phase 8
This commit is contained in:
alorig
@@ -97,5 +97,237 @@ class LinkerService:
|
||||
continue
|
||||
|
||||
return results
|
||||
|
||||
def process_product(self, content_id: int) -> Content:
|
||||
"""
|
||||
Process product content for linking (Phase 8).
|
||||
Enhanced linking for products: links to related products, categories, and service pages.
|
||||
|
||||
Args:
|
||||
content_id: Content ID to process (must be entity_type='product')
|
||||
|
||||
Returns:
|
||||
Updated Content instance
|
||||
"""
|
||||
try:
|
||||
content = Content.objects.get(id=content_id, entity_type='product')
|
||||
except Content.DoesNotExist:
|
||||
raise ValueError(f"Product content with id {content_id} does not exist")
|
||||
|
||||
# Use base process but with product-specific candidate finding
|
||||
account = content.account
|
||||
|
||||
# Check credits
|
||||
try:
|
||||
self.credit_service.check_credits(account, 'linking')
|
||||
except InsufficientCreditsError:
|
||||
raise
|
||||
|
||||
# Find product-specific link candidates (related products, categories, services)
|
||||
candidates = self._find_product_candidates(content)
|
||||
|
||||
if not candidates:
|
||||
logger.info(f"No link candidates found for product content {content_id}")
|
||||
return content
|
||||
|
||||
# Inject links
|
||||
result = self.injection_engine.inject_links(content, candidates)
|
||||
|
||||
# Update content
|
||||
content.html_content = result['html_content']
|
||||
content.internal_links = result['links']
|
||||
content.linker_version += 1
|
||||
content.save(update_fields=['html_content', 'internal_links', 'linker_version'])
|
||||
|
||||
# Deduct credits
|
||||
self.credit_service.deduct_credits_for_operation(
|
||||
account=account,
|
||||
operation_type='linking',
|
||||
description=f"Product linking for: {content.title or 'Untitled'}",
|
||||
related_object_type='content',
|
||||
related_object_id=content.id
|
||||
)
|
||||
|
||||
logger.info(f"Linked product content {content_id}: {result['links_added']} links added")
|
||||
|
||||
return content
|
||||
|
||||
def process_taxonomy(self, content_id: int) -> Content:
|
||||
"""
|
||||
Process taxonomy content for linking (Phase 8).
|
||||
Enhanced linking for taxonomies: links to related categories, tags, and content.
|
||||
|
||||
Args:
|
||||
content_id: Content ID to process (must be entity_type='taxonomy')
|
||||
|
||||
Returns:
|
||||
Updated Content instance
|
||||
"""
|
||||
try:
|
||||
content = Content.objects.get(id=content_id, entity_type='taxonomy')
|
||||
except Content.DoesNotExist:
|
||||
raise ValueError(f"Taxonomy content with id {content_id} does not exist")
|
||||
|
||||
# Use base process but with taxonomy-specific candidate finding
|
||||
account = content.account
|
||||
|
||||
# Check credits
|
||||
try:
|
||||
self.credit_service.check_credits(account, 'linking')
|
||||
except InsufficientCreditsError:
|
||||
raise
|
||||
|
||||
# Find taxonomy-specific link candidates (related taxonomies, categories, content)
|
||||
candidates = self._find_taxonomy_candidates(content)
|
||||
|
||||
if not candidates:
|
||||
logger.info(f"No link candidates found for taxonomy content {content_id}")
|
||||
return content
|
||||
|
||||
# Inject links
|
||||
result = self.injection_engine.inject_links(content, candidates)
|
||||
|
||||
# Update content
|
||||
content.html_content = result['html_content']
|
||||
content.internal_links = result['links']
|
||||
content.linker_version += 1
|
||||
content.save(update_fields=['html_content', 'internal_links', 'linker_version'])
|
||||
|
||||
# Deduct credits
|
||||
self.credit_service.deduct_credits_for_operation(
|
||||
account=account,
|
||||
operation_type='linking',
|
||||
description=f"Taxonomy linking for: {content.title or 'Untitled'}",
|
||||
related_object_type='content',
|
||||
related_object_id=content.id
|
||||
)
|
||||
|
||||
logger.info(f"Linked taxonomy content {content_id}: {result['links_added']} links added")
|
||||
|
||||
return content
|
||||
|
||||
def _find_product_candidates(self, content: Content) -> List[dict]:
|
||||
"""
|
||||
Find link candidates specific to product content.
|
||||
|
||||
Args:
|
||||
content: Product Content instance
|
||||
|
||||
Returns:
|
||||
List of candidate dicts
|
||||
"""
|
||||
candidates = []
|
||||
|
||||
# Find related products (same category, similar features)
|
||||
related_products = Content.objects.filter(
|
||||
account=content.account,
|
||||
site=content.site,
|
||||
sector=content.sector,
|
||||
entity_type='product',
|
||||
status__in=['draft', 'review', 'publish']
|
||||
).exclude(id=content.id)
|
||||
|
||||
# Use structure_data to find products with similar categories/features
|
||||
if content.structure_data:
|
||||
product_type = content.structure_data.get('product_type')
|
||||
if product_type:
|
||||
related_products = related_products.filter(
|
||||
structure_data__product_type=product_type
|
||||
)
|
||||
|
||||
# Add product candidates
|
||||
for product in related_products[:5]: # Limit to 5 related products
|
||||
candidates.append({
|
||||
'content_id': product.id,
|
||||
'title': product.title or 'Untitled Product',
|
||||
'url': f'/products/{product.id}', # Placeholder URL
|
||||
'relevance_score': 0.8,
|
||||
'anchor_text': product.title or 'Related Product'
|
||||
})
|
||||
|
||||
# Find related service pages
|
||||
related_services = Content.objects.filter(
|
||||
account=content.account,
|
||||
site=content.site,
|
||||
sector=content.sector,
|
||||
entity_type='service',
|
||||
status__in=['draft', 'review', 'publish']
|
||||
)[:3] # Limit to 3 related services
|
||||
|
||||
for service in related_services:
|
||||
candidates.append({
|
||||
'content_id': service.id,
|
||||
'title': service.title or 'Untitled Service',
|
||||
'url': f'/services/{service.id}', # Placeholder URL
|
||||
'relevance_score': 0.6,
|
||||
'anchor_text': service.title or 'Related Service'
|
||||
})
|
||||
|
||||
# Use base candidate engine for additional candidates
|
||||
base_candidates = self.candidate_engine.find_candidates(content, max_candidates=5)
|
||||
candidates.extend(base_candidates)
|
||||
|
||||
return candidates
|
||||
|
||||
def _find_taxonomy_candidates(self, content: Content) -> List[dict]:
|
||||
"""
|
||||
Find link candidates specific to taxonomy content.
|
||||
|
||||
Args:
|
||||
content: Taxonomy Content instance
|
||||
|
||||
Returns:
|
||||
List of candidate dicts
|
||||
"""
|
||||
candidates = []
|
||||
|
||||
# Find related taxonomies
|
||||
related_taxonomies = Content.objects.filter(
|
||||
account=content.account,
|
||||
site=content.site,
|
||||
sector=content.sector,
|
||||
entity_type='taxonomy',
|
||||
status__in=['draft', 'review', 'publish']
|
||||
).exclude(id=content.id)[:5] # Limit to 5 related taxonomies
|
||||
|
||||
for taxonomy in related_taxonomies:
|
||||
candidates.append({
|
||||
'content_id': taxonomy.id,
|
||||
'title': taxonomy.title or 'Untitled Taxonomy',
|
||||
'url': f'/taxonomy/{taxonomy.id}', # Placeholder URL
|
||||
'relevance_score': 0.7,
|
||||
'anchor_text': taxonomy.title or 'Related Taxonomy'
|
||||
})
|
||||
|
||||
# Find content in this taxonomy (using json_blocks categories/tags)
|
||||
if content.json_blocks:
|
||||
for block in content.json_blocks:
|
||||
if block.get('type') == 'categories':
|
||||
categories = block.get('items', [])
|
||||
for category in categories[:3]: # Limit to 3 categories
|
||||
category_name = category.get('name', '')
|
||||
if category_name:
|
||||
related_content = Content.objects.filter(
|
||||
account=content.account,
|
||||
site=content.site,
|
||||
sector=content.sector,
|
||||
categories__icontains=category_name,
|
||||
status__in=['draft', 'review', 'publish']
|
||||
).exclude(id=content.id)[:3]
|
||||
|
||||
for related in related_content:
|
||||
candidates.append({
|
||||
'content_id': related.id,
|
||||
'title': related.title or 'Untitled',
|
||||
'url': f'/content/{related.id}', # Placeholder URL
|
||||
'relevance_score': 0.6,
|
||||
'anchor_text': related.title or 'Related Content'
|
||||
})
|
||||
|
||||
# Use base candidate engine for additional candidates
|
||||
base_candidates = self.candidate_engine.find_candidates(content, max_candidates=5)
|
||||
candidates.extend(base_candidates)
|
||||
|
||||
return candidates
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user