fineprint
Read the fine print of your metrics. A decompiler for your dashboards.
从 dbt 项目的编译产物(manifest / catalog / compiled SQL)逆向合成每个指标的 完整技术口径——零数据库连接。读懂指标的小字条款,给看板的反编译器。
Reverse-engineers the full technical caliber of every metric from your dbt artifacts (manifest / catalog / compiled SQL) — zero database connection.
- 字段级血缘图 column-level lineage:跨模型、进 CTE/子查询/UNION/PIVOT, 条件与聚合语义一并入图。
- 口径卡 caliber cards:确定性公式组合器(发布权威)× LLM 解读(解释与 叙述)双通道互证,每个事实带机器证据;组合器不可证时 LLM 兜底并标注原因。 在 5 个公开真实项目(34,499 列)上组合器覆盖 99.7%,残余全部具名。
- 口径漂移检测 caliber drift:图快照逐语义点比对,SQL 改动落到受影响 指标与条件粒度。
pip install fineprint
metriclens init --project <dbt 项目目录> # 生成 metriclens.yml 模板
metriclens graph --project DIR # 建血缘图(需先 dbt compile + docs generate)
metriclens synth --project DIR # 合成口径卡(需配置 LLM 端点)
metriclens drift --project DIR # 漂移检测
fineprint 与 metriclens 两个命令等价;import 名为 metriclens。
LLM 端点经环境变量或项目根 .env 配置(METRICLENS_LLM_BASE /
METRICLENS_LLM_KEY / METRICLENS_LLM_MODEL),兼容 OpenAI 风格 API。
无 catalog 也能跑(列 schema 由 yml 声明 + 编译 SQL 拓扑推断补全);能执行
dbt docs generate 时仍建议补上,实测列集更可靠。
License: Apache-2.0
1 maintainer
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file fineprint-0.8.1.tar.gz.
File metadata
- Download URL: fineprint-0.8.1.tar.gz
- Upload date:
- Size: 74.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/7.0.0 CPython/3.13.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
4702817f8230a7ce61696171e6aeee9b42bae6704d173619b04dbe883ac9126e
|
|
| MD5 |
cfcadf34f1ce443aad7d7b1205784df0
|
|
| BLAKE2b-256 |
a76d0b522b4da4353d44070707107736d71193f0819a69d52a23a8df2071bd07
|
File details
Details for the file fineprint-0.8.1-py3-none-any.whl.
File metadata
- Download URL: fineprint-0.8.1-py3-none-any.whl
- Upload date:
- Size: 83.3 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/7.0.0 CPython/3.13.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
36bd3a646cb96bc28d149ceedd83353d6796cf7c8b5c9b2b60cae6c17f42a9b1
|
|
| MD5 |
4647fc1ef071efcca6df6e7251e576a3
|
|
| BLAKE2b-256 |
fc4443b665256eaeaeae34c65d55c80c18a091e814e13c1d3c4271db225eb875
|
