def ii_init(self, record_info_in: Sdk.RecordInfo) -> bool:
# Make sure the user provided a field to parse
# Returns a new, empty RecordCreator object that is identical to record_info_in.
record_info_out = record_info_in.clone()
# Lets the downstream tools know what the outgoing record metadata will look like
self.parent.output.init(record_info_out)
self.parent.error_output.init(record_info_out)
# Creating a new, empty record creator based on record_info_out's record layout.
self.record_creator = record_info_out.construct_record_creator()
# Instantiate a new instance of the RecordCopier class.
self.record_copier = Sdk.RecordCopier(record_info_out, record_info_in)
# Map each column of the input to where we want in the output.
for index in range(record_info_in.num_fields):
# Adding a field index mapping.
self.record_copier.add(index, index)
# Let record copier know that all field mappings have been added.
self.record_copier.done_adding()
return True
def ii_init(self, record_info_in: Sdk.RecordInfo) -> bool:
# Make sure the user provided a field to parse
if self.parent.url is None:
self.parent.display_error_msg('Select a source field')
return False
# Get information about the source path field
try:
self.SourceField = record_info_in.get_field_by_name(
self.parent.url)
except Exception as e:
self.parent.display_error_msg(
f'Invalid source field: {self.parent.url}')
return False
#match_field_type: Sdk.FieldType = self.SourceField.type
#match_field_size: int = self.SourceField.size
# Returns a new, empty RecordCreator object that is identical to record_info_in.
self.record_info_out = record_info_in.clone()
#output field config
self.fields = [['protocol', Sdk.FieldType.string, 20],
['net_location', Sdk.FieldType.string, 100],
['path', Sdk.FieldType.string, 1000],
['query', Sdk.FieldType.string, 1000],
['parsed_query', Sdk.FieldType.string, 1000],
['fragment', Sdk.FieldType.string, 100],
['hostname', Sdk.FieldType.string, 100],
['port', Sdk.FieldType.int32, 10]]
self.outRecords = []
# Adds field to record with specified name and output type.
for name, dtype, size in self.fields:
self.outRecords.append(
self.record_info_out.add_field(name, dtype, size))
# Lets the downstream tools know what the outgoing record metadata will look like
self.parent.output.init(self.record_info_out)
# Creating a new, empty record creator based on record_info_out's record layout.
self.record_creator = self.record_info_out.construct_record_creator()
# Instantiate a new instance of the RecordCopier class.
self.record_copier = Sdk.RecordCopier(self.record_info_out,
record_info_in)
# Map each column of the input to where we want in the output.
for index in range(record_info_in.num_fields):
# Adding a field index mapping.
self.record_copier.add(index, index)
# Let record copier know that all field mappings have been added.
self.record_copier.done_adding()
return True
def ii_init(self, record_info_in: Sdk.RecordInfo) -> bool:
# Make sure the user provided a field to parse
if self.parent.SourceFieldName is None:
self.parent.display_error_msg('Select a source field')
return False
# Make sure the user provided a field to parse
if self.parent.DestFieldName is None:
self.parent.display_error_msg('Select a destination field')
return False
# Get information about the source path field
self.SourceField = record_info_in.get_field_by_name(
self.parent.SourceFieldName)
#match_field_type: Sdk.FieldType = self.SourceField.type
#match_field_size: int = self.SourceField.size
# Get information about the destination path field
self.DestField = record_info_in.get_field_by_name(
self.parent.DestFieldName)
# Returns a new, empty RecordCreator object that is identical to record_info_in.
record_info_out = record_info_in.clone()
# Adds field to record with specified name and output type.
self.OutputField = record_info_out.add_field(self.output_name,
self.output_type,
self.output_size)
# Lets the downstream tools know what the outgoing record metadata will look like
self.parent.output.init(record_info_out)
self.parent.error_output.init(record_info_out)
# Creating a new, empty record creator based on record_info_out's record layout.
self.record_creator = record_info_out.construct_record_creator()
# Instantiate a new instance of the RecordCopier class.
self.record_copier = Sdk.RecordCopier(record_info_out, record_info_in)
# Map each column of the input to where we want in the output.
for index in range(record_info_in.num_fields):
# Adding a field index mapping.
self.record_copier.add(index, index)
# Let record copier know that all field mappings have been added.
self.record_copier.done_adding()
return True
def ii_init(self, record_info_in: Sdk.RecordInfo) -> bool:
# Make sure the user provided a field to parse
if self.parent.CompanyNamesField is None:
self.parent.display_error_msg('Select a field')
return False
self.CompanyNamesField = record_info_in.get_field_by_name(
self.parent.CompanyNamesField)
# Returns a new, empty RecordCreator object that is identical to record_info_in.
output_record = record_info_in.clone()
# Adds field to record with specified name and output type.
self.IdField = output_record.add_field("Glassdoor ID",
Sdk.FieldType.v_wstring, 100)
self.NameField = output_record.add_field("Glassdoor Name",
Sdk.FieldType.v_wstring, 100)
self.ReviewLinkField = output_record.add_field(
"Begin Review Search Link", Sdk.FieldType.v_wstring, 1000)
self.ReviewPagesField = output_record.add_field(
"Review Pages", Sdk.FieldType.int64)
self.InterviewLinkField = output_record.add_field(
"Begin Interview Search Link", Sdk.FieldType.v_wstring, 1000)
self.InterviewPagesField = output_record.add_field(
"Interview Pages", Sdk.FieldType.int64)
# Lets the downstream tools know what the outgoing record metadata will look like
self.parent.Output.init(output_record)
self.parent.Reviews.init(self.ReviewsRecord)
self.parent.Interviews.init(self.InterviewsRecord)
# Creating a new, empty record creator based on record_info_out's record layout.
self.OutputCreator = output_record.construct_record_creator()
# Instantiate a new instance of the RecordCopier class.
self.OutputCopier = Sdk.RecordCopier(output_record, record_info_in)
# Map each column of the input to where we want in the output.
for index in range(record_info_in.num_fields):
# Adding a field index mapping.
self.OutputCopier.add(index, index)
# Let record copier know that all field mappings have been added.
self.OutputCopier.done_adding()
return True