Carla Alessandra Lima Reis

23 papers B 3C 5Journal 2Unranked 12
YearRankTypeTitle / Venue / Authors
2015 conf
CSD@ICSE
Adailton Magalhães Lima, Rodrigo Quites Reis, Carla Alessandra Lima Reis
2014 conf
RSSE@ICSE
Thiago Jorge A. Santos, Adailton Magalhães Lima, Carla Alessandra Lima Reis, Rodrigo Quites Reis
2014 C conf
CLEI
Daniel Dias de Carvalho, Larissa Fernandes Chagas, Carla Alessandra Lima Reis
2014 C conf
SPICE
Daniel Dias de Carvalho, Larissa Fernandes Chagas, Adailton Magalhães Lima, Carla Alessandra Lima Reis
2014 C conf
SPICE
Larissa Fernandes Chagas, Daniel Dias de Carvalho, Adailton Magalhães Lima, Carla Alessandra Lima Reis
2012 C conf
CLEI
Silvia Nunes das Dôres, Patricia Matias Lopes, Carla Alessandra Lima Reis
2011 ed.
SBQS
Carla Alessandra Lima Reis, Tayana Uchôa Conte
2010 J jnl
RITA
Luciana Maria Azevedo Nascimento, Talita Vieira Ribeiro, Carla Alessandra Lima Reis, Rodrigo Quites Reis, Adailton Magalhães Lima
2010 conf
SBQS
Renata W. Monteiro, Amanda Sizo, Tácio Vinicius, Luis Felipe Guimaràes, Cláudio Roberto L. Martins, Ernani de Oliveira Sales, Carla Alessandra Lima Reis
2009 conf
SBQS
Jadielly F. Oliveira, Gabriela F. Andrade, Lorena C. Tavares, Carla Alessandra Lima Reis
2008 conf
SBES
Ernani de Oliveira Sales, Carla Alessandra Lima Reis, Rodrigo Quites Reis
2008 conf
SBQS
Vanderlene Covre, Carla Alessandra Lima Reis, Eloi Luiz Favero
2007 conf
SBQS
Carla Paxiúba, Marcelo Pereira, Carla Alessandra Lima Reis, Rodrigo Quites Reis
2002 C conf
SEKE
Carla Alessandra Lima Reis, Rodrigo Quites Reis, Heribert Schlebbe, Daltro José Nunes
2002 J jnl
Ann. Softw. Eng.
Rodrigo Quites Reis, Carla Alessandra Lima Reis, Heribert Schlebbe, Daltro José Nunes
2002 B conf
COMPSAC
Rodrigo Quites Reis, Carla Alessandra Lima Reis, Heribert Schlebbe, Daltro José Nunes
2002 conf
HCC
Carla Alessandra Lima Reis, Rodrigo Quites Reis, Marcelo M. de Abreu, Heribert Schlebbe, Daltro José Nunes
2002 B conf
COMPSAC
Carla Alessandra Lima Reis, Rodrigo Quites Reis, Heribert Schlebbe, Daltro José Nunes
2002 B conf
ICGT
Carla Alessandra Lima Reis, Rodrigo Quites Reis, Marcelo M. de Abreu, Heribert Schlebbe, Daltro José Nunes
2001 conf
SBES
Carla Alessandra Lima Reis, Rodrigo Quites Reis, Heribert Schlebbe, Daltro J. Nunes
2001 conf
SBES
Rodrigo Quites Reis, Carla Alessandra Lima Reis, Daltro José Nunes
2001 conf
CRIWG
Rodrigo Quites Reis, Carla Alessandra Lima Reis, Daltro José Nunes
1999 conf
SBES
Fábio Augusto das Dores Silva, Rodrigo Quites Reis, Carla Alessandra Lima Reis, Daltro José Nunes
redb/extractors/macho_extractors/macho_features.py
← Index redb/extractors/macho_extractors/macho_features.py python
import inspect
import json
from datetime import datetime, timezone
from typing import Any

from redb.extractors.enum import Tag
from redb.extractors.macho_extractor import MachOExtractor
from redb.models.dataclasses import MachO


class MachOFeaturesExtractor(MachOExtractor):

    def __init__(
        self,
        filepath,
        log,
        exporters=None,
        index_prefix=None,
        elastic_index=None,
        known_benign=False,
        known_malicious=False,
        macho=None,
    ):
        super().__init__(
            filepath,
            log,
            exporters,
            index_prefix,
            elastic_index,
            known_benign,
            known_malicious,
            macho,
        )
        self.elastic_index = self.index_prefix + "-macho_features"
        self.log.debug(inspect.currentframe().f_code.co_name)

    def tag(self):
        return Tag.MACHO_FEATURES.value

    def _extract_macho_features_for_arch(self, arch_name):
        """Extract basic MachO features for a specific architecture."""
        self.log.debug(inspect.currentframe().f_code.co_name)

        if not self.macho:
            return None

        try:
            # Get general info using new API
            general_info = self.macho.get_general_info()
            if not general_info:
                return None

            # Get header info for this specific architecture
            header_raw = self.macho.get_macho_header(arch=arch_name)
            header_formatted = self.macho.get_macho_header(arch=arch_name, formatted=True)

            if not header_raw or not header_formatted:
                return None

            # Get entry point using new API
            entry_point = None
            try:
                entry_point_info = self.macho.get_entry_point(arch=arch_name)
                if entry_point_info:
                    if isinstance(entry_point_info, dict):
                        entry_point = entry_point_info.get('entryoff') or entry_point_info.get('entry_address')
                    else:
                        entry_point = entry_point_info
            except Exception as e:
                self.log.debug(f"No entry point found: {e}")

            # Get UUID using new API
            uuid = None
            try:
                uuid_info = self.macho.get_uuid(arch=arch_name)
                if uuid_info:
                    uuid = str(uuid_info)
            except Exception as e:
                self.log.debug(f"No UUID found: {e}")

            # Get version info using new API with formatting
            version_info = None
            version_info_str = None
            try:
                version_info = self.macho.get_version_info(arch=arch_name)
                version_info_str = self.macho.get_version_info(arch=arch_name, formatted=True)
            except Exception as e:
                self.log.debug(f"No version info found: {e}")

            # Get segments using new API
            segments = self.macho.get_segments(arch=arch_name)
            number_of_segments = len(segments) if segments else 0

            # Get dylib info using new API
            dylib_names = self.macho.get_dylib_names(arch=arch_name)
            number_of_dylibs = len(dylib_names) if dylib_names else 0

            # Count imports using new API
            imported_functions = self.macho.get_imported_functions(arch=arch_name)
            number_of_imports = sum(len(funcs) for funcs in imported_functions.values()) if imported_functions else 0

            # Count exports using new API
            exported_symbols = self.macho.get_exported_symbols(arch=arch_name)
            number_of_exports = sum(len(symbols) for symbols in exported_symbols.values()) if exported_symbols else 0

            # Get unique load command types using new API
            load_commands_set = self.macho.get_load_commands_set(arch=arch_name, formatted=True)
            if isinstance(load_commands_set, set):
                load_commands_set = sorted(list(load_commands_set))

            # Get architectures for listing purposes
            architectures = self.macho.get_architectures()

            # Determine binary properties
            is_64bit = header_raw.get('magic') in [0xFEEDFACF, 0xCFFAEDFE]  # MH_MAGIC_64, MH_CIGAM_64
            is_signed = self._is_signed()
            
            # Create MachO dataclass
            macho_features = MachO(
                # Raw values from new API
                magic=header_raw.get('magic', 0),
                cputype=header_raw.get('cputype', 0),
                cpusubtype=header_raw.get('cpusubtype', 0),
                filetype=header_raw.get('filetype', 0),
                ncmds=header_raw.get('ncmds', 0),
                sizeofcmds=header_raw.get('sizeofcmds', 0),
                flags=header_raw.get('flags', 0),
                architecture=header_raw.get('cputype', 0),  # Use cputype as architecture
                architectures=[arch_info.get('cputype', 0) for arch_info in [self.macho.get_macho_header(arch=arch) for arch in architectures] if arch_info],
                # Human-readable values from formatted API
                magic_str=header_formatted.get('magic', ''),
                cputype_str=header_formatted.get('cputype', ''),
                cpusubtype_str=header_formatted.get('cpusubtype', ''),
                filetype_str=header_formatted.get('filetype', ''),
                flags_str=header_formatted.get('flags', '').split(', ') if header_formatted.get('flags') else [],
                architecture_str=header_formatted.get('cputype', ''),
                architectures_str=[arch_info.get('cputype', '') for arch_info in [self.macho.get_macho_header(arch=arch, formatted=True) for arch in architectures] if arch_info],
                # Additional properties
                is_64bit=is_64bit,
                is_signed=is_signed,
                number_of_load_commands=header_raw.get('ncmds', 0),
                load_commands_set=load_commands_set,
                number_of_segments=number_of_segments,
                number_of_dylibs=number_of_dylibs,
                number_of_imports=number_of_imports,
                number_of_exports=number_of_exports,
                entry_point=entry_point,
                uuid=uuid,
                version_info=version_info,
                version_info_str=version_info_str
            )
            
            return macho_features
            
        except Exception as e:
            self.log.error(f"Error extracting MachO features: {e}")
            return None

    def extract(self):
        self.log.debug(inspect.currentframe().f_code.co_name)
        try:
            if not self.macho:
                return None
            architectures = self.macho.get_architectures()
            if not architectures:
                return None

            # For FAT binaries, return features for all architectures
            # For single arch, return just the single result
            if len(architectures) > 1:
                # FAT binary - return list of features for each architecture
                results = []
                for arch_name in architectures:
                    features = self._extract_macho_features_for_arch(arch_name)
                    if features:
                        # Add architecture identifier to the result
                        if hasattr(features, '__dict__'):
                            # If it's a dataclass, we can access its dict
                            features.arch_identifier = arch_name
                        results.append(features)
                return results
            else:
                # Single architecture - return single result
                return self._extract_macho_features_for_arch(architectures[0])
        except Exception as e:
            self.log.error(f"Error extracting MachO features: {e}")
            return None

    def prepare_export_data(self, exporter_type: str) -> Any:
        if exporter_type == "ElasticsearchExporter":
            return self.extract()
        elif exporter_type == "ClickHouseExporter":
            if not self.macho:
                return None

            # Get architectures (macho is already parsed in __init__)
            try:
                architectures = self.macho.get_architectures()
            except Exception as e:
                self.log.error(f"Could not get architectures: {e}")
                return None

            data = []
            current_time = datetime.now(timezone.utc)

            # Loop through each architecture (1 for single, multiple for FAT)
            for arch_name in architectures:
                # Extract features for this specific architecture
                macho_features = self._extract_macho_features_for_arch(arch_name)
                if not macho_features:
                    continue

                # Get architecture-specific sha256
                try:
                    arch_general_info = self.macho.get_general_info(arch=arch_name)
                    arch_sha256 = arch_general_info.get('SHA256', self.sha256)
                except Exception as e:
                    self.log.warning(f"Could not get arch-specific hash for {arch_name}: {e}")
                    arch_sha256 = self.sha256

                # Architecture field (kept in features table for ORDER BY)
                architecture = macho_features.cputype    # Raw CPU type value

                data.append([
                    arch_sha256,                          # sha256 (architecture-specific)
                    architecture,                         # architecture (raw CPU type)
                    # Raw values
                    macho_features.magic,                 # magic
                    macho_features.cputype,               # cputype
                    macho_features.cpusubtype,            # cpusubtype
                    macho_features.filetype,              # filetype
                    macho_features.ncmds,                 # ncmds
                    macho_features.sizeofcmds,            # sizeofcmds
                    macho_features.flags,                 # flags
                    # Human-readable values
                    macho_features.magic_str,             # magic_str
                    macho_features.cputype_str,           # cputype_str
                    macho_features.cpusubtype_str,        # cpusubtype_str
                    macho_features.filetype_str,          # filetype_str
                    macho_features.flags_str,             # flags_str
                    # Other fields
                    macho_features.is_64bit,              # is_64bit
                    macho_features.is_signed,             # is_signed
                    macho_features.entry_point if macho_features.entry_point else None,           # entry_point
                    macho_features.uuid if macho_features.uuid else None,  # uuid
                    json.dumps(macho_features.version_info) if macho_features.version_info else "{}",  # version_info
                    json.dumps(macho_features.version_info_str) if macho_features.version_info_str else "{}",  # version_info_str
                    macho_features.load_commands_set,     # load_commands_set
                    macho_features.number_of_segments,    # number_of_segments
                    macho_features.number_of_dylibs,      # number_of_dylibs
                    macho_features.number_of_imports,     # number_of_imports
                    macho_features.number_of_exports,     # number_of_exports
                    current_time,                         # analysis_date
                ])

            column_names = [
                'sha256', 'architecture',
                # Raw values
                'magic', 'cputype', 'cpusubtype', 'filetype', 'ncmds', 'sizeofcmds', 'flags',
                # Human-readable values
                'magic_str', 'cputype_str', 'cpusubtype_str', 'filetype_str', 'flags_str',
                # Other fields
                'is_64bit', 'is_signed',
                'entry_point', 'uuid', 'version_info', 'version_info_str', 'load_commands_set',
                'number_of_segments', 'number_of_dylibs', 'number_of_imports', 'number_of_exports',
                'analysis_date'
            ]

            if not data:
                return None

            column_type_names = [
                'FixedString(64)', 'Nullable(UInt32)',
                # Raw values
                'UInt32', 'UInt32', 'UInt32', 'UInt32', 'UInt32', 'UInt32', 'UInt32',
                # Human-readable values
                'LowCardinality(String)', 'LowCardinality(String)', 'LowCardinality(String)', 'LowCardinality(String)', 'Array(LowCardinality(String))',
                # Other fields
                'UInt8', 'UInt8',
                'Nullable(UInt64)', 'Nullable(String)', 'JSON', 'JSON', 'Array(Nullable(String))',
                'UInt32', 'UInt32', 'UInt32', 'UInt32', 'DateTime64(3, \'UTC\')'
            ]

            return (data, column_names, column_type_names)
        
        return None

    def get_clickhouse_table(self) -> str:
        return "redb_macho_features"