refactor: split compare result into full compare info and fast compare info. (#224)

Author: Artanias

Makefile

@@ -1,4 +1,4 @@
-UTIL_VERSION            := 0.5.18
+UTIL_VERSION            := 0.5.19
 UTIL_NAME               := codeplag
 PWD                     := $(shell pwd)
 

src/codeplag/algorithms/compare.py

@@ -7,11 +7,11 @@
 from codeplag.consts import DEFAULT_MAX_DEPTH, DEFAULT_NGRAMS_LENGTH, DEFAULT_WEIGHTS
 from codeplag.types import (
     ASTFeatures,
-    CompareInfo,
-    FastMetrics,
+    FastCompareInfo,
+    FullCompareInfo,
     MaxDepth,
     NgramsLength,
-    StructuresInfo,
+    StructureCompareInfo,
     Threshold,
 )
 
@@ -21,11 +21,11 @@ def fast_compare(
     features_s: ASTFeatures,
     ngrams_length: NgramsLength = DEFAULT_NGRAMS_LENGTH,
     weights: tuple[float, float, float, float] = DEFAULT_WEIGHTS,
-) -> FastMetrics:
+) -> FastCompareInfo:
     """Returns comparison result of two works compared by fast algorithms.
 
     Calculates the similarity of features of two programs using four algorithms, calculates their
-    weighted average, and returns all of this  in 'FastMetrics' structure.
+    weighted average, and returns all of this  in 'FastCompareInfo' structure.
 
     Args:
     ----
@@ -48,7 +48,7 @@ def fast_compare(
         np.array([jakkar_coef, ops_res, kw_res, lits_res]), weights=weights
     )
 
-    fast_metrics = FastMetrics(
+    fast_metrics = FastCompareInfo(
         jakkar=jakkar_coef,
         operators=ops_res,
         keywords=kw_res,
@@ -65,7 +65,7 @@ def compare_works(
     ngrams_length: NgramsLength = DEFAULT_NGRAMS_LENGTH,
     max_depth: MaxDepth = DEFAULT_MAX_DEPTH,
     threshold: Threshold | None = None,
-) -> CompareInfo:
+) -> FastCompareInfo | FullCompareInfo:
     """The function returns the complex result of comparing two works.
 
     Args:
@@ -79,18 +79,18 @@ def compare_works(
 
     Returns:
     -------
-        CompareInfo, which is the result of comparing works.
+        FastCompareInfo or FullCompareInfo, which is the result of comparing works.
         This can consist of fast metrics and, if the threshold
         value has been crossed, structure metric.
         If the threshold value is not set, it returns the structure
-        metric anywhere.
+        metric anywhere (FullCompareInfo).
 
     """
-    fast_metrics = fast_compare(
+    fast_compare_info = fast_compare(
         features_f=features1, features_s=features2, ngrams_length=ngrams_length
     )
-    if threshold and (fast_metrics.weighted_average * 100.0) < threshold:
-        return CompareInfo(fast=fast_metrics)
+    if threshold and (fast_compare_info.weighted_average * 100.0) < threshold:
+        return fast_compare_info
 
     compliance_matrix = np.empty(
         (len(features1.head_nodes), len(features2.head_nodes), 2), dtype=np.int64
@@ -102,6 +102,8 @@ def compare_works(
     )
     struct_res = struct_res[0] / struct_res[1]
 
-    structure_info = StructuresInfo(similarity=struct_res, compliance_matrix=compliance_matrix)
+    structure_info = StructureCompareInfo(
+        similarity=struct_res, compliance_matrix=compliance_matrix
+    )
 
-    return CompareInfo(fast=fast_metrics, structure=structure_info)
+    return FullCompareInfo(fast=fast_compare_info, structure=structure_info)

src/codeplag/display.py

@@ -8,7 +8,7 @@
 import pandas as pd
 from typing_extensions import Self
 
-from codeplag.types import ASTFeatures, CompareInfo, NodeCodePlace
+from codeplag.types import ASTFeatures, FullCompareInfo, NodeCodePlace
 
 CHARS_CNT: Final[int] = 40
 USEFUL_CHARS: Final[int] = 100
@@ -157,7 +157,7 @@ def clear_line() -> None:
 def print_compare_result(
     features1: ASTFeatures,
     features2: ASTFeatures,
-    compare_info: CompareInfo,
+    compare_info: FullCompareInfo,
     compliance_matrix_df: pd.DataFrame | None = None,
 ) -> None:
     """Prints the pretty result of comparing two files.
@@ -166,7 +166,7 @@ def print_compare_result(
     ----
         features1 (ASTFeatures): The features of the first  source file.
         features2 (ASTFeatures): The features of the second  source file.
-        compare_info (CompareInfo): The compare metrics of two works.
+        compare_info (FullCompareInfo): The compare metrics of two works.
         compliance_matrix_df (pd.DataFrame | None, optional): DataFrame consisting
           structures similarity information of two works.
 
@@ -186,15 +186,12 @@ def print_compare_result(
         [compare_info.fast],
         index=np.array(["Similarity"]),
         columns=pd.Index(
-            (field.upper() for field in compare_info.fast._fields), name="FastMetrics:"
+            (field.upper() for field in compare_info.fast._fields), name="FastCompareInfo:"
         ),
     )
     print(main_metrics_df)
     print()
 
-    if compare_info.structure is None:
-        return
-
     additional_metrics_df = pd.DataFrame(
         compare_info.structure.similarity,
         index=np.array(["Similarity"]),

src/codeplag/handlers/check.py

@@ -37,10 +37,11 @@
 from codeplag.reporters import CSVReporter
 from codeplag.types import (
     ASTFeatures,
-    CompareInfo,
     ExitCode,
     Extension,
+    FastCompareInfo,
     Flag,
+    FullCompareInfo,
     MaxDepth,
     Mode,
     NgramsLength,
@@ -315,10 +316,10 @@ def _handle_compare_result(
         self: Self,
         work1: ASTFeatures,
         work2: ASTFeatures,
-        metrics: CompareInfo,
+        metrics: FullCompareInfo | FastCompareInfo,
         save: bool = False,
     ) -> ExitCode:
-        if metrics.structure is None:
+        if isinstance(metrics, FastCompareInfo):
             return ExitCode.EXIT_SUCCESS
         if self.reporter and save:
             self.reporter.save_result(work1, work2, metrics)
@@ -347,7 +348,7 @@ def _handle_completed_futures(
     ) -> ExitCode:
         exit_code = ExitCode.EXIT_SUCCESS
         for future in as_completed(futures):
-            metrics: CompareInfo = future.result()
+            metrics: FullCompareInfo | FastCompareInfo = future.result()
             proc_works_info = processing[future.id]  # type: ignore
             exit_code = ExitCode(
                 exit_code

src/codeplag/handlers/report.py

@@ -24,8 +24,8 @@
 from codeplag.reporters import deserialize_compare_result, read_df
 from codeplag.translate import get_translations
 from codeplag.types import (
-    CompareInfo,
     ExitCode,
+    FullCompareInfo,
     Language,
     ReportType,
     SameFuncs,
@@ -222,7 +222,7 @@ def _get_parsed_line(
     df: pd.DataFrame,
     threshold: int = DEFAULT_THRESHOLD,
     include_funcs_less_threshold: bool = True,
-) -> Generator[tuple[pd.Series, CompareInfo, SameFuncs, SameFuncs], None, None]:
+) -> Generator[tuple[pd.Series, FullCompareInfo, SameFuncs, SameFuncs], None, None]:
     for _, line in df.iterrows():
         cmp_res = deserialize_compare_result(line)
         first_heads = _deserialize_head_nodes(line.first_heads)

src/codeplag/reporters.py

@@ -14,9 +14,9 @@
 from codeplag.logger import codeplag_logger as logger
 from codeplag.types import (
     ASTFeatures,
-    CompareInfo,
-    FastMetrics,
-    StructuresInfo,
+    FastCompareInfo,
+    FullCompareInfo,
+    StructureCompareInfo,
 )
 
 
@@ -29,7 +29,7 @@ def save_result(
         self: Self,
         first_work: ASTFeatures,
         second_work: ASTFeatures,
-        compare_info: CompareInfo,
+        compare_info: FullCompareInfo,
     ) -> None: ...
 
 
@@ -48,14 +48,14 @@ def save_result(
         self: Self,
         first_work: ASTFeatures,
         second_work: ASTFeatures,
-        compare_info: CompareInfo,
+        compare_info: FullCompareInfo,
     ) -> None:
         """Updates the cache with new comparisons and writes it to the filesystem periodically.
 
         Args:
             first_work (ASTFeatures): Contains the first work metadata.
             second_work (ASTFeatures): Contains the second work metadata.
-            compare_info (CompareInfo): Contains information about comparisons
+            compare_info (FullCompareInfo): Contains information about comparisons
               between the first and second works.
         """
         if not self.reports.is_dir():
@@ -91,7 +91,7 @@ def _write_df_to_fs(self: Self) -> None:
 
     def get_compare_result_from_cache(
         self: Self, work1: ASTFeatures, work2: ASTFeatures
-    ) -> CompareInfo | None:
+    ) -> FullCompareInfo | None:
         cache_val = self.__df_report[
             (self.__df_report.first_path == str(work1.filepath))
             & (self.__df_report.second_path == str(work2.filepath))
@@ -112,10 +112,8 @@ def read_df(path: Path) -> pd.DataFrame:
 def serialize_compare_result(
     first_work: ASTFeatures,
     second_work: ASTFeatures,
-    compare_info: CompareInfo,
+    compare_info: FullCompareInfo,
 ) -> pd.DataFrame:
-    assert compare_info.structure is not None
-
     return pd.DataFrame(
         {
             "date": _get_current_date(),
@@ -139,21 +137,21 @@ def serialize_compare_result(
     )
 
 
-def deserialize_compare_result(compare_result: pd.Series) -> CompareInfo:
+def deserialize_compare_result(compare_result: pd.Series) -> FullCompareInfo:
     if isinstance(compare_result.compliance_matrix, str):
         similarity_matrix = np.array(json.loads(compare_result.compliance_matrix))
     else:
         similarity_matrix = np.array(compare_result.compliance_matrix)
 
-    compare_info = CompareInfo(
-        fast=FastMetrics(
+    compare_info = FullCompareInfo(
+        fast=FastCompareInfo(
             jakkar=float(compare_result.jakkar),
             operators=float(compare_result.operators),
             keywords=float(compare_result.keywords),
             literals=float(compare_result.literals),
             weighted_average=float(compare_result.weighted_average),
         ),
-        structure=StructuresInfo(
+        structure=StructureCompareInfo(
             compliance_matrix=similarity_matrix,
             similarity=float(compare_result.struct_similarity),
         ),

src/codeplag/types.py

@@ -113,22 +113,22 @@ def get_sha256(self: Self) -> str:
 # Compare information
 
 
-class FastMetrics(NamedTuple):
+class FastCompareInfo(NamedTuple):
     jakkar: float
     operators: float
     keywords: float
     literals: float
     weighted_average: float
 
 
-class StructuresInfo(NamedTuple):
+class StructureCompareInfo(NamedTuple):
     similarity: float
     compliance_matrix: npt.NDArray
 
 
-class CompareInfo(NamedTuple):
-    fast: FastMetrics
-    structure: StructuresInfo | None = None
+class FullCompareInfo(NamedTuple):
+    fast: FastCompareInfo
+    structure: StructureCompareInfo
 
 
 # Exceptions and errors
@@ -198,6 +198,6 @@ class ProcessingWorks(NamedTuple):
 # To fix that you need to pass right module-name to namedtuple-factory
 NodeCodePlace.__module__ = __name__
 NodeStructurePlace.__module__ = __name__
-FastMetrics.__module__ = __name__
-StructuresInfo.__module__ = __name__
-CompareInfo.__module__ = __name__
+FastCompareInfo.__module__ = __name__
+StructureCompareInfo.__module__ = __name__
+FullCompareInfo.__module__ = __name__

test/unit/codeplag/algorithms/test_compare.py

@@ -2,11 +2,11 @@
 from typing_extensions import Self
 
 from codeplag.algorithms.compare import compare_works, fast_compare
-from codeplag.types import ASTFeatures, CompareInfo
+from codeplag.types import ASTFeatures, FastCompareInfo, FullCompareInfo
 
 
 class TestCompareWorks:
-    def test_compare_works_with_structure(self: Self, first_compare_result: CompareInfo):
+    def test_compare_works_with_structure(self: Self, first_compare_result: FullCompareInfo):
         assert first_compare_result.fast.jakkar == pytest.approx(0.737, 0.001)
         assert first_compare_result.fast.operators == pytest.approx(0.667, 0.001)
         assert first_compare_result.fast.keywords == 1.0
@@ -28,12 +28,12 @@ def test_compare_works_without_structure(
             features1=first_features, features2=third_features, threshold=60
         )
 
-        assert compare_info.fast.jakkar == 0.24
-        assert compare_info.fast.operators == 0.0
-        assert compare_info.fast.keywords == 0.6
-        assert compare_info.fast.literals == 0.0
-        assert compare_info.fast.weighted_average == pytest.approx(0.218, 0.001)
-        assert compare_info.structure is None
+        assert isinstance(compare_info, FastCompareInfo)
+        assert compare_info.jakkar == 0.24
+        assert compare_info.operators == 0.0
+        assert compare_info.keywords == 0.6
+        assert compare_info.literals == 0.0
+        assert compare_info.weighted_average == pytest.approx(0.218, 0.001)
 
 
 class TestFastCompare:

test/unit/codeplag/conftest.py

@@ -6,7 +6,7 @@
 
 from codeplag.algorithms.compare import compare_works
 from codeplag.pyplag.utils import get_ast_from_filename, get_features_from_ast
-from codeplag.types import ASTFeatures, CompareInfo
+from codeplag.types import ASTFeatures, FullCompareInfo
 
 CWD = Path(os.path.dirname(os.path.abspath(__file__)))
 FILEPATH1 = CWD / "./data/test1.py"
@@ -51,7 +51,9 @@ def third_features(third_tree: ast.Module) -> ASTFeatures:
 
 
 @pytest.fixture
-def first_compare_result(first_features: ASTFeatures, second_features: ASTFeatures) -> CompareInfo:
+def first_compare_result(
+    first_features: ASTFeatures, second_features: ASTFeatures
+) -> FullCompareInfo:
     compare_info = compare_works(features1=first_features, features2=second_features)
-    assert compare_info.structure is not None
+    assert isinstance(compare_info, FullCompareInfo)
     return compare_info

test/unit/codeplag/handlers/test_report.py

@@ -18,7 +18,7 @@
     calculate_sources_total_similarity,
 )
 from codeplag.reporters import serialize_compare_result
-from codeplag.types import ASTFeatures, CompareInfo, SameHead
+from codeplag.types import ASTFeatures, FullCompareInfo, SameHead
 
 
 @pytest.mark.parametrize(
@@ -171,7 +171,7 @@ def test__get_same_funcs(
 def test__get_parsed_line(
     first_features: ASTFeatures,
     second_features: ASTFeatures,
-    first_compare_result: CompareInfo,
+    first_compare_result: FullCompareInfo,
 ):
     compare_df = serialize_compare_result(first_features, second_features, first_compare_result)
     compare_df.iloc[0].first_heads = str(compare_df.iloc[0].first_heads)