Inputting BBO Extractor into EBUScore with 2+ sections
Today I tried to import the BBO Extractor file into EBUScore using the "Import from XML" feature, but it incorrectly assumed that pairs 1-18 were NS and 19-36 were EW. As there were 18 tables (2 9-table sections), BBO Extractor had NS pairs 1-9 and 19-27, and EW 10-18 and 28-36, so EBUScore returned some very strange results! The only way I could solve the problem was to go into the XML file and manually change the pair numbers around which took a while.
I wouldn't normally bother putting the results into EBUScore but in this case it was needed to make a couple of adjustments. As it turns out, it would have been quicker for me to work out the MP scores myself (including a weighted score!) and then input them directly into the XML file.
Does anyone know how this pair number issue with multiple sections can be avoided?
Comments
I'll look into the possibility of altering the way pair numbers are assigned by BBO Extractor, so that NS pair numbers occupy a single contiguous block, followed by another contiguous block for the EW pair numbers.
Incidentally, in the near future we'll be updating BBOtoXML to enable contracts/scores to be edited, or adjustments to be made, before generating the XML. This would avoid the need to import into EBUScore if the only reason for doing so is to make score adjustments.
That would be great, thanks!
I have occasionally used BBX, but also use Chris Chambers' spreadsheet (& also have used BriAn). I don't wish to go into the relative merits at this time, but on this specific point: Chris Chambers' sheet does not recognise different sections, and treats the whole event as a single section. i.e. it numbers all the N/S pairs from 1 to N, and E/W from N+1 to 2N. So there shouldn't be any issue in importing to EBUScore.
Incidentally, I have very rarely used EBUScore for this purpose, since the output from both tools can be used for Bridgewebs & UMS uploads directly to the EBU. (But i've just checked with an 18-table event this morning, and confirm that it works)
The one issue i have had is the EBUScore limit on table numbers in a single-section event (which i think is currently 30). On a couple of occasions when I've needed to use EBUscore, I've had more than 30 tables. So, I've had to manually create separate files for each section to import into EBUScore, and there re-combine the sections.
It's a challenge to separate the BW file produced by BBX (& would be virtually impossible from Chambers), but it helps a little that it numbers the pairs in separate sections rather than as one large one!
Well, the latest release of BBO Extractor now numbers NS pairs from all sections into a single range and EW pairs from all sections into a single non-overlapping range ! This has been changed as a result of an issue reported with EBUScore at the start of this thread. So, if you had, say, two sections with 10 pairs in each section, the NS pairs for the first section would be 1-10, and for the second section would be 11-20. For the EW sections they would be 21-30 and 31-40 respectively. I guess that means it would still be possible, though tedious, to separate them out manually.
As I said earlier in this thread, we intend to provide a contract/score editing facility in the BBOtoXML utility, which should reduce the need that some people have to import the XML into EBUScore.
Thanks John.
Mine wasn't a suggestion that the old numbering method should be retained. It was intended more as a wry comment that the 'wrong' method did at least have one unintended benefit.
It is much more important that everybody has a means of importing into EBUScore than to make my life marginally easier to divide into sections! (for which the need, i expect, will be very rare). As you observe, it won't be that much more difficult anyway.