Known Issues and Limitations in V32

We are aware of the following issues and limitations in v32.0.0. We expect to address them in future versions of Hyperscience.

Some issues in v31 may also apply to v32. To see which v31 issues remain outstanding in v32, see Known Issues and Limitations in V31.

Layout Editor

Recognition of field names in Semi-structured layouts – After entering text in a field's Name field and clicking the Data Types drop-down list, the application marks the Name field in red, suggesting that it is empty.

[Fixed in v32.0.1] Unexpected warning when creating a Semi-structured layout – Entering and deleting a layout name when creating a Semi-structured layout causes the following warning message to appear:

“A layout with that name already exists in your library.”

Workaround: Start entering a layout name, and the warning message will disappear.

Layouts

[Fixed in v32.0.1] “Save changes” without creating a layout tag – In the Layout Details page, entering a layout tag name and clicking the Save changes button without actually creating a tag results in a “Select_create” tag to be added.

“Version History” tab for Additional layouts – The Layout Details page for Additional layouts contains a Version History tab, even though Additional layouts do not have versions.

Data Types

Two scrollbars when creating a pattern CFDT in Edge – Two scrollbars appear on the page when creating a pattern CFDT in Edge.

Flows

[Fixed in v32.0.4] Exporting flows without names – If a user exports a flow that does not have a name, a JSON named undefined.json is generated, but it is not downloaded to the user's machine.

Maximum retries and flow failure – When a task has reached its maximum number of retries, the flow is not marked as failed, even though it is.

Message Queue and HTTP Notifier Output Blocks in "Document Processing Notifications (V32)" – If the "Document Processing Notifications (V32)" flow contains only a HTTP Notifier Output Block or a Message Queue Notifier Output Block, the system does not display all the available settings for the block.

Workaround: Hyperscience can add a Custom Code Block in front of the Output Block, which will require a custom Document Processing Notifications flow.

Flow Studio

Masking of S3 Submission Retrieval Store in IE 11 – The value of S3 Submission Retrieval Store is never hidden in IE 11.

[Fixed in v32.0.1] Text in "Save and Deploy?" dialog box – We plan on editing the text in the "Save and Deploy?" dialog box to match the flow terminology used throughout the application, among other changes.

Trimming leading and trailing spaces – Leading and trailing spaces in flow settings are not trimmed, which may cause issues when configuring input and output connections.

"Loading" indicator when changing Transcription Automation Training setting – Unlike other flow settings that affect related settings, no "loading" indicator appears when the user enables or disables the Transcription Automation Training.

Internal settings shown in settings for custom flows – In some custom flows, editable flow settings may appear that correspond to internal settings that have been coded incorrectly.

Releases

Positioning of flow name and flow checkbox on the “Assign To Flow” dialog box – The flow name and the flow checkbox are misaligned and not centered in the Assign To Flow dialog box when assigning a release to a flow.

Identification

Marking a layout variation as incorrect during Identification Supervision and waiting for the task to complete – Marking a layout variation as incorrect during Identification Supervision and waiting for the task to complete results in the following:

The user receives a "This will end submission processing without extracting any data." warning. Instead, the user should receive a warning about the processing not continuing for the document that is marked with Layout Variation Incorrect.
The user receives a “Page not found” error message instead of being redirected to the Submission Output page.

Inconsistent location of the “Mark Layout Variation Incorrect” button – During Field ID QA, the Mark Layout Variation Incorrect button is located at the bottom of the right-hand sidebar. During Table ID QA, the Mark Layout Variation Incorrect button is located at the top of the right-hand sidebar.

Table Identification

[Fixed in v32.0.1] Opening and closing the Table ID keyboard shortcut sidebar in IE – After opening and closing the Table ID keyboard shortcut sidebar once, subsequent attempts to open the sidebar fail.

Workaround: Reload the page to open the sidebar again.

Table column names in the right-hand sidebar during Table ID Supervision – Table column names in the right-hand sidebar during Table ID Supervision use only around 50% of their field’s width.

Two users performing the same Table MVT task – If two users start performing the same Table MVT task from the Model Details page, and the first user completes the task while the second one closes the task using the X button in the upper-right corner, the second user will receive an error message.

Completed Table ID QA tasks and “Table Identification” card in the “Tasks Completed Today” section – In the Tasks Completed Today section of the Dashboard, the Table Identification card does not include a count of completed Table ID QA tasks.

Table ID Supervision tasks that take more than 4 minutes to complete – Table ID Supervision tasks that take more than 4 minutes to complete are abandoned and re-assigned to another keyer.

Selecting and resizing cells from multiple columns – Selecting and resizing cells from multiple columns causes cells to become blank if the cells are resized to an empty area. Instead, cells that are resized to an empty area should be removed.

Template-row identification

[Fixed in v32.0.1] Clearing a bounding box during template-row identification in Chrome and IE – Pressing the Esc key during template-row identification in Chrome and IE does not clear the selected bounding box.

Workaround: To clear a bounding box, use Delete or Backspace.

[Fixed in v32.0.1] “Next step” keyboard shortcut during template-row identification in Chrome and IE – Pressing the Enter or Return key during template-row identification does not complete template-row identification.

Workaround: To complete template-row identification and move to the table-identification review task, use Command + Return for Mac or Control + Enter for Windows.

[Fixed in v32.0.1] Find Missing Cells button appears during template-row identification – The Find Missing Cells for this column on all pages button ( ) appears during template-row identification, even though no rows can be automatically identified during that step of Table ID Supervision.

Table-identification review

[Fixed in v32.0.1] “Close task” keyboard shortcut during table-identification review in Chrome and IE – Pressing Option+Command+X for Mac or Alt+Control+X for Windows during table-identification review does not close the task.

Right-click and “Delete Cell” during table-identification review – Right-clicking on a cell and then clicking Delete Cell during table-identification review deletes the whole column.

Dragging the mouse during “Select in order” and “Multi select” – Dragging the mouse during Select in order (Shift + Click) and Multi select (Command + Click for Mac or Control + Click for Windows) breaks the selection process.

Row numbers and the text below them during table-identification review – The row numbers hide the text below them during table-identification review.

Pressing the “Esc” key while drawing a bounding box – Pressing the Esc key during bounding-box drawing while keeping the mouse clicked does not clear the bounding box. Instead, the bounding box remains on the page. To take another action on the table, the user needs to click a column in the right-hand sidebar.

[Fixed in v32.0.1] Overlapping table cells from different rows – If a table cell overlaps another cell from a different row, only one row is created instead of two.

Workaround: Manually create a new row and resize it to include all relevant table cells.

[Fixed in v32.0.1] Adding a table row to a multi-page document – If you:

have a multi-page document with a table that spans over two or more pages, and
you delete all identified rows on a page that is not the first page, and
you click the + button to add a row,

dragging the bottom border causes that row to stay outside the table boundaries.

Table Transcription

Workaround: Reload the page to open the sidebar again.

[Fixed in v32.0.1] “Next empty cell” keyboard shortcut and a hidden column during Table Transcription – If the right-hand sidebar hides a table column during Table Transcription and a keyer presses Control + E, the keyer isn’t moved to the next empty cell.

[Working as designed] “Override Validation” keyboard shortcut during Table Transcription – If a keyer presses F9 to override validation for a specific cell and then they click on another cell, Override Validation isn’t saved for the first cell.

Workaround: Press F9 and transcribe the cell before moving to another cell.

Classification Models

Small arrow appears when hovering over a classification model’s “Training Document Pages” – When hovering over a classification model’s Training Document Pages, a small arrow appears above the number of training document pages.

Table Models

[Fixed in v32.0.1] Table ID models and live releases that are assigned to multiple flows – If a live release is assigned to multiple flows, and the release contains a layout with a table that is also a part of other live releases, the Model Details page does not show information about the Table ID model, and the Run Training button is unavailable.

Flow Execution

Loading of Flow Runs for large flows – Flow Runs for large flows may take additional time to load, but during that time, there is no loading indicator to let the user know that the application is not frozen.

Timed-out blocks in Flow Run diagrams – Currently, there is no way to distinguish between blocks that are temporarily timed out and those that have reached their retry count.

Reporting

[Fixed in v32.0.1] Legacy data in System Transcription Sample Errors report – When Legacy data is selected in the flows filter in the System Transcription Sample Errors report (Reporting > Accuracy), the report's data cannot be downloaded.

Keyer Data Management

Appearance of disabled Save Changes button – The color of the disabled Save Changes button on the Annotations page matches that of active buttons in the application. As a result, users may think that there are changes to be saved when there are not.

[Fixed in v32.0.1] Pressing the Delete shortcut for a field without a bounding box – When a field without a bounding box is selected and the Delete shortcut is pressed, the Save button becomes enabled. However, there are no changes to be saved.

[Fixed in v32.0.1] Save Changes button when moving to another document – When moving to the next document in the list, the user is notified that any changes they made will be lost if they do not save them before proceeding. However, when the user moves to the next document, any unsaved changes they made remain intact, and the Save Changes button is still enabled.

[Addressed in v32.0.1] Focusing on selected fields – When a field is selected on the Annotations page, the interface does not bring that field into focus. We plan on changing this behavior to match that of the Supervision task interface.

Saving changes – We're fixing an issue that causes some saved changes to be reverted.

[Addressed in v32.0.1] Ordering of fields – The ordering of fields on the Annotations page is not consistent across documents that share the same layout.

[Addressed in v32.0.1] Time needed to save changes – We're working to reduce the amount of time it takes to save changes on the Annotations page.

Filtering Field Level Automation table for completed fields – The Field Level Automation table should be filtered to include only those fields that have finished processing.

Aggregating data for fields across layout versions – Data for fields in multiple layout versions is separated by layout version in Field Level Automation table. This data should be aggregated so that the field only appears once in the table.

Permissions

[Fixed in v32.0.3] Specifying task restrictions for submissions via API – Task restrictions for submissions that are specified via API are not applied to Supervision tasks.

Input Connections and Output Connections

Template Key field for Box Notifier Output connections – The Template Key setting for Box Notifier Output connections is required, but it is marked as optional in the UI.

[Fixed in v32.0.1] SSL CipherSuites and IBM Message Queue (MQ) connections – We're fixing an issue that causes input and output connections to IBM MQ servers to fail when SSL CipherSuites are enabled.

Removing a connection with validation errors – After removing an input connection with validation errors, a validation error is still shown in the input block.

[Fixed in v32.0.1] Default names of connections – If an imported connection does not have a name, the name should default to the name of the connection type and the date the connection was created. Currently, it defaults to the internal name for the connection and the timestamp for its creation date and time, with all text strings connected by underscores.

[Fixed in v32.0.1] Output key for UiPath Notifier Output connections – We plan on wrapping the API payload for UiPath Notifier Output connections in an output key. This change will make the formatting of this payload consistent with those of other output connections.

"Browse" button for Universal Folder Listener connector – We plan on adding a Browse button to the Universal Folder Listener connector, which would allow users to browse their file system as they did in the original Folder Listener connector.

File Storage

[Addressed in v32.0.1] Support for AWS Signature Version 2 in S3 file stores – We plan on adding support for AWS Signature Version 2 in requests sent to S3 file stores.

SaaS

[Addressed in v32.0.1] CSRF token in Client Libraries – We plan on adding CSRF tokens to the Java, Python, and C# versions of our Client Library.