llmstxt-standalone 0.1.0__tar.gz → 0.1.1__tar.gz

{llmstxt_standalone-0.1.0 → llmstxt_standalone-0.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: llmstxt-standalone
-Version: 0.1.0
+Version: 0.1.1
 Summary: Generate llms.txt from built HTML documentation
 Keywords: llms,documentation,markdown,mkdocs
 Author: Shaan Majid

{llmstxt_standalone-0.1.0 → llmstxt_standalone-0.1.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "llmstxt-standalone"
-version = "0.1.0"
+version = "0.1.1"
 description = "Generate llms.txt from built HTML documentation"
 readme = "README.md"
 requires-python = ">=3.10"

{llmstxt_standalone-0.1.0 → llmstxt_standalone-0.1.1}/src/llmstxt_standalone/config/derive.py

@@ -10,10 +10,13 @@ def nav_to_sections(nav: list[Any]) -> dict[str, list[str]]:
     sections: dict[str, list[str]] = {}
 
     for item in nav:
-        if isinstance(item, dict):
+        if isinstance(item, str):
+            # Bare top-level page: - index.md
+            sections.setdefault("Pages", []).append(item)
+        elif isinstance(item, dict):
             for key, value in item.items():
                 if isinstance(value, str):
-                    # Top-level page, add to "Pages" section
+                    # Top-level page with title: - Home: index.md
                     sections.setdefault("Pages", []).append(value)
                 elif isinstance(value, list):
                     # Section with children

{llmstxt_standalone-0.1.0 → llmstxt_standalone-0.1.1}/src/llmstxt_standalone/config/load.py

@@ -74,6 +74,27 @@ def _config_from_mkdocs(raw: dict[str, Any]) -> Config:
         full_output = llmstxt_config.get("full_output", DEFAULT_FULL_OUTPUT)
         content_selector = llmstxt_config.get("content_selector")
         sections = llmstxt_config.get("sections", {})
+        if not isinstance(sections, dict):
+            raise ValueError(
+                f"llmstxt 'sections' must be a mapping, got {type(sections).__name__}"
+            )
+        for section_name, pages in sections.items():
+            if not isinstance(section_name, str):
+                raise ValueError(
+                    "llmstxt 'sections' keys must be strings, "
+                    f"got {type(section_name).__name__}"
+                )
+            if not isinstance(pages, list):
+                raise ValueError(
+                    f"llmstxt 'sections.{section_name}' must be a list of strings, "
+                    f"got {type(pages).__name__}"
+                )
+            for page in pages:
+                if not isinstance(page, str):
+                    raise ValueError(
+                        f"llmstxt 'sections.{section_name}' entries must be strings, "
+                        f"got {type(page).__name__}"
+                    )
     else:
         markdown_description = ""
         full_output = DEFAULT_FULL_OUTPUT

{llmstxt_standalone-0.1.0 → llmstxt_standalone-0.1.1}/src/llmstxt_standalone/convert.py

@@ -34,9 +34,15 @@ def _autoclean(soup: BeautifulSoup | Tag) -> None:
     for element in soup.find_all("table", attrs={"class": "highlighttable"}):
         code = element.find("code")
         if code:
-            element.replace_with(
-                BeautifulSoup(f"<pre>{code.get_text()}</pre>", "html.parser")
+            # Find the root BeautifulSoup document to create new tags
+            # (soup parameter may be a Tag, which doesn't have new_tag)
+            doc = next(
+                (p for p in element.parents if isinstance(p, BeautifulSoup)), None
             )
+            if doc:
+                pre_tag = doc.new_tag("pre")
+                pre_tag.string = code.get_text()
+                element.replace_with(pre_tag)
 
 
 def _get_language(tag: Tag) -> str: