On Wed, Jul 3, 2013 at 3:11 PM, Greg Hellings <greg.helli...@gmail.com>wrote:
> Chris, > > > On Wed, Jul 3, 2013 at 2:54 PM, Chris Burrell <ch...@burrell.me.uk> wrote: > >> Hi >> >> Do versifications sometimes have different verse orders to each other. >> >> Is this something Sword copes with/is intending to cater for? >> > > My understanding is that no, we do not cater to books, chapters or verses > being out of order. This can be a problem for trying to interweave > deuterocanonical material in Esther and Daniel depending on how a > particular text labels those verses and chapters. But I don't remember > hearing anyone provide concrete examples of this actually happening. > > >> >> My aim is to be able to do a text comparison verse for verse, and so >> assuming I have the mappings between each verse, can I take two lists and >> work my way down... In other words >> > > I have written a Python script, available > http://www.crosswire.org/svn/sword-tools/trunk/versification/av11n.py, > that compares every available Sword versification to a given OSIS document > and reports the deviation from that versification. For it, I first > construct a list of all the refs in a versification, pull out ones that are > encountered in the document, and report the "missing" ones. Not sure if > that helps. > I mentioned this one specifically because the OSIS file can be a giant jumbled pile of <verse> elements that don't have to be in any particular order and our utilities will put all the verses in the same order when they construct the module. --Greg > > Alternatively, I have > http://www.crosswire.org/svn/sword-tools/trunk/modules/compare.py which > compares two installed modules, verse by verse, to see if they are exactly > identical in text throughout. > > Maybe one of those will help inspire your implementation. Or maybe not! > caveat emptor: they may not have been updated for all the API churn over > the past few months so things like .Error should be .popError and > appropriate getter/setter should be recognized. > > --Greg > > >> >> int i, j = i = 0; >> >> while(i < passageLength & j < passageLength) { >> if (v11nA[i] mapsTo v11nB[j]) { >> output both verses >> } else { >> work out with of i & j is behind >> then i++ (or j++) >> } >> } >> >> In the above, i and j only increment and j in particular doesn't jump >> around. The idea above, is that I can read the passages for v11nA and v11nB >> up front, and then process sequentially. (as opposed, to currently in >> JSword, making multiple reads to the backends for each verse.) >> >> Yes, there would be duplicates since 1 verse may map to multiple verses, >> but that's ok I think. >> >> 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
