Issues 16 and 17

Makefile

@@ -52,3 +52,6 @@ import-lmfdb:
 
 import-house-of-graphs:
 	python ./web/manage.py import_house_of_graphs
+
+import-oeis:
+	python ./web/manage.py import_oeis

README.md

@@ -48,7 +48,7 @@ Finally, populate the database.
     ```bash
     make install-scispacy
     ```
-    then configure your email `WIKIPEDIA_CONTACT_EMAIL` in [source_wikidata.py](web/slurper/source_wikidata.py)
+    then configure your email `MATHSWITCH_CONTACT_EMAIL` in [source_wikidata.py](web/slurper/source_wikidata.py)
     * This is needed
   * Then run the database population (make sure your db is cleared)
 

web/.env.example

@@ -2,7 +2,7 @@ SECRET_KEY="django-insecure-9wy9w#vf^tde0262doyy_j19=64c()_qub!1)f+fh-b^=7ndw*"
 
 # Wikidata
 WIKIDATA_PAGE_SIZE=1000
-WIKIPEDIA_CONTACT_EMAIL=my@email.com
+MATHSWITCH_CONTACT_EMAIL=my@email.com
 IMPORT_MATH_ENTITIES=True
 IMPORT_PHYSICS_ENTITIES=False
 

web/categorizer/wikidata_fetch_service.py

@@ -5,7 +5,7 @@
 import requests
 from concepts.models import Item
 
-from web.settings import WIKIPEDIA_CONTACT_EMAIL
+from web.settings import MATHSWITCH_CONTACT_EMAIL
 
 WIKIDATA_API_URL = "https://www.wikidata.org/w/api.php"
 
@@ -27,7 +27,7 @@ def __init__(self):
 
     def get_headers(self):
         return {
-            "User-Agent": f"MathSwitch/1.0 ({WIKIPEDIA_CONTACT_EMAIL})",
+            "User-Agent": f"MathSwitch/1.0 ({MATHSWITCH_CONTACT_EMAIL})",
             "Accept": "application/json",
         }
 

web/concepts/migrations/0024_alter_item_source.py

@@ -0,0 +1,32 @@
+# Generated by Django 4.2.30 on 2026-09-21 17:36
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("concepts", "0023_alter_item_source"),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name="item",
+            name="source",
+            field=models.CharField(
+                choices=[
+                    ("Wd", "Wikidata"),
+                    ("nL", "nLab"),
+                    ("MW", "MathWorld"),
+                    ("PW", "ProofWiki"),
+                    ("EoM", "Encyclopedia of Mathematics"),
+                    ("WpEN", "Wikipedia (English)"),
+                    ("AUm", "Agda Unimath"),
+                    ("LMF", "The L-functions and modular forms database"),
+                    ("HoG", "House of Graphs"),
+                    ("OEIS", "OEIS"),
+                ],
+                max_length=4,
+            ),
+        ),
+    ]

web/concepts/models.py

@@ -105,6 +105,7 @@ class Source(models.TextChoices):
         AGDA_UNIMATH = "AUm", "Agda Unimath"
         LMFDB = "LMF", "The L-functions and modular forms database"
         HOUSE_OF_GRAPHS = "HoG", "House of Graphs"
+        OEIS = "OEIS", "OEIS"
 
         @staticmethod
         def key():
@@ -118,6 +119,7 @@ def key():
                 Item.Source.AGDA_UNIMATH,
                 Item.Source.LMFDB,
                 Item.Source.HOUSE_OF_GRAPHS,
+                Item.Source.OEIS,
             ]
             return lambda item: SOURCES.index(item.source)
 

web/concepts/templates/index.html

@@ -30,6 +30,7 @@ <h2>About</h2>
   <li><a href="https://unimath.github.io/agda-unimath/">Agda Unimath</a> ({{ number_of_links.agda_unimath }} entries)</li>
   <li><a href="https://www.lmfdb.org/">LMFDB</a> ({{ number_of_links.lmfdb }} entries)</li>
   <li><a href="https://houseofgraphs.org/">House of Graphs</a> ({{ number_of_links.house_of_graphs }} entries)</li>
+  <li><a href="https://oeis.org/">OEIS</a> ({{ number_of_links.oeis }} entries)</li>
 </ul>
 <p>The entries are organized into a concept network that connects the same concept appearing in different sources.</p>
 <p>The team: <a href="https://katja.not.si">Katja Berčič</a> and Slobodan Stanojevikj.</p>

web/concepts/views.py

@@ -57,6 +57,7 @@ def home(request):
             "agda_unimath": items.filter(source=Item.Source.AGDA_UNIMATH).count(),
             "lmfdb": items.filter(source=Item.Source.LMFDB).count(),
             "house_of_graphs": items.filter(source=Item.Source.HOUSE_OF_GRAPHS).count(),
+            "oeis": items.filter(source=Item.Source.OEIS).count(),
         },
     }
     return render(request, "index.html", context)

web/slurper/management/commands/clear_oeis.py

@@ -0,0 +1,46 @@
+import logging
+import sys
+from datetime import timedelta
+
+from concepts.models import Item
+from django.core.management.base import BaseCommand
+from slurper.models import SlurperRun
+
+MIN_INTERVAL = timedelta(days=7)
+
+
+class Command(BaseCommand):
+    help = (
+        "Delete all OEIS items. Guarded by a 7-day throttle; use --force to override."
+    )
+
+    def add_arguments(self, parser):
+        parser.add_argument(
+            "--force",
+            action="store_true",
+            help="Clear even if the OEIS slurper ran within the last 7 days.",
+        )
+
+    def handle(self, *args, force=False, **options):
+        source = Item.Source.OEIS
+        if not force and not SlurperRun.can_run(source, MIN_INTERVAL):
+            if sys.stdin.isatty():
+                answer = (
+                    input(
+                        f"OEIS slurper ran within the last {MIN_INTERVAL.days} days. "
+                        f"Clear anyway? [y/N] "
+                    )
+                    .strip()
+                    .lower()
+                )
+                if answer not in ("y", "yes"):
+                    logging.info(f"[{source.label}] clear cancelled.")
+                    return
+            else:
+                logging.info(
+                    f"[{source.label}] clear skipped: ran less than "
+                    f"{MIN_INTERVAL.days} days ago (use --force to override)."
+                )
+                return
+        deleted, _ = Item.objects.filter(source=source).delete()
+        logging.info(f"[{source.label}] cleared {deleted} items.")

web/slurper/management/commands/import_oeis.py

@@ -0,0 +1,14 @@
+from django.core.management.base import BaseCommand
+from slurper import source_oeis
+
+
+class Command(BaseCommand):
+    def add_arguments(self, parser):
+        parser.add_argument(
+            "--force",
+            action="store_true",
+            help="Bypass the 7-day throttle and run anyway.",
+        )
+
+    def handle(self, *args, force=False, **options):
+        source_oeis.OEIS_SLURPER.save_items(force=force)

web/slurper/source_oeis.py

@@ -0,0 +1,103 @@
+"""
+Slurper for the On-Line Encyclopedia of Integer Sequences (OEIS).
+
+Sequence names are fetched from the OEIS names.gz dump. OEIS content is
+licensed under CC-BY-SA-4.0 by The On-Line Encyclopedia of Integer
+Sequences, https://oeis.org/.
+"""
+
+import gzip
+import logging
+from datetime import timedelta
+from typing import Iterator, Optional, Tuple
+
+import requests
+from concepts.models import Item, Link
+from slurper.models import SlurperRun
+
+from web.settings import MATHSWITCH_CONTACT_EMAIL
+
+
+class OeisSlurper:
+    NAMES_URL = "https://oeis.org/names.gz"
+    SEQUENCE_URL_PREFIX = "https://oeis.org/"
+    MIN_INTERVAL = timedelta(days=7)
+
+    def __init__(self):
+        self.source = Item.Source.OEIS
+
+    def fetch_names(self) -> bytes:
+        headers = {"User-Agent": f"MathSwitch/1.0 ({MATHSWITCH_CONTACT_EMAIL})"}
+        response = requests.get(self.NAMES_URL, headers=headers)
+        response.raise_for_status()
+        return gzip.decompress(response.content)
+
+    def parse_names(self, raw: bytes) -> Iterator[Tuple[str, str]]:
+        for line in raw.decode("utf-8", errors="replace").splitlines():
+            if not line or line.startswith("#"):
+                continue
+            identifier, _, description = line.partition(" ")
+            description = description.strip()
+            if not identifier or not description:
+                continue
+            yield identifier, description
+
+    def line_to_item(self, identifier: str, description: str) -> Item:
+        return Item(
+            source=self.source,
+            identifier=identifier,
+            url=self.SEQUENCE_URL_PREFIX + identifier,
+            name=identifier,
+            description=description,
+        )
+
+    def extract_candidate_name(self, description: str) -> Optional[str]:
+        if ":" not in description:
+            return None
+        candidate = description.split(":", 1)[0].strip()
+        return candidate or None
+
+    def save_items(self, force: bool = False):
+        if not force and not SlurperRun.can_run(self.source, self.MIN_INTERVAL):
+            logging.info(
+                f"[{self.source.label}] skipped: ran less than "
+                f"{self.MIN_INTERVAL.days} days ago (use --force to override)."
+            )
+            return
+        total_filled = 0
+        total_linked = 0
+        for identifier, description in self.parse_names(self.fetch_names()):
+            existing = Item.objects.filter(
+                source=self.source, identifier=identifier
+            ).first()
+            if existing is not None:
+                existing.description = description
+                existing.save(update_fields=["description"])
+                total_filled += 1
+                continue
+
+            candidate_name = self.extract_candidate_name(description)
+            if candidate_name is None:
+                continue
+            matches = list(
+                Item.objects.exclude(source=self.source).filter(
+                    name__iexact=candidate_name
+                )
+            )
+            if not matches:
+                continue
+
+            item = self.line_to_item(identifier, description)
+            item.save()
+            for match in matches:
+                Link.save_new(item, match, Link.Label.NAME_EQ)
+                total_linked += 1
+
+        SlurperRun.mark_ran(self.source)
+        logging.info(
+            f"[{self.source.label}] save_items finished: "
+            f"{total_filled} shells filled, {total_linked} NAME_EQ links created."
+        )
+
+
+OEIS_SLURPER = OeisSlurper()

web/slurper/source_wikidata.py

@@ -10,8 +10,8 @@
 from web.settings import (
     IMPORT_MATH_ENTITIES,
     IMPORT_PHYSICS_ENTITIES,
+    MATHSWITCH_CONTACT_EMAIL,
     WIKIDATA_PAGE_SIZE,
-    WIKIPEDIA_CONTACT_EMAIL,
 )
 
 # Wikipedia API contact email (required by Wikipedia API guidelines)
@@ -233,7 +233,7 @@ def fetch_pages(self):
 
     def get_headers(self):
         return {
-            "User-Agent": f"MathSwitch/1.0 ({WIKIPEDIA_CONTACT_EMAIL})",
+            "User-Agent": f"MathSwitch/1.0 ({MATHSWITCH_CONTACT_EMAIL})",
             "Accept": "application/json",
             "Accept-Language": "en-US,en;q=0.9",
         }
@@ -242,13 +242,13 @@ def fetch_article(self, json_item, index=None, total=None):
         global _missing_email_logged
 
         # Check if contact email is configured
-        if WIKIPEDIA_CONTACT_EMAIL is None:
+        if MATHSWITCH_CONTACT_EMAIL is None:
             if not _missing_email_logged:
                 logging.log(
                     logging.WARNING,
-                    "WIKIPEDIA_CONTACT_EMAIL is not set. "
+                    "MATHSWITCH_CONTACT_EMAIL is not set. "
                     "Wikipedia article fetching is disabled. "
-                    "Please set WIKIPEDIA_CONTACT_EMAIL at the top of "
+                    "Please set MATHSWITCH_CONTACT_EMAIL at the top of "
                     "source_wikidata.py to enable article fetching.",
                 )
                 _missing_email_logged = True