Feature/pkgsplit (#83)

* Starting to spawn base package
* More package hacking
* Bedrock and VertexAI
* Parquet split
* Updated templates
* Utils

trustgraph-flow/trustgraph/decoding/pdf/__init__.py

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

trustgraph-flow/trustgraph/decoding/pdf/__main__.py

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

trustgraph-flow/trustgraph/decoding/pdf/pdf_decoder.py

@@ -0,0 +1,87 @@
+
+"""
+Simple decoder, accepts PDF documents on input, outputs pages from the
+PDF document as text as separate output objects.
+"""
+
+import tempfile
+import base64
+from langchain_community.document_loaders import PyPDFLoader
+
+from ... schema import Document, TextDocument, Source
+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 inited")
+
+    def handle(self, msg):
+
+        print("PDF message received")
+
+        v = msg.value()
+
+        print(f"Decoding {v.source.id}...", flush=True)
+
+        with tempfile.NamedTemporaryFile(delete_on_close=False) as fp:
+
+            fp.write(base64.b64decode(v.data))
+            fp.close()
+
+            with open(fp.name, mode='rb') as f:
+
+                loader = PyPDFLoader(fp.name)
+                pages = loader.load()
+
+                for ix, page in enumerate(pages):
+
+                    id = v.source.id + "-p" + str(ix)
+                    r = TextDocument(
+                        source=Source(
+                            source=v.source.source,
+                            title=v.source.title,
+                            id=id,
+                        ),
+                        text=page.page_content.encode("utf-8"),
+                    )
+
+                    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.start(module, __doc__)
+