Function dump

beanhub_cli/inbox/main.py:380–490

(
    env: Environment,
    repo: str | None,
    config: str | None,
    workdir: str,
    unsafe_tar_extract: bool,
)

@pass_env
@handle_api_exception(logger)
def dump(
    env: Environment,
    repo: str | None,
    config: str | None,
    workdir: str,
    unsafe_tar_extract: bool,
):
    workdir_path = pathlib.Path(workdir)
    inbox_doc = load_inbox_doc(config)

    if not hasattr(tarfile, "data_filter") and not unsafe_tar_extract:
        logger.error(
            "You need to use Python >= 3.11 in order to safely unpack the downloaded tar file, or you need to pass "
            "in --unsafe-tar-extract argument to allow unsafe tar file extracting"
        )
        sys.exit(-1)
    config = ensure_auth_config(api_base_url=env.api_base_url, repo=repo)

    inbox_emails = fetch_all_emails(env=env, config=config)
    missing_email_output_files = list(
        compute_missing_emails(
            inbox_doc=inbox_doc, inbox_emails=inbox_emails, workdir_path=workdir_path
        )
    )
    if not missing_email_output_files:
        logger.info("No missing emails found, no need to update")
        return

    private_key = PrivateKey.generate()
    public_key = private_key.public_key.encode(URLSafeBase64Encoder).decode("ascii")

    with make_auth_client(base_url=env.api_base_url, token=config.token) as client:
        client.raise_on_unexpected_status = True
        # TODO: it's a bit slow to download all the emails in one request if there are tons of emails.
        #       maybe it makes more sense to break it down to smaller requests and stream line it?
        resp: CreateInboxDumpRequestResponse = create_inbox_dump_request.sync(
            body=CreateInboxDumpRequest(
                public_key=public_key,
                email_ids=[
                    inbox_email.id for inbox_email, _ in missing_email_output_files
                ],
            ),
            username=config.username,
            repo_name=config.repo,
            client=client,
        )
        dump_id = resp.id
        logger.info(
            "Created dump [green]%s[/] with public_key [green]%s[/], email_count=[green]%s[/], workdir=[green]%s[/], waiting for updates ...",
            dump_id,
            public_key,
            len(missing_email_output_files),
            workdir_path,
            extra={"markup": True, "highlighter": None},
        )

        while True:
            time.sleep(5)