EML: Issueshttps://projects.ecoinformatics.org/ecoinfo/https://projects.ecoinformatics.org/ecoinfo/ecoinfo/favicon.ico?14691340362011-06-27T18:47:45ZEcoinformatics Redmine
Redmine Bug #5431 (Resolved): Data Manager throws exception for otherEntity when attributeList is optiona...https://projects.ecoinformatics.org/ecoinfo/issues/54312011-06-27T18:47:45ZDuane Costadcosta@lternet.edu
<p>The EML schema (2.x.y) specifies that all data entities be classified as one of six different DatasetType types:</p>
<p>dataTable<br />spatialRaster<br />spatialVector<br />storedProcedure<br />view<br />otherEntity</p>
<p>Of these, only 'otherEntity' is allowed to optionally specify an 'attributeList' element; the other five entity types are required to specify an 'attributeList' element.</p>
<p>The Data Manager Library assumes that all data entities are required to include the 'attributeList' element, throwing an exception whenever it is not specified. The DML should change its behavior in accordance with the EML schema, allowing 'attributeList' to be an optional element when the DatasetType is 'otherEntity'.</p> Bug #5428 (Resolved): Leading whitespace in data URL causes download to hanghttps://projects.ecoinformatics.org/ecoinfo/issues/54282011-06-24T15:52:20ZDuane Costadcosta@lternet.edu
<p>Leading whitespace in a data URL has been demonstrated to cause the Data Manager to hang during its download operation. A data package that demonstrates this behavior is:</p>
<pre><code><a class="external" href="http://metacat.lternet.edu/knb/metacat/knb-lter-sbc.10.15">http://metacat.lternet.edu/knb/metacat/knb-lter-sbc.10.15</a></code></pre>
<p>One of the two data table entities in the above data package contains leading white space:</p>
<p>"\n <a class="external" href="http://sbc.lternet.edu/external/Ocean/Data/Monthly_Water_Sampling/LTER_monthly_bottledata.txt">http://sbc.lternet.edu/external/Ocean/Data/Monthly_Water_Sampling/LTER_monthly_bottledata.txt</a>"</p>
<p>The fix for this bug would be for the Data Manager to trim leading and trailing whitespace from the data URL.</p>
<p>A secondary enhancement would be for the Data Manager to produce a warning in the form of a quality check whenever it finds that the trimmed URL is different from the original URL. However, this secondary enhancement should be implemented in the DATAMANAGER_QUALITY branch rather than in the 'eml' trunk.</p> Bug #5317 (Resolved): Data Manager Library: Checks for collapseDelimiter instead of collapseDelim...https://projects.ecoinformatics.org/ecoinfo/issues/53172011-02-21T23:04:28ZDuane Costadcosta@lternet.edu
<p>There are two lines in the Data Manager Library source code that contain an apparent bug. The code checks for an EML element named "collapseDelimiter" when it should be checking for "collapseDelimiters". These lines are at:</p>
<p>src/org/ecoinformatics/datamanager/parser/eml/Eml200Parser.java, line 1204:</p>
<pre><code>elementName.equals("collapseDelimiter") &&</code></pre>
<p>src/org/ecoinformatics/datamanager/parser/generic/GenericDataPackageParser.java, line 1278:</p>
<pre><code>elementName.equals("collapseDelimiter") &&</code></pre>
<p>In addition, there are a large number of method names, method parameters, instance variables, and local variables throughout the DML code that are named 'collapseDelimiter' when the more appropriate name for these constructs would be 'collapseDelimiters'. Since these are only names, they do not affect the code logic, but it would be good to clean these up and rename them in accordance with the actual EML element name, 'collapseDelimiters'.</p> Bug #2737 (Resolved): Exception when attributeName contains SQL keywordshttps://projects.ecoinformatics.org/ecoinfo/issues/27372007-01-19T22:26:16ZChad Burtcburt@msi.ucsb.edu
<p>A dataset I'm working with has columns named "group" and "order". When trying to import the dataset into a sql database using the DataManager library I get this exception:</p>
<p>SQLException: ERROR: syntax error at or near "group"</p>
<p>The library has a problem with columns named after SQL keywords. While group may be easier to avoid, order would not when storing taxonomic info. There is a large list of reserved keywords for postgres including Year, Datestamp, etc that I haven't yet tested.</p>
<p>Here is a link to the problem dataset:<br /><a class="external" href="http://sbcdata.lternet.edu/catalog/metacat?action=read&qformat=sbclter&docid=knb-lter-sbc.17.2&displaymodule=entity&entitytype=dataTable&entityindex=1">http://sbcdata.lternet.edu/catalog/metacat?action=read&qformat=sbclter&docid=knb-lter-sbc.17.2&displaymodule=entity&entitytype=dataTable&entityindex=1</a></p>