Resolve filenames to GUIDs¶

In order to address MediaFiles by GUIDs, we need to extract the GUID from the SonyCiAsset.name field.

Steps¶

1. Get the list of assets from the DB
2. Filter out any name that does not start with cpb-aacip-
3. Split the name on any of:
   • _ underscore
   • . period
   • -dupe

There are 8 records with -dupe in the name. All other records are correctly filtered with _ and ..

In [3]:

from chowda.db import engine
from chowda.models import SonyCiAsset
from sqlmodel import Session, select
from re import search, split


def get_asset():
    with Session(engine) as session:
        statement = select(SonyCiAsset)
        results = session.exec(statement)

        return [
            (asset.id, split('_|\.|-dupe', asset.name)[0])
            for asset in results.all()
            if search('^cpb-aacip-', asset.name)
        ]


# assets = get_asset()

from chowda.db import engine from chowda.models import SonyCiAsset from sqlmodel import Session, select from re import search, split def get_asset(): with Session(engine) as session: statement = select(SonyCiAsset) results = session.exec(statement) return [ (asset.id, split('_|\.|-dupe', asset.name)[0]) for asset in results.all() if search('^cpb-aacip-', asset.name) ] # assets = get_asset()

Check the list¶

If needed, write the guid list to a file and check it manually.

In [4]:

# len(assets)


def write_assets():
    with open('guids.txt', 'w') as f:
        for asset in assets:
            f.write(f'{asset[1]}\n')

# len(assets) def write_assets(): with open('guids.txt', 'w') as f: for asset in assets: f.write(f'{asset[1]}\n')

Insert MediaFiles¶

Iterate through the list of assets:

• Split the GUID from the filename.
• Search the database for a matching MediaFile object.
  • Insert a new MediaFile object if it does not already exist.
• Find the SonyCiAsset object in the database.
• Link the MediaFile object to the SonyCiAsset object.

In [6]:

from chowda.models import MediaFile


def insert_media_files():
    with Session(engine) as session:
        for asset in assets:
            # Extract the GUID name
            guid = split(r'_|\.|-dupe', asset[1])[0]
            # media_file = session.get(MediaFile, guid)
            media_file = session.exec(
                select(MediaFile).where(MediaFile.guid == guid)
            ).first()
            if not media_file:
                # Create a new MediaFile with the new guid
                media_file = MediaFile(guid=guid)
                # session.add(media_file)
            ci_asset = session.get(SonyCiAsset, asset[0])
            # Add the asset to the existing MediaFile
            media_file.assets.append(ci_asset)
            # asset.media_files.append(media_file)
            session.add(media_file)
        session.commit()

from chowda.models import MediaFile def insert_media_files(): with Session(engine) as session: for asset in assets: # Extract the GUID name guid = split(r'_|\.|-dupe', asset[1])[0] # media_file = session.get(MediaFile, guid) media_file = session.exec( select(MediaFile).where(MediaFile.guid == guid) ).first() if not media_file: # Create a new MediaFile with the new guid media_file = MediaFile(guid=guid) # session.add(media_file) ci_asset = session.get(SonyCiAsset, asset[0]) # Add the asset to the existing MediaFile media_file.assets.append(ci_asset) # asset.media_files.append(media_file) session.add(media_file) session.commit()