[PR #1059] [MERGED] Make Pregel loop runnable step-by-step #1862

Closed
opened 2026-02-20 17:45:43 -05:00 by yindo · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/langchain-ai/langgraph/pull/1059
Author: @nfcampos
Created: 7/18/2024
Status: Merged
Merged: 7/19/2024
Merged by: @nfcampos

Base: mainHead: nc/18jul/loop


📝 Commits (10+)

📊 Changes

14 files changed (+1193 additions, -967 deletions)

View changed files

📝 libs/langgraph/langgraph/channels/manager.py (+2 -0)
📝 libs/langgraph/langgraph/checkpoint/base.py (+13 -8)
📝 libs/langgraph/langgraph/constants.py (+2 -1)
📝 libs/langgraph/langgraph/pregel/__init__.py (+135 -939)
libs/langgraph/langgraph/pregel/algo.py (+427 -0)
📝 libs/langgraph/langgraph/pregel/executor.py (+3 -2)
libs/langgraph/langgraph/pregel/loop.py (+423 -0)
📝 libs/langgraph/langgraph/pregel/types.py (+9 -0)
📝 libs/langgraph/poetry.lock (+39 -4)
📝 libs/langgraph/pyproject.toml (+2 -1)
📝 libs/langgraph/tests/__snapshots__/test_pregel.ambr (+2 -2)
📝 libs/langgraph/tests/test_prebuilt.py (+109 -3)
📝 libs/langgraph/tests/test_pregel.py (+15 -5)
📝 libs/langgraph/tests/test_pregel_async.py (+12 -2)

📄 Description

This lays the groundwork for lots of new ways to run graphs, eg

  • run nodes in different machines
  • place hard rate limits on particular nodes globally across all runs
  • run loop as the target for a queue topic subscription (eg. event driven architecture)

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/langchain-ai/langgraph/pull/1059 **Author:** [@nfcampos](https://github.com/nfcampos) **Created:** 7/18/2024 **Status:** ✅ Merged **Merged:** 7/19/2024 **Merged by:** [@nfcampos](https://github.com/nfcampos) **Base:** `main` ← **Head:** `nc/18jul/loop` --- ### 📝 Commits (10+) - [`5623803`](https://github.com/langchain-ai/langgraph/commit/56238036d760a49ee0608e35b7f666e2f6659e1e) WIP: Split out loop logic from Pregel class - [`60f7a7d`](https://github.com/langchain-ai/langgraph/commit/60f7a7d593d88bc55d0694dda654745a180a81d0) Lint - [`b3ee728`](https://github.com/langchain-ai/langgraph/commit/b3ee7288396d58ea7d18c74ef793c2c7b8b2d136) Pregel.stream passing all tests - [`06f8371`](https://github.com/langchain-ai/langgraph/commit/06f83710c8bce3437dde99f704202469a10f9f78) Lint - [`73b4b6e`](https://github.com/langchain-ai/langgraph/commit/73b4b6ec5b38526d5e993cdbe05bcfc4bb81ff9e) Lint - [`909bf44`](https://github.com/langchain-ai/langgraph/commit/909bf4433fc508ab4ffb918c3df502b854ccae4d) Lint - [`f974471`](https://github.com/langchain-ai/langgraph/commit/f974471f7d031164f9b09896bb8e92432a9a1498) Lint - [`7ed5f9e`](https://github.com/langchain-ai/langgraph/commit/7ed5f9e4bcdb713f97e87f26fd8027596a717b29) Implement AsyncLoop - [`29e860a`](https://github.com/langchain-ai/langgraph/commit/29e860acca3c2cf67a21f8e7cbfb5f3cc3b3c004) Update manager.py - [`ed5d114`](https://github.com/langchain-ai/langgraph/commit/ed5d114087c5c9817e02310f45262e08723cdc00) Fix ### 📊 Changes **14 files changed** (+1193 additions, -967 deletions) <details> <summary>View changed files</summary> 📝 `libs/langgraph/langgraph/channels/manager.py` (+2 -0) 📝 `libs/langgraph/langgraph/checkpoint/base.py` (+13 -8) 📝 `libs/langgraph/langgraph/constants.py` (+2 -1) 📝 `libs/langgraph/langgraph/pregel/__init__.py` (+135 -939) ➕ `libs/langgraph/langgraph/pregel/algo.py` (+427 -0) 📝 `libs/langgraph/langgraph/pregel/executor.py` (+3 -2) ➕ `libs/langgraph/langgraph/pregel/loop.py` (+423 -0) 📝 `libs/langgraph/langgraph/pregel/types.py` (+9 -0) 📝 `libs/langgraph/poetry.lock` (+39 -4) 📝 `libs/langgraph/pyproject.toml` (+2 -1) 📝 `libs/langgraph/tests/__snapshots__/test_pregel.ambr` (+2 -2) 📝 `libs/langgraph/tests/test_prebuilt.py` (+109 -3) 📝 `libs/langgraph/tests/test_pregel.py` (+15 -5) 📝 `libs/langgraph/tests/test_pregel_async.py` (+12 -2) </details> ### 📄 Description This lays the groundwork for lots of new ways to run graphs, eg - run nodes in different machines - place hard rate limits on particular nodes globally across all runs - run loop as the target for a queue topic subscription (eg. event driven architecture) --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
yindo added the pull-request label 2026-02-20 17:45:43 -05:00
yindo closed this issue 2026-02-20 17:45:43 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/langgraph#1862