mauricemengel.de

Exporting MPX, Part II

Maurice — Thu, 11 Mar 2010 16:45:37 +0000

This is the second part of my blog on exporting MuseumPlus (M+) to xml. I write this part for Frank. In this part, I describe how you get simple XML out of the RTF which produced in the first part. I had written perl, batch and xslt years ago and I was really curious if it still works, -- and yes it does! The outcome of this step is nice xml in a structure which obviously mimicks the "internal" structure of the database. I put internal in quotes since my xml format uses the screennames of the fields. It does not really look into the representation inside the database. The next transformation which I usually execute is a sanity check on the data; I also correct know bugs, for example, if one collection always confuses "Hersteller" and "Produzent" then I replace it so that in the end it looks better. I call this third step a fix, because the input format and the output format remain the same. As the fourth step, I used to transform this data to museumdat.

How it works

So, I unzip the contents of the archive, go in that directory with dos and execute the stupid batch (levelup.bat) which runs my perl program and some xslt transformations. If you want to do it too you need to adopt some paths in the batch. It should be self-explaining. I attach the relevant xslt. You need saxon as a xslt processor.

I attach the test data, the stupid batch which you might need to adapt to your system and the simple perl script.

C: cd temp levelup

Main Results

The levelup command should produce nice xml output in the format which I call MPX-lvl2 (level2). It does so in a number of steps in the middle which sometimes go wrong and which I should explain some time, but not now.

The main result is a file called "join.lvl2.mpx" which contains the combined data in mpx-lvl2. It's so short that I can quote it here (I leave out some tags, indicated by [...], full version as attachment). Please note that in the file the unicode works, just not here after I post it directly from notepad:

<?xml version="1.0" encoding="UTF-8"?>
<museumPlusExport xmlns="http://www.mpx.org/mpx"
                  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                  xsi:schemaLocation="http://www.mpx.org/mpx File://C:/Perl/site/lib/XSD/mpx-lvl2.v2.xsd">
   <personKÃ¶rperschaft kueId="4207" exportdatum="2010-03-11T17:36:49">
      <bearbDatum>13.06.2003</bearbDatum>
      <bemerkungen>DÃœ-Vermerk: Original - Archiv</bemerkungen>
      <name>Kubik, Gerhard</name>
      <nennform>Kubik, Gerhard</nennform>
      <typ>Person</typ>
      <verantwortlichkeit>EM-Medienarchiv</verantwortlichkeit>
   </personKÃ¶rperschaft>
[...]
   <sammlungsobjekt objId="240472" exportdatum="2010-03-11T17:28:19">
      <andereNr art="Produktions-Nr.">GC 2-12539</andereNr>
      <andereNr art="ID">4</andereNr>
      <andereNr art="Alte Inv. Nr.">P 0133</andereNr>
      <andereNr art="DAT-Nr.">1</andereNr>
      <bearbDatum>13.06.2003</bearbDatum>
      <besetzung>vokal/instr.: SÃ¤ngerin, Mandoline, Tamburin</besetzung>
      <credits>Staatliche Museen zu Berlin, PreuÃŸischer Kulturbesitz, Ethnologisches Museum</credits>
      <geogrBezug bezeichnung="Ethnie">Perser</geogrBezug>
      <geogrBezug bezeichnung="Land">UdSSR</geogrBezug>
      <geogrBezug bezeichnung="Ort">Tiflis</geogrBezug>
      <geogrBezug art="Region">Georgien</geogrBezug>
      <identNr art="Ident. Nr.">VII 78/0004</identNr>
      <objekttyp>Audio</objekttyp>
      <personKÃ¶rperschaftRef funktion="Produzent">Gramophone Concert Record</personKÃ¶rperschaftRef>
      <sachbegriff>Schellackplatte</sachbegriff>
      <titel art="Titel">Titel A: [Persisch]</titel>
      <verantwortlichkeit>EM-Medienarchiv</verantwortlichkeit>
      <verwaltendeInstitution>Ethnologisches Museum, Staatliche Museen zu Berlin</verwaltendeInstitution>
   </sammlungsobjekt>
   <sammlungsobjekt objId="1432496" exportdatum="2010-03-11T17:28:19">
      <andereNr>M30-37281/2</andereNr>
      <bearbDatum>22.10.2009</bearbDatum>
      <identNr>VII LP 5264</identNr>
      <maÃŸangabe typ="Geschwindigkeit (Schallplatte)">33 U/min</maÃŸangabe>
      <objekttyp>Audio</objekttyp>
      <personKÃ¶rperschaftRef>kein Eintrag</personKÃ¶rperschaftRef>
      <sachbegriff>Schallplatte</sachbegriff>
      <verantwortlichkeit>EM-Medienarchiv</verantwortlichkeit>
      <verwaltendeInstitution>Ethnologisches Museum, Staatliche Museen zu Berlin</verwaltendeInstitution>
   </sammlungsobjekt>
</museumPlusExport>

TODO: Intermediary steps

explain the intermediary steps

Attachment	Size
levelup.bat	5.27 KB
rtftable2xml.pl.txt	18.6 KB
newJoin.xsl	1.01 KB
lupmpx.4thGen.xsl	38 KB
join.lvl2.mpx	6.5 KB

Exporting MPX, Part I

Maurice — Thu, 11 Mar 2010 12:06:23 +0000

Why and what for?

This is a description how to get XML data out of MuseumPlus (M+). I write this for Rainer and his current project. The idea is to see how this export can be optimized, and how much can be done by others. Originally, I have written this export for ethnoArc. This page describes the first part of the process which could easily be done by someone else.

Technical overview and justification

Since I do not have access to the M+ internals, I have to use an existing way of exporting data and this is via M+'s reports. Strangley enough, M+ creates rtf files with the data in tables when one creates certain reports. I use those rtf, transform them to stupid xml with a perl script and then with another transformation (in XSLT) to nice xml. I described the format in an older post.

Preparations

I defined several lists (in M+ they are called "Liste") with almost all the fields in the Sammlungsmodul and PersonenKörperschaftsmodul. (In the future, I might post them to show you how this step works, but at the moment this is not necessary.) The only important thing at this point is how the lists are called (list name in M+ followed by short label):

EM-Export-Sammlung eABasic: eabasic
EM-Export-Sammlung eADatierungExtras: eadatierungextras
EM-Export-Sammlung eAext: eaext
EM-Export-Multimedia eA: mume
EM-Export-Sammlung eAoov: oov
EM-Export-Personen/Körperschaften (PerKör): perkor
EM-Export-Personen/Körperschaften (PerKör ext): perkorext

Step 1: Selecting records for export

The first step is to identify all the records you want to export. You could, for example, select all objects (records) you want to export in one group ("Objektgruppe"). Up to approx. 3000 objects work on my machine without a problem.

Create another group for and put all the Personen/Körperschaften which you need.

M+E Objektgruppen

Step 2: Export to RTF

For every of the seven lists named under export the whole gruppe.

Select the next list

M+ creates a file in C:\Temp\RFTFile.rtf
Rename that file to appropriate short list name (as given under preparations): eabasic.rtf
Repeat for first five lists
Change to the Personen/Körperschaften Gruppe you created in Step 1 and repeat for the last two lists.

Put everything in a folder and compress it using zip and send it to me (e.g. via Pool). Attached to this post, there is an example.

Troubleshooting

My M+ complains "Keine Vorlage vorhanden - Datenquellengeneriert". This is intended and no problem.
You have to close RTFFile.RTF in Word, otherwise M+ does not generate a new file.

The next step

This part of the export is really easy, just a little tedious if you deal with a lot of objects. You can easily make a mistake and not notice it. On the other hand, all you need to do is a little M+ and Windows stuff.

Attachment	Size
text-export-rtf.zip	12.32 KB

Tomcat

Maurice — Wed, 10 Mar 2010 08:40:15 +0000

My Tomcat problem (diagnosis)

I start tomcat. It works, serves static and dynamic pages for a few minutes and then seems to freeze. Every dynamic request times out.

java.exe in task manager shows 46432 K memory usage and doesn't change (with 19 threads).

Attempts

I still have this problem, after I put -Xms228M -Xmx512M in %JAVA_OPTS on my laptop and if I start tomcat from windows shell in the tomcat directory.

Looking for Log files

I cannot find any log files now from today. They are created, but empty. Somehow, CATALINA_HOME is not set under windows. So I set it again. I expect that logs will be written then. Yes.

What do they say?

Still nothing, but when I start a second instance it correctly complains in the log files that the port is blocked. The tomcat window says nothing after reporting successful startup in so-and-so many ms.

java.exe is 44820k - 44824k changing very, very slowly and 19 threads

From cygwin

Let's try it one more time from cygwin. Same thing. Now I try with a different version (and new installation) of tomcat 5.5. This time I choose the windows installer.

Another try

Antoher computer, the one at work. Hardly any java on it. I install a new java jdk, set JAVA_HOME and CATALINA_HOME and download tomcat and the server, set the admin password in tomcat-users.xml and it just works without any further ado. As far as I can tell in the little time left. So, I will try this on the laptop, too.

From my OAI Diaries

Maurice — Tue, 09 Mar 2010 09:10:51 +0000

I am trying to learn OAI for the MIMO project. I am currently looking at Jeff Young's oaicat. The problem with is that I only had a vague or no idea at all about the OAI metadata harvesting protocol, Java servlets and Tomcat, Ant etc. when I started to look at it.

Tomcat

Now I am a little more familiar with Tomcat, but I still have problems configuring it on some of my test servers. On my laptop, I use it with cygwin and it seems to hang after some time. It runs best on my webserver so far. I need to look at in detail to describe the problem better.

edit: my experiments on this front

Documentation

I find it a bit confusing that documentation of oaicat is on several different pages and not really good enough for a servlet newbe like me:

Homepage at OCLC: http://www.oclc.org/research/activities/oaicat/default.htm
Customization notes included in the war; also accessible at demo installation at http://alcme.oclc.org/oaicat
Wiki (seems not to be updated anymore): http://alcme.oclc.org/wikid/CollectionOaiCat:FrontPage
Code at Google: http://code.google.com/p/oaicat. Google code has no documentation, but the source comes with some api documentation.
Readme in the src. Seems only to be included if you checkout oaicat via svn. It has a FAQ section with 4 questions. Quite useful. I attach it below.

Code

Like the documentation is stored in different place, the code comes in several forms as well and at least the war file seems to be outdated:

download old war from google code at http://code.google.com/p/oaicat/downloads/list
checkout ant build version with subversion: http://code.google.com/p/oaicat/source/checkout
there is also a new jar distribution at http://code.google.com/p/oaicat/downloads/list

Starting oaicat

This is a quote from the oaicat demo. I have been starting tomcat from that dir and it worked correctly:

Caveat! This webapp assumes that Tomcat will be started from the CATALINA_HOME subdirectory. If these pages appear, but the servlets invoked from the forms below fail, chances are you started Tomcat from a different location. In this case, edit the webapps/oaicat/WEB-INF/web.xml file and change the following relative reference to an absolute file path.

Deploying

Today, I finally deployed the war file from http://code.google.com/p/oaicat successfully. The whole thing runs fine and I can change oaicat.properties and restart tomcat. I don't really like the html that comes with oaicat, but it's okay. I would rather be able to switch between the naked xml and the html view, and I think it would be helpful to give more info on the kind of data I have to fill in. Some of it should be trivial to adjust; I could restart the oaicat without the oaicat.xsl if nothing else is worth it.

Updating

To update one of the documentations (not the readme in src) says to copy new oaicat.jar into the deployed directory. I did this at my PC at work today and it worked like dream. I kind a miss that I don't see which version of the oaicat I am running.

Ant

The next step seems to be to understand the build process. I can checkout a more current version via svn. Procedure is described at google code. As far as I am guessing, this build process is a pre-requisite for the procedures I planning in the next section. So the next task seems to be to set up a working oaicat from source via ant.

edit: I convinced ant to build me something, now I am not sure what to do with it. It gives me many warnings (mainly deprecation), but at the end it says built successfully. The result looks very much like the .tar.gz which I can download as well. I am not sure what file I have to copy to the webserver. Probably it would be best for me if I can use eclipse, build and run on the same machine. Currently, I can't. I have this weired tomcat problem on my laptop. So this is still todo.

Setting up oaicat - The plan, part 1

My idea at the moment is to set up an oai server which basically understands the mpx which I get from the database and then to use various xslts to transform the data in museumdat and lido.

Customization notes in the readme (see above).

Looking at the demo config I: Classes

AbstractCatalog.oaiCatalogClassName=ORG.oclc.oai.server.catalog.FileSystemOAICatalog
AbstractCatalog.recordFactoryClassName=ORG.oclc.oai.server.catalog.FileRecordFactory

Can I just use them or do I need to change them? Will most likely need to change both, although FileSystemOAICatalog looks easy to adapt at first glance, although longer. Probably another illusion due to my lack of java.

Looking at the demo config II: Transformations

The demo uses file system as default. I see no reason at the moment to change that. Actually, it is my most important reason when opting for oaicat that I can use more or less my MPX output directly:

AbstractCatalog.oaiCatalogClassName=ORG.oclc.oai.server.catalog.FileSystemOAICatalog
AbstractCatalog.recordFactoryClassName=ORG.oclc.oai.server.catalog.FileRecordFactory

The oaicat demo comes with 5 records of metadata describing publications. The native format is oai_etdms. This is Electronic Theses and Dissertations Metadata Standard. Here is some background info on it http://outgoing.typepad.com/outgoing/2005/04/etdms.html. Nothing spectacular, so it is a good demonstration. The native format can be mapped to oai_dc.

So how is that reflected in the oaicat.properties:

FileSystemOAICatalog.homeDir=webapps/oaicat/WEB-INF/META/

The above seems self-explaining, the below is worth it look at it a little further:

Crosswalks.oai_dc=ORG.oclc.oai.server.crosswalk.FileMap2oai_dc
Crosswalks.oai_etdms=ORG.oclc.oai.server.crosswalk.FileMap2oai_etdms
FileMap2oai_dc.xsltName=webapps/oaicat/WEB-INF/etdms2dc.xsl

The FileMap XSLT from the last line converts from native format (etdms) to dc as the filename suggests. I will shorten it a bit:

<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
                              xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/"
                              xmlns:etdms="http://www.ndltd.org/standards/metadata/etdms/1.0/"
                              xmlns:dc="http://purl.org/dc/elements/1.1/"
                              exclude-result-prefixes="etdms">
<xsl:output method="xml"
        omit-xml-declaration="yes"
        encoding="utf-8"/>

<xsl:template match="/etdms:thesis">
    <oai_dc:dc xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
      <xsl:apply-templates/>
    </oai_dc:dc>
</xsl:template>

<xsl:template match="etdms:title">
    <dc:title><xsl:apply-templates/></dc:title>
</xsl:template>

<xsl:template match="etdms:creator">
    <dc:creator><xsl:apply-templates/></dc:creator>
</xsl:template>

...

<xsl:template match="etdms:degree">
</xsl:template>

</xsl:stylesheet>

Customization - More on the plan

There were some customization notes in one of documentation source. Also, the readme.txt (see attachment) gives some hints.

I guess I need to get my MPX running as native format first and later deal with transformations, but not even this is clear at the moment.

A new native format

Quote from the readme. It looks as if there were two options for XML. I don't see a reason to prefer the one over the other, so I will go the default of the demo as long as possible:

Q3: What are the semantics of these record objects?
    a) If FooOAICatalog returns records as byte arrays, examples can be anything such as MARC Communications Format. If FooOAICatalog returns Strings, examples might include MARC BER, or any kind of XML String. If FooOAICatalog returns DOM Documents, examples can be any XML-based metadata format. Let's assume FooOAICatalog returns records as DOM Documents containing MARCXML content.
    b) Copy oaicatjar/src/ORG/oclc/oai/server/catalog/XMLRecordFactory.java
       to com/acme/oai/server/catalog/MARCXMLDOMRecordFactory.java
       and modify the code so the class name matches the new filename.
    c) Change the methods to cast each Object nativeItem parameter to a org.w3c.dom.Document and use it to extract the relevant data for each method.
    d) Make a corresponding package/class name change to the
       AbstractCatalog.recordFactoryClassName entry in the
       webapps/oaicat/WEB-INF/oaicat.properties file to have OAICat use your custom class.

Set oaicat.properties in d) accordingly. That's the only part that seems obvious even for me. I have an idea what c) means, but I should look at it more closely.

Next step will be to look at the way the demo is working, i.e. at

oaicatjar/src/ORG/oclc/oai/server/catalog/XMLRecordFactory.java
what is the equivalent of FooOAICatalog in the demo?
Procedure: Can I get the whole thing running without a transformation? If not how can I ever debug it? If not, I would need to start with the transformation and only when I understand that move on to this part.

Transformations

That would be the next step if will ever get so far. Implement a transformation from MPX to museumdat and LIDO.

Quote from the readme:

Q4: What OAI metadatdaFormats will be supported?
    a) Examples include oai_dc, marcxml, or oai_etdms.
    b) For oai_dc, copy oaicatjar/src/ORG/oclc/oai/crosswalk/XML2oai_dc.java
       to com/acme/oai/server/catalog/MARCXMLDOM2oai_dc.java and modify
       the code so the class name matches the new filename.
    c) Change the constructor to use the appropriate schemaLocation for
       this metadataFormat.
    d) Change the methods to cast each Object nativeItem parameter to a
       org.w3c.dom.Document and use it to service the method accordingly.
       In this case, you could use the Library of Congress MARCXML to DC
       XSL stylesheet (see http://www.loc.gov/standards/marcxml/) to
       perform the crosswalk to Dublin Core.
    e) Repeat steps b, c, and d for each metadatdaFormat to be supported.
    f) Make a corresponding package/class name change to the
       Crosswalks.* entries in the webapps/oaicat/WEB-INF/oaicat.properties
       file to have OAICat use your custom classes.

Todo

It will be interesting to look at the java classes next, but now I have to do other buisness first...

Attachment	Size
README.txt	4.34 KB

XSLT Testing

Maurice — Sat, 06 Mar 2010 12:21:17 +0000

I found this on the XSL mailling list: http://assets.expectnation.com/15/event/2/Testing%20XSLT%20Presentation.pdf

The document as such is great, but now I wonder which approach is the best for me respectively is the eaisest to learn.

At the moment, I look at I http://utf-x.sourceforge.net/manual/manual.pdf, but I am really not sure this is the way to go. One question seems to be with or without Java.

http://www.proxml.be/users/paul/weblog/fc7ff/Unit_testing_XSLT.html

Restructuring this site

Maurice — Sat, 27 Feb 2010 14:39:48 +0000

Hi everyone!

I decided that this weekend I will take some time to restructure my site. It's the first site I made with Drupal and it looks terrible in everyway. This weekend I change it a little bit, so don't wonder why something might not work.

Maurice

Remote Drupal Admin

Maurice — Mon, 19 Oct 2009 13:35:05 +0000

I have too many drupal sites now and updating modules etc (i.e. standard tasks) start to get annoying. I wonder if there are already tools to control drupal remotely and if I should write some small scripts using WWW::Mechanize or something else.

Also, I wonder if I should write web tests for the drupal sites I am developing.

This is just an idea. Here are some links I find helpful:

What would be easy?

test if site is up
execute update.php, cron.php
report/show drupal error messages

oai_browser.pl -> oai_browser.exe

Maurice — Fri, 11 Sep 2009 21:26:19 +0000

Here is a windows xp executable of Tim Brody's oai_browser from cpan.

Attachment	Size
oai_browser.zip	2.44 MB

docx2txt.pl

Maurice — Thu, 03 Sep 2009 16:45:14 +0000

Recently, I installed search files http://drupal.org/project/search_files and noticed that I still cannot search my docx files. So here my first attempt to solve the problem. A little perl script that reads in docx, unzips it, and displays the content in no particular order using libxml. It's my first attempt. Version 0.001 after a few minutes.

Found a similar project: http://docx2txt.sourceforge.net

which uses

/usr/bin/unzip and
manual xml conversion (regular expressions) and
manual unicode conversion
is not available on CPAN.

I still cannot hide the exceptions from Archive::Zip.

Edit: I attach a version which should run easy enough as a script. Now called catdocx

Attachment	Size
docx2txt.pl.txt	1.52 KB
catdocx_simple.pl.txt	1.78 KB

Last time i used LibXML

Maurice — Sat, 25 Jul 2009 12:47:14 +0000

Recently, Martin has transferred recordings from 108 DATs to hard disk. Unfortunately, he didn't preserve the ids on the DATs, but instead segmented the audio according to tracks according to silence. This is not the same, since ids are sometimes one and sometimes two tracks on these dats. Thing is, the recordings were originally stored on 78s. The ids consistently describe these records, i.e. for each record has one id, no matter how many sides it has.

Although this segmentation was not intended, it provides an interesting case study, since segmentation according to silence is relatively easy to automatize. However, the result is worthless if one cannot associate the result with the original chunks. In other words: if I cannot associate tracks with ids, the audio chunks are worthless.

So far so good. The problem is that when I start have the db which stores

tape number (tape_no)
id
identifier
...

while the new files come only with two types of information:

tape number (tape_no) and the
track(track_no).

The only common field here is tape_no. As explained above, track_no is usually different from id. Unfortunately, I cannot link the two tables because a lack of common fields.

So, I try to come up with an logarithm to do exactly this. Possible? Theoretically yes, in practice it is pretty fishy.

Since a 78 can have one side or two, there is no easy rule how I come from tracks to ids. Instead, I look them up in our database. If I find two titles it usually means that the record has two sides and that means I need two tracks for one id. This is fairly reliable. THere a few cases where there are no titles. And another set of examples, where somebody put in translations, so that there are 4 titles. I didn't come across two titles where the 78 has actually only one side. But maybe I should check this again.

So, what I decided to do is what I call a "reverse loopup". This seems not to be an existing term in this context judging from google. I don't know. Perhaps it is not a good term either. You can judge yourself. What I mean is that with a normal query (lookup), I would start at either end, i.e.

either from the db side with tape and id, loop over both and access every item; or the other way round from the
file side with tape_no and track_no.

Both don't work here, so I change order in the middle of the query. I start in one direction and then change direction. That's at least how I think of it. Not quite sure, if this description is accurate. Rather not. More likely I reconstruct ids which I actually do not have from the fact that I know that an id consists of either one or two consecutive tracks.

More concretely: I loop through the tapes. For each tape, I create ids. I assume that there is a id=1. For this id, I look up in the db record with that tape and that id. In that record i check wether this record has one or two sides. Accordingly, I associate this id with one or two tracks and continue with the next id and the next track. Just a loop with two variables.

So far for the theory in which this works fine, but unfortuneately my db data is not so reliable. Once in a while, something is missing and I don't find a a tape_no and id pair. What then? I cannot be sure if that record has one or two sides. And that means that all the id-track associations afterwards may or may not be correct anymore. Now, at this moment, I could guess, but this is not good enough. I need to rely to the information I produce, so I have to abort the whole tape and continue with the next one. (Right now as I am still writing on the script, half the tapes have something missing...:-(

Now, before I do this. I try to eliminate a few missing gaps in the data manually.

A few comments:

Doing this, I notice problems in my xml data. When $id = $tape_no the info is missing. For the first 13 tapes or so. I didn't see this error so far. I guess the error is in lvl3 (fix) conversion or somewhere else on the way. I correct it in my xml and then it should not be relevant anymore.
I will attach the perl once it's done, but let me just put the sections relevant for libxml in here since I will look at it again the next time I will use libxml. This time I tried to use xpc, although it wasn't necessary.

...

use XML::LibXML;

...

$xml = '/home/mengel/EMEM-78test.lvl3.220708.mpx';

...

my $parser = XML::LibXML->new();
my $tree = $parser->parse_file($xml); #tree is doc
my $root = $tree->getDocumentElement;

...

my $xpc = XML::LibXML::XPathContext->new;
$xpc->registerNs( 'mpx' => 'http://www.mpx.org/mpx' );

...

my $xpath =
"//mpx:sammlungsobjekt[mpx:andereNr[\@art = 'ID' and . = '$id'] and mpx:andereNr[\@art ='DAT-Nr.' and . = '$tape_no' ]]";
my @nodes = $xpc->findnodes( $xpath, $root );

...

        if ( $#nodes eq 0 ) {
            my $identNr       = $nodes[0]->find('mpx:identNr');
            my @titles        = $nodes[0]->findnodes('mpx:titel[@art]');
       }

I tried XPathContext here although not necessary. Not exactly sure why. Have an idea, but this is not the topic here anyways. So far so good. I guess I continue in a comment when I find it.

Attachment	Size
78er_file_map.txt	187.23 KB
78er_missing.txt	15.51 KB
78er_warnings.txt	24.65 KB
78er_file_cache.txt	205.14 KB
78er.missing_tracks.txt	1.72 KB
Rename78.pl.txt	40.74 KB