fix test_pdf_miner script

dedoc/scripts/test_pdf_miner.py

@@ -1,14 +1,14 @@
+import json
 import os
-from pathlib import Path
 import re
+from pathlib import Path
 from tempfile import TemporaryDirectory
 
+import requests
+
 from dedoc.api.api_utils import json2txt
 from dedoc.dedoc_manager import DedocManager
-from dedoc.readers.pdf_reader.pdf_txtlayer_reader.pdfminer_reader.pdfminer_extractor import PdfminerExtractor
-
 
-import requests
 
 URL = "https://at.ispras.ru/owncloud/index.php/s/uImxYhliBHU8ei7/download"
 
@@ -20,22 +20,24 @@
             f.write(response.content)
         import zipfile
 
-        with zipfile.ZipFile((Path(tmpdir) / "pdfs.zip"), 'r') as zip_ref:
+        with zipfile.ZipFile((Path(tmpdir) / "pdfs.zip"), "r") as zip_ref:
             zip_ref.extractall((Path(tmpdir)))
         os.remove(Path(tmpdir) / "pdfs.zip")
-        pdfs_path = Path(tmpdir) / 'PdfMiner params '
+        pdfs_path = Path(tmpdir) / "PdfMiner params "
 
         manager = DedocManager()
         for file in os.listdir(pdfs_path):
             result = manager.parse(file_path=str(pdfs_path / file), parameters={"pdf_with_text_layer": "true"})
             txt_content = json2txt(paragraph=result.content.structure)
-            with (Path(tmpdir) / 'ocr.txt').open('w') as f:
+            with (Path(tmpdir) / "ocr.txt").open("w") as f:
                 f.write(txt_content)
 
             accuracy_script_path = os.path.abspath(os.path.join(os.path.dirname(__file__), "accuracy"))
-            gt_path = Path("pdf_ground_truths") / (file[:-3] + 'txt')
-            tmp_ocr_path = Path(tmpdir) / 'ocr.txt'
+            gt_path = Path("pdf_ground_truths") / (file[:-3] + "txt")
+            tmp_ocr_path = Path(tmpdir) / "ocr.txt"
             accuracy_path = Path(tmpdir) / "accuracy.txt"
+            if accuracy_path.exists():
+                accuracy_path.unlink()
             command = f"{accuracy_script_path} \"{gt_path}\" {tmp_ocr_path} >> {accuracy_path}"
             os.system(command)
 
@@ -47,5 +49,8 @@
                 acc_percent = re.findall(r"\d+\.\d+", matched[0])[0][:-1]
                 info[str(file)] = acc_percent
 
-    print(info)
+    output_dir = os.path.abspath(os.path.join(os.path.dirname(__file__), "..", "..", "resources", "benchmarks"))
+    with (Path(output_dir) / "test_pdf_miner.json").open("w") as f:
+        json.dump(info, f)
 
+    print("save result in" + output_dir)

resources/benchmarks/test_pdf_miner.json

@@ -0,0 +1,13 @@
+{
+  "\u041e\u0431\u0440\u0430\u0437\u0435\u0446 \u043f\u0440\u0438\u043c\u0435\u0440\u043d\u043e\u0433\u043e \u0437\u0430\u043f\u043e\u043b\u043d\u0435\u043d\u0438\u044f \u0443\u0432\u0435\u0434\u043e\u043c\u043b\u0435\u043d\u0438\u044f \u043e\u0431 \u043e\u0442\u0441\u0443\u0442\u0441\u0442\u0432\u0438\u0438 \u0446\u0438\u0444\u0440\u043e\u0432\u044b\u0445 \u0444\u0438\u043d\u0430\u043d\u0441\u043e\u0432\u044b\u0445 \u0430\u043a\u0442\u0438\u0432\u043e\u0432.pdf": "100.0",
+  "2023 \u0413\u043e\u043d\u043e\u0447\u043d\u0430\u044f \u0418\u043d\u0441\u0442\u0440\u0443\u043a\u0446\u0438\u044f  CR 2023.pdf": "100.0",
+  "support_182_poisk-dokumentov.pdf": "100.0",
+  "6.1 \u041e\u043f\u0438\u0441\u0430\u043d\u0438\u0435 \u043f\u0440\u043e\u0435\u043a\u0442\u0430 Thalamus.pdf": "100.0",
+  "ECPPM2020_Instructions.pdf": "100.0",
+  "NOR CHR 2023.pdf": "100.0",
+  "2-column-state.pdf": "100.0",
+  "ba-2017.pdf": "100.0",
+  "\u041c\u0435\u0436\u0434\u0443\u043d\u0430\u0440\u043e\u0434\u043d\u043e\u0435 \u0438 \u043d\u0430\u0446\u0438\u043e\u043d\u0430\u043b\u044c\u043d\u043e\u0435 \u0441\u043f\u043e\u0440\u0442\u0438\u0432\u043d\u043e\u0435 \u043f\u0440\u0430\u0432\u043e \u043f\u043e\u0440\u0442\u0444\u043e\u043b\u0438\u043e_\u0440\u0443\u0441.pdf": "100.0",
+  "Uvedoml_ESN.pdf": "100.0",
+  "instruction_gibdd.pdf": "100.0"
+}