@ -1,22 +1,22 @@
import os import os
import sys import sys
import re import re
import json
from openai import OpenAI from openai import OpenAI
client = OpenAI ( api_key = os . environ [ ' OPENAI_API_KEY ' ] ) client = OpenAI ( api_key = os . environ [ ' OPENAI_API_KEY ' ] )
# Regex patterns as constants
SEO_BLOCK_PATTERN = r ' ```+json \ s*// \ [doc-seo \ ] \ s*( \ { .*? \ }) \ s*```+ '
SEO_BLOCK_WITH_BACKTICKS_PATTERN = r ' (```+)json \ s*// \ [doc-seo \ ] \ s*( \ { .*? \ }) \ s* \ 1 '
def has_seo_description ( content ) : def has_seo_description ( content ) :
""" Check if content already has SEO description with Description field """ """ Check if content already has SEO description with Description field """
import json match = re . search ( SEO_BLOCK_PATTERN , content , flags = re . DOTALL )
# Match SEO description block with 3 or more backticks
pattern = r ' ```+json \ s*// \ [doc-seo \ ] \ s*( \ { .*? \ }) \ s*```+ '
match = re . search ( pattern , content , flags = re . DOTALL )
if not match : if not match :
return False return False
# Check if Description field exists and is not empty
try : try :
json_str = match . group ( 1 ) json_str = match . group ( 1 )
seo_data = json . loads ( json_str ) seo_data = json . loads ( json_str )
@ -24,24 +24,42 @@ def has_seo_description(content):
except json . JSONDecodeError : except json . JSONDecodeError :
return False return False
def is_content_too_short ( content ) : def has_seo_block ( content ) :
""" Check if content is less than 200 characters """ """ Check if content has any SEO block (with or without Description) """
# Remove SEO tags if present for accurate count return bool ( re . search ( SEO_BLOCK_PATTERN , content , flags = re . DOTALL ) )
# Match SEO description block with 3 or more backticks
clean_content = re . sub ( r ' ```+json \ s*// \ [doc-seo \ ].*?```+ \ s* ' , ' ' , content , flags = re . DOTALL ) def remove_seo_blocks ( content ) :
""" Remove all SEO description blocks from content """
return len ( clean_content . strip ( ) ) < 200 return re . sub ( SEO_BLOCK_PATTERN + r ' \ s* ' , ' ' , content , flags = re . DOTALL )
def is_content_too_short ( content , min_length = 200 ) :
""" Check if content is less than minimum length (excluding SEO blocks) """
clean_content = remove_seo_blocks ( content )
return len ( clean_content . strip ( ) ) < min_length
def get_content_preview ( content , max_length = 1000 ) : def get_content_preview ( content , max_length = 1000 ) :
""" Get preview of content for OpenAI """ """ Get preview of content for OpenAI (excluding SEO blocks) """
# Remove existing SEO tags if present clean_content = remove_seo_blocks ( content )
# Match SEO description block with 3 or more backticks
clean_content = re . sub ( r ' ```+json \ s*// \ [doc-seo \ ].*?```+ \ s* ' , ' ' , content , flags = re . DOTALL )
return clean_content [ : max_length ] . strip ( ) return clean_content [ : max_length ] . strip ( )
def escape_json_string ( text ) :
""" Escape special characters for JSON """
return text . replace ( ' \\ ' , ' \\ \\ ' ) . replace ( ' " ' , ' \\ " ' ) . replace ( ' \n ' , ' \\ n ' )
def create_seo_block ( description ) :
""" Create a new SEO block with the given description """
escaped_desc = escape_json_string ( description )
return f ''' ```json
/ / [ doc - seo ]
{ {
" Description " : " {escaped_desc} "
} }
` ` `
'''
def generate_description ( content , filename ) : def generate_description ( content , filename ) :
""" Generate SEO description using OpenAI with system prompt from OpenAIService.cs """ """ Generate SEO description using OpenAI """
try : try :
preview = get_content_preview ( content ) preview = get_content_preview ( content )
@ -69,154 +87,169 @@ Generate only the description text, nothing else:"""}
) )
description = response . choices [ 0 ] . message . content . strip ( ) description = response . choices [ 0 ] . message . content . strip ( )
return description return description
except Exception as e : except Exception as e :
print ( f " ❌ Error generating description: { e } " ) print ( f " ❌ Error generating description: { e } " )
return f " Learn about { os . path . splitext ( filename ) [ 0 ] } in ABP Framework documentation. " return f " Learn about { os . path . splitext ( filename ) [ 0 ] } in ABP Framework documentation. "
def add _seo_description( content , description ) : def update _seo_description( content , description ) :
""" Add or update SEO description in content """ """ Update existing SEO block with new description """
import json match = re . search ( SEO_BLOCK_WITH_BACKTICKS_PATTERN , content , flags = re . DOTALL )
# Escape special characters for JSON if not match :
escaped_desc = description . replace ( ' \\ ' , ' \\ \\ ' ) . replace ( ' " ' , ' \\ " ' ) . replace ( ' \n ' , ' \\ n ' ) return None
# Check if SEO block already exists backticks = match . group ( 1 )
pattern = r ' (```+)json \ s*// \ [doc-seo \ ] \ s*( \ { .*? \ }) \ s* \ 1 ' json_str = match . group ( 2 )
match = re . search ( pattern , content , flags = re . DOTALL )
if match : try :
# SEO block exists, update Description field seo_data = json . loads ( json_str )
backticks = match . group ( 1 ) seo_data [ ' Description ' ] = description
json_str = match . group ( 2 ) updated_json = json . dumps ( seo_data , indent = 4 , ensure_ascii = False )
try : new_block = f ''' { backticks } json
# Parse existing JSON
seo_data = json . loads ( json_str )
# Update Description
seo_data [ ' Description ' ] = description
# Convert back to formatted JSON
updated_json = json . dumps ( seo_data , indent = 4 , ensure_ascii = False )
# Replace the old block with updated one
new_block = f ''' { backticks } json
/ / [ doc - seo ] / / [ doc - seo ]
{ updated_json } { updated_json }
{ backticks } ''' { backticks } '''
return re . sub ( pattern , new_block , content , count = 1 , flags = re . DOTALL ) return re . sub ( SEO_BLOCK_WITH_BACKTICKS_PATTERN , new_block , content , count = 1 , flags = re . DOTALL )
except json . JSONDecodeError : except json . JSONDecodeError :
# If JSON is invalid, replace the whole block return None
pass
# No existing block or invalid JSON, add new block at the beginning
seo_tag = f ''' ```json
/ / [ doc - seo ]
{ {
" Description " : " {escaped_desc} "
} }
` ` `
''' def add_seo_description ( content , description ) :
return seo_tag + content """ Add or update SEO description in content """
# Try to update existing block first
updated_content = update_seo_description ( content , description )
if updated_content :
return updated_content
# No existing block or update failed, add new block at the beginning
return create_seo_block ( description ) + content
def is_file_ignored ( filepath , ignored_folders ) : def is_file_ignored ( filepath , ignored_folders ) :
""" Check if file is in an ignored folder """ """ Check if file is in an ignored folder """
path_parts = filepath . split ( ' / ' ) path_parts = filepath . split ( ' / ' )
for ignored in ignored_folders : return any ( ignored in path_parts for ignored in ignored_folders )
if ignored in path_parts :
return True def get_changed_files ( ) :
return False """ Get changed files from command line or environment variable """
if len ( sys . argv ) > 1 :
return sys . argv [ 1 : ]
changed_files_str = os . environ . get ( ' CHANGED_FILES ' , ' ' )
return [ f . strip ( ) for f in changed_files_str . strip ( ) . split ( ' \n ' ) if f . strip ( ) ]
def process_file ( filepath , ignored_folders ) :
""" Process a single markdown file. Returns (processed, skipped, skip_reason) """
if not filepath . endswith ( ' .md ' ) :
return False , False , None
# Check if file is in ignored folder
if is_file_ignored ( filepath , ignored_folders ) :
print ( f " 📄 Processing: { filepath } " )
print ( f " 🚫 Skipped (ignored folder) \n " )
return False , True , ' ignored '
print ( f " 📄 Processing: { filepath } " )
try :
# Read file with original line endings
with open ( filepath , ' r ' , encoding = ' utf-8 ' , newline = ' ' ) as f :
content = f . read ( )
# Check if content is too short
if is_content_too_short ( content ) :
print ( f " ⏭️ Skipped (content less than 200 characters) \n " )
return False , True , ' too_short '
# Check if already has SEO description
if has_seo_description ( content ) :
print ( f " ⏭️ Skipped (already has SEO description) \n " )
return False , True , ' has_description '
# Generate description
filename = os . path . basename ( filepath )
print ( f " 🤖 Generating description... " )
description = generate_description ( content , filename )
print ( f " 💡 Generated: { description } " )
# Add or update SEO description
if has_seo_block ( content ) :
print ( f " 🔄 Updating existing SEO block... " )
else :
print ( f " ➕ Adding new SEO block... " )
updated_content = add_seo_description ( content , description )
# Write back (preserving line endings)
with open ( filepath , ' w ' , encoding = ' utf-8 ' , newline = ' ' ) as f :
f . write ( updated_content )
print ( f " ✅ Updated successfully \n " )
return True , False , None
except Exception as e :
print ( f " ❌ Error: { e } \n " )
return False , False , None
def save_statistics ( processed_count , skipped_count , skipped_too_short , skipped_ignored ) :
""" Save processing statistics to file """
try :
with open ( ' /tmp/seo_stats.txt ' , ' w ' ) as f :
f . write ( f " { processed_count } \n { skipped_count } \n { skipped_too_short } \n { skipped_ignored } " )
except Exception as e :
print ( f " ⚠️ Warning: Could not save statistics: { e } " )
def save_updated_files ( updated_files ) :
""" Save list of updated files """
try :
with open ( ' /tmp/seo_updated_files.txt ' , ' w ' ) as f :
f . write ( ' \n ' . join ( updated_files ) )
except Exception as e :
print ( f " ⚠️ Warning: Could not save updated files list: { e } " )
def main ( ) : def main ( ) :
# Ignored folders from GitHub variable (or default values) # Get ignored folders from environment
IGNORED_FOLDERS_STR = os . environ . get ( ' IGNORED_FOLDERS ' , ' Blog-Posts,Community-Articles,_deleted,_resources ' ) IGNORED_FOLDERS_STR = os . environ . get ( ' IGNORED_FOLDERS ' , ' Blog-Posts,Community-Articles,_deleted,_resources ' )
IGNORED_FOLDERS = [ folder . strip ( ) for folder in IGNORED_FOLDERS_STR . split ( ' , ' ) if folder . strip ( ) ] IGNORED_FOLDERS = [ folder . strip ( ) for folder in IGNORED_FOLDERS_STR . split ( ' , ' ) if folder . strip ( ) ]
# Get changed files from environment or command line # Get changed files
if len ( sys . argv ) > 1 : changed_files = get_changed_files ( )
# Files passed as command line arguments
changed_files = sys . argv [ 1 : ]
else :
# Files from environment variable (for GitHub Actions)
changed_files_str = os . environ . get ( ' CHANGED_FILES ' , ' ' )
changed_files = [ f . strip ( ) for f in changed_files_str . strip ( ) . split ( ' \n ' ) if f . strip ( ) ]
# Statistics
processed_count = 0 processed_count = 0
skipped_count = 0 skipped_count = 0
skipped_too_short = 0 skipped_too_short = 0
skipped_ignored = 0 skipped_ignored = 0
updated_files = [ ] # Track actually updated files updated_files = [ ]
print ( " 🤖 Processing changed markdown files... \n " ) print ( " 🤖 Processing changed markdown files... \n " )
print ( f " 🚫 Ignored folders: { ' , ' . join ( IGNORED_FOLDERS ) } \n " ) print ( f " � Ignored folders: { ' , ' . join ( IGNORED_FOLDERS ) } \n " )
# Process each file
for filepath in changed_files : for filepath in changed_files :
if not filepath . endswith ( ' .md ' ) : processed , skipped , skip_reason = process_file ( filepath , IGNORED_FOLDERS )
continue
# Check if file is in ignored folder if processed :
if is_file_ignored ( filepath , IGNORED_FOLDERS ) : processed_count + = 1
print ( f " 📄 Processing: { filepath } " ) updated_files . append ( filepath )
print ( f " 🚫 Skipped (ignored folder) \n " ) elif skipped :
skipped_ignored + = 1
skipped_count + = 1 skipped_count + = 1
continue if skip_reason == ' too_short ' :
print ( f " 📄 Processing: { filepath } " )
try :
# Read file
with open ( filepath , ' r ' , encoding = ' utf-8 ' ) as f :
content = f . read ( )
# Check if content is too short (less than 200 characters)
if is_content_too_short ( content ) :
print ( f " ⏭️ Skipped (content less than 200 characters) \n " )
skipped_too_short + = 1 skipped_too_short + = 1
skipped_count + = 1 elif skip_reason == ' ignored ' :
continue skipped_ignored + = 1
# Check if already has SEO description
if has_seo_description ( content ) :
print ( f " ⏭️ Skipped (already has SEO description) \n " )
skipped_count + = 1
continue
# Generate description
filename = os . path . basename ( filepath )
print ( f " 🤖 Generating description... " )
description = generate_description ( content , filename )
print ( f " 💡 Generated: { description } " )
# Add SEO tag
updated_content = add_seo_description ( content , description )
# Write back
with open ( filepath , ' w ' , encoding = ' utf-8 ' ) as f :
f . write ( updated_content )
print ( f " ✅ Updated successfully \n " )
processed_count + = 1
updated_files . append ( filepath ) # Track this file as updated
except Exception as e :
print ( f " ❌ Error: { e } \n " )
# Print summary
print ( f " \n 📊 Summary: " ) print ( f " \n 📊 Summary: " )
print ( f " ✅ Updated: { processed_count } " ) print ( f " ✅ Updated: { processed_count } " )
print ( f " ⏭️ Skipped (total): { skipped_count } " ) print ( f " ⏭️ Skipped (total): { skipped_count } " )
print ( f " ⏭️ Skipped (too short): { skipped_too_short } " ) print ( f " ⏭️ Skipped (too short): { skipped_too_short } " )
print ( f " 🚫 Skipped (ignored folder): { skipped_ignored } " ) print ( f " 🚫 Skipped (ignored folder): { skipped_ignored } " )
# Save counts and updated files list for next step # Save statistics
with open ( ' /tmp/seo_stats.txt ' , ' w ' ) as f : save_statistics ( processed_count , skipped_count , skipped_too_short , skipped_ignored )
f . write ( f " { processed_count } \n { skipped_count } \n { skipped_too_short } \n { skipped_ignored } " ) save_updated_files ( updated_files )
# Save updated files list
with open ( ' /tmp/seo_updated_files.txt ' , ' w ' ) as f :
f . write ( ' \n ' . join ( updated_files ) )
if __name__ == ' __main__ ' : if __name__ == ' __main__ ' :
main ( ) main ( )
SALİH ÖZKARA
4 months ago