Metacat: Issueshttps://projects.ecoinformatics.org/ecoinfo/https://projects.ecoinformatics.org/ecoinfo/ecoinfo/favicon.ico?14691340362005-03-23T21:21:46ZEcoinformatics Redmine
Redmine Bug #2052 (Resolved): Metacat does not handle attribute XPATH queries correctlyhttps://projects.ecoinformatics.org/ecoinfo/issues/20522005-03-23T21:21:46ZChris Jonescjones@nceas.ucsb.edu
<p>When metacat is queried and one of the query parameters is a an XML attribute<br />expressed in XPATH, the resultant squery that is created doesn't create the<br />correct XPATH for the attribute. The final '/' before the attribute is<br />ommitted. This should be a simple fix in DBQuery.java.</p>
<p>Also, when <strong>only</strong> an attribute is used as a query parameter, an exception is<br />thrown in newPathExpressionWithOutAttribute in QuerySpecification.java.</p>
<p>An example:</p>
<p><a class="external" href="http://metacat.nceas.ucsb.edu/knb/servlet/metacat?action=query&qformat=xml&/eml/@packageId=%25">http://metacat.nceas.ucsb.edu/knb/servlet/metacat?action=query&qformat=xml&/eml/@packageId=%25</a><br />(produces returned documents)</p>
<p><a class="external" href="http://metacat.nceas.ucsb.edu/knb/servlet/metacat?action=query&qformat=xml&@packageId=%25">http://metacat.nceas.ucsb.edu/knb/servlet/metacat?action=query&qformat=xml&@packageId=%25</a><br />(produces the exception and no documents)</p> Bug #2000 (Resolved): Percentage search does't work in KNB metacathttps://projects.ecoinformatics.org/ecoinfo/issues/20002005-03-02T23:22:39ZJing Taotao@nceas.ucsb.edu
<p>KNB metacat uses oracle db to store documents. There are some limit in this<br />oracle : if search result documents is greater than 1000, it will throw a<br />exception. So user will find 0 documents if they did a % search.</p> Bug #1986 (Resolved): system_id of xml_catalog table points to knb.ecoinformatics.org for a new m...https://projects.ecoinformatics.org/ecoinfo/issues/19862005-02-18T18:53:36ZJing Taotao@nceas.ucsb.edu
<p>After new installation of metacat, administrator should registor known dtd or<br />schema files in xml_catalog. In xml_catalog table, public_id and system_id is<br />for this feature. The public_id will points to the namespace and system_id will<br />points to physical location(url) of the dtd or schema file.</p>
<p>Before metacat 1.4.0 installation, the system_id will point to local dtd or<br />schema file. For example, if user installed memtacat in<br /><a class="external" href="http://some.nceas.ucsb.edu/knb">http://some.nceas.ucsb.edu/knb</a>, the system_id for eml200 will be<br /><a class="external" href="http://some.nceas.ucsb.edu/knb/schema/eml-2.0.0/eml.xsd">http://some.nceas.ucsb.edu/knb/schema/eml-2.0.0/eml.xsd</a></p>
<p>But in metacat 1.4.0 release, no matter where the user installed metacat, the<br />system_id will point to <a class="external" href="http://knb.ecoinformatics.org/knb/">http://knb.ecoinformatics.org/knb/</a>. I think this is not<br />a good idea. If the server in nceas was down, the other metacats couldn't work.<br />So it is better to make the system_id to point to local machine.</p> Bug #1979 (Resolved): The packagetype in xml_relation table is eml200 namespace for eml 201 documenthttps://projects.ecoinformatics.org/ecoinfo/issues/19792005-02-12T01:57:09ZJing Taotao@nceas.ucsb.edu
<p>When a eml201 document insert into metacat, metacat will create record in<br />xml-relation table if the eml201 document points to one or more data files.</p>
<p>The field "packagetype" in xml-relation table should be eml201 namespace for an<br />eml201 document. Howerver, the result turns out the "packagetype is eml200<br />namespace. They are not match.</p>
<p>Except fix the bug, we also need a script to correct existed records in<br />xml-relation table.</p> Bug #1948 (Resolved): Changes in Data Registry (NRS, LTSS, OBFS) Format for ESAhttps://projects.ecoinformatics.org/ecoinfo/issues/19482005-02-09T22:39:35ZVeronique Connollyconnolly@nceas.ucsb.edu
<p>This is a request from Mark Stromberg (<a class="email" href="mailto:stromberg@berkeley.edu">stromberg@berkeley.edu</a>):</p>
<p>Here is a list of things to change for making it more general and include best<br />practices. This both includes a report on a lot of the discussion and a guide<br />for someone actually changing the HTML code in the Guide and on the web form. <br />This is aimed at the web form for ESA. So, the changes should be first<br />implemented in the current LTSS page. Then we can migrate to NRS pages, etc.? I<br />think if you did the LTSS page, NRS and OBFS could do their own editing (Me,<br />Marsh, Kevin Brown)</p>
<p>Overview and Introduction<br />We should put something like this in both the Guide and webform:</p>
<p>What is the Registry?<br />A way to discover data, or more clearly, the person with a data set. Any links<br />provided to real data sets are only potential links, as there is no data archive<br />(yet). So, much of the data registry is focused on letting a user track down<br />individuals who either authored some ecological data, or know where to find it.<br />When an ecological data archive with contributed data sets exists, the registry<br />can provide links to the actual data. For now, the registry is a means to find data.</p>
<p>Revise the format to reflect usability. Consult with Laura Downey.<br />Shade each section<br />Include boxes with examples: Good Examples .vs. Bad Examples<br />Have suggestions under "Bad Examples" and suggestions on "Good Examples" <br />Then, by each section: match the field names in the web form to the field names<br />in the Guide. Need a complete check of Guide to exactly match the web form.</p>
<p>Basic Information<br />Station Name (in Guide): change to Organization Name (webform): For ESA, no pull<br />down, but Organization. Need examples:</p>
<p>User friendly format change<br />Good: National Center for Ecological Analysis and Synthesis<br /> University of Michigan Biological Station<br /> Duke University FACE Facility</p>
<p>Bad: University of California Santa Barbara<br /> Michigan Station<br /> Duke University</p>
<p>Data Set Title<br />User friendly format change<br />Your examples, good and bad are fine; split in boxes. Continue to use the<br />related pairs; a good example and a bad example from each.</p>
<p>Data Set Originator (Guide) .vs. Principal Data Set Owner<br />User friendly format change.<br />"Originator Role" only on Guide. ..not in web form. We should explain in the<br />Guide, and maybe on the web form that the "Originator" is traditionally what<br />ecologists consider to be the "author". Additional Originators are comparable to<br />"co-authors" <br />Check for typos; "not required of you provide" in Guide.</p>
<p>Additional Originators (in Guide) yet "Associated Parties" on web form<br />User friendly format change<br />Maybe show two lines so you immediately know more than one can be there.<br />Add some description of what each role. I like the reduced number of roles, but<br />could add a user-defined role? Or add 1 or 2 more of the roles in the full<br />Morpho set?<br />Here, we might allow people to have more than one role assigned to each<br />additional originator. Also allow a field that can be filled in with "Different<br />From Above".<br />"Originator" is confusing..is it the same as PI?<br />"Metadata provider" is confusing; this is the person filling out the form?<br />Add "Data Poacher"....(Reichman).</p>
<p>I think the consensus was that we need a section called Attribution that lists<br />those who receive credit for the data's existence. We might want to use names we<br />already know like Authors, Co-authors? We need to talk more about this. Or,<br />maybe the clarification in the first field might eliminate a need for "Attribution".</p>
<p>Data Set Abstract<br />User friendly format change<br />Good Example (ok in guide) but need a bad example.</p>
<p>Keyword Information<br />User friendly format change<br />Looks good; matches</p>
<p>Temporal Coverage of Data<br />User friendly format change<br />Looks good, matches</p>
<p>Spatial Coverage of Data<br />Remove the "Use Station Coordinates" <br />Maybe install a place name look up for coordinates...see COREO site that uses<br />the place name (address, street name, etc.) that generates the Latitude and<br />Longitude. If so, put the address before this...<br />Rest looks ok..</p>
<p>Taxonomic Coverage of Data<br />User friendly format change.<br />Three fields, each needs instruction (Taxon Rank, Taxon Name, Taxonomic Authority)<br />Here, we need to add some language that is explanatory. We suggest:<br />There are two fields that will allow people to find the level of taxonomic<br />resolution associated with the data being described. You could include a list of<br />species, with "species" in Taxon Rank, and the name "Ursos arctos" in the Taxon<br />Name. Or, if the data set does not include information based on specimens<br />identified down to the species level, then "Genera" and "Ursos" If the data only<br />include determinations to a higher taxonomic concept, say "Tribes" and<br />"Oryzeae". etc.</p>
<p>Taxonomic Authority: Consensus was that we want the resource one used to<br />identify the taxon or associate a taxonomic concept. For instance, the reference<br />citation for the field guide, key, or nomenclature revision.<br />Suggest that people check in with Glasgow Name Serve or ITIS<br />(<a class="external" href="http://www.itis.usda.gove">www.itis.usda.gove</a>) before they enter their taxa names to avoid mis-spelling.<br />Provide links near data entry field, or put link in guide.</p>
<p>Methods<br />User friendly format change.<br />"Add Paragraph"- what does this do?; why not just keep writing in the box?</p>
<p>Data Set Contact<br />User friendly format change.<br />Again, is there some way we can ask people for a geographic name of where the<br />data were taken and let it generate the Lat and Long? If so, maybe this could be<br />moved up? But, then, we are asking different questions. This is about<br />discovering where to find a person. So, it is fine as it is.</p>
<p>Distribution Information<br />User friendly format change.<br />Dataset Name or Identifier: Give some examples of data identifiers, good and<br />bad. Reference to "filename and path" in Guide is jargon and they are so<br />ephemeral anyway, don't mention them. Delete reference to "station"; replace<br />with organization. Give some good and bad examples of dataset names, maybe from<br />LTER dataset names.</p>
<p>Data medium; give examples; paper, digital files, video tapes, etc.<br />Clearly point out that if any URL provided should be as permanent as possible<br />and that no guarantee that the data may be there in the future, as local<br />computers and servers change names rapidly.</p>
<p>Additional Information/Remarks (Guide) and Additional Information (web form);<br />suggest user describe the data format; Excel, Access, FoxPro, DBase, or any of a<br />large universe of data formats.</p>
<p>Submit Dataset<br />User friendly format change.<br />Very confusing. The web form says....<br />NOTE: You must submit your data in order for these to be entered in the Data<br />Registry! This phrase suggests that the actual data must be submitted somewhere.<br />Just delete the phrase?<br />Make the "Submit" button clearer. The button should say "Submit Metadata<br />Described Above", or something to that effect. The Reset button is dangerous.</p> Bug #1850 (Resolved): Unsuccessful deleting documents when revision number not specifiedhttps://projects.ecoinformatics.org/ecoinfo/issues/18502004-12-28T19:03:55ZSaurabh Gargsgarg@nceas.ucsb.edu
<p>This is regarding the delete functionality in Metacat. I tried deleting <br />documents without entering the revision number and I was not able to do it.</p>
<p>So I tried deleting obfs.2 document and I just gave obfs.2 as the docid and I <br />got a 'Document deleted.' message. But I was still able to read obfs.2. Then I <br />tried deleting by giving obfs.2.1 as the docid and I got a 'Document deleted.' <br />message. I was not able to read obfs.2 after that.</p>
<p>Now as far as I remember, specifying obfs.2 should have worked also as per the <br />functionality coded. In case, that is not what the functionality is supposed to <br />be, 'Document deleted' should not have been returned.</p>
<p>I went through the code and found the following:</p>
<p>// NEW - WHEN CLIENT ALWAYS PROVIDE ACCESSION NUMBER INCLUDING REV<br />// IN IT<br />//AccessionNumber ac = new AccessionNumber(accnum, "DELETE");<br />String docid = MetaCatUtil.getDocIdFromAccessionNumber(accnum);<br />//String rev = ac.getRev();</p>
<p>I am not sure who added the comment. Maybe the person who added the document <br />can explain what is the expected behavior.</p>
<p>When obfs.2 is specified as docid, getDocIdFromAccessionNumber(accnum) returns <br />docid as 'obfs' and hence is not able to delete anything. As none of the sql <br />statements return any error, 'Document deleted' message is returned.</p>
<p>The bug can be fixed by replacing MetaCatUtil.getDocIdFromAccessionNumber<br />(accnum) with MetaCatUtil.getDocIdFromString(accnum).<br />But I am not sure what the person who added the comment had in mind. So please <br />explain it to me so that I can change the code accordingly.</p> Bug #1818 (Resolved): clarifications for DR entry form pagehttps://projects.ecoinformatics.org/ecoinfo/issues/18182004-12-07T23:06:17ZAndrea Andreachadden@nceas.ucsb.edu
<p>1. The entry form page uses the term "registry" toward the top, when it should <br />be "repository".</p>
<p>2. It should be clear early on (before a user begins to enter metadata) that a <br />KNB username and password will be needed. I think it should say this at the top <br />of the page.</p> Bug #1768 (Resolved): Javascript error in "include_searchbox.jsp" for default skinhttps://projects.ecoinformatics.org/ecoinfo/issues/17682004-11-12T11:13:47ZOwen Jonesowen.jones@imperial.ac.uk
<p>Clicking on the common search terms produces an "error on page" (Line: 1, <br />Char: 1, Error: Object expected, Code: 0, URL: <br /><a class="external" href="http://localhost:8080/test1/index.jsp">http://localhost:8080/test1/index.jsp</a>)</p>
<p>I found that if I replace:<br /> "<a target="_top" href="javascript:keywordSearch<br />(document.searchForm, 'alpine')" class="searchsubcat">Alpine,</a>"</p>
<p>with: <br /> "<a href="#" onClick="keywordSearch(document.searchForm, 'alpine')" <br />class="searchsubcat">Alpine,</a>"</p>
<p>It works fine.</p> Bug #1755 (Resolved): Install of data-registry requires cvs checkouthttps://projects.ecoinformatics.org/ecoinfo/issues/17552004-11-03T22:40:54ZSaurabh Gargsgarg@nceas.ucsb.edu
<p>Install of data-registry does a cvs checkout from cvs.nceas.ucsb.edu. For <br />someone outside NCEAS, this requires getting a cvs username info. As a <br />standalone install, these files should be part of the metacat install.</p>
<p>(Chin from Taiwan faced this problem)</p> Bug #1718 (Resolved): spatial search does not workhttps://projects.ecoinformatics.org/ecoinfo/issues/17182004-10-15T19:59:07ZMatt Jonesjones@nceas.ucsb.edu
<p>Spatial search is not working in metacat. This is probably because of casting<br />problems, where the nodedata column is being cast to a number to do the <<br />comparison. Looking at the metadata, it appears that the LAT/LONG values are<br />sometimes empty, and sometimes have leading spaces. These will not cast<br />correctly to numeric values.</p>
<p>The simplest fix is to simply eliminate all non-numeric values before doing the<br />comparisons.</p>
<p>A more complex (and possibly better) approach is to have separate nodevalue<br />tables for each data type we need to treat differently (e.g., String, Numeric,<br />Date).</p>
<p>To see some of the offending values in the data, try a SQL command like this:</p>
<p>select nodeid, nodedata from xml_nodes <br /> where nodetype like 'TEXT'<br /> and nodedata like ' %'<br /> and parentnodeid in <br /> (select nodeid from xml_nodes <br /> where nodetype like 'ELEMENT'<br /> and nodename IN ('eastBoundingCoordinate', 'westBoundingCoordinate',</p>
<pre><code>'northBoundingCoordinate', 'southBoundingCoordinate'))</code></pre>
<p>We really need for query performance to be improved for these more complex<br />searches, so we need to make some design decisions about the overall performace<br />issues.</p> Bug #1711 (Resolved): apostrophe in text element causes xslt errorhttps://projects.ecoinformatics.org/ecoinfo/issues/17112004-09-30T22:41:10ZChad Berkleyberkley@nceas.ucsb.edu
<p>a text element that contains an apostrophe causes metacat to throw an xslt<br />error. It says that the character 0x19 is invalid and can't be transformed. it<br />seems like an apostrophe shouldn't be a problem in a tag like <para> so this<br />needs to be handled by metacat. probably "'" should be replaces with "'" <br />when a file is uploaded.</p> Bug #1703 (Resolved): spatial search does not workhttps://projects.ecoinformatics.org/ecoinfo/issues/17032004-09-28T23:49:48ZChad Berkleyberkley@nceas.ucsb.edu
<p>the spatial bounding box search dialog in morpho does not return any results<br />even when it should.</p>
<p>Also: The labels on the lat/long text boxes do not match up with the labels on<br />the xslt stylesheets for the web site. on the web site, the values are labeled<br />with directions (e.g. 'west') but then use negative values. the labels in<br />morpho just have a directional indicator but do not use negative values.</p> Bug #1551 (Resolved): Error in Result Page tmpl.https://projects.ecoinformatics.org/ecoinfo/issues/15512004-05-03T18:01:43ZSaurabh Gargsgarg@nceas.ucsb.edu
<p>To reproduce the bug:<br />Try to open a document in registry that has been modified or created using a <br />client other than registry (like Morpho) and has fields which cannot be <br />displayed in the registry. The error page which says that following document <br />cannot be changed using registry has a link back to the form with all the <br />values in it.</p>
<p>Solution: conditional statement needs to be added in the error page...</p> Bug #1538 (Resolved): Entity/Character Refeference Conversion Problemshttps://projects.ecoinformatics.org/ecoinfo/issues/15382004-04-28T16:29:39ZDan Higginshiggins@nceas.ucsb.edu
<p>In Morpho, we have run into some problem with the use of special, non-ascii<br />characters (like the 'degree' symbol or greek 'mu'). [Any character represented<br />by a byte with a decimal value > 127 is in this class of special characters.]<br />These characters have been copied from Word or PDF documents into Morpho fields<br />and then put into eml xml docs. Unfortunately, they are not not necessarily in<br />the correct format for xml documents and have caused parser problems.</p>
<p>The solution that was implemented in Morpho was to use entity/character<br />references. Any character with a value greater than 127 is written as '&#xxx;'<br />where 'xxx' is the decimal value of the character. On a Windows machine, the<br />'deg' symbol becomes '°' and 'mu' becomes 'μ'. XML parsers<br />automatically convert these character entities to the character for display, but<br />the conversion depends on the assumed character set.</p>
<p>The metacat problem is that when one submits a document containing such<br />character references (&#xxx;) and then reads the document back, one does not get<br />the character refenence, but rather the character itself! I assume this is due<br />to the XML parser. This is a violation of the idea that metacat should return<br />exactly the same data given it.</p>
<p>Morpho already handles this by converting back to character references any info<br />sent it by Metacat with character values greater than 127. But metacat actually<br />sends back the wrong character for some symbols! (e.g. a 'mu' becomes a '1/4'<br />symbol. I assume this is due to different character set assumption under linux<br />and windows. In any case, there is some data corruption here that we should<br />figureout how to avoid.</p> Bug #1372 (Resolved): Registry: Reconfigure entryForm.tmplhttps://projects.ecoinformatics.org/ecoinfo/issues/13722004-03-18T19:29:10ZSaurabh Gargsgarg@nceas.ucsb.edu
<p>entryForm.tmpl needs to be reconfigured because it has someplaces with skin <br />names have to be hard coded into it. This should be changed and made achievable <br />from cfg file.</p>