memory2 tf service#2707
Open
leshy wants to merge 11 commits into
Open
Conversation
Unify replay tf with the live service: StreamTF(MultiTBuffer, TFSpec) mirrors PubSubTF, pulling windows from a recorded tf stream on demand instead of receiving pushed messages. Lookups span buffer_size backward (or an explicit time_tolerance) plus forward_tolerance ahead; a cache miss prefetches cache_span past the query window and evicts everything first, so chronological replay costs one db query per cache_span. - TFLookup protocol (read side) + mypy conformance checks - get_pose hoisted from PubSubTF to TFSpec - MultiTBuffer: None tolerance resolves to buffer_size explicitly - map global: registration via tf stream (never Observation poses), --frame auto-detects world/map/odom via probe lookups, fail-fast when the cloud frame can't be resolved - tf lookup tests consolidated into a grid: live MultiTBuffer vs StreamTF over memory/sqlite stores
|
Preview deployment for your docs. Learn more about Mintlify Previews.
💡 Tip: Enable Workflows to automatically generate PRs for you. |
Codecov Report❌ Patch coverage is @@ Coverage Diff @@
## main #2707 +/- ##
==========================================
+ Coverage 72.05% 72.07% +0.01%
==========================================
Files 893 894 +1
Lines 80728 80867 +139
Branches 7310 7332 +22
==========================================
+ Hits 58165 58281 +116
- Misses 20614 20628 +14
- Partials 1949 1958 +9
Flags with carried forward coverage won't be shown. Click here to find out more.
... and 3 files with indirect coverage changes 🚀 New features to boost your workflow:
|
leshy
requested review from
mustafab0,
paul-nechifor and
spomichter
as code owners
Contributor
Greptile SummaryThis PR introduces
Confidence Score: 4/5Safe to merge for recordings where the tf stream has no temporal gaps; recordings with tf gaps during flight may silently produce holes in the accumulated map due to the dedup/registration inconsistency. The dimos/mapping/utils/cli/map.py — specifically the interaction between Important Files Changed
Sequence Diagram%%{init: {'theme': 'neutral'}}%%
sequenceDiagram
participant CLI as map.py (main)
participant STF as StreamTF
participant Store as Stream[TFMessage]
participant Acc as _accumulate
CLI->>Store: stream(lidar_stream, PointCloud2)
CLI->>STF: StreamTF.from_store(store)
CLI->>STF: get(world, cloud_frame, first_obs.ts) [probe]
STF->>STF: _ensure(tp-back, tp+fwd)
STF->>Store: stream.at(center, radius)
Store-->>STF: observations with TFMessages
STF->>STF: "receive_transform(*transforms)"
STF-->>CLI: Transform or None [probe result]
loop dedup frames
CLI->>STF: _position(obs) → tf lookup
STF-->>CLI: (x,y,z) or None
Note over CLI: fallback to obs.pose if tf=None
end
CLI->>Acc: "_accumulate(kept, register=_register)"
loop kept frames
Acc->>STF: register(obs) → buf.get(world, frame, ts)
STF-->>Acc: Transform or None
alt Transform available
Acc->>Acc: apply transform to cloud
else None
Acc->>Acc: skip frame (continue)
end
end
Acc-->>CLI: PointCloud2 (global map)
%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
sequenceDiagram
participant CLI as map.py (main)
participant STF as StreamTF
participant Store as Stream[TFMessage]
participant Acc as _accumulate
CLI->>Store: stream(lidar_stream, PointCloud2)
CLI->>STF: StreamTF.from_store(store)
CLI->>STF: get(world, cloud_frame, first_obs.ts) [probe]
STF->>STF: _ensure(tp-back, tp+fwd)
STF->>Store: stream.at(center, radius)
Store-->>STF: observations with TFMessages
STF->>STF: "receive_transform(*transforms)"
STF-->>CLI: Transform or None [probe result]
loop dedup frames
CLI->>STF: _position(obs) → tf lookup
STF-->>CLI: (x,y,z) or None
Note over CLI: fallback to obs.pose if tf=None
end
CLI->>Acc: "_accumulate(kept, register=_register)"
loop kept frames
Acc->>STF: register(obs) → buf.get(world, frame, ts)
STF-->>Acc: Transform or None
alt Transform available
Acc->>Acc: apply transform to cloud
else None
Acc->>Acc: skip frame (continue)
end
end
Acc-->>CLI: PointCloud2 (global map)
Reviews (2): Last reviewed commit: "Merge branch 'feat/ivan/memtf' of github..." | Re-trigger Greptile |
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
| if stream is not None: | ||
| kwargs["stream"] = stream | ||
| TFSpec.__init__(self, **kwargs) | ||
| MultiTBuffer.__init__(self, buffer_size=math.inf) |
Contributor
There was a problem hiding this comment.
Why infinity? Is it wise to never evict? Was this meant to be `self.config.cache_span?
| def _load(self, lo: float, hi: float) -> None: | ||
| for obs in self.stream.at((lo + hi) / 2, (hi - lo) / 2): | ||
| self.receive_transform(*obs.data.transforms) | ||
| self._covered = (lo, hi) |
Contributor
There was a problem hiding this comment.
Don't you need to use self._cv for modifying self._covered too?
|
|
||
| if TYPE_CHECKING: | ||
| # mypy conformance check: StreamTF satisfies the read-side tf protocol. | ||
| _lookup_impl: TFLookup = cast("StreamTF", None) |
Contributor
There was a problem hiding this comment.
Is this used anywhere?
|
|
||
| if TYPE_CHECKING: | ||
| # mypy conformance checks: the live services satisfy the read-side protocol. | ||
| _lookup_impls: tuple[type[TFLookup], ...] = (MultiTBuffer, PubSubTF) |
Contributor
There was a problem hiding this comment.
Is this used anywhere?
paul-nechifor
approved these changes
Jul 3, 2026
github-actions
Bot
added
the
ready-to-merge
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
we converted recordings to sensor frame, mem now needs a tf service for global mapping