* On the web, load different maps by making asynchronous HTTP calls. #21
This is a very strange, specializd approach; it's unclear how it'll
generalize to reading scenarios and prebaked results. Ideally we could
call abstutil::read_binary as we currently do and somehow hide this async
trickery underneath, but I'm not sure how yet. In the meantime, this
moves us forward with the hack well-contained.
Next steps for web: stop bundling in all of data/system in the .wasm,
now that we can load from HTTP.
imported, because they referenced way IDs from before the service road
import. That happened after a bad Cargo.lock merge undid the effects of
pinning to the latest seattle_traffic_signals.
When I added web support in February, it was easier to get started with
stdweb, since it has the nice cargo web tool. However, stdweb is
unmaintained, winit is deprecating support for it, and the next steps
for web (downloading maps dynamically) have better support for web-sys.
With Alvin's guidance, I got
https://github.com/dabreegster/minimal_websys_winit_glow_demo working
first. This PR cuts A/B Street over too.
I tested abst and the widgetry demo in both native and web. The only
major regression from stdweb is the canvas placement and size. I
attempted some fixes, but at this point, I'll leave it as a smaller
followup instead.
https://github.com/koute/cargo-web/issues/217. After adding some
debugging to that tool, discovered the problem was generic-array, pulled
in by the new geo version. Forcing that to the older version fixes
things.
Just a few changes to make usvg not depend on Rustybuzz (which is still
in the middle of becoming a pure Rust library). It probably only works
with English fonts, but for the moment, it moves us forward.
uber-turns (sequences of turns through a few intersections) due to OSM
turn restrictions, we have to be a little careful how we sum up the cost
for the entire sequence, only rounding at the end.
proles: account for work capacity
previously all residents went to work on the map, but if the
neighborhood is mostly residential that meant they were "competing"
heavily for the scarce jobs available on the map.
with this commit, we do a better job of spreading worker/residential
demand to account for what is available, and fill in the gap with
off-map trip origins/desitinations.
e.g. if a neighborhood is mostly residential, we have folks commute off
of map. If a neighborhood is mostly commercial, we have folks commute
into the map.
* Read ImporterConfiguration from .toml file if present. Also check for presence of dependency programs.
* Actually use config.osmconvert instead of hard-coded 'osmconvert'
* Add curl to checked dependencies
* Don't require all fields to be present in importer.toml configuration. Fill in defaults.
* Remove unneeded slice
* Use json for importer config and use serde(default) attribute to simplify application of defaults
* Extract re_dependencies_callable from main.rs and remove dependency on Job
* Revert change to abstutil, use struct-level default attribute
* Add importer.json to .gitignore
* Make gunzip_args string instead of Option<string>
85ad194d introduced a patched winit.
In the glium build, we were pulling in a second version of winit (the
unpatched version, via glutin). This caused conflicts between the data
types, breaking the build.
Apparently the proper way to specify a patch like this, usable by
transitive dependencies, is via cargo's "patch" specifiers
combining primitive transitions into sequences.
Brief context on the state/transition system: The game crate is
organized as a stack of states, with the topmost one being active.
Transitions manipulate this stack. For example, the stack might look
like: [main menu, sandbox mode, edit mode, traffic signal editor, signal
picker]
when allowing a car to start a turn. It causes
https://github.com/dabreegster/abstreet/pull/276#discussion_r470269394
and also the lakeslice scenario to gridlock (a regression that began a
few weeks ago). But keep the flag on for now, to keep the montlake
scenaro running at least.
https://dabreegster.github.io/abstreet/trafficsim/gridlock.html has
notes about the many different causes and in-progress fixes for
gridlock. This experiment hasn't been explained very well yet, but
roughly it treats a cluster of traffic signals as one, so that once a
vehicle gains access through the first light, they guarantee immediate
access through the entire sequence. This interacts with the "don't block
the box" behavior (don't start a turn if you might get stuck in the
intersection) strangely.
While attempting to get this rollback to work, I also had to manually
redraw the traffic lights for a few manually specified intersections.
They became out-of-date a few weeks ago when I cleaned up the OSM
geometry upstream and the referenced IDs changed, and I hadn't bothered
to re-time the signals. Luckily, with the new multi-signal editor,
redrawing the timing was much easier than originally!
Regenerated all data and lots of bus routes vanished. Plan to get back
to that project soon.
