Compare revisions

Javier Jardón · Jürg Billeter · Jürg Billeter · Jürg Billeter · Jürg Billeter · Chandan Singh
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -53,26 +53,20 @@ tests-fedora-28:
  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:28-master-46405991
  <<: *tests
-tests-ubuntu-18.04:
+tests-fedora-29:
-  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-ubuntu:18.04-master-46405991
+  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:29-master-47052095
  <<: *tests
-tests-python-3.7-stretch:
+tests-ubuntu-18.04:
-  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-python:3.7-stretch-master-46405991
+  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-ubuntu:18.04-master-46405991
  <<: *tests
-  variables:
-    # Note that we explicitly specify TOXENV in this case because this
-    # image has both 3.6 and 3.7 versions. python3.6 cannot be removed because
-    # some of our base dependencies declare it as their runtime dependency.
-    TOXENV: py37
 tests-centos-7.6:
  <<: *tests
  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-centos:7.6.1810-master-46405991
-overnight-fedora-28-aarch64:
+overnight-fedora-29-aarch64:
-  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:aarch64-28-master-46405991
+  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:aarch64-29-master-47052095
  tags:
    - aarch64
  <<: *tests
@@ -91,7 +85,7 @@ overnight-fedora-28-aarch64:
 tests-unix:
  # Use fedora here, to a) run a test on fedora and b) ensure that we
  # can get rid of ostree - this is not possible with debian-8
-  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:28-master-46405991
+  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:29-master-47052095
  <<: *tests
  variables:
    BST_FORCE_BACKEND: "unix"
@@ -109,7 +103,7 @@ tests-unix:
 tests-fedora-missing-deps:
  # Ensure that tests behave nicely while missing bwrap and ostree
-  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:28-master-46405991
+  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:29-master-47052095
  <<: *tests
  script:
@@ -128,7 +122,7 @@ tests-fedora-update-deps:
  # Check if the tests pass after updating requirements to their latest
  # allowed version.
  allow_failure: true
-  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:28-master-46405991
+  image: registry.gitlab.com/buildstream/buildstream-docker-images/testsuite-fedora:29-master-47052095
  <<: *tests
  script:
@@ -166,6 +160,7 @@ tests-wsl:
  script:
  - "${TEST_COMMAND}"
+  when: manual
 # Automatically build documentation for every commit, we want to know
 # if building documentation fails even if we're not deploying it.
@@ -288,6 +283,7 @@ coverage:
  dependencies:
  - tests-debian-9
  - tests-fedora-28
+  - tests-fedora-29
  - tests-fedora-missing-deps
  - tests-ubuntu-18.04
  - tests-unix

--- a/buildstream/_cas/cascache.py
+++ b/buildstream/_cas/cascache.py
@@ -35,6 +35,8 @@ from .._exceptions import CASCacheError
 from .casremote import BlobNotFound, _CASBatchRead, _CASBatchUpdate
+_BUFFER_SIZE = 65536
 # A CASCache manages a CAS repository as specified in the Remote Execution API.
 #
@@ -43,9 +45,10 @@ from .casremote import BlobNotFound, _CASBatchRead, _CASBatchUpdate
 #
 class CASCache():
-    def __init__(self, path):
+    def __init__(self, path, *, disable_exec=False):
        self.casdir = os.path.join(path, 'cas')
        self.tmpdir = os.path.join(path, 'tmp')
+        self._disable_exec = disable_exec
        os.makedirs(os.path.join(self.casdir, 'refs', 'heads'), exist_ok=True)
        os.makedirs(os.path.join(self.casdir, 'objects'), exist_ok=True)
        os.makedirs(self.tmpdir, exist_ok=True)
@@ -340,8 +343,12 @@ class CASCache():
    # Returns:
    #     (str): The path of the object
    #
-    def objpath(self, digest):
+    def objpath(self, digest, *, is_exec=False):
-        return os.path.join(self.casdir, 'objects', digest.hash[:2], digest.hash[2:])
+        if is_exec and not self._disable_exec:
+            filename = '{}.exec'.format(digest.hash[2:])
+        else:
+            filename = digest.hash[2:]
+        return os.path.join(self.casdir, 'objects', digest.hash[:2], filename)
    # add_object():
    #
@@ -358,7 +365,7 @@ class CASCache():
    #
    # Either `path` or `buffer` must be passed, but not both.
    #
-    def add_object(self, *, digest=None, path=None, buffer=None, link_directly=False):
+    def add_object(self, *, digest=None, path=None, buffer=None, link_directly=False, is_exec=False):
        # Exactly one of the two parameters has to be specified
        assert (path is None) != (buffer is None)
@@ -371,16 +378,13 @@ class CASCache():
            with contextlib.ExitStack() as stack:
                if path is not None and link_directly:
                    tmp = stack.enter_context(open(path, 'rb'))
-                    for chunk in iter(lambda: tmp.read(4096), b""):
+                    for chunk in iter(lambda: tmp.read(_BUFFER_SIZE), b""):
                        h.update(chunk)
                else:
-                    tmp = stack.enter_context(utils._tempnamedfile(dir=self.tmpdir))
+                    tmp = stack.enter_context(self._temporary_object(is_exec=is_exec))
-                    # Set mode bits to 0644
-                    os.chmod(tmp.name, stat.S_IRUSR | stat.S_IWUSR | stat.S_IRGRP | stat.S_IROTH)
                    if path:
                        with open(path, 'rb') as f:
-                            for chunk in iter(lambda: f.read(4096), b""):
+                            for chunk in iter(lambda: f.read(_BUFFER_SIZE), b""):
                                h.update(chunk)
                                tmp.write(chunk)
                    else:
@@ -393,7 +397,7 @@ class CASCache():
                digest.size_bytes = os.fstat(tmp.fileno()).st_size
                # Place file at final location
-                objpath = self.objpath(digest)
+                objpath = self.objpath(digest, is_exec=is_exec)
                os.makedirs(os.path.dirname(objpath), exist_ok=True)
                os.link(tmp.name, objpath)
@@ -602,11 +606,7 @@ class CASCache():
        for filenode in directory.files:
            # regular file, create hardlink
            fullpath = os.path.join(dest, filenode.name)
-            os.link(self.objpath(filenode.digest), fullpath)
+            os.link(self.objpath(filenode.digest, is_exec=filenode.is_executable), fullpath)
-            if filenode.is_executable:
-                os.chmod(fullpath, stat.S_IRUSR | stat.S_IWUSR | stat.S_IXUSR |
-                         stat.S_IRGRP | stat.S_IXGRP | stat.S_IROTH | stat.S_IXOTH)
        for dirnode in directory.directories:
            # Don't try to checkout a dangling ref
@@ -698,8 +698,8 @@ class CASCache():
            elif stat.S_ISREG(mode):
                filenode = directory.files.add()
                filenode.name = name
-                self.add_object(path=full_path, digest=filenode.digest)
                filenode.is_executable = (mode & stat.S_IXUSR) == stat.S_IXUSR
+                self.add_object(path=full_path, digest=filenode.digest, is_exec=filenode.is_executable)
            elif stat.S_ISLNK(mode):
                symlinknode = directory.symlinks.add()
                symlinknode.name = name
@@ -798,7 +798,7 @@ class CASCache():
        for filenode in directory.files:
            if update_mtime:
-                os.utime(self.objpath(filenode.digest))
+                os.utime(self.objpath(filenode.digest, is_exec=filenode.is_executable))
            reachable.add(filenode.digest.hash)
        for dirnode in directory.directories:
@@ -809,7 +809,7 @@ class CASCache():
        d = remote_execution_pb2.Digest()
        d.hash = directory_digest.hash
        d.size_bytes = directory_digest.size_bytes
-        yield d
+        yield False, d
        directory = remote_execution_pb2.Directory()
@@ -820,11 +820,26 @@ class CASCache():
            d = remote_execution_pb2.Digest()
            d.hash = filenode.digest.hash
            d.size_bytes = filenode.digest.size_bytes
-            yield d
+            yield filenode.is_executable, d
        for dirnode in directory.directories:
            yield from self._required_blobs(dirnode.digest)
+    # _temporary_object():
+    #
+    # Returns:
+    #     (file): A file object to a named temporary file.
+    #
+    # Create a named temporary file with 0o0644 access rights.
+    @contextlib.contextmanager
+    def _temporary_object(self, *, is_exec=False):
+        with utils._tempnamedfile(dir=self.tmpdir) as f:
+            access = stat.S_IRUSR | stat.S_IWUSR | stat.S_IRGRP | stat.S_IROTH
+            if is_exec and not self._disable_exec:
+                access |= stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH
+            os.chmod(f.name, access)
+            yield f
    # _ensure_blob():
    #
    # Fetch and add blob if it's not already local.
@@ -836,32 +851,34 @@ class CASCache():
    # Returns:
    #     (str): The path of the object
    #
-    def _ensure_blob(self, remote, digest):
+    def _ensure_blob(self, remote, digest, is_exec=False):
-        objpath = self.objpath(digest)
+        objpath = self.objpath(digest, is_exec=is_exec)
        if os.path.exists(objpath):
            # already in local repository
            return objpath
-        with utils._tempnamedfile(dir=self.tmpdir) as f:
+        with self._temporary_object(is_exec=is_exec) as f:
            remote._fetch_blob(digest, f)
-            added_digest = self.add_object(path=f.name, link_directly=True)
+            added_digest = self.add_object(path=f.name, link_directly=True, is_exec=is_exec)
            assert added_digest.hash == digest.hash
        return objpath
-    def _batch_download_complete(self, batch):
+    def _batch_download_complete(self, batch, executable):
        for digest, data in batch.send():
-            with utils._tempnamedfile(dir=self.tmpdir) as f:
+            for is_exec in executable[digest.hash]:
-                f.write(data)
+                with self._temporary_object(is_exec=is_exec) as f:
-                f.flush()
+                    f.write(data)
+                    f.flush()
-                added_digest = self.add_object(path=f.name, link_directly=True)
+                    added_digest = self.add_object(path=f.name, link_directly=True, is_exec=is_exec)
-                assert added_digest.hash == digest.hash
+                    assert added_digest.hash == digest.hash
+            del executable[digest.hash]
    # Helper function for _fetch_directory().
-    def _fetch_directory_batch(self, remote, batch, fetch_queue, fetch_next_queue):
+    def _fetch_directory_batch(self, remote, batch, fetch_queue, fetch_next_queue, executable):
-        self._batch_download_complete(batch)
+        self._batch_download_complete(batch, executable)
        # All previously scheduled directories are now locally available,
        # move them to the processing queue.
@@ -870,8 +887,9 @@ class CASCache():
        return _CASBatchRead(remote)
    # Helper function for _fetch_directory().
-    def _fetch_directory_node(self, remote, digest, batch, fetch_queue, fetch_next_queue, *, recursive=False):
+    def _fetch_directory_node(self, remote, digest, batch, fetch_queue, fetch_next_queue, executable,
-        in_local_cache = os.path.exists(self.objpath(digest))
+                              *, recursive=False, is_exec=False):
+        in_local_cache = os.path.exists(self.objpath(digest, is_exec=is_exec))
        if in_local_cache:
            # Skip download, already in local cache.
@@ -879,14 +897,17 @@ class CASCache():
        elif (digest.size_bytes >= remote.max_batch_total_size_bytes or
              not remote.batch_read_supported):
            # Too large for batch request, download in independent request.
-            self._ensure_blob(remote, digest)
+            self._ensure_blob(remote, digest, is_exec=is_exec)
            in_local_cache = True
        else:
            if not batch.add(digest):
                # Not enough space left in batch request.
                # Complete pending batch first.
-                batch = self._fetch_directory_batch(remote, batch, fetch_queue, fetch_next_queue)
+                batch = self._fetch_directory_batch(remote, batch, fetch_queue, fetch_next_queue, executable)
                batch.add(digest)
+            if digest.hash not in executable:
+                executable[digest.hash] = set()
+            executable[digest.hash].add(is_exec)
        if recursive:
            if in_local_cache:
@@ -915,13 +936,14 @@ class CASCache():
    def _fetch_directory(self, remote, dir_digest, *, excluded_subdirs=None):
        fetch_queue = [dir_digest]
        fetch_next_queue = []
+        executable = {}
        batch = _CASBatchRead(remote)
        if not excluded_subdirs:
            excluded_subdirs = []
        while len(fetch_queue) + len(fetch_next_queue) > 0:
            if not fetch_queue:
-                batch = self._fetch_directory_batch(remote, batch, fetch_queue, fetch_next_queue)
+                batch = self._fetch_directory_batch(remote, batch, fetch_queue, fetch_next_queue, executable)
            dir_digest = fetch_queue.pop(0)
@@ -934,14 +956,18 @@ class CASCache():
            for dirnode in directory.directories:
                if dirnode.name not in excluded_subdirs:
                    batch = self._fetch_directory_node(remote, dirnode.digest, batch,
-                                                       fetch_queue, fetch_next_queue, recursive=True)
+                                                       fetch_queue, fetch_next_queue,
+                                                       executable,
+                                                       recursive=True)
            for filenode in directory.files:
                batch = self._fetch_directory_node(remote, filenode.digest, batch,
-                                                   fetch_queue, fetch_next_queue)
+                                                   fetch_queue, fetch_next_queue,
+                                                   executable,
+                                                   is_exec=filenode.is_executable)
        # Fetch final batch
-        self._fetch_directory_batch(remote, batch, fetch_queue, fetch_next_queue)
+        self._fetch_directory_batch(remote, batch, fetch_queue, fetch_next_queue, executable)
    def _fetch_subdir(self, remote, tree, subdir):
        subdirdigest = self._get_subdir(tree, subdir)
@@ -960,7 +986,7 @@ class CASCache():
            tree.children.extend([tree.root])
            for directory in tree.children:
                for filenode in directory.files:
-                    self._ensure_blob(remote, filenode.digest)
+                    self._ensure_blob(remote, filenode.digest, is_exec=filenode.is_executable)
                # place directory blob only in final location when we've downloaded
                # all referenced blobs to avoid dangling references in the repository
@@ -973,22 +999,28 @@ class CASCache():
    def _send_directory(self, remote, digest, u_uid=uuid.uuid4()):
        required_blobs = self._required_blobs(digest)
+        executable = {}
        missing_blobs = dict()
        # Limit size of FindMissingBlobs request
        for required_blobs_group in _grouper(required_blobs, 512):
            request = remote_execution_pb2.FindMissingBlobsRequest(instance_name=remote.spec.instance_name)
-            for required_digest in required_blobs_group:
+            for is_exec, required_digest in required_blobs_group:
                d = request.blob_digests.add()
                d.hash = required_digest.hash
                d.size_bytes = required_digest.size_bytes
+                if required_digest.hash not in executable:
+                    executable[required_digest.hash] = set()
+                executable[required_digest.hash].add(is_exec)
            response = remote.cas.FindMissingBlobs(request)
            for missing_digest in response.missing_blob_digests:
                d = remote_execution_pb2.Digest()
                d.hash = missing_digest.hash
                d.size_bytes = missing_digest.size_bytes
-                missing_blobs[d.hash] = d
+                for is_exec in executable[missing_digest.hash]:
+                    missing_blobs[d.hash] = (is_exec, d)
        # Upload any blobs missing on the server
        self._send_blobs(remote, missing_blobs.values(), u_uid)
@@ -996,8 +1028,8 @@ class CASCache():
    def _send_blobs(self, remote, digests, u_uid=uuid.uuid4()):
        batch = _CASBatchUpdate(remote)
-        for digest in digests:
+        for is_exec, digest in digests:
-            with open(self.objpath(digest), 'rb') as f:
+            with open(self.objpath(digest, is_exec=is_exec), 'rb') as f:
                assert os.fstat(f.fileno()).st_size == digest.size_bytes
                if (digest.size_bytes >= remote.max_batch_total_size_bytes or

--- a/buildstream/_cas/casserver.py
+++ b/buildstream/_cas/casserver.py
@@ -61,7 +61,7 @@ class ArtifactTooLargeException(Exception):
 def create_server(repo, *, enable_push,
                  max_head_size=int(10e9),
                  min_head_size=int(2e9)):
-    cas = CASCache(os.path.abspath(repo))
+    cas = CASCache(os.path.abspath(repo), disable_exec=True)
    # Use max_workers default from Python 3.5+
    max_workers = (os.cpu_count() or 1) * 5

--- a/buildstream/_context.py
+++ b/buildstream/_context.py
@@ -361,14 +361,17 @@ class Context():
    #    (bool): Whether or not to use strict build plan
    #
    def get_strict(self):
+        if self._strict_build_plan is None:
+            # Either we're not overridden or we've never worked it out before
+            # so work out if we should be strict, and then cache the result
+            toplevel = self.get_toplevel_project()
+            overrides = self.get_overrides(toplevel.name)
+            self._strict_build_plan = _yaml.node_get(overrides, bool, 'strict', default_value=True)
        # If it was set by the CLI, it overrides any config
-        if self._strict_build_plan is not None:
+        # Ditto if we've already computed this, then we return the computed
-            return self._strict_build_plan
+        # value which we cache here too.
+        return self._strict_build_plan
-        toplevel = self.get_toplevel_project()
-        overrides = self.get_overrides(toplevel.name)
-        return _yaml.node_get(overrides, bool, 'strict', default_value=True)
    # get_cache_key():
    #

--- a/buildstream/_frontend/cli.py
+++ b/buildstream/_frontend/cli.py
@@ -440,6 +440,9 @@ def show(app, elements, deps, except_, order, format_):
        %{public}         Public domain data
        %{workspaced}     If the element is workspaced
        %{workspace-dirs} A list of workspace directories
+        %{deps}           A list of all dependencies
+        %{build-deps}     A list of build dependencies
+        %{runtime-deps}   A list of runtime dependencies
    The value of the %{symbol} without the leading '%' character is understood
    as a pythonic formatting string, so python formatting features apply,

--- a/buildstream/_frontend/widget.py
+++ b/buildstream/_frontend/widget.py
@@ -27,7 +27,7 @@ from ruamel import yaml
 import click
 from . import Profile
-from .. import Element, Consistency
+from .. import Element, Consistency, Scope
 from .. import _yaml
 from .. import __version__ as bst_version
 from .._exceptions import ImplError
@@ -435,6 +435,27 @@ class LogLine(Widget):
                    line = p.fmt_subst(
                        line, 'workspace-dirs', '')
+            # Dependencies
+            if "%{deps" in format_:
+                deps = [e.name for e in element.dependencies(Scope.ALL, recurse=False)]
+                line = p.fmt_subst(
+                    line, 'deps',
+                    yaml.safe_dump(deps, default_style=None).rstrip('\n'))
+            # Build Dependencies
+            if "%{build-deps" in format_:
+                build_deps = [e.name for e in element.dependencies(Scope.BUILD, recurse=False)]
+                line = p.fmt_subst(
+                    line, 'build-deps',
+                    yaml.safe_dump(build_deps, default_style=False).rstrip('\n'))
+            # Runtime Dependencies
+            if "%{runtime-deps" in format_:
+                runtime_deps = [e.name for e in element.dependencies(Scope.RUN, recurse=False)]
+                line = p.fmt_subst(
+                    line, 'runtime-deps',
+                    yaml.safe_dump(runtime_deps, default_style=False).rstrip('\n'))
            report += line + '\n'
        return report.rstrip('\n')

--- a/buildstream/_yaml.py
+++ b/buildstream/_yaml.py
@@ -365,8 +365,8 @@ _sentinel = object()
 #
 def node_get(node, expected_type, key, indices=None, *, default_value=_sentinel, allow_none=False):
    value = node.get(key, default_value)
-    provenance = node_get_provenance(node)
    if value is _sentinel:
+        provenance = node_get_provenance(node)
        raise LoadError(LoadErrorReason.INVALID_DATA,
                        "{}: Dictionary did not contain expected key '{}'".format(provenance, key))
@@ -914,6 +914,10 @@ RoundTripRepresenter.add_representer(SanitizedDict,
                                     SafeRepresenter.represent_dict)
+# Types we can short-circuit in node_sanitize for speed.
+__SANITIZE_SHORT_CIRCUIT_TYPES = (int, float, str, bool, tuple)
 # node_sanitize()
 #
 # Returnes an alphabetically ordered recursive copy
@@ -922,9 +926,21 @@ RoundTripRepresenter.add_representer(SanitizedDict,
 # Only dicts are ordered, list elements are left in order.
 #
 def node_sanitize(node):
+    # Short-circuit None which occurs ca. twice per element
+    if node is None:
+        return node
+    node_type = type(node)
+    # Next short-circuit integers, floats, strings, booleans, and tuples
+    if node_type in __SANITIZE_SHORT_CIRCUIT_TYPES:
+        return node
+    # Now short-circuit lists.  Note this is only for the raw list
+    # type, CommentedSeq and others get caught later.
+    elif node_type is list:
+        return [node_sanitize(elt) for elt in node]
-    if isinstance(node, collections.abc.Mapping):
+    # Finally ChainMap and dict, and other Mappings need special handling
+    if node_type in (dict, ChainMap) or isinstance(node, collections.Mapping):
        result = SanitizedDict()
        key_list = [key for key, _ in node_items(node)]
@@ -932,10 +948,12 @@ def node_sanitize(node):
            result[key] = node_sanitize(node[key])
        return result
+    # Catch the case of CommentedSeq and friends.  This is more rare and so
+    # we keep complexity down by still using isinstance here.
    elif isinstance(node, list):
        return [node_sanitize(elt) for elt in node]
+    # Everything else (such as commented scalars) just gets returned as-is.
    return node
@@ -1064,15 +1082,52 @@ class ChainMap(collections.ChainMap):
            return default
+# Node copying
+#
+# Unfortunately we copy nodes a *lot* and `isinstance()` is super-slow when
+# things from collections.abc get involved.  The result is the following
+# intricate but substantially faster group of tuples and the use of `in`.
+#
+# If any of the {node,list}_{chain_,}_copy routines raise a ValueError
+# then it's likely additional types need adding to these tuples.
+# When chaining a copy, these types are skipped since the ChainMap will
+# retrieve them from the source node when needed.  Other copiers might copy
+# them, so we call them __QUICK_TYPES.
+__QUICK_TYPES = (str, bool,
+                 yaml.scalarstring.PreservedScalarString,
+                 yaml.scalarstring.SingleQuotedScalarString,
+                 yaml.scalarstring.DoubleQuotedScalarString)
+# These types have to be iterated like a dictionary
+__DICT_TYPES = (dict, ChainMap, yaml.comments.CommentedMap)
+# These types have to be iterated like a list
+__LIST_TYPES = (list, yaml.comments.CommentedSeq)
+# These are the provenance types, which have to be cloned rather than any other
+# copying tactic.
+__PROVENANCE_TYPES = (Provenance, DictProvenance, MemberProvenance, ElementProvenance)
+# These are the directives used to compose lists, we need this because it's
+# slightly faster during the node_final_assertions checks
+__NODE_ASSERT_COMPOSITION_DIRECTIVES = ('(>)', '(<)', '(=)')
 def node_chain_copy(source):
    copy = ChainMap({}, source)
    for key, value in source.items():
-        if isinstance(value, collections.abc.Mapping):
+        value_type = type(value)
+        if value_type in __DICT_TYPES:
            copy[key] = node_chain_copy(value)
-        elif isinstance(value, list):
+        elif value_type in __LIST_TYPES:
            copy[key] = list_chain_copy(value)
-        elif isinstance(value, Provenance):
+        elif value_type in __PROVENANCE_TYPES:
            copy[key] = value.clone()
+        elif value_type in __QUICK_TYPES:
+            pass  # No need to copy these, the chainmap deals with it
+        else:
+            raise ValueError("Unable to be quick about node_chain_copy of {}".format(value_type))
    return copy
@@ -1080,14 +1135,17 @@ def node_chain_copy(source):
 def list_chain_copy(source):
    copy = []
    for item in source:
-        if isinstance(item, collections.abc.Mapping):
+        item_type = type(item)
+        if item_type in __DICT_TYPES:
            copy.append(node_chain_copy(item))
-        elif isinstance(item, list):
+        elif item_type in __LIST_TYPES:
            copy.append(list_chain_copy(item))
-        elif isinstance(item, Provenance):
+        elif item_type in __PROVENANCE_TYPES:
            copy.append(item.clone())
-        else:
+        elif item_type in __QUICK_TYPES:
            copy.append(item)
+        else:  # Fallback
+            raise ValueError("Unable to be quick about list_chain_copy of {}".format(item_type))
    return copy
@@ -1095,14 +1153,17 @@ def list_chain_copy(source):
 def node_copy(source):
    copy = {}
    for key, value in source.items():
-        if isinstance(value, collections.abc.Mapping):
+        value_type = type(value)
+        if value_type in __DICT_TYPES:
            copy[key] = node_copy(value)
-        elif isinstance(value, list):
+        elif value_type in __LIST_TYPES:
            copy[key] = list_copy(value)
-        elif isinstance(value, Provenance):
+        elif value_type in __PROVENANCE_TYPES:
            copy[key] = value.clone()
-        else:
+        elif value_type in __QUICK_TYPES:
            copy[key] = value
+        else:
+            raise ValueError("Unable to be quick about node_copy of {}".format(value_type))
    ensure_provenance(copy)
@@ -1112,14 +1173,17 @@ def node_copy(source):
 def list_copy(source):
    copy = []
    for item in source:
-        if isinstance(item, collections.abc.Mapping):
+        item_type = type(item)
+        if item_type in __DICT_TYPES:
            copy.append(node_copy(item))
-        elif isinstance(item, list):
+        elif item_type in __LIST_TYPES:
            copy.append(list_copy(item))
-        elif isinstance(item, Provenance):
+        elif item_type in __PROVENANCE_TYPES:
            copy.append(item.clone())
-        else:
+        elif item_type in __QUICK_TYPES:
            copy.append(item)
+        else:
+            raise ValueError("Unable to be quick about list_copy of {}".format(item_type))
    return copy
@@ -1142,22 +1206,26 @@ def node_final_assertions(node):
        # indicates that the user intended to override a list which
        # never existed in the underlying data
        #
-        if key in ['(>)', '(<)', '(=)']:
+        if key in __NODE_ASSERT_COMPOSITION_DIRECTIVES:
            provenance = node_get_provenance(node, key)
            raise LoadError(LoadErrorReason.TRAILING_LIST_DIRECTIVE,
                            "{}: Attempt to override non-existing list".format(provenance))
-        if isinstance(value, collections.abc.Mapping):
+        value_type = type(value)
+        if value_type in __DICT_TYPES:
            node_final_assertions(value)
-        elif isinstance(value, list):
+        elif value_type in __LIST_TYPES:
            list_final_assertions(value)
 def list_final_assertions(values):
    for value in values:
-        if isinstance(value, collections.abc.Mapping):
+        value_type = type(value)
+        if value_type in __DICT_TYPES:
            node_final_assertions(value)
-        elif isinstance(value, list):
+        elif value_type in __LIST_TYPES:
            list_final_assertions(value)

--- a/buildstream/utils.py
+++ b/buildstream/utils.py
@@ -235,7 +235,7 @@ def sha256sum(filename):
    try:
        h = hashlib.sha256()
        with open(filename, "rb") as f:
-            for chunk in iter(lambda: f.read(4096), b""):
+            for chunk in iter(lambda: f.read(65536), b""):
                h.update(chunk)
    except OSError as e:

--- a/man/bst-artifact-checkout.1
+++ b/man/bst-artifact-checkout.1
-.TH "BST ARTIFACT CHECKOUT" "1" "24-Jan-2019" "" "bst artifact checkout Manual"
+.TH "BST ARTIFACT CHECKOUT" "1" "12-Feb-2019" "" "bst artifact checkout Manual"
 .SH NAME
 bst\-artifact\-checkout \- Checkout contents of an artifact
 .SH SYNOPSIS

--- a/man/bst-artifact-log.1
+++ b/man/bst-artifact-log.1
-.TH "BST ARTIFACT LOG" "1" "24-Jan-2019" "" "bst artifact log Manual"
+.TH "BST ARTIFACT LOG" "1" "12-Feb-2019" "" "bst artifact log Manual"
 .SH NAME
 bst\-artifact\-log \- Show logs of an artifact
 .SH SYNOPSIS

--- a/man/bst-artifact-pull.1
+++ b/man/bst-artifact-pull.1
-.TH "BST ARTIFACT PULL" "1" "24-Jan-2019" "" "bst artifact pull Manual"
+.TH "BST ARTIFACT PULL" "1" "12-Feb-2019" "" "bst artifact pull Manual"
 .SH NAME
 bst\-artifact\-pull \- Pull a built artifact
 .SH SYNOPSIS

--- a/man/bst-artifact-push.1
+++ b/man/bst-artifact-push.1
-.TH "BST ARTIFACT PUSH" "1" "24-Jan-2019" "" "bst artifact push Manual"
+.TH "BST ARTIFACT PUSH" "1" "12-Feb-2019" "" "bst artifact push Manual"
 .SH NAME
 bst\-artifact\-push \- Push a built artifact
 .SH SYNOPSIS

--- a/man/bst-artifact-server.1
+++ b/man/bst-artifact-server.1
-.TH "BST-ARTIFACT-SERVER" "1" "24-Jan-2019" "" "bst-artifact-server Manual"
+.TH "BST-ARTIFACT-SERVER" "1" "12-Feb-2019" "" "bst-artifact-server Manual"
 .SH NAME
 bst-artifact-server \- CAS Artifact Server
 .SH SYNOPSIS

--- a/man/bst-artifact.1
+++ b/man/bst-artifact.1
-.TH "BST ARTIFACT" "1" "24-Jan-2019" "" "bst artifact Manual"
+.TH "BST ARTIFACT" "1" "12-Feb-2019" "" "bst artifact Manual"
 .SH NAME
 bst\-artifact \- Manipulate cached artifacts
 .SH SYNOPSIS

--- a/man/bst-build.1
+++ b/man/bst-build.1
-.TH "BST BUILD" "1" "24-Jan-2019" "" "bst build Manual"
+.TH "BST BUILD" "1" "12-Feb-2019" "" "bst build Manual"
 .SH NAME
 bst\-build \- Build elements in a pipeline
 .SH SYNOPSIS
@@ -32,3 +32,6 @@ Allow tracking to cross junction boundaries
 .TP
 \fB\-\-track\-save\fP
 Deprecated: This is ignored
+.TP
+\fB\-r,\fP \-\-remote TEXT
+The URL of the remote cache (defaults to the first configured cache)
--- a/man/bst-help.1
+++ b/man/bst-help.1
-.TH "BST HELP" "1" "24-Jan-2019" "" "bst help Manual"
+.TH "BST HELP" "1" "12-Feb-2019" "" "bst help Manual"
 .SH NAME
 bst\-help \- Print usage information
 .SH SYNOPSIS

--- a/man/bst-init.1
+++ b/man/bst-init.1
-.TH "BST INIT" "1" "24-Jan-2019" "" "bst init Manual"
+.TH "BST INIT" "1" "12-Feb-2019" "" "bst init Manual"
 .SH NAME
 bst\-init \- Initialize a new BuildStream project
 .SH SYNOPSIS

--- a/man/bst-shell.1
+++ b/man/bst-shell.1
-.TH "BST SHELL" "1" "24-Jan-2019" "" "bst shell Manual"
+.TH "BST SHELL" "1" "12-Feb-2019" "" "bst shell Manual"
 .SH NAME
 bst\-shell \- Shell into an element's sandbox environment
 .SH SYNOPSIS

--- a/man/bst-show.1
+++ b/man/bst-show.1
-.TH "BST SHOW" "1" "24-Jan-2019" "" "bst show Manual"
+.TH "BST SHOW" "1" "12-Feb-2019" "" "bst show Manual"
 .SH NAME
 bst\-show \- Show elements in the pipeline
 .SH SYNOPSIS
@@ -43,6 +43,9 @@ the following symbols can be used in the format string:
    %{public}         Public domain data
    %{workspaced}     If the element is workspaced
    %{workspace-dirs} A list of workspace directories
+    %{deps}           A list of all dependencies
+    %{build-deps}     A list of build dependencies
+    %{runtime-deps}   A list of runtime dependencies
 .PP
 The value of the %{symbol} without the leading '%' character is understood
 as a pythonic formatting string, so python formatting features apply,

--- a/man/bst-source-checkout.1
+++ b/man/bst-source-checkout.1
-.TH "BST SOURCE CHECKOUT" "1" "24-Jan-2019" "" "bst source checkout Manual"
+.TH "BST SOURCE CHECKOUT" "1" "12-Feb-2019" "" "bst source checkout Manual"
 .SH NAME
 bst\-source\-checkout \- Checkout sources for an element
 .SH SYNOPSIS
No results found