v1.4.0 — how much is left to dig

How much is left to dig

Break counters answer who dug the most. They were never going to answer how much is left: explosions and world edits are not credited to any player, so on a TNT-driven perimeter the counters miss almost everything.

1.4.0 answers it directly by counting the blocks that are still standing inside the shape.

/shapeboard scan bigculo
/shapeboard progress

— Big Culo progress —
█░░░░░░░░░░░░░░░░░░░ 6.96%
Progress: 6,630,915 / 95,290,463 blocks cleared, 88,659,548 remaining
Layers: 0 of 198 layers cleared
Left to dig: deepslate 37,376,792, stone 32,912,885, tuff 3,131,119

/shapeboard progress with no arguments uses the zone you are standing in, so anyone can check without knowing the shape id.

Per layer

/shapeboard layer 63

— Big Culo · layer y63 —
██░░░░░░░░░░░░░░░░░░ 11.00%
Layer y63: 479,739 of 539,012 blocks remaining

With no Y it uses the layer you are standing on. Layers are measured against what the layer actually started with, not the raw column count: y63 above began with 539,012 blocks because part of it was already air and water.

What counts

Only solid, collidable, breakable blocks. Air, water, lava, leaves, grass, decoration and bedrock are skipped, so the bar reaches 100% when the hole is genuinely finished. The default Y range now starts above the bedrock floor, since those bottom layers can never be emptied.

The denominator

By default the percentage is measured against the raw volume of the Y range. The remaining count is exact, but the percentage is not: natural caves and the air above the terrain count as "already dug", which can easily read as 40% on an untouched perimeter.

For a true reading, measure what was there before anyone touched it. On a throwaway server with the same seed, pregenerate the area, copy your shape in, and run:

/shapeboard scan <id>
/shapeboard baseline <id> fromscan

Then carry that world/shapeboard/shapes.json to the live server. It stores the real starting count, per layer included. On our own perimeter this took the reading from a meaningless 43% to the actual 7%.

Cost

The scan runs on a worker thread and reads chunk data straight from storage, so the server keeps ticking. A 797,680 column perimeter (3,298 chunks) scans in about 2 seconds. It saves the world first so freshly dug chunks are always included. Run it from cron and /shapeboard progress answers instantly from the stored snapshot.

Commands added

Command	Who	What
/shapeboard scan <id>	OP	Count the blocks still standing inside the shape
/shapeboard progress [id]	all	Progress bar and remaining count from the last scan
/shapeboard layer [y] [id]	all	One Y layer, against what that layer started with
/shapeboard range <id> <ymin> <ymax>	OP	Y slice the scan covers
/shapeboard baseline <id> <blocks>|fromscan	OP	The real starting count

Which jar

Minecraft	Jar
1.21, 1.21.1	shapeboard-1.4.0+1.21.jar
1.21.2	shapeboard-1.4.0+1.21.2.jar
1.21.3	shapeboard-1.4.0+1.21.3.jar
1.21.4	shapeboard-1.4.0+1.21.4.jar
1.21.5	shapeboard-1.4.0+1.21.5.jar
1.21.6	shapeboard-1.4.0+1.21.6.jar
1.21.7	shapeboard-1.4.0+1.21.7.jar
1.21.8	shapeboard-1.4.0+1.21.8.jar

Fully additive: 1.3.0 shapes keep working, nothing to migrate.

Full Changelog: v1.3.0...v1.4.0