Morpho: Issueshttps://projects.ecoinformatics.org/ecoinfo/https://projects.ecoinformatics.org/ecoinfo/ecoinfo/favicon.ico?14691340362014-04-05T00:52:36ZEcoinformatics Redmine
Redmine Bug #6500 (New): Give users ability to import certifates into Morpho easilyhttps://projects.ecoinformatics.org/ecoinfo/issues/65002014-04-05T00:52:36ZJing Taotao@nceas.ucsb.edu
<p>We may create a ~/.morpho/certs directory and have a simple UI for users to drop certificates to the directory.</p>
<p>On startup, morpho looks in that dir and adds any CA certs it finds to the keystore, and keeps track of them, so if the user removes one then it removes it from the truststore on next restart.</p>
<p>The d1_libclient already has some features. The d1_libclient does this all dynamically at runtime - it does not actually write entries to the trustore, just merges them for runtime use</p> Story #6084 (New): Use SOLR query instead of pathqueryhttps://projects.ecoinformatics.org/ecoinfo/issues/60842013-09-06T20:23:20Zben leinfelderleinfelder@nceas.ucsb.eduBug #5858 (Resolved): Does the system metadata need the obsoletedBy element?https://projects.ecoinformatics.org/ecoinfo/issues/58582013-02-12T05:36:53ZJing Taotao@nceas.ucsb.edu
<p>I created data package foo.1 and saved it locally. Then I modified the data package and saved it foo.2. The system metadata of foo.1 should have something like:<br /> <obsoletedBy>foo.2</obsoletedBy><br />But, the system metadata doesn't have the obsoletedBy element.</p>
<p>And the system metadata of foo.2 it does have the obsoletes element.</p> Bug #5848 (New): Ensure Morpho can edit EML contributed by DataUphttps://projects.ecoinformatics.org/ecoinfo/issues/58482013-02-06T01:31:26Zben leinfelderleinfelder@nceas.ucsb.edu
<p>Tried to open network EML 2.1.1 contributed by DataUp tool:<br /><a class="external" href="https://oneshare.unm.edu/knb/d1/mn/object/ark:/90135/q13j39xf/4/mrt-eml.xml">https://oneshare.unm.edu/knb/d1/mn/object/ark:/90135/q13j39xf/4/mrt-eml.xml</a><br />Got a few errors while it tried to parse the package (butterfly still flapping):<br />physicalList is null!<br />physicalList is null!<br />physicalList is null!<br />physicalList is null!<br />Invalid Id</p> Bug #5826 (Resolved): Show identifier conflict prompt when conflict is detectedhttps://projects.ecoinformatics.org/ecoinfo/issues/58262013-01-28T18:00:20Zben leinfelderleinfelder@nceas.ucsb.edu
<p>I removed this window from the save workflow because it seemed like there was no logical question to ask the user - if you are saving the datapackage it is a new revision, not a new datapackage (you can use the "duplicate" feature if you want a new one). We did away with the notion of the identifiers encoding the revision history, so there's only one thing to do: generate a new identifier for the new revision and use it. The only thing I think we could do is ask what <em>kind</em> of identifier we want to generate (UUID, DOI, ARK...).</p> Bug #5823 (New): Improve save-to-network performancehttps://projects.ecoinformatics.org/ecoinfo/issues/58232013-01-25T07:37:21Zben leinfelderleinfelder@nceas.ucsb.edu
<p>Jing saved data packages with binary data entities from Morpho to a Metacat MN to test the performance of those save operations. There seems to be some overhead for this no matter how small the data file.</p>
<p>------------------------------------------<br />File size Total time (s)<br />1.4 K 10<br />395 k 10<br />1.3 M 15<br />5.8 M 26<br />8.8 M 33<br />50 M 115<br />-------------------------------------------</p> Bug #5766 (New): Include data identifier in EML (rather than ecogrid:// URL)https://projects.ecoinformatics.org/ecoinfo/issues/57662012-12-12T00:07:13Zben leinfelderleinfelder@nceas.ucsb.edu
<p>We use ecogrid:// URLs in the EML distribution section for data files. But we should also be writing the plain identifier into a standard place in the EML so that we can match the data Identifier from the ORE to the EML</p> Bug #5757 (New): Add identifier scheme drop down to id conflict windowhttps://projects.ecoinformatics.org/ecoinfo/issues/57572012-12-03T22:14:44Zben leinfelderleinfelder@nceas.ucsb.edu
<p>In cases where we want to generate an identifier at the MN, we need to be able to tell the MN what type of identifier should be created. Usually we will target DOI, but only a handful of MNs will support this so we should have multiple options as well as a sensible default. Note that the scheme parameter for MN.generateIdentifier() is not optional.</p> Bug #5747 (New): Open ORE documents locallyhttps://projects.ecoinformatics.org/ecoinfo/issues/57472012-11-23T21:35:14Zben leinfelderleinfelder@nceas.ucsb.edu
<p>We should be able to open ORE documents as a complete data package. This requires that the UI move toward ORE ids being the primary identifiers for packages, of course.</p> Bug #5743 (New): Include ORE identifier in search resultshttps://projects.ecoinformatics.org/ecoinfo/issues/57432012-11-20T20:24:31Zben leinfelderleinfelder@nceas.ucsb.edu
<p>If we are truly moving toward an ORE-based packaging approach, our primary object (and identifier) should be the resource map, not the science metadata (EML). But our current infrastructure deals primarily with EML identifiers when handling packages.</p> Bug #5735 (New): Decide on primary package identifierhttps://projects.ecoinformatics.org/ecoinfo/issues/57352012-11-19T23:18:52Zben leinfelderleinfelder@nceas.ucsb.edu
<p>Morpho 1.x uses the EML datapackage ID for displaying search result records, opening, saving, etc.<br />With Morpho 2.x incorporating the DataONE DataPackage (ORE) model, we have the opportunity to switch to the ORE object identifier as the primary mechanism for identifying the overall package. This would be more consistent with the model, but requires that EVERY EML file have an ORE that packages it. <br />It's also unclear how you could open an older EML document unless we supported both. Maybe we can allow opening older EML packages, but only allow you to save ORE packages?</p>