Add Dataverse to UI. Fixes #900 (#969)

Add Dataverse to UI. Fixes #900

binderhub/app.py

@@ -33,7 +33,8 @@
 from .main import MainHandler, ParameterizedMainHandler, LegacyRedirectHandler
 from .repoproviders import (GitHubRepoProvider, GitRepoProvider,
                             GitLabRepoProvider, GistRepoProvider,
-                            ZenodoProvider, FigshareProvider)
+                            ZenodoProvider, FigshareProvider,
+                            DataverseProvider)
 from .metrics import MetricsHandler
 
 from .utils import ByteSpecification, url_path_join
@@ -393,6 +394,7 @@ def _add_slash(self, proposal):
             'gl': GitLabRepoProvider,
             'zenodo': ZenodoProvider,
             'figshare': FigshareProvider,
+            'dataverse': DataverseProvider,
         },
         config=True,
         help="""

binderhub/event-schemas/launch.json

@@ -12,7 +12,8 @@
                 "GitLab",
                 "Git",
                 "Zenodo",
-                "Figshare"
+                "Figshare",
+                "Dataverse"
             ],
             "description": "Provider for the repository being launched"
         },

binderhub/main.py

@@ -16,7 +16,8 @@
     "gl": "GitLab",
     "git": "Git repo",
     "zenodo": "Zenodo",
-    "figshare": "Figshare"
+    "figshare": "Figshare",
+    "dataverse": "Dataverse",
 }
 
 

binderhub/repoproviders.py

@@ -15,13 +15,14 @@
 import re
 import subprocess
 
+import escapism
 from prometheus_client import Gauge
 
 from tornado import gen
 from tornado.httpclient import AsyncHTTPClient, HTTPError, HTTPRequest
 from tornado.httputil import url_concat
 
-from traitlets import Dict, Unicode, Bool, default, List, observe
+from traitlets import Dict, Unicode, Bool, default, List
 from traitlets.config import LoggingConfigurable
 
 from .utils import Cache
@@ -294,6 +295,62 @@ def get_build_slug(self):
         return "figshare-{}".format(self.record_id)
 
 
+class DataverseProvider(RepoProvider):
+    name = Unicode("Dataverse")
+
+    @gen.coroutine
+    def get_resolved_ref(self):
+        client = AsyncHTTPClient()
+        req = HTTPRequest("https://doi.org/{}".format(self.spec),
+                          user_agent="BinderHub")
+        r = yield client.fetch(req)
+
+        search_url = urllib.parse.urlunparse(
+            urllib.parse.urlparse(r.effective_url)._replace(
+                path="/api/datasets/:persistentId"
+            )
+        )
+        req = HTTPRequest(search_url, user_agent="BinderHub")
+        r = yield client.fetch(req)
+        resp = json.loads(r.body)
+
+        assert resp["status"] == "OK"
+
+        self.identifier = resp["data"]["identifier"]
+        self.record_id = "{datasetId}.v{major}.{minor}".format(
+            datasetId=resp["data"]["id"],
+            major=resp["data"]["latestVersion"]["versionNumber"],
+            minor=resp["data"]["latestVersion"]["versionMinorNumber"],
+        )
+
+        # NOTE: data.protocol should be potentially prepended here
+        #  {protocol}:{authority}/{identifier}
+        self.resolved_spec = "{authority}/{identifier}".format(
+            authority=resp["data"]["authority"],
+            identifier=resp["data"]["identifier"],
+        )
+        self.resolved_ref_url = resp["data"]["persistentUrl"]
+        return self.record_id
+
+    async def get_resolved_spec(self):
+        if not hasattr(self, 'resolved_spec'):
+            await self.get_resolved_ref()
+        return self.resolved_spec
+
+    async def get_resolved_ref_url(self):
+        if not hasattr(self, 'resolved_ref_url'):
+            await self.get_resolved_ref()
+        return self.resolved_ref_url
+
+    def get_repo_url(self):
+        # While called repo URL, the return value of this function is passed
+        # as argument to repo2docker, hence we return the spec as is.
+        return self.spec
+
+    def get_build_slug(self):
+        return "dataverse-" + escapism.escape(self.identifier, escape_char="-").lower()
+
+
 class GitRepoProvider(RepoProvider):
     """Bare bones git repo provider.
 

binderhub/static/js/index.js

@@ -89,6 +89,11 @@ function updateRepoText() {
     $("#ref").prop("disabled", true);
     $("label[for=ref]").prop("disabled", true);
   }
+  else if (provider === "dataverse") {
+    text = "Dataverse DOI (10.7910/DVN/TJCLKP)";
+    $("#ref").prop("disabled", true);
+    $("label[for=ref]").prop("disabled", true);
+  }
   $("#repository").attr('placeholder', text);
   $("label[for=repository]").text(text);
   $("#ref").attr('placeholder', tag_text);
@@ -112,7 +117,7 @@ function getBuildFormValues() {
   }
 
   var ref = $('#ref').val().trim() || 'master';
-  if (providerPrefix === 'zenodo' || providerPrefix === 'figshare') {
+  if (providerPrefix === 'zenodo' || providerPrefix === 'figshare' || providerPrefix === 'dataverse') {
     ref = "";
   }
   var path = $('#filepath').val().trim();

binderhub/templates/index.html

@@ -55,6 +55,7 @@ <h4 id="form-header" class='row'>Build and launch a repository</h4>
                 <li class="dropdown-item" value="git"><a href="#">Git repository</a></li>
                 <li class="dropdown-item" value="zenodo"><a href="#">Zenodo DOI</a></li>
                 <li class="dropdown-item" value="figshare"><a href="#">Figshare DOI</a></li>
+                <li class="dropdown-item" value="dataverse"><a href="#">Dataverse DOI</a></li>
               </ul>
             </div>
             <input class="form-control" type="text" id="repository" data-lpignore="true" placeholder="GitHub repository name or link"/>

binderhub/tests/test_repoproviders.py

@@ -6,7 +6,8 @@
 
 from binderhub.repoproviders import (
     tokenize_spec, strip_suffix, GitHubRepoProvider, GitRepoProvider,
-    GitLabRepoProvider, GistRepoProvider, ZenodoProvider, FigshareProvider
+    GitLabRepoProvider, GistRepoProvider, ZenodoProvider, FigshareProvider,
+    DataverseProvider
 )
 
 
@@ -98,6 +99,30 @@ async def test_figshare(spec, resolved_spec, resolved_ref, resolved_ref_url, bui
     assert spec == resolved_spec
 
 
+@pytest.mark.parametrize('spec,resolved_spec,resolved_ref,resolved_ref_url,build_slug', [
+    ['10.7910/DVN/TJCLKP',
+     '10.7910/DVN/TJCLKP',
+     '3035124.v3.0',
+     'https://doi.org/10.7910/DVN/TJCLKP',
+     'dataverse-dvn-2ftjclkp'],
+])
+async def test_dataverse(spec, resolved_spec, resolved_ref, resolved_ref_url, build_slug):
+    provider = DataverseProvider(spec=spec)
+
+    # have to resolve the ref first
+    ref = await provider.get_resolved_ref()
+    assert ref == resolved_ref
+
+    slug = provider.get_build_slug()
+    assert slug == build_slug
+    repo_url = provider.get_repo_url()
+    assert repo_url == spec
+    ref_url = await provider.get_resolved_ref_url()
+    assert ref_url == resolved_ref_url
+    spec = await provider.get_resolved_spec()
+    assert spec == resolved_spec
+
+
 @pytest.mark.github_api
 def test_github_ref():
     namespace = 'jupyterhub/zero-to-jupyterhub-k8s'

doc/developer/repoproviders.rst

@@ -31,6 +31,8 @@ Currently supported providers, their prefixes and specs are:
     +------------+--------------------+-------------------------------------------------------------+-------------------------------------------------------------------------------------------------------------------------------------------+
     | Figshare   | ``figshare``       | ``<figshare-DOI>``                                          | `FigShare <https://figshare.com/>`_ is company that offers hosting for scholarly artifacts (such as code repositories).                   |
     +------------+--------------------+-------------------------------------------------------------+-------------------------------------------------------------------------------------------------------------------------------------------+
+    | Dataverse  | ``dataverse``      | ``<dataverse-DOI>``                                         | `Dataverse <https://dataverse.org/>`_ is open source research data repository software installed all over the world.                      |
+    +------------+--------------------+-------------------------------------------------------------+-------------------------------------------------------------------------------------------------------------------------------------------+
     | Git        | ``git``            | ``<url-escaped-url>/<commit-sha>``                          | A generic repository provider for URLs that point directly to a git repository.                                                           |
     +------------+--------------------+-------------------------------------------------------------+-------------------------------------------------------------------------------------------------------------------------------------------+
 

doc/reference/repoproviders.rst

@@ -52,6 +52,13 @@ Module: :mod:`binderhub.repoproviders`
     :members:
 
 
+:class:`DataverseProvider`
+---------------------------
+
+.. autoconfigurable:: DataverseProvider
+    :members:
+
+
 :class:`GitRepoProvider`
 ---------------------------