Add CKAN content provider

Author: u10313335

docs/source/usage.rst

@@ -39,7 +39,8 @@ where ``<source-repository>`` is:
 
   * a URL of a Git repository (``https://github.com/binder-examples/requirements``),
   * a Zenodo DOI (``10.5281/zenodo.1211089``),
-  * a SWHID_ (``swh:1:rev:999dd06c7f679a2714dfe5199bdca09522a29649``), or
+  * a SWHID_ (``swh:1:rev:999dd06c7f679a2714dfe5199bdca09522a29649``),
+  * a URL of a CKAN_ dataset (``https://demo.ckan.org/dataset/sample-dataset-1``), or
   * a path to a local directory (``a/local/directory``)
 
 of the source repository you want to build.
@@ -136,3 +137,4 @@ Command line API
 
 .. _Pytudes: https://github.com/norvig/pytudes
 .. _SWHID: https://docs.softwareheritage.org/devel/swh-model/persistent-identifiers.html
+.. _CKAN: https://ckan.org

repo2docker/app.py

@@ -152,6 +152,7 @@ def _default_log_level(self):
             contentproviders.Dataverse,
             contentproviders.Hydroshare,
             contentproviders.Swhid,
+            contentproviders.CKAN,
             contentproviders.Mercurial,
             contentproviders.Git,
         ],

repo2docker/contentproviders/__init__.py

@@ -1,4 +1,5 @@
 from .base import Local
+from .ckan import CKAN
 from .dataverse import Dataverse
 from .figshare import Figshare
 from .git import Git

repo2docker/contentproviders/ckan.py

@@ -0,0 +1,101 @@
+import re
+from datetime import datetime, timedelta, timezone
+from os import path
+from urllib.parse import urlparse
+
+from requests import Session
+
+from .. import __version__
+from .base import ContentProvider
+
+
+class CKAN(ContentProvider):
+    """Provide contents of a remote CKAN dataset."""
+
+    def __init__(self):
+        super().__init__()
+        self.session = Session()
+        self.session.headers.update(
+            {
+                "user-agent": f"repo2docker {__version__}",
+            }
+        )
+
+    def _fetch_version(self, api_url):
+        """Fetch dataset modified date and convert to epoch.
+        Borrowed from the Hydroshare provider.
+        """
+        package_show_url = f"{api_url}package_show?id={self.dataset_id}"
+        resp = self.urlopen(package_show_url).json()
+        date = resp["result"]["metadata_modified"]
+        parsed_date = datetime.strptime(date, "%Y-%m-%dT%H:%M:%S.%f")
+        epoch = parsed_date.replace(tzinfo=timezone(timedelta(0))).timestamp()
+        # truncate the timestamp
+        return str(int(epoch))
+
+    def _request(self, url, **kwargs):
+        return self.session.get(url, **kwargs)
+
+    urlopen = _request
+    url_regex = r"/dataset/[a-z0-9_\\-]*$"
+
+    def detect(self, source, ref=None, extra_args=None):
+        """Trigger this provider for things that resolve to a CKAN dataset."""
+        parsed_url = urlparse(source)
+        if not parsed_url.netloc:
+            return None
+
+        api_url = parsed_url._replace(
+            path=re.sub(self.url_regex, "/api/3/action/", parsed_url.path)
+        ).geturl()
+
+        status_show_url = f"{api_url}status_show"
+        resp = self.urlopen(status_show_url)
+        if resp.status_code == 200:
+            self.dataset_id = parsed_url.path.rsplit("/", maxsplit=1)[1]
+            self.version = self._fetch_version(api_url)
+            return {
+                "dataset_id": self.dataset_id,
+                "api_url": api_url,
+                "version": self.version,
+            }
+        else:
+            return None
+
+    def fetch(self, spec, output_dir, yield_output=False):
+        """Fetch a CKAN dataset."""
+        dataset_id = spec["dataset_id"]
+
+        yield f"Fetching CKAN dataset {dataset_id}.\n"
+        package_show_url = f"{spec['api_url']}package_show?id={dataset_id}"
+        resp = self.urlopen(
+            package_show_url,
+            headers={"accept": "application/json"},
+        )
+
+        dataset = resp.json()
+
+        yield "Fetching CKAN resources.\n"
+
+        resources = dataset["result"]["resources"]
+
+        for resource in resources:
+            file_url = resource["url"]
+            fname = file_url.rsplit("/", maxsplit=1)[-1]
+            if fname == "":
+                fname = resource["id"]
+
+            yield f"Requesting {file_url}\n"
+            resp = self._request(file_url, stream=True)
+            resp.raise_for_status()
+
+            dst_fname = path.join(output_dir, fname)
+            with open(dst_fname, "wb") as dst:
+                yield f"Fetching {fname}\n"
+                for chunk in resp.iter_content(chunk_size=None):
+                    dst.write(chunk)
+
+    @property
+    def content_id(self):
+        """A unique ID to represent the version of the content."""
+        return f"{self.dataset_id}.v{self.version}"

tests/unit/contentproviders/test_ckan.py

@@ -0,0 +1,58 @@
+import os
+from contextlib import contextmanager
+from tempfile import NamedTemporaryFile, TemporaryDirectory
+
+import pytest
+
+from repo2docker.contentproviders import CKAN
+
+test_ckan = CKAN()
+test_hosts = [
+    (
+        [
+            "http://demo.ckan.org/dataset/sample-dataset-1",
+        ],
+        {
+            "dataset_id": "sample-dataset-1",
+            "api_url": "http://demo.ckan.org/api/3/action/",
+            "version": "1707387710",
+        },
+    )
+]
+
+
+@pytest.mark.parametrize("test_input, expected", test_hosts)
+def test_detect_ckan(test_input, expected):
+    assert CKAN().detect(test_input[0]) == expected
+
+    # Don't trigger the CKAN content provider
+    assert CKAN().detect("/some/path/here") is None
+    assert CKAN().detect("https://example.com/path/here") is None
+    assert CKAN().detect("https://data.gov.tw/dataset/6564") is None
+
+
+@contextmanager
+def ckan_file():
+    with NamedTemporaryFile() as file:
+        file.write(b"some content")
+        yield file.name
+
+
+def test_ckan_fetch(requests_mock):
+    with ckan_file() as ckan_path:
+        mock_response = {"result": {"resources": [{"url": f"file://{ckan_path}"}]}}
+        requests_mock.get(
+            "http://demo.ckan.org/api/3/action/package_show?id=1234", json=mock_response
+        )
+        requests_mock.get(f"file://{ckan_path}", content=open(ckan_path, "rb").read())
+        with TemporaryDirectory() as d:
+            ckan = CKAN()
+            spec = {
+                "dataset_id": "1234",
+                "api_url": "http://demo.ckan.org/api/3/action/",
+            }
+            output = []
+            for l in ckan.fetch(spec, d):
+                output.append(l)
+            expected = {ckan_path.rsplit("/", maxsplit=1)[1]}
+            assert expected == set(os.listdir(d))