CSV metadata table unreachable about fredracor

[cmil]

The immediate cause of the problem is that for each play we currently provide two URLs for the digital source, one is an HTML page and there other the actual TEI from Théâtre Classique. See for instance:

fredracor/tei/abeille-argelie.xml

Lines 28 to 31 in a9fa55c

	<bibl type="digitalSource">
	<name>Théâtre Classique</name>
	<idno type="URL">http://theatre-classique.fr/pages/programmes/edition.php?t=../documents/ABEILLE_ARGELIE.xml</idno>
	<idno type="URL">http://theatre-classique.fr/pages/documents/ABEILLE_ARGELIE.xml</idno>

This is unexpected in the XQuery and causes the exception.

While this should be caught on the API level, the question remains whether it makes sense to keep these two digital source URLs. I would suggest to loose the reference to the HTML page and just keep the TEI URL, since these are actually the documents we use for the import to FreDraCor. What do you think @lehkost?

[cmil]

@lucagiovannini7 The exception is fixed in version 0.91.1 of the API. We should still fix the TEI.

[cmil]

I created a new issue to take care of the URLs, so we can close this one.