Re: [Tei-council] first draft release notes

Hi Syd, That is the standard way that those profiles work, by importing html.xsl (which of course imports all the common and html xslt), isn't it? The HTML is just a pass through for those doing readme files that have some html embedded. The makefile I was referring to was the P5 one at: https://github.com/TEIC/TEI/blob/dev/P5/Makefile#L271-L273 But that was just how I was figured out which XSLT profile is being called for the release notes. You are right that grabbing version is a problem, but you've omitted the _real_ problem which is that xslt is run over every xml file in the directory. So you can't pass it the current version. Instead, you would need to get it from the filename itself. So we'd have to get it from the filename. We should be able to do something like: <xsl:variable name="filename" select="(tokenize($document-uri,'/'))[last()]"/> to get the filename and then remove the version number from that. Note, if doing this that it would have to be tei:gi[not(ancestor::tei:ref)] or something, right? (Since there are readme files that already have the refs around them. ;-) Just thinking out loud, James -- Dr James Cummings, James.Cummings@newcastle.ac.uk Senior Lecturer in Late-Medieval Literature and Digital Humanities School of English, Newcastle University ________________________________ From: Tei-council <tei-council-bounces@lists.tei-c.org> on behalf of Syd Bauman <s.bauman@northeastern.edu> Sent: 25 January 2019 17:36:38 To: TEI Council Subject: Re: [Tei-council] first draft release notes Boy, I never looked at it before, but profiles/readme/html/to.xsl looks like a very weird animal. It has templates that match the HTML namespace and ones that match the TEI namespace. That file imports html/html.xsl, which itself imports: common/verbatim.xsl, html/html_param.xsl, and common/common.xsl, which itself imports: common/common_param.xsl, common/common_core.xsl, common/common_textstructure.xsl, common/common_header.xsl, common/common_linking.xsl, common/common_msdescription.xsl, common/common_figures.xsl, common/common_textcrit.xsl, common/common_gaiji.xsl, common/i18n.xsl, and common/functions.xsl I've just looked through all of those (quickly), and none seem to have a template that matches <gi> explicitly. So I think it's probably quite reasonable to just tuck it in here. How would the code in to.xsl get the version number (to point to the Vault)? A parameter? Read P5/VERSION (in which case it needs to be updated before this is run)?
Not sure it does work automatically, one would think that it would follow the same thing that happens with the Guidelines but looking back through some earlier ones I'm not sure it does.
Looking at the makefile which calls the ant script, I think that the release notes are generated using the 'readme' profile of the stylesheets. So:
https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%...
If we put a template in to handle tei:gi in that stylesheet that would probably do it. (Anyone else see a problem with that?)
And yes, I think release notes should always point to the Vault of that version of the the Guidelines because if they point to the live site then by next version they might be out of date.
Tei-council mailing list Tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

That is the standard way that those profiles work, ...
Yes, indeed. I was just listing 'em as that's the set of files I looked through.
The HTML is just a pass through for those doing readme files that have some html embedded.
Ummm ... OK. But of the 35 readme-*.xml files we have in P5/RelaseNotes/, not one has a single non-TEI element.
The makefile I was referring to was the P5 one at: https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%...
Wow. Thanks. (Not so much for the Makefile pointer, but for the cool demo of highlighting lines in GitHub source!)
You are right that grabbing version is a problem, but you've omitted the _real_ problem which is that xslt is run over every xml file in the directory. So you can't pass it the current version. Instead, you would need to get it from the filename itself.
Oh, good point!
Note, if doing this that it would have to be tei:gi[not(ancestor::tei:ref)] or something, right? (Since there are readme files that already have the refs around them. ;-)
Sure, but my plan was to just nuke those <ref> elements. (After all, there are only 35 files ...)
Just thinking out loud,
Thanks for doing so. I will get on that v. important version number bit in a few mins. Also need to add code for <att> and <ident>, which also occur inside <ref> in our collection of readmes.
participants (2)
-
James Cummings
-
Syd Bauman