Hi Chris,
We've had this in progress for a while and a facility for
updating/converting Bibles from Michael's site is already mostly
implemented. It's essentially run by a Python script that grabs the USFM
files when they're new, validates them against Michael's hashes, calls
usfm2osis.py to convert to OSIS, generates or updates a .conf, calls
osis2mod to generate a Sword module, and adds or updates the module
files in the wycliffe repository (which was set up for these Bibles and
other minority language Bibles from Wycliffe, the Bible League, and
similar).
The only remaining issues are that I haven't addressed .conf generation
and at the moment there's a moratorium on release of all modules that
don't predictably and exactly fit within their versification system.
Since these don't all predictably fit within any versification system, I
won't be working on this at all until someone (possibly me) fixes the
mod2osis bug causing the obliteration of remapped verses.
--Chris
On 10/28/2013 03:35 PM, Chris Burrell wrote:
Hello
There are around 260 Bibles available on ftp.ebible.org
<http://ftp.ebible.org>. I've started automating a tool that takes the
OSIS files produced by Michael's Haiola and converts them into Sword
modules/conf files. These are mainly translations for Papua New Guinea.
There are still some issues to sort out like choosing the right
versification and some possible data cleansing issues.
In the meantime, I was wondering if CrossWire would like to host these
and if so what the process would be.
* registering the initials for each version. At the moment, Michael has
them labelled by their ethnologue language code, which only creates 1
collision with the current modules (the Klingon translation!)
* who would maintain these conversion going forward (happy for that to
be me, or someone else).
* whether we would consider new versification schemes to accommodate
some of the translations. For example, the NRSVA scheme which is the
only scheme which supports 4 Macc. only contains the Old Testament (at
least in JSword) . Some of the translations seem to be based on the
NRSVA versification for the OT but also have a NT. What would be the
preferred approach for coping with this kind of scenario?
(Also, if anybody has any ideas on how to detect the most appropriate
versification for a particular OSIS text, please do let me know - the
only 2 ideas(fuzzy matching a full list of verses, or identifying verses
that differentiate the versification schemes - both have their own set
of issues)
Chris
_______________________________________________
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page
_______________________________________________
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page
