attempt to fix some logic

This commit is contained in:
overcuriousity 2025-09-17 00:05:48 +02:00
parent 47ce7ff883
commit f2db739fa1
6 changed files with 218 additions and 142 deletions

View File

@ -41,7 +41,7 @@ class GraphManager:
self.correlation_index = {}
# Compile regex for date filtering for efficiency
self.date_pattern = re.compile(r'^\d{4}-\d{2}-\d{2}[ T]\d{2}:\d{2}:\d{2}')
self.EXCLUDED_KEYS = ['confidence', 'provider', 'timestamp', 'type','crtsh_cert_validity_period_days']
self.EXCLUDED_KEYS = ['confidence', 'provider', 'timestamp', 'type','crtsh_cert_validity_period_days','crtsh_cert_source']
def __getstate__(self):
"""Prepare GraphManager for pickling, excluding compiled regex."""
@ -112,7 +112,7 @@ class GraphManager:
def _create_enhanced_correlation_node_and_edges(self, value, correlation_data):
"""
UPDATED: Create correlation node and edges with detailed provider tracking.
UPDATED: Create correlation node and edges with raw provider data (no formatting).
"""
correlation_node_id = f"corr_{hash(str(value)) & 0x7FFFFFFF}"
nodes = correlation_data['nodes']
@ -120,13 +120,14 @@ class GraphManager:
# Create or update correlation node
if not self.graph.has_node(correlation_node_id):
# Determine the most common provider/attribute combination
# Use raw provider/attribute data - no formatting
provider_counts = {}
for source in sources:
# Keep original provider and attribute names
key = f"{source['provider']}_{source['attribute']}"
provider_counts[key] = provider_counts.get(key, 0) + 1
# Use the most common provider/attribute as the primary label
# Use the most common provider/attribute as the primary label (raw)
primary_source = max(provider_counts.items(), key=lambda x: x[1])[0] if provider_counts else "unknown_correlation"
metadata = {
@ -303,18 +304,18 @@ class GraphManager:
return is_new_node
def add_edge(self, source_id: str, target_id: str, relationship_type: str,
confidence_score: float = 0.5, source_provider: str = "unknown",
raw_data: Optional[Dict[str, Any]] = None) -> bool:
"""Add or update an edge between two nodes, ensuring nodes exist."""
confidence_score: float = 0.5, source_provider: str = "unknown",
raw_data: Optional[Dict[str, Any]] = None) -> bool:
"""
UPDATED: Add or update an edge between two nodes with raw relationship labels.
"""
if not self.graph.has_node(source_id) or not self.graph.has_node(target_id):
return False
new_confidence = confidence_score
if relationship_type.startswith("c_"):
edge_label = relationship_type
else:
edge_label = f"{source_provider}_{relationship_type}"
# UPDATED: Use raw relationship type - no formatting
edge_label = relationship_type
if self.graph.has_edge(source_id, target_id):
# If edge exists, update confidence if the new score is higher.
@ -324,7 +325,7 @@ class GraphManager:
self.graph.edges[source_id, target_id]['updated_by'] = source_provider
return False
# Add a new edge with all attributes.
# Add a new edge with raw attributes
self.graph.add_edge(source_id, target_id,
relationship_type=edge_label,
confidence_score=new_confidence,
@ -417,73 +418,45 @@ class GraphManager:
def get_graph_data(self) -> Dict[str, Any]:
"""
Export graph data formatted for frontend visualization.
UPDATED: Fixed certificate validity styling logic for unified data model.
SIMPLIFIED: No certificate styling - frontend handles all visual styling.
"""
nodes = []
for node_id, attrs in self.graph.nodes(data=True):
node_data = {'id': node_id, 'label': node_id, 'type': attrs.get('type', 'unknown'),
'attributes': attrs.get('attributes', []), # Ensure attributes is a list
'description': attrs.get('description', ''),
'metadata': attrs.get('metadata', {}),
'added_timestamp': attrs.get('added_timestamp')}
# UPDATED: Fixed certificate validity styling logic
node_type = node_data['type']
attributes_list = node_data['attributes']
if node_type == 'domain' and isinstance(attributes_list, list):
# Check for certificate-related attributes
has_certificates = False
has_valid_certificates = False
has_expired_certificates = False
for attr in attributes_list:
attr_name = attr.get('name', '').lower()
attr_provider = attr.get('provider', '').lower()
attr_value = attr.get('value')
# Look for certificate attributes from crt.sh provider
if attr_provider == 'crtsh' or 'cert' in attr_name:
has_certificates = True
# Check certificate validity
if attr_name == 'cert_is_currently_valid':
if attr_value is True:
has_valid_certificates = True
elif attr_value is False:
has_expired_certificates = True
# Also check for certificate expiry indicators
elif 'expires_soon' in attr_name and attr_value is True:
has_expired_certificates = True
elif 'expired' in attr_name and attr_value is True:
has_expired_certificates = True
# Apply styling based on certificate status
if has_expired_certificates and not has_valid_certificates:
# Red for expired/invalid certificates
node_data['color'] = {'background': '#ff6b6b', 'border': '#cc5555'}
elif not has_certificates:
# Grey for domains with no certificates
node_data['color'] = {'background': '#c7c7c7', 'border': '#999999'}
# Default green styling is handled by the frontend for domains with valid certificates
node_data = {
'id': node_id,
'label': node_id,
'type': attrs.get('type', 'unknown'),
'attributes': attrs.get('attributes', []), # Raw attributes list
'description': attrs.get('description', ''),
'metadata': attrs.get('metadata', {}),
'added_timestamp': attrs.get('added_timestamp')
}
# Add incoming and outgoing edges to node data
if self.graph.has_node(node_id):
node_data['incoming_edges'] = [{'from': u, 'data': d} for u, _, d in self.graph.in_edges(node_id, data=True)]
node_data['outgoing_edges'] = [{'to': v, 'data': d} for _, v, d in self.graph.out_edges(node_id, data=True)]
node_data['incoming_edges'] = [
{'from': u, 'data': d} for u, _, d in self.graph.in_edges(node_id, data=True)
]
node_data['outgoing_edges'] = [
{'to': v, 'data': d} for _, v, d in self.graph.out_edges(node_id, data=True)
]
nodes.append(node_data)
edges = []
for source, target, attrs in self.graph.edges(data=True):
edges.append({'from': source, 'to': target,
'label': attrs.get('relationship_type', ''),
'confidence_score': attrs.get('confidence_score', 0),
'source_provider': attrs.get('source_provider', ''),
'discovery_timestamp': attrs.get('discovery_timestamp')})
edges.append({
'from': source,
'to': target,
'label': attrs.get('relationship_type', ''),
'confidence_score': attrs.get('confidence_score', 0),
'source_provider': attrs.get('source_provider', ''),
'discovery_timestamp': attrs.get('discovery_timestamp')
})
return {
'nodes': nodes, 'edges': edges,
'nodes': nodes,
'edges': edges,
'statistics': self.get_statistics()['basic_metrics']
}

View File

@ -355,7 +355,7 @@ class CrtShProvider(BaseProvider):
'not_before': cert_data.get('not_before'),
'not_after': cert_data.get('not_after'),
'entry_timestamp': cert_data.get('entry_timestamp'),
'source': 'crt.sh'
'source': 'crtsh'
}
try:
@ -367,8 +367,9 @@ class CrtShProvider(BaseProvider):
metadata['is_currently_valid'] = self._is_cert_valid(cert_data)
metadata['expires_soon'] = (not_after - datetime.now(timezone.utc)).days <= 30
metadata['not_before'] = not_before.strftime('%Y-%m-%d %H:%M:%S UTC')
metadata['not_after'] = not_after.strftime('%Y-%m-%d %H:%M:%S UTC')
# UPDATED: Keep raw date format or convert to standard format
metadata['not_before'] = not_before.isoformat()
metadata['not_after'] = not_after.isoformat()
except Exception as e:
self.logger.logger.debug(f"Error computing certificate metadata: {e}")

View File

@ -155,12 +155,7 @@ class DNSProvider(BaseProvider):
def _query_record(self, domain: str, record_type: str, result: ProviderResult) -> None:
"""
Query a specific type of DNS record for the domain and add results to ProviderResult.
Args:
domain: Domain to query
record_type: DNS record type (A, AAAA, CNAME, etc.)
result: ProviderResult to populate
UPDATED: Query DNS records with minimal formatting - keep raw values.
"""
try:
self.total_requests += 1
@ -180,13 +175,14 @@ class DNSProvider(BaseProvider):
elif record_type == 'SOA':
target = str(record.mname).rstrip('.')
elif record_type in ['TXT']:
# TXT records are treated as attributes, not relationships
# UPDATED: Keep raw TXT record value
txt_value = str(record).strip('"')
dns_records.append(f"TXT: {txt_value}")
continue
elif record_type == 'SRV':
target = str(record.target).rstrip('.')
elif record_type == 'CAA':
# UPDATED: Keep raw CAA record format
caa_value = f"{record.flags} {record.tag.decode('utf-8')} \"{record.value.decode('utf-8')}\""
dns_records.append(f"CAA: {caa_value}")
continue
@ -200,8 +196,8 @@ class DNSProvider(BaseProvider):
'value': target,
'ttl': response.ttl
}
relationship_type = f"{record_type.lower()}_record"
confidence = 0.8 # Standard confidence for DNS records
relationship_type = f"{record_type.lower()}_record" # Raw relationship type
confidence = 0.8
# Add relationship
result.add_relationship(
@ -213,7 +209,7 @@ class DNSProvider(BaseProvider):
raw_data=raw_data
)
# Add DNS record as attribute to the source domain
# UPDATED: Keep raw DNS record format
dns_records.append(f"{record_type}: {target}")
# Log relationship discovery
@ -226,7 +222,7 @@ class DNSProvider(BaseProvider):
discovery_method=f"dns_{record_type.lower()}_record"
)
# Add DNS records as a consolidated attribute
# Add DNS records as a consolidated attribute (raw format)
if dns_records:
result.add_attribute(
target_node=domain,
@ -241,5 +237,5 @@ class DNSProvider(BaseProvider):
except Exception as e:
self.failed_requests += 1
self.logger.logger.debug(f"{record_type} record query failed for {domain}: {e}")
# Re-raise the exception so the scanner can handle it
raise e

View File

@ -211,14 +211,7 @@ class ShodanProvider(BaseProvider):
def _process_shodan_data(self, ip: str, data: Dict[str, Any]) -> ProviderResult:
"""
Process Shodan data to extract relationships and attributes.
Args:
ip: IP address queried
data: Raw Shodan response data
Returns:
ProviderResult with relationships and attributes
UPDATED: Process Shodan data with raw attribute names and values.
"""
result = ProviderResult()
@ -271,9 +264,10 @@ class ShodanProvider(BaseProvider):
confidence=0.9
)
elif isinstance(value, (str, int, float, bool)) and value is not None:
# UPDATED: Keep raw Shodan field names (no "shodan_" prefix)
result.add_attribute(
target_node=ip,
name=f"shodan_{key}",
name=key, # Raw field name from Shodan API
value=value,
attr_type='shodan_info',
provider=self.name,

View File

@ -382,7 +382,6 @@ class GraphManager {
graphData.nodes.forEach(node => {
if (node.type === 'large_entity' && node.attributes) {
// UPDATED: Handle unified data model - look for 'nodes' attribute in the attributes list
const nodesAttribute = this.findAttributeByName(node.attributes, 'nodes');
if (nodesAttribute && Array.isArray(nodesAttribute.value)) {
nodesAttribute.value.forEach(nodeId => {
@ -394,15 +393,30 @@ class GraphManager {
});
const filteredNodes = graphData.nodes.filter(node => {
// Only include nodes that are NOT members of large entities, but always include the container itself
return !this.largeEntityMembers.has(node.id) || node.type === 'large_entity';
});
console.log(`Filtered ${graphData.nodes.length - filteredNodes.length} large entity member nodes from visualization`);
// Process only the filtered nodes
// FIXED: Process nodes with proper certificate coloring
const processedNodes = filteredNodes.map(node => {
return this.processNode(node);
const processed = this.processNode(node);
// FIXED: Apply certificate-based coloring here in frontend
if (node.type === 'domain' && Array.isArray(node.attributes)) {
const certInfo = this.analyzeCertificateInfo(node.attributes);
if (certInfo.hasExpiredOnly) {
// Red for domains with only expired/invalid certificates
processed.color = { background: '#ff6b6b', border: '#cc5555' };
} else if (!certInfo.hasCertificates) {
// Grey for domains with no certificates
processed.color = { background: '#c7c7c7', border: '#999999' };
}
// Valid certificates use default green (handled by processNode)
}
return processed;
});
const mergedEdges = {};
@ -439,24 +453,19 @@ class GraphManager {
const existingNodeIds = this.nodes.getIds();
const existingEdgeIds = this.edges.getIds();
// Add new nodes with fade-in animation
const newNodes = processedNodes.filter(node => !existingNodeIds.includes(node.id));
const newEdges = processedEdges.filter(edge => !existingEdgeIds.includes(edge.id));
// Update existing data
this.nodes.update(processedNodes);
this.edges.update(processedEdges);
// After data is loaded, apply filters
this.updateFilterControls();
this.applyAllFilters();
// Highlight new additions briefly
if (newNodes.length > 0 || newEdges.length > 0) {
setTimeout(() => this.highlightNewElements(newNodes, newEdges), 100);
}
// Auto-fit view for small graphs or first update
if (processedNodes.length <= 10 || existingNodeIds.length === 0) {
setTimeout(() => this.fitView(), 800);
}
@ -470,6 +479,46 @@ class GraphManager {
}
}
analyzeCertificateInfo(attributes) {
let hasCertificates = false;
let hasValidCertificates = false;
let hasExpiredCertificates = false;
for (const attr of attributes) {
const attrName = (attr.name || '').toLowerCase();
const attrProvider = (attr.provider || '').toLowerCase();
const attrValue = attr.value;
// Look for certificate attributes from crtsh provider
if (attrProvider === 'crtsh' || attrName.startsWith('cert_')) {
hasCertificates = true;
// Check certificate validity using raw attribute names
if (attrName === 'cert_is_currently_valid') {
if (attrValue === true) {
hasValidCertificates = true;
} else if (attrValue === false) {
hasExpiredCertificates = true;
}
}
// Check for expiry indicators
else if (attrName === 'cert_expires_soon' && attrValue === true) {
hasExpiredCertificates = true;
}
else if (attrName.includes('expired') && attrValue === true) {
hasExpiredCertificates = true;
}
}
}
return {
hasCertificates,
hasValidCertificates,
hasExpiredCertificates,
hasExpiredOnly: hasExpiredCertificates && !hasValidCertificates
};
}
/**
* UPDATED: Helper method to find an attribute by name in the standardized attributes list
* @param {Array} attributes - List of StandardAttribute objects
@ -496,7 +545,7 @@ class GraphManager {
size: this.getNodeSize(node.type),
borderColor: this.getNodeBorderColor(node.type),
shape: this.getNodeShape(node.type),
attributes: node.attributes || [], // Keep as standardized attributes list
attributes: node.attributes || [],
description: node.description || '',
metadata: node.metadata || {},
type: node.type,
@ -509,19 +558,33 @@ class GraphManager {
processedNode.borderWidth = Math.max(2, Math.floor(node.confidence * 5));
}
// Handle merged correlation objects (similar to large entities)
// FIXED: Certificate-based domain coloring
if (node.type === 'domain' && Array.isArray(node.attributes)) {
const certInfo = this.analyzeCertificateInfo(node.attributes);
if (certInfo.hasExpiredOnly) {
// Red for domains with only expired/invalid certificates
processedNode.color = '#ff6b6b';
processedNode.borderColor = '#cc5555';
} else if (!certInfo.hasCertificates) {
// Grey for domains with no certificates
processedNode.color = '#c7c7c7';
processedNode.borderColor = '#999999';
}
// Green for valid certificates (default color)
}
// Handle merged correlation objects
if (node.type === 'correlation_object') {
const metadata = node.metadata || {};
const values = metadata.values || [];
const mergeCount = metadata.merge_count || 1;
if (mergeCount > 1) {
// Display as merged correlation container
processedNode.label = `Correlations (${mergeCount})`;
processedNode.title = `Merged correlation container with ${mergeCount} values: ${values.slice(0, 3).join(', ')}${values.length > 3 ? '...' : ''}`;
processedNode.borderWidth = 3; // Thicker border for merged nodes
processedNode.borderWidth = 3;
} else {
// Single correlation value
const value = Array.isArray(values) && values.length > 0 ? values[0] : (metadata.value || 'Unknown');
const displayValue = typeof value === 'string' && value.length > 20 ? value.substring(0, 17) + '...' : value;
processedNode.label = `${displayValue}`;
@ -532,6 +595,7 @@ class GraphManager {
return processedNode;
}
/**
* Process edge data with styling and metadata
* @param {Object} edge - Raw edge data

View File

@ -868,20 +868,14 @@ class DNSReconApp {
return html;
}
/**
* NEW: Organized attributes section with provider/semantic grouping (no formatting)
*/
generateOrganizedAttributesSection(attributes, nodeType) {
if (!Array.isArray(attributes) || attributes.length === 0) {
return '';
}
// Group attributes intelligently
const groups = this.groupAttributesByProviderAndType(attributes, nodeType);
let html = '';
// Sort groups by priority
const sortedGroups = Object.entries(groups).sort((a, b) => {
const priorityOrder = { 'high': 0, 'medium': 1, 'low': 2 };
return priorityOrder[a[1].priority] - priorityOrder[b[1].priority];
@ -904,22 +898,10 @@ class DNSReconApp {
`;
groupData.attributes.forEach(attr => {
// Format the value appropriately
let displayValue = '';
if (attr.value === null || attr.value === undefined) {
displayValue = 'N/A';
} else if (Array.isArray(attr.value)) {
displayValue = attr.value.length > 0 ? `Array (${attr.value.length} items)` : 'Empty Array';
} else if (typeof attr.value === 'object') {
displayValue = 'Object';
} else {
displayValue = String(attr.value);
}
html += `
<div class="attribute-item-compact">
<span class="attribute-key-compact">${this.escapeHtml(attr.name || 'Unknown')}</span>
<span class="attribute-value-compact">${this.escapeHtml(displayValue)}</span>
<span class="attribute-value-compact">${this.formatAttributeValue(attr)}</span>
</div>
`;
});
@ -930,12 +912,49 @@ class DNSReconApp {
return html;
}
/**
* NEW: Group attributes by provider and semantic meaning (no formatting)
*/
formatAttributeValue(attr) {
const value = attr.value;
const name = attr.name || '';
if (value === null || value === undefined) {
return 'N/A';
}
if (Array.isArray(value)) {
if (value.length === 0) {
return 'Empty Array';
}
// Special handling for DNS records and similar arrays
if (name === 'dns_records' || name.includes('record') || name.includes('hostname')) {
// Show DNS records as a readable list
if (value.length <= 5) {
return this.escapeHtml(value.join(', '));
} else {
const preview = value.slice(0, 3).join(', ');
return this.escapeHtml(`${preview} ... (+${value.length - 3} more)`);
}
}
// For other arrays
if (value.length <= 3) {
return this.escapeHtml(value.join(', '));
} else {
const preview = value.slice(0, 2).join(', ');
return this.escapeHtml(`${preview} ... (${value.length} total)`);
}
}
if (typeof value === 'object') {
return 'Object';
}
return this.escapeHtml(String(value));
}
groupAttributesByProviderAndType(attributes, nodeType) {
const groups = {
'DNS Records': { icon: '🔍', priority: 'high', attributes: [] },
'DNS Records': { icon: '📋', priority: 'high', attributes: [] },
'Certificate Information': { icon: '🔒', priority: 'high', attributes: [] },
'Network Information': { icon: '🌐', priority: 'high', attributes: [] },
'Provider Data': { icon: '📊', priority: 'medium', attributes: [] },
@ -943,23 +962,29 @@ class DNSReconApp {
};
for (const attr of attributes) {
const provider = attr.provider?.toLowerCase() || '';
const name = attr.name?.toLowerCase() || '';
const provider = (attr.provider || '').toLowerCase();
const name = (attr.name || '').toLowerCase();
const type = (attr.type || '').toLowerCase();
let assigned = false;
// DNS-related attributes
if (provider === 'dns' || ['dns', 'record', 'ptr', 'mx', 'cname', 'ns', 'txt', 'soa'].some(keyword => name.includes(keyword))) {
// DNS-related attributes (better detection)
if (provider === 'dns' ||
name === 'dns_records' ||
name.includes('record') ||
['ptr', 'mx', 'cname', 'ns', 'txt', 'soa'].some(keyword => name.includes(keyword))) {
groups['DNS Records'].attributes.push(attr);
assigned = true;
}
// Certificate-related attributes
else if (provider === 'crtsh' || ['cert', 'certificate', 'ssl', 'tls', 'issuer', 'validity', 'san'].some(keyword => name.includes(keyword))) {
else if (provider === 'crtsh' || name.startsWith('cert_') ||
['certificate', 'ssl', 'tls', 'issuer', 'validity', 'san'].some(keyword => name.includes(keyword))) {
groups['Certificate Information'].attributes.push(attr);
assigned = true;
}
// Network/Shodan attributes
else if (provider === 'shodan' || ['port', 'service', 'banner', 'asn', 'organization', 'country', 'city', 'network'].some(keyword => name.includes(keyword))) {
else if (provider === 'shodan' ||
['port', 'service', 'banner', 'asn', 'organization', 'country', 'city', 'network'].some(keyword => name.includes(keyword))) {
groups['Network Information'].attributes.push(attr);
assigned = true;
}
@ -985,6 +1010,33 @@ class DNSReconApp {
return groups;
}
formatEdgeLabel(relationshipType, confidence) {
if (!relationshipType) return '';
// UPDATED: No formatting of relationship type - use raw values
const confidenceText = confidence >= 0.8 ? '●' : confidence >= 0.6 ? '◐' : '○';
return `${relationshipType} ${confidenceText}`;
}
createEdgeTooltip(edge) {
let tooltip = `<div style="font-family: 'Roboto Mono', monospace; font-size: 11px;">`;
tooltip += `<div style="color: #00ff41; font-weight: bold; margin-bottom: 4px;">${edge.label || 'Relationship'}</div>`;
tooltip += `<div style="color: #999; margin-bottom: 2px;">Confidence: ${(edge.confidence_score * 100).toFixed(1)}%</div>`;
// UPDATED: Use raw provider name (no formatting)
if (edge.source_provider) {
tooltip += `<div style="color: #999; margin-bottom: 2px;">Provider: ${edge.source_provider}</div>`;
}
if (edge.discovery_timestamp) {
const date = new Date(edge.discovery_timestamp);
tooltip += `<div style="color: #666; font-size: 10px;">Discovered: ${date.toLocaleString()}</div>`;
}
tooltip += `</div>`;
return tooltip;
}
/**
* UPDATED: Enhanced correlation details showing the correlated attribute clearly (no formatting)
*/
@ -996,7 +1048,7 @@ class DNSReconApp {
let html = '';
// Show what attribute is being correlated
// Show what attribute is being correlated (raw names)
const primarySource = metadata.primary_source || 'unknown';
html += `
@ -1013,7 +1065,7 @@ class DNSReconApp {
<span class="attribute-value-compact"><code>${this.escapeHtml(String(value))}</code></span>
</div>
<div class="attribute-item-compact">
<span class="attribute-key-compact">Attribute Type</span>
<span class="attribute-key-compact">Attribute Source</span>
<span class="attribute-value-compact">${primarySource}</span>
</div>
<div class="attribute-item-compact">
@ -1050,10 +1102,6 @@ class DNSReconApp {
return html;
}
/**
* UPDATED: Generate large entity details using unified data model
*/