[core] support v11n conversion to another book

[chriswep]

putting Slack conversation for reference:

David Instone-Brewer 10:46 AM
Sorry it took me some time to get to these Qs.
On the error log, what I found is:
"no valid book id" for Man, Oda, S3Y, Esg, 2Ma & 1Es
You suggested that this is caused by them being non-OSIS abbreviations, but there are lots of other non-Osis (eg Deu should be Deut - see https://wiki.crosswire.org/OSIS_Book_Abbreviations)
YOu can indeed ignore these rules for standard Protestant Bibles
It would be good to add a filter to your process "If this book exists in this Bible, then..."
I suggest that this is part of your first filter, ie "If the SourceRef exists in this Bible, then...."
2. "moving to a different book is currently not supported"
You only need to process this kind of movement when dealing with Bibles that include Greek Esther AND Hebrew Esther (like the NRSV) or BIbles that have the extra material of Daniel in extra chapters instead of extra books (like many Latin-based Bibles).
3. "invalid chapter number in Est.A:1" (etc)
The sad fact is that some Bibles put the extra material of Greek Esther into chapters called A, B etc. This was Jerome's solution because it allows these chapters to be interspersed within the book (as they should be) instead of making a disjointerd appendix of extra numbered chapters.
(edited)

David Instone-Brewer 10:56 AM
Does this help? Sorry - I thought your questions were going to involve a lot of corrections. If I'd realised they were more general I would have dealt with them earlier.

Chris Metz 12:25 PM
thank you David for your answers. i think i will have to address this issue properly on a later time, however here a few questions for clarification:
book-ids
you are right, the problem wasn't really osis-compatibility, since i was already detecting different types of book-abbreviations. however the books mentioned are not part of the book-list on the end of your rules files at all (and i also can't find most of them in the crosswire wiki)
would it also be safe to do "If StandardRef book exists in this Bible, then ..." (i assume so, however you only explicitly mentioned the SourceRef and there are caes like Dan > S3Y

Chris Metz 12:41 PM
2. i'm hitting serveral walls with this - i guess thats what it means to get experience in a specialized field (like bible software) 🙂 - anyway, if i understand correctly book-moving only happens for versions with apocryphal books? however since BibleEngine is used as a base for STEPapp this will be an important/necessary use case longterm, so at some point there needs to be a major refactoring to support book-movements, right?
3. as far as i can see you never use A/B/C in the standardRef but convert them to subverses right? this would actually play nice with BibleEngine (with some minor refactoring) and would naturally put the content in the right place without loosing the technical efficiency and simpleness of ordering by integer numbers.
one thing i wondered: if you convert Est:A to 1:1.1 to 1:1.17 is it intended that you put it between the original verse one and two (since you renumber verse 1 to subverse zero and not 1:1.18)

David Instone-Brewer 1:23 PM
I've just noticed these two latest messages - sorry! Slack must have thought I'd seen them.
I now have a complete list af abbreviations at the end.
The "Standard" that I'm working with is NRSV, which puts the additional material of Daniel into three additional small books. If you are dealing with a Bible that has none of the additions, the rules should automatically exclude any lines referring to these books (ie S3Y, Bel, Sus.
These additional books exist in the StandardRef - they exist there by definition. If the Source Bible doesn't have this material, these books will be empty at the end of the process, so they can be discarded.
However, some of the Tests do ask questions about these books that may be non-existant in the SourceRef, eg "Sus.1:64=Last" which doesn't make much sense if Sus doesn't exist. Your idea of adding "If Sus exists" is a good fix for this.
2) Yes, some 'book movements' are needed for some additional apocryphal material
3) Yes, I use numerical subverses for machines, and convert these to "a" "b" etc for notes that humans will read (because this is traditional).
Yes, one reason for using ".0" for the original text of a verse before adding subverses is to make sure that the ".0" material occurs BEFORE the additional subverses. The numbering of subverses is designed so that numerical sort will result in the correct order of text.
If I fail to respond to something (like I did this time) don't hesitate to give me a poke. I don't intentionally want to leave you hanging.

David Instone-Brewer 6:18 PM
I've had a change of heart wrt Esther
As you know, the extra material in Esther can be inserted into the Standard versification as subverses,
or it can be added as extra chapters 11-15
or it can be added as extra chapters called A-F (which unfortunately don't match the extra numbered chapters).
Initially I had the Callouts and Notes often referring to the extra chaps A-F.
I've now changed that to extra chaps 11-15 (except in Callouts where the original has chaps A-F) because I think this is more common and certainly more understandable. I've also upgraded the mappings so that all three options are shown, eg (Hebrew= -- • Latin=11:7/A:6 • Greek=1:1f)
I've also added new Phrases to translate:

1. %In some Bibles only the start of this verse is present%.

• This replaces notes such as:
  "%Normally in this Bible this verse includes words that are at% 11:19, 21-22, 24, 26-28"

2. %In some Bibles this verse contains extra text%.

• This is added to Bibles (eg Hebrew and Protestant Bibles) that do not have these extra subverses
  This doesn't get rid of those pesky chapters A-F from the tests, unfortunately.

David Instone-Brewer 7:15 PM
I've also removed the concept of subverses of subverses (which onluy occurred at Est.1:1.17.1 and Est.1:1.17.2) by creating Est.1:1.18

David Instone-Brewer 7:58 PM
Another programmer was having problems with refs like Est.A:1
I had another look at this, and realised that the big problem is that these refs exist in the Tests.
I had a look at the issue and found a way to remove them.
There are now no tests such as Est.A:1=Exist or Est.A:1=NotExist
I'm sorry that I didn't think of this before!