Re: [tei-council] Git question

Ah, thanks, that answers at least two of my questions. And (incidentally) confirms my supposition that you haven't actually tested this branch at all! Trust the games night went well and you let him win some. On 17/12/15 23:09, James Cummings wrote:

Lou,

Apologies, out this evening for an it services games night with Sebastian.

Could you check bottom of tei-pm.odd in the simple repo. There were some additional changes that I might not have copied across. Mea culpa.

From phone,

James

-- Dr James Cummings, Academic IT, University of Oxford

-----Original Message----- From: Lou Burnard [lou.burnard@retired.ox.ac.uk] Received: Thursday, 17 Dec 2015, 21:33 To: tei-council@lists.tei-c.org [tei-council@lists.tei-c.org] Subject: Re: [tei-council] Git question

Either this source is not the right source, or the processing-model branch of P5 is some distance from ready for prime time.

a) the text of TD is not valid against p5odds.rnc (can't have a specGrp floating outside a div) b) the source for <paramSpec> refers to a non existent class ("att.identifiable" should be "att.identified") c) the source for <model> uses <paramList> but this has not been defined as a possible child of <valItem> (I am adding <paramList> to model.descLike for the moment)

With these changes made, I can at least generate a p5.xml which is valid. However...

d) the attribute @degree on <certainty> is now past its sell-by and so schematron squawks. Gave it another year to run, for now. e) the attribute @name of <param> has no datatype. Gave it data.name for now. f) the <remarks> child of <paramList> needs both @xml:lang and @versionDate values (sigh); in the same elementSpec there is a <desc> with an attribute "@verisionDate"

With these changes, I get a clean run up to the point of processing the exemplars, where it falls over when it hits DTD generation, of course, but we know what to do about that.

I can't see how <model> ever gets into the content of any other element.

It would be nice to know that I am actually working on the right source, since I would have thought these problems should show up in the most rudimentary testing. Unfortunately, James has now disappeared into holiday land, so this may have to wait until the new year.

In the meantime I'm committing the corrections listed here to the processing-model branch.

On 17/12/15 20:24, Lou Burnard wrote:

...

Getting it to build (i.e. produce a P5 schema) seems like a necessary first step... I'll fix the errors and check them into your branch first.

On 17/12/15 15:11, James Cummings wrote:

...

Lou,

Have at it! I wasn't planning to touch it in the next few days. If something looks amiss you can always try to disentangle Sebastian's original logic in the TEI-Simple repo. Or Magdalena, now also on Council, I'm sure will be happy to comment. :-)

-James

On 17/12/15 15:02, Lou Burnard wrote:

...

Ah yes, sorry, I was looking in the wrong place. There they are!

How about if I check out this stuff, hack your prose mercilessly, and then check it into the lb42-pure-odd-2 branch for you? (Or into dev, if we've actually managed to merge by then)

L

On 17/12/15 14:44, James Cummings wrote:

...

Lou, Basically correct, but I swear I added the specs. Yes, I've just checked here, done a git status and a git push and it says I'm all up to date. they should be in the processing-model branch. On github.com it shows there being a model.xml spec file on the processing-model branch but not on the dev branch, so no, they aren't in dev yet.

In the TD chapter I included the XIncludes to the specs and a new div of prose and the specGrpRef as you noted.

So you are entirely right except your last point.

Thanks for looking at this!

-James

On 17/12/15 13:06, Lou Burnard wrote:

...

James, sfaics in your "processing-model" branch, the only changes are

a) you added a new section #TDPM into chapter TD b) you added a new specGrpRef into section #TDBuild b) you added a bunch of new specs (which are not, however, included in your branch so are presumably already lurking in the dev branch )

Is that right, or am I entirely missing something?

L

On 15/12/15 16:49, James Cummings wrote: > That looked so promising. We've both done changes in the TD > chapter which it thinks are conflicts (but probably aren't > really, I just plopped in a whole new section). > > (and for the record rather than sort it out, I just did a new > clone of the whole repository. ;-) ) > > Ok, I think I'll just do my changes again. > > Just to double check before I do... it is still best to do them > off of the lb42-pureodd-2 branch... right? > > -James > > > On 15/12/15 14:22, Hugh Cayless wrote: >> There’s a few things you could try. Depends a bit on what you’ve >> done and how your changes interact with Lou’s. >> >> I think what I’d try first is: >> >> git fetch >> git checkout lb42-pureodd-2 >> git checkout your_branch >> git rebase lb42-pureodd-2 >> >> That should re-run the commits you’ve made in your branch on top >> of the new pureodd branch and hopefully give you a branch that >> acts like you started from there in the first place. >> >> I'd make a copy of your repo first though, just in case things >> don’t turn out the way you hope. >> >>> On Dec 15, 2015, at 9:07 , James Cummings >>> wrote: >>> >>> >>> Hiya, >>> >>> Is there a way to copy the changes I've made on one branch to >>> another? Or move one branch (off another branch) to be tracking >>> a different branch instead? (e.g. I made the 'processing-model' >>> branch off of Lou's original lb42-pureodd branch and I want to >>> move it to his current one.) >>> >>> Suggestions? Or do I have to go do all the changes again? >>> >>> -James >>> -- >>> Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT >>> Services, University of Oxford >>> -- >>> tei-council mailing list >>> tei-council@lists.tei-c.org >>> http://lists.lists.tei-c.org/mailman/listinfo/tei-council >>> >>> PLEASE NOTE: postings to this list are publicly archived >

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived