From 2ac363e7cfffc8ca78a644cf9b5d5107305cf672 Mon Sep 17 00:00:00 2001 From: Cyber MacGeddon Date: Tue, 5 Aug 2025 09:56:11 +0100 Subject: [PATCH] Updated schemas and specs --- .gitignore | 1 + docs/tech-specs/STRUCTURED_DATA.md | 2 +- docs/tech-specs/STRUCTURED_DATA_SCHEMAS.md | 4 ++-- trustgraph-base/trustgraph/schema/knowledge/__init__.py | 2 +- .../trustgraph/schema/knowledge/{extraction.py => object.py} | 0 5 files changed, 5 insertions(+), 4 deletions(-) rename trustgraph-base/trustgraph/schema/knowledge/{extraction.py => object.py} (100%) diff --git a/.gitignore b/.gitignore index ef901963..c464fe27 100644 --- a/.gitignore +++ b/.gitignore @@ -13,4 +13,5 @@ trustgraph-flow/trustgraph/flow_version.py trustgraph-ocr/trustgraph/ocr_version.py trustgraph-parquet/trustgraph/parquet_version.py trustgraph-vertexai/trustgraph/vertexai_version.py +trustgraph-mcp/trustgraph/mcp_version.py vertexai/ \ No newline at end of file diff --git a/docs/tech-specs/STRUCTURED_DATA.md b/docs/tech-specs/STRUCTURED_DATA.md index 75e84554..a3f5025b 100644 --- a/docs/tech-specs/STRUCTURED_DATA.md +++ b/docs/tech-specs/STRUCTURED_DATA.md @@ -56,7 +56,7 @@ The structured data integration requires the following technical components: 3. **Object Extraction Module** - Enhanced knowledge extractor flow integration - - Builds on existing prototype in `trustgraph-flow/trustgraph/extract/objects` + - Builds on existing prototype in `trustgraph-flow/trustgraph/extract/object/row/` - Identifies and extracts structured objects from unstructured sources - Maintains provenance and confidence scores diff --git a/docs/tech-specs/STRUCTURED_DATA_SCHEMAS.md b/docs/tech-specs/STRUCTURED_DATA_SCHEMAS.md index 0dc47ad1..1e758e10 100644 --- a/docs/tech-specs/STRUCTURED_DATA_SCHEMAS.md +++ b/docs/tech-specs/STRUCTURED_DATA_SCHEMAS.md @@ -82,7 +82,7 @@ class StructuredQueryResponse(Record): ``` #### 2.2 Object Extraction Output -New file: `knowledge/extraction.py` +New file: `knowledge/object.py` ```python from pulsar.schema import Record, String, Map, Double @@ -136,4 +136,4 @@ The schemas will be used by new flow modules: 2. Update existing services to recognize new schema types 3. Implement flow modules that use these schemas 4. Add gateway/rev-gateway endpoints for new services -5. Create unit tests for schema validation \ No newline at end of file +5. Create unit tests for schema validation diff --git a/trustgraph-base/trustgraph/schema/knowledge/__init__.py b/trustgraph-base/trustgraph/schema/knowledge/__init__.py index c51f925f..3359b25c 100644 --- a/trustgraph-base/trustgraph/schema/knowledge/__init__.py +++ b/trustgraph-base/trustgraph/schema/knowledge/__init__.py @@ -5,4 +5,4 @@ from .knowledge import * from .nlp import * from .rows import * from .structured import * -from .extraction import * \ No newline at end of file +from .object import * diff --git a/trustgraph-base/trustgraph/schema/knowledge/extraction.py b/trustgraph-base/trustgraph/schema/knowledge/object.py similarity index 100% rename from trustgraph-base/trustgraph/schema/knowledge/extraction.py rename to trustgraph-base/trustgraph/schema/knowledge/object.py