Recently, we’ve been experimenting with the new blob indexer feature. To set up blob indexing, create an Azure blob datasource, a search index (if you don’t have one already), then create an indexer that connects that datasource to the target index. With Azure Search we try to help you build really great search applications over your data.
The function extracts text from the PDF file using pdf.js and using the supplies rules extracts metadata from the text & stores the result (text + metadata) in a DocumentDB collection which can then be used as a datasource for Azure Search.

You might also want to add a URL reference to the actual image file so you can allow users to open it directly from your application. One file type we have not yet added support for, but is a common ask, is of images. If nothing happens, download the GitHub extension for Visual Studio and try again. If it works you should see output in the Function logs like so: The rules.json file contains the regular expressions rules that are matched against the extracted text and stored as metadata. In you Azure Function you will need to supply a few, Note: When setting your DocumentDB connection string as the data source, you will need to include the Database name in the string like so, After a brief moment (give it a minute) you should now be able to run the. download the GitHub extension for Visual Studio, search/PDF2Search.postman_collection.json, You must have an active Azure Subscription, if you do not you can always start with a, Create a public or private (depending on your needs) blob container called, For testing when creating your collection start with the smallest/cheapest configuration which would be. If nothing happens, download GitHub Desktop and try again. In the words of Nuno Coimbra, senior developer at ALS: Blob indexer allowed us to follow an almost “shoot and forget” approach to document data extraction and indexing, taking from our hands that kind of plumbing.

When it comes to availability and scalability, this is a much easier solution for us to maintain.

This technique is called Optical Character Recognition (OCR) and I want to show you how this can be used to help enhance the content in your Azure Search index. This Azure Function binds to an Azure Storage Blob container and triggers when a PDF file is stored.

Note that this demo requires writing to an Azure Storage Account, which you will be billed monthly for the storage written to, and by default provisions a Basic Azure Search service which is billed hourly.

The included search/PDF2Search.postman_collection.json Postman collection contains the basics required to create the data source (DocumentDB), the index (search schema) and the indexer (reads from data source and indexes data using the configured index) as well as a very simple search query.

Recently we released the Azure Search Indexer for Azure Blob Storage which allows extraction of text from common file types such as Office, PDF and HTML.

Azure Cognitive Search is available in the new Microsoft Azure portal.

Azure Stack est un portefeuille de produits qui étendent les services et fonctionnalités Azure vers l’environnement de votre choix, du centre de données aux emplacements de périphérie et aux bureaux distants. This sample is just a starting point.

Access Visual Studio, Azure credits, Azure DevOps, and many other resources for creating, deploying, and managing applications. We hope that you will find the Azure Search blob indexer useful. On the right hand side of the application, we have some geospatial information coming back as a part of the search so if we want to include maps in our search, we can implement that as well. Through capabilities like the Azure Search Indexer, we have tried to make it convenient to ingest data from common data sources to enable this full text search support.