Add flag for force creating page

osolmaz · osolmaz · commit 03b60aebda5a · 2024-09-13T15:06:57.000+02:00
diff --git a/docs/conversion.md b/docs/conversion.md
@@ -125,9 +125,11 @@ HTML->JSON-DOC tasks
 - [x] Convert line breaks `<br>`
 - [x] Convert `<caption>` and `<figcaption>`
 - [x] Force_page=true
+- [x] Add an argument to CLI for force-creating  a page
+    - for some reason, pandoc doesn't create a top-level `<html>` and `<body>` when converting from docx to HTML
 - [ ] Residual strings, newlines or empty paragraphs in the final output list (in progress)
 - [ ] Make sure `<a>` conversion is consistent
 - [ ] Cleanup empty blocks at the end
 - [ ] Table cells with colspan/rowspan
 - [ ] Add test for `<code>` and `<pre>`
-- [ ] Table thead/tbody/tfoot ordering
+- [ ] Table thead/tbody/tfoot ordering
diff --git a/jsondoc/bin/convert_jsondoc.py b/jsondoc/bin/convert_jsondoc.py
@@ -65,6 +65,7 @@ def convert_to_jsondoc(
     indent: int | None = None,
     source_format: str | None = None,
     target_format: str | None = None,
+    force_page: bool = False,
 ):
     """
     Convert to and from JSON-DOC format.
@@ -127,6 +128,7 @@ def convert_to_jsondoc(
                     input_content if input_content is not None else input_file,
                     "html",
                     format=source_format,
+                    extra_args=["--wrap=none"],
                 )
             except RuntimeError as e:
                 # Handle different error message from Pandoc
@@ -143,7 +145,7 @@ def convert_to_jsondoc(
                 else:
                     raise e
 
-        jsondoc = html_to_jsondoc(html_content)
+        jsondoc = html_to_jsondoc(html_content, force_page=force_page)
 
         # Serialize the jsondoc
         serialized_jsondoc = jsondoc_dump_json(jsondoc, indent=indent)
@@ -189,6 +191,12 @@ def main():
         help="Number of spaces for indentation in the output JSON file",
         default=None,
     )
+    parser.add_argument(
+        "--force-page",
+        action="store_true",
+        help="Force the creation of a page even if the input doesn't "
+        "contain a top-level HTML structure",
+    )
     args = parser.parse_args()
 
     try:
@@ -198,6 +206,7 @@ def main():
             indent=args.indent,
             source_format=args.source_format,
             target_format=args.target_format,
+            force_page=args.force_page,
         )
     except (ValueError, RuntimeError) as e:
         print(e)
diff --git a/jsondoc/convert/html.py b/jsondoc/convert/html.py
@@ -64,6 +64,7 @@ class ConvertOutput(BaseModel):
     and then we concatenate the result with prev_objects and next_objects:
     >>> final_objects = prev_objects + reconciled_objects + next_objects
     """
+
     main_object: BlockBase | RichTextBase
     prev_objects: List[BlockBase | RichTextBase] = []
     next_objects: List[BlockBase | RichTextBase] = []
@@ -359,6 +360,7 @@ class DefaultOptions:
         default_title = False
         keep_inline_images_in = []
         strip = None
+        force_page = False
 
     class Options(DefaultOptions):
         pass
@@ -379,16 +381,14 @@ def convert(self, html: str | bytes) -> Page | BlockBase | List[BlockBase]:
         soup = BeautifulSoup(html, "html.parser")
         return self.convert_soup(soup)
 
-    def convert_soup(
-        self, soup: BeautifulSoup, force_page=False
-    ) -> Page | BlockBase | List[BlockBase]:
+    def convert_soup(self, soup: BeautifulSoup) -> Page | BlockBase | List[BlockBase]:
 
         children = self.process_tag(soup, convert_as_inline=False, children_only=True)
         children = run_final_block_transformations(children)
         is_page = self._is_soup_page(soup)
 
         ret = None
-        if is_page or force_page:
+        if is_page or self.options["force_page"]:
             title = self._get_html_title(soup)
             # Ensure that children is a list
             if not isinstance(children, list):
