trustgraph-ocr 0.21.11__tar.gz

trustgraph_ocr-0.21.11/PKG-INFO

@@ -0,0 +1,21 @@
+Metadata-Version: 2.1
+Name: trustgraph-ocr
+Version: 0.21.11
+Summary: TrustGraph provides a means to run a pipeline of flexible AI processing components in a flexible means to achieve a processing pipeline.
+Home-page: https://github.com/trustgraph-ai/trustgraph
+Download-URL: https://github.com/trustgraph-ai/trustgraph/archive/refs/tags/v0.21.11.tar.gz
+Author: trustgraph.ai
+Author-email: security@trustgraph.ai
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: GNU General Public License v3 or later (GPLv3+)
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+Requires-Dist: trustgraph-base<0.22,>=0.21
+Requires-Dist: pulsar-client
+Requires-Dist: prometheus-client
+Requires-Dist: boto3
+Requires-Dist: pdf2image
+Requires-Dist: pytesseract
+
+See https://trustgraph.ai/

trustgraph_ocr-0.21.11/README.md

@@ -0,0 +1 @@
+See https://trustgraph.ai/

trustgraph_ocr-0.21.11/scripts/pdf-ocr

@@ -0,0 +1,6 @@
+#!/usr/bin/env python3
+
+from trustgraph.decoding.ocr import run
+
+run()
+

trustgraph_ocr-0.21.11/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

trustgraph_ocr-0.21.11/setup.py

@@ -0,0 +1,47 @@
+import setuptools
+import os
+import importlib
+
+with open("README.md", "r") as fh:
+    long_description = fh.read()
+
+# Load a version number module
+spec = importlib.util.spec_from_file_location(
+    'version', 'trustgraph/ocr_version.py'
+)
+version_module = importlib.util.module_from_spec(spec)
+spec.loader.exec_module(version_module)
+
+version = version_module.__version__
+
+setuptools.setup(
+    name="trustgraph-ocr",
+    version=version,
+    author="trustgraph.ai",
+    author_email="security@trustgraph.ai",
+    description="TrustGraph provides a means to run a pipeline of flexible AI processing components in a flexible means to achieve a processing pipeline.",
+    long_description=long_description,
+    long_description_content_type="text/markdown",
+    url="https://github.com/trustgraph-ai/trustgraph",
+    packages=setuptools.find_namespace_packages(
+        where='./',
+    ),
+    classifiers=[ 
+        "Programming Language :: Python :: 3",
+        "License :: OSI Approved :: GNU General Public License v3 or later (GPLv3+)",
+        "Operating System :: OS Independent",
+    ],
+    python_requires='>=3.8',
+    download_url = "https://github.com/trustgraph-ai/trustgraph/archive/refs/tags/v" + version + ".tar.gz",
+    install_requires=[
+        "trustgraph-base>=0.21,<0.22",
+        "pulsar-client",
+        "prometheus-client",
+        "boto3",
+        "pdf2image",
+        "pytesseract",
+    ],
+    scripts=[
+        "scripts/pdf-ocr",
+    ]
+)

trustgraph_ocr-0.21.11/trustgraph/decoding/ocr/__init__.py

@@ -0,0 +1,3 @@
+
+from . pdf_decoder import *
+

trustgraph_ocr-0.21.11/trustgraph/decoding/ocr/__main__.py

@@ -0,0 +1,7 @@
+#!/usr/bin/env python3
+
+from . pdf_decoder import run
+
+if __name__ == '__main__':
+    run()
+

trustgraph_ocr-0.21.11/trustgraph/decoding/ocr/pdf_decoder.py

@@ -0,0 +1,83 @@
+
+"""
+Simple decoder, accepts PDF documents on input, outputs pages from the
+PDF document as text as separate output objects.
+"""
+
+import tempfile
+import base64
+import pytesseract
+from pdf2image import convert_from_bytes
+
+from ... schema import Document, TextDocument, Metadata
+from ... schema import document_ingest_queue, text_ingest_queue
+from ... log_level import LogLevel
+from ... base import ConsumerProducer
+
+module = ".".join(__name__.split(".")[1:-1])
+
+default_input_queue = document_ingest_queue
+default_output_queue = text_ingest_queue
+default_subscriber = module
+
+class Processor(ConsumerProducer):
+
+    def __init__(self, **params):
+
+        input_queue = params.get("input_queue", default_input_queue)
+        output_queue = params.get("output_queue", default_output_queue)
+        subscriber = params.get("subscriber", default_subscriber)
+
+        super(Processor, self).__init__(
+            **params | {
+                "input_queue": input_queue,
+                "output_queue": output_queue,
+                "subscriber": subscriber,
+                "input_schema": Document,
+                "output_schema": TextDocument,
+            }
+        )
+
+        print("PDF OCR inited")
+
+    async def handle(self, msg):
+
+        print("PDF message received")
+
+        v = msg.value()
+
+        print(f"Decoding {v.metadata.id}...", flush=True)
+
+        blob = base64.b64decode(v.data)
+
+        pages = convert_from_bytes(blob)
+
+        for ix, page in enumerate(pages):
+
+            try:
+                text = pytesseract.image_to_string(page, lang='eng')
+            except Exception as e:
+                print(f"Page did not OCR: {e}")
+                continue
+
+            r = TextDocument(
+                metadata=v.metadata,
+                text=text.encode("utf-8"),
+            )
+
+            await self.send(r)
+
+        print("Done.", flush=True)
+
+    @staticmethod
+    def add_args(parser):
+
+        ConsumerProducer.add_args(
+            parser, default_input_queue, default_subscriber,
+            default_output_queue,
+        )
+
+def run():
+
+    Processor.launch(module, __doc__)
+

trustgraph_ocr-0.21.11/trustgraph/ocr_version.py

@@ -0,0 +1 @@
+__version__ = "0.21.11"

trustgraph_ocr-0.21.11/trustgraph/tesseract_version.py

@@ -0,0 +1 @@
+__version__ = "0.0.0"

trustgraph_ocr-0.21.11/trustgraph_ocr.egg-info/PKG-INFO

@@ -0,0 +1,21 @@
+Metadata-Version: 2.1
+Name: trustgraph-ocr
+Version: 0.21.11
+Summary: TrustGraph provides a means to run a pipeline of flexible AI processing components in a flexible means to achieve a processing pipeline.
+Home-page: https://github.com/trustgraph-ai/trustgraph
+Download-URL: https://github.com/trustgraph-ai/trustgraph/archive/refs/tags/v0.21.11.tar.gz
+Author: trustgraph.ai
+Author-email: security@trustgraph.ai
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: GNU General Public License v3 or later (GPLv3+)
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+Requires-Dist: trustgraph-base<0.22,>=0.21
+Requires-Dist: pulsar-client
+Requires-Dist: prometheus-client
+Requires-Dist: boto3
+Requires-Dist: pdf2image
+Requires-Dist: pytesseract
+
+See https://trustgraph.ai/

trustgraph_ocr-0.21.11/trustgraph_ocr.egg-info/SOURCES.txt

@@ -0,0 +1,13 @@
+README.md
+setup.py
+scripts/pdf-ocr
+trustgraph/ocr_version.py
+trustgraph/tesseract_version.py
+trustgraph/decoding/ocr/__init__.py
+trustgraph/decoding/ocr/__main__.py
+trustgraph/decoding/ocr/pdf_decoder.py
+trustgraph_ocr.egg-info/PKG-INFO
+trustgraph_ocr.egg-info/SOURCES.txt
+trustgraph_ocr.egg-info/dependency_links.txt
+trustgraph_ocr.egg-info/requires.txt
+trustgraph_ocr.egg-info/top_level.txt

trustgraph_ocr-0.21.11/trustgraph_ocr.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

trustgraph_ocr-0.21.11/trustgraph_ocr.egg-info/requires.txt

@@ -0,0 +1,6 @@
+trustgraph-base<0.22,>=0.21
+pulsar-client
+prometheus-client
+boto3
+pdf2image
+pytesseract

trustgraph_ocr-0.21.11/trustgraph_ocr.egg-info/top_level.txt

@@ -0,0 +1,2 @@
+scripts
+trustgraph