TLDR-518: Fix tabby partially read (#372)

* Fix tabby partially read * Add more tests * Fix tabby page slice parameters * Fix extract table in tabby with page range parameter --------- Co-authored-by: Nasty <[email protected]>
ispras · Nov 21, 2023 · 9c9c50e · 9c9c50e
1 parent fa396ef
commit 9c9c50e
Show file tree

Hide file tree

Showing 3 changed files with 18 additions and 6 deletions.
diff --git a/dedoc/readers/pdf_reader/pdf_txtlayer_reader/pdf_tabby_reader.py b/dedoc/readers/pdf_reader/pdf_txtlayer_reader/pdf_tabby_reader.py
@@ -117,13 +117,13 @@ def __extract(self, path: str, parameters: dict, warnings: list)\
                 return all_lines, all_tables, all_tables_on_images, all_attached_images, document_metadata
 
         # in java tabby reader page numeration starts with 1, end_page is included
-        # first_tabby_page = first_page + 1 if first_page is not None else 1
-        # last_tabby_page = None if last_page is not None and last_page > page_count else last_page
-        # document = self.__process_pdf(path=path, start_page=first_tabby_page, end_page=last_tabby_page) TODO TLDR-518
+        first_tabby_page = first_page + 1 if first_page is not None else 1
+        last_tabby_page = page_count if (last_page is None) or (last_page is not None and last_page > page_count) else last_page
+        self.logger.info(f"Reading PDF pages from {first_tabby_page} to {last_tabby_page}")
+        document = self.__process_pdf(path=path, start_page=first_tabby_page, end_page=last_tabby_page)
 
-        document = self.__process_pdf(path=path)
         pages = document.get("pages", [])
-        for page in pages[first_page:last_page]:
+        for page in pages:
             page_lines = self.__get_lines_with_location(page, file_hash)
             if page_lines:
                 all_lines.extend(page_lines)

diff --git a/dedoc/readers/pdf_reader/pdf_txtlayer_reader/tabbypdf/jars/ispras_tbl_extr.jar b/dedoc/readers/pdf_reader/pdf_txtlayer_reader/tabbypdf/jars/ispras_tbl_extr.jar
diff --git a/tests/api_tests/test_api_format_pdf_page_limit.py b/tests/api_tests/test_api_format_pdf_page_limit.py
@@ -42,7 +42,7 @@ def test_auto_tabby(self) -> None:
 
     def __check_out_of_limit(self, reader: str) -> None:
         text_expected = ""
-        for pages in ("10:11", ):
+        for pages in (":-1", "-1:0", "0:0", "10:11", "11:"):
             self.__check(pages, text_expected, reader=reader)
 
     def __check_limit(self, reader: str, check_partially: bool = False) -> None:
@@ -56,6 +56,18 @@ def __check_limit(self, reader: str, check_partially: bool = False) -> None:
         text_expected = "\n".join(self.lines[0:2])
         self.__check("1:2", text_expected, reader=reader, check_partially=check_partially)
 
+        text_expected = self.lines[0]
+        self.__check("1:1", text_expected, reader=reader, check_partially=check_partially)
+
+        text_expected = self.lines[1]
+        self.__check("2:2", text_expected, reader=reader, check_partially=check_partially)
+
+        text_expected = "\n".join(self.lines[1:3])
+        self.__check("2:3", text_expected, reader=reader, check_partially=check_partially)
+
+        text_expected = "\n".join(self.lines[4:8])
+        self.__check("5:8", text_expected, reader=reader, check_partially=check_partially)
+
         text_expected = self.lines[8]
         self.__check("9:", text_expected, reader=reader, check_partially=False)