Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 22:56, Marshall Schor wrote: > > I guess we've always (probably incorrectly) thought that those use-cases which > wanted unique, fixed identifiers for every feature structure would just add > them > as a feature, with whatever kind of id was wanted... I didn't want to do that

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
I guess we've always (probably incorrectly) thought that those use-cases which wanted unique, fixed identifiers for every feature structure would just add them as a feature, with whatever kind of id was wanted... -Marshall On 8/11/2016 4:44 PM, Richard Eckart de Castilho wrote: > On 11.08.2016, a

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 22:37, Marshall Schor wrote: > > I had the thought that perhaps it would be quite easy to add Type System > (only) > serialization directly into Form 6 using all of its fancy compression > techniques. Only those types used would need to be serialized. This would > make > it v

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 22:29, Marshall Schor wrote: > > Yes, it seems to me also, that reinitializing a CAS's type system and index > definition is unusual, and sort of major surgery. It's not clear to me what > happens if this CAS is part of a CAS pool - I suspect things could easily > break > down

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
I had the thought that perhaps it would be quite easy to add Type System (only) serialization directly into Form 6 using all of its fancy compression techniques. Only those types used would need to be serialized. This would make it very convenient and efficient to store the type system (which woul

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
Yes, it seems to me also, that reinitializing a CAS's type system and index definition is unusual, and sort of major surgery. It's not clear to me what happens if this CAS is part of a CAS pool - I suspect things could easily break down because there's an assumption made that all CASes in a CAS po

[jira] [Created] (UIMA-5054) JCas returning generic class instead of JCas cover class

2016-08-11 Thread Richard Eckart de Castilho (JIRA)
Richard Eckart de Castilho created UIMA-5054: Summary: JCas returning generic class instead of JCas cover class Key: UIMA-5054 URL: https://issues.apache.org/jira/browse/UIMA-5054 Project:

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 22:04, Marshall Schor wrote: > > What should a load call using auto-detection, detecting a XMI or XCAS input, > do > if the tsiInputStream is not null? > > Current impl: tsiInputStream is ignored for XMI and XCAS. > > More consistent Impl might be to use read the TSI info an

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
What should a load call using auto-detection, detecting a XMI or XCAS input, do if the tsiInputStream is not null? Current impl: tsiInputStream is ignored for XMI and XCAS. More consistent Impl might be to use read the TSI info and use it to set up the CAS's type system and index specs. What do

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 21:53, Richard Eckart de Castilho wrote: > > On 11.08.2016, at 21:34, Marshall Schor wrote: >> >> load ( InputStream casInputStream, TypeSystem compressedForm6originalTS) ? >> >> Form 6 is always "lenient", if it is passed an original (meaning the one that >> corresponds to t

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
OK, sorry I missed that other case. Thanks for the feedback. -Marshall On 8/11/2016 3:53 PM, Richard Eckart de Castilho wrote: > On 11.08.2016, at 21:34, Marshall Schor wrote: >> I see - you're using the TS part of the TSI to do lenient loading for >> Compressed >> Form 6. >> >> What about an A

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 21:34, Marshall Schor wrote: > > I see - you're using the TS part of the TSI to do lenient loading for > Compressed > Form 6. > > What about an API which expresses this more directly? > > E.g. > > load ( InputStream casInputStream, TypeSystem compressedForm6originalTS) ? >

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
I see - you're using the TS part of the TSI to do lenient loading for Compressed Form 6. What about an API which expresses this more directly? E.g. load ( InputStream casInputStream, TypeSystem compressedForm6originalTS) ? Form 6 is always "lenient", if it is passed an original (meaning the one

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
> On 11.08.2016, at 21:17, Marshall Schor wrote: > > up the CAS, once (build/commit the type system, create the index repositories, > etc.), and then load loop would run potentially significantly faster. You assume that I use the TSI the reinit the CAS - but that is not what I do. I don't touch

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
Re: skipping re-reading the separate TSI file info: OK, makes sense. One thing I wonder, in terms of "useless work avoiding": If you're in a loop like this: casMgr = read(casMgrFile); for (file in directory) { load(file, casMgr, CAS, boolean) } it seems it would make more sense to rea

[jira] [Work started] (UIMA-5053) DUCC ducc_watcher optional admin script to determine status and send notifications

2016-08-11 Thread Lou DeGenaro (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5053?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Work on UIMA-5053 started by Lou DeGenaro. -- > DUCC ducc_watcher optional admin script to determine status and send > notification

[jira] [Created] (UIMA-5053) DUCC ducc_watcher optional admin script to determine status and send notifications

2016-08-11 Thread Lou DeGenaro (JIRA)
Lou DeGenaro created UIMA-5053: -- Summary: DUCC ducc_watcher optional admin script to determine status and send notifications Key: UIMA-5053 URL: https://issues.apache.org/jira/browse/UIMA-5053 Project: U

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 19:43, Marshall Schor wrote: > > I'm working on this now. > > I note that the new load(InputStream, CasMgrSerialzer, CAS, boolean) method is > "public". Is there some code (perhaps in DkPro) that needs this form? > > If not, I'll remove this method and make the reading to cre

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
I'm working on this now. I note that the new load(InputStream, CasMgrSerialzer, CAS, boolean) method is "public". Is there some code (perhaps in DkPro) that needs this form? If not, I'll remove this method and make the reading to create the CasMgrSerializer "lzay" - not done until needed. Not

Build failed in Jenkins: UIMAJ-SDK_java8 #336

2016-08-11 Thread Apache Jenkins Server
See Changes: [rec] [UIMA-4685] Implement lenient loading for COMPRESSED_FILTERED_TSI and add tests for lenient loading of XMI, XCAS, and COMPRESSED_FILTERED_TSI. [rec] [UIMA-4685] Adding back COMPRESSED_FILTERED_TSI, refactoring to all

Build failed in Jenkins: UIMAJ-SDK_java8 » Apache UIMA Java SDK: uimaj #336

2016-08-11 Thread Apache Jenkins Server
See -- [INFO] [INFO] [INFO] Building

Creating UIMA-AS deployment descriptors programmatically

2016-08-11 Thread Richard Eckart de Castilho
On 16.07.2015, at 23:10, Jaroslaw Cwiklik wrote: > > The UIMA-AS *does* have an API to generate deployment descriptors although > its not documented. Its an internal API for now and most likely will be > documented in the next release of UIMA-AS. The API is implemented by > DeploymentDescriptorFa

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 18:05, Marshall Schor wrote: > re: CASImpl reinit change, reading the additional passed-in stream for the > TSI info: > > If both this stream is non-null and the header says the primary stream already > has the TSI info, the primary stream's info is used. > > The code was chan

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
re: CASImpl reinit change, reading the additional passed-in stream for the TSI info: If both this stream is non-null and the header says the primary stream already has the TSI info, the primary stream's info is used. The code was changed to always read the additional passed-in stream for the TSI

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
OK, I've looked thru the rest of the changes and think I understand them :-). I'll take a pass at making these minor fixups and documenting things. -Marshall On 8/11/2016 10:21 AM, Richard Eckart de Castilho wrote: > On 11.08.2016, at 16:08, Marshall Schor wrote: >> Re: BinaryCasSerDes6 changes

Jenkins build is back to normal : UIMA-DUCC #970

2016-08-11 Thread Apache Jenkins Server
See

Jenkins build is back to normal : UIMA-DUCC » Apache UIMA DUCC: uima-ducc-orchestrator #970

2016-08-11 Thread Apache Jenkins Server
See

[jira] [Closed] (UIMA-5048) DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests

2016-08-11 Thread Jerry Cwiklik (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5048?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jerry Cwiklik closed UIMA-5048. --- Resolution: Fixed Modified PM to ignore tasks for which there is no command line (jobs, Services). Thi

[jira] [Commented] (UIMA-5052) DUCC check_ducc should complain if ssh is not viable

2016-08-11 Thread Lou DeGenaro (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5052?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15417351#comment-15417351 ] Lou DeGenaro commented on UIMA-5052: > add ssh_operational T/F test to ducc_util.py scr

[jira] [Work started] (UIMA-5052) DUCC check_ducc should complain if ssh is not viable

2016-08-11 Thread Lou DeGenaro (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5052?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Work on UIMA-5052 started by Lou DeGenaro. -- > DUCC check_ducc should complain if ssh is not viable > -

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 16:08, Marshall Schor wrote: > > Re: BinaryCasSerDes6 changes > > BinaryCasSerDes6 has many public constructors, which are variants but end up > calling one common constructor. > This update changes the the arg list for the common constructor, which means > any > callers of th

[jira] [Created] (UIMA-5052) DUCC check_ducc should complain if ssh is not viable

2016-08-11 Thread Lou DeGenaro (JIRA)
Lou DeGenaro created UIMA-5052: -- Summary: DUCC check_ducc should complain if ssh is not viable Key: UIMA-5052 URL: https://issues.apache.org/jira/browse/UIMA-5052 Project: UIMA Issue Type: Bug

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 16:08, Richard Eckart de Castilho wrote: > > I was thinking that some of the method signatures might > profit from some argument reordering, i.e. moving > common parameters more to the front (e.g. format in save > and CAS/leniently in load)... The argument "tgtTs" in BinaryCas

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
Re: BinaryCasSerDes6 changes BinaryCasSerDes6 has many public constructors, which are variants but end up calling one common constructor. This update changes the the arg list for the common constructor, which means any callers of this need updating. Since this is a public API, it's possible that

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 15:50, Marshall Schor wrote: > > nice! And thanks for finding the other missing parts / bugs in my attempt! > > Are you still working on this (e.g. I noticed some Javadocs haven't been > updated). > > I'm happy to help out with doc updates (tutorials guide and even Javadocs

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Marshall Schor
nice! And thanks for finding the other missing parts / bugs in my attempt! Are you still working on this (e.g. I noticed some Javadocs haven't been updated). I'm happy to help out with doc updates (tutorials guide and even Javadocs) - let me know. -Marshall On 8/11/2016 9:34 AM, Richard Ecka

[jira] [Commented] (UIMA-5049) uima-ducc-demo adjustments

2016-08-11 Thread Lou DeGenaro (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15417272#comment-15417272 ] Lou DeGenaro commented on UIMA-5049: send submitter output to ~duccy/spool > uima-ducc

[jira] [Updated] (UIMA-5049) uima-ducc-demo adjustments

2016-08-11 Thread Lou DeGenaro (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5049?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lou DeGenaro updated UIMA-5049: --- Summary: uima-ducc-demo adjustments (was: uima-ducc-vm adjustments) > uima-ducc-demo adjustments > ---

[jira] [Created] (UIMA-5051) Deprecate uimaFIT CasIOUtil in favor of uimaj-core CasIOUtils

2016-08-11 Thread Richard Eckart de Castilho (JIRA)
Richard Eckart de Castilho created UIMA-5051: Summary: Deprecate uimaFIT CasIOUtil in favor of uimaj-core CasIOUtils Key: UIMA-5051 URL: https://issues.apache.org/jira/browse/UIMA-5051 Pro

[jira] [Commented] (UIMA-5043) Provide method to access individual external override settings

2016-08-11 Thread Richard Eckart de Castilho (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5043?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15417256#comment-15417256 ] Richard Eckart de Castilho commented on UIMA-5043: -- A read-only interface

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 11.08.2016, at 13:06, Richard Eckart de Castilho wrote: > The "leniently" flag is currently only used for XMI/XCAS... I think this flag > should also be > used for COMPRESSED_FILTERED_TSI and when enabled in other context should > throw an exception? Partially addressed. The flag is now also

[jira] [Created] (UIMA-5050) IndexCorruptionReportingTest fails in Eclipse

2016-08-11 Thread Richard Eckart de Castilho (JIRA)
Richard Eckart de Castilho created UIMA-5050: Summary: IndexCorruptionReportingTest fails in Eclipse Key: UIMA-5050 URL: https://issues.apache.org/jira/browse/UIMA-5050 Project: UIMA

[jira] [Commented] (UIMA-5043) Provide method to access individual external override settings

2016-08-11 Thread Burn Lewis (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5043?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15417182#comment-15417182 ] Burn Lewis commented on UIMA-5043: -- I've tried putting the functions on UIMAFramework (sin

Jenkins build is back to normal : UIMA-DUCC » Apache UIMA DUCC: uima-ducc-parent #969

2016-08-11 Thread Apache Jenkins Server
See

Build failed in Jenkins: UIMA-DUCC #969

2016-08-11 Thread Apache Jenkins Server
See Changes: [degenaro] UIMA-5048 DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests [degenaro] UIMA-5048 DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests

Build failed in Jenkins: UIMA-DUCC » Apache UIMA DUCC: uima-ducc-orchestrator #969

2016-08-11 Thread Apache Jenkins Server
See Changes: [degenaro] UIMA-5048 DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests -- [INFO]

[jira] [Work started] (UIMA-5049) uima-ducc-vm adjustments

2016-08-11 Thread Lou DeGenaro (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5049?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Work on UIMA-5049 started by Lou DeGenaro. -- > uima-ducc-vm adjustments > > > Key: UIMA-50

[jira] [Created] (UIMA-5049) uima-ducc-vm adjustments

2016-08-11 Thread Lou DeGenaro (JIRA)
Lou DeGenaro created UIMA-5049: -- Summary: uima-ducc-vm adjustments Key: UIMA-5049 URL: https://issues.apache.org/jira/browse/UIMA-5049 Project: UIMA Issue Type: Bug Components: DUCC

[jira] [Assigned] (UIMA-5048) DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests

2016-08-11 Thread Lou DeGenaro (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5048?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lou DeGenaro reassigned UIMA-5048: -- Assignee: Jerry Cwiklik (was: Lou DeGenaro) > DUCC Orchestrator (OR) record Process Manager (PM

[jira] [Updated] (UIMA-5048) DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests

2016-08-11 Thread Lou DeGenaro (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5048?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lou DeGenaro updated UIMA-5048: --- Description: On uima-ducc-demo we saw one Job that caused PM to OOM. According to the PM log, the req

[jira] [Work started] (UIMA-5048) DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests

2016-08-11 Thread Lou DeGenaro (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5048?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Work on UIMA-5048 started by Lou DeGenaro. -- > DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests > --

[jira] [Created] (UIMA-5048) DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests

2016-08-11 Thread Lou DeGenaro (JIRA)
Lou DeGenaro created UIMA-5048: -- Summary: DUCC Orchestrator (OR) record Process Manager (PM) Job CommandLine requests Key: UIMA-5048 URL: https://issues.apache.org/jira/browse/UIMA-5048 Project: UIMA

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
> On 11.08.2016, at 10:08, Richard Eckart de Castilho wrote: > > On 10.08.2016, at 23:29, Richard Eckart de Castilho wrote: >> So I had a first look at whether I can replace parts of the DKPro Core >> BinaryCasReader/ >> BinaryCasWriter code with calls to CasIoUtils, but so far that doesn't se

Re: Any more fixes / commits needed for UIMA SDK 2.9.0?

2016-08-11 Thread Richard Eckart de Castilho
On 10.08.2016, at 23:37, Richard Eckart de Castilho wrote: > > On 08.08.2016, at 18:33, Marshall Schor wrote: >> >> On 8/8/2016 3:49 AM, Richard Eckart de Castilho wrote: >>> Has the loading of index information been addressed yet? >> Yes, the CasIOUtils methods now forward to the base implemen

Re: CasIOUtils class - some meta-questions

2016-08-11 Thread Richard Eckart de Castilho
On 10.08.2016, at 23:29, Richard Eckart de Castilho wrote: > > On 04.08.2016, at 11:30, Peter Klügl wrote: >> >> Am 04.08.2016 um 11:24 schrieb Richard Eckart de Castilho: >>> I think todo would be ok. >>> >>> Btw. I plan to check if/how the DKPro Core BinaryCasReader/Writer can be >>> refact

[jira] [Commented] (UIMA-5043) Provide method to access individual external override settings

2016-08-11 Thread Richard Eckart de Castilho (JIRA)
[ https://issues.apache.org/jira/browse/UIMA-5043?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15416697#comment-15416697 ] Richard Eckart de Castilho commented on UIMA-5043: -- Well, there are differ