TLDR-428 add flake8 tests and fix styles (#3)

.flake8

@@ -0,0 +1,20 @@
+[flake8]
+
+max-line-length = 160
+max-complexity = 10
+inline-quotes = "
+
+application-import-names = dedocutils, tests
+import-order-style = pycharm
+
+exclude =
+    .git,
+    __pycache__,
+    .idea,
+    build,
+    dedocutils/text_detection/doctr_text_detector/doctr,
+    *__init__.py
+
+# ANN101 - type annotations for self
+ignore =
+    ANN101

.github/workflows/test_on_push.yaml

@@ -23,7 +23,10 @@ jobs:
       run: |
         python3 -m pip install --upgrade pip
         pip3 install .[dev,torch]
+    - name: Run lint
+      run: |
+        pip3 install .[lint]
+        flake8 .
     - name: Run tests
       run: |
-        python3 -m unittest -v -f tests/test_style.py
         python3 -m unittest -v -f tests/unit_tests/test*

dedocutils/data_structures/bbox.py

@@ -1,6 +1,6 @@
 from collections import OrderedDict
 from dataclasses import dataclass
-from typing import Tuple, Dict
+from typing import Dict, Tuple
 
 
 @dataclass
@@ -46,12 +46,12 @@ def from_two_points(top_left: Tuple[int, int], bottom_right: Tuple[int, int]) ->
 
     def have_intersection_with_box(self, box: "BBox", threshold: float = 0.3) -> bool:
         # determine the (x, y)-coordinates of the intersection rectangle
-        xA = max(self.x_top_left, box.x_top_left)
-        yA = max(self.y_top_left, box.y_top_left)
-        xB = min(self.x_top_left + self.width, box.x_top_left + box.width)
-        yB = min(self.y_top_left + self.height, box.y_top_left + box.height)
+        x_min = max(self.x_top_left, box.x_top_left)
+        y_min = max(self.y_top_left, box.y_top_left)
+        x_max = min(self.x_top_left + self.width, box.x_top_left + box.width)
+        y_max = min(self.y_top_left + self.height, box.y_top_left + box.height)
         # compute the area of intersection rectangle
-        inter_a_area = max(0, xB - xA) * max(0, yB - yA)
+        inter_a_area = max(0, x_max - x_min) * max(0, y_max - y_min)
         # compute the area of both the prediction and ground-truth
         # rectangles
         box_b_area = float(box.width * box.height)

dedocutils/line_segmentation/clustering_line_segmenter.py

@@ -23,12 +23,12 @@ def segment(self, bboxes: List[BBox], parameters: Optional[dict] = None) -> List
         for i, pred_item in enumerate(pred):
             bbox = bboxes[i]
             cluster = pred_item
-            sorted_bboxes[cluster]['bbox'].append(bbox)
-            sorted_bboxes[cluster]['avgY'] += bbox.y_top_left
+            sorted_bboxes[cluster]["bbox"].append(bbox)
+            sorted_bboxes[cluster]["avgY"] += bbox.y_top_left
 
         for el in sorted_bboxes:
-            el['avgY'] /= len(el['bbox'])
-        sorted_bboxes = sorted(sorted_bboxes, key=lambda x: x['avgY'])
-        sorted_bboxes = [el['bbox'] for el in sorted_bboxes]
+            el["avgY"] /= len(el["bbox"])
+        sorted_bboxes = sorted(sorted_bboxes, key=lambda x: x["avgY"])
+        sorted_bboxes = [el["bbox"] for el in sorted_bboxes]
         sorted_bboxes = [sorted(line, key=lambda x: x.x_top_left) for line in sorted_bboxes]
         return sorted_bboxes

dedocutils/line_segmentation/intersection_line_segmenter.py

@@ -1,4 +1,4 @@
-from typing import List, Tuple, Optional
+from typing import List, Optional, Tuple
 
 import numpy as np
 
@@ -21,7 +21,7 @@ def segment(self, bboxes: List[BBox], parameters: Optional[dict] = None) -> List
             box_group, lines = self.__segment_lines(np.array(box_group))
 
         res_lines = []
-        for line_id, line in enumerate(lines):
+        for line in lines:
             res_lines.append([BBox(box[0][0], box[0][1], box[1][0] - box[0][0], box[1][1] - box[0][1]) for box in line])
 
         return res_lines

dedocutils/preprocessing/adaptive_binarizer.py

@@ -1,4 +1,5 @@
-from typing import Tuple, Optional
+from typing import Optional, Tuple
+
 import cv2
 import numpy as np
 

dedocutils/preprocessing/skew_corrector.py

@@ -1,4 +1,4 @@
-from typing import Tuple, Optional
+from typing import Optional, Tuple
 
 import cv2
 import numpy as np

dedocutils/text_detection/doctr_text_detector/doctr_text_detector.py

@@ -68,4 +68,4 @@ def _set_device(self, on_gpu: bool) -> None:
             self.location = lambda storage, loc: storage.cuda()
         else:
             self.device = torch.device("cpu")
-            self.location = 'cpu'
+            self.location = "cpu"

dedocutils/text_detection_recognition/tesseract_detector_recognizer.py

@@ -1,9 +1,9 @@
-from typing import Optional, List
+from typing import List, Optional
 
 import numpy as np
 import pytesseract
-from dedocutils.data_structures import BBox
 
+from dedocutils.data_structures import BBox
 from dedocutils.data_structures.text_with_bbox import TextWithBBox
 from dedocutils.text_detection_recognition.abstract_detector_recognizer import AbstractDetectorRecognizer
 

pyproject.toml

@@ -43,11 +43,20 @@ torch = [
 ]
 dev = [
     "build==0.10.0",
-    "twine==4.0.2",
-    "pycodestyle==2.7.0",
-    "flake8==3.9.2",
-    "flake8-annotations==2.6.2",
-    "pyflakes==2.3.1"
+    "twine==4.0.2"
+]
+lint = [
+    "flake8==5.0.4",
+    "flake8-absolute-import==1.0.0.1",
+    "flake8-annotations==2.9.1",
+    "flake8-bugbear==23.3.12",
+    "flake8-builtins==2.1.0",
+    "flake8-import-order==0.18.2",
+    "flake8-print==5.0.0",
+    "flake8-quotes==3.3.2",
+    "flake8-use-fstring==1.4",
+    "pycodestyle==2.9.0",
+    "pep8-naming==0.13.3"
 ]
 
 [tool.setuptools.packages.find]

tests/unit_tests/test_classes.py

@@ -2,6 +2,7 @@
 import unittest
 
 import cv2
+
 from dedocutils.text_detection import DoctrTextDetector