refactor content negotiation

rogue_scholar_api/__init__.py

@@ -158,30 +158,11 @@ async def post(slug, suffix=None):
     elif slug in prefixes and suffix:
         doi = f"https://doi.org/{slug}/{suffix}"
         if format_ in formats:
-            content_types = {
-                "bibtex": "application/x-bibtex",
-                "ris": "application/x-research-info-systems",
-                "csl": "application/vnd.citationstyles.csl+json",
-                "citation": f"text/x-bibliography; style={style}; locale={locale}",
-            }
-            content_type = content_types.get(format_)
-            metadata = get_doi_metadata_from_ra(doi, headers={"Accept": content_type})
-            if not metadata:
-                logger.warning(f"Metadata not found for {doi}")
+            response = get_doi_metadata_from_ra(doi, format_, style, locale)
+            if not response:
+                logger.warning("Metadata not found")
                 return {"error": "Metadata not found."}, 404
-            if format_ == "csl":
-                filename = f"{slug}-{suffix}.json"
-            elif format_ == "ris":
-                filename = f"{slug}-{suffix}.ris"
-            elif format_ == "bibtex":
-                filename = f"{slug}-{suffix}.bib"
-            else:
-                filename = f"{slug}-{suffix}.txt"
-            options = {
-                "Content-Type": content_type,
-                "Content-Disposition": f"attachment; filename={filename}",
-            }
-            return (metadata, 200, options)
+            return (response["data"], 200, response["options"])
         else:
             try:
                 response = (

rogue_scholar_api/utils.py

@@ -1,13 +1,47 @@
 import requests
+import re
+from typing import Optional
 
 
-def get_doi_metadata_from_ra(doi: str, headers) -> str:
-    """use DOI content negotiation to get metadata in various formats"""
-    response = requests.get(doi, headers=headers)
-    response.encoding = 'UTF-8'
+def sanitize_suffix(str):
+    # Regular expression to only allow certain characters in DOI suffix
+    # taken from https://www.crossref.org/blog/dois-and-matching-regular-expressions/
+    m = re.match(r"^\[-._;\(\)/:A-Z0-9\]+$", str)
+    print(m)
+    return m
+
+
+def get_doi_metadata_from_ra(
+    doi: str, format_: str = "bibtex", style: str = "apa", locale: str = "en-US"
+) -> Optional[dict]:
+    """use DOI content negotiation to get metadata in various formats.
+    format_ can be bibtex, ris, csl, citation, with bibtex as default."""
+
+    content_types = {
+        "bibtex": "application/x-bibtex",
+        "ris": "application/x-research-info-systems",
+        "csl": "application/vnd.citationstyles.csl+json",
+        "citation": f"text/x-bibliography; style={style}; locale={locale}",
+    }
+    content_type = content_types.get(format_) or "application/x-bibtex"
+    response = requests.get(doi, headers={"Accept": content_type})
+    response.encoding = "UTF-8"
     if response.status_code >= 400:
-        return "Metadata not found"
-    return response.text.strip()
+        return None
+    basename = doi.replace("/", "-")
+    if format_ == "csl":
+        ext = "json"
+    elif format_ == "ris":
+        ext = "ris"
+    elif format_ == "bibtex":
+        ext = "bib"
+    else:
+        ext = "txt"
+    options = {
+        "Content-Type": content_type,
+        "Content-Disposition": f"attachment; filename={basename}.{ext}",
+    }
+    return {"doi": doi, "data": response.text.strip(), "options": options}
 
 
 # def format_metadata(meta: dict, to: str = "bibtex"):

tests/test-utils.py

@@ -7,10 +7,9 @@
 def test_get_doi_metadata_bibtex():
     "get doi metadata in bibtex format"
     doi = "https://doi.org/10.53731/ybhah-9jy85"
-    headers = {"Accept": "application/x-bibtex"}
-    bibtex = get_doi_metadata_from_ra(doi, headers=headers)
+    result = get_doi_metadata_from_ra(doi)
     assert (
-        bibtex
+        result["data"]
         == """@article{Fenner_2023,
 \tdoi = {10.53731/ybhah-9jy85},
 \turl = {https://doi.org/10.53731%2Fybhah-9jy85},
@@ -26,9 +25,36 @@ def test_get_doi_metadata_bibtex():
 def test_get_doi_metadata_citation():
     "get doi metadata as formatted citation"
     doi = "https://doi.org/10.53731/ybhah-9jy85"
-    headers = {"Accept": "text/x-bibliography; style=apa"}
-    citation = get_doi_metadata_from_ra(doi, headers=headers)
+    result = get_doi_metadata_from_ra(doi, format_="citation")
     assert (
-        citation
+        result["data"]
         == "Fenner, M. (2023). The rise of the (science) newsletter. https://doi.org/10.53731/ybhah-9jy85"
     )
+
+
+# def test_sanitize_cool_suffix():
+#     "sanitize cool suffix"
+#     suffix = "sfzv4-xdb68"
+#     sanitized_suffix = sanitize_suffix(suffix)
+#     assert sanitized_suffix == "sfzv4-xdb68"
+
+
+# def test_sanitize_semantic_suffix():
+#     "sanitize semantic suffix"
+#     suffix = "dini-blog.20230724"
+#     sanitized_suffix = sanitize_suffix(suffix)
+#     assert sanitized_suffix == "dini-blog.20230724"
+
+
+# def test_sanitize_sici_suffix():
+#     "sanitize sici suffix"
+#     suffix = "0002-8231(199412)45:10<737:TIODIM>2.3.TX;2-M"
+#     sanitized_suffix = sanitize_suffix(suffix)
+#     assert sanitized_suffix == "0002-8231(199412)45:10<737:TIODIM>2.3.TX;2-M"
+
+
+# def test_sanitize_invalid_suffix():
+#     "sanitize invalid suffix"
+#     suffix = "000 333"
+#     sanitized_suffix = sanitize_suffix(suffix)
+#     assert sanitized_suffix == "0002-8231(199412)45:10<737:TIODIM>2.3.TX;2-M"