aso2101 / satavahana-inscriptions-data Goto Github PK

Starting with Bhaja, Bedse, Karle, and Shailarvadi:

Add a <profileDesc> that contains one (only one for now) keyword identifying the type of inscription.

    <profileDesc>
      <textClass>
	<keywords>
	  <term>Donative</term>
	</keywords>
      </textClass>
    </profileDesc>

The <profileDesc> follows the <fileDesc> in the TEI header.

For possible values see here.

• Move all person entities to individual files (from listPerson.xml). Most will be in a directory named after the place where they are attested, but a few (mostly kings and other high-profile people) will be in a "Multiple" directory.
• Determine a standard set of tags/descriptions to use in the person descriptions, alongside important "vocabulary" terms. List these on the wiki for this project.
• Settle upon a convention for Middle Indic vs. Sanskrit names (I prefer Sanskrit names where they're possible to reconstruct, but we might do: attested name (@type="attested"); probable Middle Indic reconstruction (@xml:lang="mi" @type="reconstructed"); probable Sanskrit reconstruction (@xml:lang="sa" @type="reconstructed"). Put these conventions on the wiki.

I think it's probably a good idea to have only text data in the satavahana-inscriptions-data package, and keep the images elsewhere. This will keep the size of the package down, and if the images are on a separate server, we'll be able to take advantage of the IIIF API (or something similar) down the road.

I don't think it's necessarily (or advisable) to move on this now, but @wsalesky, if you have any experience with integrating an eXist app with an external image collection I'd be interested to know how it works.

Refer to earlier conversation with Arlo about the use of source, resp, and wit in entries.

Standardize the practice in these inscriptions: the Sannati data is already okay.

Also: possibly define sigla for bibliography keys.

The bibliographic entries are all found in contextual/listBibl.xml at the moment, and (now that I've corrected the @when attribute) they all conform to the EpiDoc schema. However, it may be better to convert the <bibl> elements to <biblStruct> elements. This would probably involve a minimal number of additions to the already-existing EpiDoc schema. Below I simply describe the structure of the data that's currently there to help us make a decision.

I use the @type attribute to specify whether it is a book, article, incollection, inbook, dissertation or report. (Those are the only values currently represented.) Each of those types has different required elements:

For book: <author>+, <title> (I also have <title level="m"> but this should be changed I think), <pubPlace>, <publisher>, and <date>; <biblScope unit="vol">* and <note>* are optional.

For article: <author>+, <title level="a">, <title level="j">, <biblScope unit="vol">, (with@fromand@to), and ; `* is optional.

For dissertation: <author>, <title>, <publisher> (the university), <pubPlace>, and <date>. <note>* is optional.

For report: <title>, <publisher>, and <pubPlace> (it should really have a <date> and an <author>+, too, although the latter will be an institution). <note>* is optional.

For incollection: <author>+, <title level="a">, <title level="m">, <editor>+, <biblScope unit="pp"> with @from and @to, <publisher>, <pubPlace>, and <date>. <note>* is optional.

inbook is the same as incollection but without the requirement of <editor>+.

Each <author> and <editor> element should have the structure <name><surname></surname><forename></forename></name>. Each <date> element should have a @when attribute for sorting.

"0000" is not a valid value for notBefore-custom and notAfter-custom; change to "0001".

I've made a minor change to pre-install.xql since an installation of eXist was unhappy with unencoded UTF-8 filenames when creating the xml database. It works for me, but if you can't load the data collection in a future attempt @wsalesky check on pre-install.xql. If you can load it, then feel free to close this issue.

Starting with Bhaja, Bedse, Karle, and Shailarvadi:

insert a <msContents> elenent, as below, between the <msIdentifier> tag and the <physDesc> tag in the <msDesc> element.

	  <msContents>
	    <msItem>
	      <textLang xml:lang="eng" mainLang="pra-Brah">Middle Indic, Southern Brāhmī script</textLang>
	    </msItem>
	  </msContents>

Starting with the Bhaja, Bedse, Karle, and Shailarvadi inscriptions:

Convert the inline apparatus (contained in <div type="edition">) into an external apparatus (contained in <div type="apparatus">). See here for some guidelines.

The hash-tags used to refer to sources in the inline apparatus should be converted to bibl:.

this page shows an error

On my local instance there are no issues, but I'm noticing the following kinds of errors when trying to install SAI-data on the server:

Deployment incomplete, 350 issues found: pre-install.xql: /usr/share/existdb/webapp/WEB-INF/data/fs/db/apps/SAI-data/pre-install.xql -> /usr/share/existdb/webapp/WEB-INF/data/fs.journal/txn.22492/oper.1716fe54-5f4d-4bf3-917e-f1a6d6fb7e61/db/apps/SAI-data/pre-install.xql; controller.xql: /usr/share/existdb/webapp/WEB-INF/data/fs/db/apps/SAI-data/controller.xql -> /usr/share/existdb/webapp/WEB-INF/data/fs.journal/txn.22493/oper.d41c0a72-da3a-4fe0-b48d-dafe7ab0b205/db/apps/SAI-data/controller.xql; README.md: /usr/share/existdb/webapp/WEB-INF/data/fs/db/apps/SAI-data/README.md -> /usr/share/existdb/webapp/WEB-INF/data/fs.journal/txn.22494/oper.fd7f1f8f-ebc7-49e4-9bf0-c2fc0529d48c/db/apps/SAI-data/README.md; SaB8-3-sarma1993.jpg

and so on for all of the images in the collection. And now they don't appear on the server.

Any ideas what this could be @wsalesky ?

(This is not a high priority since I plan on moving the images to a separate server.)