#!/usr/bin/env python
"""Classes for exporting data from AFF4 to the rest of the world.

ExportConverters are defined in `export_converters` subfolder.

ExportConverters convert various complex RDFValues to simple RDFValues
(without repeated fields, without recursive field definitions) that can
easily be written to a relational database or just to a set of files.
"""

from grr_response_core.lib.util import collection
from grr_response_server import export_converters_registry
from grr_response_server.export_converters import base


class Error(Exception):
  """Errors generated by export converters."""


class NoConverterFound(Error):
  """Raised when no converter is found for particular value."""


class ExportError(Error):
  """Unspecified error while exporting."""


def GetMetadata(client_id, client_full_info):
  """Builds ExportedMetadata object for a given client id and ClientFullInfo."""

  metadata = base.ExportedMetadata()

  last_snapshot = None
  if client_full_info.HasField("last_snapshot"):
    last_snapshot = client_full_info.last_snapshot

  metadata.client_urn = client_id
  metadata.client_age = client_full_info.metadata.first_seen

  if last_snapshot is not None:
    kb = client_full_info.last_snapshot.knowledge_base
    os_release = last_snapshot.os_release
    os_version = last_snapshot.os_version

    metadata.hostname = kb.fqdn
    metadata.os = kb.os
    metadata.os_release = os_release
    metadata.os_version = os_version
    metadata.usernames = ",".join(user.username for user in kb.users)

    addresses = last_snapshot.GetMacAddresses()
    if addresses:
      metadata.mac_address = "\n".join(last_snapshot.GetMacAddresses())
    metadata.hardware_info = last_snapshot.hardware_info
    metadata.kernel_version = last_snapshot.kernel

    ci = last_snapshot.cloud_instance
    if ci is not None:
      if ci.cloud_type == ci.InstanceType.AMAZON:
        metadata.cloud_instance_type = metadata.CloudInstanceType.AMAZON
        metadata.cloud_instance_id = ci.amazon.instance_id
      elif ci.cloud_type == ci.InstanceType.GOOGLE:
        metadata.cloud_instance_type = metadata.CloudInstanceType.GOOGLE
        metadata.cloud_instance_id = ci.google.unique_id

  system_labels = set()
  user_labels = set()
  for l in client_full_info.labels:
    if l.owner == "GRR":
      system_labels.add(l.name)
    else:
      user_labels.add(l.name)

  metadata.labels = ",".join(sorted(system_labels | user_labels))
  metadata.system_labels = ",".join(sorted(system_labels))
  metadata.user_labels = ",".join(sorted(user_labels))

  return metadata


def ConvertValuesWithMetadata(metadata_value_pairs, options=None):
  """Converts a set of RDFValues into a set of export-friendly RDFValues.

  Args:
    metadata_value_pairs: Tuples of (metadata, rdf_value), where metadata is an
      instance of ExportedMetadata and rdf_value is an RDFValue subclass
      instance to be exported.
    options: rdfvalue.ExportOptions instance that will be passed to
      ExportConverters.

  Yields:
    Converted values. Converted values may be of different types.

  Raises:
    NoConverterFound: in case no suitable converters were found for a value in
                      metadata_value_pairs. This error is only raised after
                      all values in metadata_value_pairs are attempted to be
                      converted. If there are multiple value types that could
                      not be converted because of the lack of corresponding
                      converters, only the last one will be specified in the
                      exception message.
  """
  no_converter_found_error = None
  metadata_value_groups = collection.Group(
      metadata_value_pairs, lambda pair: pair[1].__class__.__name__
  )
  for metadata_values_group in metadata_value_groups.values():
    _, first_value = metadata_values_group[0]
    converters_classes = export_converters_registry.GetConvertersByValue(
        first_value
    )
    if not converters_classes:
      no_converter_found_error = "No converters found for value: %s" % str(
          first_value
      )
      continue

    converters = [cls(options) for cls in converters_classes]
    for converter in converters:
      for result in converter.BatchConvert(metadata_values_group):
        yield result

  if no_converter_found_error is not None:
    raise NoConverterFound(no_converter_found_error)


def ConvertValues(default_metadata, values, options=None):
  """Converts a set of RDFValues into a set of export-friendly RDFValues.

  Args:
    default_metadata: base.ExportedMetadata instance with basic information
      about where the values come from. This metadata will be passed to
      exporters.
    values: Values to convert. They should be of the same type.
    options: rdfvalue.ExportOptions instance that will be passed to
      ExportConverters.

  Returns:
    Converted values. Converted values may be of different types
    (unlike the source values which are all of the same type). This is due to
    the fact that multiple ExportConverters may be applied to the same value
    thus generating multiple converted values of different types.

  Raises:
    NoConverterFound: in case no suitable converters were found for the values.
  """
  batch_data = [(default_metadata, obj) for obj in values]
  return ConvertValuesWithMetadata(batch_data, options=options)