tolquane 1.3.0__tar.gz

tolquane-1.3.0/.gitignore

@@ -0,0 +1,38 @@
+# Python
+__pycache__/
+*.py[cod]
+*.egg-info/
+build/
+dist/
+.eggs/
+
+# Environments and secrets
+.env
+keys.env
+.venv*/
+venv/
+.python-version
+
+# Tooling caches
+.pytest_cache/
+.mypy_cache/
+.ruff_cache/
+.coverage
+htmlcov/
+
+# Editors and OS
+.idea/
+.vscode/
+.DS_Store
+site/
+
+# Tolquane Web (frontend)
+web/node_modules/
+src/tolquane/web/static/
+web/dist/
+web/test-results/
+web/playwright-report/
+web/e2e/screenshots/
+# The workspace a Playwright run seeds and throws away.
+web/e2e/.workspace/
+.claude/worktrees/

tolquane-1.3.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

tolquane-1.3.0/PKG-INFO

@@ -0,0 +1,185 @@
+Metadata-Version: 2.5
+Name: tolquane
+Version: 1.3.0
+Summary: Parallel programming with composable building blocks: nodes, pipelines, farms and all-to-all, on threads, processes or a network.
+Project-URL: Homepage, https://github.com/robtacconelli/Tolquane
+Project-URL: Repository, https://github.com/robtacconelli/Tolquane
+Project-URL: Issues, https://github.com/robtacconelli/Tolquane/issues
+Project-URL: Design, https://github.com/robtacconelli/Tolquane/blob/main/DESIGN.md
+Project-URL: Changelog, https://github.com/robtacconelli/Tolquane/blob/main/CHANGELOG.md
+Author: Roberto Tacconelli
+License-Expression: Apache-2.0
+License-File: LICENSE
+Keywords: building-blocks,dataflow,farm,fastflow,parallel,pipeline,streaming
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Topic :: Software Development :: Libraries
+Classifier: Topic :: System :: Distributed Computing
+Classifier: Typing :: Typed
+Requires-Python: >=3.11
+Provides-Extra: ai
+Requires-Dist: anthropic>=1.0; extra == 'ai'
+Requires-Dist: openai>=2.0; extra == 'ai'
+Provides-Extra: cloudpickle
+Requires-Dist: cloudpickle>=3.0; extra == 'cloudpickle'
+Provides-Extra: dev
+Requires-Dist: anthropic>=1.0; extra == 'dev'
+Requires-Dist: cloudpickle>=3.0; extra == 'dev'
+Requires-Dist: fastapi>=0.115; extra == 'dev'
+Requires-Dist: httpx>=0.27; extra == 'dev'
+Requires-Dist: mkdocs-material>=9.5; extra == 'dev'
+Requires-Dist: mkdocstrings[python]>=0.26; extra == 'dev'
+Requires-Dist: mypy>=1.11; extra == 'dev'
+Requires-Dist: openai>=2.0; extra == 'dev'
+Requires-Dist: pytest-timeout>=2.3; extra == 'dev'
+Requires-Dist: pytest>=8.3; extra == 'dev'
+Requires-Dist: ruff>=0.6; extra == 'dev'
+Requires-Dist: uvicorn[standard]>=0.30; extra == 'dev'
+Provides-Extra: docs
+Requires-Dist: mkdocs-material>=9.5; extra == 'docs'
+Requires-Dist: mkdocstrings[python]>=0.26; extra == 'docs'
+Provides-Extra: msgpack
+Requires-Dist: msgpack>=1.0; extra == 'msgpack'
+Provides-Extra: numpy
+Requires-Dist: numpy>=1.26; extra == 'numpy'
+Provides-Extra: processes
+Requires-Dist: cloudpickle>=3.0; extra == 'processes'
+Provides-Extra: web
+Requires-Dist: fastapi>=0.115; extra == 'web'
+Requires-Dist: uvicorn[standard]>=0.30; extra == 'web'
+Description-Content-Type: text/markdown
+
+# Tolquane
+
+[![CI](https://github.com/robtacconelli/Tolquane/actions/workflows/ci.yml/badge.svg)](https://github.com/robtacconelli/Tolquane/actions/workflows/ci.yml)
+[![License: Apache-2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
+
+Parallel programming with composable building blocks, in Python.
+
+Nodes speak on channels. Pipelines, farms and all-to-all blocks compose them, and the
+same graph runs on threads, processes or across a network. Tolquane is the successor of
+[BBFlow](https://github.com/robtacconelli/BBFlow), a Java implementation of the
+[FastFlow](https://github.com/fastflow/fastflow) building blocks, rebuilt from scratch
+to be simple to use and impossible to hang.
+
+> **1.1.** Nodes, pipelines, farms with every emitter and collector policy and any block
+> as a worker, ordered farms, node fusion, all-to-all, feedback loops that terminate by
+> rule, batching, sessions, deadlock detection, an optimizer that cuts threads; threads,
+> processes, coroutine pools, a distributed runtime over TCP with a one-command launcher,
+> and a deterministic sync runtime; a run report that names the bottleneck; an AI builder
+> that writes, checks and runs flows from a sentence; and Tolquane Web, a local GUI with
+> the blocks on a canvas, live runs and schedules. See [DESIGN.md](DESIGN.md) for the
+> design and the liveness rules, [docs/api-card.md](docs/api-card.md) for the whole API
+> on one page, [examples/](examples/) for flows in the house style, and
+> [CHANGELOG.md](CHANGELOG.md).
+
+## What it looks like
+
+```python
+import tolquane as tq
+
+@tq.source
+def numbers():
+    yield from range(1, 101)          # a source is a generator
+
+@tq.node
+def double(x: int) -> int:
+    return x * 2                       # the return value is sent downstream
+
+@tq.sink
+def show(x: int) -> None:
+    print(x)
+
+graph = numbers >> tq.farm(double, workers=4) >> show
+tq.run(graph)                          # threads by default
+tq.run(graph, runtime="processes")     # same graph, farm workers in child processes
+tq.run(graph, runtime="sync")          # same graph, one thread, deterministic
+tq.run(graph, deploy="deploy.toml", group="G1")   # same graph, this host's share of it
+```
+
+```toml
+# deploy.toml: each host runs `tolquane run flow.py --deploy deploy.toml --group <name>`
+[groups.G1]
+endpoint = "10.0.0.1:7000"
+nodes = ["numbers", "double.emitter", "double.collector", "show"]
+[groups.G2]
+endpoint = "10.0.0.2:7000"
+nodes = ["double.[0-9]*"]           # the workers, on the other machine
+```
+
+A function is a node. Return `tq.SKIP` to drop an item; `None` is an ordinary value.
+A generator function yields zero or many items. Farms come with round-robin, broadcast,
+scatter, on-demand and key-based emitters, first-come, round-robin, gather and ordered
+collectors. `tq.all2all` joins two farms worker to worker, `tq.feedback` wires a block
+back onto itself with a loop that closes when nothing is left in flight, and
+`tq.session` keeps a graph running while you push items in and read results out.
+
+## Runtimes
+
+| Runtime | Use when |
+|---|---|
+| `sync` (available) | Tests and debugging: one node runs at a time in a fixed order, and a deadlock is reported the moment it happens. |
+| `threads` (available) | I/O-bound stages, numpy and C work, and full parallelism on free-threaded CPython 3.14t. |
+| `processes` (available) | CPU-bound pure Python on a GIL build: farm workers in child processes, everything else in the parent. |
+| async nodes (available) | Network-heavy stages: `async def` nodes run on an event loop, and a farm of them is one pool running `workers` coroutines at a time on one thread. |
+| distributed (available) | Two or more machines: a deploy file cuts the graph into groups, and edges crossing a group become TCP channels with backpressure, resend and an optional shared secret. |
+
+## The AI builder
+
+```
+pip install "tolquane[ai]"
+export ANTHROPIC_API_KEY=...        # or OPENAI_API_KEY with --provider openai
+tolquane build "read urls.txt, fetch each with 8 workers, write url, status and size to status.csv"
+```
+
+The builder writes one short, commented `flow.py` in the house style, checks its wiring,
+runs it on the deterministic runtime with a sample it makes up (or `--sample file`),
+fixes what fails, then asks you what to change. Claude Opus 5 is the default; GPT works
+through `--provider openai`. Ten flows it wrote, unedited, with their transcripts, are in
+[examples/generated/](examples/generated/); none of the ten needed a correction.
+The same loop is a function: `tolquane.ai.build(description, workdir=".")`.
+
+Generated code runs on your machine, in a subprocess, with a timeout. Keys are read
+from the environment and never stored. `tolquane check`, `run`, `explain` and `draw`
+work on any file that defines `build(source=None)`.
+
+## Tolquane Web
+
+```
+pip install "tolquane[web]"
+tolquane web
+```
+
+A local page for the flows in one directory: the blocks on a canvas, the Python beside
+it and editable both ways, runs with live per-node counts and tapped items, a run
+history, cron schedules and the AI builder in a side panel. Flows run in child
+processes, so a hung or crashing flow cannot take the server down, and the file stays a
+plain `flow.py` that runs with `python flow.py` anywhere Tolquane is installed. Since 1.3
+it also has accounts with roles, parameters and environment variables for a run or a
+schedule, a git history of every flow, and webhooks, mail and retries when a schedule
+ends. The tour, with screenshots, is [docs/web-user.md](docs/web-user.md).
+
+## Principles
+
+- No hangs. EOS is a message, every edge is bounded, every node waits on one inbox,
+  errors cancel the graph, and a watchdog reports deadlocks by name.
+- No magic. One way to write a node; cardinality is checked at build time, not guessed.
+- No dependencies. Pure Python 3.11+, optional extras for msgpack, numpy and cloudpickle.
+
+## Development
+
+```
+uv venv .venv --python 3.13
+uv pip install --python .venv/bin/python -e ".[dev]"
+.venv/bin/ruff check . && .venv/bin/ruff format --check . && .venv/bin/mypy && .venv/bin/pytest
+```
+
+## License
+
+Apache-2.0. See [LICENSE](LICENSE).

tolquane-1.3.0/README.md

@@ -0,0 +1,128 @@
+# Tolquane
+
+[![CI](https://github.com/robtacconelli/Tolquane/actions/workflows/ci.yml/badge.svg)](https://github.com/robtacconelli/Tolquane/actions/workflows/ci.yml)
+[![License: Apache-2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
+
+Parallel programming with composable building blocks, in Python.
+
+Nodes speak on channels. Pipelines, farms and all-to-all blocks compose them, and the
+same graph runs on threads, processes or across a network. Tolquane is the successor of
+[BBFlow](https://github.com/robtacconelli/BBFlow), a Java implementation of the
+[FastFlow](https://github.com/fastflow/fastflow) building blocks, rebuilt from scratch
+to be simple to use and impossible to hang.
+
+> **1.1.** Nodes, pipelines, farms with every emitter and collector policy and any block
+> as a worker, ordered farms, node fusion, all-to-all, feedback loops that terminate by
+> rule, batching, sessions, deadlock detection, an optimizer that cuts threads; threads,
+> processes, coroutine pools, a distributed runtime over TCP with a one-command launcher,
+> and a deterministic sync runtime; a run report that names the bottleneck; an AI builder
+> that writes, checks and runs flows from a sentence; and Tolquane Web, a local GUI with
+> the blocks on a canvas, live runs and schedules. See [DESIGN.md](DESIGN.md) for the
+> design and the liveness rules, [docs/api-card.md](docs/api-card.md) for the whole API
+> on one page, [examples/](examples/) for flows in the house style, and
+> [CHANGELOG.md](CHANGELOG.md).
+
+## What it looks like
+
+```python
+import tolquane as tq
+
+@tq.source
+def numbers():
+    yield from range(1, 101)          # a source is a generator
+
+@tq.node
+def double(x: int) -> int:
+    return x * 2                       # the return value is sent downstream
+
+@tq.sink
+def show(x: int) -> None:
+    print(x)
+
+graph = numbers >> tq.farm(double, workers=4) >> show
+tq.run(graph)                          # threads by default
+tq.run(graph, runtime="processes")     # same graph, farm workers in child processes
+tq.run(graph, runtime="sync")          # same graph, one thread, deterministic
+tq.run(graph, deploy="deploy.toml", group="G1")   # same graph, this host's share of it
+```
+
+```toml
+# deploy.toml: each host runs `tolquane run flow.py --deploy deploy.toml --group <name>`
+[groups.G1]
+endpoint = "10.0.0.1:7000"
+nodes = ["numbers", "double.emitter", "double.collector", "show"]
+[groups.G2]
+endpoint = "10.0.0.2:7000"
+nodes = ["double.[0-9]*"]           # the workers, on the other machine
+```
+
+A function is a node. Return `tq.SKIP` to drop an item; `None` is an ordinary value.
+A generator function yields zero or many items. Farms come with round-robin, broadcast,
+scatter, on-demand and key-based emitters, first-come, round-robin, gather and ordered
+collectors. `tq.all2all` joins two farms worker to worker, `tq.feedback` wires a block
+back onto itself with a loop that closes when nothing is left in flight, and
+`tq.session` keeps a graph running while you push items in and read results out.
+
+## Runtimes
+
+| Runtime | Use when |
+|---|---|
+| `sync` (available) | Tests and debugging: one node runs at a time in a fixed order, and a deadlock is reported the moment it happens. |
+| `threads` (available) | I/O-bound stages, numpy and C work, and full parallelism on free-threaded CPython 3.14t. |
+| `processes` (available) | CPU-bound pure Python on a GIL build: farm workers in child processes, everything else in the parent. |
+| async nodes (available) | Network-heavy stages: `async def` nodes run on an event loop, and a farm of them is one pool running `workers` coroutines at a time on one thread. |
+| distributed (available) | Two or more machines: a deploy file cuts the graph into groups, and edges crossing a group become TCP channels with backpressure, resend and an optional shared secret. |
+
+## The AI builder
+
+```
+pip install "tolquane[ai]"
+export ANTHROPIC_API_KEY=...        # or OPENAI_API_KEY with --provider openai
+tolquane build "read urls.txt, fetch each with 8 workers, write url, status and size to status.csv"
+```
+
+The builder writes one short, commented `flow.py` in the house style, checks its wiring,
+runs it on the deterministic runtime with a sample it makes up (or `--sample file`),
+fixes what fails, then asks you what to change. Claude Opus 5 is the default; GPT works
+through `--provider openai`. Ten flows it wrote, unedited, with their transcripts, are in
+[examples/generated/](examples/generated/); none of the ten needed a correction.
+The same loop is a function: `tolquane.ai.build(description, workdir=".")`.
+
+Generated code runs on your machine, in a subprocess, with a timeout. Keys are read
+from the environment and never stored. `tolquane check`, `run`, `explain` and `draw`
+work on any file that defines `build(source=None)`.
+
+## Tolquane Web
+
+```
+pip install "tolquane[web]"
+tolquane web
+```
+
+A local page for the flows in one directory: the blocks on a canvas, the Python beside
+it and editable both ways, runs with live per-node counts and tapped items, a run
+history, cron schedules and the AI builder in a side panel. Flows run in child
+processes, so a hung or crashing flow cannot take the server down, and the file stays a
+plain `flow.py` that runs with `python flow.py` anywhere Tolquane is installed. Since 1.3
+it also has accounts with roles, parameters and environment variables for a run or a
+schedule, a git history of every flow, and webhooks, mail and retries when a schedule
+ends. The tour, with screenshots, is [docs/web-user.md](docs/web-user.md).
+
+## Principles
+
+- No hangs. EOS is a message, every edge is bounded, every node waits on one inbox,
+  errors cancel the graph, and a watchdog reports deadlocks by name.
+- No magic. One way to write a node; cardinality is checked at build time, not guessed.
+- No dependencies. Pure Python 3.11+, optional extras for msgpack, numpy and cloudpickle.
+
+## Development
+
+```
+uv venv .venv --python 3.13
+uv pip install --python .venv/bin/python -e ".[dev]"
+.venv/bin/ruff check . && .venv/bin/ruff format --check . && .venv/bin/mypy && .venv/bin/pytest
+```
+
+## License
+
+Apache-2.0. See [LICENSE](LICENSE).

tolquane-1.3.0/benchmarks/README.md

@@ -0,0 +1,120 @@
+# Benchmarks
+
+Two scripts reproduce the shapes measured in the BBFlow thesis so the numbers can be
+compared as the runtime evolves.
+
+| Script | Thesis reference | What it measures |
+|---|---|---|
+| `pipeline2.py N [runtime] [capacity]` | Table 1, Figures 10 and 11 | Per-item channel overhead: producer to consumer, N items |
+| `farm_scaling.py [items] [max_workers] [work]` | Figures 13 to 15 | Speedup and efficiency of a farm of CPU-bound workers |
+| `farm_throughput.py [items] [workers] [batch]` | Table 5 in spirit | Hand-off cost through emitter, workers and collector with trivial work |
+
+## Baseline, 2026-09-05 (Phase 1, no batching yet)
+
+Machine: 16 hardware threads, Linux. Interpreters installed with uv.
+
+Two-node pipeline, 1,000,000 integers:
+
+| Interpreter | Runtime | Time | Rate |
+|---|---|---|---|
+| 3.13 | threads | 2.40 s | 0.42 M items/s |
+| 3.13 | sync | 2.52 s | 0.40 M items/s |
+| 3.14t | threads | 3.06 s | 0.33 M items/s |
+| 3.14t | sync | 4.35 s | 0.23 M items/s |
+
+For scale, the thesis measured 159 ms for the same 1M items in BBFlow (Java, with
+class preloading) and 76 ms in FastFlow (C++). Python's per-item cost is about 2.4 µs
+here, almost all of it condition-variable traffic per item. Transparent batching on
+channels (Phase 2) is the lever that changes this.
+
+Farm of CPU-bound pure Python workers, 800 items, 20,000 loop iterations each:
+
+| Interpreter | 1 worker | 2 | 4 | 8 |
+|---|---|---|---|---|
+| 3.13 (GIL) speedup | 0.97 | 0.97 | 0.89 | 0.90 |
+| 3.14t (free-threaded) speedup | 0.96 | 1.79 | 2.95 | 4.52 |
+| 3.14t efficiency | 0.96 | 0.90 | 0.74 | 0.57 |
+
+The GIL build shows none, as it must for pure Python work; the free-threaded build
+shows the farm scaling the way the thesis's Java farm did. The processes runtime
+(Phase 4) is the answer on GIL builds.
+
+## Phase 2, batching (2026-09-05)
+
+Channels now carry up to `batch` items per hand-off (default 32), flushed when the
+batch is full, after one millisecond, or before the producer blocks on anything.
+
+Two-node pipeline, 1,000,000 integers, threads:
+
+| Interpreter | batch 1 | batch 32 |
+|---|---|---|
+| 3.13 | 2.54 s | 2.10 s |
+| 3.14t | 3.27 s | 2.32 s |
+
+Farm of four trivial workers (emitter, workers, collector: three hops), 300,000 integers:
+
+| Interpreter | batch 1 | batch 32 |
+|---|---|---|
+| 3.13 (GIL) | 3377 ms | 2788 ms |
+| 3.14t | 3311 ms | 684 ms |
+
+The two-node case is bounded by interpreter work in the producer and consumer loops
+(about 2 µs per item: a dozen Python calls and attribute loads). Batching removes the
+lock and condition traffic per item, which is why the three-hop farm gains 5x on the
+free-threaded build while the pipeline gains 20%. On the GIL build the six threads of
+the farm still serialize on the interpreter, so the gain is small; the processes
+runtime (Phase 4) is the answer there, and for array work it is sending arrays, not
+scalars.
+
+Farm scaling (800 items of 20,000 loop iterations) on 3.14t with the default batch:
+speedup 1.34, 2.86 and 3.88 with 2, 4 and 8 workers.
+
+## Phase 4, processes (2026-09-05)
+
+`runtime="processes"` puts every farm worker in its own spawned child; the rest of the
+graph stays in the parent. Farm of CPU-bound workers, 3,000 items of 20,000 loop
+iterations each (about 2.4 s sequential):
+
+| Runtime | 1 worker | 2 | 4 | 8 |
+|---|---|---|---|---|
+| 3.13 threads (GIL) | 0.97 | 0.98 | 0.91 | 0.92 |
+| 3.13 processes | 0.97 | 1.87 | 3.51 | 5.47 |
+| 3.14t threads | 0.98 | 1.83 | 3.51 | 5.56 |
+
+Speedup over the sequential loop. Processes on the GIL build now match the
+free-threaded interpreter's threads. Efficiency at 8 workers is 0.68 on both; the
+missing part is process startup (about 0.1 s per run, all children at once), pickling
+each item across the pipe, and the emitter and collector sharing one core with the
+parent's threads. On a 0.5 s workload the same farm reaches only 3.3x, so keep
+processes for work that runs longer than a second or send bigger items.
+
+## Phase 5, TCP between groups (2026-09-05)
+
+Two groups as two processes on loopback, one cut edge between them (thesis Tables 2
+and 5). `benchmarks/network_pipeline.py 1000000 <batch>` and
+`benchmarks/network_farm.py 300000 3 32`.
+
+| Shape | batch 1 | batch 32 | batch 256 |
+|---|---|---|---|
+| producer to consumer, 1,000,000 integers | 45.5 s | 3.5 s | 2.5 s |
+| farm of 3 with the workers on the other side, 300,000 integers | | 4.0 s | |
+
+For scale, the thesis measured 1.2 s for the same million items through BBFlow's
+buffered Java channel on loopback and 4.5 s unbuffered. The batch is the whole story
+here: one frame per item costs a system call and a round trip of acknowledgements per
+item, one frame per 32 items amortizes both. The farm shape pays twice (emitter to
+worker and worker to collector both cross the wire), and its per-item Python work in
+the parent group runs on one core next to the two proxies per worker.
+
+## Phase 6, coroutine pools (2026-09-05)
+
+`tq.farm(fetch, workers=N)` where `fetch` awaits 20 ms per item, 2000 items, 3.13:
+
+| workers | 1 | 10 | 100 | 1000 |
+|---|---|---|---|---|
+| time | 41.3 s | 4.21 s | 0.44 s | 0.08 s |
+
+Wall time is the waiting time divided by the pool size until the fixed cost of the
+pool (about 40 µs per item on one thread) takes over. The sync runtime's speed is
+unchanged by the scheduler rework of this phase: 2.50 s for the 1,000,000-item
+pipeline against 2.52 s in the Phase 1 baseline.