ChilliPeep commited on
Commit
34d3ef3
·
verified ·
1 Parent(s): 733de97

Deploy demo

Browse files
README.md CHANGED
@@ -1,13 +1,25 @@
1
  ---
2
- title: Techdebt Ml
3
- emoji: 😻
4
  colorFrom: yellow
5
- colorTo: gray
6
  sdk: gradio
7
- sdk_version: 6.30.0
8
- python_version: '3.12'
9
  app_file: app.py
10
  pinned: false
 
 
 
 
 
 
11
  ---
12
 
13
- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
 
 
 
 
 
 
1
  ---
2
+ title: TechDebt ML
3
+ emoji: 🧹
4
  colorFrom: yellow
5
+ colorTo: red
6
  sdk: gradio
7
+ sdk_version: 6.29.1
8
+ python_version: "3.12"
9
  app_file: app.py
10
  pinned: false
11
+ license: mit
12
+ short_description: Predicts technical debt risk per file in Python repos
13
+ models:
14
+ - sentence-transformers/all-MiniLM-L6-v2
15
+ datasets:
16
+ - ChilliPeep/techdebt-ml-dataset
17
  ---
18
 
19
+ # TechDebt ML — demo
20
+
21
+ Paste a public GitHub repo with Python code and get, per file, the predicted probability
22
+ that it carries technical debt.
23
+
24
+ Source code, training notebook and known limitations:
25
+ [github.com/Chillipeeper1/techdebt-ml](https://github.com/Chillipeeper1/techdebt-ml)
app.py ADDED
@@ -0,0 +1,295 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """
2
+ app.py - Gradio demo for TechDebt ML (Hugging Face Spaces, ZeroGPU)
3
+
4
+ Clones a public GitHub repo, runs the same v3 pipeline as src/predict.py
5
+ (tabular features + code embeddings) and shows the per-file risk report.
6
+
7
+ Only the embeddings run on the GPU. If the visitor's ZeroGPU quota is used up,
8
+ they are computed on CPU instead (same model, a few seconds slower).
9
+ """
10
+
11
+ import json
12
+ import os
13
+ import re
14
+ import select
15
+ import subprocess
16
+ import sys
17
+ import tempfile
18
+ import time
19
+ from pathlib import Path
20
+
21
+ # src/ lives next to this file on the Space, and one level up in the GitHub repo
22
+ HERE = Path(__file__).resolve().parent
23
+ for candidate in (HERE / "src", HERE.parent / "src"):
24
+ if candidate.is_dir():
25
+ sys.path.insert(0, str(candidate))
26
+ break
27
+
28
+ try:
29
+ import spaces # must be imported before torch on ZeroGPU
30
+ except ImportError: # running locally without the HF `spaces` package
31
+ spaces = None
32
+
33
+ import gradio as gr
34
+ import numpy as np
35
+ import pandas as pd
36
+ import torch
37
+ from sentence_transformers import SentenceTransformer
38
+
39
+ import predict # loads the v3 model, PCA and the CPU embedding model once at startup
40
+ from fast_git import git_features
41
+
42
+ EXAMPLES_DIR = HERE / "examples"
43
+ GITHUB_URL_RE = re.compile(r"^https?://github\.com/([\w.-]+)/([\w.-]+?)(?:\.git)?/?$")
44
+
45
+ MAX_FILES = 150 # larger repos are cut down to their biggest .py files
46
+ CLONE_TIMEOUT_S = 90
47
+ PER_FILE_TIMEOUT_S = 10 # radon can hang on some huge files
48
+
49
+ # ZeroGPU wants models placed on cuda at import time; the CPU copy in predict stays as fallback
50
+ GPU_EMBED_MODEL = SentenceTransformer("all-MiniLM-L6-v2").to("cuda") if torch.cuda.is_available() else None
51
+
52
+ # column in the report -> header shown in the UI
53
+ SHOWN_COLUMNS = {
54
+ "file": "file",
55
+ "danger": "risk",
56
+ "sloc": "SLOC",
57
+ "cyclomatic_complexity": "avg CC",
58
+ "max_complexity": "max CC",
59
+ "maintainability_index": "MI",
60
+ "max_nesting_depth": "nesting",
61
+ "num_functions": "funcs",
62
+ "commit_count": "commits×",
63
+ "author_count": "authors×",
64
+ "churn_rate": "churn×",
65
+ "days_since_change": "age×",
66
+ }
67
+
68
+ EXAMPLE_REPOS = [
69
+ "https://github.com/psf/requests",
70
+ "https://github.com/pallets/click",
71
+ "https://github.com/Chillipeeper1/techdebt-ml",
72
+ ]
73
+
74
+
75
+ def parse_github_url(url: str) -> tuple[str, str]:
76
+ match = GITHUB_URL_RE.match(url.strip())
77
+ if not match:
78
+ raise gr.Error("Please paste a public GitHub repo URL, e.g. https://github.com/psf/requests")
79
+ return match.group(1), match.group(2)
80
+
81
+
82
+ def clone_repo(owner: str, name: str, dest: Path):
83
+ # Full history is required: commit count, authors and churn are model features.
84
+ try:
85
+ subprocess.run(
86
+ ["git", "clone", "--quiet", "--single-branch", "--no-tags",
87
+ f"https://github.com/{owner}/{name}.git", str(dest)],
88
+ check=True, capture_output=True, text=True, timeout=CLONE_TIMEOUT_S,
89
+ env={**os.environ, "GIT_TERMINAL_PROMPT": "0"},
90
+ )
91
+ except subprocess.TimeoutExpired:
92
+ raise gr.Error(f"Cloning took more than {CLONE_TIMEOUT_S}s. Try a smaller repo.")
93
+ except subprocess.CalledProcessError:
94
+ raise gr.Error(f"Could not clone {owner}/{name}. Is it public and does it exist?")
95
+
96
+
97
+ def pick_files(repo_path: Path) -> tuple[list[str], int]:
98
+ files = predict.find_python_files(repo_path)
99
+ total = len(files)
100
+ if total > MAX_FILES:
101
+ files = sorted(files, key=lambda f: (repo_path / f).stat().st_size, reverse=True)[:MAX_FILES]
102
+ return files, total
103
+
104
+
105
+ def code_metrics(repo_path: Path, files: list[str], progress) -> tuple[dict, list[str]]:
106
+ """Runs metrics_worker.py over the files, killing and restarting it when a
107
+ file takes more than PER_FILE_TIMEOUT_S. Returns ({file: output}, skipped)."""
108
+ outputs, skipped = {}, []
109
+ pending = list(files)
110
+ while pending:
111
+ proc = subprocess.Popen(
112
+ [sys.executable, str(HERE / "metrics_worker.py"), str(repo_path)],
113
+ stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.DEVNULL,
114
+ )
115
+ proc.stdin.write("\n".join(pending).encode())
116
+ proc.stdin.close()
117
+ buffer = b""
118
+ try:
119
+ while pending:
120
+ if b"\n" not in buffer:
121
+ ready, _, _ = select.select([proc.stdout], [], [], PER_FILE_TIMEOUT_S)
122
+ chunk = os.read(proc.stdout.fileno(), 1 << 20) if ready else b""
123
+ if not chunk: # timed out or worker died on pending[0]
124
+ skipped.append(pending.pop(0))
125
+ break
126
+ buffer += chunk
127
+ continue
128
+ line, buffer = buffer.split(b"\n", 1)
129
+ outputs[pending.pop(0)] = json.loads(line)
130
+ progress(0.2 + 0.5 * len(outputs) / len(files), desc=f"Code metrics ({len(outputs)}/{len(files)})")
131
+ finally:
132
+ proc.kill()
133
+ proc.wait()
134
+ proc.stdout.close()
135
+ return outputs, skipped
136
+
137
+
138
+ def extract_features(repo_path: Path, files: list[str], repo_name: str, progress):
139
+ """Same features as predict.feature_extraction, but with git activity from
140
+ fast_git and code metrics from a killable worker process."""
141
+ progress(0.1, desc="Reading git history")
142
+ git_feats = git_features(repo_path, files)
143
+ metrics, skipped = code_metrics(repo_path, files, progress)
144
+
145
+ results, functions, index = {}, [], []
146
+ for file in files:
147
+ out = metrics.get(file)
148
+ if out is None or "error" in out:
149
+ continue
150
+ features = {**out["features"], **git_feats[file], "repo": repo_name}
151
+ key = f"{repo_name}/{file}".replace("\\", "/")
152
+ results[key] = features
153
+ index.append((key, len(out["functions"])))
154
+ functions.extend(out["functions"])
155
+ return results, functions, index, skipped
156
+
157
+
158
+ def file_embeddings(functions: list[str], index: list, model) -> dict:
159
+ """Same as predict.generar_embeddings_reducidos, with the model as a parameter:
160
+ encodes every function, then mean-pools them per file (zeros if none)."""
161
+ encoded = model.encode(functions, batch_size=64) if functions else np.empty((0, 384))
162
+ per_file, cursor = {}, 0
163
+ for key, count in index:
164
+ per_file[key] = encoded[cursor: cursor + count].mean(axis=0) if count else np.zeros(384)
165
+ cursor += count
166
+ return per_file
167
+
168
+
169
+ def _embed_on_gpu(functions, index):
170
+ return file_embeddings(functions, index, GPU_EMBED_MODEL)
171
+
172
+
173
+ if spaces is not None and GPU_EMBED_MODEL is not None:
174
+ _embed_on_gpu = spaces.GPU(duration=30)(_embed_on_gpu)
175
+
176
+
177
+ def embed(functions: list[str], index: list) -> tuple[dict, str]:
178
+ if GPU_EMBED_MODEL is not None:
179
+ try:
180
+ return _embed_on_gpu(functions, index), "GPU"
181
+ except Exception as e: # typically the visitor's daily ZeroGPU quota
182
+ print(f"GPU embedding failed, falling back to CPU: {e}")
183
+ return file_embeddings(functions, index, predict.EMBED_MODEL), "CPU"
184
+
185
+
186
+ def analyze_repo(repo_path: Path, repo_name: str, progress) -> tuple[pd.DataFrame, list[str]]:
187
+ files, total = pick_files(repo_path)
188
+ if not files:
189
+ raise gr.Error("No .py files found. TechDebt ML only supports Python codebases.")
190
+
191
+ notes = []
192
+ if total > len(files):
193
+ notes.append(f"The repo has {total} .py files; only the {len(files)} largest were analyzed.")
194
+
195
+ results, functions, index, skipped = extract_features(repo_path, files, repo_name, progress)
196
+ if skipped:
197
+ notes.append(f"Skipped {len(skipped)} file(s) that took over {PER_FILE_TIMEOUT_S}s to parse: "
198
+ + ", ".join(f"`{f}`" for f in skipped[:5]))
199
+ if not results:
200
+ raise gr.Error("Could not extract features from any file.")
201
+
202
+ progress(0.75, desc=f"Embedding {len(functions)} functions and predicting")
203
+ embeddings, device = embed(functions, index)
204
+ if device == "CPU" and GPU_EMBED_MODEL is not None:
205
+ notes.append("GPU quota used up, embeddings were computed on CPU.")
206
+ report = predict.full_report(results, embeddings)
207
+ report["file"] = report["file"].str.removeprefix(f"{repo_name}/")
208
+ return report, notes
209
+
210
+
211
+ def example_cache_path(owner: str, name: str) -> Path:
212
+ return EXAMPLES_DIR / f"{owner}__{name}.csv"
213
+
214
+
215
+ def build_outputs(report: pd.DataFrame, owner: str, name: str, notes: list[str], elapsed: float | None):
216
+ high = int((report["danger"] >= 0.5).sum())
217
+ summary = [
218
+ f"### {owner}/{name}",
219
+ f"**{len(report)}** files analyzed · **{high}** with risk ≥ 0.5 · "
220
+ f"median risk **{report['danger'].median():.2f}**",
221
+ ]
222
+ if elapsed is None:
223
+ summary.append("_Pre-computed example (instant). Any other URL is analyzed live._")
224
+ else:
225
+ summary.append(f"_Analyzed live in {elapsed:.0f}s._")
226
+ summary += [f"- {n}" for n in notes]
227
+
228
+ top = report.head(15)[["file", "danger"]].iloc[::-1]
229
+
230
+ csv_path = Path(tempfile.mkdtemp(prefix="techdebt_report_")) / f"techdebt_{owner}_{name}.csv"
231
+ report.to_csv(csv_path, index=False)
232
+
233
+ shown = report[list(SHOWN_COLUMNS)].round(2).rename(columns=SHOWN_COLUMNS)
234
+ return "\n\n".join(summary), top, shown, str(csv_path)
235
+
236
+
237
+ def run(url: str, progress=gr.Progress()):
238
+ owner, name = parse_github_url(url)
239
+
240
+ cached = example_cache_path(owner, name)
241
+ if cached.exists():
242
+ return build_outputs(pd.read_csv(cached), owner, name, [], None)
243
+
244
+ started = time.monotonic()
245
+ with tempfile.TemporaryDirectory(prefix="techdebt_") as tmp:
246
+ progress(0, desc=f"Cloning {owner}/{name}")
247
+ clone_repo(owner, name, Path(tmp) / name)
248
+ report, notes = analyze_repo(Path(tmp) / name, name, progress)
249
+
250
+ return build_outputs(report, owner, name, notes, time.monotonic() - started)
251
+
252
+
253
+ DESCRIPTION = """
254
+ # TechDebt ML
255
+
256
+ Predicts, **per file**, the probability that a Python file carries technical debt.
257
+ It combines code metrics (Radon + AST), the file's git activity relative to its own repo,
258
+ and embeddings of its most complex functions (`all-MiniLM-L6-v2`), fed to a Random Forest
259
+ trained on 2,503 files from 22 public repos.
260
+
261
+ Paste a **public GitHub repo** with Python code. Live analysis clones the full history,
262
+ so it takes from ~10s to a couple of minutes depending on the repo.
263
+ [Source code](https://github.com/Chillipeeper1/techdebt-ml) ·
264
+ [Known limitations](https://github.com/Chillipeeper1/techdebt-ml/tree/main/docs/decisions)
265
+ """
266
+
267
+ with gr.Blocks(title="TechDebt ML") as demo:
268
+ gr.Markdown(DESCRIPTION)
269
+ with gr.Row():
270
+ url = gr.Textbox(label="GitHub repo URL", placeholder="https://github.com/owner/repo", scale=4)
271
+ button = gr.Button("Analyze", variant="primary", scale=1)
272
+ gr.Examples(EXAMPLE_REPOS, inputs=url, label="Try one (pre-computed, instant)")
273
+
274
+ summary = gr.Markdown()
275
+ chart = gr.BarPlot(x="file", y="danger", label="Top 15 riskiest files",
276
+ y_lim=[0, 1], x_label_angle=-45, sort="-y", height=380)
277
+ table = gr.Dataframe(label="All analyzed files (highest risk first)")
278
+ gr.Markdown(
279
+ "CC = cyclomatic complexity · MI = maintainability index · "
280
+ "**×** = relative to the repo's median file (2.0 = twice the typical file). "
281
+ "The CSV has every feature the model uses."
282
+ )
283
+ download = gr.File(label="Full report (CSV)")
284
+
285
+ gr.Markdown(
286
+ "A high score is a prompt to look closer, not a verdict. "
287
+ "Files in folders named `test` are skipped, same as the CLI."
288
+ )
289
+
290
+ outputs = [summary, chart, table, download]
291
+ button.click(run, inputs=url, outputs=outputs)
292
+ url.submit(run, inputs=url, outputs=outputs)
293
+
294
+ if __name__ == "__main__":
295
+ demo.queue(default_concurrency_limit=2).launch()
examples/Chillipeeper1__techdebt-ml.csv ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ file,danger,loc,sloc,comment_ratio,blank_ratio,cyclomatic_complexity,max_complexity,maintainability_index,halstead_effort,num_functions,avg_function_len,max_nesting_depth,is_tested,dependency_count,commit_count,author_count,days_since_change,churn_rate,repo
2
+ src/predict.py,0.57,281,205,0.046263345195729534,0.2277580071174377,3.6,7,41.39601673862559,4930.789048557579,10,20.8,4,0,17,1.5,1.0,0.9746835443037974,2.890625,techdebt-ml
3
+ src/repo_search.py,0.52,292,195,0.09931506849315068,0.2328767123287671,3.111111111111111,6,51.947356266687024,3574.2223636474364,9,24.22222222222222,3,0,7,1.5,1.0,1.0,2.6875,techdebt-ml
4
+ tests/test_feature_lib.py,0.45,160,112,0.025,0.29375,3.0,5,48.57537705455356,1728.7502218695658,12,2.5,1,1,5,1.0,1.0,1.0,1.5625,techdebt-ml
5
+ src/feature_extraction.py,0.4,78,63,0.14102564102564102,0.08974358974358974,17.0,17,64.11771733993977,1282.0406359950857,1,70.0,6,0,5,1.5,1.0,1.0,2.5520833333333335,techdebt-ml
6
+ src/source_repos.py,0.25,15,11,0.0,0.26666666666666666,0.0,0,63.9732329289583,82.49268997791475,0,0.0,0,0,2,0.5,1.0,1.0,0.078125,techdebt-ml
7
+ src/feature_lib.py,0.19,190,152,0.07894736842105263,0.17894736842105263,2.4,6,47.82040322245754,2994.7935319735184,15,15.333333333333334,3,0,5,1.0,1.0,0.9746835443037974,1.0,techdebt-ml
8
+ src/file_embedding.py,0.17,153,112,0.0,0.2679738562091503,0.0,0,35.91452980437473,1721.5407704102943,0,0.0,3,0,9,0.5,1.0,0.9746835443037974,0.796875,techdebt-ml
9
+ src/hf_upload.py,0.15,27,25,0.0,0.07407407407407407,0.0,0,65.3795898712936,35.19781364344408,0,0.0,0,0,2,0.5,1.0,1.0,0.140625,techdebt-ml
10
+ src/embedding_test.py,0.08,32,27,0.0,0.15625,1.0,1,67.2293414982706,15.509775004326936,1,2.0,1,0,2,0.5,1.0,0.9746835443037974,0.16666666666666666,techdebt-ml
examples/pallets__click.csv ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ file,danger,loc,sloc,comment_ratio,blank_ratio,cyclomatic_complexity,max_complexity,maintainability_index,halstead_effort,num_functions,avg_function_len,max_nesting_depth,is_tested,dependency_count,commit_count,author_count,days_since_change,churn_rate,repo
2
+ tests/test_compat.py,0.99,270,225,0.13703703703703704,0.12222222222222222,3.0833333333333335,4,58.38203460551011,622.505329527493,12,1.0,2,1,9,4.0,2.6666666666666665,1.0,3.472,click
3
+ src/click/shell_completion.py,0.96,801,454,0.22971285892634208,0.19850187265917604,2.575,10,43.57881522831033,7886.297733279147,40,19.175,6,0,18,13.75,10.0,0.8387096774193549,12.904,click
4
+ src/click/_winconsole.py,0.96,292,223,0.06164383561643835,0.1917808219178082,2.16,7,43.50839835299815,2904.565396356504,25,10.56,4,0,29,5.75,3.0,0.43010752688172044,9.488,click
5
+ src/click/_compat.py,0.95,590,411,0.09830508474576272,0.2,2.510204081632653,21,32.41820208606701,7642.443443862683,49,11.755102040816327,4,0,14,10.5,4.0,0.5806451612903226,23.784,click
6
+ src/click/decorators.py,0.95,627,300,0.2966507177033493,0.22966507177033493,2.1363636363636362,10,52.746122309232156,5263.344085907805,22,24.272727272727273,6,0,15,14.25,6.333333333333333,1.021505376344086,14.688,click
7
+ tests/test_termui.py,0.95,1984,1477,0.1159274193548387,0.17036290322580644,2.6530612244897958,17,10.785013892429287,24431.874907066154,98,5.091836734693878,4,1,21,16.75,8.666666666666666,0.5591397849462365,25.416,click
8
+ src/click/testing.py,0.95,798,451,0.2731829573934837,0.17794486215538846,2.3902439024390243,19,40.971086579830434,6074.847468633201,41,31.804878048780488,6,0,20,12.0,6.666666666666667,0.8817204301075269,13.432,click
9
+ tests/test_shell_completion.py,0.94,617,482,0.05348460291734198,0.16531604538087522,3.5714285714285716,9,23.28086307554391,9656.210246782257,42,1.6428571428571428,2,1,19,7.25,6.666666666666667,0.8817204301075269,7.824,click
10
+ tests/test_options.py,0.94,3945,3217,0.10266159695817491,0.13079847908745248,3.5970149253731343,13,0.0,50585.12683044013,134,3.425373134328358,4,1,16,34.75,17.333333333333332,0.1827956989247312,49.52,click
11
+ src/click/utils.py,0.94,676,317,0.3210059171597633,0.21005917159763313,2.764705882352941,12,45.35208196896837,6235.585301901369,34,20.941176470588236,4,0,26,15.0,7.333333333333333,0.3118279569892473,12.656,click
12
+ src/click/termui.py,0.92,1024,447,0.384765625,0.177734375,4.148148148148148,21,41.19924554476005,7104.835440791835,27,33.2962962962963,5,0,31,20.25,11.666666666666666,0.46236559139784944,22.784,click
13
+ src/click/types.py,0.92,1437,823,0.22894919972164232,0.20250521920668058,2.6116504854368934,18,16.379594512263477,18489.539878910222,103,19.54368932038835,6,0,26,21.75,11.333333333333334,0.34408602150537637,31.64,click
14
+ tests/test_testing.py,0.91,840,571,0.08333333333333333,0.23095238095238096,4.368421052631579,15,18.052590520986374,11612.581927998759,38,3.4473684210526314,4,1,14,11.75,8.0,0.8817204301075269,9.728,click
15
+ src/click/_termui_impl.py,0.91,972,610,0.2037037037037037,0.1697530864197531,3.978723404255319,19,19.87412789579279,39746.57621505823,47,28.23404255319149,6,0,38,17.5,9.0,0.5591397849462365,26.088,click
16
+ tests/test_arguments.py,0.9,1229,920,0.06265256305939788,0.19772172497965826,3.4098360655737703,8,10.38871981322617,20350.007984101718,61,2.639344262295082,4,1,9,16.0,6.666666666666667,0.1827956989247312,15.552,click
17
+ tests/test_commands.py,0.9,700,538,0.06428571428571428,0.19142857142857142,3.962962962962963,26,27.58020402791504,6429.701108987239,27,2.185185185185185,4,1,4,11.5,6.666666666666667,0.7634408602150538,10.832,click
18
+ tests/test_basic.py,0.89,986,733,0.0486815415821501,0.21196754563894524,4.476190476190476,11,10.295207289577604,24869.389698705712,42,2.0476190476190474,3,1,9,14.0,6.0,0.3763440860215054,16.8,click
19
+ src/click/core.py,0.88,4038,1975,0.33407627538385337,0.17855373947498762,4.22360248447205,35,0.0,139100.5450469338,161,43.192546583850934,10,0,64,64.75,25.666666666666668,0.1827956989247312,112.72,click
20
+ src/click/exceptions.py,0.87,378,229,0.1455026455026455,0.24074074074074073,2.657142857142857,13,46.9906608001246,4580.507679657695,35,14.714285714285714,5,0,14,9.0,5.333333333333333,1.4838709677419355,6.912,click
21
+ src/click/parser.py,0.84,533,331,0.22889305816135083,0.16135084427767354,4.375,10,40.79188795676464,11417.608584709487,24,32.666666666666664,7,0,20,7.5,4.666666666666667,1.6666666666666667,12.584,click
22
+ tests/test_formatting.py,0.84,632,472,0.1518987341772152,0.13765822784810128,2.9565217391304346,8,45.653005905441425,2794.6177056740407,23,2.130434782608696,3,1,3,8.0,6.0,1.0,8.528,click
23
+ src/click/__init__.py,0.83,144,121,0.0,0.11805555555555555,7.0,7,57.005748291254356,46.285714285714285,1,69.0,2,0,72,11.75,2.6666666666666665,0.9139784946236559,5.248,click
24
+ src/click/formatting.py,0.83,320,197,0.1875,0.175,2.6470588235294117,7,51.91024993233019,4989.098430337277,17,27.0,5,0,8,5.5,3.0,1.4838709677419355,4.624,click
25
+ tests/test_defaults.py,0.83,567,363,0.19576719576719576,0.19753086419753085,4.388888888888889,8,43.09415859790043,7408.418224384461,18,1.1666666666666667,4,1,6,5.25,2.3333333333333335,0.41935483870967744,5.792,click
26
+ docs/conf.py,0.82,59,50,0.05084745762711865,0.1016949152542373,0.0,0,100.0,0.0,0,0.0,0,0,2,6.75,2.3333333333333335,1.4408602150537635,6.56,click
27
+ tests/test_context.py,0.8,782,557,0.05370843989769821,0.2289002557544757,3.5833333333333335,9,26.13343350156297,8265.68566073093,24,6.625,4,1,12,7.75,4.333333333333333,3.5698924731182795,9.432,click
28
+ examples/termui/termui.py,0.78,169,129,0.005917159763313609,0.17159763313609466,2.7,8,42.253629046745274,855.5566984373538,10,12.6,5,0,4,4.0,0.6666666666666666,21.462365591397848,2.744,click
29
+ tests/test_chain.py,0.77,245,180,0.0,0.23673469387755103,2.7,3,42.0840388193055,313.33977418572846,10,1.3,4,1,3,5.0,2.0,5.376344086021505,5.352,click
30
+ examples/imagepipe/imagepipe.py,0.74,288,199,0.10416666666666667,0.17708333333333334,2.888888888888889,5,48.95065720286186,1761.5766022803014,18,9.777777777777779,4,0,5,2.25,0.6666666666666666,21.580645161290324,4.0,click
31
+ examples/repo/repo.py,0.72,166,113,0.1686746987951807,0.15060240963855423,2.1,5,67.33204405466785,193.39108448262166,10,9.9,3,0,4,3.75,0.6666666666666666,1.8387096774193548,3.088,click
32
+ tests/test_info_dict.py,0.71,345,319,0.002898550724637681,0.06376811594202898,2.5,5,46.01173723486477,179.29323028347633,8,1.0,2,1,2,2.0,2.3333333333333335,0.3225806451612903,3.016,click
33
+ examples/aliases/aliases.py,0.7,143,86,0.11888111888111888,0.22377622377622378,1.875,7,66.73409260824235,133.25742227201613,16,9.0625,4,0,3,4.0,1.6666666666666667,1.8387096774193548,2.192,click
34
+ tests/test_imports.py,0.69,80,70,0.0,0.125,5.0,5,62.329586036881174,48.6,1,11.0,3,1,4,7.75,4.333333333333333,1.8387096774193548,1.88,click
35
+ examples/complex/complex/cli.py,0.62,60,43,0.0,0.23333333333333334,2.5,4,51.55307621578065,76.10749561002055,8,7.5,4,0,3,2.75,1.0,12.344086021505376,1.04,click
36
+ src/click/_textwrap.py,0.61,188,131,0.0,0.18085106382978725,8.833333333333334,31,50.21074725034268,11637.307462280256,6,52.166666666666664,8,0,6,1.75,1.0,1.6236559139784945,1.696,click
37
+ examples/validation/validation.py,0.56,48,34,0.0,0.1875,3.25,4,74.46900661627932,563.1625072679101,4,10.5,4,0,2,3.0,1.3333333333333333,1.7849462365591398,0.96,click
38
+ tests/conftest.py,0.51,8,5,0.0,0.375,1.0,1,100.0,0.0,1,2.0,1,0,2,2.75,1.3333333333333333,12.827956989247312,3.536,click
39
+ src/click/globals.py,0.47,67,28,0.0,0.3283582089552239,1.6666666666666667,3,84.63937839905921,54.0,6,7.333333333333333,3,0,4,2.75,1.0,10.010752688172044,1.0,click
40
+ tests/test_normalization.py,0.45,63,48,0.015873015873015872,0.2222222222222222,4.0,8,59.129075445325064,195.34452978042594,3,1.0,2,1,1,1.5,1.3333333333333333,6.838709677419355,0.856,click
41
+ tests/test_deprecations.py,0.38,255,213,0.043137254901960784,0.13333333333333333,1.6153846153846154,3,56.198721731289254,51.2,13,2.923076923076923,3,1,8,1.25,0.6666666666666666,0.1827956989247312,2.28,click
42
+ examples/colors/colors.py,0.35,39,29,0.0,0.15384615384615385,4.0,4,100.0,0.0,1,15.0,2,0,1,1.75,1.0,22.182795698924732,0.808,click
43
+ tests/test_stream_lifecycle.py,0.33,540,323,0.11666666666666667,0.24444444444444444,3.5185185185185186,7,36.49326840190735,4568.884549703529,27,5.111111111111111,3,1,13,0.5,0.3333333333333333,1.4623655913978495,4.336,click
44
+ tests/test_utils/test_echo.py,0.33,172,125,0.046511627906976744,0.21511627906976744,7.0,25,44.311102246774745,574.3048115887916,6,5.333333333333333,2,1,7,0.5,0.6666666666666666,1.010752688172043,1.552,click
45
+ examples/naval/naval.py,0.32,72,43,0.0,0.2361111111111111,1.0,1,100.0,0.0,8,2.625,1,0,1,1.5,0.6666666666666666,25.419354838709676,1.216,click
46
+ examples/inout/inout.py,0.3,30,12,0.0,0.16666666666666666,4.0,4,97.41779417468767,1.0,1,24.0,4,0,1,1.25,1.3333333333333333,25.903225806451612,0.416,click
47
+ tests/test_abort_interrupt.py,0.29,179,86,0.0,0.29608938547486036,1.8461538461538463,5,68.53298782365535,321.0894851215667,13,8.538461538461538,3,1,3,0.25,0.3333333333333333,0.43010752688172044,1.432,click
48
+ tests/test_utils/test_echo_via_pager.py,0.28,259,211,0.09266409266409266,0.138996138996139,2.0,5,60.69927824428672,152.92539048396907,8,6.0,2,1,10,0.5,0.6666666666666666,0.5806451612903226,2.12,click
49
+ tests/test_types/test_Path.py,0.28,190,138,0.0,0.22631578947368422,2.2222222222222223,3,52.18810870570842,595.1026719031277,9,9.777777777777779,2,1,9,0.5,0.6666666666666666,0.34408602150537637,1.52,click
50
+ src/click/_utils.py,0.26,45,18,0.0,0.35555555555555557,1.2,2,100.0,0.0,5,6.0,2,0,3,1.0,0.6666666666666666,0.45161290322580644,0.44,click
51
+ tests/test_utils/test_open_file.py,0.25,117,83,0.017094017094017096,0.27350427350427353,2.25,5,48.81880582209984,497.809234850837,8,2.75,2,1,6,0.5,0.6666666666666666,0.8817204301075269,1.848,click
52
+ examples/complex/complex/commands/cmd_status.py,0.25,11,7,0.0,0.2727272727272727,1.0,1,100.0,0.0,1,4.0,1,0,2,1.25,1.3333333333333333,25.903225806451612,0.2,click
53
+ tests/typing/typing_path.py,0.23,34,23,0.08823529411764706,0.23529411764705882,0.0,0,78.17971375002185,42.793987519471216,0,0.0,0,0,4,0.25,0.3333333333333333,0.34408602150537637,0.272,click
54
+ tests/test_utils/test_style.py,0.22,127,120,0.05511811023622047,0.05511811023622047,2.0,3,77.40470878002992,12.63309714925922,3,1.6666666666666667,1,1,2,0.75,0.6666666666666666,1.0,1.032,click
55
+ tests/test_types/test_convert_type.py,0.21,108,71,0.23148148148148148,0.17592592592592593,2.4,3,82.73801938511949,39.863137138648355,5,2.6,1,1,4,0.25,0.3333333333333333,0.3763440860215054,0.864,click
56
+ tests/test_types/test_File.py,0.19,31,21,0.25806451612903225,0.16129032258064516,1.0,1,100.0,2.3774437510817346,1,19.0,1,1,2,0.25,0.3333333333333333,0.3763440860215054,0.248,click
57
+ examples/completion/completion.py,0.19,56,32,0.08928571428571429,0.3392857142857143,1.7142857142857142,4,76.65469465101006,43.47325716263743,7,3.857142857142857,3,0,3,0.75,0.3333333333333333,21.096774193548388,0.512,click
58
+ tests/test_command_decorators.py,0.17,92,68,0.0,0.2608695652173913,3.4,5,44.1786456575496,192.64744592513122,5,1.0,2,1,2,1.0,1.0,12.204301075268818,0.736,click
59
+ tests/test_utils/test_sentinel.py,0.15,97,75,0.07216494845360824,0.14432989690721648,5.666666666666667,11,65.77598084172764,655.3945306455817,3,1.0,2,1,8,0.5,0.6666666666666666,0.45161290322580644,0.776,click
60
+ tests/test_types/test_FloatRange.py,0.14,42,32,0.0,0.23809523809523808,1.6666666666666667,2,64.41888735626253,15.509775004326936,3,3.6666666666666665,1,1,2,0.25,0.3333333333333333,0.3763440860215054,0.336,click
61
+ tests/test_utils/test_KeepOpenFile.py,0.13,10,8,0.0,0.2,3.0,3,75.15517382051522,2.3774437510817346,1,6.0,2,1,1,0.5,0.6666666666666666,0.9247311827956989,0.096,click
62
+ tests/typing/typing_prompt.py,0.13,50,24,0.16,0.36,1.25,2,92.91946736989439,2.3774437510817346,4,3.75,2,0,2,0.25,0.3333333333333333,0.8602150537634409,0.4,click
63
+ tests/test_types/test_Choice.py,0.12,7,5,0.0,0.2857142857142857,2.0,2,79.7423427057672,2.3774437510817346,1,1.0,1,1,1,0.25,0.3333333333333333,0.3763440860215054,0.056,click
64
+ tests/test_types/test_ParamType.py,0.12,24,13,0.0,0.2916666666666667,3.0,3,88.79358865017207,83.25989365817458,2,1.0,3,1,2,0.25,0.3333333333333333,0.3763440860215054,0.192,click
65
+ tests/test_exceptions/test_FileError.py,0.12,6,4,0.0,0.3333333333333333,2.0,2,81.85633424453339,2.3774437510817346,1,1.0,1,1,1,0.25,0.3333333333333333,0.3763440860215054,0.048,click
66
+ tests/test_custom_classes.py,0.12,112,63,0.0,0.2767857142857143,2.3333333333333335,3,70.49873164207317,195.40466561840492,6,1.0,3,1,1,0.25,0.3333333333333333,24.236559139784948,0.896,click
67
+ examples/complex/complex/commands/cmd_init.py,0.11,13,9,0.0,0.23076923076923078,2.0,2,73.1756852057256,2.3774437510817346,1,5.0,2,0,2,1.25,1.0,25.419354838709676,0.232,click
68
+ tests/test_types/test_FuncParamType.py,0.11,22,16,0.0,0.2727272727272727,2.0,2,73.1756852057256,2.3774437510817346,1,1.0,2,1,2,0.25,0.3333333333333333,0.3763440860215054,0.176,click
69
+ tests/test_utils/test_make_default_short_help.py,0.11,63,57,0.015873015873015872,0.07936507936507936,4.0,4,76.3330396377789,15.509775004326936,1,1.0,2,1,2,0.75,1.0,0.3118279569892473,0.552,click
70
+ tests/test_utils/test__detect_program_name.py,0.11,27,20,0.0,0.25925925925925924,1.6666666666666667,2,73.041182281749,2.3774437510817346,3,2.6666666666666665,2,1,3,0.25,0.3333333333333333,1.021505376344086,0.216,click
71
+ tests/test_types/test_IntRange.py,0.1,38,32,0.0,0.15789473684210525,2.0,2,69.4458918400094,15.509775004326936,2,1.0,1,1,2,0.25,0.3333333333333333,0.3763440860215054,0.304,click
72
+ tests/typing/typing_aliased_group.py,0.1,46,30,0.021739130434782608,0.30434782608695654,2.6666666666666665,6,66.7458196448301,54.81181356564148,6,7.0,4,0,3,0.75,1.0,6.043010752688172,0.384,click
73
+ tests/typing/typing_group_kw_options.py,0.1,11,6,0.0,0.45454545454545453,1.0,1,100.0,0.0,1,2.0,1,0,2,0.25,0.3333333333333333,12.688172043010752,0.088,click
74
+ tests/typing/typing_progressbar.py,0.1,24,16,0.0,0.3333333333333333,1.6666666666666667,2,100.0,0.0,3,4.0,2,1,4,0.25,0.3333333333333333,7.537634408602151,0.192,click
75
+ tests/test_utils/test_LazyFile.py,0.1,11,9,0.0,0.18181818181818182,3.0,3,74.03933453534945,2.3774437510817346,1,7.0,2,1,1,0.5,0.6666666666666666,0.9247311827956989,0.104,click
76
+ tests/test_utils/test_prompt.py,0.09,100,62,0.12,0.25,2.1666666666666665,3,70.71511903450501,68.53238859703687,6,1.0,3,1,5,0.5,0.6666666666666666,0.8602150537634409,0.8,click
77
+ tests/typing/typing_confirmation_option.py,0.08,14,7,0.0,0.42857142857142855,1.0,1,100.0,0.0,1,2.0,1,0,2,0.75,1.0,6.043010752688172,0.128,click
78
+ tests/typing/typing_version_option.py,0.08,16,7,0.0,0.375,1.0,1,100.0,0.0,1,2.0,1,0,2,0.75,1.0,6.043010752688172,0.144,click
79
+ tests/test_parser.py,0.08,32,25,0.0,0.21875,2.0,2,63.37475596230147,12.63309714925922,3,1.0,1,1,4,0.75,1.0,12.333333333333334,0.32,click
80
+ tests/typing/typing_password_option.py,0.06,14,8,0.0,0.42857142857142855,1.0,1,100.0,0.0,1,2.0,1,0,3,0.25,0.3333333333333333,12.741935483870968,0.112,click
81
+ tests/typing/typing_edit.py,0.06,10,6,0.1,0.3,0.0,0,100.0,0.0,0,0.0,0,0,3,0.25,0.3333333333333333,0.5591397849462365,0.08,click
82
+ tests/test_utils/test_confirm.py,0.06,49,38,0.0,0.22448979591836735,7.5,13,49.652669125649695,236.60752504759634,2,1.0,3,1,1,0.25,0.3333333333333333,1.021505376344086,0.392,click
83
+ tests/typing/typing_help_option.py,0.05,13,7,0.0,0.38461538461538464,1.0,1,100.0,0.0,1,3.0,1,0,2,0.25,0.3333333333333333,12.741935483870968,0.104,click
84
+ tests/typing/typing_simple_example.py,0.05,17,9,0.0,0.35294117647058826,2.0,2,100.0,0.0,1,4.0,2,0,2,0.5,0.6666666666666666,10.010752688172044,0.136,click
85
+ tests/typing/typing_options.py,0.05,16,9,0.0,0.375,2.0,2,100.0,0.0,1,3.0,2,0,2,0.75,1.0,6.043010752688172,0.144,click
86
+ tests/test_utils/test_format_filename.py,0.04,9,7,0.0,0.2222222222222222,6.0,6,69.01468614621594,29.718046888521677,1,1.0,1,1,1,0.25,0.3333333333333333,1.021505376344086,0.072,click
87
+ tests/test_utils/test__expand_args.py,0.04,17,13,0.058823529411764705,0.17647058823529413,7.0,7,79.3806344361897,68.53238859703687,1,1.0,1,1,2,0.25,0.3333333333333333,1.021505376344086,0.136,click
88
+ tests/test_types/__init__.py,0.02,0,0,0.0,0.0,0.0,0,100.0,0.0,0,0.0,0,0,0,0.25,0.3333333333333333,0.3763440860215054,0.0,click
89
+ tests/test_exceptions/__init__.py,0.02,0,0,0.0,0.0,0.0,0,100.0,0.0,0,0.0,0,0,0,0.25,0.3333333333333333,0.3763440860215054,0.0,click
90
+ tests/test_utils/__init__.py,0.0,0,0,0.0,0.0,0.0,0,100.0,0.0,0,0.0,0,0,0,0.25,0.3333333333333333,1.021505376344086,0.0,click
91
+ examples/complex/complex/__init__.py,0.0,0,0,0.0,0.0,0.0,0,100.0,0.0,0,0.0,0,0,0,0.25,0.3333333333333333,48.645161290322584,0.0,click
92
+ examples/complex/complex/commands/__init__.py,0.0,0,0,0.0,0.0,0.0,0,100.0,0.0,0,0.0,0,0,0,0.25,0.3333333333333333,48.645161290322584,0.0,click
examples/psf__requests.csv ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ file,danger,loc,sloc,comment_ratio,blank_ratio,cyclomatic_complexity,max_complexity,maintainability_index,halstead_effort,num_functions,avg_function_len,max_nesting_depth,is_tested,dependency_count,commit_count,author_count,days_since_change,churn_rate,repo
2
+ src/requests/models.py,0.91,1184,720,0.23057432432432431,0.16554054054054054,3.9285714285714284,21,16.424806214955883,23812.207112680357,56,35.05357142857143,8,0,28,3.1666666666666665,3.0,0.76875,8.75,requests
3
+ src/requests/utils.py,0.9,1155,637,0.24935064935064935,0.2077922077922078,4.413043478260869,19,22.5697327312295,35511.65780635283,46,20.67391304347826,6,0,31,3.1666666666666665,3.5,0.775,8.343023255813954,requests
4
+ tests/test_utils.py,0.89,1013,830,0.044422507403751234,0.1362290227048371,2.324675324675325,6,18.409681277306113,9958.25440156326,77,5.64935064935065,5,1,18,12.0,15.5,0.95,14.290697674418604,requests
5
+ tests/test_requests.py,0.87,3094,2440,0.048157724628312866,0.16063348416289594,2.859437751004016,8,0.0,109553.96156774365,249,14.012048192771084,7,1,32,45.833333333333336,47.0,0.76875,58.74418604651163,requests
6
+ src/requests/adapters.py,0.87,748,424,0.28475935828877,0.15240641711229946,3.8636363636363638,20,46.100395693040014,1842.6083666723753,22,55.54545454545455,6,0,28,3.0,3.5,0.775,6.034883720930233,requests
7
+ docs/conf.py,0.82,385,84,0.5584415584415584,0.24415584415584415,0.0,0,100.0,0.0,0,0.0,0,0,3,6.666666666666667,7.5,1.1375,4.738372093023256,requests
8
+ setup.py,0.79,9,6,0.1111111111111111,0.3333333333333333,0.0,0,99.90348152172656,2.3774437510817346,0,0.0,1,0,2,29.0,21.5,1.5375,6.773255813953488,requests
9
+ tests/test_lowlevel.py,0.79,428,284,0.10514018691588785,0.23130841121495327,3.4285714285714284,6,44.153518212593404,2802.577552678092,14,21.928571428571427,2,1,6,4.5,6.0,1.58125,4.494186046511628,requests
10
+ tests/testserver/server.py,0.73,176,137,0.045454545454545456,0.18181818181818182,2.2,4,52.34542381192951,190.66666666666666,15,19.8,4,0,4,2.5,4.0,5.875,1.7674418604651163,requests
11
+ src/requests/sessions.py,0.7,920,479,0.30869565217391304,0.17608695652173914,4.225806451612903,15,35.48718644321636,10488.742764470222,31,49.61290322580645,5,0,24,1.8333333333333333,3.0,0.775,6.372093023255814,requests
12
+ tests/test_testserver.py,0.69,165,112,0.04242424242424243,0.21818181818181817,2.4166666666666665,4,49.208813990549096,111.4694169837047,12,18.333333333333332,3,1,6,2.8333333333333335,3.0,1.58125,2.0174418604651163,requests
13
+ src/requests/__init__.py,0.54,219,137,0.2237442922374429,0.1598173515981735,6.5,10,69.51805483330125,455.0171144033819,2,23.0,3,0,19,1.5,2.5,0.775,1.680232558139535,requests
14
+ src/requests/_types.py,0.51,188,126,0.11702127659574468,0.2127659574468085,1.4166666666666667,3,52.3424609035083,5317.0824764809,12,4.5,2,0,10,1.1666666666666667,1.0,0.76875,1.2209302325581395,requests
15
+ tests/test_structures.py,0.51,91,69,0.0,0.21978021978021978,2.1176470588235294,3,41.14719594099197,520.9561949440348,17,6.235294117647059,2,1,2,1.1666666666666667,2.0,1.01875,1.680232558139535,requests
16
+ src/requests/__version__.py,0.5,14,10,0.21428571428571427,0.07142857142857142,0.0,0,100.0,0.0,0,0.0,0,0,0,2.3333333333333335,1.0,0.93125,0.36046511627906974,requests
17
+ tests/conftest.py,0.48,67,43,0.05970149253731343,0.2835820895522388,1.2,2,72.68279430886432,15.509775004326936,5,8.2,2,0,8,1.0,2.0,0.95625,0.5174418604651163,requests
18
+ src/requests/status_codes.py,0.38,128,100,0.1640625,0.09375,6.0,6,83.6791512823983,112.92631877271643,1,16.0,4,0,1,0.6666666666666666,1.5,1.0,0.8255813953488372,requests
19
+ src/requests/auth.py,0.38,354,238,0.11864406779661017,0.20056497175141244,2.739130434782609,19,42.76044354466967,7330.84285060853,23,23.47826086956522,8,0,16,0.8333333333333334,1.0,0.775,2.604651162790698,requests
20
+ src/requests/packages.py,0.38,23,15,0.17391304347826086,0.17391304347826086,0.0,0,86.66611985445543,89.62406251802892,0,0.0,3,0,2,1.0,1.5,5.49375,0.4011627906976744,requests
21
+ src/requests/exceptions.py,0.36,162,45,0.18518518518518517,0.3950617283950617,1.2857142857142858,5,73.32411200242946,196.0,28,3.9285714285714284,3,0,5,0.6666666666666666,1.0,1.0,1.0,requests
22
+ src/requests/cookies.py,0.32,625,343,0.2624,0.1952,2.3214285714285716,9,38.08781688720079,9077.945691552755,56,15.375,7,0,13,0.5,1.0,0.775,4.563953488372093,requests
23
+ src/requests/compat.py,0.3,115,66,0.2608695652173913,0.1826086956521739,4.0,4,79.26713569292579,39.863137138648355,1,10.0,4,0,18,1.1666666666666667,1.0,0.775,0.8546511627906976,requests
24
+ src/requests/api.py,0.29,180,38,0.5611111111111111,0.22777777777777777,1.0,1,100.0,0.0,8,17.875,1,0,6,0.3333333333333333,0.5,1.0,1.1395348837209303,requests
25
+ tests/__init__.py,0.25,14,6,0.2857142857142857,0.21428571428571427,0.0,0,100.0,0.0,0,0.0,1,0,2,1.1666666666666667,2.5,8.4875,0.3372093023255814,requests
26
+ src/requests/help.py,0.23,132,99,0.11363636363636363,0.16666666666666666,4.333333333333333,6,61.949053948140005,550.713459778579,3,29.333333333333332,5,0,13,0.8333333333333334,1.0,0.775,1.0348837209302326,requests
27
+ tests/compat.py,0.2,23,19,0.0,0.17391304347826086,1.0,1,100.0,0.0,1,10.0,1,0,4,0.6666666666666666,1.0,10.15625,0.2616279069767442,requests
28
+ src/requests/_internal_utils.py,0.17,51,25,0.0,0.21568627450980393,2.5,3,100.0,0.0,2,12.0,2,0,2,0.6666666666666666,0.5,1.0,0.36627906976744184,requests
29
+ src/requests/structures.py,0.16,130,62,0.26153846153846155,0.26153846153846155,1.368421052631579,2,76.39007946622573,42.793987519471216,19,8.68421052631579,3,0,5,0.3333333333333333,0.5,1.0,0.9883720930232558,requests
30
+ tests/test_adapters.py,0.16,8,5,0.0,0.25,2.0,2,78.01508585192973,2.3774437510817346,1,1.0,1,1,1,0.3333333333333333,1.0,0.95625,0.06976744186046512,requests
31
+ src/requests/certs.py,0.15,18,3,0.5555555555555556,0.2777777777777778,0.0,0,64.3595984114544,2.3774437510817346,0,0.0,1,0,1,0.3333333333333333,0.5,1.58125,0.10465116279069768,requests
32
+ docs/_themes/flask_theme_support.py,0.12,86,73,0.7674418604651163,0.13953488372093023,1.0,1,100.0,0.0,1,79.0,1,0,2,0.16666666666666666,0.5,35.16875,0.5,requests
33
+ tests/utils.py,0.1,17,15,0.0,0.11764705882352941,3.0,3,69.19993388914534,2.3774437510817346,1,12.0,3,0,2,0.5,1.0,10.15625,0.12209302325581395,requests
34
+ tests/test_help.py,0.09,27,13,0.0,0.3333333333333333,1.8,2,86.45398139061467,27.0,5,2.8,2,1,2,0.8333333333333334,1.0,7.2125,0.47093023255813954,requests
35
+ src/requests/hooks.py,0.09,48,25,0.1875,0.2708333333333333,4.0,6,84.90733657858264,68.33680652339717,2,9.5,4,0,5,0.5,0.5,1.0,0.38372093023255816,requests
36
+ tests/test_packages.py,0.06,13,7,0.0,0.46153846153846156,1.0,1,100.0,0.0,3,2.0,1,1,1,0.16666666666666666,0.5,21.38125,0.0755813953488372,requests
37
+ tests/test_hooks.py,0.05,22,15,0.0,0.3181818181818182,1.6666666666666667,2,67.28666784000676,6.965784284662087,3,1.3333333333333333,1,1,2,0.5,1.0,10.15625,0.23255813953488372,requests
38
+ tests/testserver/__init__.py,0.0,0,0,0.0,0.0,0.0,0,100.0,0.0,0,0.0,0,0,0,0.16666666666666666,0.5,24.20625,0.0,requests
fast_git.py ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """
2
+ fast_git.py - same output as feature_lib.feature_git_raw, but fast on big repos.
3
+
4
+ feature_git_raw calls commit.stats (one `git diff` per commit) for every commit of
5
+ every file, so a commit touching 50 files is diffed 50 times. On a repo with
6
+ thousands of commits that takes many minutes. Here:
7
+
8
+ * churn: ONE `git log --numstat` over the whole history, diffing each commit
9
+ against its first parent (or the empty tree for root commits) with renames
10
+ off, which is exactly what GitPython's commit.stats does.
11
+ * per-file history: one `git log -- <file>` per file, which applies the same
12
+ history simplification as repo.iter_commits(paths=file).
13
+
14
+ Raw author names (%an, no mailmap) and the committer-local date (%cI) match
15
+ commit.author.name and commit.committed_datetime.
16
+ """
17
+
18
+ import subprocess
19
+ from concurrent.futures import ThreadPoolExecutor
20
+ from datetime import date
21
+ from pathlib import Path
22
+
23
+
24
+ def _git(repo_path: Path, *args: str) -> str:
25
+ return subprocess.run(
26
+ ["git", "-C", str(repo_path), *args],
27
+ check=True, capture_output=True, text=True, encoding="utf-8", errors="replace",
28
+ ).stdout
29
+
30
+
31
+ def churn_by_commit(repo_path: Path) -> dict[str, dict[str, int]]:
32
+ out = _git(
33
+ repo_path, "log", "HEAD", "--no-renames", "--diff-merges=first-parent",
34
+ "--numstat", "--format=%x00%H",
35
+ )
36
+ churn = {}
37
+ for block in out.split("\x00")[1:]:
38
+ lines = block.strip("\n").split("\n")
39
+ files = {}
40
+ for line in lines[1:]:
41
+ parts = line.split("\t", 2)
42
+ if len(parts) != 3:
43
+ continue
44
+ insertions, deletions, path = parts
45
+ # GitPython counts binary files ("-") as 0
46
+ files[path] = (0 if insertions == "-" else int(insertions)) + (0 if deletions == "-" else int(deletions))
47
+ churn[lines[0]] = files
48
+ return churn
49
+
50
+
51
+ def file_history(repo_path: Path, file: str) -> list[tuple[str, str, str]]:
52
+ out = _git(repo_path, "log", "HEAD", "--format=%H%x09%cI%x09%an", "--", file)
53
+ return [tuple(line.split("\t", 2)) for line in out.splitlines() if line]
54
+
55
+
56
+ def git_features(repo_path: Path, files: list[str], workers: int = 4) -> dict[str, dict]:
57
+ churn = churn_by_commit(repo_path)
58
+ posix_files = [f.replace("\\", "/") for f in files]
59
+
60
+ with ThreadPoolExecutor(max_workers=workers) as pool:
61
+ histories = list(pool.map(lambda f: file_history(repo_path, f), posix_files))
62
+
63
+ features = {}
64
+ for file, posix, history in zip(files, posix_files, histories):
65
+ authors = set()
66
+ churn_total = 0
67
+ for sha, _, author in history:
68
+ authors.add(author)
69
+ churn_total += churn.get(sha, {}).get(posix, 0)
70
+
71
+ features[file] = {
72
+ "commit_count": len(history),
73
+ "author_count": len(authors),
74
+ "days_since_change": (date.today() - date.fromisoformat(history[0][1][:10])).days if history else 0,
75
+ "churn_rate": churn_total,
76
+ }
77
+ return features
metrics_worker.py ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """
2
+ metrics_worker.py - computes the per-file code features in a separate process.
3
+
4
+ Radon can hang on some files (e.g. a 7k-line test file in fastapi never finishes
5
+ in radon.raw.analyze), and a stuck thread can't be killed. app.py runs this
6
+ worker as a subprocess so it can kill it when a file takes too long.
7
+
8
+ Protocol: argv[1] = repo path, stdin = one relative file path per line,
9
+ stdout = one JSON object per file, in the same order.
10
+ """
11
+
12
+ import json
13
+ import sys
14
+ from pathlib import Path
15
+
16
+ HERE = Path(__file__).resolve().parent
17
+ for candidate in (HERE / "src", HERE.parent / "src"):
18
+ if candidate.is_dir():
19
+ sys.path.insert(0, str(candidate))
20
+ break
21
+
22
+ from feature_lib import embedding_extract_functions, features_ats, features_radon
23
+
24
+
25
+ def main():
26
+ repo_path = Path(sys.argv[1])
27
+ for file in sys.stdin.read().splitlines():
28
+ try:
29
+ code = (repo_path / file).read_text(encoding="utf-8")
30
+ features = {}
31
+ features.update(features_radon(code))
32
+ features.update(features_ats(code))
33
+ out = {"file": file, "features": features, "functions": embedding_extract_functions(code)}
34
+ except Exception as e:
35
+ out = {"file": file, "error": f"{type(e).__name__}: {e}"}
36
+ print(json.dumps(out, default=float), flush=True)
37
+
38
+
39
+ if __name__ == "__main__":
40
+ main()
models/features_v3.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0045765fd8165c4c9ceaf89f323d64745e2bd611c048ff3a71a0d3ece72e830d
3
+ size 291
models/pca_v3.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e29e6ee614256d9ac9bb3efcf53188224b6a93b86d088022222f9b945676cccc
3
+ size 65918
models/random_forest_v3.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2fe5bd76d6ef4d75398b6721e03a0f8d7f601ce7e5894aa82503a8751809f4e
3
+ size 2996649
requirements.txt ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ # ZeroGPU supports torch 2.8 - 2.13 (the `spaces` package is preinstalled there)
2
+ torch==2.13.0
3
+ # Must match the version the v3 .pkl artifacts were trained with
4
+ scikit-learn==1.9.0
5
+ sentence-transformers
6
+ pandas
7
+ numpy
8
+ gitpython
9
+ radon
10
+ joblib
src/feature_lib.py ADDED
@@ -0,0 +1,190 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import ast
2
+ from radon.complexity import cc_visit
3
+ from radon import metrics, raw
4
+ from datetime import date
5
+ import numpy as np
6
+
7
+ class MyAnalyzer(ast.NodeVisitor):
8
+ def __init__(self):
9
+ self.dependency_counts = 0
10
+ self.max_nesting_depth = 0
11
+ self.current_depth = 0
12
+ self.is_tested = 0
13
+
14
+ def visit_Import(self, node):
15
+ self.dependency_counts += 1
16
+ self.generic_visit(node)
17
+ def visit_ImportFrom(self, node):
18
+ self.dependency_counts += 1
19
+ self.generic_visit(node)
20
+
21
+ def visit_If(self, node):
22
+ self.current_depth += 1
23
+
24
+ if self.current_depth > self.max_nesting_depth:
25
+ self.max_nesting_depth = self.current_depth
26
+ self.generic_visit(node)
27
+ self.current_depth -= 1
28
+
29
+ def visit_For(self, node):
30
+ self.current_depth += 1
31
+
32
+ if self.current_depth > self.max_nesting_depth:
33
+ self.max_nesting_depth = self.current_depth
34
+ self.generic_visit(node)
35
+ self.current_depth -= 1
36
+ def visit_While(self, node):
37
+ self.current_depth += 1
38
+
39
+ if self.current_depth > self.max_nesting_depth:
40
+ self.max_nesting_depth = self.current_depth
41
+ self.generic_visit(node)
42
+ self.current_depth -= 1
43
+ def visit_Try(self, node):
44
+ self.current_depth += 1
45
+
46
+ if self.current_depth > self.max_nesting_depth:
47
+ self.max_nesting_depth = self.current_depth
48
+ self.generic_visit(node)
49
+ self.current_depth -= 1
50
+
51
+ def visit_FunctionDef(self, node):
52
+ self.current_depth += 1
53
+ if self.current_depth > self.max_nesting_depth:
54
+ self.max_nesting_depth = self.current_depth
55
+
56
+ if node.name.startswith("test_"):
57
+ self.is_tested = 1
58
+ self.generic_visit(node)
59
+ self.current_depth -= 1
60
+
61
+
62
+ def visit_ClassDef(self, node):
63
+ self.current_depth += 1
64
+
65
+ if self.current_depth > self.max_nesting_depth:
66
+ self.max_nesting_depth = self.current_depth
67
+ if node.name.startswith("Test"):
68
+ self.is_tested = 1
69
+ self.generic_visit(node)
70
+ self.current_depth -= 1
71
+
72
+
73
+ def features_radon(file):
74
+ result_raw_loc = raw.analyze(file).loc # feat
75
+ result_raw_sloc = raw.analyze(file).sloc # feat
76
+ result_raw_comments = raw.analyze(file).comments
77
+ result_raw_multi = raw.analyze(file).multi
78
+ result_raw_blank = raw.analyze(file).blank
79
+
80
+ comment_ratio = (
81
+ (result_raw_comments + result_raw_multi) / result_raw_loc
82
+ if result_raw_comments > 0 else 0
83
+ )
84
+ blank_ratio = result_raw_blank / result_raw_loc if result_raw_loc > 0 else 0
85
+ # Cyclomatic complexity
86
+ result_cc = cc_visit(file)
87
+ if len(result_cc) > 0:
88
+
89
+ sum_func_lines = 0
90
+ complexities = [score.complexity for score in result_cc]
91
+ max_complexity = max(complexities) # feat
92
+ average_cc = sum(complexities) / len(complexities) # feat
93
+ for score in result_cc:
94
+ func_lines = score.endline - score.lineno + 1
95
+ sum_func_lines += func_lines
96
+
97
+ num_functions = len(result_cc) # feat
98
+
99
+ avg_function_len = sum_func_lines / num_functions # feat
100
+ else:
101
+ num_functions = 0
102
+ avg_function_len = 0
103
+ average_cc = 0
104
+ max_complexity = 0
105
+
106
+ # Maintainability Index
107
+ result_mi = metrics.mi_visit(file, multi=True) # feat
108
+
109
+ # Halstead metrics (effort specifically)
110
+ result_hl = metrics.h_visit(file).total.effort # feat
111
+
112
+ return {
113
+ "loc": result_raw_loc,
114
+ "sloc": result_raw_sloc,
115
+ "comment_ratio": comment_ratio,
116
+ "blank_ratio": blank_ratio,
117
+ "cyclomatic_complexity": average_cc,
118
+ "max_complexity": max_complexity,
119
+ "maintainability_index": result_mi,
120
+ "halstead_effort": result_hl,
121
+ "num_functions": num_functions,
122
+ "avg_function_len": avg_function_len,
123
+ }
124
+
125
+ def features_ats(file):
126
+ # ast dependency counts and max nesting depth
127
+ result_ast = ast.parse(file)
128
+ analyze = MyAnalyzer()
129
+ analyze.visit(result_ast)
130
+ max_nesting_depth = analyze.max_nesting_depth # feat
131
+ dependency_counts = analyze.dependency_counts #feat
132
+ nesting_depth = analyze.current_depth
133
+ is_tested = analyze.is_tested # feat
134
+
135
+ return {
136
+ "max_nesting_depth": max_nesting_depth,
137
+ "is_tested": is_tested,
138
+ "dependency_count": dependency_counts
139
+ }
140
+
141
+
142
+
143
+ def feature_git_raw(file, repo):
144
+ author_list = set()
145
+ churn_rate_total = 0
146
+
147
+ file_commits = list(repo.iter_commits(paths=file))
148
+ commit_count = len(file_commits)
149
+
150
+ for commit in file_commits:
151
+ author_list.add(commit.author.name)
152
+ churn_rate = commit.stats.files
153
+ change_file = file.replace("\\", "/")
154
+ if change_file in churn_rate:
155
+ churn_rate_total += (
156
+ churn_rate[change_file]['deletions']
157
+ + churn_rate[change_file]['insertions']
158
+ )
159
+
160
+ author_count = len(author_list)
161
+
162
+ if commit_count > 0:
163
+ days_since_change = (date.today() - file_commits[0].committed_datetime.date()).days
164
+ else:
165
+ days_since_change = 0
166
+
167
+ return {
168
+ "commit_count": commit_count,
169
+ "author_count": author_count,
170
+ "days_since_change": days_since_change,
171
+ "churn_rate": churn_rate_total,
172
+ }
173
+
174
+ def normalization_relative_to_repo(df_features, repo_col="repo"):
175
+ relative_cols = ["commit_count", "author_count", "churn_rate", "days_since_change"]
176
+ for col in relative_cols:
177
+ medians = df_features.groupby(repo_col)[col].transform("median")
178
+ medians = medians.replace(0, 1)
179
+ df_features[col] = df_features[col] / medians
180
+ return df_features
181
+
182
+ def embedding_extract_functions(code, min_lines=2):
183
+ functions = cc_visit(code)
184
+ lines = code.splitlines()
185
+ texts = []
186
+ for f in functions:
187
+ body = "\n".join(lines[f.lineno - 1 : f.endline])
188
+ if len(body.splitlines()) >= min_lines:
189
+ texts.append(body)
190
+ return texts
src/predict.py ADDED
@@ -0,0 +1,281 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """
2
+ predict.py - analyzes a local repo and predicts the technical debt per file
3
+
4
+ Uses model v3: tabular features (Radon + AST + relative git activity) combined
5
+ with PCA-reduced code embeddings (see docs/decisions/004-embeddings-tradeoff.md).
6
+ """
7
+
8
+ import argparse
9
+ import sys
10
+ import os
11
+ import shutil
12
+ import stat
13
+ import tempfile
14
+ import re
15
+ from pathlib import Path
16
+ from datetime import datetime
17
+
18
+ import joblib
19
+ import numpy as np
20
+ import pandas as pd
21
+ from git import Repo
22
+ from git.exc import InvalidGitRepositoryError, NoSuchPathError
23
+ from sentence_transformers import SentenceTransformer
24
+
25
+ from feature_lib import (
26
+ features_ats,
27
+ features_radon,
28
+ feature_git_raw,
29
+ normalization_relative_to_repo,
30
+ embedding_extract_functions,
31
+ )
32
+
33
+ SRC_DIR = Path(__file__).resolve().parent
34
+ MODELS_DIR = SRC_DIR.parent / "models"
35
+ REPORTS_DIR = SRC_DIR.parent / "temp_reports"
36
+
37
+ if str(SRC_DIR) not in sys.path:
38
+ sys.path.insert(0, str(SRC_DIR))
39
+
40
+ # Oficial Artifacts: v3 (embeddings + PCA)
41
+ feat_list = joblib.load(MODELS_DIR / "features_v3.pkl")
42
+ MODEL = joblib.load(MODELS_DIR / "random_forest_v3.pkl")
43
+ PCA_MODEL = joblib.load(MODELS_DIR / "pca_v3.pkl")
44
+ EMBED_MODEL = SentenceTransformer("all-MiniLM-L6-v2")
45
+
46
+ CONFIG = {
47
+ "skip_dirs": ["test", ".venv"],
48
+ }
49
+
50
+
51
+ def validate_repo(path: str) -> tuple[Path, Repo]:
52
+ repo_path = Path(path).resolve()
53
+
54
+ if not repo_path.exists():
55
+ print(f"Error: the path '{repo_path}' does not exist.")
56
+ sys.exit(1)
57
+
58
+ try:
59
+ repo_git = Repo(repo_path)
60
+ except (InvalidGitRepositoryError, NoSuchPathError):
61
+ print(f"Error: '{repo_path}' is not a valid git repository.")
62
+ sys.exit(1)
63
+
64
+ return repo_path, repo_git
65
+
66
+
67
+ def find_python_files(repo_path: Path) -> list[str]:
68
+ return [
69
+ str(f.relative_to(repo_path)) for f in repo_path.rglob("*.py")
70
+ if not any(part in CONFIG["skip_dirs"] for part in f.parts)
71
+ ]
72
+
73
+
74
+ def print_summary(repo_path: Path, files: list[str]):
75
+ print("\n" + "=" * 60)
76
+ print(f" FILES FOUND IN {repo_path.name}")
77
+ print("=" * 60)
78
+ print(f" Total: {len(files)} .py files")
79
+ print("=" * 60)
80
+
81
+ for file in files[:15]:
82
+ print(f" {file}")
83
+
84
+ if len(files) > 15:
85
+ print(f" ... and {len(files) - 15} more")
86
+
87
+ print("=" * 60)
88
+
89
+
90
+ def feature_extraction(path, files, repo_git, repo_name):
91
+ resultados = {}
92
+ todas_las_funciones = []
93
+ indice_por_archivo = []
94
+
95
+ for archivo in files:
96
+ try:
97
+ with open(path / archivo, "r", encoding="utf-8") as f:
98
+ feat_file = f.read()
99
+
100
+ features = {}
101
+ features.update(features_radon(feat_file))
102
+ features.update(features_ats(feat_file))
103
+ features.update(feature_git_raw(archivo, repo_git))
104
+ features["repo"] = repo_name
105
+
106
+ # Ruta normalizada — evita el desajuste backslash/forward-slash
107
+ # entre este dict y las claves de embeddings_por_archivo.
108
+ clave = f"{repo_name}/{archivo}".replace("\\", "/")
109
+ resultados[clave] = features
110
+
111
+ funciones_archivo = embedding_extract_functions(feat_file)
112
+ indice_por_archivo.append((clave, len(funciones_archivo)))
113
+ todas_las_funciones.extend(funciones_archivo)
114
+
115
+ except Exception as e:
116
+ print(f"corrupted file: {archivo} , {e}")
117
+ continue
118
+
119
+ return resultados, todas_las_funciones, indice_por_archivo
120
+
121
+
122
+ def generar_embeddings_reducidos(todas_las_funciones, indice_por_archivo):
123
+ """Genera embeddings por función (batched), hace mean pooling por archivo.
124
+ NO aplica PCA aquí — eso ocurre en full_report, sobre la matriz completa."""
125
+ if todas_las_funciones:
126
+ embeddings_funciones = EMBED_MODEL.encode(todas_las_funciones, batch_size=64)
127
+ else:
128
+ embeddings_funciones = np.empty((0, 384))
129
+
130
+ embeddings_por_archivo = {}
131
+ cursor = 0
132
+ for clave, cantidad in indice_por_archivo:
133
+ if cantidad == 0:
134
+ embeddings_por_archivo[clave] = np.zeros(384)
135
+ continue
136
+ bloque = embeddings_funciones[cursor: cursor + cantidad]
137
+ embeddings_por_archivo[clave] = bloque.mean(axis=0)
138
+ cursor += cantidad
139
+
140
+ return embeddings_por_archivo
141
+
142
+
143
+ def full_report(results: dict, embeddings_por_archivo: dict) -> pd.DataFrame:
144
+ """Combina features tabulares (normalizados relativo al repo) con embeddings
145
+ reducidos por PCA (ya ajustado en entrenamiento — aquí solo .transform())."""
146
+ df = pd.DataFrame.from_dict(results, orient="index")
147
+ df.index.name = "file"
148
+
149
+ df = normalization_relative_to_repo(df, repo_col="repo")
150
+
151
+ X_tabular = df[feat_list].values
152
+ matriz_embeddings = np.array([embeddings_por_archivo[clave] for clave in df.index])
153
+ X_embeddings_reducido = PCA_MODEL.transform(matriz_embeddings) # transform, NUNCA fit_transform
154
+
155
+ X_final = np.hstack([X_tabular, X_embeddings_reducido])
156
+ df.insert(0, "danger", MODEL.predict_proba(X_final)[:, 1].round(3))
157
+
158
+ return df.sort_values("danger", ascending=False).reset_index()
159
+
160
+
161
+ def save_csv_report(report: pd.DataFrame, repo_name: str, destination: Path = REPORTS_DIR) -> Path:
162
+ destination.mkdir(parents=True, exist_ok=True)
163
+ clean_name = re.sub(r"[^A-Za-z0-9._-]", "_", repo_name) or "repo"
164
+ timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
165
+ csv_path = destination / f"report_{clean_name}_{timestamp}.csv"
166
+ report.to_csv(csv_path, index=False, encoding="utf-8")
167
+ return csv_path
168
+
169
+
170
+ def _delete_tree(path: Path):
171
+ def _remove_readonly(func, path, _exc_info):
172
+ os.chmod(path, stat.S_IWRITE)
173
+ func(path)
174
+
175
+ shutil.rmtree(path, onerror=_remove_readonly)
176
+
177
+
178
+ def fetch_and_clone_repo(max_attempts: int = 5):
179
+ try:
180
+ from repo_search import get_random_repo
181
+ except EnvironmentError as e:
182
+ print(f"Error: {e}")
183
+ return None
184
+
185
+ tried_urls = set()
186
+
187
+ for attempt in range(1, max_attempts + 1):
188
+ repo_info = get_random_repo()
189
+ if repo_info is None:
190
+ print(f" [{attempt}/{max_attempts}] No repo was obtained. Retrying...")
191
+ continue
192
+
193
+ clone_url = repo_info["clone_url"]
194
+ if clone_url in tried_urls:
195
+ continue
196
+ tried_urls.add(clone_url)
197
+
198
+ tmp_dir = Path(tempfile.mkdtemp(prefix="techdebt_"))
199
+ dest = tmp_dir / repo_info["name"]
200
+
201
+ try:
202
+ print(f" [{attempt}/{max_attempts}] Cloning {repo_info['full_name']} ...")
203
+ repo_git = Repo.clone_from(clone_url, dest)
204
+ except Exception as e:
205
+ print(f" Clone failed: {e}")
206
+ _delete_tree(tmp_dir)
207
+ continue
208
+
209
+ files = find_python_files(dest)
210
+ if not files:
211
+ print(" The repo has no usable .py files. Trying another one...")
212
+ repo_git.close()
213
+ _delete_tree(tmp_dir)
214
+ continue
215
+
216
+ return tmp_dir, dest, repo_git, repo_info["name"], files
217
+
218
+ return None
219
+
220
+
221
+ def main():
222
+ parser = argparse.ArgumentParser(
223
+ description="Analyzes a repository and predicts the technical debt per file. "
224
+ "Without arguments, it looks for and clones a random GitHub repo."
225
+ )
226
+ parser.add_argument(
227
+ "repo_path", nargs="?", default=None,
228
+ help="Path to a local repo (optional). If omitted, a random GitHub repo "
229
+ "is looked up and cloned.",
230
+ )
231
+ args = parser.parse_args()
232
+
233
+ tmp_dir = None
234
+ repo_git = None
235
+ try:
236
+ if args.repo_path:
237
+ repo_path, repo_git = validate_repo(args.repo_path)
238
+ repo_name = repo_path.name
239
+ files = find_python_files(repo_path)
240
+ if not files:
241
+ print(f"No .py files were found in '{repo_path}'.")
242
+ return
243
+ else:
244
+ print("Looking for a random repo on GitHub...")
245
+ result = fetch_and_clone_repo()
246
+ if result is None:
247
+ print("Could not obtain a valid GitHub repo after several attempts.")
248
+ return
249
+ tmp_dir, repo_path, repo_git, repo_name, files = result
250
+
251
+ print_summary(repo_path, files)
252
+
253
+ results, functions_list, index_list = feature_extraction(
254
+ repo_path, files, repo_git, repo_name
255
+ )
256
+ print(f"\nFeatures extracted from {len(results)} files.")
257
+
258
+ if not results:
259
+ print("Could not extract features from any file.")
260
+ return
261
+
262
+ embeddings_por_archivo = generar_embeddings_reducidos(functions_list, index_list)
263
+ report = full_report(results, embeddings_por_archivo)
264
+
265
+ print("\n" + "=" * 60)
266
+ print(" RISK + RAW VALUES (highest to lowest risk)")
267
+ print("=" * 60)
268
+ print(report.to_string(index=False))
269
+
270
+ csv_path = save_csv_report(report, repo_name)
271
+ print(f"\nReport saved to: {csv_path}")
272
+ finally:
273
+ if repo_git is not None:
274
+ repo_git.close()
275
+ if tmp_dir is not None:
276
+ _delete_tree(tmp_dir)
277
+ print("\nTemporary repo deleted.")
278
+
279
+
280
+ if __name__ == "__main__":
281
+ main()