remove PyMuPDF dependency

deepset-ai · May 7, 2024 · 2a64135 · 2a64135
1 parent 3c2a4fe
commit 2a64135
Show file tree

Hide file tree

Showing 5 changed files with 5 additions and 344 deletions.
diff --git a/.github/workflows/license_compliance.yml b/.github/workflows/license_compliance.yml
@@ -42,10 +42,9 @@ jobs:
           # Exclusions in the vanilla distribution must be explicitly motivated
           #
           # - tqdm is MLP but there are no better alternatives
-          # - PyMuPDF is optional
           # - pinecone-client is optional
           # - psycopg2 is optional
-          exclude: "(?i)^(PyMuPDF|tqdm|pinecone-client|psycopg2).*"
+          exclude: "(?i)^(tqdm|pinecone-client|psycopg2).*"
 
       # We keep the license inventory on FOSSA
       - name: Send license report to Fossa
@@ -199,7 +198,7 @@ jobs:
           # Special cases:
           # - pyzmq is flagged because dual-licensed, but we assume using BSD
           # - tqdm is MLP but there are no better alternatives
-          exclude: "(?i)^(astroid|certifi|chardet|num2words|nvidia-|pathspec|pinecone-client|psycopg2|pylint|PyMuPDF|pyzmq|tqdm).*"
+          exclude: "(?i)^(astroid|certifi|chardet|num2words|nvidia-|pathspec|pinecone-client|psycopg2|pylint|pyzmq|tqdm).*"
 
       - name: Print report
         if: ${{ always() }}
@@ -272,7 +271,7 @@ jobs:
           # Special cases:
           # - pyzmq is flagged because dual-licensed, but we assume using BSD
           # - tqdm is MLP but there are no better alternatives
-          exclude: "(?i)^(astroid|certifi|chardet|num2words|nvidia-|pathspec|pinecone-client|psycopg2|pylint|PyMuPDF|pyzmq|tqdm).*"
+          exclude: "(?i)^(astroid|certifi|chardet|num2words|nvidia-|pathspec|pinecone-client|psycopg2|pylint|pyzmq|tqdm).*"
 
       - name: Print report
         if: ${{ always() }}

diff --git a/haystack/nodes/file_converter/__init__.py b/haystack/nodes/file_converter/__init__.py
@@ -9,16 +9,6 @@
 from haystack.nodes.file_converter.txt import TextConverter
 from haystack.nodes.file_converter.azure import AzureConverter
 from haystack.nodes.file_converter.parsr import ParsrConverter
-
-
-try:
-    with LazyImport() as fitz_import:
-        # Try to use PyMuPDF, if not available fall back to xpdf
-        from haystack.nodes.file_converter.pdf import PDFToTextConverter  # type: ignore
-
-    fitz_import.check()
-except (ModuleNotFoundError, ImportError):
-    from haystack.nodes.file_converter.pdf_xpdf import PDFToTextConverter  # type: ignore  # pylint: disable=reimported,ungrouped-imports
-
+from haystack.nodes.file_converter.pdf_xpdf import PDFToTextConverter
 from haystack.nodes.file_converter.markdown import MarkdownConverter
 from haystack.nodes.file_converter.image import ImageToTextConverter
diff --git a/haystack/nodes/file_converter/pdf.py b/haystack/nodes/file_converter/pdf.py