Last active
March 30, 2023 13:24
-
-
Save alexcg1/99c722d80129ce29286f8cf8516feda9 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from docarray import BaseDoc, DocArray | |
from docarray.documents import TextDoc | |
from docarray.typing import AnyUrl | |
from jina import Deployment, Executor, requests | |
class TextChunk(TextDoc): | |
tags: dict = {} | |
class PDFDocument(BaseDoc): | |
texts: DocArray[TextChunk] = DocArray() | |
path: AnyUrl | None | |
class Extractor(Executor): | |
def __init__( | |
self, | |
content_types: list = ['text', 'table', 'image', 'metadata'], | |
**kwargs, | |
): | |
super().__init__(**kwargs) | |
self.content_types = content_types | |
@requests(on='/extract') | |
def add_chunks(self, docs: DocArray[PDFDocument], **kwargs): | |
for doc in docs: | |
self._extract_text(doc) | |
def _extract_text(self, doc: PDFDocument, **kwargs): | |
for _ in range(0, 90): | |
doc.texts.append(TextChunk(text='foo')) | |
dep = Deployment( | |
uses=Extractor, | |
uses_with={'content_types': ['text']}, | |
) | |
docs = DocArray([PDFDocument(path='foo.png')]) | |
print(f'input: {docs[0]}') | |
# above returns PDFDocument(id='5dede5812eb41a28953075b3bc52934c', texts=<DocArray (length=0)>, path=AnyUrl('foo.png', host_type='domain')) | |
with dep: | |
output = dep.post(inputs=docs, on='/extract') # this works fine | |
print(f'output: {output[0]}') | |
# above returns AnyDoc(id='5dede5812eb41a28953075b3bc52934c') |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment