An Overview of ModeShape
4 likes4,695 views
The document provides an overview of Modeshape, highlighting its compatibility with the JCR API and its advantages for managing various types of content. It details the flexible structure, query capabilities, storage options, and built-in sequencers for automatic content extraction. Additionally, the document explains Modeshape's integration features, including support for federation and remote access through RESTful and WebDAV services.
![Why use JCR?
Flexible schema (part 1)
Each JCR node has:
• Name, path, & identifier
• Properties
- name & value(s) Status Report.pdf
-
jcr:primaryType nt:file
property types jcr:mixinTypes [acme:tagged]
• Child nodes
jcr:createdBy jsmith
jcr:created 2010-07-31T08:10:14
- same-name-siblings acme:tags ["status", "projectX"]
- may be ordered
• One or more node types
- primary type and mixins
- dictate whether properties and children can be added/removed
- can be changed at any time on each node
8](https://image.slidesharecdn.com/2010-09modeshape-100922075725-phpapp01/85/An-Overview-of-ModeShape-8-320.jpg)
![Why use JCR?
Sample JCR-SQL2 queries
SELECT * FROM [car:Car] WHERE [car:model] LIKE ‘%Toyota%’ AND [car:year] >= 2006
SELECT [jcr:primaryType],[jcr:created],[jcr:createdBy] FROM [nt:file]
WHERE NAME() LIKE $name
SELECT file.*,content.* FROM [nt:file] AS file
JOIN [nt:resource] AS content ON ISCHILDNODE(content,file)
WHERE NAME(file) LIKE 'q*.2.vdb'
13](https://image.slidesharecdn.com/2010-09modeshape-100922075725-phpapp01/85/An-Overview-of-ModeShape-13-320.jpg)
![Why use ModeShape?
Enhanced JCR-SQL2
ModeShape accepts all valid JCR-SQL2 queries, plus:
• Additional joins (FULL OUTER, CROSS)
• Subqueries in criteria
• SELECT [DISTINCT]
• UNION/INTERSECT/EXCEPT [ALL]
• LIMIT and OFFSET
• DEPTH and PATH criteria
• REFERENCE criteria
• IN and NOT IN
• BETWEEN val1 [EXCLUSIVE] AND val2 [EXCLUSIVE]
• Arithmetic expressions in criteria
24](https://image.slidesharecdn.com/2010-09modeshape-100922075725-phpapp01/85/An-Overview-of-ModeShape-24-320.jpg)
![Why use ModeShape?
Sample enhanced JCR-SQL2 queries
SELECT * FROM [car:Car] WHERE [car:model] LIKE ‘%Toyota%’ AND [car:year] >= 2006
SELECT [jcr:primaryType],[jcr:created],[jcr:createdBy] FROM [nt:file]
WHERE PATH() LIKE $path
SELECT file.*,content.* FROM [nt:file] AS file
JOIN [nt:resource] AS content ON ISCHILDNODE(content,file)
WHERE PATH(file) LIKE '/files/q*.2.vdb'
SELECT [jcr:primaryType],[jcr:created],[jcr:createdBy] FROM [nt:file]
WHERE PATH() IN (
SELECT [vdb:originalFile] FROM [vdb:virtualDatabase]
WHERE [vdb:version] <= $maxVersion
AND CONTAINS([vdb:description],'xml OR xml maybe')
)
25](https://image.slidesharecdn.com/2010-09modeshape-100922075725-phpapp01/85/An-Overview-of-ModeShape-25-320.jpg)
An Overview of ModeShape
- 1. An Overview of ModeShape September 2010 modeshape.org
- 2. Topics • Why use JCR? • Why use ModeShape? • Productization 2
- 3. What is JCR? JSR-170 and JSR-283 • Standard programmatic API: javax.jcr • Abstracts where/how the content is stored • Best features from databases and file systems Image credit: David Nuescheler 3
- 4. Typical uses of JCR • Content management systems • Portals (e.g., GateIn) • Knowledge bases • Test management systems • Artifact repositories (e.g., BRMS, EDS, ...) 4
- 5. Why use a JCR repository? • Flexible data structure - because your data will change • Hierarchical graph-based storage - natural organization - flexible schema - handles wide variety of content • Search & query • Versioning, events, access control, referential integrity, locking • Standard API 5
- 6. Fundamental elements of JCR Repository Workspacen Namespace Registry Node Type Manager Sessionm Sessionmm Session Sessionm Version Sessionm Sessionm Nodes Manager Lock Manager other services 6
- 7. Why use JCR? Hierarchical organization • Great for natural hierarchies - URLs and file systems - Dates - Composition - Catalogs - ... many others • Examples - Files and folders (artifacts) - Database schemas - Web services - XML, YAML, JSON, etc. 7
- 8. Why use JCR? Flexible schema (part 1) Each JCR node has: • Name, path, & identifier • Properties - name & value(s) Status Report.pdf - jcr:primaryType nt:file property types jcr:mixinTypes [acme:tagged] • Child nodes jcr:createdBy jsmith jcr:created 2010-07-31T08:10:14 - same-name-siblings acme:tags ["status", "projectX"] - may be ordered • One or more node types - primary type and mixins - dictate whether properties and children can be added/removed - can be changed at any time on each node 8
- 9. Why use JCR? Flexible schema (part 2) JCR node types: • define the allowed properties - name (or unrestricted with “*”) LONG, DOUBLE, DECIMAL, DATE, - type BOOLEAN, NAME, - number of values PATH, URI, - searchability REFERENCE, WEAK_REFERENCE, - mandatory UNDEFINED • define the allowed child nodes - name (or unrestricted with “*”) - required node types - same-name-siblings allowed • inheritance and mixed together 9
- 10. Why use JCR? Query languages • XPath From JCR 1.0, deprecated in JCR 2.0 • JCR-SQL • JCR-SQL2 New in JCR 2.0 • JCR-JQOM 10
- 11. Why use JCR? JCR-SQL2 is based upon SQL JCR-SQL2 SQL node type table property column node row Nodes appear as rows in those tables that correspond to the node’s types - a node can appear as a row in multiple tables - often a complete picture of a node requires joining multiple tables - other joins are possible, such as ancestors, descendants, equivalent properties 11
- 12. Why use JCR? JCR-SQL2 features • Select • Joins (INNER, LEFT/RIGHT OUTER) • Property & name criteria • Child & descendant node criteria • Comparison operators (=,<,<=,>,>=,<>, LIKE) • Logical operators (AND, OR, NOT) • Full-text search criteria • Variables ModeShape accepts all valid JCR-SQL2 queries 12
- 13. Why use JCR? Sample JCR-SQL2 queries SELECT * FROM [car:Car] WHERE [car:model] LIKE ‘%Toyota%’ AND [car:year] >= 2006 SELECT [jcr:primaryType],[jcr:created],[jcr:createdBy] FROM [nt:file] WHERE NAME() LIKE $name SELECT file.*,content.* FROM [nt:file] AS file JOIN [nt:resource] AS content ON ISCHILDNODE(content,file) WHERE NAME(file) LIKE 'q*.2.vdb' 13
- 14. Why use JCR? Other benefits • Versioning • Notification (events) • Access control • Referential integrity • Locking • Standard API 14
- 15. Using JCR 2.0 repositories • Uses JCR 2.0 API & Java SE ServiceLoader • No implementation dependencies Java code: Properties parameters = new Properties(); parameters.load(...); // load from a file Repository repository = null; for (RepositoryFactory factory : ServiceLoader.load(RepositoryFactory.class)) { repository = factory.getRepository(parameters); if (repository != null) break; } Properties for ModeShape: org.modeshape.jcr.URL = file:path/to/configRepository.xml?repositoryName=MyRepository or org.modeshape.jcr.URL = jndi:/jcr/local?repositoryName=MyRepository 15
- 16. Why use ModeShape? • Supports JCR 2.0 • Storage options • Access existing content thru JCR • Unified repository via federation • Automatically derives content • Lightweight, embeddable & clusterable • Plays well with JBoss AS 16
- 17. Why use ModeShape? Storage options JCR Client Application • Store content where it JCR API makes sense for your application ModeShape JCR Repository • Multiple connectors out-of-the-box ? • Easy enough to write your own ? Transient Infinispan JBoss JDBC File SVN Other in-memory data grid Cache Database System repository systems storage for storage for storage for storage 17
- 18. Why use ModeShape? Storage examples • Store in RDBMS using Hibernate - Conventional and well-understood • Store in Infinispan - Scalability, fault tolerance and speed of a data grid • Store artifacts on file system - Images, documents, downloads, static pages, etc. - Serve files via file system, bypassing JCR altogether - Manipulate content through JCR (e.g., CMS) • Transient in-memory - Lightweight and fast for embedded uses - Often used with imported data 18
- 19. Why use ModeShape? Accessing other data JCR Client Application • Use JCR API to JCR API access data from other systems ModeShape JCR Repository • Leave the data where it is ? • Just another connector ? JCR JDBC File SVN Other repositories Database System repository systems schema 19
- 20. Why use ModeShape? Access examples • Access files already stored on file system - Use the JCR API within your application - Use search and query • Access database metadata (live) - Leverage session-based caching - Consistent model • Access (& manipulate) content in other JCRs - Got data already in another JCR system? - Makes most sense with federation ... 20
- 21. Why use ModeShape? Federation JCR Client • Use JCR API to Application access a virtual, JCR API unified repository ModeShape • Leave the data Federated where it is Repository ? Transient JBoss Infinispan JDBC JDBC File SVN Other in-memory Cache data grid Database Database System repository systems storage for storage for storage for storage schema 21
- 22. Why use ModeShape? Federation examples • Own some content, access other data - Use the JCR API consistently throughout your application - Single virtual and unified repository - Leverage JCR’s flexible schema - Use search and query across all data - Leave existing data where it lives • Use the best data store for each kind of data - Use the JCR API consistently throughout your application - Single virtual and unified repository - Match access and capabilities with requirements 22
- 23. Why use ModeShape? Query languages • XPath From JCR 1.0, deprecated in JCR 2.0 • JCR-SQL • JCR-SQL2 New in JCR 2.0 • JCR-JQOM • Full-text search Like web search engines, based on JCR-SQL2’s full-text search expression grammar Separate parsers, but all use the same canonical AST, planner, optimizer, and processor 23
- 24. Why use ModeShape? Enhanced JCR-SQL2 ModeShape accepts all valid JCR-SQL2 queries, plus: • Additional joins (FULL OUTER, CROSS) • Subqueries in criteria • SELECT [DISTINCT] • UNION/INTERSECT/EXCEPT [ALL] • LIMIT and OFFSET • DEPTH and PATH criteria • REFERENCE criteria • IN and NOT IN • BETWEEN val1 [EXCLUSIVE] AND val2 [EXCLUSIVE] • Arithmetic expressions in criteria 24
- 25. Why use ModeShape? Sample enhanced JCR-SQL2 queries SELECT * FROM [car:Car] WHERE [car:model] LIKE ‘%Toyota%’ AND [car:year] >= 2006 SELECT [jcr:primaryType],[jcr:created],[jcr:createdBy] FROM [nt:file] WHERE PATH() LIKE $path SELECT file.*,content.* FROM [nt:file] AS file JOIN [nt:resource] AS content ON ISCHILDNODE(content,file) WHERE PATH(file) LIKE '/files/q*.2.vdb' SELECT [jcr:primaryType],[jcr:created],[jcr:createdBy] FROM [nt:file] WHERE PATH() IN ( SELECT [vdb:originalFile] FROM [vdb:virtualDatabase] WHERE [vdb:version] <= $maxVersion AND CONTAINS([vdb:description],'xml OR xml maybe') ) 25
- 26. Why use ModeShape? Use the data you already have • Repositories frequently store files - Many of those files contain structured information • How do you use what’s inside those files? - Download and parse in your application? - Rely upon full-text search? - Do it whenever the files change? • ModeShape sequencers get at that info - Parsers for various file formats - Extract the information and place into graph form using node types - Store derived information inside the repository - Access, navigate,and query 26
- 27. Why use ModeShape? Sequencer examples • Web service artifacts - WSDL, XSD, WS Policy, ... • Data management artifacts - DDL, Teiid models & VDBs, services, data sources, ... • Software artifacts - JAR, WAR, EAR, POM, manifests, Java source/classfiles, ... • Rule and process artifacts - Business/technical rules, functions, flows, DSLs, ... 27
- 28. Why use ModeShape? Automatically extract content 5) access 4) store 1) upload 2) notify 3) derive Sequencers 28
- 29. Why use ModeShape? ModeShape sequencers (out-of-the-box) • ZIP, JAR, EAR, and WAR files • Java class and source files • DDL files • Teiid Relational models and VDBs • Text files (CSV and delimited) Sequencer API • JCR Compact Node Definition files sequencer implementation • XML documents • Microsoft Office® documents • Image metadata 29
- 30. Why use ModeShape? ModeShape components • JCR engine • RESTful service • WebDAV service • JDBC driver • JBoss AS 5.1 kit • Eclipse plugin 30
- 31. Why use ModeShape? RESTful service • Remote access to repository • Standard HTTP methods - GET repositories - GET workspaces - GET, PUT, POST, DELETE nodes - POST queries • JSON responses (HTML, XML planned) • Uses RESTEasy • Deployed as a WAR 31
- 32. Why use ModeShape? WebDAV service • Remote access to repository • Can mount as network share • Download, upload, create or delete files • Navigate, create and delete folders • Deployed as a WAR 32
- 33. Why use ModeShape? JDBC driver • Query repository via JDBC using JCR-SQL2 • Connects to one workspace • Can be used in same VM as ModeShape jdbc:jcr:jndi:{path/in/jndi}?repositoryName={name}&user=... • Or in remote VM jdbc:jcr:http://{hostname}:{port}/modeshape-rest/?repositoryName={name}&user=... • Supports result set metadata • Supports database metadata 33
- 34. Why use ModeShape? JBoss AS kit • JCR service (the engine) • JOPR plugin for management & monitoring • REST service (WAR) • WebDAV service (WAR) • JAAS authentication & authorization • JDBC data source (1 per workspace) • JCR API on classpath • Sequencers • Disk-based storage by default (HSQLDB & Lucene) • Packaged as ZIP • Script to install into a profile 34
- 35. Why use ModeShape? JOPR plugin 35
- 36. Productization SOA-P 5.1 EDS • Includes Teiid & ModeShape • Slightly customized ModeShape JBoss AS kit - modified configuration and select connectors, sequencers • Repository for data service artifacts - relational model files - virtual database (VDB) files • Other files can be managed (but not sequenced) - XSD, WSDL, CSV, docs, etc. • Publish and unpublish from JBoss Dev Studio • Can access repository content from within VDBs 36
- 37. Productization BRMS 5.1 • Offer ModeShape as an option (in place of Jackrabbit) • Same binaries as in SOA-P 5.1 • Enable future enhancements (if desired): - sequencers for rules, flows, DSLs, etc. - query via JDBC - connectors & federation - RESTful access 37
- 38. Questions?