[jira] [Resolved] (JENA-634) Integrate JSON-LD into RIOT
[ https://issues.apache.org/jira/browse/JENA-634?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Andy Seaborne resolved JENA-634. Resolution: Done Note: JSONLD implements simple literals and xsd:strings in RDF 1.1 style - they are the same term. > Integrate JSON-LD into RIOT > --- > > Key: JENA-634 > URL: https://issues.apache.org/jira/browse/JENA-634 > Project: Apache Jena > Issue Type: Improvement > Components: RIOT >Reporter: Andy Seaborne >Assignee: Andy Seaborne > Fix For: Jena 2.11.2 > > > JSON-LD engine used is > [com.github.jsonld-java:jsonld-java|https://github.com/jsonld-java/]. This > is the 'core' module. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Commented] (JENA-633) optimizer sequences instead of joins / subqueries with limits project more results than they should
[
https://issues.apache.org/jira/browse/JENA-633?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13891304#comment-13891304
]
ASF subversion and git services commented on JENA-633:
--
Commit 1564534 from [~andy.seaborne] in branch 'jena/trunk'
[ https://svn.apache.org/r1564534 ]
JENA-633 : JSON-LD Support.
> optimizer sequences instead of joins / subqueries with limits project more
> results than they should
> ---
>
> Key: JENA-633
> URL: https://issues.apache.org/jira/browse/JENA-633
> Project: Apache Jena
> Issue Type: Bug
> Components: ARQ, Jena
>Affects Versions: Jena 2.11.0
>Reporter: Joshua Taylor
> Labels: optimization, sparql
> Attachments: SubqueryLimitBug.java
>
>
> A subquery with `limit 1` specified can `leak` more values to the surrounding
> query. Rob Vesse tracked this down to a difference between the basic algebra
> and the optimized algebra [1] . One query has the following basic and
> optimized Ops; the "join" in the former shouldn't get turned into a
> "sequence" in the latter.
> {noformat}
> Basic:
> (join
> (bgp (triple ?resource
> ))
> (slice _ 1
> (project (?resource ?p)
> (order (?p)
> (bgp (triple ?resource ?p))
> Optimized:
> (sequence
> (bgp (triple ?resource
> ))
> (project (?resource ?p)
> (top (1 ?p)
> (bgp (triple ?resource ?p)
> {noformat}
> [1]
> http://answers.semanticweb.com/questions/26343/outer-query-results-include-more-values-than-subquery-with-limit-n-produces/26344
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
[jira] [Commented] (JENA-494) Variables in SERVICE can be written by scoping transforms and not recovered.
[
https://issues.apache.org/jira/browse/JENA-494?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13890896#comment-13890896
]
Andy Seaborne commented on JENA-494:
Good fix - we should close this.
I wonder if the ideal, longer term solution is to introduce an algebra operator
{{(rename ...)}} as a general capability.
> Variables in SERVICE can be written by scoping transforms and not recovered.
>
>
> Key: JENA-494
> URL: https://issues.apache.org/jira/browse/JENA-494
> Project: Apache Jena
> Issue Type: Bug
> Components: ARQ
>Affects Versions: Jena 2.10.1
>Reporter: Andy Seaborne
>Assignee: Rob Vesse
> Fix For: Jena 2.11.2
>
>
> {noformat}
> SELECT ?z
> {
>SELECT ?z {
> SERVICE { ?c ?p ?z }
> ?c ?q ?z
>}
> }
> {noformat}
> ==>
> {noformat}
> (project (?z)
> (project (?z)
> (sequence
> (service
> (bgp (triple ?/c ?/p ?z)))
> (bgp (triple ?/c ?/q ?z)
> {noformat}
> but the remote request is
> {noformat}
> SELECT * { ?c ?p ?o }
> {noformat}
> so the {{?/c}} and {{?c}} do not join. Note that {{?z}} is not renamed
> because it emerges from the inner projection.
> From:
> http://mail-archives.apache.org/mod_mbox/jena-users/201307.mbox/%3C20130717190649.GA25207%40netestate.de%3E
> Renaming is done by TransformScopeRename, looking for projections which are
> not the top level as it walk back up the tree. The renaming is done by
> Rename.renameVars.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
[jira] [Commented] (JENA-636) No configuration option for setting output encoding in schemagen
[ https://issues.apache.org/jira/browse/JENA-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13890792#comment-13890792 ] Ian Dickinson commented on JENA-636: I should have been clearer. Yes, it's schemagen not the plugin itself - there's currently no way to set an output encoding other than platform default. The action here is to add a configuration option to schemagen, and ensure that it has an effect on the output stream! > No configuration option for setting output encoding in schemagen > > > Key: JENA-636 > URL: https://issues.apache.org/jira/browse/JENA-636 > Project: Apache Jena > Issue Type: Bug > Components: Cmd line tools >Affects Versions: Jena 2.11.1 >Reporter: Ian Dickinson >Assignee: Ian Dickinson >Priority: Minor > > User reported a problem that Java source files generated by schemagen are > only created with character encoding cp1252, whereas his source files need to > be in UTF-8 to cope with international characters in the ontology content. > At present, there is no configuration option to set the character encoding on > the output file, so this would need to be added to schemagen. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Commented] (JENA-635) Add LSB init daemon support to fuseki startup script
[ https://issues.apache.org/jira/browse/JENA-635?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13890789#comment-13890789 ] Ian Dickinson commented on JENA-635: OK, I have checked and - for me at least - it does run from the command line. If the lsb init functions are available from the usual place it will load them, otherwise it default to using echo. I also tweaked the script so that the start file can be specified in the environment or via /etc/default > Add LSB init daemon support to fuseki startup script > > > Key: JENA-635 > URL: https://issues.apache.org/jira/browse/JENA-635 > Project: Apache Jena > Issue Type: Improvement > Components: Fuseki >Reporter: Ian Dickinson >Assignee: Ian Dickinson >Priority: Minor > Attachments: fuseki-init.d.patch > > > I have made some improvements to the Fuseki startup script, so that it works > more easily as an init.d service script. Specifically, I have: > * allowed default values for variables to be loaded from /etc/default, so > that changing local setup does not require editing the script > * switched to using lsb_daemon_log rather than echo for output > * added fallbacks so that echo will be used if the lsb init functions are not > available > * switched to using 'ps -p' rather than 'kill -0' to find out if a process is > running > * made the location of the PID file more flexible. > This is a call for review of these changes, and in particular test reports > from other environments. I have tested on Ubuntu 12.04LTS, 13.10 and 14.04. I > have not tested on MaxOS, cygwin or other Linux distros, and would welcome > test reports and implementation improvements. > The full revised script is in a gist: > https://gist.github.com/ephemerian/8802368 -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (JENA-635) Add LSB init daemon support to fuseki startup script
[ https://issues.apache.org/jira/browse/JENA-635?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ian Dickinson updated JENA-635: --- Attachment: fuseki-init.d.patch Revised to ensure that it starts from the command line > Add LSB init daemon support to fuseki startup script > > > Key: JENA-635 > URL: https://issues.apache.org/jira/browse/JENA-635 > Project: Apache Jena > Issue Type: Improvement > Components: Fuseki >Reporter: Ian Dickinson >Assignee: Ian Dickinson >Priority: Minor > Attachments: fuseki-init.d.patch > > > I have made some improvements to the Fuseki startup script, so that it works > more easily as an init.d service script. Specifically, I have: > * allowed default values for variables to be loaded from /etc/default, so > that changing local setup does not require editing the script > * switched to using lsb_daemon_log rather than echo for output > * added fallbacks so that echo will be used if the lsb init functions are not > available > * switched to using 'ps -p' rather than 'kill -0' to find out if a process is > running > * made the location of the PID file more flexible. > This is a call for review of these changes, and in particular test reports > from other environments. I have tested on Ubuntu 12.04LTS, 13.10 and 14.04. I > have not tested on MaxOS, cygwin or other Linux distros, and would welcome > test reports and implementation improvements. > The full revised script is in a gist: > https://gist.github.com/ephemerian/8802368 -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (JENA-635) Add LSB init daemon support to fuseki startup script
[ https://issues.apache.org/jira/browse/JENA-635?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ian Dickinson updated JENA-635: --- Attachment: (was: fuseki-init.d.patch) > Add LSB init daemon support to fuseki startup script > > > Key: JENA-635 > URL: https://issues.apache.org/jira/browse/JENA-635 > Project: Apache Jena > Issue Type: Improvement > Components: Fuseki >Reporter: Ian Dickinson >Assignee: Ian Dickinson >Priority: Minor > Attachments: fuseki-init.d.patch > > > I have made some improvements to the Fuseki startup script, so that it works > more easily as an init.d service script. Specifically, I have: > * allowed default values for variables to be loaded from /etc/default, so > that changing local setup does not require editing the script > * switched to using lsb_daemon_log rather than echo for output > * added fallbacks so that echo will be used if the lsb init functions are not > available > * switched to using 'ps -p' rather than 'kill -0' to find out if a process is > running > * made the location of the PID file more flexible. > This is a call for review of these changes, and in particular test reports > from other environments. I have tested on Ubuntu 12.04LTS, 13.10 and 14.04. I > have not tested on MaxOS, cygwin or other Linux distros, and would welcome > test reports and implementation improvements. > The full revised script is in a gist: > https://gist.github.com/ephemerian/8802368 -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Commented] (JENA-636) No configuration option for setting output encoding in schemagen
[
https://issues.apache.org/jira/browse/JENA-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13890764#comment-13890764
]
Andy Seaborne commented on JENA-636:
It may not (just) be the plugin. schemagen.java has:
{noformat}
m_output = new PrintStream( new FileOutputStream( out ) );
{noformat}
which looks suspicious. It opens the PrintStream with the platform default
character encoding.
See also : FileUtils.asPrintWriterUTF8.
> No configuration option for setting output encoding in schemagen
>
>
> Key: JENA-636
> URL: https://issues.apache.org/jira/browse/JENA-636
> Project: Apache Jena
> Issue Type: Bug
> Components: Cmd line tools
>Affects Versions: Jena 2.11.1
>Reporter: Ian Dickinson
>Assignee: Ian Dickinson
>Priority: Minor
>
> User reported a problem that Java source files generated by schemagen are
> only created with character encoding cp1252, whereas his source files need to
> be in UTF-8 to cope with international characters in the ontology content.
> At present, there is no configuration option to set the character encoding on
> the output file, so this would need to be added to schemagen.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
[jira] [Commented] (JENA-634) Integrate JSON-LD into RIOT
[
https://issues.apache.org/jira/browse/JENA-634?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13890739#comment-13890739
]
Andy Seaborne commented on JENA-634:
Migration go code into {{org.apache.jena.riot}} done.
Remaining:
* content negotiation for Fuseki.
* clean up tests
> Integrate JSON-LD into RIOT
> ---
>
> Key: JENA-634
> URL: https://issues.apache.org/jira/browse/JENA-634
> Project: Apache Jena
> Issue Type: Improvement
> Components: RIOT
>Reporter: Andy Seaborne
>Assignee: Andy Seaborne
> Fix For: Jena 2.11.2
>
>
> JSON-LD engine used is
> [com.github.jsonld-java:jsonld-java|https://github.com/jsonld-java/]. This
> is the 'core' module.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
[jira] [Commented] (JENA-635) Add LSB init daemon support to fuseki startup script
[ https://issues.apache.org/jira/browse/JENA-635?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13890716#comment-13890716 ] Ian Dickinson commented on JENA-635: Good thought, I'll investigate. > Add LSB init daemon support to fuseki startup script > > > Key: JENA-635 > URL: https://issues.apache.org/jira/browse/JENA-635 > Project: Apache Jena > Issue Type: Improvement > Components: Fuseki >Reporter: Ian Dickinson >Assignee: Ian Dickinson >Priority: Minor > Attachments: fuseki-init.d.patch > > > I have made some improvements to the Fuseki startup script, so that it works > more easily as an init.d service script. Specifically, I have: > * allowed default values for variables to be loaded from /etc/default, so > that changing local setup does not require editing the script > * switched to using lsb_daemon_log rather than echo for output > * added fallbacks so that echo will be used if the lsb init functions are not > available > * switched to using 'ps -p' rather than 'kill -0' to find out if a process is > running > * made the location of the PID file more flexible. > This is a call for review of these changes, and in particular test reports > from other environments. I have tested on Ubuntu 12.04LTS, 13.10 and 14.04. I > have not tested on MaxOS, cygwin or other Linux distros, and would welcome > test reports and implementation improvements. > The full revised script is in a gist: > https://gist.github.com/ephemerian/8802368 -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Commented] (JENA-634) Integrate JSON-LD into RIOT
[ https://issues.apache.org/jira/browse/JENA-634?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13890715#comment-13890715 ] ASF subversion and git services commented on JENA-634: -- Commit 1564316 from [~andy.seaborne] in branch 'jena/trunk' [ https://svn.apache.org/r1564316 ] JENA-634. Add JSON-LD readers/writers to RIOT > Integrate JSON-LD into RIOT > --- > > Key: JENA-634 > URL: https://issues.apache.org/jira/browse/JENA-634 > Project: Apache Jena > Issue Type: Improvement > Components: RIOT >Reporter: Andy Seaborne >Assignee: Andy Seaborne > Fix For: Jena 2.11.2 > > > JSON-LD engine used is > [com.github.jsonld-java:jsonld-java|https://github.com/jsonld-java/]. This > is the 'core' module. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
Integrating in JSON-LD : new dependencies
JENA-634 I'm in the process of integrating reader/writer adapters for JSON-LD. This adds readers and writers; the writing is mechanical and it's not (yet) possible to provide a specific context for a particular appearance. The core JSON-LD engine is com.github.jsonld-java:jsonld-java which as the following dependencies: \- com.github.jsonld-java:jsonld-java:jar:0.2 +- com.fasterxml.jackson.core:jackson-core:jar:2.2.1 +- com.fasterxml.jackson.core:jackson-databind:jar:2.2.1 | \- com.fasterxml.jackson.core:jackson-annotations:jar:2.2.1 \- org.apache.httpcomponents:httpclient-cache:jar:4.2.5 \- commons-logging:commons-logging:jar:1.1.1 The Jena adapter only uses the core module - the Jena adapter in their code base is adapted from my github one but it's easier for version tracking reasons to have the adapter in Jena. At the moment httpclient-cache is excluded and v4.2.3 included to be compatible with the rest of Jena. commons-logging is excluded and jcl-over-slf4j used. jsonld-java are discussing going to 0.3 but I don't think it changes the dependencies except that it is jackson 2.3.1. com.fasterxml.jackson.* is licensed under Apache License 2.0. No changed to N&L for binary aggregations (apache-jena, jena-fuseki) are necessary. The apache-jena binaries increase by about 1.2/1.3M. Still to do: HTTP Content Negotiation in Fuseki. Andy
[jira] [Commented] (JENA-635) Add LSB init daemon support to fuseki startup script
[ https://issues.apache.org/jira/browse/JENA-635?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13890635#comment-13890635 ] Andy Seaborne commented on JENA-635: At the moment, the script works when used from the command line as well as an init.d script. "echo" helps for that use case. Would it be possible to detect and adjust accordingly? > Add LSB init daemon support to fuseki startup script > > > Key: JENA-635 > URL: https://issues.apache.org/jira/browse/JENA-635 > Project: Apache Jena > Issue Type: Improvement > Components: Fuseki >Reporter: Ian Dickinson >Assignee: Ian Dickinson >Priority: Minor > Attachments: fuseki-init.d.patch > > > I have made some improvements to the Fuseki startup script, so that it works > more easily as an init.d service script. Specifically, I have: > * allowed default values for variables to be loaded from /etc/default, so > that changing local setup does not require editing the script > * switched to using lsb_daemon_log rather than echo for output > * added fallbacks so that echo will be used if the lsb init functions are not > available > * switched to using 'ps -p' rather than 'kill -0' to find out if a process is > running > * made the location of the PID file more flexible. > This is a call for review of these changes, and in particular test reports > from other environments. I have tested on Ubuntu 12.04LTS, 13.10 and 14.04. I > have not tested on MaxOS, cygwin or other Linux distros, and would welcome > test reports and implementation improvements. > The full revised script is in a gist: > https://gist.github.com/ephemerian/8802368 -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Created] (JENA-636) No configuration option for setting output encoding in schemagen
Ian Dickinson created JENA-636: -- Summary: No configuration option for setting output encoding in schemagen Key: JENA-636 URL: https://issues.apache.org/jira/browse/JENA-636 Project: Apache Jena Issue Type: Bug Components: Cmd line tools Affects Versions: Jena 2.11.1 Reporter: Ian Dickinson Assignee: Ian Dickinson Priority: Minor User reported a problem that Java source files generated by schemagen are only created with character encoding cp1252, whereas his source files need to be in UTF-8 to cope with international characters in the ontology content. At present, there is no configuration option to set the character encoding on the output file, so this would need to be added to schemagen. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (JENA-635) Add LSB init daemon support to fuseki startup script
[ https://issues.apache.org/jira/browse/JENA-635?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ian Dickinson updated JENA-635: --- Attachment: fuseki-init.d.patch Patch file for changes to fuseki init script > Add LSB init daemon support to fuseki startup script > > > Key: JENA-635 > URL: https://issues.apache.org/jira/browse/JENA-635 > Project: Apache Jena > Issue Type: Improvement > Components: Fuseki >Reporter: Ian Dickinson >Assignee: Ian Dickinson >Priority: Minor > Attachments: fuseki-init.d.patch > > > I have made some improvements to the Fuseki startup script, so that it works > more easily as an init.d service script. Specifically, I have: > * allowed default values for variables to be loaded from /etc/default, so > that changing local setup does not require editing the script > * switched to using lsb_daemon_log rather than echo for output > * added fallbacks so that echo will be used if the lsb init functions are not > available > * switched to using 'ps -p' rather than 'kill -0' to find out if a process is > running > * made the location of the PID file more flexible. > This is a call for review of these changes, and in particular test reports > from other environments. I have tested on Ubuntu 12.04LTS, 13.10 and 14.04. I > have not tested on MaxOS, cygwin or other Linux distros, and would welcome > test reports and implementation improvements. > The full revised script is in a gist: > https://gist.github.com/ephemerian/8802368 -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Updated] (JENA-635) Add LSB init daemon support to fuseki startup script
[ https://issues.apache.org/jira/browse/JENA-635?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ian Dickinson updated JENA-635: --- Assignee: Ian Dickinson > Add LSB init daemon support to fuseki startup script > > > Key: JENA-635 > URL: https://issues.apache.org/jira/browse/JENA-635 > Project: Apache Jena > Issue Type: Improvement > Components: Fuseki >Reporter: Ian Dickinson >Assignee: Ian Dickinson >Priority: Minor > > I have made some improvements to the Fuseki startup script, so that it works > more easily as an init.d service script. Specifically, I have: > * allowed default values for variables to be loaded from /etc/default, so > that changing local setup does not require editing the script > * switched to using lsb_daemon_log rather than echo for output > * added fallbacks so that echo will be used if the lsb init functions are not > available > * switched to using 'ps -p' rather than 'kill -0' to find out if a process is > running > * made the location of the PID file more flexible. > This is a call for review of these changes, and in particular test reports > from other environments. I have tested on Ubuntu 12.04LTS, 13.10 and 14.04. I > have not tested on MaxOS, cygwin or other Linux distros, and would welcome > test reports and implementation improvements. > The full revised script is in a gist: > https://gist.github.com/ephemerian/8802368 -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Created] (JENA-635) Add LSB init daemon support to fuseki startup script
Ian Dickinson created JENA-635: -- Summary: Add LSB init daemon support to fuseki startup script Key: JENA-635 URL: https://issues.apache.org/jira/browse/JENA-635 Project: Apache Jena Issue Type: Improvement Components: Fuseki Reporter: Ian Dickinson Priority: Minor I have made some improvements to the Fuseki startup script, so that it works more easily as an init.d service script. Specifically, I have: * allowed default values for variables to be loaded from /etc/default, so that changing local setup does not require editing the script * switched to using lsb_daemon_log rather than echo for output * added fallbacks so that echo will be used if the lsb init functions are not available * switched to using 'ps -p' rather than 'kill -0' to find out if a process is running * made the location of the PID file more flexible. This is a call for review of these changes, and in particular test reports from other environments. I have tested on Ubuntu 12.04LTS, 13.10 and 14.04. I have not tested on MaxOS, cygwin or other Linux distros, and would welcome test reports and implementation improvements. The full revised script is in a gist: https://gist.github.com/ephemerian/8802368 -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Created] (JENA-634) Integrate JSON-LD into RIOT
Andy Seaborne created JENA-634: -- Summary: Integrate JSON-LD into RIOT Key: JENA-634 URL: https://issues.apache.org/jira/browse/JENA-634 Project: Apache Jena Issue Type: Improvement Components: RIOT Reporter: Andy Seaborne Assignee: Andy Seaborne Fix For: Jena 2.11.2 JSON-LD engine used is [com.github.jsonld-java:jsonld-java|https://github.com/jsonld-java/]. This is the 'core' module. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
