TLDR-844 add auth key to grobid (#503)

dedoc/readers/article_reader/article_reader.py

@@ -28,6 +28,9 @@ def __init__(self, config: Optional[dict] = None) -> None:
         else:
             self.grobid_url = f"http://{os.environ.get('GROBID_HOST', 'localhost')}:{os.environ.get('GROBID_PORT', '8070')}"
         self.url = f"{self.grobid_url}/api/processFulltextDocument"
+
+        auth_key = os.environ.get("GROBID_AUTH_KEY", "")
+        self.request_headers = {"Authorization": auth_key} if auth_key else {}
         self.grobid_is_alive = False
 
     def read(self, file_path: str, parameters: Optional[dict] = None) -> UnstructuredDocument:
@@ -48,7 +51,7 @@ def read(self, file_path: str, parameters: Optional[dict] = None) -> Unstructure
         with open(file_path, "rb") as file:
             files = {"input": file}
             try:
-                response = requests.post(self.url, files=files, data={"teiCoordinates": "figure"})
+                response = requests.post(self.url, files=files, data={"teiCoordinates": "figure"}, headers=self.request_headers)
                 if response.status_code != 200:
                     warning = f"GROBID returns code {response.status_code}."
                     self.logger.warning(warning)
@@ -106,7 +109,7 @@ def __update_grobid_alive(self, grobid_url: str, max_attempts: int = 2) -> None:
         attempt = max_attempts
         while attempt > 0:
             try:
-                response = requests.get(f"{grobid_url}/api/isalive")
+                response = requests.get(f"{grobid_url}/api/isalive", headers=self.request_headers)
                 if response.status_code == 200:
                     self.logger.info(f"GROBID up on {grobid_url}.")
                     self.grobid_is_alive = True

dedoc/readers/pdf_reader/pdf_base_reader.py

@@ -2,7 +2,6 @@
 from collections import namedtuple
 from typing import Dict, Iterator, List, Optional, Set, Tuple
 
-import numpy as np
 from dedocutils.data_structures.bbox import BBox
 from numpy import ndarray
 
@@ -13,7 +12,6 @@
 from dedoc.readers.pdf_reader.data_classes.line_with_location import LineWithLocation
 from dedoc.readers.pdf_reader.data_classes.pdf_image_attachment import PdfImageAttachment
 from dedoc.readers.pdf_reader.data_classes.tables.scantable import ScanTable
-from dedoc.readers.pdf_reader.pdf_image_reader.table_recognizer.gost_frame_recognizer import GOSTFrameRecognizer
 
 ParametersForParseDoc = namedtuple("ParametersForParseDoc", [
     "orient_analysis_cells",
@@ -45,6 +43,7 @@ def __init__(self, *, config: Optional[dict] = None, recognized_extensions: Opti
 
         from dedoc.readers.pdf_reader.pdf_image_reader.line_metadata_extractor.metadata_extractor import LineMetadataExtractor
         from dedoc.readers.pdf_reader.pdf_image_reader.paragraph_extractor.scan_paragraph_classifier_extractor import ScanParagraphClassifierExtractor
+        from dedoc.readers.pdf_reader.pdf_image_reader.table_recognizer.gost_frame_recognizer import GOSTFrameRecognizer
         from dedoc.readers.pdf_reader.pdf_image_reader.table_recognizer.table_recognizer import TableRecognizer
         from dedoc.readers.pdf_reader.utils.line_object_linker import LineObjectLinker
         from dedoc.attachments_extractors.concrete_attachments_extractors.pdf_attachments_extractor import PDFAttachmentsExtractor
@@ -153,8 +152,8 @@ def _parse_document(self, path: str, parameters: ParametersForParseDoc) -> (
             metadata["rotated_page_angles"] = page_angles
         return all_lines_with_paragraphs, mp_tables, attachments, warnings, metadata
 
-    def _process_document_with_gost_frame(self, images: Iterator[np.ndarray], first_page: int, parameters: ParametersForParseDoc, path: str) -> \
-            Tuple[Tuple[List[LineWithLocation], List[ScanTable], List[PdfImageAttachment], List[float]], Dict[int, Tuple[np.ndarray, BBox, Tuple[int, ...]]]]:
+    def _process_document_with_gost_frame(self, images: Iterator[ndarray], first_page: int, parameters: ParametersForParseDoc, path: str) -> \
+            Tuple[Tuple[List[LineWithLocation], List[ScanTable], List[PdfImageAttachment], List[float]], Dict[int, Tuple[ndarray, BBox, Tuple[int, ...]]]]:
         from joblib import Parallel, delayed
         from dedoc.readers.pdf_reader.pdf_txtlayer_reader.pdf_txtlayer_reader import PdfTxtlayerReader
 
@@ -170,7 +169,7 @@ def _process_document_with_gost_frame(self, images: Iterator[np.ndarray], first_
         return result, gost_analyzed_images
 
     def _shift_all_contents(self, lines: List[LineWithMeta], unref_tables: List[ScanTable], attachments: List[PdfImageAttachment],
-                            gost_analyzed_images: Dict[int, Tuple[np.ndarray, BBox, Tuple[int, ...]]]) -> None:
+                            gost_analyzed_images: Dict[int, Tuple[ndarray, BBox, Tuple[int, ...]]]) -> None:
         # shift unref_tables
         for scan_table in unref_tables:
             for location in scan_table.locations: