Solidify logic for picking rspm vs mran

- MRAN doesn't seem to have R 4.1 specific snapshots, so let's
  default to RSPM for anything 4.1+.
- Otherwise, snapshot dates in 2022 will result in using rspm

repo2docker/buildpacks/r.py

@@ -185,28 +185,50 @@ class RBuildPack(PythonBuildPack):
 
         return super().get_packages().union(packages)
 
-    def get_cran_mirror_url(self, snapshot_date):
+    def get_rspm_snapshot_url(self, snapshot_date, max_days_prior=7):
-        # Call the API to find out if we have a snapshot available for the given date.
+        for i in range(max_days_prior):
-        # If so, use the URL for that snapshot. If not, fall back to MRAN.
+            snapshots = requests.post(
-        snapshots = requests.post(
+                "https://packagemanager.rstudio.com/__api__/url",
-            "https://packagemanager.rstudio.com/__api__/url",
+                # Ask for midnight UTC snapshot
-            # Ask for midnight UTC snapshot
+                json={
-            json={
+                    "repo": "all",
-                "repo": "all",
+                    "snapshot": (snapshot_date - datetime.timedelta(days=i)).strftime("%Y-%m-%dT00:00:00Z"),
-                "snapshot": snapshot_date.strftime("%Y-%m-%dT00:00:00Z"),
+                },
-            },
+            ).json()
-        ).json()
+            # Construct a snapshot URL that will give us binary packages for Ubuntu Bionic (18.04)
-        # Construct a snapshot URL that will give us binary packages for Ubuntu Bionic (18.04)
+            if "upsi" in snapshots:
-        if "upsi" in snapshots:
+                return (
-            return (
+                    "https://packagemanager.rstudio.com/all/__linux__/bionic/"
-                "https://packagemanager.rstudio.com/all/__linux__/bionic/"
+                    + snapshots["upsi"]
-                + snapshots["upsi"]
+                )
-            )
+        raise ValueError('No snapshot found for {} or {} days prior in packagemanager.rstudio.com'.format(
+            snapshot_date.strftime("%Y-%m-%d"), max_days_prior
+        ))
+
+    def get_mran_snapshot_url(self, snapshot_date, max_days_prior=7):
+        for i in range(max_days_prior):
+            try_date = snapshot_date - datetime.timedelta(days=i)
+            # Fall back to MRAN if packagemanager.rstudio.com doesn't have it
+            url = "https://mran.microsoft.com/snapshot/{}".format(
+                try_date.isoformat()
+            )
+            r = requests.head(url)
+            if r.ok:
+                return url
+        else:
+            raise ValueError('No snapshot found for {} or {} days prior in mran.microsoft.com'.format(
+                snapshot_date.strftime("%Y-%m-%d"), max_days_prior
+            ))
+
+    def get_cran_mirror_url(self, snapshot_date):
+        # Date after which we will use rspm + binary packages instead of MRAN + source packages
+        rspm_cutoff_date = datetime.date(2022, 1, 1)
+
+        if snapshot_date >= rspm_cutoff_date or self.r_version >= V('4.1'):
+            return self.get_rspm_snapshot_url(snapshot_date)
+        else:
+            return self.get_mran_snapshot_url(snapshot_date)
 
-        # Fall back to MRAN if packagemanager.rstudio.com doesn't have it
-        return "https://mran.microsoft.com/snapshot/{}".format(
-            snapshot_date.isoformat()
-        )
 
     def get_devtools_snapshot_date(self):
         """

tests/r/r4.1/runtime.txt

@@ -1 +1 @@
-r-4.1-2021-07-07
+r-4.1-2021-10-22

tests/r/r4.1/verify

@@ -1,8 +1,12 @@
 #!/usr/bin/env Rscript
 library('ggplot2')
 
-print(version)
 # Fail if version is not 4.1
 if (!(version$major == "4" && as.double(version$minor) >= 1 && as.double(version$minor) < 2)) {
   quit("yes", 1)
 }
+
+# R 4.1 always uses rspm snapshots
+if (!(startsWith(options()$repos["CRAN"], "https://packagemanager.rstudio.com"))) {
+  quit("yes", 1)
+}

tests/unit/test_r.py

@@ -64,28 +64,46 @@ def test_mran_date(tmpdir, runtime, expected):
     assert r.checkpoint_date == date(*expected)
 
 
-@pytest.mark.parametrize("expected", ["2019-12-29", "2019-12-26"])
+def test_snapshot_rspm_date():
-def test_mran_latestdate(tmpdir, expected):
+    test_dates = {
+        # Even though there is no snapshot specified in the interface at https://packagemanager.rstudio.com/client/#/repos/1/overview
+        # For 2021 Oct 22, the API still returns a valid URL that one can install
+        # packages from - probably some server side magic that repeats our client side logic.
+        # No snapshot for this date from
+        date(2021, 10, 22): date(2021, 10, 22),
+        # Snapshot exists for this date
+        date(2022, 1, 1): date(2022, 1, 1),
+    }
+
+    r = buildpacks.RBuildPack()
+    for requested, expected in test_dates.items():
+        snapshot_url = r.get_rspm_snapshot_url(requested)
+        print(snapshot_url)
+        assert snapshot_url.startswith(
+            "https://packagemanager.rstudio.com/all/__linux__/bionic/" + expected.strftime("%Y-%m-%d")
+        )
+
+    with pytest.raises(ValueError):
+        r.get_rspm_snapshot_url(date(1691, 9, 5))
+
+
+@pytest.mark.parametrize("expected", [date(2019, 12, 29), date(2019, 12, 26)])
+@pytest.mark.parametrize("requested", [date(2019, 12, 31)])
+def test_snapshot_mran_date(requested, expected):
     def mock_request_head(url):
         r = Response()
-        if url == "https://mran.microsoft.com/snapshot/" + expected:
+        if url == "https://mran.microsoft.com/snapshot/" + expected.isoformat():
             r.status_code = 200
         else:
             r.status_code = 404
             r.reason = "Mock MRAN no snapshot"
         return r
 
-    tmpdir.chdir()
-
-    with open("DESCRIPTION", "w") as f:
-        f.write("")
-
     with patch("requests.head", side_effect=mock_request_head):
-        with patch("datetime.date") as mockdate:
+        r = buildpacks.RBuildPack()
-            mockdate.today.return_value = date(2019, 12, 31)
+        assert r.get_mran_snapshot_url(
-            r = buildpacks.RBuildPack()
+            requested
-            r.detect()
+        ) == "https://mran.microsoft.com/snapshot/{}".format(expected.isoformat())
-    assert r.checkpoint_date.isoformat() == expected
 
 
 def test_install_from_base(tmpdir):