KNIME Analytics Platform 4.4

KNIME Analytics Platform 4.4.4

Release date: January 19, 2022
(includes changes of 4.4.3, which was not released publicly)

Bug Fixes

• AP-18100: Generic Web Services Client fails on certain WSDLs with the 4.4 update
• AP-17734: Date&Time Widget: flow variable takes presence over manually changed values
• AP-17729: FlowObjectStack not properly updated on configure under special circumstances
• AP-17717: Credentials Widget does not work with the refresh button without a default value set
• BD-1154: (Big Data Extensions): Metadata browsing fails with Local Big Data Environment database
• AP-18232: Excel Writer fails to append to existing file using SMB Connector
• AP-18227: NPE in NodeContainerEditPart on Component Update
• AP-18211: XML Reader leaks input streams when using XPath with invalid reader settings
• AP-18196: Update log4j version used by xmlbeans to 2.17.1; prevents CVE-2021-44228 and others (though not exploitable in KNIME)
• AP-18126: Custom/KNIME URL file system does not properly handle HTTP responses with compressed Content-Encoding
• AP-18121: Index Reader node does not use connected file system
• AP-17693: Tableau: Error when writing hyperlog on Server
• AP-17576: Refresh Button Widget: Activated view / widget nodes not defined in the layout are missing after refresh button execution
• AP-17559: Excel Writer fails when writing to XLSX with Evaluate Formulas option enabled
• AP-17469: Refresh Button Widget causes String Widget to complain about an empty input
• AP-17457: Send to Power BI: Fix NPE if Power BI sends an unexpected error
• AP-17365: Filtered columns are still considered as duplicate column names
• AP-17072: Salesforce Simple Query node does not properly URLencode the where clause
• AP-15679: Weird Prompt comes up in New File Handling framework
• BD-1153: (Big Data Extensions): PySpark: Invalid Python code inserted when clicking on input column
• BD-1150: (Big Data Extensions): Databricks File System Connector/Create Databricks Environment nodes should handle HTTP 401 response on invalid credentials

KNIME Analytics Platform 4.4.2

Release date: October 21, 2021

Bug Fixes

• AP-17454: XPath node fails to parse XML documents with more than 2000 lines on non-English locale
• AP-17443: Deeply Nested Components won't show views
• AP-17416: DB metadata browser should expand only first hierarchy level
• AP-17403: Call workflow table-based does not recognise Container input variable
• AP-17630: Dragging and dropping XML files to a workflow calls deprecated XML Reader
• AP-17544: JSON/XML/Image Writer node dialog does not detect port-based file system connection
• AP-17520: DB functions starting with $ can make KNIME unresponsive
• AP-17484: Microsoft Authentication node claims success but does not provide authentication
• AP-17470: Excel Reader only hides first column in a group of consecutive hidden columns when reading XLSB files
• AP-17461: Refresh button deactivates interactive filters widget
• AP-17456: SAP Reader (Theobald Software) fails when reading missing date and time fields
• AP-17438: Excel Reader shifts data if column on the left of a hidden column contains missing values
• AP-17430: Call Workflow (Table Based) node doesn't delete jobs after configure
• AP-17425: Fix "Include only explicitly installed" button of Conda Env Prop node on target machine
• AP-17407: Table Row to Variable node no longer able to handle columns of missing type
• AP-17362: Widgets: flow variable-controlled configs are not transferred to view value on re-execution
• AP-17234: Workflow Writer fails on opening written workflow if Explorer is not initialised
• AP-17172: Typo in Workflow Editor Settings
• AP-17003: Non-deterministic output in String Manipulation Multi Column Node
• AP-16812: File Upload Widget does not percent-encode illegal characters in knime:// URLs
• AP-16741: Cannot update mountpoint-relative components when opening workflow from server as local copy
• AP-16740: Memory leak when hiliting is enabled leads to executors running out of heap space
• AP-16066: Heatmaps in the Same Component Plot the same Columns
• AP-13839: Number to String and String to Number check included columns, even if exclude list is enforced

KNIME Analytics Platform 4.4.1

Release date: August 25, 2021

New nodes

• AP-17218: XML Reader (new file handling)
• AP-13421: JSON Writer (new file handling)
• AP-13420: XML Writer (new file handling)
• BD-1116: (Big Data Extensions): Spark Logistic Regression Learner (based on spark.ml)
• BD-1115: (Big Data Extensions): Spark Linear Regression Learner (based on spark.ml)

Enhancements

• AP-16893: SVM Learner: Better error message when target column contains constant value
• AP-16540: SAP Reader (Theobald Software) with support for Kerberos authentication

Bug Fixes

• AP-17355: Save As button sometimes disabled (but shouldn't)
• AP-17348: Problems deserializing AdapterCells containing blobs after restoring workflow
• AP-17317: Amazon S3 Connection (legacy) throws NoClassDefFoundError
• AP-17301: Path To String and Path To String (Variable) nodes do not work with workflow-relative paths if executed on KNIME Server
• AP-17259: External Tool node Error / Output View does not work anymore in 4.4.0
• AP-17253: Windows: KNIME 4.4. up to 30% slower than KNIME 4.3 (not using "server" VM)
• AP-17250: Arrow Python serializer fails with environments created by preference page on Windows
• AP-17243: JSON To XML Node generates an invalid XML
• AP-17210: Component views only work for top-level components
• AP-17201: Reader and writer node dialogs do not open in RWE anymore
• AP-17160: Drag and drop of shortlinks from workflows etc. not working anymore
• AP-17158: Dialog of Create Date&Time Range node cannot be closed
• AP-17126: Fix Metadata Browser on database connections without multi DB support
• AP-17102: Workbench focus lost when configuring node
• AP-17371: Parquet Reader and Writer fail with SMB Connector
• AP-17370: FTP Connector: Unable to list directory if it's name contains spaces
• AP-17369: FTP Connector: concurrency issue with the createDirectory method
• AP-17368: FTP Connector: "Execute failed: 150" error when running Table Reader node
• AP-17367: FTP Connector: OutputStream is extremely slow
• AP-17330: Deep learning: Installation tests do not use variable-controlled Python environment
• AP-17295: AP can't extract creation date from workflow
• AP-17230: NPE in Decision Tree to Ruleset if tree contains only the root node
• AP-17206: File Reader node description should refer to the Complex File Reader
• AP-17183: Python Object Writer has wrong default script
• AP-17155: Double Widget node overrides user input in scientific notation when lowercase 'e' is used
• AP-17152: Filter event fired on re-load after re-execution (Refresh Button) event does not contain filter information
• AP-17146: Typo: "Specifiy" in the Setup Component Wizard
• AP-17128: Deep learning conda environments specified via executor.epf are ignored
• AP-17108: No Focus on dialog when opening node configuration
• AP-17078: Closing and saving a workflow sometimes throws NPE
• AP-17021: MultiLayerPerceptron Predictor node: Node configuration is not in sync with node description
• AP-16977: Simple Streaming Executor can not handle inactive ports
• AP-16961: File Chooser Widget node leads to following error when clicking the refresh button
• AP-16872: Node view values not updated on config- or input-spec-change
• AP-16723: Default configuration of Strings to Document node can lead to NPE errors
• AP-16688: Conda Env Prop node needs to clean up after itself when conda env re-creation fails
• AP-16565: Python: Dialog panes of scripting nodes throw NullPointerException/IndexOutOfBoundsException
• AP-16333: Double Configuration node overrides user input in scientific notation if lowercase 'e' is used
• AP-16299: Credentials Widget does not save KNIME Server Login
• AP-15951: Opening workflow with “External Tool” node with invalid dialog settings causes errors even if configured using flow variable
• AP-15664: Keras: Unmaterialized port object does not use variable-controlled Python environment
• AP-15366: Component on server can't be overwritten from KNIME Explorer
• AP-13732: Keras: LSTM layers cause OpenMP Error #15 (Mac only)
• WEBP-824: Downstream metanodes prevent re-execution of single page

KNIME Analytics Platform 4.4.0

_{(see highlight summary)}

Release date: June 30, 2021

New nodes

• AP-17077: MongoDB Connector (multiple new nodes, extension rewritten)
• AP-10494: Snowflake Connector
• AP-15952: Refresh Button Widget (for Dynamic Data Apps)
• AP-13331: Workflow Reader
• AP-16816: Workflow Summary Extractor
• AP-3325: Variable Creator
• AP-12355: Container Input (File)
• AP-10537: H2O Integration: H2O AutoML Learner
• AP-16680: Container Input (Variable) (revised)
• AP-16669: Container Output (File)
• AP-16612: DB Table Creator with new table selection component
• AP-16808: JSON Reader (new file handling)
• AP-16567: File Reader (new file handling)
• AP-16566: File Reader (Complex Format)
• AP-15932: Compress Files/Folder table input (new file handling)
• AP-14932: Transfer files table input (new file handling)
• AP-13864: Image Writer (Table) (new file handling)
• AP-13418: Table Writer (new file handling)
• AP-13414: Table Reader (new file handling)

Enhancements

• AP-15826: Upgrade to Java 11 and Eclipse 2021-03
• AP-17014: Cleanup Preference Pages and Wizards
• AP-16967: Keep original exceptions as cause in KafkaConnectionValidator.java#checkConnection
• AP-16946: Node Repository: Move Conda folder up from Python to Scripts folder
• AP-16940: Enable R nodes to directly use conda.environment flow variables
• AP-16843: Amazon S3 Connector: Add SSE-C support
• AP-16842: Add timeout option(s) to "Send to Tableau Server" node
• AP-16779: Rename node category and change node description … of 'Single Page Application'.
• AP-16718: Update buillt-in mysql5 driver to version 5.1.49 to support TLS 1.2
• AP-16712: Chromium update to version 91
• AP-16701: Update Jackson to 2.12.1
• AP-16662: Show only selected data on start scatter plot
• AP-16659: Support custom client id and secret in Google Authentication node
• AP-16605: Additional date and time functions for Column Expression node
• AP-16556: Table Validator with new option to validate only on execution (needed in Try-Catch constructs)
• AP-16539: Arrow: Unmap MappedByteBuffer when they are not needed anymore
• AP-16516: Metadata support for databases without schemas e.g. Firebird
• AP-16482: JS View Image Generation using CEF
• AP-16448: Integrated Deployment: Capture Start/End node with new option to propagate 'scope' variables
• AP-16447: Try-Catch node with new option to propagate 'scope' variables
• AP-16446: Loops with new option to feed back modified variables to their loop start nodes
• AP-16440: Support Path type in Column Expressions node
• AP-16433: Java Snippet to support Java 11 language feature
• AP-16365: Return missing values if FS does not support permission look ups
• AP-16267: Add single page re-execution in the AP for the CEF
• AP-16180: Always allow String as KNIME type in transformation tab of Parquet/ORC Reader
• AP-16067: Expose file size, creation and modify date for KNIME Server file system
• AP-16030: Always allow String as KNIME type in transformation tab of Excel Reader
• AP-16028: Allow to set the KNIME type of unknown columns in transformation tab of reader nodes
• AP-16027: Always allow String as KNIME type in transformation tab of CSV Reader
• AP-15971: Update H2O to 3.32.1.2
• AP-15884: Show Nodefactory class in Nodemonitor
• AP-15876: Allow merging of different FSLocationCells
• AP-15871: More Java to KNIME Converters for Numeric Values
• AP-15808: Add executable selection tab to all Python nodes
• AP-15802: Name of conda env flow variable should be user-defined
• AP-15387: Enable compression in Apache Arrow Backend
• AP-15202: User defined Hub view
• AP-15123: Append File Rights Attributes to Files/Folders Meta Info (new filehandling)
• AP-15078: Integrated Deployment: String representation of a workflow should be its custom workflow name if set
• AP-15006: Send to Tableau: Add option to authenticate with "Personal Access Token"
• AP-14999: Joiner node to be able to match different column types
• AP-14778: GET Request node to support query retry with delay
• AP-14724: UI Improvement “Workflow Combiner”
• AP-14723: Small UI Improvement “Capture Workflow End”
• AP-14577: Join with Match Any
• AP-14401: Revised help menu
• AP-14326: Power-BI: Allow for Table Relationships
• AP-14006: Add OS agnostic line break option to csv reader
• AP-13961: Support reading of password protected Excel files
• AP-13949: Create option in reader nodes to add file identifier when reading multiple files without loop
• AP-13933: Support multi database in db nodes
• AP-13899: Support multiple projects when working with BigQuery
• AP-13049: Add success message after successfull deployment of preferences
• AP-12193: DB Loader support for Snowflake
• AP-8195: Change line thickness in Java Script based line plot
• AP-7397: REST nodes should have an option to disable chunked transfers
• AP-5213: "Run only during execute" in Java Edit Variable node is confusing
• BD-1117: (Big Data Extensions): Add support for Spark 3.0 to KNIME Livy client
• BD-1103: (Big Data Extensions): Update Create Local Big Data Environment to Spark 3.0.2
• BD-1102: (Big Data Extensions): Add support for Spark 3.0 to KNIME Extension for Apache Spark

Bug Fixes

• AP-16763: Layout and Configuration Editor is not working on Linux (due to some lib changes on the OS level)
• AP-16730: Table to PDF & HTML cannot write pngs
• AP-16486: External NPM version dependencies cause IE11 to fail
• AP-16040: Twitter Streaming node failure
• AP-17061: Test Data Generator fails with only a few rows using fast tables
• AP-17059: Boolean Sets with only a few rows cause unsafe memory access in fast tables
• AP-17031: Bundle org.knime.ext.ftp.filehandling fails to initialize
• AP-16950: Active Learner Loop Start/End nodes cause NPE / error dialog in case "favorite nodes" view is open
• AP-16943: Chrome Driver fails to render images on output ports on Windows
• AP-16939: Workflow Reader: Incorrect node message when no workflow is specified
• AP-16903: Move file name column to the end rather than prepending it in reader nodes
• AP-16888: DB Loader loads wrong values for boolean columns in MySQL
• AP-16887: NullPointerException in Call Local Workflow (row based) node when report is generated
• AP-16880: Excel Reader cannot handle numeric cells with missing values but without cell type
• AP-16875: Improve loading indicator – Define a max-height for the loading indicator
• AP-16844: SSH Connector: Fix file attribute cache issues after copy/move
• AP-16827: The transformation tab resets the column order when the dialog is reopened
• AP-16825: DynamoDB nodes lose rows when table spec changes in-between read batches
• AP-16824: FTP Connector cannot browse or list files on Windows IIS FTP server
• AP-16813: NPE in Parameter Optimization Loop Start
• AP-16807: Annotation text on Linux sometimes not readable (font not installed)
• AP-16783: Path to URI with presigned URLs should use normalize
• AP-16774: Extract system properties exports application-wide ('static') proxy passwords in cleartext
• AP-16768: NullPointerException in Files/Folders Meta/Info node on paths from SSH Connector
• AP-16761: Add better error message for SPA not working in Bundled Chromium and SWT Browser
• AP-16734: Workflow Writer / Deploy to Server nodes do not work if the server URL contains a port and contains a trailing slash
• AP-16697: Fix different behavior of Clock.systemDefaultZone().instant()
• AP-16682: Community converter can overwrite KNIME converter
• AP-16660: Make WordWrapJLabel a fixed size label
• AP-16643: MacOS Big Sur: Various screen artifacts (distorted table, broken splash screen, etc...)
• AP-16627: Context menus of components linking to a disconnected Hub space cannot be opened
• AP-16585: Conda Environment Propagation Dialog forever loads if Python is not set up
• AP-16580: Fix Locales for extract date & time
• AP-16570: Python: File extension mapping of pickled files (.pkl) maps to deprecated Python Object Reader node
• AP-16549: Workflow executor fails if directly connected to workflow input port (i.e. in a metanode)
• AP-16538: "Workflow Executor" exports variables in reverse order
• AP-16488: Race condition (and memory leak) when data is retrieved and evicted from the cache at the same time
• AP-16471: Google nodes that use the Google Authentication output port sometimes fail with "Unable to clone input data at port"
• AP-16456: Create Temp Dir throws NPE for Relative to FS after workflow is closed
• AP-16395: DB Transaction End node shouldn't fail with inactive input ports
• AP-16289: Conda Environment Propagation: Kernel queue is not cleared upon failure during environment creation
• AP-16214: Python: Deprecated port type selectable in component setup
• AP-16211: NodeSettingsMigrationManager#getSourceVariableSettings returns destination variable settings instead
• AP-16160: NPE when closing the Workflow Coach
• AP-16124: Column selection set to "Document" instead of "Preprocessed Document" in workflow example
• AP-16110: Mountpoint file system cannot connect to server workflow repository when running workflow on KNIME Server
• AP-16062: Add data to Reference Reader nodes recursively (for workflow ports)
• AP-16054: Domain Calculator throws exception if Path column is included
• AP-15972: KNIME Server Connector and Relative to file systems treat workflows differently
• AP-15969: NPE on DnD of an URI lacking a host
• AP-15886: Installing extension not on update sites throws NPE
• AP-15882: Problem with connections reaching into a capture scope whose source is the parent workflow (i.e. metanode port)
• AP-15879: Fix virtual scope context retrieval from metanode
• AP-15877: Parallel Chunk Loop should retain port objects which lead into a capture scope for later usage
• AP-15862: TRF based Readers don't recalculate the spec if spec affecting settings are controlled via flow variable
• AP-15722: Wrapping/scrolling of status message doesn't work properly
• AP-15623: FileReader throws wrong exception text
• AP-15431: SettingsModelReader/Writer show filter options though they are not available
• AP-15066: Workflow Writer displays confusing warning on escaped default workflow name
• AP-15030: 'Search'-menu entry in main menu (should be removed)
• AP-14672: Text Output Widget internals not reset consistently with other nodes
• AP-14627: TensorFlow 2 Reader: Reading a network in SavedModel format from TensorFlow Hub fails with ProviderMismatchException
• AP-14554: Check if “Custom workflow name” in ”Capture Workflow End“ node has illegal characters
• AP-13893: Problem with credentials flow variables when injected via reference reader node
• AP-13769: Configure in Joiner slow for wide tables
• AP-13500: Handle missing port object types when loading workflow segments
• AP-13171: Send to Power BI node: Upload with special column name fails
• WEBP-779: Progress reset on refresh for Data-Apps — Include full set of re-executed node IDs in addition to the re-executing node IDs