gitmole 0.6.8__tar.gz → 0.6.9__tar.gz

{gitmole-0.6.8 → gitmole-0.6.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gitmole
-Version: 0.6.8
+Version: 0.6.9
 Summary: Offline git repository analysis with a terminal report: hotspots, coupling, ownership, code age, secrets, repo health.
 License: MIT
 Project-URL: Homepage, https://github.com/antvinni/gitmole

{gitmole-0.6.8 → gitmole-0.6.9}/gitmole/__init__.py

@@ -1,3 +1,3 @@
 """gitmole: offline git repository analysis with a terminal report."""
 
-__version__ = "0.6.8"
+__version__ = "0.6.9"

{gitmole-0.6.8 → gitmole-0.6.9}/gitmole/functions.py

@@ -30,10 +30,20 @@ def select_files(repo: str, ignore=(), types_spec: str = None) -> list:
     return [f for f in files if lizard.get_reader_for(f) is not None]
 
 
+NAME_CAP = 200        # a deeply nested fixture gives lizard a dotted name of megabytes; nobody reads past this
+LONG_NAME_CAP = 500
+
+
+def _cut(text: str, cap: int) -> str:
+    return text if len(text) <= cap else text[:cap - 1] + "…"
+
+
 def csv_row(info, fn) -> list:
-    """The columns `lizard --csv` prints, so the loader does not care which produced the file."""
+    """The columns `lizard --csv` prints, so the loader does not care which produced the file. Names
+    are cut to what a table can show, so one pathological fixture cannot make the file unreadable."""
+    name = _cut(fn.name, NAME_CAP)
     return [fn.nloc, fn.cyclomatic_complexity, fn.token_count, fn.parameter_count, fn.length,
-            f"{fn.name}@{fn.start_line}-{fn.end_line}@{info.filename}", info.filename, fn.name, fn.long_name, fn.start_line, fn.end_line]
+            f"{name}@{fn.start_line}-{fn.end_line}@{info.filename}", info.filename, name, _cut(fn.long_name, LONG_NAME_CAP), fn.start_line, fn.end_line]
 
 
 def write_duplicates(dup: Duplicates, fh) -> None:

{gitmole-0.6.8 → gitmole-0.6.9}/gitmole/load.py

@@ -3,6 +3,7 @@ from __future__ import annotations
 
 import csv
 import io
+import sys
 import json
 import os
 import re
@@ -145,13 +146,21 @@ def parse_authors_log(text: str) -> list:
     ]
 
 
+NAME_CAP = 200   # a function name a table can show; deeply nested fixtures give lizard dotted names of megabytes
+csv.field_size_limit(min(sys.maxsize, 2**31 - 1))   # an older functions.csv may still carry such a name
+
+
+def _cut(name: str, cap: int = NAME_CAP) -> str:
+    return name if len(name) <= cap else name[:cap - 1] + "…"
+
+
 def parse_functions(text: str) -> list:
     """lizard --csv rows: nloc, ccn, tokens, params, length, location, file, function, long name, start, end."""
     rows = []
     for r in csv.reader(io.StringIO(text)):
         if len(r) < 11:
             continue
-        rows.append({"file": _rel(r[6]), "function": r[7] or "(anonymous)", "ccn": _num(r[1]), "nloc": _num(r[0]), "params": _num(r[3]),
+        rows.append({"file": _rel(r[6]), "function": _cut(r[7]) or "(anonymous)", "ccn": _num(r[1]), "nloc": _num(r[0]), "params": _num(r[3]),
                      "start": _num(r[9]), "end": _num(r[10])})
     return rows
 

{gitmole-0.6.8 → gitmole-0.6.9}/gitmole.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gitmole
-Version: 0.6.8
+Version: 0.6.9
 Summary: Offline git repository analysis with a terminal report: hotspots, coupling, ownership, code age, secrets, repo health.
 License: MIT
 Project-URL: Homepage, https://github.com/antvinni/gitmole

{gitmole-0.6.8 → gitmole-0.6.9}/tests/test_functions.py

@@ -67,6 +67,17 @@ class FunctionsScript(unittest.TestCase):
         self.assertIn('"tracked"', csv, "functions are still measured")
         self.assertIsNone(dup, "no duplicates.txt: the finder did not run, so nothing pretends it did")
 
+    def test_a_huge_nested_name_is_cut_before_it_is_written(self):
+        from types import SimpleNamespace
+        name = ".".join("a" for _ in range(100_000))
+        fn = SimpleNamespace(nloc=5, cyclomatic_complexity=3, token_count=40, parameter_count=1, length=5, name=name,
+                             long_name=name + "( )", start_line=1, end_line=5)
+        row = functions.csv_row(SimpleNamespace(filename="a.py"), fn)
+        self.assertEqual(len(row[7]), functions.NAME_CAP)
+        self.assertTrue(row[7].endswith("…"))
+        self.assertLessEqual(len(row[8]), functions.LONG_NAME_CAP)
+        self.assertLessEqual(len(row[5]), functions.NAME_CAP + len("@1-5@a.py"))
+
     def test_csv_matches_lizards_own_layout(self):
         with tempfile.TemporaryDirectory() as d:
             make_repo(d)

{gitmole-0.6.8 → gitmole-0.6.9}/tests/test_load.py

@@ -138,6 +138,19 @@ class ParseFunctions(unittest.TestCase):
     def test_empty(self):
         self.assertEqual(load.parse_functions(""), [])
 
+    def test_a_row_with_a_huge_long_name_does_not_abort_the_report(self):
+        # ruff: lizard wrote a long name of several hundred kilobytes for one function, over csv's default field limit
+        huge = "f( " + "a, " * 100_000 + ")"
+        rows = load.parse_functions(f'5,3,40,1,5,"f@1-5@a.rs","a.rs","f","{huge}",1,5\n')
+        self.assertEqual((rows[0]["function"], rows[0]["file"], rows[0]["ccn"]), ("f", "a.rs", 3))
+
+    def test_a_huge_function_name_is_cut_to_something_a_table_can_show(self):
+        # a test fixture of deeply nested functions gives lizard a dotted name of megabytes
+        name = ".".join("a" for _ in range(100_000))
+        rows = load.parse_functions(f'5,3,40,1,5,"{name}@1-5@a.py","a.py","{name}","{name}( )",1,5\n')
+        self.assertEqual(len(rows[0]["function"]), load.NAME_CAP)
+        self.assertTrue(rows[0]["function"].endswith("…"))
+
     def test_a_nameless_function_is_called_anonymous(self):
         # lizard names Go function literals with an empty string where it names JavaScript's "(anonymous)"
         rows = load.parse_functions('136,47,926,1,270,"@316-585@completions.go","completions.go",""," c * Command",316,585\n')