Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 7 additions & 2 deletions src/tracealign/io/sefaria.py
Original file line number Diff line number Diff line change
Expand Up @@ -52,11 +52,16 @@ def _fetch_json(url: str) -> dict:


def _build_url(ref: str, version: str | None) -> str:
"""Build the v3 API URL for a reference, optionally pinning a version."""
"""Build the v3 API URL for a reference, optionally pinning a version.

The Sefaria v3 API accepts ``version=<language>|<versionTitle>``. The
language token must be the full name (``hebrew``), not the ISO code
(``he``); the API silently returns an empty versions array otherwise.
"""
encoded_ref = quote(ref.replace(" ", "_"), safe="._")
url = f"{SEFARIA_API_BASE}/{encoded_ref}"
if version is not None:
url += f"?version=he|{quote(version, safe='')}"
url += f"?version=hebrew|{quote(version, safe='')}"
return url


Expand Down
17 changes: 17 additions & 0 deletions tests/io/test_sefaria_fetch.py
Original file line number Diff line number Diff line change
Expand Up @@ -30,3 +30,20 @@ def test_fetch_json_uses_fixture():
result = sefaria._fetch_json("https://example.invalid/")
assert result["ref"] == "Pirkei Avot 1"
assert result["versions"][0]["versionTitle"].startswith("William Davidson")


def test_build_url_uses_hebrew_language_token():
"""Pin: Sefaria API requires 'hebrew|' not 'he|' for the version language."""
url = sefaria._build_url("Pirkei Avot 1", version="Vilna Edition")
assert "version=hebrew%7CVilna" in url or "version=hebrew|Vilna" in url
# Negative: the ISO-code form must not slip back in
assert "version=he|" not in url
assert "version=he%7C" not in url


def test_build_url_without_version_has_no_query():
url = sefaria._build_url("Pirkei Avot 1", version=None)
assert "?" not in url
assert url.startswith("https://www.sefaria.org/api/v3/texts/")
# The reference must be URL-safe: no raw whitespace ends up in the URL.
assert " " not in url
Loading