Changes to improve AMCache.hve format support #2790 (#3701)

ACKNOWLEDGEMENTS

@@ -135,6 +135,7 @@ Copied with permission granted by Jerome Marty.
 
 Copied with permission granted by Antoine Brodin.
 * PartitionsEx-WebCacheV01.dat
+* win10-Amcache.hve
 
 Copied with permission granted by Rob Lee.
 Copyright SANS Institute - Digital Forensics and Incident Response.

plaso/lib/definitions.py

@@ -137,6 +137,7 @@
 TIME_DESCRIPTION_LAST_SHUTDOWN = 'Last Shutdown Time'
 TIME_DESCRIPTION_LAST_USED = 'Last Used Time'
 TIME_DESCRIPTION_LAST_VISITED = 'Last Visited Time'
+TIME_DESCRIPTION_LINK_TIME = 'Link Time'
 TIME_DESCRIPTION_MODIFICATION = 'Content Modification Time'
 TIME_DESCRIPTION_PURCHASED = 'Purchased'
 TIME_DESCRIPTION_RECORDED = 'Event Recorded'

plaso/parsers/winreg_plugins/amcache.py

@@ -1,8 +1,11 @@
 # -*- coding: utf-8 -*-
 """Windows Registry plugin to parse the AMCache.hve Root key."""
 
+import re
+
 from dfdatetime import filetime as dfdatetime_filetime
 from dfdatetime import posix_time as dfdatetime_posix_time
+from dfdatetime import time_elements as dfdatetime_time_elements
 
 from dfwinreg import errors as dfwinreg_errors
 
@@ -95,6 +98,14 @@ class AMCachePlugin(interface.WindowsRegistryPlugin):
       interface.WindowsRegistryKeyPathFilter('\\Root')])
 
   # Contains: {value name: attribute name}
+  _APPLICATION_SUB_KEY_VALUES = {
+      'LowerCaseLongPath': 'full_path',
+      'ProductName': 'product_name',
+      'ProductVersion': 'file_version',
+      'ProgramId': 'program_identifier',
+      'Publisher': 'company_name',
+      'Size': 'file_size'}
+
   _FILE_REFERENCE_KEY_VALUES = {
       '0': 'product_name',
       '1': 'company_name',
@@ -106,7 +117,7 @@ class AMCachePlugin(interface.WindowsRegistryPlugin):
       '100': 'program_identifier',
       '101': 'sha1'}
 
-  _AMCACHE_COMPILATION_TIME = 'f'
+  _AMCACHE_LINK_TIME = 'f'
   _AMCACHE_FILE_MODIFICATION_TIME = '11'
   _AMCACHE_FILE_CREATION_TIME = '12'
   _AMCACHE_ENTRY_WRITE_TIME = '17'
@@ -115,6 +126,13 @@ class AMCachePlugin(interface.WindowsRegistryPlugin):
 
   _AMCACHE_P_FILES = 'Files'
 
+  # Date and time string formatted as: "MM/DD/YYYY hh:mm:ss"
+  # for example "04/07/2014 15:18:49"
+  # TODO: determine if this is true for other locales.
+  _LINK_DATE_TIME_RE = re.compile(
+      r'([0-9][0-9])/([0-9][0-9])/([0-9][0-9][0-9][0-9]) '
+      r'([0-9][0-9]):([0-9][0-9]):([0-9][0-9])')
+
   _PRODUCT_KEY_VALUES = {
       '0': 'name',
       '1': 'version',
@@ -126,8 +144,7 @@ class AMCachePlugin(interface.WindowsRegistryPlugin):
       'f': 'product_code',
       '10': 'package_code',
       '11': 'msi_product_code',
-      '12': 'msi_package_code',
-  }
+      '12': 'msi_package_code'}
 
   def _GetValueDataAsObject(
       self, parser_mediator, key_path, value_name, registry_value):
@@ -167,6 +184,112 @@ def _GetValueDataAsObject(
 
     return value_object
 
+  def _ParseApplicationSubKey(self, parser_mediator, application_sub_key):
+    """Parses a Root\\InventoryApplicationFile\\%NAME%|%IDENTIFIER% key.
+
+    Args:
+      parser_mediator (ParserMediator): mediates interactions between parsers
+          and other components, such as storage and dfVFS.
+      application_sub_key (dfwinreg.WinRegistryKey): application sub key of the
+          InventoryApplicationFile Windows Registry key.
+    """
+    event_data = AMCacheFileEventData()
+
+    for value_name, attribute_name in self._APPLICATION_SUB_KEY_VALUES.items():
+      value = application_sub_key.GetValueByName(value_name)
+      if value:
+        value_data = self._GetValueDataAsObject(
+            parser_mediator, application_sub_key.path, value_name, value)
+
+        setattr(event_data, attribute_name, value_data)
+
+    install_date_value = application_sub_key.GetValueByName('InstallDate')
+    if install_date_value:
+      date_time = self._ParseDateStringValue(
+         parser_mediator, application_sub_key.path, install_date_value)
+      event = time_events.DateTimeValuesEvent(
+          date_time, definitions.TIME_DESCRIPTION_LINK_TIME)
+      parser_mediator.ProduceEventWithEventData(event, event_data)
+
+    install_date_msi_value = application_sub_key.GetValueByName(
+        'InstallDateMsi')
+    if install_date_msi_value:
+      date_time = self._ParseDateStringValue(
+         parser_mediator, application_sub_key.path, install_date_msi_value)
+      event = time_events.DateTimeValuesEvent(
+          date_time, definitions.TIME_DESCRIPTION_LINK_TIME)
+      parser_mediator.ProduceEventWithEventData(event, event_data)
+
+    link_date_value = application_sub_key.GetValueByName('LinkDate')
+    if link_date_value:
+      date_time = self._ParseDateStringValue(
+         parser_mediator, application_sub_key.path, link_date_value)
+      event = time_events.DateTimeValuesEvent(
+          date_time, definitions.TIME_DESCRIPTION_LINK_TIME)
+      parser_mediator.ProduceEventWithEventData(event, event_data)
+
+  def _ParseDateStringValue(self, parser_mediator, key_path, registry_value):
+    """Parses a date and time string value.
+
+    Args:
+      parser_mediator (ParserMediator): mediates interactions between parsers
+          and other components, such as storage and dfvfs.
+      key_path (str): key path.
+      registry_value (dfwinreg.WinRegistryValue): Windows Registry value.
+
+    Returns:
+      dfdatetime_time_elements.TimeElements: date and time value or None
+          if not available.
+    """
+    if not registry_value.DataIsString():
+      parser_mediator.ProduceExtractionWarning((
+          'unsupported {0:s} with value data type: {1:s} in key: '
+          '{2:s}').format(
+              registry_value.name, registry_value.data_type_string, key_path))
+      return None
+
+    date_time_string = registry_value.GetDataAsObject()
+    if not date_time_string:
+      parser_mediator.ProduceExtractionWarning(
+          'missing {0:s} value data in key: {1:s}'.format(
+              registry_value.name, key_path))
+      return None
+
+    re_match = self._LINK_DATE_TIME_RE.match(date_time_string)
+    if not re_match:
+      parser_mediator.ProduceExtractionWarning(
+          'unsupported {0:s} value data: {1!s} in key: {2:s}'.format(
+              registry_value.name, date_time_string, key_path))
+      return None
+
+    month, day_of_month, year, hours, minutes, seconds= re_match.groups()
+
+    try:
+      year = int(year, 10)
+      month = int(month, 10)
+      day_of_month = int(day_of_month, 10)
+      hours = int(hours, 10)
+      minutes = int(minutes, 10)
+      seconds = int(seconds, 10)
+    except (TypeError, ValueError):
+      parser_mediator.ProduceExtractionWarning(
+          'invalid {0:s} date time value: {1!s} in key: {2:s}'.format(
+              registry_value.name, date_time_string, key_path))
+      return None
+
+    time_elements_tuple = (year, month, day_of_month, hours, minutes, seconds)
+
+    try:
+      date_time = dfdatetime_time_elements.TimeElements(
+          time_elements_tuple=time_elements_tuple)
+    except ValueError:
+      parser_mediator.ProduceExtractionWarning(
+          'invalid {0:s} date time value: {1!s} in key: {2:s}'.format(
+              registry_value.name, time_elements_tuple, key_path))
+      return None
+
+    return date_time
+
   def _ParseFileKey(self, parser_mediator, file_key):
     """Parses a Root\\File key.
 
@@ -220,42 +343,54 @@ def _ParseFileReferenceKey(self, parser_mediator, file_reference_key):
 
       setattr(event_data, attribute_name, value_data)
 
-    amcache_time_value = file_reference_key.GetValueByName(
+    write_time_value = file_reference_key.GetValueByName(
         self._AMCACHE_ENTRY_WRITE_TIME)
-    if amcache_time_value:
-      timestamp = amcache_time_value.GetDataAsObject()
-      amcache_time = dfdatetime_filetime.Filetime(timestamp=timestamp)
+    if write_time_value:
+      timestamp = write_time_value.GetDataAsObject()
+      date_time = dfdatetime_filetime.Filetime(timestamp=timestamp)
       event = time_events.DateTimeValuesEvent(
-          amcache_time, definitions.TIME_DESCRIPTION_MODIFICATION)
+          date_time, definitions.TIME_DESCRIPTION_MODIFICATION)
       parser_mediator.ProduceEventWithEventData(event, event_data)
 
     creation_time_value = file_reference_key.GetValueByName(
         self._AMCACHE_FILE_CREATION_TIME)
     if creation_time_value:
       timestamp = creation_time_value.GetDataAsObject()
-      creation_time = dfdatetime_filetime.Filetime(timestamp=timestamp)
+      date_time = dfdatetime_filetime.Filetime(timestamp=timestamp)
       event = time_events.DateTimeValuesEvent(
-          creation_time, definitions.TIME_DESCRIPTION_CREATION)
+          date_time, definitions.TIME_DESCRIPTION_CREATION)
       parser_mediator.ProduceEventWithEventData(event, event_data)
 
     modification_time_value = file_reference_key.GetValueByName(
         self._AMCACHE_FILE_MODIFICATION_TIME)
     if modification_time_value:
       timestamp = modification_time_value.GetDataAsObject()
-      modification_time = dfdatetime_filetime.Filetime(timestamp=timestamp)
+      date_time = dfdatetime_filetime.Filetime(timestamp=timestamp)
       event = time_events.DateTimeValuesEvent(
-          modification_time, definitions.TIME_DESCRIPTION_MODIFICATION)
+          date_time, definitions.TIME_DESCRIPTION_MODIFICATION)
       parser_mediator.ProduceEventWithEventData(event, event_data)
 
-    compilation_time_value = file_reference_key.GetValueByName(
-        self._AMCACHE_COMPILATION_TIME)
-    if compilation_time_value:
-      timestamp = compilation_time_value.GetDataAsObject()
-      link_time = dfdatetime_posix_time.PosixTime(timestamp=timestamp)
+    link_time_value = file_reference_key.GetValueByName(self._AMCACHE_LINK_TIME)
+    if link_time_value:
+      timestamp = link_time_value.GetDataAsObject()
+      date_time = dfdatetime_posix_time.PosixTime(timestamp=timestamp)
       event = time_events.DateTimeValuesEvent(
-          link_time, definitions.TIME_DESCRIPTION_CHANGE)
+          date_time, definitions.TIME_DESCRIPTION_LINK_TIME)
       parser_mediator.ProduceEventWithEventData(event, event_data)
 
+  def _ParseInventoryApplicationFileKey(
+      self, parser_mediator, inventory_application_file_key):
+    """Parses a Root\\InventoryApplicationFile key.
+
+    Args:
+      parser_mediator (ParserMediator): mediates interactions between parsers
+          and other components, such as storage and dfVFS.
+      inventory_application_file_key (dfwinreg.WinRegistryKey): the
+          InventoryApplicationFile Windows Registry key.
+    """
+    for application_sub_key in inventory_application_file_key.GetSubkeys():
+      self._ParseApplicationSubKey(parser_mediator, application_sub_key)
+
   def _ParseProgramKey(self, parser_mediator, program_key):
     """Parses a program key (a sub key of Root\\Programs) for events.
 
@@ -268,13 +403,11 @@ def _ParseProgramKey(self, parser_mediator, program_key):
 
     for value_name, attribute_name in self._PRODUCT_KEY_VALUES.items():
       value = program_key.GetValueByName(value_name)
-      if not value:
-        continue
-
-      value_data = self._GetValueDataAsObject(
-          parser_mediator, program_key.path, value_name, value)
+      if value:
+        value_data = self._GetValueDataAsObject(
+            parser_mediator, program_key.path, value_name, value)
 
-      setattr(event_data, attribute_name, value_data)
+        setattr(event_data, attribute_name, value_data)
 
     installation_time_value = program_key.GetValueByName(
         self._AMCACHE_P_INSTALLATION_TIME)
@@ -312,6 +445,9 @@ def _ParseRootKey(self, parser_mediator, root_key):
       if sub_key.name == 'File':
         self._ParseFileKey(parser_mediator, sub_key)
 
+      elif sub_key.name == 'InventoryApplicationFile':
+        self._ParseInventoryApplicationFileKey(parser_mediator, sub_key)
+
       elif sub_key.name == 'Programs':
         self._ParseProgramsKey(parser_mediator, sub_key)
 
@@ -328,7 +464,6 @@ def _ParseSubKey(self, parser_mediator, registry_key):
     for sub_key in registry_key.GetSubkeys():
       self._ParseSubKey(parser_mediator, sub_key)
 
-
   def ExtractEvents(self, parser_mediator, registry_key, **kwargs):
     """Extracts events from a Windows Registry key.
 

tests/parsers/winreg_plugins/amcache.py

@@ -6,6 +6,7 @@
 
 from dfwinreg import regf as dfwinreg_regf
 
+from plaso.lib import definitions
 from plaso.parsers.winreg_plugins import amcache
 
 from tests.parsers.winreg_plugins import test_lib
@@ -54,7 +55,8 @@ def testProcess(self):
         'data_type': 'windows:registry:amcache',
         'date_time': '1992-06-19 22:22:17',
         'full_path': expected_full_path,
-        'sha1': '82274eef0911a948f91425f5e5b0e730517fe75e'}
+        'sha1': '82274eef0911a948f91425f5e5b0e730517fe75e',
+        'timestamp_desc': definitions.TIME_DESCRIPTION_LINK_TIME}
 
     self.CheckEventValues(storage_writer, events[0], expected_event_values)
 
@@ -68,12 +70,43 @@ def testProcess(self):
             'c:\\program files (x86)\\fileinsight'],
         'name': 'FileInsight - File analysis tool',
         'publisher': 'McAfee Inc.',
+        'timestamp_desc': definitions.TIME_DESCRIPTION_INSTALLATION,
         'uninstall_key': [
             'HKEY_LOCAL_MACHINE\\Software\\Wow6432Node\\Microsoft\\Windows\\'
             'CurrentVersion\\Uninstall\\FileInsight']}
 
     self.CheckEventValues(storage_writer, events[1285], expected_event_values)
 
+  def testProcessWindows101(self):
+    """Tests the Process function on a Windows 10 1807 AMCache.hve file."""
+    test_file_entry = self._GetTestFileEntry(['win10-Amcache.hve'])
+
+    file_object = test_file_entry.GetFileObject()
+
+    registry_file = dfwinreg_regf.REGFWinRegistryFile(
+        ascii_codepage='cp1252', emulate_virtual_keys=False)
+    registry_file.Open(file_object)
+
+    registry_key = registry_file.GetKeyByPath('\\Root')
+
+    plugin = amcache.AMCachePlugin()
+    storage_writer = self._ParseKeyWithPlugin(
+        registry_key, plugin, file_entry=test_file_entry)
+
+    self.assertEqual(storage_writer.number_of_events, 236)
+    self.assertEqual(storage_writer.number_of_extraction_warnings, 0)
+    self.assertEqual(storage_writer.number_of_recovery_warnings, 0)
+
+    events = list(storage_writer.GetSortedEvents())
+
+    expected_event_values = {
+        'data_type': 'windows:registry:amcache',
+        'date_time': '1997-01-10 22:26:24',
+        'full_path': 'c:\\windows\\system32\\svchost.exe',
+        'timestamp_desc': definitions.TIME_DESCRIPTION_LINK_TIME}
+
+    self.CheckEventValues(storage_writer, events[0], expected_event_values)
+
 
 if __name__ == '__main__':
   unittest.main()