Pretalx data

deploy/playbooks/04_cron.yml

@@ -0,0 +1,9 @@
+- name: Scheduled tasks using the bot user
+  hosts: intbot_app
+
+  tasks:
+    - name: "Download pretalx data every hour"
+      ansible.builtin.cron:
+        name: "Download pretalx data every hour"
+        minute: "5"   # run on the 5th minute of every hour
+        job: "make prod/cron/pretalx"

deploy/templates/app/Makefile.app.j2

@@ -1,18 +1,22 @@
+MAKE_APP="docker compose run app make"
 
 echo:
 	"Dummy target, to not run something accidentally"
 
 prod/migrate:
-	docker compose run app make in-container/migrate
+	$(MAKE_APP) in-container/migrate
 
 prod/shell:
-	docker compose run app make in-container/shell
+	$(MAKE_APP) in-container/shell
 
 prod/db_shell:
-	docker compose run app make in-container/db_shell
+	$(MAKE_APP) in-container/db_shell
 
 prod/manage:
-	docker compose run app make in-container/manage ARG=$(ARG)
+	$(MAKE_APP) in-container/manage ARG=$(ARG)
+
+prod/cron/pretalx:
+	$(MAKE_APP) in-container/manage ARG="download_pretalx_data --event=europython-2025"
 
 logs:
 	docker compose logs -f

intbot/core/admin.py

@@ -1,6 +1,6 @@
 import json
 
-from core.models import DiscordMessage, Webhook
+from core.models import DiscordMessage, PretalxData, Webhook
 from django.contrib import admin
 from django.utils.html import format_html
 
@@ -26,12 +26,12 @@ class WebhookAdmin(admin.ModelAdmin):
         "processed_at",
     ]
 
-    def pretty_meta(self, obj):
+    def pretty_meta(self, obj: Webhook):
         return format_html("<pre>{}</pre>", json.dumps(obj.meta, indent=4))
 
     pretty_meta.short_description = "Meta"
 
-    def pretty_content(self, obj):
+    def pretty_content(self, obj: Webhook):
         return format_html("<pre>{}</pre>", json.dumps(obj.content, indent=4))
 
     pretty_content.short_description = "Content"
@@ -61,11 +61,38 @@ class DiscordMessageAdmin(admin.ModelAdmin):
         "sent_at",
     ]
 
-    def content_short(self, obj):
+    def content_short(self, obj: DiscordMessage):
         # NOTE(artcz) This can create false shortcuts, but for most messages is
         # good enough, because most of them are longer than 20 chars
         return f"{obj.content[:10]}...{obj.content[-10:]}"
 
 
+class PretalxDataAdmin(admin.ModelAdmin):
+    list_display = [
+        "uuid",
+        "resource",
+        "created_at",
+        "modified_at",
+    ]
+    list_filter = [
+        "created_at",
+        "resource",
+    ]
+    readonly_fields = fields = [
+        "uuid",
+        "resource",
+        "pretty_content",
+        "created_at",
+        "modified_at",
+        "processed_at",
+    ]
+
+    def pretty_content(self, obj: PretalxData):
+        return format_html("<pre>{}</pre>", json.dumps(obj.content, indent=4))
+
+    pretty_content.short_description = "Content"
+
+
 admin.site.register(Webhook, WebhookAdmin)
 admin.site.register(DiscordMessage, DiscordMessageAdmin)
+admin.site.register(PretalxData, PretalxDataAdmin)

intbot/core/integrations/pretalx.py

@@ -0,0 +1,89 @@
+import logging
+from typing import Any
+
+import httpx
+from core.models import PretalxData
+from django.conf import settings
+
+logger = logging.getLogger(__name__)
+
+PRETALX_EVENTS = [
+    "europython-2022",
+    "europython-2023",
+    "europython-2024",
+    "europython-2025",
+]
+
+ENDPOINTS = {
+    # Questions need to be passed to include answers in the same endpoint,
+    # saving us later time with joining the answers.
+    PretalxData.PretalxResources.submissions: "submissions/?questions=all",
+    PretalxData.PretalxResources.speakers: "speakers/?questions=all",
+}
+
+
+JsonType = dict[str, Any]
+
+
+def get_event_url(event):
+    assert event in PRETALX_EVENTS
+
+    return f"https://pretalx.com/api/events/{event}/"
+
+
+def fetch_pretalx_data(
+    event: str, resource: PretalxData.PretalxResources
+) -> list[JsonType]:
+    headers = {
+        "Authorization": f"Token {settings.PRETALX_API_TOKEN}",
+        "Content-Type": "application/json",
+    }
+
+    base_url = get_event_url(event)
+    endpoint = ENDPOINTS[resource]
+    url = f"{base_url}{endpoint}"
+
+    # Pretalx paginates the output, so we will need to do multiple requests and
+    # then merge multiple pages to one big dictionary
+    results = []
+    page = 0
+
+    # This takes advantage of the fact that url will contain a url to the
+    # next page, until there is more data to fetch. If this is the last page,
+    # then the url will be None (falsy), and thus stop the while loop.
+    while url:
+        page += 1
+        response = httpx.get(url, headers=headers)
+
+        if response.status_code != 200:
+            raise Exception(f"Error {response.status_code}: {response.text}")
+
+        logger.info("Fetching data from %s, page %s", url, page)
+
+        data = response.json()
+        results += data["results"]
+        url = data["next"]
+
+    return results
+
+
+def download_latest_submissions(event: str) -> PretalxData:
+    data = fetch_pretalx_data(event, PretalxData.PretalxResources.submissions)
+
+    pretalx_data = PretalxData.objects.create(
+        resource=PretalxData.PretalxResources.submissions,
+        content=data,
+    )
+
+    return pretalx_data
+
+
+def download_latest_speakers(event: str) -> PretalxData:
+    data = fetch_pretalx_data(event, PretalxData.PretalxResources.speakers)
+
+    pretalx_data = PretalxData.objects.create(
+        resource=PretalxData.PretalxResources.speakers,
+        content=data,
+    )
+
+    return pretalx_data

intbot/core/management/commands/download_pretalx_data.py

@@ -0,0 +1,28 @@
+from core.integrations.pretalx import (
+    PRETALX_EVENTS,
+    download_latest_speakers,
+    download_latest_submissions,
+)
+from django.core.management.base import BaseCommand
+
+
+class Command(BaseCommand):
+    help = "Downloads latest pretalx data"
+
+    def add_arguments(self, parser):
+        # Add keyword argument event
+        parser.add_argument(
+            "--event",
+            choices=PRETALX_EVENTS,
+            help="slug of the event (for example `europython-2025`)",
+            required=True,
+        )
+
+    def handle(self, **kwargs):
+        event = kwargs["event"]
+
+        self.stdout.write(f"Downloading latest speakers from pretalx... {event}")
+        download_latest_speakers(event)
+
+        self.stdout.write(f"Downloading latest submissions from pretalx... {event}")
+        download_latest_submissions(event)

intbot/core/migrations/0005_add_pretalx_data_model.py

@@ -0,0 +1,43 @@
+# Generated by Django 5.1.4 on 2025-04-18 11:43
+
+import uuid
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+    dependencies = [
+        ("core", "0004_add_inbox_item_model"),
+    ]
+
+    operations = [
+        migrations.CreateModel(
+            name="PretalxData",
+            fields=[
+                (
+                    "id",
+                    models.BigAutoField(
+                        auto_created=True,
+                        primary_key=True,
+                        serialize=False,
+                        verbose_name="ID",
+                    ),
+                ),
+                ("uuid", models.UUIDField(default=uuid.uuid4)),
+                (
+                    "resource",
+                    models.CharField(
+                        choices=[
+                            ("submissions", "Submissions"),
+                            ("speakers", "Speakers"),
+                            ("schedule", "Schedule"),
+                        ],
+                        max_length=255,
+                    ),
+                ),
+                ("content", models.JSONField()),
+                ("created_at", models.DateTimeField(auto_now_add=True)),
+                ("modified_at", models.DateTimeField(auto_now=True)),
+                ("processed_at", models.DateTimeField(blank=True, null=True)),
+            ],
+        ),
+    ]

intbot/core/models.py

@@ -81,3 +81,32 @@ def summary(self) -> str:
 
     def __str__(self):
         return f"{self.uuid} {self.author}: {self.content[:30]}"
+
+
+class PretalxData(models.Model):
+    """
+    Table to store raw data download from pretalx for later parsing.
+
+    We first download data from pretalx to this table, and then fire a separate
+    background task that pulls data from this table and stores in separate
+    "business" tables, like "Proposal" or "Speaker".
+    """
+
+    class PretalxResources(models.TextChoices):
+        submissions = "submissions", "Submissions"
+        speakers = "speakers", "Speakers"
+        schedule = "schedule", "Schedule"
+
+    uuid = models.UUIDField(default=uuid.uuid4)
+    resource = models.CharField(
+        max_length=255,
+        choices=PretalxResources.choices,
+    )
+    content = models.JSONField()
+
+    created_at = models.DateTimeField(auto_now_add=True)
+    modified_at = models.DateTimeField(auto_now=True)
+    processed_at = models.DateTimeField(blank=True, null=True)
+
+    def __str__(self):
+        return f"{self.uuid}"

intbot/intbot/settings.py

@@ -197,6 +197,9 @@ def get(name) -> str:
 ZAMMAD_GROUP_SPONSORS = get("ZAMMAD_GROUP_SPONSORS")
 ZAMMAD_GROUP_GRANTS = get("ZAMMAD_GROUP_GRANTS")
 
+# Pretalx
+PRETALX_API_TOKEN = get("PRETALX_API_TOKEN")
+
 
 if DJANGO_ENV == "dev":
     DEBUG = True
@@ -282,6 +285,8 @@ def get(name) -> str:
     ZAMMAD_GROUP_HELPDESK = "TestZammad Helpdesk"
     ZAMMAD_GROUP_BILLING = "TestZammad Billing"
 
+    PRETALX_API_TOKEN = "Test-Pretalx-API-token"
+
 
 elif DJANGO_ENV == "local_container":
     DEBUG = False

intbot/tests/test_admin.py

@@ -2,7 +2,7 @@
 Sanity checks (mostly) if the admin resources are available
 """
 
-from core.models import DiscordMessage, Webhook
+from core.models import DiscordMessage, PretalxData, Webhook
 
 
 def test_admin_for_webhooks_sanity_check(admin_client):
@@ -32,3 +32,35 @@ def test_admin_for_discordmessages_sanity_check(admin_client):
     assert str(dm.uuid).encode() in response.content
     assert dm.channel_id.encode() in response.content
     assert dm.channel_name.encode() in response.content
+
+
+def test_admin_list_for_pretalx_data(admin_client):
+    """Simple sanity check if the page loads correctly"""
+    url = "/admin/core/pretalxdata/"
+    pd = PretalxData.objects.create(
+        resource=PretalxData.PretalxResources.speakers,
+        content={},
+    )
+    assert pd.uuid
+
+    response = admin_client.get(url)
+
+    assert response.status_code == 200
+    assert str(pd.uuid).encode() in response.content
+    assert pd.get_resource_display().encode() in response.content
+
+
+def test_admin_change_for_pretalx_data(admin_client):
+    """Simple sanity check if the page loads correctly"""
+    url = "/admin/core/pretalxdata/"
+    pd = PretalxData.objects.create(
+        resource=PretalxData.PretalxResources.speakers,
+        content={},
+    )
+    assert pd.uuid
+
+    response = admin_client.get(f"{url}{pd.pk}/change/")
+
+    assert response.status_code == 200
+    assert str(pd.uuid).encode() in response.content
+    assert pd.get_resource_display().encode() in response.content