[eml-dev] [Bug 2212] New: - Several bug fixes for the eml2nbii.xsl stylesheet (transforming EML metadata to the federal standard, BDP)
bugzilla-daemon@ecoinformatics.org
bugzilla-daemon at ecoinformatics.org
Fri Sep 30 08:42:27 PDT 2005
http://bugzilla.ecoinformatics.org/show_bug.cgi?id=2212
Summary: Several bug fixes for the eml2nbii.xsl stylesheet
(transforming EML metadata to the federal standard, BDP)
Product: EML
Version: 2.0.1
Platform: All
OS/Version: All
Status: NEW
Severity: normal
Priority: P2
Component: eml - general bugs
AssignedTo: jones at nceas.ucsb.edu
ReportedBy: isangil at lternet.edu
QAContact: eml-dev at ecoinformatics.org
TARGET: EML 2.1.0
Several bug fixes improved the stylesheet eml2nbii.xsl. This stylesheet converts
EML documents into the federal standard format BDP.
The stylesheet was delivered to eml-dev, and resides in the LNO CVS.
Here is a log of the changes/fixes
*
* There are about 50 issues (some critical metadata loss) solved in this
revision. Mostly are
* metadata sections that were not completely mapped from the EML to the BDP
standard.
*
* Additionally, there are a few other improvements in the mapping.
* This revision, 3000+ lines, previous 1600.
*
** Itemized list of changes. NOTE, line numbers refer to the previous version
(revision 1.21) of "eml2nbii.xsl"
*
* line 51: Make all possible metadata transformations, that is, set the
variable $show_optional to true (1)
* line 141: Loop over all entities in EML; Previosuly, only dataTable,
spatialRaster and spatialVector considered. "view", "storedProcedure" and
"otherEntity" added.
* hard coded content values changed to english (dataTable to
tabular digital data, etc)
* line 148: BDP "Sername" (series) contents not mapped. Added EML"series"
content, if populated in the EML instance.
* line 158: BDP "pubPlace" contents not mapped. Added EML "pubPlace" content
when used.
* line 161: BDP "publish" contents not mapped. Added EML "publisher" individual
and org. name when used.
* line 172: BDP "onlink" contents not mapped. Added EML
"distribution/online/url" when used.
* line 178: Odd hard coded value removed. It is possible that the author meant
some action.
* line 196: BDP "purpose" contents not mapped. Added EML "purpose"
* line 200: BDP "supplinf" contents not mapped. Added EML "additionalInfo"
* line 243: Handling multiple single data time elements addressed (for each..),
gelogical Scale / Alternatetimescale, etc. About 300+ lines of code replace
about 120 of them
* line 333: Currentness element hard coded to "as it was when data was compiled"
added
* line 330: Comment change
* line 335: Added all possible EML maintenance contents to map into BDP's update
field. Also, there was a bug in the conditional statement.
* line 367: Added bounding altitudes mapping to the geographic coverage mapping
(Not considered before)
* line 369: Added the G-poligon mapping.
* line 539: Added taxonomic keywords, previously forced to NONE.
* line 562: Added the case of optional classification system elements and other
taxonomic-related metadata groups (230 lines of code)
* line 607, 620: Access element code group duplicated for use and access
constraints elements in BDP. Not quite accurate granularity. Removed duplicate code
* line 607: Access element incomplete, the EML intellectual Rights section is
relevant here. Add code.
* line 633: Added EML contact info. Need handle refs.
* line 641: Added in "datacred" (credit) info from EML "associated parties" and
"funding" agency.
* line 663: Addition of metadata: Attemting to use section 2 of BDP: Data
Quality Info. Some sections from EML methods can be used here. quite a bit of
code added
* sections 3 and 4 of BDP skipped for now.
* line 791: Misleading/Obsolete comments refined.
* line 792: Include all EML dataset entities; add storedProcedure, view and
otherEntity types.
* line 794: Not sure about the loop logic. Need testing.
* line 886: Textdomain from attribute type nominal or ordinal mapped correctly
as is to "unrepresentable domain"
* line 911: Unit type added to BDPs "range domain" from EML's attributes of type
interval or ratio.
* line 915: EMLs date time attribute type mapped correctly to unrepresentable
domain.
* line 921: THIS MAY NEED A CHANGE. accuracy missing. It is not exactly the
precision, but if EMLs attr. precision exists, it is mapped to BDPs attr.
accuracy for the sake of preserving metadata.
* line 931: This conditional is out of place: there is no need of a EML physical
element to populate this element. All you need is a "publisher" or a "contact".
An distribution/online element helps.
* line 936: Distribution element. The distributor should be the publisher. If
the publisher is not specified, it could default to the contact. Options included
* line 960: Contact org mapped.
* line 972: Contact person mapped
* line 984: Contact person mapped
* line 1058: Phone mapped better
* line 1072: Fax mapped
* line 1078: Email mapped
* line 1102: In the previous version, the BDP element distribution info is only
mapped if the EML physical element is populated. We consider the case where the
EML "distribution" exists,
* this opens a new case (lots of code)
* line 1122: Sometimes the "data format " -geoform- is placed depending on the
entity we have in EML (data table, spatial vector, etc)
* line 1257: Clarification added to the hard coded text.
* line 1338: In practice, both metadata and data have the same publication date
(pubDate), therefore pubdate of EML data mapped to pub date of BDP metadata
* line 1351: Try to map the metadataprovider as the metadta creator, otherwise
(metadata provider not present in EML) default to the EML contact
* line 1386: add contact organization or contact person name even though we
placed already one or the other.
* line 1475: map better the phone
* line 1490: map the fax
* line 1495: most importantly, map the email.
*
More information about the Eml-dev
mailing list