Overview

This document is intended to be a thorough description of the data repository design; concentrating on its external interfaces. The intended audience is repository coders as well as implementers of other components that depend on the repository.  It should be kept up to date with any API changes so it can serve as a reference manual.

...

• Workflow and lifecycle management with automated enforcement.
• Isolated administrative domains so separate groups can share a repository.
• Fine-grained access control for reading as well as modification operations.

Concepts and Internal Structure

This section describes the internal conceptual data model.

Resource Instance

The primary purpose of the repository is to store, edit, and retrieve resource instances for its clients. The term resource instance comes from the eagle-i data model: a resource is an indivisible unit of something described in the eagle-i database.  A resource instance in the repository is the corresponding graph of RDF statements that represents this resource as abstract data.  It is rooted at one subject URI.

...

• The /update service creates, modifies, or deletes exactly one single instance transactionally.
• The repository maintains metadata about each instance, described in more detail below.
• Resource instances typically live in workspaces, described in more detail below.
• The dissemination service retrieves the contents of a single instance in various formats.
• The /harvest service reports on changes to instances.

Embedded Instances (within Resource Instance)

Early in the development process, the definition of a resource instance was extended to encompass embedded instances (EI). This change only affects the rules governing what statements belong in the instance's graph, although it has a profound effect on the behavior of the repository. Embedded instances are essentially regular resource instances which are considered part of a "parent" resource instance. The precise definition of an embedded instance is as follows:

...

See the description of the /harvest service for full details. Essentially, since EIs do not have an independent presence in the "instance" model of the repository, they are not reported on individually when the harvest service reports changes. A change to an EI, even deletion of the EI, is reported as a change to its parent. Likewise, creation of an EI is also reported as a change to its parent.

Resource Property Hiding

The repository is required to hide statements with certain predicates when exporting resource instances in these contexts:

...

• Properties that are effectively provenance metadata added by users in the resource acquisition and curation process, but are not suitable for public viewing. They may contain confidential information or comments that the curators do not want publicized.
• Properties whose values inherently contain confidential information, or information such as e-mail addresses, physical locations, and phone numbers that administrators have been directed to hide.

Contacts and E-mail

The "contact" issue is closely related to property hiding. The essential problem is that for every resource instance, it is desired to have a means for the agent viewing that resource. For example, a user in a Web browser viewing a semantic Web-style dissemination page of the resource instance to contact the agent who "owns" (or is otherwise responsible for) the resource. E-mail is one means of implementing this contact, but certainly not the only one. The contact could be in the form of a telephone number, street address, or even a redirect to another Web site which might include indirect contact info of its own. The purpose is to put a potential consumer of the resource in touch with its owner.

...

The mechanism of contact-property hiding is implemented through access controls, explained in more detail later in this document.

Internal Ontology and Metadata

There is a separate ontology document describing the repository's internal data model and administrative metadata, refer to repository-internal.n3 for more information. Note that some statements described by that ontology appear as publicly-readable metadata statements, while others are private and never exposed outside of the repository codebase.

The "ontology" graph is considered read-only in normal operation. All internal metadata, (administrative metadata) is stored in a separate, distinct, named graph which should only be available to the repository's internal operations.

Named Graphs

The repository design takes full advantage of the named graph abstraction provided by most modern RDF database engines (Sesame, Virtuoso, Anzo, AllegroGraph). Every statement in the RDF database belongs to exactly one named graph. Since this is typically implemented by adding a fourth column to each triple for the named graph's URI, these databases are often called quad-stores instead of triple-stores. The data repository design takes advantage of named graphs to:

...

Internally, we collect some metadata about each named graph including access control rules, and a type that documents the purpose of the named graph.

Named Graph Types:

1. Ontology---contains ontology descriptions, so reasoning may be required. The name of the graph is the namespace URI. Also implies that the graph is part of the "TBox" for inferencing.
2. Metadata---contains the repository's own metadata; public and private metadata are in separate graphs for easier containment of query results.
3. Workspace---holds resource instances still in the process of editing and curation. Not visible to the public, and access to some workspaces is restricted to a subset of users.
4. Published--Published resource description data.
5. Internal---Contains data that is only to be used internally by repository code and never exposed.

...

Relationships---it would be helpful to record metadata about related named graphs, although the most compelling case for this is the ontologies that use owl:includes to embed other ontology graphs. Since Sesame does not interpret OWL by itself, and we have no plans to add this sort of functionality for the initial repository implementation, this will be considered later.

Views

The repository provides views to give clients an easier way to query over a useful set of named graphs. A view is just a way of describing a dataset (a collection of named graphs). The repository server has a built-in set of views, each named by a simple keyword. You can use a view with the SPARQL Protocol and with a resource dissemination request. It is a equivalent to building up a dataset out of named graphs but it is a lot less trouble, and guaranteed to be stable whereas graph names could change. The views are:

• published---all resource instances and user records visible to the public, and all relevant ontologies and metadata.
• published-resources ---just resource instances visible to the public, and all relevant ontologies and metadata.
• metadata---all named graphs of type Metadata visible to the authenticated user.
• ontology---all named graphs of type Ontology visible to the authenticated user
• metadata+ontology---all named graphs of types Metadata and Ontology visible to the authenticated user
• null---all statements in the internal RDF database regardless of named graph. Administrators only.NOTE: This is the ONLY way to see any statements that do not belong to any named graph, i.e. the "null context" in Sesame. If we are lucky this will be a small or empty set.
• user--- those graphs which that the current user has permission to read.
• user-resources---the graphs containing or related to eagle-i resources which that the current user has permission to read. Note that the graph containing instances of repository users (i.e. user,s of type foaf:Person) is , ironically, NOT part of this data set, since they are not eagle-i resources.
• public---all graphs which that are visible to the public, i.e. to the anonymous user, plus inferred statements. Note that this is not the same thing same  as 'published'.
• all---all named graphs including the ones internal to the repository; administrators only.

Important Note: You may have noticed that according to the definition, the user view is the same as the all view for an administrator user, so why bother creating an all view? It is intended to be specified when you have a query that really must cover all of the named graphs to work properly; if a non-administrator attempts it, it will fail with a permission error, instead of misleadingly returning a subset of the graphs.

Workspaces

A workspace is just another way to describe a data set, by starting with a named graph. It is effectively a special kind of view. The name of a workspace is the URI of its base named graph, which must be of type workspace or published. When you specify that as the workspace, the repository server automatically adds these other graphs to the data set:

...

You can specify a workspace instead of a view in SPARQL Protocol requests, and in resource dissemination requests.

Inferencing

The repository supports inferencing in some very specific cases. Since the repository's RDF data is very frequently modified, it does only the minimal inferencing needed by its users in order to keep the performance bearable.

...

See the RDF Semantics page for an overview of the greater body of inference rules (of which we implement only implement a small subset). The repository implements two different kinds of inferencing:

1. "TBox" inferencing:
   • TBox is the terminology source, i.e. the ontology graphs.
   • It consists of named graphs whose type is ontology.
   • Upon * any change to a TBox graph, inferencing is redone on the entire graph, and on all the ABox graphs.
   • All inferred TBox statements are added as inferred statements directly to their TBox graph.
   • Inference is done independently on every TBox graph they are expected to be completely disjointed. Thus, there should only be one or two TBox graphs.
   • Following entailment rule rdfs11, all inferred rdfs:subClassOf relationships are created as direct statements.
   • Following entailment rule rdfs5, all inferred subPropertyOf relationships are created as direct statements.
   • Following entailment rule rdfs9, all inferred rdf:type properties are added.
2. " ABox"   inferencing:
   • ABox is the body of assertions, i.e. the statements about resource instances.
   • All non-TBox named graphs are part of the ABox.
   • All statements created from inference on ABox statements are added to a special named graph, http://eagle-i.org/ont/repo/1.0/NG_Inferred.;
     • This makes it easy to include or exclude the inferred statements in the dataset of SPARQL query
     • You can detect whether or not a statement is inferred by adding a GRAPH keyword to the query and testing its home graph.
   • Any change to a TBox graph only causes re-inferencing of the subject of each statement where a named instance was changed. This is possible because the inferred statements only depend on the asserted rdf:type properties of a a subject and the TBox graphs.
   • Following entailment rule rdfs9, all inferred rdf:type properties are added to the graph for inferred statements.

...

This inferencing scheme ensures very fast performance by assuming the TBox graphs never change under normal operations, which ought to be true. The data model ontology graph is only modified when a new version of the ontology is released. Likewise, the repository's internal ontology graph remains unchanged once the repository is installed. When the TBox graphs are changed, be aware that you will probably see a delay of many seconds or perhaps minutes, as all the TBox and ABox inferencing is re-done.

Inferred statements are not normally written when an entire graph is dumped. See the /graph service for details.

Users and Authentication

Authentication is managed entirely by the Java Servlet container. We rely on the container to supply an authenticated user name (a short text string) and whether that user has the "superuser" role. The container's role is only used for bootstrapping; normally roles are recorded in the RDF database and they take precedence over the container's role map.

Users Are Described in Two Databases

Each login user is (ideally) recorded in both the RDBMS used by the servlet container (or possibly some other external DB) and the RDF database. This is necessary because the servlet container, which is doing the authentication, only has access to the RDBMS through a plugin, but the repository authorization mechanism expects an RDF expression of the user as well. All of the services that modify users keep the RDBMS and RDF databases synchronized, and can cope with users found in one and not the other.

The RDBMS description of a user contains:

1. Username---*Present present in RDF as well, this is the common key. Since it is the permanent description of a user this is is  *immutable.
2. Password---Only only in RDBMS.
3. Membership in Superuser role. No other roles.

The RDF description of a user contains:

1. Username of the corresponding RDBMS user.
2. Roles including Superuser if present in RDBMS
3. Various descriptive and provenance metadata.

...

When a user is present in the RDBMS but not in the RDF, they are considered undocumented. Upon loginlogging in, an undocumented user is given the URI corresponding to his/her highest known role:
:Role_Superuser if the RDBMS indicates that role, or :Role_Anonymous otherwise. (Arguably, the default role could also be :Role_Authenticated, but without RDF data for the user they are not fully authenticated, and this is incentive to fix the discrepancy.)

...

An Administrator (superuser); can become documented by logging in, and either running the \whoami service with create=true or using the Admin UI to edit and save their own user information. An Administrator can fix an ordinary undocumented user by using the Admin UI to save their descriptive metadata; even if it is all blank, a user record will be created. Importing users also straightens out the mapping automatically.

About Roles

Roles are a way to characterize a group of users, for example, to grant them some access rights in the access-control system. Functionally, the role is part of a user's authentication, i.e., "who" they are.
A role is defined by a URI, the subject of a :Role instance. It should also have a locally unique, short text-string name (the :label of its :Role instance).

Each Role is independent of other Roles. Roles cannot be "nested". \   This is a necessary limitation that simplified the implementation considerably.

The Superuser role is built into the system because its privileges are hardcodedhard coded.

There are a couple of special Roles whose membership is implicit, that is, it never needs to be granted explicitly:

1. Anonymous---unauthenticated users are implicitly assigned this role so they can be given explicit access. This role is never visibly asserted by a user, it is only for describing access controls: For example, "The Published graph is readable by the Anonymous role".
2. Authenticated ---any user who logs in is authenticated and implicitly belongs to this role; the opposite of anonymous. This role is never explicitly asserted by a user, it is only for describing access controls.

Username and Password

A repository user is identified uniquely (within the scope of ONE repository instance) by a short symbolic username. This is a character string composed of characters from a certain restricted subset of the ASCII character set, in order to avoid problems of character translation and metacharacter interpretation in both the protocol layer (HTTP) and OS tools such as command shells. The password, which is paired with a username to serve as login credentials, is likewise restricted to the same range of characters as the username.

...

Note that although the HTTP protocol allows any graphic characters in the ISO-8859-1 codeset (modulo ':'), and linear whitespace, and even chars WITH special MIME RFC-2047 encoding rules, these are often implemented wrongly by HTTP clients and also invite encoding and metacharacter problems with OS and scripting tools.To avoid these troubles we simply restrict the available characters.

Authentication policy

All of the servlet URLs in this interface except the public dissemination request (/i) require authentication. The dissemination request makes use of an authenticated user and roles when they are available, to access data that would be invisible to an anonymous request, but it is never required.

Access Control (Authorization)

This is just an outline of the access control system. It is implemented as statements stored in the internal metadata graph. The access controls applying to an instance or other object are not to be directly visible in the repository API, except through administrative UI.

These types of access can be granted:

• Read---in context of named graph, allows the graph to be included in dataset of a query; for workflow transitions and hidden/contact property groups, this is the only relevant access grant.
• Add---add triples to a graph or instance.
• Remove---remove triples from a graph or instance.
• Admin---change the access controls on a resource. (Not really used in practice; only the Superuser role can change access.)

On the following types of * resources*:

• Named Graphs, including workspaces.
  • Read lets people download and run queries against the graph.
  • Add, Remove lets you modify it with /graph Usually reserved for admins.
• Resource Instances (ignores explicit Read access; that comes from it's home named graph.)
  • Add, Remove let you modify it with /update, this is usually granted temporarily to a specific user as part of the workflow process.
  • ignores Read grants, read access reverts to its home graph.
• Property Groups - sets of properties on a resource instance identified by statements in the data model ontology, see the data model configuration manual, particularly datamodel.hideProperty.predicate, datamodel.hideProperty.object.
  • Read lets the Dissemination and Harvest service report on these properties.
  • All other access types are ignored.
• Workflow Transitions
  • Read gives access to take (push) the transition.
  • All other access types are ignored.

...

The resource is the URI of the instance, named graph, or workflow transition of interest. The access-type names one of the four types of access described above: read, add, remove, admin. Finally, the accessor is the URI of the Principal to be granted the access, either a Role or an Agent (user).

We anticipate having a relatively small number of these access grants. Although named graphs and workflow transitions need elaborate access descriptions, there are only a few of those--on the order of dozens. Resource instances are of course more numerous but most of them have no access grants, deriving their read/query access from the named graph they reside in. The workflow claim service adds temporary grants to give the claim owner read/write access to be able to edit the instance while it is claimed.

Provenance Metadata

The repository automatically records provenance metadata about objects when they are created and modified by users' actions. Provenance means information about the history and origin of the data, in this case the authenticated identity responsible and time of the latest change. The following properties are recorded for these types of objects, and can be obtained by querying with a view or dataset that includes the named graph containing public administrative metadata.

...

• dcterms:modified---literal date of last modification, encoded as xsd:dateTime
• dcterms:contributor---the URI of the Agent (authenticated user) who last modified it
• dcterms:source---description of the file or URI last loaded into this NG, if that is how it was created.  This record is used to compare it against the source later to decide whether an update is necessary. It is a node (possibly blank node) with the following properties:
  • dcterms:identifier---the resolvable URI of the resource loaded, most likely a URL in either the file or http scheme.
  • dcterms:modified---last-modification date of the resource, for later comparison when deciding whether to decache the repository copy of an external file, a literal xsd:dateTime.
    Resource Instance:
  • dcterms:created---literal date when resource was created, encoded as xsd:dateTime
    • Set automatically by the Repository.
  • dcterms:creator---the URI of the Agent (should be an authorized user) who created the instance.
    • The true meaning is the user who authored the data in the inital version of this instance.
    • Usually, this is the same as the user directly responsible for creating the instance.
    • However, when a different user uploads, a spreadsheet created by other RNAVs, for example, the Repository user is a mediator. The actual value of dcterms:creator comes from the uploaded data.
  • (optional) dcterms:mediator---ONLY when dcterms:creator does not refer to the authenticated user who created the data, this is the URI of the Agent (authenticated user) who created this instance in the Repository.
    • Set automatically by the Repository.
  • dcterms:modified---literal date when resource was last modified, encoded as xsd:dateTime
    • Set automatically by the Repository.
  • dcterms:contributor---the URI of the Agent (authenticated user) who last modified this instance.
    • Set automatically by the Repository.

Sesame Triplestore Extensions

Some repository features are implemented as extensions to the Sesame RDF database (also known as triplestore). This means they are available both internally to the repository implementation and externally whenever an API to Sesame, its SPARQL query engine, is exposed.

...

To invoke it you must have the repository's URI namespace defined as a prefix. For example,

 PREFIX repo:<[http://eagle-i.org/ont/repo/1.0/]>
..query text...
 ORDER BYrepo:upperCaseStr(?label) 

Workflow

The repository includes a "workflow" control system that directs the complete life cycle of each resource instance, and mediates access by users at each life cycle stage. The word "workflow" is often used to describe process-management and administration systems, but in this case it is really just a minimal implementation of states and extended access control.

...

1. Workflow State---A node on the state map. Every resource instance has exactly one current state. A newly created resource is initialized to a fixed "New" state.
2. Transition---Description of an arc on the map, or a transition from an initial state to a final state.
3. Claim---Assertion on a resource instance that a specific user (the claimant) has taken possession of it, in order to prepare it for the next workflow transition.
   • Claiming typically causes some side-effects such as changing access controls.
   • A claim may be resolved by taking a transition to another state, or by releasing it.
4. Pool---The set of resource instances available for claiming to a specific user. A pool is always computed by a query, it is not materialized.

Bootstrapping and Configuration

Bootstrapping refers to how a new repository node first starts up. The process is not trivial, since so much of its operation depends on the RDF database (triple-store, actually a quad-store) which is completely empty when a new repository is launched.

...

1. Install the webapp by dropping the WAR file into place, and adding configuration:
   1. Add one system property to the Web server's JVM to indicate the configuration properties.
   2. Create configuration properties file with at least the minimum required properties.
2. Create authentication DB and add the initial Administrator user.
3. Start the servlet container---the repository webapp automatically initializes the following named graphs if they are found to be empty:
   1. The internal Repository ontology (read-only)
   2. Internal Repository administrative metadata (e.g. Roles, Named Graphs, WF Transitions, etc.)
   3. SPARQL Query macros.
4. Create an RDF metadata wrapper for the initial administrator user - this is done automatically by the post-install procedure, finish-install.sh. See the Administrator Guide.
5. Load the eagle-i data model ontology. Although this is not necessary for any of the Repository's internal functions, the eagle-i apps will expect it to be there.
6. Browse to the Admin UI and log in as the Administrator user. You can create user logins and assign roles; modulo any customizations of workflow, workspaces and such. With that, the repository is open for business.

Configuration properties

The configuration properties are loaded from the file configuration.properties in the repository home directory. It is read by Apache Commons Configuration, which allows system properties and other variables to be interpolated into the values. See the Apache Web site for complete documentation.

See the Configuration section in the Administrator Guide for a complete list of configuration properties.

Basic Repository REST API

These are the possible HTTP requests in the repository API.

Webapp Mount Point

The repository's webapp must be mounted at the web server's root, so that it can resolve the canonical URI of resource data instances.

Format Identifiers and MIME Types

This section lists the formats that the repo repository can use for output and, in some cases, input, of data. The MIME type is how you describe it in the API, whether through explicit args or HTTP headers.

...

Identify and/or "Create" Current User ( /whoami)

This request returns a tabular report on the current authenticated user's identity, to compose a friendly display in a Web UI. This is simply a SPARQL query wrapped in a servlet to hide the details of the internal data.

...

Access:
Open to authenticated users.

Create New Instance Identifier(s) (/new)

This call creates one or more globally unique, resolvable, URIs for new resource instances. It does not add any data to the repository; the instances will not exist until a user inserts some statements about them. The URI namespace is the default namespace from the configuration properties, followed immediately by the unique identifier. Note that ETL tools may request thousands of URIs at once so the mechanism to produce unique IDs must be able to handle that.

...

Access: Requires an authenticated user.

Disseminate Instance (/i, /repository/resource)

The disseminate service returns the RDF content of an instance; it is how the URI is resolved to implement the Linked Open Data paradigm of the Semantic Web. Note that there are actually three valid ways to construct a request for any given data instance:

...

About HTML dissemination:
When the negotiated format is text/html, and unless either of the forceRDF or forceXML args was given, the dissemination process creates an intermediate XML document and transforms it into XHTML with the configured XSLT stylesheet. See description of the eaglei.repository.instance.xslt in the Repoistory Administrator Guide.

If no XSLT stylesheet is configured, the intermediate XML document is delivered instead, with a media content type of application/xml. Note that this means, to obtain correct XHTML output, you MUST configure an XSLT stylesheet.

The content of the intermediate XML format is described in a W3C XML Schema document that may be downloaded from a running repository at for example,

 [https://localhost:8443/repository/schemas/instance.xsd] 

We provide an example transformation stylesheet that produces very simple HTML, intended to be the basis of custom stylesheets. It is available for download at .:{{https://localhost:8443/repository/styles/example.xsl
}}We manage the transformation within the repository, instead of adding an xml-stylesheet processing instruction to the XML, for compelling reasons:

...

There is also access control on some indivdual individual properties of the resource: those properties identified as hidden and contact properties by the data model ontology (and its configuration, see that separate document). The access controls on the resource URI configured as datamodel.hideProperty.object regulate hidden proertiesproperties, and datamodel.contactProperty.object regulates contact properties.

Anyone with READ access on the URI gets to see the properties. To expose them to the world you'd give access to the Anonymous role. Normally only Curator, RNAV, and Lab User roles would be granted access to hidden and contact properties since they have to see and manipulate them through the data tools.

Update a Single Instance (/update)

This service actually implements three different kinds of requests:

...

• Content-Type---header on the value of the content entity in a POST request. This takes precedence because it allows for different content-types in insert and delete args.
• format---query argument value.

Wiki Markup*URL:* {{/repository/update}} \ [_/instance-ID_\] (POST only)

Args:
{{ uri---}}optional way to explicitly specify the complete URI, instead of assuming that the URI's namespace matches the hostname, context, and servlet path ("/i") of this webserver.
format---the default expected format for insert and delete graphs. If the args specify a content-type header, that overrides this value. Only recognizes triples even if the format supports quads.
action=(update|create|gettoken)---Update to modify an existing instance, create adds a new one. See below for details about gettoken.
token=uri---When action is update or create, this must be supplied. The value is the URI returned by the last gettoken request.
workspace=uri---choose workspace named graph where new instance is created. Only necessary when action=create. Optional, default is the default workspace. DO NOT specify a workspace when action=update.
delete---graph of statements to remove from the instance; subject must be the instance URI. Deletes are done before inserts. Graph may include wildcard URIs in predicate and/or object to match all values in that part of a statement.
insert---graph of statements to add to instance; subject must be the instance URI.
bypassSanity---(boolean, default false, deprecated) NOTE: It is best if you pretend this option does not exist. When true, it skips some of the sanity tests on the resulting instance graph, mostly the ones checking the integrity of Embedded instances. Requires Administrator privilege. This was added to make the data migration from broken old EI data possible, it should rarely if ever be needed.

The delete wildcard URI is http://eagle-i.org/ont/repo/1.0/MatchAnything

Result:
HTTP status indicates success or failure. Any modifications are transactional; on success the entire change was made, and upon failure nothing gets changed.

...

We may eventually decide to implement a quota on the count of statements that may be added or deleted, as a protection against DOS attacks and runaway clients.

SPARQL Protocol Endpoint (/sparql)

This servlet is a complete implementation of the standard SPARQL Protocol, implemented according to the description in http://www.w3.org/TR/2008/REC-rdf-sparql-protocol-20080115/. It also extends the protocol to support specific features of the repository REST API, such as views and workspaces.

...

Args:
query---the text of the SPARQL query document.
default-graph-uri---add this URI to the list of default graphs in the query dataset, overriding any built-in default. Read access required. Conflicts with other explicit methods of specifying datasets: the *-graph-uri, view, and workspace arguments are all mutually exclusive. Can be specified multiple times to include multiple graphs.
named-graph-uri---add this URI to the list of named graphs in the query dataset, overriding any built-in default. Read access required. NOTE: Sesame's query engine does not seem to pay attention to named graphs, just default graphs, so this argument should be deprecated.
format---explicitly names a result format, overriding the default from HTTP content negotiation. Value must be one of the specified MIME type strings. This is a nonstandard extension to the SPARQL Protocol.
view---symbolic name of a pre-packaged set of named graphs presenting a "view" of the data; see secton on Views in the Concepts section above. Conflicts with other explicit methods of specifying datasets: the *-graph-uri, {{view, and workspace arguments are all mutually exclusive. This is a nonstandard extension to the SPARQL Protocol.
workspace---URI of the workspace named graph to take the place of the default graph. Relevant metadata and ontology graphs are included automatically. Conflicts with other explicit methods of specifying datasets: the *-graph-uri, view, and workspace arguments are all mutually exclusive. This is a nonstandard extension to the SPARQL Protocol.
time---sets an explicit time limit on the query, overriding any built-in default. Value is in seconds. This is a nonstandard extension to the SPARQL Protocol.
inferred---boolean arg.  When false, any inferred statements are excluded from the results. The default is true. This is a nonstandard extension to the SPARQL Protocol.

...

Access:
Requires Read access to all of the named graphs included in the dataset. When workspace or view arg with a dynamic view is supplied, the dataset is automatically constructed to include only readable named graphs out of a set that depends on the chosen view or workspace.. If a set of graphs is explicitly requested, it is an error if any of them are not readable.
Access NOTE: It MAY be desireable to configure the servlets to allow unauthenticated access to /sparql. This would only allow query access to named graphs which the anonymous role can read.

List Named Graphs (/listGraphs)

Returns an enumeration of all named graphs in the repository matching the search criteria, and which the current authenticated user can read. This request is equivalent to a SPARQL tuple query, except it hides the details of internal administrative metadata structures and bypasses most users' lack of read access to that metadata.

...

Access:
Requires authenticated user; reports on all graphs and contexts regardless of your access..

Dump Contents of a Named Graph (/graph - GET)

Creates a serialization of all statements in the indicated graph (or, optionally, all named graphs in the repository). This is typically used for administrative tasks such as backing-up and migrating the contents of the repository.

...

Request must include either the name argument or all.

URL: /repository/graph (GET)

Args:
format---ps output serialization format, overrides what HTTP content negotiation would decide.
name=URI---Name of graph to dump - mutually exclusive with all.
all---dumps all (accessible) graphs. Format must be one that encodes quads, e.g. TriG or TriX, or an error will occur.
inferred---include inferred statements as well. By default they are left out. This argument does not need a value, it is boolean so just including its name as a query arg will assert it. (SEE WARNING BELOW)

...

Access:
Requires read access on the selected graph(s). When all is specified, non-readable graphs are ignored.

Load Serialized RDF into Named Graph (/graph - POST, PUT)

Import a serialized RDF document into a specific named graph, optionally replacing all previous contents. The repository records some provenance metadata about when and who did the import, and a description of the source material.

...

Note on File Format and Character Set: The request specifies the file format and/or character set of the serialized RDF data as a Content-Type value, for example:

 text/rdf+n3; charset="ISO-8859-1"

. The character set defaults to Unicode UTF-8, so if your source data is not in that character set you must declare it. This can be provided in several different places,they are searched in this order of priority, and the first one found is the only one considered:

...

Args:
name=named-graph-URI---Required required name of the affected graph. Mutually excluseive exclusive with all.
all---loads serialized data into all (accessible) graphs. Format must be one that encodes quads, e.g. TriG or TriX.
action=(add|replace|delete)---Requiredrequired, determines whether new statements are added to existing contents of the graph, replace the contents, or are deleted from teh the graph.
format---MIME type input serialization format, overrides the Content-Type in HTTP protocol.
type=(metadata|ontology|workspace|internal|published)---Keyword keyword representing the type of content in this named graph.
label=text---value for rdfs:label property of the named graph, which appears as its title.
content=---The immediate content of serialized RDF to load into the graph. ONLY used when method is POST, but then it is required.
source=string---The file or URL to record as the source of the data being ingested into the graph. Only applies to add and replace actions. Overrides any filename specified with the content, which would otherwise be recorded as the source identifier. It is recorded as the dcterms:identifier property of the dcterms:source node in the metadata.
sourceModified=dateTimeStamp---last-modified time for the source of data being ingested. Value must be parsed as XSD date-time expression. It is recorded as the dcterms:modified property of the dcterms:source node in the metadata.

Result:
HTTP status indicates success.

Access:
If the graph already exists, requires add access (and remove access action is replace or delete). If there is no existing graph by that name, only the administrator can create a new one. For multi-graph (all) load, requires superuser access.

Logout (/logout)

Destroys any lingering session state and credentials for the current user. Users of shared or public computers need a way to positively end a session so that subsequent users do not get their access to the repository.

NOTE: Beware of this if you are using a Web browser to access the repository . The repository uses HTTP Basic Authentication to identify users. Most web browsers cache the last Basic Authentication credentials for a site and don't offer an easy way to clear it.

For Mozilla Firefox 3, try the Web Developer add-on and select Miscellaneous->Clear Private Data->HTTP Authentication from its menus.

URL: /repository/logout (POST only)

Args: none.

Result:
HTTP status indicates success. Succeeds even if there was no session to destroy.

Access:
Requires a login and established session.

Logout (/logout)

Destroys any lingering session state and credentials for the current user. Users of shared or public computers need a way to positively end a session so that subsequent users do not get their access to the repository.

NOTE: Beware of this if you are using a Web browser to access the repository The repository uses HTTP Basic Authentication to identify users. Most web browsers cache the last Basic Authentication credentials for a site and don't offer an easy way to clear it.

For Mozilla Firefox 3, try the Web Developer add-on, and select Miscellaneous-> Clear Private Data-> HTTP Authentication from its menu.

...

Access:
Requires a login and established session.

Check or Load Data Model Ontology (/model) - Admin Only

This service either reports on or loads the Data Model Ontology. The GET request returns a report of the version of the ontology loaded, and the version that is available to be automatically loaded from within the repository's codebase. The POST request give a choice of selective update or forcibly replacing the ontology.

...

The value of loaded is the result of this query over the ontology view:

 select * where {<http://purl.obolibrary.org/obo/ero.owl> owl:versionInfo ?loaded} 

URL: /repository/admin/model (POST)

...

Loads a new copy of the data model ontology, replacing the entire existing contents (if any) of its named graph. For more details about how the graph name and data model source are determined, see The Data Model Configuration Properties Guide.

When action=update, only load the ontology if the version of the current ontology is different from the version of the available one (also if there is nothing loaded yet). Note that it does not attempt to judge if the available version is later, just different.

...

The source argument determines where the ontology data comes from. When source=jar, which is recommended, the ontology is the collection of all OWL files in the maven package eagle-i-model-owl}}as it was loaded into the repository's source base. This ought to reflect the version of the ontology for which the sofware was designed. When {{source=model, the EIOntModel is used as the data source. Although there should be no difference between them, because the EIOntModel is loaded in to Jena from the same source files, in practice there appears to be some differences.

Manage Internal Graphs (/internal)

URL: /repository/internal (GET, POST)

...

This service is intended for repository internal maintenance and testing procedures ONLY. If{ you are planning to use it in a user application, you are very likely making a bad mistake. Its API is unstable and will probably change without any notice.

GET Method:

name=uri - optional, URI of the internal graph to report on, default is to list all known intenral graphs.
  format=mimetype - optional, default is negotiated, must be tuple query result format.

Result is a tabular query response with teh the following columns:

1. name--- URI of the graph
2. loaded---version of initializer that was loaded, if any
3. available---version of initializer that is avaialble and would be loaded.

...

This is a multi-function service, depending on the value of action: When action=load, replaces contents of an internal graph with hte initializer in the webapp. For action=read, returns contents of initializer for given graph in the result docuemnt. For action=query, runs given SPARQL query over contents of initializer(s) for given graphs (this is the ONLY form that can take multiple graphs). For action=decache, tells repository server to decache all in-memory data derived from RDF content in case, for example, the internal graphs have been modified or updated. All of these actions require Administrator privilege.

Harvest Resource Metadata (/harvest)

This service harvests the "metadata", which is actually the resource instance data from the repository. It is intended to be used by an external search index application to keep its index up-to-date by incremental updates, since that is much more efficient than periodically rebuilding the index.

...

The set of properties returned when detail=full is automatically filtered to remove hidden and contact properties as dictated by the access privileges of the user making the request. The filtering follows the same access-control rules as the Dissemination service.

URL: /harvest; (GET or POST method)

...

Precise last-modified timestamp: the /harvest service also adds a non-standard HTTP header to its responses to describe the last-modified time in full precision down to the millisecond level. This allows an application to give that precise time as the after argument value on the next incremental harvest to be sure of capturing any changes since that last /harvest call. The header name and date format looks like:

 X-Precise-Last-Modified:Mon, 10 Jan 2011 20:49:10.770 GMT 

Note that the time format is essentially the HTTP time format with ".mmm" appended to the time for milliseconds.  It must be converted to the XML timestamp format to serve as an argument value to "after". By the time it was obvious what a stupid choice this was, it was too late in the development cycle to change easily, but perhaps in the future an XML-format precise timestamp header can be added.

Access control:
Read permission is required on all desired graphs.

Export and Import of Users, Resource Instances (/export, /import)

This service lets you export the users OR data contents of one repository and import them into a different one, or a later instance of the "same" repository. It is intended to migrate and mirror data between different releases of the software and ontology, or different deployments created for testing. Although the same service can handle both "user" and "resource" instances, these are two separate operations and cannot be combined.

...

Args:
format---the MIME type (and charset) of the input document's serialization format. This overrides any content-type on the content arg's entity.
type=(user|resource|transition|role|grant)---which type of data is to be imported, one MUST be chosen and there is no default. MUST match the type of the exported data.
content=stream..---the serialized RDF data to import, must have been generated by export of the same type of data.
transform=(yes|no)---Required, no default. When 'yes', URIs of imported instances are transformed into new, unique URIs resolvable by the importing repository. When 'no', URIs are left as they are.
duplicate=(abort|ignore|replace)---how to handle an import that would result in a duplicate object. Default is abort.
newgraph=(abort|create)---how to handle a resource instance import that would result in creating a new named graph for the data. Must be superuser to choose 'create'. Default is abort.
{{include=list.. }}---Explicit list of user or resource instances to include in the import. Format is a series of resource URIs separated by commas (",") and optional spaces. Users may also be referenced by username. IMPORTANT NOTE: The URIs in the include (and exclude) lists are matched against URIs in the imported source file, NOT the repository.
{{exclude=list.. }}---Explicit list of user or resource instances to exclude from the import. Mutually exclusive with include, format is the same. Useful for avoiding conflicts, for example, when loading a user export into a repository where the initial admin user already exists.
ignoreACL=(true|false)---Ignore all access control grants on imported objects. The imported objects will not have any specific access grants.

Send Contact E-

...

mail  (/emailContact)

This service has not been rigorously designed, so it does not have a specification of behavior. This section only documents how to call it, the internal actions it takes are still under discussion.

URL: /emailContact (POST method only)
Args:
uri=subject-URI---required
client_ip=IP addr---required
from_name=personal name---required
from_email=mailbox---required
{{ subject=text
{{ message=text
test_mode---default false, true if present.

Sends an email message to the designated contacts for the resource identified by URI. The exact rules for determining that contact address have not been formally specified yet. When a contact cannot be determined, the mail is sent to the configured postmaster for the repository. See the Admin Guide for Repository Installation and Administrator Guide for details about configuring that and the rest of the e-mail service.

Workflow Services REST API

Show Transitions

/repository/workflow/transitions

Method: GET or POST
Args:
workspace=URI---restrict results to transitions applying to given workspace (includes wildcards). Default is to list all.
format=mime---type of result.

...

• Subject URI
• Label
• Description
• Workspace URI
• Workspace Label
• initial-state URI
• initial-state Label
• final-state URI
• final-state Label
• allowed - boolean literal, true if current user has permission on this transition

Show Resources

/repository/workflow/resources

Method: GET, POST
Args:
state=URI|all---workflow state by which to filter, or 'all' for wildcard.
type=URI--- include only instances of this type (by inference counts)Default is no restriction.
unclaimed=(true|false)---when true, unclaimed resources are included in the report. Default is true.
owner=(self|all|none)---show, in addition to any selected unclaimed resources:

• self---those claimed by current user
• all---instances claimed by anyone
• none---includes no claimed resources. Default is self. Note that the combination unclaimed=false and owner=none is illegal.
workspace=URI---restrict results to resources in a given workspace; optional, default is all available worksaces.
  format=mime---type of result, one of the SPARQL query formats.
  detail=(brief|full)---columns to include in results, see description below.
  addPattern=SPARQL-text---add this triple pattern to SPARQL query. May refer to variables as described below. Value is a literal hunk of triple pattern ready to be inserted into the patterns of a SELECT query. It may refer to the result variables mentioned below, e.g. ?r_type.
  addResults=query-variable..---add these variables to results columns. value is expected to be a space-separated series of query variables (which must appear in the triple pattern), e.g. "?lab ?labname"
  addModifiers=modifiers..---add a SPARQL modifiers modifier clause to the query. Modifers include ORDER BY, LIMIT, and OFFSET and can be used to implement pagination.

...

• The brief level of detail includes columns:
  • r_subject - URI of resource instance
  • r_label - label of resource instance
  • r_type - URI of instance's asserted type
• The full level of detail adds the fields:
  • r_created - created date from provenance
  • r_owner - URI of workflow claimant if any
  • r_ownerLabel - the rdfs:label of r_owner if it is bound (and has a label)
  • r_state - URI of workflow state
• Any query variables you specified in the addResults list are added to either result.

Claim Resource Instance

/repository/workflow/claim

Method: POST
Args:
uri=URI---subject to claim
user=UserURI---optional, user who asserts the claim; default is authenticated user.

...

1. Adds insert and delete access to the instance for the claiming user.
2. Sets :hasWorkflowOwner property to the claimant (in an internal metadata graph)

Release (claimed) Resource Instance

/repository/workflow/release

Method: POST
Args:
uri=URI---subject to release

...

Removes insert and delete access to the instance for the current user.
Removes :hasWorkflowOwner property (in an internal metadata graph)

Transition on Resource Instance

/repository/workflow/push

Method: POST
Args:
uri=URI---subject to transition
transition=URI---indicates transition to take

...

• Implied release of any current claim (and all applicable side-effects of that)
• Executes action associated with this transaction, if any.
• Resource's workflow state will become the final state of the transition.

Admin UI Support Services

There are also some REST services created to support the administrative GUI (the JSP pages). They are NOT intended to be part of the regular API since their exact function and parameter structure is dictated by the JSPs and is thus still somewhat fluid. Most of them also require Administrator access.

...

/repository/admin/updateUser
/repository/admin/updateGrants
/repository/admin/updateRole
/repository/admin/updateTransition
/repository/admin/updateNamedGraph

Site Home Page

Since the repository webapp is installed as the ROOT webapp (so it can have control over the URL space for the purpose of resolving all of its resource instances), it also controls the site's top-level "home" (index) page. The site home page is not part of the API, however, so it does not need to control the content of that page.

In order to give the site administrator control over the content of the site home page, the repo lets you redirect it to any URL, under the control of a configuration property. See eaglei.repository.index.url in the configuration properties.

Services to Support Admin GUI Pages

These REST services were included to support the administrative GUI, which is implemented as JSPs. They are targeted precisely at serving as the back end of the JSPs, so their range of function is strictly limited and they are effectively "write-only". Typically you call on the service and the HTTP status says whether it succeeded or failed; it does not return a detailed report.

...

The services follow. Red marks required args.

Update User Account

/repository/admin/updateUser

Method: POST
Args:
username=text---account name to change, only not required for create
only_password=(boolean)---do not process changes to user metadata, just update password
password---new password
password_confirm---must match new password
old_password---current (old) password, required for non-admin user
first---first name of person (metadata)
last---last name of person (metadata)
mailbox---email address of person
role---URI of role to add to user, may be multiple; when specified, this list of roles replaces the user's current roles.

...

• Cannot create new users
• Can only change your own user account
• Must supply old password to change password
• Cannot change roles

Update Role

/repository/admin/updateRole

Method: POST
Args:
action= create | update | delete---(keyword)
uri=URI---required except for create, identifies the role
label= text---immediate label of role
comment= text---lenghty description of the role

Update Grants

/repository/admin/updateGrants

Method: POST
Args:
action= add | remove---(keyword)
uri = URI---subject of the access grant
access= URI---access type to be added or removed
agent = URI of person or Role in the grant

Update Workflow Transition

/repository/admin/updateTransition

Method: POST
Args:
action= create | update | delete}}---(keyword)
uri = URI---required except for create, identifies the role role
uri=URI--subject to release
{{ label=text}}---required to create
comment=text (---description)
initial= URI of initial state (required to create)
final=URI of final state (required to create)
tAction=java class of transition action provider
tActionParameter* = text or URI param to pass to action method
order= sorting key, usually a decimal integer

Update Named Graph

/repository/admin/updateNamedGraph

Method: POST
Args:
uri*=URI names the graph or context
label=text
type=URI of a NamedGraph Type

Interactive Web UI Pages

Although the repository is mostly accessed by other programs through  the HTTP REST API, there are a few pages available to interactive users.

HTML Dissemination View of a Resource Instance

The dissemination servlet that makes resource instances resolvable also generates an HTML page for interactive users when content negotiation selects the text/html format.  The HTML renders properties and the values of object properties as links to their URIs, so they can be resolved easily. 

...

See the eaglei.repository.instance.xslt configuration property for instructions on adding an XSLT transformation stage into the pipeline that delivers HTML content. (Sorry, we haven't washed away all of the Cocoon kool-aid yet..)

Administrative UI pages (/admin)

Visit this page for a directory of all administrative UI functions.  Users without superuser privilege will only see a partial list.  It also displays server-status and configuration details.

 http://_hostname_/repository/admin 

Although almost all repository administration is done by manipulating RDF statements in an internal administrative-metadata graph, there is a need for a Web-based UI to make it easier for site administrators to see and change the current state. Except as noted, these pages are only available to users with the Administrator role:

• Interactive SPARQL query workbench (all auththenticated users).
• List Named Graphs with the option to edit their descriptive metadata and ACLs.
  • Note that new Named Graphs are only added with the /graph service.
• List, add, delete, and modify access Roles - edit the metadata of each Role.
• List, add, and modify User login accounts:
  • Display and change membership in roles.
  • Edit user metadata and change password.
  • Disable a user from logging in at all, or reinstate disabled user.
• List, add, delete, and modify Workflow Transitions.
• Report versions of Data Model Ontology: loaded, and available.
• Get a list of Published Resources. This is handy for troubleshooting (all users).

Search Engine Control

• {{/robots.txt}---direct robots to a site map, warn them away from other pages.
• site maps---HTML and XML site maps to direct crawlers to pages we want them to index.