3.0

In this major release, Canopy integrates Elasticsearch (ES) version 8.0 into our application.

This upgrade significantly improves search functionality within the Canopy application and provides a more robust user experience compared to the previous ES.

Elasticsearch does not exhibit backward compatibility. Consequently, all new functionalities introduced in the Canopy Release Version 3.0.0 and forward will not be applicable to projects created in the previous versions. To clearly distinguish between projects created in these versions, users will see the following information:

1. Major Release Project Indicator: Projects created prior to the Version 3.0 release will be visually distinguished by a blue dot in the top right corner of their respective project names. This dot signifies that the new functionalities introduced in Version 3.0 are not applicable to these projects.

2. When users hover over the blue dot in the project dropdown, a tooltip displays a message: “This project may lack features from newer releases. Created with v1.”

Canopy now automatically detects handwritten signatures within documents, enabling users to quickly and easily identify and extract signature data during the review process.

Canopy’s signature detection uses a computer vision algorithm trained to identify signatures within an image, multi-page TIFF, or image-based PDF.

Canopy’s new Elasticsearch integration significantly enhances search performance and delivers a more robust user search experience.

This upgrade introduces changes to Canopy’s Field Search compared to previous versions.

For a detailed comparison of Field Search functionality in Version 3.0 and earlier versions, users can refer to the Image Classification and Field Search documentation.

This release improves the document viewing experience and document index with the following updates:

1. Enhancing Large Text Display: The document viewer can now load documents of any length, regardless of character count, ensuring immediate, complete access to all document content. Users can now seamlessly scroll through documents, select data, and easily add entities.

(Prior to this release, document view had a text display up to 500,000 characters. Documents exceeding this limit were not fully loaded, and users received a notification stating, “Document is not fully loaded, please download document for review.”)

2. Highlighting all PII Elements: The document viewer now highlights all PII elements within a document, regardless of the total PII count. This enhancement ensures effortless, complete detection and entry of all PII elements within entities.

(Prior to this release, document view highlighted up to 1000 PII elements).

3. Indexing the Entire Document: Canopy now fully indexes all text extracted from documents, regardless of character count, ensuring consistent and complete text index and search results.

(Prior to this release, PDF, Excel, Microsoft Access, and SQL database documents were indexed up to 10,000 characters per document; thus, documents exceeding that limit received a “Partially Indexed” status.)

4. Maximizing Screen Space: The document viewer has been resized to utilize the entire screen. Users can now see more text on the screen, enhancing readability and reducing scrolling.

This release improves the user’s search experience by significantly increasing the terms supported in a single search query. Users can now include up to 65,536 terms to search for relevant documents, enabling more comprehensive searches across document fields.

(Prior to this release, users could include up to 1,000 terms in a single search query.)

Users can enter up to 65,536 search terms in a single query.

Bulk Search allows users to define complex searches using multiple rows of queries. Users can take the following two steps to bulk search:

1. Run Test Search: Enter your search terms or queries in each row and click Run Test Search to get the results for that specific row.

2. Full Search: Once you’ve Run Test Search, click Search to retrieve documents matching all the queries across all rows. Users can use up to 65,536 search terms combined across all rows.

After 64,800 rows, Canopy disables Bulk Search after you Run Test Search. (Prior to this release, Canopy disabled Bulk Search after 900 rows.)

Users can search for and view up to 65,536 entity-related documents on the entity list view.

(Prior to this release, users could search for and view up to 1,000 entity-related documents.)

To find entity-related documents on the entity list view:

1. Navigate to and select Entities from the menu.
2. Choose Raw Entity View.
3. Select an entity or a group of entities whose documents you wish to view.
4. Click View Documents to retrieve documents containing the selected entities.

Canopy’s improved search helps users find documents easily and efficiently. Improvements include the following updates:

1. Built-in Stemmer for Keyword Search: Our built-in English stemmer automatically finds variations of your search terms. For example, typing the base word “run” in the search bar will return documents containing “running,” “runner,” “ran,” etc.

2. Phrase Search: Use double quotes to enclose exact phrases for precise searches. Searching for “red delicious apple” (in quotes) will only return results containing that exact phrase. Searching for red delicious apple (without quotes) will return results containing those words in any order.

3. No Case-Sensitive Search: Search is no longer case-sensitive. Whether you type “Apple,” “apple,” or “APPLE,” you’ll get the same results.

4. OCR-Performed Boolean Search: Searching for OCR-Performed documents is now more flexible and powerful with Canopy’s Boolean search. Use meta.ocr_performed:true to find documents that have been through OCR and meta.ocr_performed:false to find those that haven’t.

Learn more about the various search options in the Search Guide documentation.

Canopy now supports the TIFF document pagination and synchronized text-page views in the document viewer. This enhancement allows users to navigate multi-page TIFF files, streamlining the review process seamlessly. Additionally, synchronized text and page views ensure that all the content from TIFF documents is accurately aligned in text view, allowing for efficient entity detection.

Resolved an issue where the Impact Assessment report has a content overflow, and users need to scroll horizontally to view the entire report. The Impact Assessment report now fits within the pane, enabling clear visibility of all report content without horizontal scrolling.