@tscircuit/fanout-solver 0.0.17 → 0.0.19

package/README.md

@@ -42,6 +42,11 @@ and treats each bus-layer decision atomically.
   alternate instead of favoring one axis; square grids distribute equally
   across north, south, east, and west.
 - Enumerates combinations of the copper layers implied by `layerCount`.
+- Keeps a bounded beam of route alternatives for multi-connection buses, so
+  grouped power/signal lanes can backtrack across layer and track choices
+  before committing a prefix.
+- Keys route-prefix caches by both bus and layer, preserving plan uniqueness
+  when grouped-layer search changes bus order.
 - Prefers depth-cycled layer assignments: matching north/south (or east/west)
   bus depths share a layer, and deeper pairs cycle through every available
   escape layer. This forces a small stackup to reuse routing channels.
@@ -61,9 +66,17 @@ and treats each bus-layer decision atomically.
   breakout corridor.
 - Chamfers orthogonal routing corners into 45° segments before validating and
   emitting the fanout.
-- Verifies pad, via, trace, and already-routed fanout clearance.
-- Treats an obstacle whose `connectedTo` list contains the connection name as
-  electrically connected copper rather than a foreign keepout.
+- Verifies oriented-pad, via, trace, and already-routed fanout clearance on
+  every complete candidate, independent of the routing strategy that produced
+  it.
+- Resolves `netConnectionName`, connection, port, trace, and obstacle metadata
+  into electrical-net identities. Same-net copper may merge; different-net
+  pads, traces, and vias must retain clearance on every layer they occupy.
+- `allowSameNetMerges` lets grouped branches such as VCC or GND reuse connected
+  copper instead of reserving artificial clearance from one another. It is
+  opt-in; different electrical nets remain hard obstacles.
+- Audits route continuity, unique connection coverage, boundary exits, and
+  retained downstream endpoints before marking a solution complete.
 - Emits supplied fanout traces, via obstacles, and moved breakout endpoints in a
   new `SimpleRouteJson`. The returned problem is ready for a downstream
   autorouter to finish.
@@ -177,6 +190,8 @@ bus-layer combination search.
 - `busLayerAssignments`: the selected layer for every bus
 - `busDirections`: the direction shared by each bus
 - `attempts`: score and success metadata for every tried layer combination
+- `validation`: the final geometry/connectivity report, including the number of
+  independently validated breakouts
 
 ## Dataset 01
 
@@ -198,15 +213,18 @@ parameters. Each sample has one shared boundary around all of its footprints,
 and component bounds come from the exact footprinter-generated copper pad
 extents.
 
-## SRJ19 benchmark
+## SRJ29 benchmark
 
-The repository also loads all 200 samples from
-[`tscircuit/dataset-srj19`](https://github.com/tscircuit/dataset-srj19) as a
-pinned development dependency. The adapter keeps the complete obstacle field
-but selects only connections that touch the BGA, producing progressively larger
-fanout problems with opposite-side passive overlays. Every adapted problem uses
-the same six-layer stackup (`top`, `inner1` through `inner4`, and `bottom`) so
-benchmark improvements are directly comparable.
+The repository loads all 200 samples from the derivative
+[`tscircuit/dataset-srj29-bga-decoupling`](https://github.com/tscircuit/dataset-srj29-bga-decoupling)
+as a pinned development dependency. The adapter keeps the complete obstacle
+field, including opposite-layer capacitor pads and bodies. VCC and GND are
+grouped onto opposite boundary corridors, while the capacitor pad remains the
+downstream endpoint of every power connection; a local capacitor or plane via
+alone cannot count as a solved BGA pin. Remaining edge signals are grouped by
+direction. Every adapted problem uses the same six-layer stackup (`top`,
+`inner1` through `inner4`, and `bottom`) so benchmark improvements are directly
+comparable.
 
 Run the full benchmark with:
 
@@ -219,22 +237,23 @@ Use `--sample sample001`, `--limit 10`, or
 default and print progress as they finish. `--concurrency 8` runs isolated
 samples in parallel, and `--sample-timeout-seconds 600` prevents a difficult
 sample from blocking the remaining work. Each run writes the full ordered
-results to `benchmark-results/srj19.json` and
-`benchmark-results/srj19.md`. Partial solutions are reported as benchmark
-results instead of failing the command, making current completion rates a
-baseline for solver improvements. `bun run benchmark:srj19` is an alias for the
-same command.
-
-The `SRJ19 Benchmark` GitHub Actions workflow runs the complete dataset on a
+results to `benchmark-results/srj29.json` and
+`benchmark-results/srj29.md`. A row is marked solved only when every input
+connection has a validated breakout, all downstream endpoints remain attached,
+and the independent DRC audit passes. Partial solutions are reported as
+benchmark results instead of failing the command, making current completion
+rates a baseline for solver improvements. `bun run benchmark:srj29` is an alias
+for the same command.
+
+The `SRJ29 Benchmark` GitHub Actions workflow runs the complete dataset on a
 Blacksmith 32-vCPU ARM runner with 32 sample processes by default. It can be
 started manually with an optional sample id, or for a pull request by adding
 `[BENCHMARK TEST]` to its title. The workflow publishes the Markdown summary and
 uploads both reports as an artifact.
 
-Run `bun run start` and open the `datasets/srj19` Cosmos fixture to step the
-selected sample through `GenericSolverDebugger`. The page has Previous/Next,
-sample dropdown, and range controls and stores the selection in the `sample`
-URL parameter.
+Run `bun run start` and inspect the SRJ29 fixtures to step through the selected
+sample. The derivative dataset also publishes dedicated Cosmos pages for the
+first ten samples.
 
 ## Dataset 02