search

TEIC / TEI

The Text Encoding Initiative Guidelines
https://www.tei-c.org
Other
269 stars 88 forks source link

the description of att.namespaceable is (pretty much) nonsense #2395

Closed sydb closed 1 year ago

sydb commented 1 year ago

@martindholmes realized that the description of @ns provided by att.namespaceable is bonkers. (But he has to go to a meeting, so I am generating this ticket on his behalf.)

That description is

<desc versionDate="2012-10-21" xml:lang="en">specifies the namespace to which this element belongs</desc>

Surely the @ns is the namespace of the element being defined (or declared or specified) by the <elementSpec> it is on (or by all of the <elementSpec> children of the <schemaSpec> it is on), not the <elementSpec> or <schemaSpec> itself.

Furthermore, it is probably worth warning the user that this namespace does not become the namespace of attribute definitions within.

Oh, and it is missing the period we usually put at the end. (445 English <desc>s do not end in period; 1,449 do.)

sydb commented 1 year ago

Pushed up the “sydb_2395_desc_of_ns” branch with a fix. I have not created a PR yet, because the build seems to be broken on that branch, and for the moment I can’t tell why. (It happens in the detest target, with new error messages about "en" constructs needing to be first — but they are first in the only spec that got changed.) Too hungry to think, hopefully will look at it after dinner.

sydb commented 1 year ago

This ticket is finished, I think. Just waiting on review of PR #2397 from @hcayless, @martindholmes, @martinascholger, or anyone else, for that matter.

sydb commented 1 year ago

2397 has been merged, so @martindholmes’ well-spotted problem has been fixed in ~6 weeks.