commit 5b7122a2a1533c433659d18e1a668b1938aa36b7 · bretton.dev/coves

+6 -3

aggregators/kagi-news/src/html_parser.py

···

       78
       78
        
                   Perspective(

     

       79
       79
        
                       actor=p['actor'],

     

       80
       80
        
                       description=p['description'],

     

       81
       81
       -
                       source_url=p['source_url']

     

       81
       81
       +
                       source_url=p['source_url'],

     

       82
       82
       +
                       source_name=p.get('source_name', '')

     

       82
       83
        
                   )

     

       83
       84
        
                   for p in parsed['perspectives']

     

       84
       85
        
               ]

     
···

       230
       231
        
               actor, rest = full_text.split(':', 1)

     

       231
       232
        
               actor = actor.strip()

     

       232
       233
        
       

     

       233
       233
       -
               # Find the <a> tag for source URL

     

       234
       234
       +
               # Find the <a> tag for source URL and name

     

       234
       235
        
               a_tag = li.find('a')

     

       235
       236
        
               source_url = a_tag['href'] if a_tag and a_tag.get('href') else ""

     

       237
       237
       +
               source_name = a_tag.get_text(strip=True) if a_tag else ""

     

       236
       238
        
       

     

       237
       239
        
               # Extract description (between colon and source link)

     

       238
       240
        
               # Remove the source citation part in parentheses

     
···

       250
       252
        
               return {

     

       251
       253
        
                   'actor': actor,

     

       252
       254
        
                   'description': description,

     

       253
       253
       -
                   'source_url': source_url

     

       255
       255
       +
                   'source_url': source_url,

     

       256
       256
       +
                   'source_name': source_name

     

       254
       257
        
               }

     

       255
       258
        
       

     

       256
       259
        
           def _extract_sources(self, soup: BeautifulSoup) -> List[Dict]:

+1

aggregators/kagi-news/src/models.py

···

       20
       20
        
           actor: str

     

       21
       21
        
           description: str

     

       22
       22
        
           source_url: str

     

       23
       23
       +
           source_name: str = ""  # Name of the source (e.g., "The Straits Times")

     

       23
       24
        
       

     

       24
       25
        
       

     

       25
       26
        
       @dataclass

+11 -7

aggregators/kagi-news/src/richtext_formatter.py

···

       42
       42
        
                   builder.add_bold("Highlights:")

     

       43
       43
        
                   builder.add_text("\n")

     

       44
       44
        
                   for highlight in story.highlights:

     

       45
       45
       -
                       builder.add_text(f"• {highlight}\n")

     

       45
       45
       +
                       builder.add_text(f"• {highlight}\n\n")

     

       46
       46
        
                   builder.add_text("\n")

     

       47
       47
        
       

     

       48
       48
        
               # Perspectives (if present)

     
···

       53
       53
        
                       # Bold the actor name

     

       54
       54
        
                       actor_with_colon = f"{perspective.actor}:"

     

       55
       55
        
                       builder.add_bold(actor_with_colon)

     

       56
       56
       -
                       builder.add_text(f" {perspective.description} (")

     

       56
       56
       +
                       builder.add_text(f" {perspective.description}")

     

       57
       57
        
       

     

       58
       58
       -
                       # Add link to source

     

       59
       59
       -
                       source_link_text = "Source"

     

       60
       60
       -
                       builder.add_link(source_link_text, perspective.source_url)

     

       61
       61
       -
                       builder.add_text(")\n")

     

       58
       58
       +
                       # Add link to source if available

     

       59
       59
       +
                       if perspective.source_url:

     

       60
       60
       +
                           builder.add_text(" (")

     

       61
       61
       +
                           source_link_text = perspective.source_name if perspective.source_name else "Source"

     

       62
       62
       +
                           builder.add_link(source_link_text, perspective.source_url)

     

       63
       63
       +
                           builder.add_text(")")

     

       64
       64
       +
       

     

       65
       65
       +
                       builder.add_text("\n\n")

     

       62
       66
        
                   builder.add_text("\n")

     

       63
       67
        
       

     

       64
       68
        
               # Quote (if present)

     
···

       74
       78
        
                   for source in story.sources:

     

       75
       79
        
                       builder.add_text("• ")

     

       76
       80
        
                       builder.add_link(source.title, source.url)

     

       77
       77
       -
                       builder.add_text(f" - {source.domain}\n")

     

       81
       81
       +
                       builder.add_text(f" - {source.domain}\n\n")

     

       78
       82
        
                   builder.add_text("\n")

     

       79
       83
        
       

     

       80
       84
        
               # Kagi News attribution