Asynchronous method introspection

asynchronous-method-introspection

Summary

Dynamically discover, enumerate, and validate asynchronous conversion methods across a modular converter registry to build a unified job manifest of all available source-to-target chemical identifier transformations. This skill enables automatic Job discovery without manual configuration, ensuring that each conversion route is backed by a real async method on the corresponding converter instance.

When to use

Apply this skill when you have a plugin-based converter architecture (e.g., web services and compute libraries in separate directories) and you need to automatically discover all available (source_attribute, target_attribute, converter_name) conversion triples without hardcoding them. Use it before running batch annotation workflows to confirm which Jobs are actually available and executable.

When NOT to use

• Converters are statically defined and do not change at runtime — use a static configuration file instead
• You only need a subset of converters; introspecting the entire registry will waste time and resources
• Method signatures are not standardized across converters — inconsistent async/sync calling conventions will cause validation failures

Inputs

• Converter module paths (web and compute directories)
• Converter class definitions with init method defining conversions list
• Converter instances with method stubs or method templates

Outputs

• Unified Job registry (list of validated (source_attribute, target_attribute, converter_name) tuples)
• Serialized Job manifest file (structured listing of all available conversions)
• Validated method catalog (mapping of Job triples to actual callable async/sync methods)

How to apply

Load the ConverterBuilder module to recursively discover all converter classes from MSMetaEnhancer/libs/converters/web/ and MSMetaEnhancer/libs/converters/compute/ directories. For each discovered converter instance, extract its conversions list (defined as tuples of source_attribute, target_attribute, and conversion_method_name) and invoke create_top_level_conversion_methods() to trigger dynamic method generation. Then iterate over each converter's conversions list and validate that the named conversion_method exists as an async method (for WebConverters) or sync method (for ComputeConverters) on that instance. Aggregate all validated (source_attr, target_attr, converter_name) triples into a unified Job registry and serialize the result into a structured manifest file that can be queried downstream to determine which conversion routes are safe to request.

Related tools

• MSMetaEnhancer (Host framework containing ConverterBuilder and converter registry; provides async annotation orchestration) — https://github.com/RECETOX/MSMetaEnhancer
• ConverterBuilder (Module that auto-discovers, instantiates, and introspects all converter classes) — https://github.com/RECETOX/MSMetaEnhancer
• CIR (Example web converter providing chemical identifier conversions via external service) — https://cactus.nci.nih.gov/chemical/structure_documentation
• CTS (Example web converter providing chemical identifier conversions via external service) — https://cts.fiehnlab.ucdavis.edu/
• PubChem (Example web converter providing chemical identifier conversions via external service) — https://pubchem.ncbi.nlm.nih.gov/
• IDSM (Example web converter providing chemical identifier conversions via external service) — https://idsm.elixir-czech.cz/
• BridgeDb (Example web converter providing chemical identifier conversions via external service) — https://bridgedb.github.io/
• RDKit (Example compute converter providing chemical identifier conversions via local computation)
• Python (Language used to implement introspection and async method discovery)

Examples

from MSMetaEnhancer.libs.utils.ConverterBuilder import ConverterBuilder
from MSMetaEnhancer.libs.converters.web import CTS, CIR, IDSM, PubChem, BridgeDb
from MSMetaEnhancer.libs.converters.compute import RDKit
ConverterBuilder.register([CTS, CIR, IDSM, PubChem, BridgeDb, RDKit])
jobs_manifest = ConverterBuilder.discover_and_validate_jobs()
print(jobs_manifest)

Evaluation signals

• All discovered converter instances have a conversions attribute that is a non-empty list of tuples (source, target, method_name)
• For each Job triple (source_attr, target_attr, converter_name), the corresponding converter instance has a callable method with the specified method_name that is either async (WebConverters) or sync (ComputeConverters)
• The serialized Job manifest contains ≥ 1 entry per registered converter and is syntactically valid (JSON, YAML, or structured text format)
• Invoking a method from the registry on a real converter instance does not raise AttributeError or MethodNotFoundError
• The Job registry is consistent across multiple runs of the introspection routine (no non-deterministic discovery)

Limitations

• Dynamically generated methods must follow a strict naming convention (method_name derived from source and target attributes); converters that deviate will not be discovered
• Async method introspection cannot validate method signatures or return types at discovery time; validation failures only surface at runtime
• The skill assumes converters are stateless during introspection; stateful converters that require initialization parameters may not be introspectable
• Network-dependent converters (e.g., web services) are not validated to confirm they are actually reachable; only local method presence is checked

Evidence

• [other] Load the ConverterBuilder module which automatically discovers and instantiates all available converters from MSMetaEnhancer/libs/converters/web/ and MSMetaEnhancer/libs/converters/compute/ directories.: "Load the ConverterBuilder module which automatically discovers and instantiates all available converters from MSMetaEnhancer/libs/converters/web/ and MSMetaEnhancer/libs/converters/compute/"
• [other] For each discovered converter (CIR, CTS, IDSM, PubChem, BridgeDb, RDKit, and any custom converters), extract the conversions list defined in its init method as tuples of (source_attribute, target_attribute, conversion_method_name).: "For each discovered converter (CIR, CTS, IDSM, PubChem, BridgeDb, RDKit, and any custom converters), extract the conversions list defined in its init method as tuples of (source_attribute,"
• [other] Invoke the create_top_level_conversion_methods() call on each converter to trigger dynamic method generation based on the conversions list.: "Invoke the create_top_level_conversion_methods() call on each converter to trigger dynamic method generation based on the conversions list."
• [other] Validate that each Job's conversion_method exists as an async method (for WebConverters) or sync method (for ComputeConverters) on the corresponding converter instance.: "Validate that each Job's conversion_method exists as an async method (for WebConverters) or sync method (for ComputeConverters) on the corresponding converter instance."
• [readme] The app uses asynchronous implementation of annotation process allowing for optimal fetching speed.: "The app uses asynchronous implementation of annotation process allowing for optimal fetching speed."
• [readme] ConverterBuilder.register([CTS, CIR, IDSM, PubChem, BridgeDb, RDKit]): "ConverterBuilder.register([CTS, CIR, IDSM, PubChem, BridgeDb, RDKit])"