Add breadcrumbs

src/catleg/catleg.py

@@ -18,12 +18,12 @@
 app.add_typer(lf, name="lf", help="Commands for querying the raw Legifrance API")
 
 
-def _article(aid_or_url: str):
+def _article(aid_or_url: str, breadcrumbs: bool):
     article_id = article_id_or_url(aid_or_url)
     if article_id is None:
         raise ValueError(f"Sorry, I do not know how to process {aid_or_url}")
 
-    skel = asyncio.run(askel(article_id))
+    skel = asyncio.run(askel(article_id, breadcrumbs=breadcrumbs))
     return skel
 
 
@@ -35,14 +35,20 @@ def article(
             help="An article ID or Legifrance URL, for instance 'LEGIARTI000033971416' "
             "or 'https://www.legifrance.gouv.fr/codes/article_lc/LEGIARTI000033971416'."
         ),
-    ]
+    ],
+    nb: Annotated[
+        bool,
+        typer.Option(
+            help="If specified, do not print breadcrumbs (table of contents headers)"
+        ),
+    ] = False,
 ):
     """
     Output an article.
     By default, outputs markdown-formatted text.
     """
 
-    print(_article(aid_or_url))
+    print(_article(aid_or_url, not nb))
 
 
 @app.command()

src/catleg/skeleton.py

@@ -35,20 +35,43 @@ async def markdown_skeleton(textid: str, sectionid: str) -> str:
     return "\n\n".join(parts)
 
 
-async def article_skeleton(articleid: str) -> str:
+async def article_skeleton(articleid: str, breadcrumbs: bool = True) -> str:
     """
     Return an article skeleton (markdown-formatted law article).
+
+    Parameters
+    ----------
+    articleid: str
+       Legifrance article identifier
+    breadcrumbs: bool
+       if True, emits breadcrumbs (table of contents headers) before
+       outputting the article itself
+
+    Returns
+    -------
+    str
+       Markdown-formatted article
     """
     back = get_backend("legifrance")
     # This uses the Legifrance API directly, not the backend abstraction
     raw_article_json = await back.query_article_legi(articleid)
+    return _article_skeleton(raw_article_json=raw_article_json, breadcrumbs=breadcrumbs)
+
+
+# separate network calls and processing to ease unit testing
+def _article_skeleton(raw_article_json, breadcrumbs: bool = True):
     article_json = raw_article_json["article"]
     article = _article_from_legifrance_reply(raw_article_json)
     if article is None:
         raise RuntimeError(
             "Could not extract article from json reply %s", raw_article_json
         )
+
     parts = []
+    if breadcrumbs:
+        for i, toc_entry in enumerate(article_json["context"]["titresTM"], start=1):
+            parts.append(f"{'#' * i} {toc_entry['titre']}")
+
     # level: code (1) + length of section hierarchy + article (1)
     level = 1 + len(article_json["context"]["titresTM"]) + 1
     parts.append(f"{'#' * level} Article {article_json['num']} | {article.id}")

tests/test_skeleton.py

@@ -1,8 +1,10 @@
 from io import StringIO
 
+import pytest
+
 from catleg.parse_catala_markdown import parse_catala_file
 from catleg.query import _article_from_legifrance_reply
-from catleg.skeleton import _formatted_atricle
+from catleg.skeleton import _article_skeleton, _formatted_atricle
 
 from .test_legifrance_queries import _json_from_test_file
 
@@ -85,3 +87,12 @@ def test_no_article_renumbering_in_catala_file_parsing():
     assert "1. Le bénéfice ou revenu imposable est constitué" in article.text
     assert "2. Le revenu global net annuel" in article.text
     assert "3. Le bénéfice ou revenu net de chacune des catégories" in article.text
+
+
+@pytest.mark.parametrize("breadcrumbs", [False, True])
+def test_article_skeleton(breadcrumbs: bool):
+    article_json = _json_from_test_file("LEGIARTI000044983201.json")
+    askel = _article_skeleton(article_json)
+    assert "excédent du produit brut" in askel
+    if breadcrumbs:
+        assert "## Première Partie : Impôts d'État" in askel