i3 3.7
i3 3.6
i3 3.5
i3 3.4
i3 3.3
i3 3.2
i3 3.1
i3 3.0
i3 2.x and 1.x
|
i3 3.7
Findwise i3 3.7.0 focuses on a large portion of new functionalities and improvements but also on dependency upgrades to remove possible vulnerabilities. We have scanned our services and fixed most of found vulnerabilities. Backend upgraded to Spring Boot 2.7.0. Elasticsearch engine is now upgraded to 7.17.4 and SOLR wit Lucene to 8.11.1. We have migrated Index, Process and Search services to next gen blueprint service allowing configuration as a code the same way as already was available in Connect service. Search service has now user management through GUI similiar like all other services. We have extended i3 framework with AI Facade and AI Models introducing support for searching with Embeddings within i3 with accurate documentation how to setup it. We have prepared default setup for i3 Docker Swarm single node and multi nodes deployments with accurate documentation. All default templates are now automatically added on Elasticsearch container startup. All service URLs are configured now through environment variables with sensible default values to quickly starting applications regardless of environment and regardless whether containerized or not.
Release Highlights
- [I3-2632] Upgrade Elasticsearch to 7.17.4 and consequently Solr (because of Lucene) Task
- Easy deployment: As a configurer, service URLs are configured through environment variables with sensible default value as fallback, to quickly starting applications regardless of environment and regardless of whether containerized or not
- Easy deployment: As a deployer, I can start i3 and Elasticsearch in docker containers and be sure Elasticsearch templates are loaded before any content is submitted to Elasticsearch
- Easy deployment: As a consultant, I have available default settings for Docker Swarm to run i3 cluster
- Embeddings: As a consultant, I can deploy and run i3 AI Facade
- Embeddings: As a developer, I can extend i3 AI Facade with features and implementations
- Embeddings: As a consultant I can submit fields from Process document to text embeddings
- Embeddings: Create search feature for calculating embeddings vector
- Programmatic configuration: As a consultant, I can configure index with bundled programmatic configuration
- Programmatic configuration: As a consultant, I can configure process with bundled programmatic configuration
- Programmatic configuration: As a consultant, I can configure search with bundled programmatic configuration
Changelog
CONNECTOR SERVICE - 2.4.0
- Connect 2 - Broken blueprints: [I3-2226] As a consultant, I can start Connect 2 with broken configuration Story
- Easy deployment: [I3-1662] As a configurer, service URLs are configured through environment variables with sensible default value as fallback, to quickly starting applications regardless of environment and regardless of whet
her containerized or not Story
- Programmatic configuration: [I3-2589] Move field validation from setters to PostConstruct methods Task
- [I3-2581] IndexServiceSinkIT sometimes fails Bug
- [I3-2260] Remove deprecated ACL class Bug
- [I3-2661] Strategy CODE_CONFIG_ONLY_BUT_CODE_CONFIG_USERS_ONCE marks blueprints as broken instead of preventing application start Bug
- [I3-2650] Upgrade tika Task
- [I3-2648] Upgrade Google API client Task
- [I3-2646] Update Spring Boot to v2.6.7 or higher Task
- [I3-2685] Supporting CLOB type columns in JDBC connector Task
- [I3-2503] Upgrade norconex connector to version 3 Task
- [I3-2694] Connect - error 404 when trying to start a non-existing connector via REST Bug
- [I3-2696] CSV connector doesn't treat \t separator as tab Bug
PROCESSING SERVICE - 1.20.0
- Easy deployment: [I3-1662] As a configurer, service URLs are configured through environment variables with sensible default value as fallback, to quickly starting applications regardless of environment and regardless of whet
her containerized or not Story
- Embeddings: [I3-2267] As a consultant I can submit document fields and receive their corresponding vectors Story
- Embeddings: [I3-2269] As a consultant I can submit fields from Process document to text embeddings Story
- Embeddings: [I3-2270] As a consultant I can submit vectors corresponding to document fields to Elasticsearch Story
- Programmatic configuration: [I3-2589] Move field validation from setters to PostConstruct methods Task
- Programmatic configuration: [I3-2196] As a consultant, I can configure process with bundled programmatic configuration Story
- [I3-2575] Perform i3 bug fix release v3.6.1 with Elasticsearch 7.17.1 and Solr 8.11.1 Task
- [I3-2649] Find a replacement for Boilerpipe processing stage Task
- [I3-2650] Upgrade tika Task
- [I3-2646] Update Spring Boot to v2.6.7 or higher Task
- [I3-2632] Upgrade Elasticsearch to 7.17.4 and consequently Solr (because of Lucene) Task
- [I3-2706] Certain processing stages don't have required parameters marked properly. Bug
- Easy deployment: [I3-1662] As a configurer, service URLs are configured through environment variables with sensible default value as fallback, to quickly starting applications regardless of environment and regardless of whet
her containerized or not Story
- Programmatic configuration: [I3-2589] Move field validation from setters to PostConstruct methods Task
- Programmatic configuration: [I3-2197] As a consultant, I can configure index with bundled programmatic configuration Story
- [I3-2575] Perform i3 bug fix release v3.6.1 with Elasticsearch 7.17.1 and Solr 8.11.1 Task
- [I3-2661] Strategy CODE_CONFIG_ONLY_BUT_CODE_CONFIG_USERS_ONCE marks blueprints as broken instead of preventing application start Bug
- [I3-2646] Update Spring Boot to v2.6.7 or higher Task
- [I3-2632] Upgrade Elasticsearch to 7.17.4 and consequently Solr (because of Lucene) Task
- [I3-2699] Starting without MongoDB running doesn't produce broken collection (no collections at all instead) Bug
- Easy deployment: [I3-1662] As a configurer, service URLs are configured through environment variables with sensible default value as fallback, to quickly starting applications regardless of environment and regardless of whet
her containerized or not Story
- Embeddings: [I3-2658] Create search feature for calculating embeddings vector Story
- Programmatic configuration: [I3-2589] Move field validation from setters to PostConstruct methods Task
- Programmatic configuration: [I3-2195] As a consultant, I can configure search with bundled programmatic configuration Story
- [I3-2575] Perform i3 bug fix release v3.6.1 with Elasticsearch 7.17.1 and Solr 8.11.1 Task
- [I3-2646] Update Spring Boot to v2.6.7 or higher Task
- [I3-2632] Upgrade Elasticsearch to 7.17.4 and consequently Solr (because of Lucene) Task
SEARCH EDITOR PORTAL - 2.13.0
- Easy deployment: [I3-1662] As a configurer, service URLs are configured through environment variables with sensible default value as fallback, to quickly starting applications regardless of environment and regardless of whet
her containerized or not Story
- Programmatic configuration: [I3-2589] Move field validation from setters to PostConstruct methods Task
- [I3-2575] Perform i3 bug fix release v3.6.1 with Elasticsearch 7.17.1 and Solr 8.11.1 Task
- [I3-2646] Update Spring Boot to v2.6.7 or higher Task
- [I3-2632] Upgrade Elasticsearch to 7.17.4 and consequently Solr (because of Lucene) Task
i3 3.6
Mostly focused on dependency upgrades to remove possible vulnerabilities. Backend upgraded to Spring Boot 2.6.3 and frontend to React 17.0.2. Spring Boot upgrade resulted with update on how we handle now properties in all services. Elasticsearch engine is now upgraded to 7.16.3 and SOLR wit Lucene to 8.10.1. In Index we removed legacy REST version 1 API. In Connect we fixed inconsistency in plugin names. There is also included a bunch of tasks about technical debt and bugfixes in all services which we took care of.
Changelog - TBD
i3 3.5
Findwise i3 3.5.0 focuses on further development of the new Connect service. It has been extended with a variety of new connectors (Yammer, Jira, Google Drive, cloud Confluence) and received a lot of usability and performance improvements. There was introduced Index Service Sink retry mechanism. Index service received UI refreshment. All remainings after old index service 2 were removed. Blueprints framework was extended with Display interface. All i3 dependencies were updated to latest versions. Docker images of i3 services are now based on Eclipse Temurin image.
Release Highlights
- [I3-1965] As a consultant, I can fetch content from Yammer that includes a large history and that can continue to get deltas for recent messages Story
- [I3-2126] As a consultant, I can fetch issues from cloud Jira Story
- [I3-2129] As a consultant, I can fetch content from Google Drive Story
- [I3-2127] As a consultant, I can fetch content from cloud Confluence Story
- [I3-2160] As a consultant, I can use the latest versions of i3 dependencies Story
- [I3-2139] As a consultant, I want Index Service Sink to retry document submission when index service is unavailable Story
- [I3-2211] Extend blueprint framework with configuration how is it displayed in the UI Task
- Technical debt: [I3-2081] Remove all remains related to old index2 Task
- [I3-2230] New collections overview Task
- [I3-1898] As a consultant, I can use the latest search engine version Story
- [I3-2160] As a consultant, I can use the latest versions of i3 dependencies Story
Changelog
- Connector service 2.2.0
- Technical debt: [I3-1448] Prevent maven filtering for binary files Task
- Technical debt: [I3-1455] Handle favicon in spring-boot manner Task
- Technical debt: [I3-2152] Configure autoformatting pre commit hook in Connect 2 Task
- [I3-2134] As a consultant, a connect sink is configured on the same level as a batching sink, to make it easier to configure otherwise hidden required parameters Story
- [I3-2211] Extend blueprint framework with configuration how is it displayed in the UI Task
- [I3-1965] As a consultant, I can fetch content from Yammer that includes a large history and that can continue to get deltas for recent messages Story
- [I3-2126] As a consultant, I can fetch issues from cloud Jira Story
- [I3-2129] As a consultant, I can fetch content from Google Drive Story
- [I3-2127] As a consultant, I can fetch content from cloud Confluence Story
- [I3-2160] As a consultant, I can use the latest versions of i3 dependencies Story
- [I3-1898] As a consultant, I can use the latest search engine version Story
- [I3-2306] As a consultant I am clearly informed about a configurable class removal Story
- [I3-2139] As a consultant, I want Index Service Sink to retry document submission when index service is unavailable Story
- [I3-2438] Remove redundant dependency on jcifs Task
- [I3-2447] Norconex unable to pick up existing state or hangs on finishing when running in Docker Bug
- Indexing Service 4.3.0
- Technical debt: [I3-1448] Prevent maven filtering for binary files Task
- Technical debt: [I3-2081] Remove all remains related to old index2 Task
- Technical debt: [I3-1455] Handle favicon in spring-boot manner Task
- [I3-2211] Extend blueprint framework with configuration how is it displayed in the UI Task
- [I3-2231] Rerun/reindex usability improvements Task
- [I3-2232] Collection search improvements Task
- [I3-2230] New collections overview Task
- [I3-2194] Use BufferedInputStream when reading JSON documents from index api to increase read performance Task
- [I3-2303] Add id attribute to collection status label on collections overview and collection overview Task
- [I3-2308] Wrong link on create document functionality cancel button Bug
- [I3-2316] Index - slashes are not encoded Bug
- [I3-2318] batch size, not properly comparing size KB vs bytes Bug
- [I3-2160] As a consultant, I can use the latest versions of i3 dependencies Story
- [I3-1898] As a consultant, I can use the latest search engine version Story
- [I3-2406] Collection links do not work on custom context path Bug
- [I3-2306] As a consultant I am clearly informed about a configurable class removal Story
- [I3-2408] Find document by id from the Indexsvc Manage API returns HTTP 500 for a non-existent document Bug
- Processing service 1.18.0
- Technical debt: [I3-1448] Prevent maven filtering for binary files Task
- Technical debt: [I3-2081] Remove all remains related to old index2 Task
- Technical debt: [I3-1455] Handle favicon in spring-boot manner Task
- [I3-2211] Extend blueprint framework with configuration how is it displayed in the UI Task
- [I3-2230] New collections overview Task
- [I3-2065] Elasticsearch writer field mapping not working correctly Bug
- [I3-2160] As a consultant, I can use the latest versions of i3 dependencies Story
- [I3-1898] As a consultant, I can use the latest search engine version Story
- [I3-2306] As a consultant I am clearly informed about a configurable class removal Story
- Search Service 2.17.0
- Technical debt: [I3-1448] Prevent maven filtering for binary files Task
- Technical debt: [I3-1371] Remove the legacy event logger from Search Task
- Technical debt: [I3-2000] Remove legacy Jellyfish support Task
- Technical debt: [I3-1455] Handle favicon in spring-boot manner Task
- [I3-2211] Extend blueprint framework with configuration how is it displayed in the UI Task
- [I3-2160] As a consultant, I can use the latest versions of i3 dependencies Story
- [I3-1898] As a consultant, I can use the latest search engine version Story
- [I3-2306] As a consultant I am clearly informed about a configurable class removal Stor
- Search editor portal 2.11.0
- Technical debt: [I3-1448] Prevent maven filtering for binary files Task
- Technical debt: [I3-2242] Remove the legacy event logger from SEP Task
- Technical debt: [I3-2081] Remove all remains related to old index2 Task
- Technical debt: [I3-1455] Handle favicon in spring-boot manner Task
- [I3-2286] Synonyms module in i3 SEP uses Graph Synonyms for Elasticsearch Task
- [I3-2160] As a consultant, I can use the latest versions of i3 dependencies Story
- [I3-2306] As a consultant I am clearly informed about a configurable class removal Story
- [I3-2413] Unable to enter created synonym entry Bug
i3 3.4
Findwise i3 3.4.0 focuses on further development of the new Connect service. It has been extended with a variety of new connectors (RSS, CSV, Sharepoint, MongoDB) and received a lot of usability and performance improvements. Delta crawling has been made way more flexible. It is the first service that received a new way of programmatic configuration. Index 4 received bugfixes and older version migration improvements that makes the upgrade process easier and more reliable. Search service now enables to use search features in completers, thus making search results usable as query completion suggestions. Click signals can now carry non-clicked hits attributes, making signals more useful for machine learning purposes.
Release Highlights
- Connect 2.0: As a consultant, I can fetch content from rss feeds
- Connect 2.0: As a configurer, I can view and navigate all items discovered by a connector
- Connect 2.0: As a consultant, I can store hierarchical message data about a crawl, to be able to create connectors that support incremental crawling on multiple levels
- Connect 2.0: As a consultant, I can more efficiently fetch content from Exchange by making it possible to only fetch changes between executions
- Connect 2.0: As a consultant, I can fetch content from csv files
- Connect 2.0: As a consultant, I can fetch content from MongoDB database
- Connect 2.0: As a consultant, I can use SQL code editor when configuring database connectors instead of single line text field
- Connect 2.0: As a consultant, I can connect to popular databases without adding drivers on my own
- Connect 2.0: As a consultant, I can fetch content from SharePoint
- Contributed: As a consultant, I can use any search feature in a completer, to create highly adaptable type-ahead solutions
- Signals: As a consultant, I can find the IDs of all top documents in a click signal, so that it can be used to build machine learning models for result re-ranking
- As a consultant, I can configure connect with bundled programmatic configuration
Changelog
-
- Connect 2.0: [I3-2133] Remove index dependencies from connect2 Bug
- Connect 2.0: [I3-1923] Move connector reset and delete into hamburger menu at the connectors overview Task
- Connect 2.0: [I3-1782] As a consultant, I can fetch content from rss feeds Story
- Connect 2.0: [I3-1770] As a configurer, I can view and navigate all items discovered by a connector Story
- Connect 2.0: [I3-2068] Increase the number of saved connector run statistics records Task
- Connect 2.0: [I3-2084] As a consultant, I can store hierarchical message data about a crawl, to be able to create connectors that support incremental crawling on multiple levels Story
- Connect 2.0: [I3-2148] As a consultant, I can more efficiently fetch content from Exchange by making it possible to only fetch changes between executions Story
- Connect 2.0: [I3-2151] Stopping norconex web crawler results in exceptions after the crawler has stopped Bug
- Connect 2.0: [I3-2180] Write ADR and update "Extend connect" guide regarding message passing architecture consequences of Connect 2 Task
- Connect 2.0: [I3-2165] FilesystemProcessor doesn't convert backslashes Bug
- Connect 2.0: [I3-2137] A more useful connector overview Task
- Connect 2.0: [I3-2122] As a consultant, I can fetch content from csv files Story
- Connect 2.0: [I3-2123] As a consultant, I can fetch content from MongoDB database Story
- Connect 2.0: [I3-1987] As a consultant, I can use SQL code editor when configuring database connectors instead of single line text field Story
- Connect 2.0: [I3-2201] Connect - rss connector stopped and failed when id value not found in item Bug
- Connect 2.0: [I3-2130] Use Server Sent Events to improve page refresh experience Task
- Connect 2.0: [I3-2179] Sorting by id on connector items view does not work correctly Bug
- Connect 2.0: [I3-2121] MongoDB indices are created for commonly queried fields Task
- Connect 2.0: [I3-1788] As a consultant, I can fetch content from SharePoint Story
- Connect 2.0: [I3-2218] Connector overview is not refreshed frequently enough in some edge cases Bug
- [I3-2132] Update documentation links Task
- [I3-2182] Decouple blueprint persistence from blueprint service Task
- [I3-2166] As a consultant, I can configure connect with bundled programmatic configuration Story
- [I3-2206] Incorrect id for errorMessages in the UI Bug
- [I3-2214] Connect - csv connector stopped and failed when record size does not match the header size Bug
- [I3-2221] Some unit tests are not executed due to jUnit 4/5 mismatch Bug
- [I3-2222] MongoDB indices integration tests interfere with Spring Data repositories integration tests Bug
- [I3-2224] Connector's endTime is sometimes not calculated after connector has stopped Bug
- [I3-2223] Notifications don't fade away correctly in Pipeline Testing Bug
- [I3-2234] It is not possible to stop a running connector gracefully randomly due to MongoInterruptedException Bug
- [I3-2239] Issue with handling endedTime/duration for connectors Bug
- [I3-2249] Unsuccessful validation in @PostConstruct creates a broken blueprint on save Bug
- [I3-2248] Connector - missing worker-json.js script file after open MongoDB connector configuration page Bug
- [I3-2252] Adjustments to the REST Documentation Bug
- [I3-2255] Filesystem ZIP processor should add signatures to ContentMessages Bug
- [I3-2258] ACL format mismatch in different connectors Bug
-
- [I3-2132] Update documentation links Task
- [I3-2169] Index Service 3 importer demands collection mapping whereas in fact it is a list Bug
- [I3-2149] Change in HTTP methods made Index REST v2 operations non-backwards compatible Bug
- [I3-2167] Wrong repositories FQNs and redundant credentials migration in IndexService3RepositoryToIndexService4Repository configurable change, missing User migration Bug
- [I3-2168] Index Service 3 importer thread leakage Bug
- [I3-2199] Binary attachments missing after migrating collection with Index Service 3 importer Bug
- [I3-2198] Unable to search through collections migrated from Index 3 with DOCUMENT handler Bug
- [I3-2235] Searching summary or logs of index collection returns only 50 records on the result list Bug
-
Processing service 1.17.0
- [I3-2132] Update documentation links Task
- [I3-2219] Upgrade graalvlm to version based on UPL license Bug
- [I3-2223] Notifications don't fade away correctly in Pipeline Testing Bug
-
- Contributed: [I3-2146] As a consultant, I can use any search feature in a completer, to create highly adaptable type-ahead solutions Story
- Signals: [I3-2138] As a consultant, I can find the IDs of all top documents in a click signal, so that it can be used to build machine learning models for result re-ranking Story
- [I3-2132] Update documentation links Task
i3 3.3
Findwise i3 3.3.0 introduces completely rewritten connector service. The main driver for the rewrite was to make the service rock solid yet flexible, while following standards made up by other i3 services. Connect 2.0, similarly as other components, is based on blueprints framework for configuration management, has a React interactive UI and is fully manageable through REST Configuration and Management APIs. Connector implementation framework is way more flexible, making it ready to cope with the craziest custom requirements. This release includes massive UI performance optimization for all the services, which makes i3 operation and administration much more convenient with complex configurations. The UI stack has been upgraded in order to keep it secure and up to date. As usual, search engines support was updated too - i3 3.3.0 relies on Elasticsearch 7.10.0 and Apache Solr 8.7.0.
Release Highlights
- Connect 2.0: As a consultant, I can sink fetched content to a logging endpoint
- Connect 2.0: As a configurer, I can view created connectors
- Connect 2.0: As a configurer, I can create a new connector
- Connect 2.0: As a configurer, I can start a created connector and stop a running connector
- Connect 2.0: As a configurer, I can delete a created not-running connector
- Connect 2.0: As a configurer, I can have items previously discovered but are un-discovered after the current run removed
- Connect 2.0: As a consultant, I can sink fetched content to the indexing service
- Connect 2.0: As a configurer, I can set a created connector to run on a specific schedule
- Connect 2.0: As a configurer, I can view created connectors from a REST API
- Connect 2.0: As a consultant, I can fetch content from the local filesystem
- Connect 2.0: As a consultant, I can demo the connector service by producing fake but representative people data
- Connect 2.0: As a consultant, I can demo the connector service by producing fake but representative vehicle data
- Connect 2.0: As an i3 Developer I have a REST Management API to execute different actions on service entities that does not belong to REST Config API
- Connect 2.0: As a configurer, I can start created connector and stop a running connector from a REST API
- Connect 2.0: As a configurer, I can log in to the connector application
- Connect 2.0: As a consultant, I can fetch content from data source using jdbc
- Connect 2.0: As a consultant, I can fetch content from a shared Windows drive (smbj)
- Connect 2.0: As a configurer, I can be sure reported connector run errors are persisted between application restart
- Connect 2.0: As a configurer, I can set a created connector to auto-run on startup
- Connect 2.0: As a consultant, I can fetch content from web pages using norconex in a simplified approach
- Connect 2.0: As a configurer, I can fetch incremental changes from systems that support it
- Connect 2.0: As a consultant, I can fetch content from LDAP
- Connect 2.0: As a consultant, I can fetch content from web pages using norconex in a full-configurability approach
- Connect 2.0: As a configurer, I can be sure discovered items are persisted between application restart
- Connect 2.0: As a configurer, I can reuse credentials between a source and processor in a connector
- Connect 2.0: As a consultant, I can fetch content from Exchange
- Connect 2.0: As a configurer, I can be sure execution data is persisted between application restart
- Connect 2.0: As a configurer, I can clear the state that carries the IDs of all items discovered by a connector, so that a subsequent run will be identical to running the connector for the first time
- Connect 2.0: As a configurer, I can see live throughput of discovered content for all currently running connectors
- Connect 2.0: As a configurer, I can view all errors associated with the current run for a currently running connector
- Connect 2.0: As a configurer, I can view details about current and previous connector runs
- Connect 2.0: As a configurer, I can decide where persistent data is stored, to accomodate environments with limited storage options
- Connect 2.0: As connect system, items unchanged between runs are skipped (differential crawling)
- Connect 2.0: As a consultant, I can configure connect programmatically, so that configuration can be easily version managed and modified or upgraded with the type safety
- Connect 2.0: As a tester, I can test connect2 using bundled test plugins without these being exposed to a configurer in production
- Contributed: As a search user, I can use the CollapseQParserPlugin for more accurate and faster grouping on fields with many unique values
- As a search configurer, searchers with exhaustive nested configuration are displayed within seconds and performing changes are similarly responsive as editing a seacher with minimal configuration
- As an i3 stakeholder in an organization using i3, I can read guides that explains how to develop against, and how to operate i3
Changelog
-
- Connect 2.0: [I3-1766] As a consultant, I can sink fetched content to a logging endpoint Story
- Connect 2.0: [I3-1866] Rename artifactId connect2 to connect Task
- Connect 2.0: [I3-1759] As a configurer, I can view created connectors Story
- Connect 2.0: [I3-1760] As a configurer, I can create a new connector Story
- Connect 2.0: [I3-1761] As a configurer, I can start a created connector and stop a running connector Story
- Connect 2.0: [I3-1762] As a configurer, I can delete a created not-running connector Story
- Connect 2.0: [I3-1765] As a configurer, I can have items previously discovered but are un-discovered after the current run removed Story
- Connect 2.0: [I3-1767] As a consultant, I can sink fetched content to the indexing service Story
- Connect 2.0: [I3-1763] As a configurer, I can set a created connector to run on a specific schedule Story
- Connect 2.0: [I3-1776] As a configurer, I can view created connectors from a REST API Story
- Connect 2.0: [I3-1780] As a consultant, I can fetch content from the local filesystem Story
- Connect 2.0: [I3-1778] As a consultant, I can demo the connector service by producing fake but representative people data Story
- Connect 2.0: [I3-1779] As a consultant, I can demo the connector service by producing fake but representative vehicle data Story
- Connect 2.0: [I3-1906] As an i3 Developer I have a REST Management API to execute different actions on service entities that does not belong to REST Config API Story
- Connect 2.0: [I3-1777] As a configurer, I can start created connector and stop a running connector from a REST API Story
- Connect 2.0: [I3-1774] As a configurer, I can log in to the connector application Story
- Connect 2.0: [I3-1785] As a consultant, I can fetch content from data source using jdbc Story
- Connect 2.0: [I3-1781] As a consultant, I can fetch content from a shared Windows drive (smbj) Story
- Connect 2.0: [I3-1795] As a configurer, I can be sure reported connector run errors are persisted between application restart Story
- Connect 2.0: [I3-1938] Rename Post run processor concept to State Task
- Connect 2.0: [I3-1764] As a configurer, I can set a created connector to auto-run on startup Story
- Connect 2.0: [I3-1789] As a consultant, I can fetch content from web pages using norconex in a simplified approach Story
- Connect 2.0: [I3-1798] As a configurer, I can fetch incremental changes from systems that support it Story
- Connect 2.0: [I3-1786] As a consultant, I can fetch content from LDAP Story
- Connect 2.0: [I3-1988] SQL connector: JDBC driver parameter should be a dropdown of driver classes available on the classpath Task
- Connect 2.0: [I3-1790] As a consultant, I can fetch content from web pages using norconex in a full-configurability approach Story
- Connect 2.0: [I3-1768] As a configurer, I can be sure discovered items are persisted between application restart Story
- Connect 2.0: [I3-1999] As a configurer, I can reuse credentials between a source and processor in a connector Story
- Connect 2.0: [I3-1936] It is not possible to change effectively schedule type in connector after chosen CronSchedule Bug
- Connect 2.0: [I3-1787] As a consultant, I can fetch content from Exchange Story
- Connect 2.0: [I3-1794] As a configurer, I can be sure execution data is persisted between application restart Story
- Connect 2.0: [I3-2014] Error when sinking unpacked binaries in filesystem connector to indexsvc Bug
- Connect 2.0: [I3-1769] As a configurer, I can clear the state that carries the IDs of all items discovered by a connector, so that a subsequent run will be identical to running the connector for the first time Story
- Connect 2.0: [I3-2012] Connect2 - wrong url when ID contains special characters Bug
- Connect 2.0: [I3-1924] Improve new connector creation process Task
- Connect 2.0: [I3-1931] Error messages accumulate indefinitely Bug
- Connect 2.0: [I3-1772] As a configurer, I can see live throughput of discovered content for all currently running connectors Story
- Connect 2.0: [I3-1773] As a configurer, I can view all errors associated with the current run for a currently running connector Story
- Connect 2.0: [I3-1771] As a configurer, I can view details about current and previous connector runs Story
- Connect 2.0: [I3-2058] Connect2 - The on-demand schedule is not the default choice Bug
- Connect 2.0: [I3-2060] Introduce Server Send Events on connectors summary page Task
- Connect 2.0: [I3-1990] As a configurer, I can decide where persistent data is stored, to accomodate environments with limited storage options Story
- Connect 2.0: [I3-2016] When requesting a repository by connector ID, return a wrapper repository that removes the need to include the same connector ID on each method invocation Task
- Connect 2.0: [I3-2075] Error in browser console after refresh page Bug
- Connect 2.0: [I3-2073] As connect system, items unchanged between runs are skipped (differential crawling) Story
- Connect 2.0: [I3-2061] It's possible to delete all users Bug
- Connect 2.0: [I3-2067] It is possible to make configuration changes to a repository that is currently in use by a running connector Bug
- Connect 2.0: [I3-2062] Connect 2 code review/refactoring/polishing Task
- Connect 2.0: [I3-2092] Connect - regex filename filter doesn't work properly Bug
- Connect 2.0: [I3-2090] ExchangeConnector does not work with stateful repositories Bug
- Connect 2.0: [I3-2095] Building the connect2 docker image fails Bug
- Connect 2.0: [I3-2097] Write ADR for repositories Task
- Connect 2.0: [I3-2089] Copies of processors are created redundantly Bug
- Connect 2.0: [I3-2101] Stopping executor outlives the execution Bug
- Connect 2.0: [I3-2103] Change the id generation scheme for statistics Task
- Connect 2.0: [I3-2107] Connector summary is not updated with custom context path Bug
- Connect 2.0: [I3-2108] The AdvancedNorconexMessageSource validation is incorrect Bug
- Connect 2.0: [I3-2094] As a consultant, I can configure connect programmatically, so that configuration can be easily version managed and modified or upgraded with the type safety Story
- Connect 2.0: [I3-2088] ConcurrentConnectorExecutionStrategy integration tests Task
- Connect 2.0: [I3-2109] Connect - connector status on the overview and in the statistics are inconsistent Bug
- Connect 2.0: [I3-2105] The ConcurrentConnectorExecutionStrategy logs a confusing warning Bug
- Connect 2.0: [I3-2104] Confusing SMB max depth message Bug
- Connect 2.0: [I3-2111] Connect - can set -1/0 value in undiscovered since period value field Bug
- Connect 2.0: [I3-2117] Norconex connectors report delete messages massively Bug
- Connect 2.0: [I3-2118] More dropdown is not marked Active when viewing Rest Documentation Bug
- Connect 2.0: [I3-1996] As a tester, I can test connect2 using bundled test plugins without these being exposed to a configurer in production Story
- [I3-1505] Introduce dependencies' management artifact for non-boot projects Task
- [I3-1525] As a search configurer, searchers with exhaustive nested configuration are displayed within seconds and performing changes are similarly responsive as editing a seacher with minimal configuration Story
- [I3-1989] Drop the superadmin parameter in the User class throughout services Task
- [I3-1993] Consistent character encoding Task
-
- Connect 2.0: [I3-1924] Improve new connector creation process Task
- Connect 2.0: [I3-1931] Error messages accumulate indefinitely Bug
- Index 4.0: [I3-1876] Migrate changes made in 3.3 codebase to Index 4 Task
- Index 4.0: [I3-2064] Index4 uses lots of CPU, even when idle Bug
- [I3-787] Move common frontend components to commons-js Task
- [I3-1505] Introduce dependencies' management artifact for non-boot projects Task
- [I3-1808] Update React and its necessary plugins and dependencies Task
- [I3-1525] As a search configurer, searchers with exhaustive nested configuration are displayed within seconds and performing changes are similarly responsive as editing a seacher with minimal configuration Story
- [I3-1911] Index / Missing users module in configuration Bug
- [I3-1992] Index generating Paging URLs are not encoded using an URL safe encoder Bug
- [I3-1966] Upgrade Solr and Elasticsearch Task
- [I3-1989] Drop the superadmin parameter in the User class throughout services Task
- [I3-1993] Consistent character encoding Task
-
Processing service 1.16.0
- Connect 2.0: [I3-1924] Improve new connector creation process Task
- Connect 2.0: [I3-1931] Error messages accumulate indefinitely Bug
- [I3-787] Move common frontend components to commons-js Task
- [I3-1505] Introduce dependencies' management artifact for non-boot projects Task
- [I3-1808] Update React and its necessary plugins and dependencies Task
- [I3-1525] As a search configurer, searchers with exhaustive nested configuration are displayed within seconds and performing changes are similarly responsive as editing a seacher with minimal configuration Story
- [I3-1913] Process service / wrong js script address in pipeline testing module Bug
- [I3-1986] Include i3 version 3.2 release changes into 3.3 release Task
- [I3-1966] Upgrade Solr and Elasticsearch Task
- [I3-1989] Drop the superadmin parameter in the User class throughout services Task
- [I3-1993] Consistent character encoding Task
-
- Connect 2.0: [I3-1924] Improve new connector creation process Task
- Connect 2.0: [I3-1931] Error messages accumulate indefinitely Bug
- Connect 2.0: [I3-2062] Connect 2 code review/refactoring/polishing Task
- Contributed: [I3-1556] As a search user, I can use the CollapseQParserPlugin for more accurate and faster grouping on fields with many unique values Story
- [I3-787] Move common frontend components to commons-js Task
- [I3-1865] Search REST API always responds with ISO-8859-1 charset Bug
- [I3-1505] Introduce dependencies' management artifact for non-boot projects Task
- [I3-1808] Update React and its necessary plugins and dependencies Task
- [I3-1525] As a search configurer, searchers with exhaustive nested configuration are displayed within seconds and performing changes are similarly responsive as editing a seacher with minimal configuration Story
- [I3-1912] Search service / Validation error Bug
- [I3-1966] Upgrade Solr and Elasticsearch Task
- [I3-1989] Drop the superadmin parameter in the User class throughout services Task
- [I3-1993] Consistent character encoding Task
- [I3-2054] Groovy features in Search Service are impossible to save Bug
- [I3-2076] Processing: Copying pipeline mismatches GUI blueprint version Bug
-
Search editor portal 2.10.0
- [I3-1505] Introduce dependencies' management artifact for non-boot projects Task
- [I3-1966] Upgrade Solr and Elasticsearch Task
- [I3-1989] Drop the superadmin parameter in the User class throughout services Task
- [I3-1993] Consistent character encoding Task
i3 3.2
Findwise i3 3.2 is the next even more stable version of i3 3.1. Focus of this release has been to make our content inbox, the indexing service, stateless. That means that in-memory state is persisted to a robust database which is the primary enabler to be able make it fully distributable in coming versions; that is to run it in parellel to (1) share the workload and (2) provide redundancy. Index 4 as the new version is called is backwards compatible and can be a drop-in replacement to Index 3.
Release Highlights
- Index 4 is a drop-in replacement to index 3 if the built-in data migration steps are performed
- Stateless queueing mechanism
- Ghost documents and invisible adds issues resolved
- Solves deadlock issue where index 3 can hang, causing connect and process to fail with it
Changelog
- Indexing Service 4.0
- [I3-1835] - Stateless indexsvc mongodb repo improvements Task
- [I3-1838] - Add integration tests Task
- [I3-1861] - Reduce technical debt and embrace i3 coding standards Task
- [I3-1863] - Get rid of overridden classes from the org.springframework.hateoas package Task
- [I3-1867] - Clarify the responsibilities of IndexServiceCollection and IndexServiceRepository Task
- [I3-1868] - Unify naming concerning checking out a batch Task
- [I3-1869] - Harden synchronization in InMemoryProcessingQueue Task
- [I3-1870] - Make all the methods of REST API v2.1 support versioning Task
- [I3-1871] - Correct HTTP methods for REST API v2.x Task
- [I3-1872] - Retrofit the REST API v2.0 with adding default versions numbers Task
- [I3-1877] - Improve the strategy of assigning default versions to documents saved or updated via REST API v2 Task
- [I3-1893] - Implement better ordering in MongoDocumentReadOperations Task
- [I3-1902] - Ensure empty batches are checked in (InMemoryBatchManager) Task
- [I3-1904] - Ensure adherence to blueprint rules Task
- [I3-1905] - Add default value to version field Task
- [I3-1908] - Facade beans registered in each application context Task
- [I3-1909] - Replace indexsvc3 with indexsvc4 as the name for default repository Task
- [I3-1915] - Enforce document version > 0 Task
- [I3-1935] - Revise and improve upon InMemoryProcessingQueue unit tests Task
- [I3-1937] - InMemoryProcessingQueue does not consider "reserved" documents from queue Task
- [I3-1940] - Improve package structures Task
- [I3-1941] - Improve tests and test coverage Task
- [I3-1945] - Make delete[Documents]ByIdsAndVersions race-condition-proof Task
- [I3-1967] - Make Index 4 processing api backwards-compatible Task
- [I3-1501] - long values not supported in index service API Bug
- [I3-1925] - Fix nullpointer exception in InMemoryBatchRepository#checkIn Bug
- [I3-1948] - Joda.time related warnings being displayed each time an instance of the MongoTemplate is created Bug
- [I3-1961] - ProcessingRestController#hasBatches throws if the collection does not exist Bug
- [I3-1978] - Index Service 2 import utility throws on convertion from mongo representation to domain objects Bug
- [I3-1980] - In memory batch tracking memory leaks Bug
- [I3-1982] - Unhandled document status in the rest v1 converter Bug
- [I3-1983] - Sequence number generator instantiated independently for both rest APIs Bug
- [I3-1985] - Versions cannot be sent along with documents to processing service Bug
- Processing service 1.15.2
- [I3-1968] - Rename index 3 input source to index4 input source and mark in it's blueprint documentation that it also supports index 3 Task
i3 3.1
Findwise i3 3.1.0 release introduces new query completer called Precision Search. E-commerce-oriented mechanism based on a graph database lets user to explore product categories and properties before performing actual search request. Other highlights include new search features for both Solr and Elasticsearch, better hierarchical document support, configuration enhancements thanks to multivalue placeholders, more powerful customization thanks to autowiring, Spring beans overriding and default values assignments. Search solutions based on Nordic languages will benefit from better compound words support. Deployment got simplified thanks to publishing official i3 services Docker images for the first time. Besides that, a bunch of smaller bugfixes and dependency updates are included. i3 version 3.1.0 is based on Elasticsearch 7.8.0 and Apache Solr 8.5.1.
Release Highlights
- As a consultant, I can read documentation that reflects changes introduced in Spring Boot and JDK 11 migration
- As a search user, I can collapse Solr documents on a specific field
- As an i3 developer, I can build and run i3 e2e tests on Java 11
- As a searcher user, I can apply a filter on numeric values using a slider type of interaction
- As a search user, I can get an overview of all selected Solr filters
- As a configurer, application-wide sensible default values are assigned to out-of-the-box plugins where practical, so that I can more easily configure applications with less effort
- As a consultant, I can configure precision search using graph
- As a consultant, I can create facets spanning dynamic ranges
- As processing system, DocumentPath parameters replaces custom string parsing
- As a search user, I can query searchers across domains, so that I can create javascript search pages that reside on a different domain than the search service
- As processing system, the precision search graph can be populated by a processing stage in the processing pipeline
- As a search system, basic query engine in place for precision search graph is created
- Implement context-aware completion provider
- As a search user, I can get suggestions for words that starts with or is equal to a query term from defined starting points in the graph
- As a product owner, I can run automated integration tests between i3 and graph system - vol. 1
- As a configurer, I can autowire dependencies into i3 plugins
- As a consultant, I can use ready-made docker images of i3 applications in my project
- As a search user, I can use the latest Elasticsearch search engine and Elastic stack
- As a search user, I can use the latest Solr search engine
- As a search user, I can get suggestions from nearby nodes in the graph, based on defined query template
- As a search user, I can search with and get hits for compounded words, regardless of inflection
- As a configurer, I can use placeholders to supply environment specific settings that are multi-valued
- As a consultant, I can override default beans in i3 applications with my own custom beans
- As a search user, I can limit the number of suggestion returned for the graph-based query suggestions
- As a search user, I can search for and with compound words, so that search result quality is improved
- As a search user, I can read documentation on how to apply graph-based query suggestions
- As a consultant, I can demo the latest i3 features from demo environment
Changelog
-
- [I3-1635] Prevent unnecessarily high number of connections made by JDBC connectors Bug
- [I3-1685] Upgrade dependency versions Task
- [I3-1793] Images containing shell scripts might fail to start up if built on Windows machine with standard line feeding Bug
- [I3-1806] Upgrade docker base image to the latest version of adoptopenjdk Task
- [I3-1812] i3 containers run as root Bug
- [I3-1820] Running process and index with Oracle JRE distribution doesn't work Bug
-
- [I3-1480] Increase the page size of reindex operations and the size of writing batches to backing store in index service Task
- [I3-1619] Count index service document size in bytes instead of chars for a more accurate byte size measurement Bug
- [I3-407] Antlr4QueryVisitor logging not really helpful Bug
- [I3-478] When processing tries to pull information from the index service queue and the token is wrong, this should be logged clearly (warn) Bug
- [I3-1685] Upgrade dependency versions Task
- [I3-1538] As a search user, I can use the latest Elasticsearch search engine and Elastic stack Story
- [I3-1555] As a search user, I can use the latest Solr search engine Story
- [I3-1722] Process polling for batches can end up waiting indefinately Bug
- [I3-1606] As a consultant, I can override default beans in i3 applications with my own custom beans Story
- [I3-1793] Images containing shell scripts might fail to start up if built on Windows machine with standard line feeding Bug
- [I3-1316] Update solr and kibana port on all default configuration to the default port 8983 and 5601 accordingly, throughout all components Task
- [I3-1806] Upgrade docker base image to the latest version of adoptopenjdk Task
- [I3-1812] i3 containers run as root Bug
- [I3-1820] Running process and index with Oracle JRE distribution doesn't work Bug
-
Processing service 1.15.0
- [I3-1551] Pipelines with stage ElasticsearchHighLevelRestWriter becomes broken after application restart Bug
- [I3-478] When processing tries to pull information from the index service queue and the token is wrong, this should be logged clearly (warn) Bug
- [I3-1116] REST API - enter pipeline which not exist cause 500 Bug
- [I3-1184] As processing system, DocumentPath parameters replaces custom string parsing Story
- [I3-1536] As processing system, the precision search graph can be populated by a processing stage in the processing pipeline Story
- [I3-1685] Upgrade dependency versions Task
- [I3-1538] As a search user, I can use the latest Elasticsearch search engine and Elastic stack Story
- [I3-1555] As a search user, I can use the latest Solr search engine Story
- [I3-1722] Process polling for batches can end up waiting indefinately Bug
- [I3-1606] As a consultant, I can override default beans in i3 applications with my own custom beans Story
- [I3-1751] Upgrade precision search to use the latest features from develop Task
- [I3-1609] Concurrent pipelines with passwords are failed on restart and concurrent runs Bug
- [I3-1793] Images containing shell scripts might fail to start up if built on Windows machine with standard line feeding Bug
- [I3-1316] Update solr and kibana port on all default configuration to the default port 8983 and 5601 accordingly, throughout all components Task
- [I3-1806] Upgrade docker base image to the latest version of adoptopenjdk Task
- [I3-1812] i3 containers run as root Bug
- [I3-1820] Running process and index with Oracle JRE distribution doesn't work Bug
- [I3-1822] Incorrect formatting in Compound word spliter stage error messages Bug
-
- [I3-1430] Unable to access search context in ElasticGroovyQueryFeature BREAKING CHANGE, Bug
- [I3-1645] After removing MultiIndexElasticClientConfiguration from the ElasticHighLevelRestSearchClient it cannot be selected again BREAKING CHANGE, Bug
- [I3-1552] As a search user, I can collapse Solr documents on a specific field Story
- [I3-1558] As a searcher user, I can apply a filter on numeric values using a slider type of interaction Story
- [I3-1557] As a search user, I can get an overview of all selected Solr filters Story
- [I3-1325] As a consultant, I can configure precision search using graph Story
- [I3-117] As a consultant, I can create facets spanning dynamic ranges Story
- [I3-1646] XSS vulnerability: urlencoded query parameter names are returned decoded in Search service response Bug
- [I3-1567] Exception thrown when creating new Application after removing one Bug
- [I3-1664] As a search user, I can query searchers across domains, so that I can create javascript search pages that reside on a different domain than the search service Story
- [I3-1023] Search - Search Result Feedback - we can send signal with a future date in timeOfFeedback field Bug
- [I3-1627] Implement context-aware completion provider Story
- [I3-1546] As a search user, I can get suggestions for words that starts with or is equal to a query term from defined starting points in the graph Story
- [I3-1685] Upgrade dependency versions Task
- [I3-1538] As a search user, I can use the latest Elasticsearch search engine and Elastic stack Story
- [I3-1555] As a search user, I can use the latest Solr search engine Story
- [I3-1548] As a search user, I can get suggestions from nearby nodes in the graph, based on defined query template Story
- [I3-1740] Disable all the validation Task
- [I3-1737] Transition from Category to Category must be aware of selected Properties Task
- [I3-1738] Transition from Property To Category without selected category should proceed properly Task
- [I3-1454] Wrong security provider example URL on the documentation page, double class name change entry Bug
- [I3-1430] Unable to access search context in ElasticGroovyQueryFeature Bug
- [I3-1645] After removing MultiIndexElasticClientConfiguration from the ElasticHighLevelRestSearchClient it cannot be selected again Bug
- [I3-1743] Transition from Gp to GC should be changed to handle two queries Bug
- [I3-1606] As a consultant, I can override default beans in i3 applications with my own custom beans Story
- [I3-1751] Upgrade precision search to use the latest features from develop Task
- [I3-1718] As a search user, I can limit the number of suggestion returned for the graph-based query suggestions Story
- [I3-1756] Unify exception handling between suggesting starting points and nearby nodes Task
- [I3-1752] Add limit clause to the nearby nodes queries to prevent from loading too much into memory Task
- [I3-1793] Images containing shell scripts might fail to start up if built on Windows machine with standard line feeding Bug
- [I3-1316] Update solr and kibana port on all default configuration to the default port 8983 and 5601 accordingly, throughout all components Task
- [I3-1806] Upgrade docker base image to the latest version of adoptopenjdk Task
- [I3-1812] i3 containers run as root Bug
- [I3-1826] Error page doesn't work on non-root context-path Bug
- [I3-1821] ElasticDateBoostFeature time unit weeks results in error from Elasticsearch Bug
- [I3-1824] Elasticsearch SearchHitConverter - forcing fields to be multivalued does not work for nested fields Bug
- [I3-1820] Running process and index with Oracle JRE distribution doesn't work Bug
-
Search editor portal 2.9.0
- [I3-1685] Upgrade dependency versions Task
- [I3-1538] As a search user, I can use the latest Elasticsearch search engine and Elastic stack Story
- [I3-1555] As a search user, I can use the latest Solr search engine Story
- [I3-1793] Images containing shell scripts might fail to start up if built on Windows machine with standard line feeding Bug
- [I3-1316] Update solr and kibana port on all default configuration to the default port 8983 and 5601 accordingly, throughout all components Task
- [I3-1806] Upgrade docker base image to the latest version of adoptopenjdk Task
- [I3-1812] i3 containers run as root Bug
- [I3-1820] Running process and index with Oracle JRE distribution doesn't work Bug
i3 3.0
Findwise i3 version 3 introduces major changes compared to version 2.5 with some breaking changes. The driver for these changes are Oracle communicating end of life for the most popular Java version out there; Java 8. This is great news because this allowed us to switch how we package and run our applications to Spring Boot. As a direct benefit we have an add-on/plug-in eco-system to all i3 applications that truly allows creating bespoke functionality in a standardized way.
Release Highlights
Epics
- Migrate i3 to build and run on Java 11
- i3 components will never be rebuilt in client projects and i3 components should be packed based on the greatest community support
Stories
- As a team member, I can run connect as spring boot application, to build upon an active community and reduce custom packaging and scripts
- As a consultant, I can extend i3 applications with custom plugins without modifying i3 components, to uphold a clear separation between product and extension
- As a consultant, I can extend i3 applications with custom spring beans without modifying i3 components, to uphold a clear separation between product and extension
- As a consultant, I can override i3 application classes without modifying i3 components, so that I can alter existing functionality to tailor i3 to my needs without changing i3 code
- As a consultant, I can read the documentation on how to use i3 with Elasticsearch TLS security and possibly find supporting tooling
- As a consultant, I can build and run connect on Java 11
- As a consultant, I can build and run index on Java 11
- As a consultant, I can build and run process on Java 11
- As a consultant, I can build and run search on Java 11
- As a consultant, I can build and run the search editor portal on Java 11
- As a consultant, I can build and run reference on Java 11
- As a team member, I can run index as spring boot application, to build upon an active community and reduce custom packaging and scripts
- As a team member, I can run process as spring boot application, to build upon an active community and reduce custom packaging and scripts
- As a team member, I can run search as spring boot application, to build upon an active community and reduce custom packaging and scripts
- As a team member, I can run SEP as spring boot application, to build upon an active community and reduce custom packaging and scripts
- As a consultant, I can extend i3 applications with web resources without modifying i3 components, to uphold a clear separation between product and extension
- As a consultant, I can package i3 Spring boot applications with docker
- As a consultant, I can build i3-extras on Java 11
- As an i3 developer, I can build commons on Java 11
- As a consultant, I can provide single ribbon.html for all components
- Upgrade to Spring and Spring Security 5.2
- As a consultant, service names are verb-based to easier understand what each service does
- As a consultant, I can run i3 with Java 11 in Docker
- As an i3 developer, I can build i3 parent POM with Java 11
- As a consultant, I can read a documentation that reflects changes introduced in JDK 11 migration
|