Morpho: Issueshttps://projects.ecoinformatics.org/ecoinfo/https://projects.ecoinformatics.org/ecoinfo/ecoinfo/favicon.ico?14691340362007-11-06T23:42:02ZEcoinformatics Redmine
Redmine Bug #2998 (Resolved): Morpho uses 'allowFirst' by default when building access controlhttps://projects.ecoinformatics.org/ecoinfo/issues/29982007-11-06T23:42:02ZChris Jonescjones@nceas.ucsb.edu
<p>When inserting access control sections into EML documents created by Morpho, Morpho defaults to using a permission order of 'allowFirst' rather than 'denyFirst'. This creates an unintended problem when a deny statement is inserted for public access. For instance, the data-managers group below will be denied access after it has been granted access in the allow statement because it, too, belongs to the 'public' (everyone) group.</p>
<p><access authSystem="ldap://ldap.ecoinformatics.org:389/dc=ecoinformatics,dc=org" <br /> order="allowFirst"><br /> <deny><br /> <principal>public</principal><br /> <permission>read</permission><br /> </deny><br /> <allow><br /> <principal>cn=data-managers,o=PISCO,dc=ecoinformatics,dc=org</principal><br /> <permission>all</permission><br /> </allow><br /></access></p>
<p>If the above had a permission order set to 'denyFirst', public read access would be denied, and then the allow statement would 'punch through' and provide all access to the data-managers group (which is the intention of this access statement).</p>
<p>Morpho should be configured to default to 'denyFirst' when creating acls to avoid this situation.</p> Bug #1539 (Resolved): <keywordSet> is transformed incorrectly from EML2b6 to EML2https://projects.ecoinformatics.org/ecoinfo/issues/15392004-04-28T17:48:31ZChris Jonescjones@nceas.ucsb.edu
<p>The tansformation stylesheets from EML2b6 to EML2 incorrectly handle keywords.</p>
<p>An example:</p>
<p>In EML2b6 we have:</p>
<pre><code>&lt;keywordSet&gt;<br /> &lt;keyword&gt;giant kelp&lt;/keyword&gt;<br /> &lt;keyword&gt;biomass&lt;/keyword&gt;<br /> &lt;/keywordSet&gt;<br /> &lt;keywordSet&gt;<br /> &lt;keyword&gt;Historical_kelp&lt;/keyword&gt;<br /> &lt;keywordThesaurus&gt;SBCLTERCategories&lt;/keywordThesaurus&gt;<br /> &lt;/keywordSet&gt;</code></pre>
<p>transformed to EML2 we get:</p>
<pre><code>&lt;keywordSet&gt;<br /> &lt;keyword keywordType="theme"&gt;giant kelp&lt;/keyword&gt;<br /> &lt;keyword keywordType="theme"&gt;biomass&lt;/keyword&gt;<br /> &lt;keyword keywordType="theme"&gt;Historical_kelp&lt;/keyword&gt;<br /> &lt;keywordThesaurus&gt;SBCLTERCategories&lt;/keywordThesaurus&gt;<br /> &lt;/keywordSet&gt;</code></pre>
<p>The keywordSet's are combined, which incorrectly associates freetext keywords<br />with a thesaurus.</p> Bug #1537 (Resolved): whitespace causes incorrect XSL conversion of eml2beta6 to eml2 in <unit>https://projects.ecoinformatics.org/ecoinfo/issues/15372004-04-28T03:37:13ZChris Jonescjones@nceas.ucsb.edu
<p>The attribute dialogs in Morpho 1.4 allowed a researcher to enter a freetext<br />'unit' for typing a specific data column for tabular data. Morpho 1.4 seems to<br />introduce a leading whitespace in the value of the unit tag like:</p>
<p><unit> ton</unit></p>
<p>On XSL transformation using $MORPHO_HOME/xsl/eml2attr.xsl, the unit value is<br />compared to the standard unit dictionary that ships with EML2.0.0. If the<br />standard unit name is found, a <standardUnit> tag is created with the previous<br />value. If not, the default is to produce a <customUnit> tag.</p>
<p>Because of the introduced leading whitespace, ' ton' != 'ton', and therefore an<br />orphaned <customUnit> ton</customUnit> tag is generated, with no reference to<br />it's STMML unit definition.</p> Bug #461 (Resolved): Data Viewer window renders tabb seperated files incorrectlyhttps://projects.ecoinformatics.org/ecoinfo/issues/4612002-04-07T10:05:22ZChris Jonescjones@nceas.ucsb.edu
<p>When you create a new data package that includes a data file that is TAB<br />delimited, the ascii import works well, and the copy of the data file sitting in<br />the user's local profile data store is identical to the original. However, when<br />it's viewed by pressing the 'View Data' button within the editor, the<br />spreadsheet viewer incorrectly renders lines with missing values. It actually<br />just shifts cell values to the left if there are missing values to the left:</p>
<p>a <TAB> 1 <TAB> x<br />b <TAB> 2 <TAB> y<br />c <TAB> <TAB> z<br />A <TAB> 4 <TAB> X<br /> <TAB> 5 <TAB> Y<br />C <TAB> 6 <TAB> Z</p>
<p>gets rendered as:</p>
<p>a 1 x<br />b 2 y<br />c z<br />A 4 X<br />5 Y<br />C 6 Z</p> Bug #410 (Resolved): need to use more descriptive relationships in building triple statementshttps://projects.ecoinformatics.org/ecoinfo/issues/4102002-02-02T09:35:08ZChris Jonescjones@nceas.ucsb.edu
<p>Morpho and other clients that build triple statements in eml-dataset docs <br />should use more descriptive relationship strings. They should also come from <br />an standard list. Perhaps:</p>
<p>is data entity in<br />provides access control rules for<br />provides data entity information for<br />provides project information for<br />provides protocol information for<br />provides attribute information for<br />provides physical file information for</p>
<p>etc. Lastly, some relationships for data files such as: <br />isDataFileFor(C:\Documents and Settings\cjones\DESKTOP\allsp.txt) <br />are being generated by the package wizard. Seems like the path should be nixed <br />from the relationship...</p>