context-window-packer-py
Pack context chunks into a budget by relevance and priority. Sort by score (descending), break ties by priority, accept each chunk if it fits, otherwise drop. Zero runtime dependencies.
Python port of @mukundakatta/context-window-packer.
Install
pip install context-window-packer-py
Quick start
from context_window_packer import pack
chunks = [
{"text": "Pluto was reclassified...", "priority": 1, "relevance": 0.9},
{"text": "Mars is the fourth planet...", "priority": 0, "relevance": 0.4},
{"text": "Cats are independent...", "priority": 0, "relevance": 0.1},
]
result = pack(chunks, budget_tokens=20)
result.chunks # list of dicts that fit
result.tokens # int -- tokens consumed by kept chunks
result.dropped # int -- count of chunks not kept
API
pack(chunks, budget_tokens, model=None) -> PackResult
Each chunk is a mapping with at minimum text. Optional fields:
| Field | Default | Meaning |
|---|---|---|
text |
"" |
The chunk content. |
tokens |
ceil(len(text)/4) |
Pre-counted tokens (skips estimator). |
relevance / score |
0 |
Higher = more relevant; primary sort key (descending). |
priority |
0 |
Tie-breaker for relevance (descending). |
PackResult is a dataclass with:
chunks: list[dict]-- chunks that fit, in pack order.tokens: int-- total tokens consumed.dropped: int-- count of chunks dropped.
model is reserved for future per-family token estimators (currently uses the standard ceil(chars/4) heuristic, matching the JS sibling).
License
MIT
1 maintainer
Metadata
1 maintainer
Release files for context-window-packer-py 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| context_window_packer_py-0.1.0.tar.gz | 4.7 kB | Without provenance | Details for context_window_packer_py-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| context_window_packer_py-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for context_window_packer_py-0.1.0-py3-none-any.whl |
Total release size: 9.7 kB
Release files / context_window_packer_py-0.1.0.tar.gz
| Download URL | context_window_packer_py-0.1.0.tar.gz |
|---|---|
| Size | 4.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
db83737b7303dad40b7a64c689034da27ef4eccaa83ed27df21fee2ed627f6f4
|
|
BLAKE2b-256 checksum How to use checksums |
f809b3c539cc32caccb90fdf3c3808a81bac19af5cb54deef11ff5b66a393a12
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.4
|
Release files / context_window_packer_py-0.1.0-py3-none-any.whl
| Download URL | context_window_packer_py-0.1.0-py3-none-any.whl |
|---|---|
| Size | 5.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
4b7a0b39e69ceb73182ba86f08bc9a531fedd07af9a83ab47ed6b69f258760a2
|
|
BLAKE2b-256 checksum How to use checksums |
7d0eb0c395daf44532203f40ffce46850472c0986432f8d52ac1b90d3b15b3f0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.4
|
Security for context-window-packer-py 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
