github.com/lmmx/syncdoc...
November 13, 2025 at 2:42 PM
syncdoc-migrate 📤 (WIP) github.com/lmmx/syncdoc...
Asked Claude how to express a code action to modify a program and it suggested a declarative code generator from TOML program schema... 🤔🪽 cool idea but then the source would only be visible via the deployed artifact (or I suppose you could edit one crate to generate the other)
November 13, 2025 at 11:56 AM
Asked Claude how to express a code action to modify a program and it suggested a declarative code generator from TOML program schema... 🤔🪽 cool idea but then the source would only be visible via the deployed artifact (or I suppose you could edit one crate to generate the other)