Graph Explorer Change Log

August 4, 2026 · View on GitHub

Release 3.2.0

This release makes styling your graph easier to manage and share. Node and edge styles now live together in one panel, previews everywhere show your nodes and edges exactly as they appear on the canvas, and a new Styles settings page lets you save your styles to a file and selectively load them on another machine or browser. This release also makes Graph Explorer reliable across multiple browser tabs, with connections now scoped per-tab and your data no longer lost between tabs, plus a set of fixes to connections, the proxy server, and the Schema View.

Styling

  • One Styles panel. Node and edge styles used to be two separate sidebar tabs. They're now combined into a single Styles panel with a Nodes/Edges switch, in both the Graph View and the Schema View. A gear button in the panel header jumps to the Styles settings page.
  • Canvas-accurate previews everywhere. Node and edge previews now render the true shape, color, border, icon, and label instead of a generic blue circle. This applies across search results, the legend, the Schema View sidebar, and the style dialogs. The node and edge style dialogs each show a prominent live preview that updates as you customize.
  • New Styles settings page. A dedicated page in Settings is the home for saving, loading, and resetting your styles.
  • Save and share your styles. Save your styles to a file and load them on another machine or browser to reproduce your look. Loading is selective, so you preview each type's current style beside the incoming one and choose exactly which to apply.
  • Reset your styles. A single Reset action on the Styles settings page returns every type to the defaults.

Reliable Across Multiple Tabs

Graph Explorer stores all your data in your browser. This release makes that dependable when you have more than one tab open.

  • Per-tab connections. The active connection is now scoped to each tab, so switching connections in one tab no longer changes what every other tab is viewing. When you reopen Graph Explorer after closing all tabs, it resumes the connection you most recently used.
  • No more cross-tab data loss. Connections, schemas, styles, and sessions saved in two tabs at once now merge instead of one tab's copy overwriting the other's.
  • Save-failure recovery. When a write to browser storage fails, such as when storage is full, a "Changes not saved" indicator now appears in the nav bar with the failure details. When storage is full, it also offers to back up your configuration to a file, rather than losing the change silently on reload.

Bug Fixes

  • Clicking a connection that's already active no longer resets it, so you no longer lose your in-progress graph by re-selecting the connection you're on.
  • Blank nodes now show the default icon on the canvas, matching the details panel (thanks @arnavnagzirkar!).
  • The proxy server now stops with a clear message when its port is already in use, and shuts down cleanly instead of lingering in the background.
  • Removed six rounded node shapes that rendered as rounded diamonds and made connected edges disappear at the size nodes are drawn on the canvas. Any node already using one now falls back to its straight-sided equivalent.
  • Fixed a thin inset gap around dialog footers, so the footer background and divider now reach the dialog edges.
  • Connection URLs with stray whitespace or line breaks are now cleaned up, both when you save a connection (thanks @bakar-dev!) and when an existing connection is used, so a pasted URL with a trailing newline no longer breaks the connection.

Schema View

  • Node types and edge types in the Schema View details panel are now clickable, selecting that type in the graph and keeping the sidebar and canvas highlight in sync (thanks @1wos!).
  • The Schema View sidebar can now collapse and expand, and remembers its width and active tab across sessions, matching the Graph View sidebar.
  • The Schema View can auto-open the Details panel when you select a single type.

New Contributors

Welcome and thank you to our first-time contributors!

All Changes

Full Changelog: https://github.com/aws/graph-explorer/compare/v3.1.0...v3.2.0

Release 3.1.0

This release adds a built-in icon library, faster schema sync on large graphs, a new proxy security option, and several compatibility fixes.

Built-in Icon Library

Node styling now includes a searchable picker with thousands of Lucide icons (thanks @jkemmererupgrade!). Select an icon from the grid and it's stored as a symbolic reference (like lucide:plane) rather than a base64-encoded data URI. Custom uploaded icons continue to work as before.

Faster Schema Sync

Two changes combine to speed up schema sync on large Gremlin graphs. First, the schema queries now use traversal patterns that run natively on Neptune's DFE engine across all tested versions (1.2.1.0 through 1.4.7.0). Second, the summary API now requests only the fields Graph Explorer actually uses, skipping unused structural metadata.

Bug Fixes

  • Neighbor counts no longer fail on Neptune 1.3.x. The query now uses groupCount().by(label) which works consistently across all tested engines.
  • Databases with non-root URL paths (like BlazeGraph's /blazegraph/namespace/kb/sparql) now work correctly. The proxy previously replaced the entire base URL path when constructing endpoint URLs.
  • Node borders now render as expected. Per-type styles were missing a border opacity override, making borders invisible regardless of user settings.
  • SPARQL query results now recognize ?s ?p ?o variable names as shorthand for ?subject ?predicate ?object (thanks @vishwakt!).

Other

  • New PROXY_SERVER_ALLOWED_DB_ORIGINS environment variable restricts which database origins the proxy will forward to, mitigating SSRF risk. This only applies to proxy-routed requests. Direct browser connections bypass this check entirely.

New Contributors

Welcome and thank you to our first-time contributors!

All Changes

Full Changelog: https://github.com/aws/graph-explorer/compare/v3.0.3...v3.1.0

Release 3.0.3

This patch release blocks cross-origin requests by default, improves performance for larger schemas, adds inferred edge connections in search and expansion, and adds a diagnostic logging setting.

  • Schema Explorer — edge connections are now inferred from graph exploration, so the Schema Explorer shows relationships discovered through search and neighbor expansion without waiting for the database summary API
  • Performance — schema operations are significantly faster for graphs with many vertex or edge types, reducing unnecessary re-renders when the schema is already up to date
  • New feature — a diagnostic logging toggle in settings enables verbose console logging at runtime, even in production builds, for easier troubleshooting
  • Documentation — getting started restructured as a hands-on tutorial, new configuration reference, new Neptune public endpoints guide, new architecture documentation
  • Infrastructure — hardened Docker image and CI workflows, migrated to oxfmt and oxlint for faster formatting and linting, TypeScript upgraded to 6.0, proxy server runs natively without a build step

Cross-Origin Request Blocking

The proxy server now blocks cross-origin requests by default instead of allowing all origins. Since the proxy server serves both the API and the UI from the same origin in all standard deployments (Docker, SageMaker, ECS Fargate), CORS is not needed.

If you serve the UI from a different origin than the proxy server, set the PROXY_SERVER_CORS_ORIGIN environment variable to the UI origin. See the Security documentation for details.

All Changes

Release 3.0.2

This patch release fixes a bug where schema sync did not automatically trigger when switching connections and adds a new configuration option for controlling allowed origins.

  • Bug fix — schema sync now automatically triggers when switching to a connection that has no cached schema
  • Configuration — new PROXY_SERVER_CORS_ORIGIN environment variable lets you explicitly control which origins are allowed to connect to the proxy server. See the Security documentation for details.

All Changes

Release 3.0.1

This patch release improves error handling, refactors the proxy server for testability, and hardens the application protection mechanisms. Error messages surface richer diagnostics — status codes, response bodies, and cause chains — so troubleshooting is more useful. We also upgraded to Vite 8, cutting build time by 60% and bundle size by 5%.

  • Bug fixes — new empty state when no connections are configured instead of misleading "No Schema Available", Docker entrypoint now respects custom config directories (thanks @theneelshah!), Podman container permissions fix
  • Error handling — the error details dialog now surfaces status codes, response bodies, and cause chains so you can diagnose issues without digging through logs. Invalid proxy requests return proper 400 errors instead of cryptic 500s. Connection failures and CORS mismatches get targeted error messages.
  • Tooling & docs — Vite 8 upgrade cuts build time by 60% and bundle size by 5%, reorganized docs into guides and references, updated README
  • Application hardening — tighter CORS defaults, supply chain hardening, automated vulnerability scanning, least-privilege CI permissions, and a new security policy for reporting vulnerabilities
  • Proxy server — previously untestable parts of the proxy server now have 170+ tests (up from 56), making future changes safer and more reliable

HTTPS Configuration

Previous versions had a bug where the Docker entrypoint ignored custom config directory paths, which could cause your HTTPS settings to be silently skipped. While fixing this, we also discovered that when HTTPS was enabled but certificate generation or discovery failed, the server would silently fall back to HTTP instead of reporting the problem. The server now exits with a clear error in this scenario. If you were unknowingly relying on this fallback behavior, ensure your certificates are in place before upgrading or explicitly disable HTTPS. See the HTTPS Connections documentation for details.

New Contributors

Welcome and thank you to our first-time contributor!

All Changes

Release 3.0.0

Graph Explorer 3.0 is here! This release brings one of the most requested features — the ability to visualize your graph database schema — along with a fresh navigation experience and a handful of quality-of-life improvements.

If you're upgrading from a previous version, all your existing connections, preferences, and configuration data will carry over unchanged.

Schema Explorer

You can now see the shape of your graph at a glance. The new Schema Explorer renders your vertex types, edge types, and their connections as an interactive schema graph. Drill into any type from the sidebar to see property details and data types. Edge connection discovery maps out how your types relate, giving you a bird's-eye view of your entire graph structure.

The schema is built by sampling your graph data, so it reflects what Graph Explorer has seen so far and may not capture every type or property in your database.

Redesigned Navigation

Getting around Graph Explorer just got easier. The navigation bar has been rebuilt as a clean, static top bar for moving between the Graph Explorer, Data Explorer, and Schema Explorer. Sidebar tabs have been refreshed to match the new look.

Data Explorer Improvements

The Data Explorer now includes a vertex type switcher for faster browsing across types and a new export button to download data as CSV or JSON (thanks @dwrth!).

Other Improvements

  • Rewritten RDF prefix generation and replacement logic
  • Standardized terminology across Gremlin, openCypher, and SPARQL
  • Very large and very small numbers now display using scientific notation
  • Keyword search labels are clearer and more descriptive

New Contributors

Welcome and thank you to our first-time contributors! 🎉

All Changes

Full Changelog: https://github.com/aws/graph-explorer/compare/v2.5.2...v3.0.0

Release 2.5.2

This release bumps the Node & PNPM versions and cleans up the Docker image.

All Changes

Full Changelog: https://github.com/aws/graph-explorer/compare/v2.5.1...v2.5.2

Release 2.5.1

This release includes a fix for a regression that caused neighbor expansion in SPARQL databases to perform poorly.

All Changes

Full Changelog: https://github.com/aws/graph-explorer/compare/v2.5.0...v2.5.1

Release 2.5.0

This release focuses on improving the graph exploration experience with better debugging tools, enhanced node interactions, and performance optimizations.

New Features

  • Raw JSON Response Viewer: View query results as formatted JSON with syntax highlighting and copy functionality for better debugging
  • Enhanced Node Expansion: Expand single or multiple selected nodes simultaneously through an improved context menu
  • Graph View Improvements: Added toggle buttons to the empty state and updated the re-layout button icon for clarity

Improvements

  • Better Context Menu: Reorganized options with new abilities to center/zoom to selected items and remove all selected items
  • Performance: Faster app startup by lazy-loading Cytoscape and other heavy dependencies (38% reduction in initial bundle size)
  • UI Polish: Updated table and notification styling, plus fixed node stacking issues in graph rendering
  • Stability: Resolved race conditions that caused inconsistent behavior during app initialization

All Changes

Full Changelog: https://github.com/aws/graph-explorer/compare/v2.4.1...v2.5.0

Release v2.4.1

This release includes several important bug fixes and improvements, notably:

  • Added ability to manually refresh node or edge data from UI
  • Updated graph data to mirror the most recent data from searches and queries
  • Updated handling of multi-label nodes when patching the schema
  • Fixed auto-open details panel behavior when selecting entities
  • Fixed representation of default values in node & edge styles
  • Fixed several layout issues around long labels

All Changes

Full Changelog: https://github.com/aws/graph-explorer/compare/v2.4.0...v2.4.1

Release v2.4.0

This release introduces support for SPARQL queries within the query editor. Now, all three query engines are supported: Gremlin, openCypher, and SPARQL. This does not mean we are done with the query editor. We have many exciting ideas being considered for future releases.

SPARQL Query Support

  • Support for SELECT, ASK, DESCRIBE, and CONSTRUCT queries
  • DESCRIBE and CONSTRUCT queries will result in fully materialized vertex and edge results
  • SELECT and ASK queries will result in raw statements, but do not materialize results as vertices or edges
  • Support for RDF resources without a defined rdf:type
  • Updated display name defaults to use rdfs:label if it is available

Other notable changes

  • Added support for vertices that have no label in openCypher
  • Hide properties that don't have a value for the given vertex or edge
  • Added confirmation dialog when deleting a connection (thanks @dwrth)
  • Added ability to horizontally scroll toolbars if space is limited (thanks @Ansh2004P)
  • Added zoom to fit toolbar button (thanks @cnaples79)
  • Updated the strings used to represent no value, no type, and empty value to be more clear
  • Updated handling of neighbor counts when neighbors have more than one type or label
  • Updated handling of date values, specifically in openCypher connections
  • Updated behavior of auto open details panel when a node is selected
  • Fixed many bugs

All Changes

New Contributors

Full Changelog: https://github.com/aws/graph-explorer/compare/v2.3.1...v2.4.0

Release v2.3.1

This release resolves a few important issues from the previous release.

All changes

Full Changelog: https://github.com/aws/graph-explorer/compare/v2.3.0...v2.3.1

Release v2.3

This release improves the accuracy when rendering query results by preserving result names, properly rendering grouped data structures like maps, embedding source and target node information within edge results, and refining the overall results interface for better clarity.

Major changes

  • Updated query and search UI with nested values
  • Added support for named results
  • Added nested source & target vertex results inside of edge results
  • Added support for grouped data structures like maps and paths
  • Updated vertex and edge results to highlight when added to the graph
  • Fixed table view filter and sort reset when toggled off/on (thanks @dwrth 🎉)
  • Updated Gremlin expand neighbors query performance by removing sorting

All changes

New Contributors

Release v2.2

This release brings significant performance improvements for the query editor, restoring graph session, and other operations that typically need to query for details about many nodes & edges all at once.

In prior releases, this would manifest in multiple queries per node or edge, which could bog down the browser and overload the servers. With this release, similar types of requests are now batched together to reduce the number of requests down to around 3 requests in typical situations. This leads to better utilization of network bandwidth, reduces the load on both browser and server, and results in a more responsive UI.

Major changes

  • Updated query logic to prefer batching similar requests in groups of 100 instead of individual requests (#1044, #1048, #1065, #1068)

Other changes

  • Updated roadmap to move SPARQL query editor up in priority (#1066)
  • Updated imperative query logic to check for a cache value first (#1047)
  • Updated Tanstack Query logic to simplify query construction (#1015, #1040, #1042)
  • Fixed rendering boolean values for Gremlin connections (#1034)
  • Added troubleshooting steps for mismatched proxy server to documentation (#1062)
  • Fixed brittleness of some tests (#1070)

Release v2.1

This release continues the momentum of the last release, introducing query editor support for openCypher connections. This enables the use of the query editor with Neptune Analytics for the first time.

As before, if you are a Neptune user concerned about data integrity, check out the instructions on configuring IAM permissions to restrict mutations, ensuring that users can enforce read-only operations.

New features & major changes

Other changes

  • Added close button to table view (#1026)
  • Updated label for properties count in connection detail (#1030)
  • Updated dependencies to latest versions and some minor refactorings (#1014, #1023, #1025, #1027)

Release v2.0

Graph Explorer v2.0 marks a significant milestone with the introduction of the new query editor for Gremlin connections. This powerful feature allows users to enter any valid Gremlin query and visualize the returned nodes, edges, and scalar values directly from the database. You can review the results in the sidebar and choose to add all nodes & edges to the graph or add results individually.

For Neptune users concerned about data integrity, the README now includes instructions on configuring IAM permissions to restrict mutations, ensuring users can enforce read-only operations.

The release also changes the default strategy for representing node & edge labels in the UI. The app no longer performs any transformations on these names, providing a more intuitive and accurate representation of your data. For RDF datasets, namespace prefixes will continue to be substituted.

Additionally, graph layout options now have a more intuitive organization, and new directions have been added for hierarchical, subway, and klay layouts.

New features & major changes

  • Added query editor for Gremlin connections (#686, #949, #947, #956, #957, #974, #958, #989, #991)
  • Changed node & edge label strategy to better represent true database names (#996)
  • Updated graph layout options to include better descriptions and new directions for hierarchical and subway (#973)

Other changes

  • Updated too many requests error message for accuracy (#990)
  • Fixed multiple selection box behavior (#987)
  • Fixed resizing columns in table view (#988)
  • Fixed search result row separators to ensure they are always rendered properly (#997)

Release v1.16.0

This release of Graph Explorer introduces significant usability improvements and performance enhancements. Users now have greater control over their exploration experience with the ability to customize default neighbor expansion limits and resize the sidebar. The connection interface has been refined with more intuitive placement of the sync button near the last sync timestamp.

Performance has been enhanced across multiple areas, particularly for RDF datasets where neighbor expansion is now faster. The openCypher expand neighbor query has been optimized for scenarios without limits, and the neighbor count query now shows the full count of neighbors without restrictions. Under the hood, the application has undergone significant architectural improvements, transitioning from Recoil to Jotai for state management and implementing the React Compiler to boost overall rendering performance. These technical updates, along with numerous dependency updates and bug fixes, result in a more responsive and reliable experience.

New Features

  • Added ability to customize default neighbor expansion limit (#925)
  • Added ability to resize the sidebar (#938)

Other changes

  • Improved Improved connection details by moving the sync button near the last sync timestamp (thanks @enumura1, #901, #908)
  • Improved performance of neighbor expansion in RDF datasets (#942)
  • Improved openCypher expand neighbor query to be faster when no limit is provided (#924)
  • Removed limit on neighbor count query in order to always show the full count of neighbors (#924)
  • Fixed issue where a schema sync would not automatically run when a connection was changed (#919)
  • Updated localForage Recoil integration to be async and use Suspense (#883)
  • Removed Recoil state debugging tool that was never used (#909)
  • Updated the state management layer to use Jotai instead of Recoil (#896, #920, #934)
  • Updated to use the React Compiler to improve performance and simplify code (#916)
  • Updated dependencies and minor refactoring of code (#922, #926, #930, #931, #932, #933)

Release v1.15.0

Graph Explorer now offers session persistence, allowing you to seamlessly continue your work. With a single click, you can restore your previous graph visualization instead of starting from scratch. This feature retrieves the most current information for all nodes and edges from your last session, ensuring you're working with up-to-date data.

This release also resolves a couple of long standing issues. The first is that double click expansion is now much more reliable and consistent across all three query languages. The second is that SPARQL neighbor counts and expansion is much more accurate and reliable.

And as always, there are many additional small fixes and improvements.

New Features

  • Added ability to restore the graph from the previous session (#826, #840)
  • Added add /status endpoint to verify proxy health (thanks @ssheladiya, #833)
  • Added ability to see full error details from errors in the UI (#858)
  • Added ability to cancel a long running schema sync (#869)

Other changes

  • Updated graph manipulation logic to lay foundations for query editor (#864, #848, #853, #850, #843, #842, #839, #837, #838)
  • Updated styling of the connections screen (#828)
  • Updated namespaces sidebar to use tabs instead of dropdown (#830)
  • Fixed rendering performance issues adding to the graph and showing the entity filters or node & edge style sidebars (#892)
  • Fixed issues with filtered neighbor expansion and neighbor counts in RDF databases (#870)
  • Fixed issue representing metadata classes as valid instance types during schema sync in RDF databases (#903)
  • Fixed issue where double click expansion was inconsistent (#841)
  • Fixed issue with table exports (#860)
  • Fixed issue with long node titles or descriptions pushing the "add to graph" button off the screen (#824)
  • Fixed issue rendering search results with really long titles (#904)
  • Fixed issue representing an edge connecting nodes with multiple labels (#839)
  • Fixed proxy server error when a request is aborted mid-stream (#873)
  • Updated icons in context menu to be more consistent (#906)
  • Updated styling of buttons, input, textarea, and select fields (#847, #832, #834)
  • Updated tests to be less fragile (865)
  • Updated HMR behavior to ignore test files (#835)
  • Fixed Docker build by not removing gzip (#868)
  • Updated dependencies (#827, #849, #866, #876, #900)

Release v1.14.1

This release is a minor bug fix release, with a primary focus on surfacing schema sync errors to aid in diagnosing issues.

  • Improved reliability of schema syncing, retrying on failure (#813)
  • Improved handling of errors when fetching data from the database (#812)
  • Changed to allow editing and deleting default connections (#801, #821)
  • Changed service type label to be "Neptune Analytics" (#811)
  • Fixed issue where nodes and edges without any labels were causing the app to crash (#799)
  • Fixed broken link in documentation (thanks @ssheladiya, #803)

Release v1.14.0

This release includes a major new feature: the ability to load and save graphs. It required a massive effort to update the data management logic to allow this feature to exist. Some of these changes made the graph rendering logic slightly more efficient, so you may notice some small performance improvements.

Saving and Loading Graphs

Graph Explorer now supports saving and loading graphs as files. To save the current graph, click the "Save graph to file" button in the graph toolbar. You can load a previously saved graph file by clicking the "Load graph from file" button in the graph toolbar and choosing the file to load.

Graph Explorer will verify that you are currently connected to the right database and then read all the node & edge IDs in the file. It will then execute the required queries to get up to date information from the database and load the nodes & edges in to the graph. Any existing nodes & edges in your graph will be unchanged.

File Contents

The saved graph file contains the following information in plain JSON format:

  • The active connection database URL and query engine
  • The list of node & edge IDs that are currently rendered
  • Some metadata to identify the file type, source, and version

All Changes

  • Added ability to save the rendered graph to a file, allowing for reloading the graph later or sharing the graph with other users who have the same connection (#756, #758, #761, #762, #767, #768, #769, #770, #775, #781, #786, #793)
  • Updated UI labels to refer to node & edge "labels" instead of "types" (#766)
  • Improved neighbor count retrieval to be more efficient (#744)
  • Removed unused hidden flag from schema types (#737)
  • Improved entity filtering logic reducing re-renders (#739, 741)
  • Added attribute count to node labels list in the connection screen (#743)
  • Improved pagination controls by using a single shared component (#742)
  • Updated styling across the app (#777, #743, #780)
    • Rounded style for search inputs, panels, toast notifications, and more
    • Searchable list items style consistent with connection style
    • Softer grays
    • More consistent shadows
    • More consistent menus (context menus, select dropdowns, etc.)
    • Graph legend is now consistent with other panels
  • Updated dependencies and remove unused dependencies (#764, #776, #782, #783)

Release 1.13.0

This release is a maintenance release with improvements to app startup, default connection handling, and Neptune error handling.

  • Removed URL parameter configFile support for the default connection (#724)
  • Improved app startup UI to be more consistent (#723)
  • Improved error logs in the browser console (#721)
  • Improved default connection handling, including fallbacks for invalid data (#734)
  • Updated dependencies (#718, #720)

Release 1.12.1

  • Fixed issue where the edge's display name value was not being displayed properly (#716)

Release 1.12.0

This release is mostly a maintenance release, with a few new features and bug fixes.

Consistent node and edge information

The biggest noticeable change is the consistency of node and edge information across the app. Previously, the node and edge information was rendered differently in the search results, node details, edge details, and expand options. Now, they are all consistent, with the same information displayed.

These changes improve the support for multiple node labels and date values in Gremlin.

All changes

  • Improved consistency of rendering node information across search results, node details header, edge details header, and expand options header (#697)
  • Improved edge properties in the details sidebar, which now includes the source and target vertex IDs and types (#698)
  • Fixed date rendering on Gremlin connections (#698)
  • Fixed a bug where the automatic open details feature would not open the details sidebar when selecting nodes (#679)
  • Improved styling on checkboxes across the app, and specifically the export options popover and entity filters sidebar (#676)
  • Improved styling on icon buttons (#674, #675, #683)
  • Improved styling & behavior on tooltips (#709)
  • Improved styling & behavior color inputs (#707)
  • Updated documentation to reorganize and extend the troubleshooting tips (#681)
  • Updated documentation to add a roadmap outline (#696)
  • Consolidated logic around the display of nodes and edges to a single location for simplicity and consistency (#698)
  • Updated dependencies, including Node v22 (#680)

Release 1.11.0

This release includes a big change to search, improvements to the Docker image size and security, and as always, there are many additional small fixes and improvements.

Search Sidebar

With this release, we've moved search out of the top navigation bar and in to the sidebar. When new users open Graph Explorer for the first time, they will be greeted with a set of search results ready to be added to the graph and explored.

For existing users, you'll benefit from a better experience with search. Your filters will remain active as you navigate around and it will do its best to keep your attribute selection when you change the node type. There's also a new "Add All" button, when you want to add all the search results to the graph with a single click.

All Changes

  • Improved search discoverability and ergonomics by moving the UI in to the sidebar (#665, #669, #670)
  • Improved UI responsiveness by using map instead of array for large data sets (#658)
  • Improved connection selection can now happen on any part of the connection row (#657)
  • Improved style for the sidebar buttons (#651)
  • Improved styles in the context menu to be easier to read (#670)
  • Improved Docker image size, reducing it by 196 MB (#619)
  • Improved query when searching across all node types (#607)
  • Improved query generation by removing empty lines (#608)
  • Fixed scrolling on search result details #657
  • Fixed Docker image containing more files than necessary. (#613)
  • Fixed conflict when a node has a property named "id" that prevented changing the display attribute. (#626)
  • Fixed style issue where buttons have a halo around them after being clicked (#650)
  • Fixed security vulnerabilities in the Docker image from dev dependencies remaining in the image (#616)
  • Fixed minor layout issue on connections screen (#648, #670)
  • Updated multiple dependencies (#611, #614, #616, #624)

Release 1.10.1

  • Improved support for databases with thousands of node & edge types (#599)
  • Improved logging on server and around schema sync (#604)
  • Fixed context menu styles (#600)
  • Fixed alignment of close button in search panel (#603)

Release 1.10.0

This release includes some new features to make managing and supporting Graph Explorer more friendly.

Major Changes

  • Added backup & restore options for Graph Explorer local data (#549)
  • Added about screen that includes a link to submit feedback (#549)
  • Improved logging & error handling to aid in support
    • Added global error page if the React app crashes (#547)
    • Added optional server logging of database queries when using the proxy server which can be enabled within settings (#574, #575)
    • Improved handling of server errors with more consistent logging (#557)
    • Improved SageMaker Lifecycle script handling of CloudWatch log driver failures (#550)
    • Improved parsing of environment values in proxy server resulting in an error when the values are invalid (#574)

Bug Fixes and Minor Changes

  • Fixed performance issue in styling sidebar panels when many node & edge types exist (#542)
  • Fixed issue with upper case characters in RDF URIs (#544)
  • Improved app styles (#543, #548)
  • Changed Node to run in production mode (#558)
  • Removed hosting production server using the client side Vite configuration, requiring the use of the proxy server (#565)
  • Transition to Tailwind instead of EmotionCSS for styles, which should make updating the UI much simpler (#543)
  • Updated multiple dependencies (#555, #557)

Release 1.9.0

This release includes many fixes and enhancements that make using Graph Explorer a more pleasant experience, especially for users with larger databases.

Major Changes

  • Improved error experience in search and Data Explorer UI that includes a retry button (#477)
  • Improved empty state experience in Data Explorer UI (#477)
  • Improved search using openCypher which will now execute a single request when searching across all labels (#493, #532)
  • Improved error messages for node expansion (#502)
  • Improved Gremlin schema sync performance on larger databases, thanks to @dsaban-lightricks for his great suggestion in issue #225 (#498)
  • Reduced chance of throttling issues when a large amount of requests are executed in parallel by batching requests in groups of 10 (#489)
  • Reduced unnecessary search queries when no search term is provided by ignoring attribute and exact match changes (#473)
  • Improved diagnostic logging in Neptune Notebooks by adding CloudWatch logs (#517)

Bug Fixes and Minor Changes

  • Fixed issue with default connections when Neptune Notebook instance is restarted (#508)
  • Fixed expanding a node on old versions of Gremlin (#503)
  • Fixed default selection of expand type to be the first available type for expansion (#501)
  • Fixed some SPARQL endpoints by using application/sparql-results+json accept header for SPARQL requests (#499)
  • Fixed CORS issue for some SPARQL and Gremlin endpoints due to queryId in the request headers (#529) (#499)
  • Fixed text wrapping for labels in edge styling sidebar (#499)
  • Fixed potential error when the request body is very large by increasing the body size limit for proxy server (#488)
  • Fixed issue when selecting an item in search results that resulted in errors in the browser console (#474)
  • Added type checking and linting to server code (https://github.com/aws/graph-explorer/pull/522)
  • Added environment values to override HTTP and HTTPS ports for proxy server (#500)
    • PROXY_SERVER_HTTP_PORT default is 80
    • PROXY_SERVER_HTTPS_PORT default is 443
  • Updated development scripts updated to be more consistent with the industry (#487, #525)
    • Run the dev environment pnpm dev
    • Build & run the production environment pnpm build && pnpm start
    • Clean build artifacts with pnpm clean
  • Fixed running in production mode locally with Node (#500)
  • Updated multiple dependencies (#475, #486, #490, #492, #491, #522, #523)

Release 1.8.0

Major Changes

Bug Fixes and Minor Changes

Release 1.7.0

This release includes the following feature enhancements and bug fixes:

Major Changes

Bug Fixes and Minor Changes

Release 1.6.0

This release includes the following feature enhancements and bug fixes:

Features

Bug Fixes and Minor Changes

Release 1.5.1

This release includes the following feature enhancements and bug fixes:

Bug fixes

Release 1.5.0

This release includes the following feature enhancements and bug fixes:

Features

Bug fixes

Release 1.4.0

This release includes the following feature enhancements and bug fixes:

Features

Bug fixes

Release 1.3.1

This patch release includes bugfixes for Release 1.3.0.

Bug fixes

Release 1.3.0

This release includes the following feature enhancements and bug fixes:

Features

Bug fixes

Release 1.2.0

This release includes the following feature enhancements and bug fixes:

Features

Bug fixes

Release 1.1.0

This release includes the following feature enhancements and bug fixes:

Features

Bug fixes

Release 1.0.0

The first release of Graph Explorer, an Apache 2.0 React-based web application that enables users to visualize both property graph and RDF data and explore connections between data without having to write graph queries. Major features include:

  • Ability to connect to graph databases like Amazon Neptune, or open-source endpoints like Gremlin Server or Blazegraph
  • Search and preview starting nodes once connected to a graph database
  • Visualize results in the Graph View & expand to view neighbors
  • Customize the Graph View with preferred layouts, colors, icons, labels, and more
  • Use Expand filters to customize the result set by filter text or count of results when expanding
  • Scroll through the Table View to view the data in the Graph View in tabular format
  • Use Table View filters when you need to highlight a subset of the data
  • Download a CSV or JSON of the nodes and edges in visualization, or download as a PNG file