2026-07-26 - 2026-08-02
Overview
7 Releases published by 1 user
Published
v0.3.36
v0.3.36
Published
v0.3.35
v0.3.35
Published
v0.3.34
v0.3.34
Published
v0.3.33
v0.3.33
Published
v0.3.32
v0.3.32
Published
v0.3.31
v0.3.31
Published
v0.3.30
v0.3.30
15 Pull requests merged by 1 user
Merged
#91 verify is what tells you which keys to purge
Merged
#93 fix(project): a module resref names a file, not a resource
Merged
#90 Purge the edge after a repair, or verify answers per PoP (#89)
Merged
#87 fix(nwsync): declare Frame_Content_Size on every blob, and verify what is published
Merged
#84 fix(module): run scripts/fetch-upstream-manifests
Merged
#82 fix(nwsync): write emit and assemble summaries to stderr
Merged
#80 feat(nwsync): emit blobs in parallel with a bounded worker pool
Merged
#78 fix(nwsync): stream emit so peak memory tracks the largest resource (#76)
Merged
#77 docs: describe the on-disk shape of an emitted NWSync tree
Merged
#73 nwsync: upload sink, key-addressed CLI, fail-closed publication marker
Merged
#71 feat(nwsync): emit blobs and per-artifact NSYM, assemble merged manifests
Merged
#70 feat(ci): prune release assets outside the newest two tags (#52)
Merged
#69 fix(erf): align restype numbers with upstream neverwinter.nim
Merged
#68 docs(agents): note that tea reads stdin and hangs without </dev/null
Merged
#67 depot: replace --target local with --out DIR
26 Issues closed from 1 user
Closed
#54 Map: NWSync publishing — emit at artifact birth, assemble at module release
Closed
#75 Acceptance gate: a real NWN client syncs from an assembled manifest on the Bunny zone
Closed
#88 Run the verifying backfill: every blob published by emitter 1 is still unreadable by a client
Closed
#89 Replacing a blob does not purge the pull-zone cache, so verify can pass while clients still get the bad bytes
Closed
#85 Nothing verifies a published NWSync blob except a player's client
Closed
#86 Blobs under 256 bytes omit the zstd frame content size, and no client can sync past one
Closed
#83 Flip the module build's project-script name to fetch-upstream-manifests
Closed
#81 crucible nwsync emit/assemble write their summary to stdout, corrupting any caller that captures it
Closed
#79 nwsync emit: upload blobs in parallel with a bounded worker count
Closed
#76 nwsync emit holds ~5x the artifact size in RAM; OOM-killed on haks over ~1.4 GB
Closed
#53 crucible nwsync: emit blobs and per-artifact NSYM, assemble the merged manifest
Closed
#74 Provision the NWSync write credential on the two emitter repos
Closed
#61 Bunny write credential for sow-module CI
Closed
#72 Correct the platform record: assemble runs in sow-module release CI
Closed
#60 The emit upload path and NWCompressedBuffer round-trip
Closed
#59 Run upstream nwn_nwsync_write and capture reference output for one real hak
Closed
#65 Wire the consumer repos to the settled crucible nwsync interface
Closed
#52 Prune old Crucible release assets in the release workflow
Closed
#64 Five restype numbers in internal/erf disagree with upstream
Closed
#66 depot: replace --target local with --out DIR, matching crucible nwsync
Closed
#63 A module release does not record which topdata it shipped against
Closed
#56 Where the nwsync verbs live in the command surface
Closed
#62 Where the per-artifact NSYM lives, now that topdata is not a hak
Closed
#57 Restype resolution for emit: what Crucible has, what upstream's table says
Closed
#55 Bunny zone: how a blob is uploaded and how existence is checked without mounting
Closed
#58 Does sow-module CI know the full artifact set at release time?
26 Issues created by 1 user
Opened
#53 crucible nwsync: emit blobs and per-artifact NSYM, assemble the merged manifest
Opened
#54 Map: NWSync publishing — emit at artifact birth, assemble at module release
Opened
#56 Where the nwsync verbs live in the command surface
Opened
#55 Bunny zone: how a blob is uploaded and how existence is checked without mounting
Opened
#57 Restype resolution for emit: what Crucible has, what upstream's table says
Opened
#58 Does sow-module CI know the full artifact set at release time?
Opened
#59 Run upstream nwn_nwsync_write and capture reference output for one real hak
Opened
#61 Bunny write credential for sow-module CI
Opened
#60 The emit upload path and NWCompressedBuffer round-trip
Opened
#62 Where the per-artifact NSYM lives, now that topdata is not a hak
Opened
#64 Five restype numbers in internal/erf disagree with upstream
Opened
#63 A module release does not record which topdata it shipped against
Opened
#65 Wire the consumer repos to the settled crucible nwsync interface
Opened
#66 depot: replace --target local with --out DIR, matching crucible nwsync
Opened
#72 Correct the platform record: assemble runs in sow-module release CI
Opened
#74 Provision the NWSync write credential on the two emitter repos
Opened
#75 Acceptance gate: a real NWN client syncs from an assembled manifest on the Bunny zone
Opened
#76 nwsync emit holds ~5x the artifact size in RAM; OOM-killed on haks over ~1.4 GB
Opened
#79 nwsync emit: upload blobs in parallel with a bounded worker count
Opened
#81 crucible nwsync emit/assemble write their summary to stdout, corrupting any caller that captures it
Opened
#83 Flip the module build's project-script name to fetch-upstream-manifests
Opened
#85 Nothing verifies a published NWSync blob except a player's client
Opened
#86 Blobs under 256 bytes omit the zstd frame content size, and no client can sync past one
Opened
#88 Run the verifying backfill: every blob published by emitter 1 is still unreadable by a client
Opened
#89 Replacing a blob does not purge the pull-zone cache, so verify can pass while clients still get the bad bytes
Opened
#92 Spot-check hak precedence in-world: the 13 duplicate resrefs and sow_top's 2DAs
1 Unresolved Conversation
Open
#29
Cosmetic: Rename nwn-tool.yaml instances to crucible.yaml