sqlas 1.1.0__tar.gz → 1.1.1__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (23) hide show
  1. {sqlas-1.1.0/sqlas.egg-info → sqlas-1.1.1}/PKG-INFO +42 -14
  2. {sqlas-1.1.0 → sqlas-1.1.1}/README.md +35 -7
  3. {sqlas-1.1.0 → sqlas-1.1.1}/pyproject.toml +7 -7
  4. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/__init__.py +4 -4
  5. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/production.py +0 -2
  6. {sqlas-1.1.0 → sqlas-1.1.1/sqlas.egg-info}/PKG-INFO +42 -14
  7. {sqlas-1.1.0 → sqlas-1.1.1}/LICENSE +0 -0
  8. {sqlas-1.1.0 → sqlas-1.1.1}/setup.cfg +0 -0
  9. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/context.py +0 -0
  10. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/core.py +0 -0
  11. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/correctness.py +0 -0
  12. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/evaluate.py +0 -0
  13. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/py.typed +0 -0
  14. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/quality.py +0 -0
  15. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/response.py +0 -0
  16. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/runner.py +0 -0
  17. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas/safety.py +0 -0
  18. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas.egg-info/SOURCES.txt +0 -0
  19. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas.egg-info/dependency_links.txt +0 -0
  20. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas.egg-info/requires.txt +0 -0
  21. {sqlas-1.1.0 → sqlas-1.1.1}/sqlas.egg-info/top_level.txt +0 -0
  22. {sqlas-1.1.0 → sqlas-1.1.1}/tests/test_context.py +0 -0
  23. {sqlas-1.1.0 → sqlas-1.1.1}/tests/test_sqlas.py +0 -0
{sqlas-1.1.0/sqlas.egg-info → sqlas-1.1.1}/PKG-INFO RENAMED
@@ -1,13 +1,13 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: sqlas
3
- Version: 1.1.0
4
- Summary: SQLAS — SQL Agent Scoring Framework. A RAGAS-equivalent evaluation library for Text-to-SQL and SQL AI agents. 20 production-grade metrics across 8 categories.
5
- Author: SQLAS Contributors
3
+ Version: 1.1.1
4
+ Summary: SQLAS — SQL Agent Scoring Framework. Production-grade evaluation for Text-to-SQL and SQL AI agents. 20 metrics across 8 categories.
5
+ Author-email: Pradip Tivhale <pradiptivhale@gmail.com>
6
6
  License: MIT
7
- Project-URL: Homepage, https://github.com/sqlas-framework/sqlas
8
- Project-URL: Documentation, https://github.com/sqlas-framework/sqlas#readme
9
- Project-URL: Repository, https://github.com/sqlas-framework/sqlas
10
- Project-URL: Changelog, https://github.com/sqlas-framework/sqlas/blob/main/CHANGELOG.md
7
+ Project-URL: Homepage, https://github.com/thepradip/SQLAS
8
+ Project-URL: Documentation, https://github.com/thepradip/SQLAS#readme
9
+ Project-URL: Repository, https://github.com/thepradip/SQLAS
10
+ Project-URL: Changelog, https://github.com/thepradip/SQLAS/blob/main/CHANGELOG.md
11
11
  Keywords: sql,agent,evaluation,llm,text-to-sql,ragas,mlflow,benchmark,monitoring
12
12
  Classifier: Development Status :: 5 - Production/Stable
13
13
  Classifier: Intended Audience :: Developers
@@ -35,21 +35,39 @@ Provides-Extra: all
35
35
  Requires-Dist: mlflow>=3.0; extra == "all"
36
36
  Dynamic: license-file
37
37
 
38
- # SQLAS — SQL Agent Scoring Framework
38
+ <p align="center">
39
+ <img src="assets/sqlas_logo.png" alt="SQLAS Logo" width="280"/>
40
+ </p>
39
41
 
40
- **A RAGAS-equivalent evaluation library for Text-to-SQL and SQL AI agents.**
42
+ <h1 align="center">SQLAS — SQL Agent Scoring Framework</h1>
41
43
 
42
- SQLAS evaluates SQL agents across **20 production metrics** in **8 categories**, aligned with industry best practices (Spider, BIRD, Arize, MLflow).
44
+ <p align="center">
45
+ <strong>Production-grade evaluation framework for Text-to-SQL and SQL AI agents. 20 metrics. 8 categories. Any LLM.</strong>
46
+ </p>
43
47
 
44
- **Author:** SQLAS Contributors
48
+ <p align="center">
49
+ <a href="https://pypi.org/project/sqlas/"><img src="https://img.shields.io/pypi/v/sqlas?style=flat-square&color=orange" alt="PyPI"/></a>
50
+ <img src="https://img.shields.io/badge/python-3.10+-blue?style=flat-square" alt="Python"/>
51
+ <img src="https://img.shields.io/badge/license-MIT-green?style=flat-square" alt="License"/>
52
+ </p>
53
+
54
+ SQLAS scores your SQL agent the way production demands — execution accuracy, semantic correctness, context quality, cost efficiency, safety, and more. Built on industry benchmarks (Spider, BIRD) and real-world observability patterns (Arize, MLflow).
55
+
56
+ **Author:** [Pradip Tivhale](https://github.com/thepradip)
45
57
 
46
58
  ---
47
59
 
48
60
  ## Install
49
61
 
50
62
  ```bash
63
+ # From PyPI
51
64
  pip install sqlas
52
65
 
66
+ # From source
67
+ git clone https://github.com/thepradip/SQLAS.git
68
+ cd SQLAS
69
+ pip install .
70
+
53
71
  # With MLflow integration
54
72
  pip install sqlas[mlflow]
55
73
 
@@ -260,9 +278,9 @@ recall, details = context_recall(
260
278
 
261
279
  ---
262
280
 
263
- ## RAGAS Mapping
281
+ ## Metric Mapping (vs. RAG Evaluation Standards)
264
282
 
265
- | RAGAS Metric | SQLAS Equivalent | Description |
283
+ | Standard Metric | SQLAS Equivalent | Description |
266
284
  |---|---|---|
267
285
  | Faithfulness | `faithfulness` | Claims grounded in SQL result data |
268
286
  | Answer Relevance | `answer_relevance` | Response answers the question |
@@ -317,6 +335,16 @@ def judge(prompt):
317
335
 
318
336
  ---
319
337
 
338
+ ## Example: SQL AI Agent (LangGraph + SQLAS)
339
+
340
+ See [**thepradip/SQL-AI-Agent**](https://github.com/thepradip/SQL-AI-Agent) — a full-stack NL-to-SQL application powered by LangGraph that uses SQLAS for:
341
+
342
+ - **Pre-execution safety gate** — `read_only_compliance`, `safety_score`, `schema_compliance` block unsafe queries
343
+ - **Post-response quality scoring** — full `evaluate()` scores every query on 20 metrics
344
+ - **Evaluation suite** — 25 test cases across 4 difficulty tiers scored by SQLAS
345
+
346
+ ---
347
+
320
348
  ## License
321
349
 
322
- MIT License - SQLAS Contributors
350
+ MIT License - [Pradip Tivhale](https://github.com/thepradip)
{sqlas-1.1.0 → sqlas-1.1.1}/README.md RENAMED
@@ -1,18 +1,36 @@
1
- # SQLAS — SQL Agent Scoring Framework
1
+ <p align="center">
2
+ <img src="assets/sqlas_logo.png" alt="SQLAS Logo" width="280"/>
3
+ </p>
2
4
 
3
- **A RAGAS-equivalent evaluation library for Text-to-SQL and SQL AI agents.**
5
+ <h1 align="center">SQLAS — SQL Agent Scoring Framework</h1>
4
6
 
5
- SQLAS evaluates SQL agents across **20 production metrics** in **8 categories**, aligned with industry best practices (Spider, BIRD, Arize, MLflow).
7
+ <p align="center">
8
+ <strong>Production-grade evaluation framework for Text-to-SQL and SQL AI agents. 20 metrics. 8 categories. Any LLM.</strong>
9
+ </p>
6
10
 
7
- **Author:** SQLAS Contributors
11
+ <p align="center">
12
+ <a href="https://pypi.org/project/sqlas/"><img src="https://img.shields.io/pypi/v/sqlas?style=flat-square&color=orange" alt="PyPI"/></a>
13
+ <img src="https://img.shields.io/badge/python-3.10+-blue?style=flat-square" alt="Python"/>
14
+ <img src="https://img.shields.io/badge/license-MIT-green?style=flat-square" alt="License"/>
15
+ </p>
16
+
17
+ SQLAS scores your SQL agent the way production demands — execution accuracy, semantic correctness, context quality, cost efficiency, safety, and more. Built on industry benchmarks (Spider, BIRD) and real-world observability patterns (Arize, MLflow).
18
+
19
+ **Author:** [Pradip Tivhale](https://github.com/thepradip)
8
20
 
9
21
  ---
10
22
 
11
23
  ## Install
12
24
 
13
25
  ```bash
26
+ # From PyPI
14
27
  pip install sqlas
15
28
 
29
+ # From source
30
+ git clone https://github.com/thepradip/SQLAS.git
31
+ cd SQLAS
32
+ pip install .
33
+
16
34
  # With MLflow integration
17
35
  pip install sqlas[mlflow]
18
36
 
@@ -223,9 +241,9 @@ recall, details = context_recall(
223
241
 
224
242
  ---
225
243
 
226
- ## RAGAS Mapping
244
+ ## Metric Mapping (vs. RAG Evaluation Standards)
227
245
 
228
- | RAGAS Metric | SQLAS Equivalent | Description |
246
+ | Standard Metric | SQLAS Equivalent | Description |
229
247
  |---|---|---|
230
248
  | Faithfulness | `faithfulness` | Claims grounded in SQL result data |
231
249
  | Answer Relevance | `answer_relevance` | Response answers the question |
@@ -280,6 +298,16 @@ def judge(prompt):
280
298
 
281
299
  ---
282
300
 
301
+ ## Example: SQL AI Agent (LangGraph + SQLAS)
302
+
303
+ See [**thepradip/SQL-AI-Agent**](https://github.com/thepradip/SQL-AI-Agent) — a full-stack NL-to-SQL application powered by LangGraph that uses SQLAS for:
304
+
305
+ - **Pre-execution safety gate** — `read_only_compliance`, `safety_score`, `schema_compliance` block unsafe queries
306
+ - **Post-response quality scoring** — full `evaluate()` scores every query on 20 metrics
307
+ - **Evaluation suite** — 25 test cases across 4 difficulty tiers scored by SQLAS
308
+
309
+ ---
310
+
283
311
  ## License
284
312
 
285
- MIT License - SQLAS Contributors
313
+ MIT License - [Pradip Tivhale](https://github.com/thepradip)
{sqlas-1.1.0 → sqlas-1.1.1}/pyproject.toml RENAMED
@@ -4,11 +4,11 @@ build-backend = "setuptools.build_meta"
4
4
 
5
5
  [project]
6
6
  name = "sqlas"
7
- version = "1.1.0"
8
- description = "SQLAS — SQL Agent Scoring Framework. A RAGAS-equivalent evaluation library for Text-to-SQL and SQL AI agents. 20 production-grade metrics across 8 categories."
7
+ version = "1.1.1"
8
+ description = "SQLAS — SQL Agent Scoring Framework. Production-grade evaluation for Text-to-SQL and SQL AI agents. 20 metrics across 8 categories."
9
9
  readme = "README.md"
10
10
  license = {text = "MIT"}
11
- authors = [{name = "SQLAS Contributors"}]
11
+ authors = [{name = "Pradip Tivhale", email = "pradiptivhale@gmail.com"}]
12
12
  requires-python = ">=3.10"
13
13
  keywords = ["sql", "agent", "evaluation", "llm", "text-to-sql", "ragas", "mlflow", "benchmark", "monitoring"]
14
14
  classifiers = [
@@ -35,10 +35,10 @@ dev = ["pytest>=7.0", "build", "twine"]
35
35
  all = ["mlflow>=3.0"]
36
36
 
37
37
  [project.urls]
38
- Homepage = "https://github.com/sqlas-framework/sqlas"
39
- Documentation = "https://github.com/sqlas-framework/sqlas#readme"
40
- Repository = "https://github.com/sqlas-framework/sqlas"
41
- Changelog = "https://github.com/sqlas-framework/sqlas/blob/main/CHANGELOG.md"
38
+ Homepage = "https://github.com/thepradip/SQLAS"
39
+ Documentation = "https://github.com/thepradip/SQLAS#readme"
40
+ Repository = "https://github.com/thepradip/SQLAS"
41
+ Changelog = "https://github.com/thepradip/SQLAS/blob/main/CHANGELOG.md"
42
42
 
43
43
  [tool.setuptools.packages.find]
44
44
  include = ["sqlas*"]
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/__init__.py RENAMED
@@ -1,8 +1,8 @@
1
1
  """
2
2
  SQLAS — SQL Agent Scoring Framework
3
- A RAGAS-equivalent evaluation library for Text-to-SQL and SQL AI agents.
3
+ Production-grade evaluation framework for Text-to-SQL and SQL AI agents.
4
4
 
5
- Author: SQLAS Contributors
5
+ Author: Pradip Tivhale
6
6
 
7
7
  Usage:
8
8
  from sqlas import evaluate, SQLASScores, TestCase, WEIGHTS
@@ -27,8 +27,8 @@ from sqlas.safety import safety_score, read_only_compliance
27
27
  from sqlas.context import context_precision, context_recall, entity_recall, noise_robustness
28
28
  from sqlas.runner import run_suite
29
29
 
30
- __version__ = "1.1.0"
31
- __author__ = "SQLAS Contributors"
30
+ __version__ = "1.1.1"
31
+ __author__ = "Pradip Tivhale"
32
32
 
33
33
  __all__ = [
34
34
  # Core
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/production.py RENAMED
@@ -6,8 +6,6 @@ Production Execution Metrics.
6
6
  Author: SQLAS Contributors
7
7
  """
8
8
 
9
- import re
10
-
11
9
 
12
10
  def data_scan_efficiency(sql: str, result_row_count: int = 0) -> tuple[float, dict]:
13
11
  """
{sqlas-1.1.0 → sqlas-1.1.1/sqlas.egg-info}/PKG-INFO RENAMED
@@ -1,13 +1,13 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: sqlas
3
- Version: 1.1.0
4
- Summary: SQLAS — SQL Agent Scoring Framework. A RAGAS-equivalent evaluation library for Text-to-SQL and SQL AI agents. 20 production-grade metrics across 8 categories.
5
- Author: SQLAS Contributors
3
+ Version: 1.1.1
4
+ Summary: SQLAS — SQL Agent Scoring Framework. Production-grade evaluation for Text-to-SQL and SQL AI agents. 20 metrics across 8 categories.
5
+ Author-email: Pradip Tivhale <pradiptivhale@gmail.com>
6
6
  License: MIT
7
- Project-URL: Homepage, https://github.com/sqlas-framework/sqlas
8
- Project-URL: Documentation, https://github.com/sqlas-framework/sqlas#readme
9
- Project-URL: Repository, https://github.com/sqlas-framework/sqlas
10
- Project-URL: Changelog, https://github.com/sqlas-framework/sqlas/blob/main/CHANGELOG.md
7
+ Project-URL: Homepage, https://github.com/thepradip/SQLAS
8
+ Project-URL: Documentation, https://github.com/thepradip/SQLAS#readme
9
+ Project-URL: Repository, https://github.com/thepradip/SQLAS
10
+ Project-URL: Changelog, https://github.com/thepradip/SQLAS/blob/main/CHANGELOG.md
11
11
  Keywords: sql,agent,evaluation,llm,text-to-sql,ragas,mlflow,benchmark,monitoring
12
12
  Classifier: Development Status :: 5 - Production/Stable
13
13
  Classifier: Intended Audience :: Developers
@@ -35,21 +35,39 @@ Provides-Extra: all
35
35
  Requires-Dist: mlflow>=3.0; extra == "all"
36
36
  Dynamic: license-file
37
37
 
38
- # SQLAS — SQL Agent Scoring Framework
38
+ <p align="center">
39
+ <img src="assets/sqlas_logo.png" alt="SQLAS Logo" width="280"/>
40
+ </p>
39
41
 
40
- **A RAGAS-equivalent evaluation library for Text-to-SQL and SQL AI agents.**
42
+ <h1 align="center">SQLAS — SQL Agent Scoring Framework</h1>
41
43
 
42
- SQLAS evaluates SQL agents across **20 production metrics** in **8 categories**, aligned with industry best practices (Spider, BIRD, Arize, MLflow).
44
+ <p align="center">
45
+ <strong>Production-grade evaluation framework for Text-to-SQL and SQL AI agents. 20 metrics. 8 categories. Any LLM.</strong>
46
+ </p>
43
47
 
44
- **Author:** SQLAS Contributors
48
+ <p align="center">
49
+ <a href="https://pypi.org/project/sqlas/"><img src="https://img.shields.io/pypi/v/sqlas?style=flat-square&color=orange" alt="PyPI"/></a>
50
+ <img src="https://img.shields.io/badge/python-3.10+-blue?style=flat-square" alt="Python"/>
51
+ <img src="https://img.shields.io/badge/license-MIT-green?style=flat-square" alt="License"/>
52
+ </p>
53
+
54
+ SQLAS scores your SQL agent the way production demands — execution accuracy, semantic correctness, context quality, cost efficiency, safety, and more. Built on industry benchmarks (Spider, BIRD) and real-world observability patterns (Arize, MLflow).
55
+
56
+ **Author:** [Pradip Tivhale](https://github.com/thepradip)
45
57
 
46
58
  ---
47
59
 
48
60
  ## Install
49
61
 
50
62
  ```bash
63
+ # From PyPI
51
64
  pip install sqlas
52
65
 
66
+ # From source
67
+ git clone https://github.com/thepradip/SQLAS.git
68
+ cd SQLAS
69
+ pip install .
70
+
53
71
  # With MLflow integration
54
72
  pip install sqlas[mlflow]
55
73
 
@@ -260,9 +278,9 @@ recall, details = context_recall(
260
278
 
261
279
  ---
262
280
 
263
- ## RAGAS Mapping
281
+ ## Metric Mapping (vs. RAG Evaluation Standards)
264
282
 
265
- | RAGAS Metric | SQLAS Equivalent | Description |
283
+ | Standard Metric | SQLAS Equivalent | Description |
266
284
  |---|---|---|
267
285
  | Faithfulness | `faithfulness` | Claims grounded in SQL result data |
268
286
  | Answer Relevance | `answer_relevance` | Response answers the question |
@@ -317,6 +335,16 @@ def judge(prompt):
317
335
 
318
336
  ---
319
337
 
338
+ ## Example: SQL AI Agent (LangGraph + SQLAS)
339
+
340
+ See [**thepradip/SQL-AI-Agent**](https://github.com/thepradip/SQL-AI-Agent) — a full-stack NL-to-SQL application powered by LangGraph that uses SQLAS for:
341
+
342
+ - **Pre-execution safety gate** — `read_only_compliance`, `safety_score`, `schema_compliance` block unsafe queries
343
+ - **Post-response quality scoring** — full `evaluate()` scores every query on 20 metrics
344
+ - **Evaluation suite** — 25 test cases across 4 difficulty tiers scored by SQLAS
345
+
346
+ ---
347
+
320
348
  ## License
321
349
 
322
- MIT License - SQLAS Contributors
350
+ MIT License - [Pradip Tivhale](https://github.com/thepradip)
{sqlas-1.1.0 → sqlas-1.1.1}/LICENSE RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/setup.cfg RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/context.py RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/core.py RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/correctness.py RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/evaluate.py RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/py.typed RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/quality.py RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/response.py RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/runner.py RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas/safety.py RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas.egg-info/SOURCES.txt RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/sqlas.egg-info/requires.txt RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/tests/test_context.py RENAMED
File without changes
{sqlas-1.1.0 → sqlas-1.1.1}/tests/test_sqlas.py RENAMED
File without changes