2024-07-10 23:20:06 +01:00
|
|
|
|
|
|
|
|
# VERSION=$(shell git describe | sed 's/^v//')
|
2024-10-04 17:28:07 +01:00
|
|
|
|
2024-10-15 00:31:08 +01:00
|
|
|
VERSION=0.0.0
|
2024-07-10 23:20:06 +01:00
|
|
|
|
2024-08-12 17:39:54 +01:00
|
|
|
DOCKER=podman
|
|
|
|
|
|
2026-02-23 15:56:29 +00:00
|
|
|
all: containers
|
2024-07-10 23:20:06 +01:00
|
|
|
|
2024-09-30 20:50:20 +01:00
|
|
|
# Not used
|
|
|
|
|
wheels:
|
2024-10-04 22:12:39 +01:00
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph/
|
2024-09-30 19:36:09 +01:00
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph-base/
|
|
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph-flow/
|
|
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph-vertexai/
|
|
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph-bedrock/
|
2024-09-30 14:00:29 +01:00
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph-embeddings-hf/
|
2024-09-30 22:20:26 +01:00
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph-cli/
|
2025-03-20 09:29:40 +00:00
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph-ocr/
|
Add universal document decoder with multi-format support (#705)
Add universal document decoder with multi-format support
using 'unstructured'.
New universal decoder service powered by the unstructured
library, handling DOCX, XLSX, PPTX, HTML, Markdown, CSV, RTF,
ODT, EPUB and more through a single service. Tables are preserved
as HTML markup for better downstream extraction. Images are
stored in the librarian but excluded from the text
pipeline. Configurable section grouping strategies
(whole-document, heading, element-type, count, size) for non-page
formats. Page-based formats (PDF, PPTX, XLSX) are automatically
grouped by page.
All four decoders (PDF, Mistral OCR, Tesseract OCR, universal)
now share the "document-decoder" ident so they are
interchangeable. PDF-only decoders fetch document metadata to
check MIME type and gracefully skip unsupported formats.
Librarian changes: removed MIME type whitelist validation so any
document format can be ingested. Simplified routing so text/plain
goes to text-load and everything else goes to document-load.
Removed dual inline/streaming data paths — documents always use
document_id for content retrieval.
New provenance entity types (tg:Section, tg:Image) and metadata
predicates (tg:elementTypes, tg:tableCount, tg:imageCount) for
richer explainability.
Universal decoder is in its own package (trustgraph-unstructured)
and container image (trustgraph-unstructured).
2026-03-23 12:56:35 +00:00
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph-unstructured/
|
2025-07-02 18:19:23 +01:00
|
|
|
pip3 wheel --no-deps --wheel-dir dist trustgraph-mcp/
|
2024-09-30 14:00:29 +01:00
|
|
|
|
2024-10-04 15:47:46 +01:00
|
|
|
packages: update-package-versions
|
2024-09-30 20:50:20 +01:00
|
|
|
rm -rf dist/
|
2025-07-23 21:22:08 +01:00
|
|
|
cd trustgraph && python -m build --sdist --outdir ../dist/
|
|
|
|
|
cd trustgraph-base && python -m build --sdist --outdir ../dist/
|
|
|
|
|
cd trustgraph-flow && python -m build --sdist --outdir ../dist/
|
|
|
|
|
cd trustgraph-vertexai && python -m build --sdist --outdir ../dist/
|
|
|
|
|
cd trustgraph-bedrock && python -m build --sdist --outdir ../dist/
|
|
|
|
|
cd trustgraph-embeddings-hf && python -m build --sdist --outdir ../dist/
|
|
|
|
|
cd trustgraph-cli && python -m build --sdist --outdir ../dist/
|
|
|
|
|
cd trustgraph-ocr && python -m build --sdist --outdir ../dist/
|
Add universal document decoder with multi-format support (#705)
Add universal document decoder with multi-format support
using 'unstructured'.
New universal decoder service powered by the unstructured
library, handling DOCX, XLSX, PPTX, HTML, Markdown, CSV, RTF,
ODT, EPUB and more through a single service. Tables are preserved
as HTML markup for better downstream extraction. Images are
stored in the librarian but excluded from the text
pipeline. Configurable section grouping strategies
(whole-document, heading, element-type, count, size) for non-page
formats. Page-based formats (PDF, PPTX, XLSX) are automatically
grouped by page.
All four decoders (PDF, Mistral OCR, Tesseract OCR, universal)
now share the "document-decoder" ident so they are
interchangeable. PDF-only decoders fetch document metadata to
check MIME type and gracefully skip unsupported formats.
Librarian changes: removed MIME type whitelist validation so any
document format can be ingested. Simplified routing so text/plain
goes to text-load and everything else goes to document-load.
Removed dual inline/streaming data paths — documents always use
document_id for content retrieval.
New provenance entity types (tg:Section, tg:Image) and metadata
predicates (tg:elementTypes, tg:tableCount, tg:imageCount) for
richer explainability.
Universal decoder is in its own package (trustgraph-unstructured)
and container image (trustgraph-unstructured).
2026-03-23 12:56:35 +00:00
|
|
|
cd trustgraph-unstructured && python -m build --sdist --outdir ../dist/
|
2025-07-23 21:22:08 +01:00
|
|
|
cd trustgraph-mcp && python -m build --sdist --outdir ../dist/
|
2024-09-30 20:50:20 +01:00
|
|
|
|
2024-09-30 21:07:18 +01:00
|
|
|
pypi-upload:
|
|
|
|
|
twine upload dist/*-${VERSION}.*
|
|
|
|
|
|
2025-01-28 19:36:05 +00:00
|
|
|
CONTAINER_BASE=docker.io/trustgraph
|
2024-07-10 23:20:06 +01:00
|
|
|
|
2024-09-30 16:38:50 +01:00
|
|
|
update-package-versions:
|
2024-09-30 22:20:26 +01:00
|
|
|
mkdir -p trustgraph-cli/trustgraph
|
2024-10-08 20:33:14 +01:00
|
|
|
mkdir -p trustgraph/trustgraph
|
2024-09-30 19:36:09 +01:00
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph-base/trustgraph/base_version.py
|
|
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph-flow/trustgraph/flow_version.py
|
|
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph-vertexai/trustgraph/vertexai_version.py
|
|
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph-bedrock/trustgraph/bedrock_version.py
|
2024-09-30 16:38:50 +01:00
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph-embeddings-hf/trustgraph/embeddings_hf_version.py
|
2024-09-30 22:20:26 +01:00
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph-cli/trustgraph/cli_version.py
|
2025-03-20 09:29:40 +00:00
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph-ocr/trustgraph/ocr_version.py
|
Add universal document decoder with multi-format support (#705)
Add universal document decoder with multi-format support
using 'unstructured'.
New universal decoder service powered by the unstructured
library, handling DOCX, XLSX, PPTX, HTML, Markdown, CSV, RTF,
ODT, EPUB and more through a single service. Tables are preserved
as HTML markup for better downstream extraction. Images are
stored in the librarian but excluded from the text
pipeline. Configurable section grouping strategies
(whole-document, heading, element-type, count, size) for non-page
formats. Page-based formats (PDF, PPTX, XLSX) are automatically
grouped by page.
All four decoders (PDF, Mistral OCR, Tesseract OCR, universal)
now share the "document-decoder" ident so they are
interchangeable. PDF-only decoders fetch document metadata to
check MIME type and gracefully skip unsupported formats.
Librarian changes: removed MIME type whitelist validation so any
document format can be ingested. Simplified routing so text/plain
goes to text-load and everything else goes to document-load.
Removed dual inline/streaming data paths — documents always use
document_id for content retrieval.
New provenance entity types (tg:Section, tg:Image) and metadata
predicates (tg:elementTypes, tg:tableCount, tg:imageCount) for
richer explainability.
Universal decoder is in its own package (trustgraph-unstructured)
and container image (trustgraph-unstructured).
2026-03-23 12:56:35 +00:00
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph-unstructured/trustgraph/unstructured_version.py
|
2024-10-04 22:12:39 +01:00
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph/trustgraph/trustgraph_version.py
|
2025-07-02 18:19:23 +01:00
|
|
|
echo __version__ = \"${VERSION}\" > trustgraph-mcp/trustgraph/mcp_version.py
|
2024-09-30 16:38:50 +01:00
|
|
|
|
2026-04-13 20:51:09 +01:00
|
|
|
containers: container-base container-flow \
|
|
|
|
|
container-bedrock container-vertexai \
|
|
|
|
|
container-hf container-ocr \
|
|
|
|
|
container-unstructured container-mcp
|
|
|
|
|
|
2026-04-22 15:19:57 +01:00
|
|
|
some-containers: container-base container-flow container-unstructured
|
2024-07-10 23:20:06 +01:00
|
|
|
|
|
|
|
|
push:
|
2025-01-28 19:36:05 +00:00
|
|
|
${DOCKER} push ${CONTAINER_BASE}/trustgraph-base:${VERSION}
|
|
|
|
|
${DOCKER} push ${CONTAINER_BASE}/trustgraph-flow:${VERSION}
|
|
|
|
|
${DOCKER} push ${CONTAINER_BASE}/trustgraph-bedrock:${VERSION}
|
|
|
|
|
${DOCKER} push ${CONTAINER_BASE}/trustgraph-vertexai:${VERSION}
|
|
|
|
|
${DOCKER} push ${CONTAINER_BASE}/trustgraph-hf:${VERSION}
|
2025-03-20 09:29:40 +00:00
|
|
|
${DOCKER} push ${CONTAINER_BASE}/trustgraph-ocr:${VERSION}
|
Add universal document decoder with multi-format support (#705)
Add universal document decoder with multi-format support
using 'unstructured'.
New universal decoder service powered by the unstructured
library, handling DOCX, XLSX, PPTX, HTML, Markdown, CSV, RTF,
ODT, EPUB and more through a single service. Tables are preserved
as HTML markup for better downstream extraction. Images are
stored in the librarian but excluded from the text
pipeline. Configurable section grouping strategies
(whole-document, heading, element-type, count, size) for non-page
formats. Page-based formats (PDF, PPTX, XLSX) are automatically
grouped by page.
All four decoders (PDF, Mistral OCR, Tesseract OCR, universal)
now share the "document-decoder" ident so they are
interchangeable. PDF-only decoders fetch document metadata to
check MIME type and gracefully skip unsupported formats.
Librarian changes: removed MIME type whitelist validation so any
document format can be ingested. Simplified routing so text/plain
goes to text-load and everything else goes to document-load.
Removed dual inline/streaming data paths — documents always use
document_id for content retrieval.
New provenance entity types (tg:Section, tg:Image) and metadata
predicates (tg:elementTypes, tg:tableCount, tg:imageCount) for
richer explainability.
Universal decoder is in its own package (trustgraph-unstructured)
and container image (trustgraph-unstructured).
2026-03-23 12:56:35 +00:00
|
|
|
${DOCKER} push ${CONTAINER_BASE}/trustgraph-unstructured:${VERSION}
|
2025-07-03 17:00:59 +01:00
|
|
|
${DOCKER} push ${CONTAINER_BASE}/trustgraph-mcp:${VERSION}
|
2024-07-10 23:20:06 +01:00
|
|
|
|
2025-09-11 12:32:04 +01:00
|
|
|
# Individual container build targets
|
2026-04-13 20:51:09 +01:00
|
|
|
container-%: update-package-versions
|
|
|
|
|
${DOCKER} build \
|
|
|
|
|
-f containers/Containerfile.${@:container-%=%} \
|
|
|
|
|
-t ${CONTAINER_BASE}/trustgraph-${@:container-%=%}:${VERSION} .
|
Add universal document decoder with multi-format support (#705)
Add universal document decoder with multi-format support
using 'unstructured'.
New universal decoder service powered by the unstructured
library, handling DOCX, XLSX, PPTX, HTML, Markdown, CSV, RTF,
ODT, EPUB and more through a single service. Tables are preserved
as HTML markup for better downstream extraction. Images are
stored in the librarian but excluded from the text
pipeline. Configurable section grouping strategies
(whole-document, heading, element-type, count, size) for non-page
formats. Page-based formats (PDF, PPTX, XLSX) are automatically
grouped by page.
All four decoders (PDF, Mistral OCR, Tesseract OCR, universal)
now share the "document-decoder" ident so they are
interchangeable. PDF-only decoders fetch document metadata to
check MIME type and gracefully skip unsupported formats.
Librarian changes: removed MIME type whitelist validation so any
document format can be ingested. Simplified routing so text/plain
goes to text-load and everything else goes to document-load.
Removed dual inline/streaming data paths — documents always use
document_id for content retrieval.
New provenance entity types (tg:Section, tg:Image) and metadata
predicates (tg:elementTypes, tg:tableCount, tg:imageCount) for
richer explainability.
Universal decoder is in its own package (trustgraph-unstructured)
and container image (trustgraph-unstructured).
2026-03-23 12:56:35 +00:00
|
|
|
|
2026-04-13 23:14:37 +01:00
|
|
|
# Multi-arch: build both platforms sequentially into one manifest (local use)
|
2026-04-13 20:51:09 +01:00
|
|
|
manifest-%: update-package-versions
|
|
|
|
|
-@${DOCKER} manifest rm \
|
|
|
|
|
${CONTAINER_BASE}/trustgraph-${@:manifest-%=%}:${VERSION}
|
|
|
|
|
${DOCKER} build --platform linux/amd64,linux/arm64 \
|
|
|
|
|
-f containers/Containerfile.${@:manifest-%=%} \
|
|
|
|
|
--manifest \
|
|
|
|
|
${CONTAINER_BASE}/trustgraph-${@:manifest-%=%}:${VERSION} .
|
|
|
|
|
|
2026-04-13 23:14:37 +01:00
|
|
|
# Multi-arch: build a single platform image (for parallel CI)
|
|
|
|
|
platform-%-amd64: update-package-versions
|
|
|
|
|
${DOCKER} build --platform linux/amd64 \
|
|
|
|
|
-f containers/Containerfile.${@:platform-%-amd64=%} \
|
|
|
|
|
-t ${CONTAINER_BASE}/trustgraph-${@:platform-%-amd64=%}:${VERSION}-amd64 .
|
|
|
|
|
|
|
|
|
|
platform-%-arm64: update-package-versions
|
|
|
|
|
${DOCKER} build --platform linux/arm64 \
|
|
|
|
|
-f containers/Containerfile.${@:platform-%-arm64=%} \
|
|
|
|
|
-t ${CONTAINER_BASE}/trustgraph-${@:platform-%-arm64=%}:${VERSION}-arm64 .
|
|
|
|
|
|
|
|
|
|
# Push a single platform image
|
|
|
|
|
push-platform-%-amd64:
|
|
|
|
|
${DOCKER} push \
|
|
|
|
|
${CONTAINER_BASE}/trustgraph-${@:push-platform-%-amd64=%}:${VERSION}-amd64
|
|
|
|
|
|
|
|
|
|
push-platform-%-arm64:
|
|
|
|
|
${DOCKER} push \
|
|
|
|
|
${CONTAINER_BASE}/trustgraph-${@:push-platform-%-arm64=%}:${VERSION}-arm64
|
|
|
|
|
|
|
|
|
|
# Combine per-platform images into a multi-arch manifest
|
|
|
|
|
combine-manifest-%:
|
|
|
|
|
-@${DOCKER} manifest rm \
|
|
|
|
|
${CONTAINER_BASE}/trustgraph-${@:combine-manifest-%=%}:${VERSION}
|
|
|
|
|
${DOCKER} manifest create \
|
|
|
|
|
${CONTAINER_BASE}/trustgraph-${@:combine-manifest-%=%}:${VERSION} \
|
|
|
|
|
docker://${CONTAINER_BASE}/trustgraph-${@:combine-manifest-%=%}:${VERSION}-amd64 \
|
|
|
|
|
docker://${CONTAINER_BASE}/trustgraph-${@:combine-manifest-%=%}:${VERSION}-arm64
|
|
|
|
|
${DOCKER} manifest push \
|
|
|
|
|
${CONTAINER_BASE}/trustgraph-${@:combine-manifest-%=%}:${VERSION}
|
|
|
|
|
|
2026-04-13 20:51:09 +01:00
|
|
|
# Push a container
|
|
|
|
|
push-container-%:
|
|
|
|
|
${DOCKER} push \
|
|
|
|
|
${CONTAINER_BASE}/trustgraph-${@:push-container-%=%}:${VERSION}
|
|
|
|
|
|
2026-04-13 23:14:37 +01:00
|
|
|
# Push a manifest (from local multi-arch build)
|
2026-04-13 20:51:09 +01:00
|
|
|
push-manifest-%:
|
|
|
|
|
${DOCKER} manifest push \
|
|
|
|
|
${CONTAINER_BASE}/trustgraph-${@:push-manifest-%=%}:${VERSION}
|
2025-09-11 12:32:04 +01:00
|
|
|
|
2024-07-10 23:20:06 +01:00
|
|
|
clean:
|
|
|
|
|
rm -rf wheels/
|
|
|
|
|
|
2024-07-15 19:42:16 +01:00
|
|
|
set-version:
|
2024-08-30 17:47:35 +01:00
|
|
|
echo '"${VERSION}"' > templates/values/version.jsonnet
|
2024-08-13 17:30:59 +01:00
|
|
|
|
2024-09-28 11:55:30 +01:00
|
|
|
docker-hub-login:
|
|
|
|
|
cat docker-token.txt | \
|
2025-01-06 21:45:42 +00:00
|
|
|
${DOCKER} login -u trustgraph --password-stdin registry-1.docker.io
|
2024-09-28 11:55:30 +01:00
|
|
|
|