Updates for text_extraction_benchmarks (#546)

Author: hellhoundisalab

Dockerfile

@@ -9,6 +9,8 @@ ENV RESOURCES_PATH "/dedoc_root/resources"
 COPY requirements.txt .
 RUN pip3 install --no-cache-dir -r requirements.txt
 RUN apt-get update && apt-get install -y --fix-missing --no-install-recommends fontforge
+RUN apt install -y libutf8proc-dev
+RUN ln -s /usr/lib/x86_64-linux-gnu/libutf8proc.so /usr/lib/libutf8proc.so.1
 
 RUN mkdir /dedoc_root
 RUN mkdir /dedoc_root/dedoc

dedoc/extensions.py

@@ -63,7 +63,7 @@
     eml_like_format={".eml"},
     mhtml_like_format={".mhtml", ".mht", ".mhtml.gz", ".mht.gz"},
     archive_like_format={".zip", ".tar", ".tar.gz", ".rar", ".7z"},
-    image_like_format={".png"},
+    image_like_format={".png", ".jpg", ".jpeg", ".tiff", ".tif"},
     pdf_like_format={".pdf"},
     csv_like_format={".csv", ".tsv"},
     txt_like_format={".txt", ".txt.gz"},