package: add a content_hash method for packages

This calculates a hash which depends on the complete content of the
package including sources and the associated `package.py` file.

2 files changed, 56 insertions, 0 deletions

diff --git a/lib/spack/spack/package.py b/lib/spack/spack/package.py
index 19a565e019..8ea6f4415c 100644
--- a/lib/spack/spack/package.py
+++ b/lib/spack/spack/package.py
@@ -33,9 +33,11 @@ Homebrew makes it very easy to create packages.  For a complete
 rundown on spack and how it differs from homebrew, look at the
 README.
 """
+import base64
 import contextlib
 import copy
 import functools
+import hashlib
 import inspect
 import itertools
 import os
@@ -74,6 +76,7 @@ from spack import directory_layout
 from spack.util.executable import which
 from spack.stage import Stage, ResourceStage, StageComposite
 from spack.util.environment import dump_environment
+from spack.util.package_hash import package_hash
 from spack.version import Version
 
 """Allowed URL schemes for spack packages."""
@@ -1161,6 +1164,28 @@ class PackageBase(with_metaclass(PackageMeta, object)):
             if self.spec.satisfies(spec))
         return sorted(patchesToApply, key=lambda p: p.path_or_url)
 
+    def content_hash(self, content=None):
+        """Create a hash based on the sources and logic used to build the
+        package. This includes the contents of all applied patches and the
+        contents of applicable functions in the package subclass."""
+        hashContent = list()
+        source_id = fs.for_package_version(self, self.version).source_id()
+        if not source_id:
+            # TODO? in cases where a digest or source_id isn't available,
+            # should this attempt to download the source and set one? This
+            # probably only happens for source repositories which are
+            # referenced by branch name rather than tag or commit ID.
+            message = 'Missing a source id for {s.name}@{s.version}'
+            tty.warn(message.format(s=self))
+            hashContent.append(''.encode('utf-8'))
+        else:
+            hashContent.append(source_id.encode('utf-8'))
+        hashContent.extend(':'.join((p.sha256, str(p.level))).encode('utf-8')
+                           for p in self.patches_to_apply())
+        hashContent.append(package_hash(self.spec, content))
+        return base64.b32encode(
+            hashlib.sha256(bytes().join(sorted(hashContent))).digest()).lower()
+
     @property
     def namespace(self):
         namespace, dot, module = self.__module__.rpartition('.')
diff --git a/lib/spack/spack/test/packages.py b/lib/spack/spack/test/packages.py
index 9533627b24..0fd115dac2 100644
--- a/lib/spack/spack/test/packages.py
+++ b/lib/spack/spack/test/packages.py
@@ -29,6 +29,7 @@ from llnl.util.filesystem import join_path
 from spack.repository import Repo
 from spack.util.naming import mod_to_class
 from spack.spec import Spec
+from spack.util.package_hash import package_content
 
 
 @pytest.mark.usefixtures('config', 'builtin_mock')
@@ -67,6 +68,36 @@ class TestPackage(object):
         assert 'Pmgrcollective' == mod_to_class('PmgrCollective')
         assert '_3db' == mod_to_class('3db')
 
+    def test_content_hash_all_same_but_patch_contents(self):
+        spec1 = Spec("hash-test1@1.1")
+        spec2 = Spec("hash-test2@1.1")
+        content1 = package_content(spec1)
+        content1 = content1.replace(spec1.package.__class__.__name__, '')
+        content2 = package_content(spec2)
+        content2 = content2.replace(spec2.package.__class__.__name__, '')
+        assert spec1.package.content_hash(content=content1) != \
+            spec2.package.content_hash(content=content2)
+
+    def test_content_hash_different_variants(self):
+        spec1 = Spec("hash-test1@1.2 +variantx")
+        spec2 = Spec("hash-test2@1.2 ~variantx")
+        content1 = package_content(spec1)
+        content1 = content1.replace(spec1.package.__class__.__name__, '')
+        content2 = package_content(spec2)
+        content2 = content2.replace(spec2.package.__class__.__name__, '')
+        assert spec1.package.content_hash(content=content1) == \
+            spec2.package.content_hash(content=content2)
+
+    def test_all_same_but_archive_hash(self):
+        spec1 = Spec("hash-test1@1.3")
+        spec2 = Spec("hash-test2@1.3")
+        content1 = package_content(spec1)
+        content1 = content1.replace(spec1.package.__class__.__name__, '')
+        content2 = package_content(spec2)
+        content2 = content2.replace(spec2.package.__class__.__name__, '')
+        assert spec1.package.content_hash(content=content1) != \
+            spec2.package.content_hash(content=content2)
+
     # Below tests target direct imports of spack packages from the
     # spack.pkg namespace
     def test_import_package(self):