Merge pull request #45 from nexB/extract-symlinks

Extract symlinks correctly

Author: pombredanne

CHANGELOG.rst

@@ -1,6 +1,24 @@
 Changelog
 =========
 
+v32.0.0 
+--------
+
+This is a minor release with bug fixes and an output change.
+
+- We no longer support Python 3.6, only 3.7 and up.
+
+- "utils.extract_tar" function now behaves correctly with links and return
+  either a list of error message strings (the previous default) but with updated
+  messages or a list of ExtractEvent to better track extraction errors and warnings.
+  The behavious is driven by the "as_events" argument.
+
+- In all places where extract is callable (Image, Layer) there is a new
+  "skip_symlinks" argument defaulting to True. If True, we skip symlinks and links.
+  The same applies with the "as_events" available in these places as these
+  functions now return a list (rather than nothing before).
+
+
 v31.1.0
 --------
 

azure-pipelines.yml

@@ -11,23 +11,23 @@ jobs:
       parameters:
           job_name: ubuntu18_cpython
           image_name: ubuntu-18.04
-          python_versions: ['3.6', '3.7', '3.8', '3.9', '3.10']
+          python_versions: ['3.7', '3.8', '3.9', '3.10']
           test_suites:
               all: venv/bin/pytest -n 2 -vvs
 
     - template: etc/ci/azure-posix.yml
       parameters:
           job_name: ubuntu20_cpython
           image_name: ubuntu-20.04
-          python_versions: ['3.6', '3.7', '3.8', '3.9', '3.10']
+          python_versions: ['3.7', '3.8', '3.9', '3.10']
           test_suites:
               all: venv/bin/pytest -n 2 -vvs
 
     - template: etc/ci/azure-posix.yml
       parameters:
           job_name: macos1015_cpython
           image_name: macos-10.15
-          python_versions: ['3.6', '3.7', '3.8', '3.9', '3.10']
+          python_versions: ['3.7', '3.8', '3.9', '3.10']
           test_suites:
               all: venv/bin/pytest -n 2 -vvs
 
@@ -43,7 +43,7 @@ jobs:
 #      parameters:
 #          job_name: win2019_cpython
 #          image_name: windows-2019
-#          python_versions: ['3.6', '3.7', '3.8', '3.9', '3.10']
+#          python_versions: ['3.7', '3.8', '3.9', '3.10']
 #          test_suites:
 #              all: venv\Scripts\pytest -n 2 -vvs
 #

requirements-dev.txt

@@ -1,24 +1,31 @@
-aboutcode-toolkit==7.0.1
-bleach==4.1.0
+aboutcode-toolkit==7.0.2
+black==22.6.0
+bleach==5.0.1
 build==0.7.0
 commonmark==0.9.1
-docutils==0.18.1
+docutils==0.19
 et-xmlfile==1.1.0
 execnet==1.9.0
 iniconfig==1.1.1
-jeepney==0.7.1
-keyring==23.4.1
-openpyxl==3.0.9
+isort==5.10.1
+jeepney==0.8.0
+keyring==23.7.0
+mypy-extensions==0.4.3
+openpyxl==3.0.10
+pathspec==0.9.0
 pep517==0.12.0
-pkginfo==1.8.2
+pkginfo==1.8.3
+platformdirs==2.5.2
 py==1.11.0
-pytest==7.0.1
+pytest==7.1.2
 pytest-forked==1.4.0
 pytest-xdist==2.5.0
-readme-renderer==34.0
+readme-renderer==35.0
 requests-toolbelt==0.9.1
-rfc3986==1.5.0
-rich==12.3.0
+rfc3986==2.0.0
+rich==12.5.1
 secretstorage==3.3.2
-tomli==1.2.3
-twine==3.8.0
+tomli==2.0.1
+tqdm==4.64.0
+twine==4.0.1
+typing_extensions==4.3.0

requirements.txt

@@ -2,78 +2,78 @@ attrs==21.4.0
 banal==1.0.6
 beautifulsoup4==4.11.1
 binaryornot==0.4.4
-boolean.py==3.8
-certifi==2021.10.8
-cffi==1.15.0
-chardet==4.0.0
-charset-normalizer==2.0.12
-click==8.0.4
-colorama==0.4.4
-commoncode==30.2.0
+boolean.py==4.0
+certifi==2022.6.15
+cffi==1.15.1
+chardet==5.0.0
+charset-normalizer==2.1.0
+click==8.1.3
+colorama==0.4.5
+commoncode==31.0.0b4
 construct==2.10.68
-cryptography==36.0.2
-debian-inspector==30.0.0
+cryptography==37.0.4
+debian-inspector==31.0.0b1
 dockerfile-parse==1.2.0
 dparse2==0.6.1
-extractcode==30.0.0
+extractcode==31.0.0
 extractcode-7z==16.5.210531
 extractcode-libarchive==3.5.1.210531
 fasteners==0.17.3
 fingerprints==1.0.3
-ftfy==6.0.3
+ftfy==6.1.1
 future==0.18.2
 gemfileparser==0.8.0
 html5lib==1.1
 idna==3.3
-importlib-metadata==4.8.3
+importlib-metadata==4.12.0
 inflection==0.5.1
 intbitset==3.0.1
 isodate==0.6.1
-jaraco.functools==3.4.0
+jaraco.functools==3.5.1
 javaproperties==0.8.1
-Jinja2==3.0.3
+Jinja2==3.1.2
 jsonstreams==0.6.0
-license-expression==21.6.14
-lxml==4.8.0
-MarkupSafe==2.0.1
+libfwsi-python==20220123
+license-expression==30.0.0
+lxml==4.9.1
+MarkupSafe==2.1.1
 more-itertools==8.13.0
 normality==2.3.3
 packagedcode-msitools==0.101.210706
-packageurl-python==0.9.9
+packageurl-python==0.10.0
 packaging==21.3
 parameter-expansion-patched==0.3.1
-patch==1.16
-pdfminer.six==20220506
-pefile==2021.9.3
+pdfminer.six==20220524
+pefile==2022.5.30
 pip-requirements-parser==31.2.0
 pkginfo2==30.0.0
 pluggy==1.0.0
-plugincode==21.1.21
+plugincode==31.0.0b1
 ply==3.11
 publicsuffix2==2.20191221
 pyahocorasick==2.0.0b1
 pycparser==2.21
 pygmars==0.7.0
 Pygments==2.12.0
 pymaven-patch==0.3.0
-pyparsing==3.0.8
+pyparsing==3.0.9
 pytz==2022.1
 PyYAML==6.0
-rdflib==5.0.0
-regipy==2.2.2
-requests==2.27.1
+rdflib==6.2.0
+regipy==3.0.2
+requests==2.28.1
 rpm-inspector-rpm==4.16.1.3.210404
 saneyaml==0.5.2
 six==1.16.0
-soupsieve==2.3.1
+soupsieve==2.3.2.post1
 spdx-tools==0.7.0a3
 text-unidecode==1.3
 toml==0.10.2
-typecode==21.6.1
+typecode==30.0.0
 typecode-libmagic==5.39.210531
-urllib3==1.26.9
+urllib3==1.26.11
 urlpy==0.5
 wcwidth==0.2.5
 webencodings==0.5.1
-xmltodict==0.12.0
-zipp==3.6.0
+xmltodict==0.13.0
+zipp==3.8.1

setup.cfg

@@ -54,7 +54,7 @@ zip_safe = false
 
 setup_requires = setuptools_scm[toml] >= 4
 
-python_requires = >=3.6.*
+python_requires = >=3.7.*
 
 install_requires =
     click >= 6.7, !=7.0, !=8.0.3

src/container_inspector/image.py

@@ -356,14 +356,23 @@ def bottom_layer(self):
         """
         return self.layers[0]
 
-    def extract_layers(self, extracted_location):
+    def extract_layers(self, extracted_location, as_events=False, skip_symlinks=True):
         """
         Extract all layer archives to the `extracted_location` directory.
         Each layer is extracted to its own directory named after its `layer_id`.
+        Skip symlinks and links if ``skip_symlinks`` is True.
+        Return a list of ExtractEvent if ``as_events`` is True or a list of message strings otherwise.
         """
+        all_events = []
         for layer in self.layers:
             exloc = os.path.join(extracted_location, layer.layer_id)
-            layer.extract(extracted_location=exloc)
+            events = layer.extract(
+                extracted_location=exloc,
+                skip_symlinks=skip_symlinks,
+                as_events=as_events,
+            )
+            all_events.extend(events)
+        return events
 
     def get_layers_resources(self, with_dir=False):
         """
@@ -450,41 +459,53 @@ def get_installed_packages(self, packages_getter):
                 yield purl, package, layer
 
     @staticmethod
-    def extract(archive_location, extracted_location, skip_symlinks=False):
+    def extract(archive_location, extracted_location, as_events=False, skip_symlinks=False):
         """
         Extract the image archive tarball at ``archive_location`` to
-        ``extracted_location``. Skip symlinks and links if ``skip_symlinks`` is True.
+        ``extracted_location``.
+        Skip symlinks and links if ``skip_symlinks`` is True.
+        Return a list of ExtractEvent if ``as_events`` is True or a list of message strings otherwise.
         """
-        utils.extract_tar(
+        return utils.extract_tar(
             location=archive_location,
             target_dir=extracted_location,
             skip_symlinks=skip_symlinks,
+            as_events=as_events,
         )
 
     @staticmethod
     def get_images_from_tarball(
         archive_location,
         extracted_location,
         verify=True,
+        skip_symlinks=False,
     ):
         """
-        Return a list of Images found in the tarball at `archive_location` that
-        will be extracted to `extracted_location`. The tarball must be in the
+        Return a list of Images found in the tarball at ``archive_location`` that
+        will be extracted to ``extracted_location``. The tarball must be in the
         format of a "docker save" command tarball.
 
-        If `verify` is True, perform extra checks on the config data and layers
+        If ``verify`` is True, perform extra checks on the config data and layers
         checksums.
+        Skip symlinks and links if ``skip_symlinks`` is True.
+        Ignore the extract events from extraction.
         """
         if TRACE:
             logger.debug(
-                f'get_images_from_tarball: {archive_location} , '
+                f'get_images_from_tarball: {archive_location} '
                 f'extracting to: {extracted_location}'
             )
 
-        Image.extract(
+        # TODO: do not ignore extract events
+        _events = Image.extract(
             archive_location=archive_location,
             extracted_location=extracted_location,
+            skip_symlinks=skip_symlinks,
         )
+        if TRACE:
+            logger.debug(f'get_images_from_tarball: events')
+            for e in _events:
+                logger.debug(str(e))
 
         return Image.get_images_from_dir(
             extracted_location=extracted_location,
@@ -1071,16 +1092,19 @@ def __attrs_post_init__(self, *args, **kwargs):
         if not self.size:
             self.size = os.path.getsize(self.archive_location)
 
-    def extract(self, extracted_location, skip_symlinks=True):
+    def extract(self, extracted_location, as_events=False, skip_symlinks=False):
         """
         Extract this layer archive in the `extracted_location` directory and set
         this Layer ``extracted_location`` attribute to ``extracted_location``.
+        Skip symlinks and links if ``skip_symlinks`` is True.
+        Return a list of ExtractEvent if ``as_events`` is True or a list of message strings otherwise.
         """
         self.extracted_location = extracted_location
-        utils.extract_tar(
+        return utils.extract_tar(
             location=self.archive_location,
             target_dir=extracted_location,
             skip_symlinks=skip_symlinks,
+            as_events=as_events,
         )
 
     def get_resources(self, with_dir=False, walker=os.walk):

src/container_inspector/rootfs.py

@@ -31,14 +31,16 @@ class InconsistentLayersError(Exception):
     pass
 
 
-def rebuild_rootfs(img, target_dir):
+def rebuild_rootfs(img, target_dir, skip_symlinks=True):
     """
     Extract and merge or "squash" all layers of the `image` Image in a single
     rootfs in `target_dir`. Extraction is done in sequence from the bottom (root
     or initial) layer to the top (or latest) layer and the "whiteouts"
     unionfs/overlayfs procedure is applied at each step as per the OCI spec:
     https://github.com/opencontainers/image-spec/blob/master/layer.md#whiteouts
 
+    Skip symlinks and links if ``skip_symlinks`` is True.
+
     Return a list of deleted "whiteout" files.
     Raise an Exception on errrors.
 
@@ -73,8 +75,15 @@ def rebuild_rootfs(img, target_dir):
         # 1. extract a layer to temp.
         # Note that we are not preserving any special file and any file permission
         extracted_loc = tempfile.mkdtemp('container_inspector-docker')
-        layer.extract(extracted_location=extracted_loc)
-        if TRACE: logger.debug(f'  Extracted layer to: {extracted_loc}')
+        # TODO: do not ignore extract events
+        _events = layer.extract(
+            extracted_location=extracted_loc,
+            skip_symlinks=skip_symlinks,
+        )
+        if TRACE:
+            logger.debug(f'  Extracted layer to: {extracted_loc} with skip_symlinks: {skip_symlinks}')
+            for ev in _events:
+                logger.debug(f'  {ev}')
 
         # 2. find whiteouts in that layer.
         whiteouts = list(find_whiteouts(extracted_loc))