black errors

zeptofine · Nov 3, 2023 · 8421b8a · 8421b8a
1 parent f6f8b50
commit 8421b8a
Show file tree

Hide file tree

Showing 7 changed files with 39 additions and 80 deletions.
diff --git a/imdataset_creator/__main__.py b/imdataset_creator/__main__.py
@@ -25,31 +25,21 @@
 )
 
 CPU_COUNT = int(cpu_count())
-logging.basicConfig(
-    level=logging.INFO, format="%(message)s", datefmt="[%X]", handlers=[RichHandler()]
-)
+logging.basicConfig(level=logging.INFO, format="%(message)s", datefmt="[%X]", handlers=[RichHandler()])
 app = typer.Typer(pretty_exceptions_show_locals=True, pretty_exceptions_short=True)
 log = logging.getLogger()
 
 
 @app.command()
 def main(
-    config_path: Annotated[
-        Path, Option(help="Where the dataset config is placed")
-    ] = Path("config.json"),
-    database_path: Annotated[Path, Option(help="Where the database is placed")] = Path(
-        "filedb.arrow"
+    config_path: Annotated[Path, Option(help="Where the dataset config is placed")] = Path(
+        "config.json"
     ),
-    threads: Annotated[int, Option(help="multiprocessing threads")] = CPU_COUNT
-    * 3
-    // 4,
+    database_path: Annotated[Path, Option(help="Where the database is placed")] = Path("filedb.arrow"),
+    threads: Annotated[int, Option(help="multiprocessing threads")] = CPU_COUNT * 3 // 4,
     chunksize: Annotated[int, Option(help="imap chunksize")] = 5,
-    population_chunksize: Annotated[
-        int, Option(help="chunksize when populating the df")
-    ] = 100,
-    population_interval: Annotated[
-        int, Option(help="save interval in secs when populating the df")
-    ] = 60,
+    population_chunksize: Annotated[int, Option(help="chunksize when populating the df")] = 100,
+    population_interval: Annotated[int, Option(help="save interval in secs when populating the df")] = 60,
     simulate: Annotated[bool, Option(help="stops before conversion")] = False,
     verbose: Annotated[bool, Option(help="prints converted files")] = False,
     sort_by: Annotated[str, Option(help="Which database column to sort by")] = "path",
@@ -159,16 +149,12 @@ def main(
         files: list[File]
         if db_cfg.rules:
             filter_t = p.add_task("filtering", total=0)
-            files = [
-                resolved[file] for file in db.filter(set(resolved)).get_column("path")
-            ]
+            files = [resolved[file] for file in db.filter(set(resolved)).get_column("path")]
             p.update(filter_t, total=len(files), completed=len(files))
         else:
             files = list(resolved.values())
 
-        scenarios = list(
-            db_cfg.parse_files(p.track(files, description="parsing files"))
-        )
+        scenarios = list(db_cfg.parse_files(p.track(files, description="parsing files")))
         if len(scenarios) != len(files):
             p.log(f"{len(files) - len(scenarios)} files are completed")
 

diff --git a/imdataset_creator/config_handler.py b/imdataset_creator/config_handler.py
@@ -13,34 +13,23 @@
 class ConfigHandler:
     def __init__(self, cfg: MainConfig):
         # generate `Input`s
-        self.inputs: list[Input] = [
-            Input.from_cfg(folder["data"]) for folder in cfg["inputs"]
-        ]
+        self.inputs: list[Input] = [Input.from_cfg(folder["data"]) for folder in cfg["inputs"]]
         # generate `Output`s
-        self.outputs: list[Output] = [
-            Output.from_cfg(folder["data"]) for folder in cfg["output"]
-        ]
+        self.outputs: list[Output] = [Output.from_cfg(folder["data"]) for folder in cfg["output"]]
         # generate `Producer`s
         self.producers: list[Producer] = [
-            Producer.all_producers[p["name"]].from_cfg(p["data"])
-            for p in cfg["producers"]
+            Producer.all_producers[p["name"]].from_cfg(p["data"]) for p in cfg["producers"]
         ]
 
         # generate `Rule`s
-        self.rules: list[Rule] = [
-            Rule.all_rules[r["name"]].from_cfg(r["data"]) for r in cfg["rules"]
-        ]
+        self.rules: list[Rule] = [Rule.all_rules[r["name"]].from_cfg(r["data"]) for r in cfg["rules"]]
 
     @overload
-    def gather_images(
-        self, sort=True, reverse=False
-    ) -> Generator[tuple[Path, list[Path]], None, None]:
+    def gather_images(self, sort=True, reverse=False) -> Generator[tuple[Path, list[Path]], None, None]:
         ...
 
     @overload
-    def gather_images(
-        self, sort=False, reverse=False
-    ) -> Generator[tuple[Path, PathGenerator], None, None]:
+    def gather_images(self, sort=False, reverse=False) -> Generator[tuple[Path, PathGenerator], None, None]:
         ...
 
     def gather_images(
@@ -62,8 +51,7 @@ def get_outputs(self, file: File) -> list[OutputScenario]:
         return [
             OutputScenario(str(pth), output.filters)
             for output in self.outputs
-            if not (pth := output.folder / Path(output.format_file(file))).exists()
-            or output.overwrite
+            if not (pth := output.folder / Path(output.format_file(file))).exists() or output.overwrite
         ]
 
     def parse_files(self, files: Iterable[File]) -> Generator[FileScenario, None, None]:
@@ -72,16 +60,5 @@ def parse_files(self, files: Iterable[File]) -> Generator[FileScenario, None, No
                 yield FileScenario(file, out_s)
 
     def __repr__(self):
-        i = ",\n".join(map(_repr_indent, map(repr, self.inputs)))
-        o = ",\n".join(map(_repr_indent, map(repr, self.outputs)))
-        p = ",\n".join(map(_repr_indent, map(repr, self.producers)))
-        r = ",\n".join(map(_repr_indent, map(repr, self.rules)))
-        attrs = ",\n".join(
-            [
-                _repr_indent(f"inputs=[\n{i}\n]"),
-                _repr_indent(f"outputs=[\n{o}\n]"),
-                _repr_indent(f"producers=[\n{p}\n]"),
-                _repr_indent(f"rules=[\n{r}\n]"),
-            ]
-        )
+        attrs = ",".join(str(x) for x in (self.inputs, self.outputs, self.producers, self.rules))
         return "\n".join([f"{self.__class__.__name__}(", attrs, ")"])
diff --git a/imdataset_creator/datarules/dataset_builder.py b/imdataset_creator/datarules/dataset_builder.py
@@ -282,7 +282,9 @@ def filter(self, lst) -> DataFrame:  # noqa: A003
             )
 
         vdf: DataFrame = self.__df.filter(pl.col("path").is_in(lst))
-        for matcher in combine_matchers([rule.matcher for rule in self.rules]):
+        for matcher in combine_matchers(rule.matcher for rule in self.rules):
+            if not len(vdf):
+                break
             vdf = matcher(vdf, self.__df) if isinstance(matcher, DataFrameMatcher) else vdf.filter(matcher)
         return vdf
 

diff --git a/imdataset_creator/gui/config_inputs.py b/imdataset_creator/gui/config_inputs.py
@@ -4,7 +4,7 @@
 import functools
 from copy import deepcopy as objcopy
 
-from PySide6.QtCore import QRect, QSize, Qt, Signal, Slot
+from PySide6.QtCore import QMimeData, QRect, QSize, Qt, Signal, Slot
 from PySide6.QtGui import QAction, QDrag, QDragEnterEvent, QDragLeaveEvent, QDragMoveEvent, QMouseEvent
 from PySide6.QtWidgets import (
     QCheckBox,
@@ -141,6 +141,17 @@ def mouseDoubleClickEvent(self, event: QMouseEvent) -> None:
         self.toggle_group()
         event.accept()
 
+    # def mousePressEvent(self, event: QMouseEvent):
+    #     print(event)
+    #     if event.button() == Qt.MouseButton.LeftButton:
+    #         drag = QDrag(self)
+    #         mimedata = QMimeData()
+    #         mimedata.setText(self.checkbox.text())
+    #         drag.setMimeData(mimedata)
+    #         drop_action = drag.exec()
+
+    #     event.accept()
+
     @property
     def n(self) -> int:
         return self._n

diff --git a/imdataset_creator/gui/main_window.py b/imdataset_creator/gui/main_window.py
@@ -232,15 +232,13 @@ def add_lock(self):
         self.locks += 1
         if self.locks == 1:
             log.info("locked")
-        ...
 
     @Slot()
     def remove_lock(self):
         """Unlocks the UI"""
         self.locks -= 1
         if self.locks == 0:
             log.info("unlocked")
-        ...
 
     @catch_building
     @Slot()

diff --git a/imdataset_creator/image_filters/destroyers.py b/imdataset_creator/image_filters/destroyers.py
@@ -58,16 +58,11 @@ def run(
             ksize = ksize + (ksize % 2 == 0)  # ensure ksize is odd
             return cv2.GaussianBlur(img, (ksize, ksize), 0)
 
-        if (
-            algorithm == BlurAlgorithm.ISOTROPIC
-            or algorithm == BlurAlgorithm.ANISOTROPIC
-        ):
+        if algorithm == BlurAlgorithm.ISOTROPIC or algorithm == BlurAlgorithm.ANISOTROPIC:
             sigma1: float = ri * self.scale
             ksize1: int = 2 * int(4 * sigma1 + 0.5) + 1
             if algorithm == BlurAlgorithm.ANISOTROPIC:
-                return cv2.GaussianBlur(
-                    img, (ksize1, ksize1), sigmaX=sigma1, sigmaY=sigma1
-                )
+                return cv2.GaussianBlur(img, (ksize1, ksize1), sigmaX=sigma1, sigmaY=sigma1)
 
             sigma2: float = ri * self.scale
             ksize2: int = 2 * int(4 * sigma2 + 0.5) + 1
@@ -179,16 +174,12 @@ def run(self, img: ndarray):
         enc_img: ndarray
         if algorithm == CompressionAlgorithms.JPEG:
             quality = randint(*self.jpeg_quality_range)
-            enc_img = cv2.imencode(
-                ".jpg", img, [int(cv2.IMWRITE_JPEG_QUALITY), quality]
-            )[1]
+            enc_img = cv2.imencode(".jpg", img, [int(cv2.IMWRITE_JPEG_QUALITY), quality])[1]
             return cv2.imdecode(enc_img, 1)
 
         if algorithm == CompressionAlgorithms.WEBP:
             quality = randint(*self.webp_quality_range)
-            enc_img = cv2.imencode(
-                ".webp", img, [int(cv2.IMWRITE_WEBP_QUALITY), quality]
-            )[1]
+            enc_img = cv2.imencode(".webp", img, [int(cv2.IMWRITE_WEBP_QUALITY), quality])[1]
             return cv2.imdecode(enc_img, 1)
 
         if algorithm in [
@@ -218,9 +209,7 @@ def run(self, img: ndarray):
                 codec = "mpeg2video"
 
             compressor = (
-                ffmpeg.input(
-                    "pipe:", format="rawvideo", pix_fmt="bgr24", s=f"{width}x{height}"
-                )
+                ffmpeg.input("pipe:", format="rawvideo", pix_fmt="bgr24", s=f"{width}x{height}")
                 .output("pipe:", format=container, vcodec=codec, **output_args)
                 .global_args("-loglevel", "error")
                 .global_args("-max_muxing_queue_size", "300000")
@@ -241,9 +230,7 @@ def run(self, img: ndarray):
                 newimg = np.frombuffer(out, np.uint8)
                 if len(newimg) != height * width * 3:
                     log.warning("New image size does not match")
-                    newimg = newimg[
-                        : height * width * 3
-                    ]  # idrk why i need this sometimes
+                    newimg = newimg[: height * width * 3]  # idrk why i need this sometimes
 
                 return newimg.reshape((height, width, 3))
             except subprocess.TimeoutExpired as e:

diff --git a/imdataset_creator/image_filters/resizer.py b/imdataset_creator/image_filters/resizer.py
@@ -156,7 +156,5 @@ def run(self, img: np.ndarray) -> np.ndarray:
     def from_cfg(cls, cfg: RandomRotateData):
         return cls(
             rotate_chance=cfg["rotate_chance"],
-            rotate_directions=[
-                RandomRotateDirections[d] for d in cfg["rotate_directions"]
-            ],
+            rotate_directions=[RandomRotateDirections[d] for d in cfg["rotate_directions"]],
         )