commit e030d73a787e8485e5440fa2d62ba932987397e1
parent ebc69d1af2abdbd964ad07838a55fc4bf62d65e8
Author: Yuval Langer <yuval.langer@gmail.com>
Date: Sat, 8 Jan 2022 20:58:02 +0200
Convert from org-mode to markdown and add titles.
Diffstat:
9 files changed, 27 insertions(+), 206 deletions(-)
diff --git a/content/making-quicksilver-and-friends-play-nice-with-wasm32.md b/content/making-quicksilver-and-friends-play-nice-with-wasm32.md
@@ -1,190 +0,0 @@
----
-layout: default.liquid
-
-title: Making Quicksilver and friends play nice with wasm32.
-is_draft: true
----
-
-## Disclaimer:
-
-This technology is indistinguishable from magic. I am merely recalling which
- spells and incantations worked for me. If you have anything to add, you can
- reach me on <https://gitgud.io/yuvallanger/kaka.farm/> or
- <https://gitlab.com/yuvallanger/kaka.farm/>.
-
-Several months ago, writing a Flappy Bird clone called [Rectangly
-Rect](https://gitgud.io/yuvallanger/rectangly-rect/), I have done a bunch of
-asking around and found exactly which parts don't work and how to replace them.
-Yesterday, trying to adapt YasamSim to the web, I have re-discovered those
-workarounds, and decided to write this down.
-
-## `println!`!
-
-First thing, drop all of your `println!`s. For some esoteric reason, this
-function throws a wrench into the web's machinery. Same goes for
-`std::time::Instance::now()`. For now I just dropped all calls to `now()`,
-maybe I could ask the browser manually with whatever function Javascript has,
-or maybe there is a more standardized `std` alternative for the web - I don't
-know.
-
-In order to replace `println!`, I had to add to `Cargo.toml` a general
-dependency for the crate `log`, an entry for every target that is not wasm32
-for the crate `env_logger`, and an entry for the crate `web_logger` for the
-wasm32 target:
-
-```Cargo.toml
-[dependencies]
-log = "0.4"
-
-[target.'cfg(not(wasm32))'.dependencies]
-env_logger = "0.6"
-
-[target.'cfg(wasm32)'.dependencies]
-web_logger = "0.1"
-```
-
-In `src/logging.rs`, conditionally compile a different `init_logger()` function
-for each platform, wasm32 and not-wasm32:
-
-```src/logging.rs
-#[cfg(target_arch = "wasm32")]
-pub fn init_logger() {
- ::web_logger::init();
-}
-
-#[cfg(not(target_arch = "wasm32"))]
-pub fn init_logger() {
- ::env_logger::init();
-}
-```
-
-In `src/main.rs`, call the `init_logger()` defined in the `logging.rs`
-sub-module at the head of your `main()` function:
-
-```src/main.rs
-mod logging;
-
-fn main() {
- logging::init_logger();
- …
-}
-```
-
-Now you can call `info!()`, `error!()`, `warn!()`, etc., as described in <https://docs.rs/log/0.4/log/>.
-
-If you also debug it in your native target, you can also provide the `RUST_LOG`
-environment variable, as per <https://docs.rs/env_logger/0.6/env_logger/>'s
-documentation, in your command line incantations:
-
-```
-$ RUST_LOG=DEBUG cargo run
-```
-
-## Sequentialize SPECS.
-
-For some more esoteric reasons, probably something to do with threads, I had to
-rewrite how I run my `specs::System`s , and
-how `specs::System`s written.
-
-### Dispatching `specs::Dispatcher`.
-
-One normally builds a dependency graph of `specs::System`s using something
-like:
-
-```src/main.rs
-fn make_specs_dispatcher() -> specs::Dispatcher<'static, 'static> {
- specs::DispatcherBuilder::new()
- .with(
- SystemFoo,
- "system_foo",
- &[],
- )
- .with(
- SystemBar,
- "system_bar",
- &["system_foo"],
- )
- .build()
-}
-
-struct OurGameState {
- specs_world: specs::World,
- specs_dispatcher: specs::Dispatcher,
-}
-
-
-impl State for OurGameState {
- fn new() -> Result<World> {
- let specs_world = make_specs_world_and_register_components(); // Implemented elsewhere…
- let specs_dispatcher = make_specs_dispatcher();
-
- Ok(
- OurGameState {
- specs_world,
- specs_dispatcher,
- }
- )
- }
-
- fn update(&mut self, window: &mut Window) -> Result<()> {
- let system_foo = SystemFoo;
- let system_bar = SystemBar;
-
- system_foo.run_now(&selfworld.res);
- system_bar.run_now(&world.res);
-
- world.maintain();
-
- Ok(())
- }
-
- [imagine the rest of the quicksilver::lifecycle::State methods implemented here…]
-}
-```
-
-In this example `SystemBar` depends on the state of the `specs::World` left by
-`SystemFoo` after it does its thing.
-
-Instead of using this `Dispatcher` as described in
-<https://slide-rs.github.io/specs/03_dispatcher.html>, you do this in your
-`quicksilver::lifecycle::State::update()`
-
-```
-struct OurGameState {
- specs_world: specs::World,
-}
-
-impl State for OurGameState {
- fn new() -> Result<World> {
- let specs_world = make_specs_world_and_register_components(); // Implemented elsewhere…
-
- Ok(
- OurGameState {
- specs_world,
- }
- )
- }
-
- fn update(&mut self, window: &mut Window) -> Result<()> {
- let system_foo = SystemFoo;
- let system_bar = SystemBar;
-
- system_foo.run_now(&selfworld.res);
- system_bar.run_now(&world.res);
-
- world.maintain();
-
- Ok(())
- }
-
- [imagine the rest of the quicksilver::lifecycle::State methods implemented here…]
-}
-```
-
-But in order to sequentialize how you deal with `specs`, you'd need to change one more thing:
-
-### Lay off `specs::LazyUpdater`.
-
-If you do anything with `specs::LazyUpdate`, you would have to convert it into
-another form, interacting with your `Component` `Storage`s directly with
-`WriteStorage` or whatever.
diff --git a/content/pages/Commands.md b/content/pages/Commands.md
@@ -1,3 +1,3 @@
-Commands I infrequently need.
+title: Commands I infrequently need.
Dump streams, vlecs (personal use, of course), videos - mplayer -dumpfile [file] -dumpstream [url]
diff --git a/content/pages/Python-versus-c.md b/content/pages/Python-versus-c.md
@@ -0,0 +1,5 @@
+title: Python versus C.
+
+## Modulus
+
+[C/Python different behaviour of the modulo operation](http://stackoverflow.com/questions/1907565/c-python-different-behaviour-of-the-modulo-operation)
diff --git a/content/pages/Python-versus-c.org b/content/pages/Python-versus-c.org
@@ -1,3 +0,0 @@
-** Modulus
-
-[[http://stackoverflow.com/questions/1907565/c-python-different-behaviour-of-the-modulo-operation][C/Python different behaviour of the modulo operation]]
-\ No newline at end of file
diff --git a/content/pages/Python.md b/content/pages/Python.md
@@ -0,0 +1,14 @@
+title: Python.
+
+# Python
+
+## Getting a dictionary of the attribute name and attribute value pairs
+
+```python
+class a:
+ b = 'b'
+ def __init__(self, c):
+ self.c = c
+ def who(self):
+ return(scipy.who(vars(self)))
+```
diff --git a/content/pages/Python.org b/content/pages/Python.org
@@ -1,11 +0,0 @@
-* Python
-** Getting a dictionary of the attribute name and attribute value pairs
-
-#+BEGIN_SRC python
-class a:
- b = 'b'
- def __init__(self, c):
- self.c = c
- def who(self):
- return(scipy.who(vars(self)))
-#+END_SRC
diff --git a/content/pages/Vim.md b/content/pages/Vim.md
@@ -1,3 +1,5 @@
+title: Vim.
+
How many lines / Count marked lines
"In visual mode, press g C-g."
diff --git a/content/pages/Words.md b/content/pages/Words.md
@@ -1,3 +1,5 @@
+title: Words on the tip of my tongue.
+
# Words on the tip of my tongue #
* Rhetoric
diff --git a/content/pages/weave.md b/content/pages/weave.md
@@ -1,3 +1,6 @@
+title: Weave stuff.
+
+
# Accessing an array and its metadata through C/C++ in *scipy.weave.inline()*
* Accessing an array *foo* with *foo.ndim == 3*:
