allow only approved repos

admin.py

@@ -36,6 +36,7 @@ regluit imports
 from regluit import payment
 from regluit.core import models
 from regluit.marc.models import MARCRecord
+from regluit.api.models import AllowedRepo
 from regluit.core.lookups import (
     PublisherNameLookup,
     WorkLookup,
@@ -233,6 +234,9 @@ class MARCRecordAdmin(ModelAdmin):
     date_hierarchy = 'created'
     form = MARCRecordAdminForm
 
+class AllowedRepoAdmin(ModelAdmin):
+    list_display = ('org', 'repo_name')
+
 admin_site = RegluitAdmin("Admin")
 
 admin_site.register(User, UserAdmin)
@@ -258,6 +262,7 @@ admin_site.register(models.CeleryTask, CeleryTaskAdmin)
 admin_site.register(models.Press, PressAdmin)
 admin_site.register(models.Gift, GiftAdmin)
 admin_site.register(MARCRecord, MARCRecordAdmin)
+admin_site.register(AllowedRepo, AllowedRepoAdmin)
 admin_site.register(models.Relation, RelationAdmin)
 
 # payments

api/migrations/0001_initial.py

@@ -0,0 +1,34 @@
+# -*- coding: utf-8 -*-
+import datetime
+from south.db import db
+from south.v2 import SchemaMigration
+from django.db import models
+
+
+class Migration(SchemaMigration):
+
+    def forwards(self, orm):
+        # Adding model 'AllowedRepo'
+        db.create_table('api_allowedrepo', (
+            ('id', self.gf('django.db.models.fields.AutoField')(primary_key=True)),
+            ('org', self.gf('django.db.models.fields.CharField')(max_length=39)),
+            ('repo_name', self.gf('django.db.models.fields.CharField')(max_length=100)),
+        ))
+        db.send_create_signal('api', ['AllowedRepo'])
+
+
+    def backwards(self, orm):
+        # Deleting model 'AllowedRepo'
+        db.delete_table('api_allowedrepo')
+
+
+    models = {
+        'api.allowedrepo': {
+            'Meta': {'object_name': 'AllowedRepo'},
+            'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'}),
+            'org': ('django.db.models.fields.CharField', [], {'max_length': '39'}),
+            'repo_name': ('django.db.models.fields.CharField', [], {'max_length': '100'})
+        }
+    }
+
+    complete_apps = ['api']

api/models.py

@@ -0,0 +1,31 @@
+from django.db import models
+
+
+
+#https://github.com/GITenberg/Adventures-of-Huckleberry-Finn_76/raw/master/metadata.yaml
+
+def repo_allowed(repo_url):
+    if not repo_url.startswith('https://github.com/'):
+        return (False, "repo url must start with 'https://github.com/'")
+    try:
+        (org,repo_name,raw,branch,filename) = repo_url[19:].split('/')
+    except ValueError:
+        return (False, "repo url must be well formed, metadata at top repo level")
+    if not raw == 'raw':
+        return (False, "repo url must point at 'raw' file")
+    if not filename == 'metadata.yaml':
+        return (False, "repo filename must be 'metadata.yaml'")
+    if not branch == 'master':
+        return (False, "repo branch must be 'master'")
+    allowed_repos = AllowedRepo.objects.filter(org=org)
+    for allowed_repo in allowed_repos:
+        if allowed_repo.repo_name == '*':
+            return (True, '*')
+        if allowed_repo.repo_name == repo_name:
+            return (True, '*')
+    return  (False, "no allowed repos in that org")
+
+class AllowedRepo(models.Model):
+    org = models.CharField(max_length=39) 
+    repo_name = models.CharField(max_length=100)
+    

api/tests.py

@@ -18,6 +18,7 @@ import regluit.core.isbn
 
 from regluit.core import bookloader, models
 from regluit.utils.localdatetime import now
+from regluit.api import models as apimodels
 
 class ApiTests(TestCase):
     work_id=None
@@ -125,3 +126,21 @@ class ApiTests(TestCase):
         r = self.client.get('/api/widget/%s/'%self.work_id)
         self.assertEqual(r.status_code, 200)
 
+class AllowedRepoTests(TestCase):
+    def setUp(self):
+        apimodels.AllowedRepo.objects.create(org='test',repo_name='test')
+        apimodels.AllowedRepo.objects.create(org='star',repo_name='*')
+
+    def test_urls(self):
+        #good
+        self.assertTrue(apimodels.repo_allowed('https://github.com/test/test/raw/master/metadata.yaml')[0])
+        self.assertTrue(apimodels.repo_allowed('https://github.com/star/test/raw/master/metadata.yaml')[0])
+        # bad urls
+        self.assertFalse(apimodels.repo_allowed('auhf8peargp8ge')[0])
+        self.assertFalse(apimodels.repo_allowed('')[0])
+        self.assertFalse(apimodels.repo_allowed('http://github.com/test/test/raw/master/metadata.yaml')[0])
+        self.assertFalse(apimodels.repo_allowed('https://github.com/test/test/raw/master/samples/metadata.yaml')[0])
+        self.assertFalse(apimodels.repo_allowed('https://github.com/test/test/raw/branch/metadata.yaml')[0])
+        self.assertFalse(apimodels.repo_allowed('https://github.com/test/test/master/metadata.yaml')[0])
+        self.assertFalse(apimodels.repo_allowed('https://github.com/test/test/raw/master/metadata.json')[0])
+

api/views.py

@@ -17,6 +17,7 @@ from django.http import (
 import regluit.core.isbn
 from regluit.core.bookloader import load_from_yaml
 from regluit.api import opds
+from regluit.api.models import repo_allowed
 
 from regluit.core import models
 
@@ -70,6 +71,9 @@ def load_yaml(request):
     repo_url = request.POST.get('repo_url', None)
     if not repo_url:
         return HttpResponse('needs repo_url')
+    (allowed,reason) =repo_allowed(repo_url)
+    if not allowed:
+        return HttpResponse('repo_url not allowed: '+reason)
     try:
         work_id = load_from_yaml(repo_url)
         return HttpResponseRedirect(reverse('work', args=[work_id]))