plugin: bake API to get and validate a project path

A project path is a path relative to a project directory.

A project path can not also refer to the parent directory in the first
path component, or point to symbolic links, fifos, sockets and
block/character devices.

4 files changed, 147 insertions, 4 deletions

diff --git a/.pylintrc b/.pylintrc
index 93f9eeadf..153a9ef3e 100644
--- a/.pylintrc
+++ b/.pylintrc
@@ -184,7 +184,7 @@ ignore-on-opaque-inference=yes
 # List of class names for which member attributes should not be checked (useful
 # for classes with dynamically set attributes). This supports the use of
 # qualified names.
-ignored-classes=optparse.Values,thread._local,_thread._local,contextlib.closing,gi.repository.GLib.GError
+ignored-classes=optparse.Values,thread._local,_thread._local,contextlib.closing,gi.repository.GLib.GError,pathlib.PurePath
 
 # List of module names for which member attributes should not be checked
 # (useful for modules/projects where namespaces are manipulated during runtime
diff --git a/buildstream/_exceptions.py b/buildstream/_exceptions.py
index e55d942fd..5187357c5 100644
--- a/buildstream/_exceptions.py
+++ b/buildstream/_exceptions.py
@@ -1,5 +1,5 @@
 #
-#  Copyright (C) 2016 Codethink Limited
+#  Copyright (C) 2018 Codethink Limited
 #
 #  This program is free software; you can redistribute it and/or
 #  modify it under the terms of the GNU Lesser General Public
@@ -16,6 +16,7 @@
 #
 #  Authors:
 #        Tristan Van Berkom <tristan.vanberkom@codethink.co.uk>
+#        Tiago Gomes <tiago.gomes@codethink.co.uk>
 
 from enum import Enum
 
@@ -206,6 +207,13 @@ class LoadErrorReason(Enum):
     # Try to load a directory not a yaml file
     LOADING_DIRECTORY = 18
 
+    # A project path leads outside of the project directory
+    PROJ_PATH_INVALID = 19
+
+    # A project path points to a file of the not right kind (e.g. a
+    # socket)
+    PROJ_PATH_INVALID_KIND = 20
+
 
 # LoadError
 #
diff --git a/buildstream/_yaml.py b/buildstream/_yaml.py
index 0e090e2e7..33ee444aa 100644
--- a/buildstream/_yaml.py
+++ b/buildstream/_yaml.py
@@ -1,5 +1,5 @@
 #
-#  Copyright (C) 2016 Codethink Limited
+#  Copyright (C) 2018 Codethink Limited
 #
 #  This program is free software; you can redistribute it and/or
 #  modify it under the terms of the GNU Lesser General Public
@@ -22,6 +22,7 @@ import collections
 import string
 from copy import deepcopy
 from contextlib import ExitStack
+from pathlib import Path
 
 from ruamel import yaml
 from ruamel.yaml.representer import SafeRepresenter, RoundTripRepresenter
@@ -392,6 +393,95 @@ def node_get(node, expected_type, key, indices=None, default_value=_get_sentinel
     return value
 
 
+# node_get_project_path()
+#
+# Fetches a project path from a dictionary node and validates it
+#
+# Paths are asserted to never lead to a directory outside of the project
+# directory. In addition, paths can not point to symbolic links, fifos,
+# sockets and block/character devices.
+#
+# The `check_is_file` and `check_is_dir` parameters can be used to
+# perform additional validations on the path. Note that an exception
+# will always be raised if both parameters are set to ``True``.
+#
+# Args:
+#    node (dict): A dictionary loaded from YAML
+#    key (str): The key whose value contains a path to validate
+#    project_dir (str): The project directory
+#    check_is_file (bool): If ``True`` an error will also be raised
+#                          if path does not point to a regular file.
+#                          Defaults to ``False``
+#    check_is_dir (bool): If ``True`` an error will be also raised
+#                         if path does not point to a directory.
+#                         Defaults to ``False``
+# Returns:
+#    (str): The project path
+#
+# Raises:
+#    (LoadError): In case that the project path is not valid or does not
+#                 exist
+#
+def node_get_project_path(node, key, project_dir, *,
+                          check_is_file=False, check_is_dir=False):
+    path_str = node_get(node, str, key)
+    path = Path(path_str)
+    project_dir_path = Path(project_dir)
+
+    provenance = node_get_provenance(node, key=key)
+
+    if (project_dir_path / path).is_symlink():
+        raise LoadError(LoadErrorReason.PROJ_PATH_INVALID_KIND,
+                        "{}: Specified path '{}' must not point to "
+                        "symbolic links "
+                        .format(provenance, path_str))
+
+    if path.parts and path.parts[0] == '..':
+        raise LoadError(LoadErrorReason.PROJ_PATH_INVALID,
+                        "{}: Specified path '{}' first component must "
+                        "not be '..'"
+                        .format(provenance, path_str))
+
+    try:
+        if sys.version_info[0] == 3 and sys.version_info[1] < 6:
+            full_resolved_path = (project_dir_path / path).resolve()
+        else:
+            full_resolved_path = (project_dir_path / path).resolve(strict=True)
+    except FileNotFoundError:
+        raise LoadError(LoadErrorReason.MISSING_FILE,
+                        "{}: Specified path '{}' does not exist"
+                        .format(provenance, path_str))
+
+    is_inside = project_dir_path in full_resolved_path.parents or (
+        full_resolved_path == project_dir_path)
+
+    if path.is_absolute() or not is_inside:
+        raise LoadError(LoadErrorReason.PROJ_PATH_INVALID,
+                        "{}: Specified path '{}' must not lead outside of the "
+                        "project directory"
+                        .format(provenance, path_str))
+
+    if full_resolved_path.is_socket() or (
+            full_resolved_path.is_fifo() or
+            full_resolved_path.is_block_device()):
+        raise LoadError(LoadErrorReason.PROJ_PATH_INVALID_KIND,
+                        "{}: Specified path '{}' points to an unsupported "
+                        "file kind"
+                        .format(provenance, path_str))
+
+    if check_is_file and not full_resolved_path.is_file():
+        raise LoadError(LoadErrorReason.PROJ_PATH_INVALID_KIND,
+                        "{}: Specified path '{}' is not a regular file"
+                        .format(provenance, path_str))
+
+    if check_is_dir and not full_resolved_path.is_dir():
+        raise LoadError(LoadErrorReason.PROJ_PATH_INVALID_KIND,
+                        "{}: Specified path '{}' is not a directory"
+                        .format(provenance, path_str))
+
+    return path_str
+
+
 # node_items()
 #
 # A convenience generator for iterating over loaded key/value
diff --git a/buildstream/plugin.py b/buildstream/plugin.py
index 155a9500e..836b60834 100644
--- a/buildstream/plugin.py
+++ b/buildstream/plugin.py
@@ -1,5 +1,5 @@
 #
-#  Copyright (C) 2017 Codethink Limited
+#  Copyright (C) 2018 Codethink Limited
 #
 #  This program is free software; you can redistribute it and/or
 #  modify it under the terms of the GNU Lesser General Public
@@ -335,6 +335,51 @@ class Plugin():
         """
         return _yaml.node_get(node, expected_type, member_name, default_value=default)
 
+    def node_get_project_path(self, node, key, *,
+                              check_is_file=False, check_is_dir=False):
+        """Fetches a project path from a dictionary node and validates it
+
+        Paths are asserted to never lead to a directory outside of the
+        project directory. In addition, paths can not point to symbolic
+        links, fifos, sockets and block/character devices.
+
+        The `check_is_file` and `check_is_dir` parameters can be used to
+        perform additional validations on the path. Note that an
+        exception will always be raised if both parameters are set to
+        ``True``.
+
+        Args:
+           node (dict): A dictionary loaded from YAML
+           key (str): The key whose value contains a path to validate
+           check_is_file (bool): If ``True`` an error will also be raised
+                                 if path does not point to a regular file.
+                                 Defaults to ``False``
+           check_is_dir (bool): If ``True`` an error will also be raised
+                                if path does not point to a directory.
+                                Defaults to ``False``
+
+        Returns:
+           (str): The project path
+
+        Raises:
+           :class:`.LoadError`: In the case that the project path is not
+                                valid or does not exist
+
+        *Since: 1.2*
+
+        **Example:**
+
+        .. code:: python
+
+          path = self.node_get_project_path(node, 'path')
+
+        """
+
+        return _yaml.node_get_project_path(node, key,
+                                           self.__project.directory,
+                                           check_is_file=check_is_file,
+                                           check_is_dir=check_is_dir)
+
     def node_validate(self, node, valid_keys):
         """This should be used in :func:`~buildstream.plugin.Plugin.configure`
         implementations to assert that users have only entered