ODT: add more test data and improve conversion

marph91 · Jan 23, 2025 · 3a075f6 · 3a075f6
1 parent b26cbb3
commit 3a075f6
Show file tree

Hide file tree

Showing 3 changed files with 9 additions and 6 deletions.
diff --git a/src/markdown_lib/common.py b/src/markdown_lib/common.py
@@ -240,6 +240,7 @@ def get_inline_tags(text: str, start_characters: list[str]) -> list[str]:
 }
 
 # fmt: off
+INTERMEDIATE_FORMAT = "html"
 PANDOC_OUTPUT_FORMAT = (
     # https://pandoc.org/chunkedhtml-demo/8.22-markdown-variants.html
     # Don't use "commonmark_x". There is too much noise.
@@ -267,21 +268,22 @@ def get_inline_tags(text: str, start_characters: list[str]) -> list[str]:
 def html_to_markdown(text_html: bytes | str, custom_filter: list | None = None):
     # some needed preprocessing
     soup = BeautifulSoup(text_html, "html.parser")
+    if custom_filter is not None:
+        for filter_ in custom_filter:
+            filter_(soup)
     markdown_lib.html_filter.div_checklists(soup)
     markdown_lib.html_filter.highlighting(soup)
     markdown_lib.html_filter.iframes_to_links(soup)
+    markdown_lib.html_filter.merge_single_element_lists(soup)
     markdown_lib.html_filter.streamline_tables(soup)
     markdown_lib.html_filter.whitespace_in_math(soup)
-    if custom_filter is not None:
-        for filter_ in custom_filter:
-            filter_(soup)
     text_html_filtered = str(soup)
 
     # writer: json ast -> markdown
     text_md = pypandoc.convert_text(
         text_html_filtered,
         PANDOC_OUTPUT_FORMAT,
-        format="html",
+        format=INTERMEDIATE_FORMAT,
         extra_args=[
             # don't create artificial line breaks
             "--wrap=none",
@@ -307,7 +309,7 @@ def markup_to_markdown(
         # reader: x -> HTML
         text_html = pypandoc.convert_text(
             text,
-            "html",
+            INTERMEDIATE_FORMAT,
             format=format_,
             sandbox=True,
             extra_args=[

diff --git a/test/data b/test/data
diff --git a/test/test_convert.py b/test/test_convert.py
@@ -211,6 +211,7 @@ def test_formats(self, test_input):
             ["html", ["default_format/html"]],
             ["latex", ["default_format/latex"]],
             ["mediawiki", ["default_format/mediawiki"]],
+            ["odt", ["default_format/odt"]],
             ["txt2tags", ["default_format/txt2tags-2"]],
             ["vimwiki", ["default_format/vimwiki"]],
         ]
+ −		reference_data/default_format/odt/odt/100000000000030C0000024955B04860EE4EBDAD.jpg
+ −		reference_data/default_format/odt/odt/1000000100000B5F000004495E83AB1D.png
+ −		reference_data/default_format/odt/odt/2000001100001EAF00001EB7A6252779.svm
+34 −0		reference_data/default_format/odt/odt/Handout-Muster-Uni-LibreOffice-neu.md
+29 −0		reference_data/default_format/odt/odt/LibreOffice Writer Extended.md
+178 −0		reference_data/default_format/odt/odt/Scribbr-Hausarbeit-Vorlage-Open-Office-1.md
+ −		test_data/default_format/odt/Handout-Muster-Uni-LibreOffice-neu.odt
+ −		test_data/default_format/odt/LibreOffice Writer Extended.odt
+ −		test_data/default_format/odt/Scribbr-Hausarbeit-Vorlage-Open-Office-1.odt
+3 −0		test_data/sources.md