adsfolio.pdf.identify

Cached PDF identification workflow.

class adsfolio.pdf.identify.IdentifyWorkflowHelpers(parallel_map_paths, to_sidecar_key, format_elapsed, profile_sync_payload, estimate_ads_queries_for_targets, should_prompt_ads_permission, confirm_ads_queries, progress_reporter_factory, build_progress, parallel_extract_first_page_previews, write_unresolved_csv, print_unresolved_ai_prompt, resolve_from_ai_csv, ads_block_risk_threshold)

Bases: object

Callbacks retained by the PDF CLI while its commands share them.

Parameters:

• parallel_map_paths (collections.abc.Callable[[...], dict[pathlib.Path, Any]])

• to_sidecar_key (collections.abc.Callable[[pathlib.Path, pathlib.Path], str])

• format_elapsed (collections.abc.Callable[[float], str])

• profile_sync_payload (collections.abc.Callable[[Any], dict[str, Any]])

• estimate_ads_queries_for_targets (collections.abc.Callable[[...], dict[str, Any]])

• should_prompt_ads_permission (collections.abc.Callable[[], bool])

• confirm_ads_queries (collections.abc.Callable[[...], tuple[str, int]])

• progress_reporter_factory (collections.abc.Callable[[], Any])

• build_progress (collections.abc.Callable[[...], Any | None])

• parallel_extract_first_page_previews (collections.abc.Callable[[...], dict[pathlib.Path, str]])

• write_unresolved_csv (collections.abc.Callable[[...], pathlib.Path | None])

• print_unresolved_ai_prompt (collections.abc.Callable[[pathlib.Path], None])

• resolve_from_ai_csv (collections.abc.Callable[[...], dict[pathlib.Path, dict[str, Any]]])

• ads_block_risk_threshold (int)

ads_block_risk_threshold: int

build_progress: collections.abc.Callable[[...], Any | None]

confirm_ads_queries: collections.abc.Callable[[...], tuple[str, int]]

estimate_ads_queries_for_targets: collections.abc.Callable[[...], dict[str, Any]]

format_elapsed: collections.abc.Callable[[float], str]

parallel_extract_first_page_previews: collections.abc.Callable[[...], dict[pathlib.Path, str]]

parallel_map_paths: collections.abc.Callable[[...], dict[pathlib.Path, Any]]

print_unresolved_ai_prompt: collections.abc.Callable[[pathlib.Path], None]

profile_sync_payload: collections.abc.Callable[[Any], dict[str, Any]]

progress_reporter_factory: collections.abc.Callable[[], Any]

resolve_from_ai_csv: collections.abc.Callable[[...], dict[pathlib.Path, dict[str, Any]]]

should_prompt_ads_permission: collections.abc.Callable[[], bool]

to_sidecar_key: collections.abc.Callable[[pathlib.Path, pathlib.Path], str]

write_unresolved_csv: collections.abc.Callable[[...], pathlib.Path | None]

adsfolio.pdf.identify.build_cached_candidate_lookup(identifier, registry, documents)

Build a unique cached-document lookup for every strategy candidate kind.

Parameters:

• identifier (adsfolio.pdf.identifier.PdfIdentifier)

• registry (Any)

• documents (list[dict[str, Any]])

Return type:

collections.abc.Callable[[str, Any], tuple[bool, dict[str, Any] | None]]

adsfolio.pdf.identify.build_checked_record(identifier, document, path, sha256)

Build a record, or note why the PDF does not prove document.

Parameters:

• identifier (Any)

• document (dict[str, Any])

• path (pathlib.Path)

• sha256 (str)

Return type:

dict[str, Any] | None

adsfolio.pdf.identify.cached_ads_record(registry, identifier, path, doi='', bibcode='', sha256='')

Return cache-hit status and a unique folder-cache identify record.

A cached document that the PDF does not prove is reported as a cache hit without a record, and the reason is kept in identifier.identity_failures.

Parameters:

• registry (Any)

• identifier (Any)

• path (pathlib.Path)

• doi (str)

• bibcode (str)

• sha256 (str)

Return type:

tuple[bool, dict[str, Any] | None]

adsfolio.pdf.identify.print_cache_status(cached)

Print cache age and document count for an identify run.

Parameters:

cached (adsfolio.pubcache.CachedPublications | None)

Return type:

None

adsfolio.pdf.identify.run_identify(client, directory, *, helpers, recursive, debut_year, end_year, refresh_cache, output_unresolved, skip_unresolved_directory, apply, verbose_parser, strategies, allow_ads, ai_csv, verbose, profile=None, prompt_for_ai_csv=True, providers=(DoiProvider.ADS, DoiProvider.CROSSREF, DoiProvider.DATACITE), paths=None)

Identify PDFs and persist their ADS metadata.

paths, when given, are the PDFs to identify in place of a scan of directory. They must lie inside directory. The unresolved directory is not skipped for them, because the user chose them.

Parameters:

• client (Any)

• directory (pathlib.Path)

• helpers (adsfolio.pdf.identify.IdentifyWorkflowHelpers)

• recursive (bool)

• debut_year (int | None)

• end_year (int | None)

• refresh_cache (bool)

• output_unresolved (pathlib.Path | None)

• skip_unresolved_directory (bool)

• apply (bool)

• verbose_parser (bool)

• strategies (tuple[adsfolio.pdf.identify_settings.IdentifyStrategy, ...])

• allow_ads (bool)

• ai_csv (pathlib.Path | None)

• verbose (int)

• profile (adsfolio.pdf.prefetch.PublicationProfile | None)

• prompt_for_ai_csv (bool)

• providers (tuple[adsfolio.pdf.doi_providers.DoiProvider, ...])

• paths (collections.abc.Sequence[pathlib.Path] | None)

Return type:

None