Improvement to pattern matching strings within the CSV Import file

[sniffer66]

I posted up a script up to log football scores into Guardian a while ago

viewtopic.php?f=6&t=22924&start=60

It's using a similar process to a script I use to pull in tennis in play stats.

The issue we have is when team\player names in BF differ from those used on other websites i.e Man Utd, Manchester United, Manchester Utd etc or where tennis players have the same surname (the Williams sisters)
To get around that issue you can trim the string in the CSV to something like "Man". But that creates an issue when another Manchester team are playing
Currently you can only use a wildcard character at the start and end of the string in the CSV i.e *Man*.
My suggestion is to allow the use of multiple wildcards for pattern matching

i.e Current = *Manchester*

If you have a match where Manchester City are playing Manchester United the string "Manchester" would end up matching both teams and the goals assigned for each team would be incorrect

Ideally you would pattern match using *Man*Ci* (<any string>Man<any string>Ci<any string>) and *Man*Uni* (<any string>Man<any string>Uni<any string>)

This would ensure 100% matching from CSV to team/player name

[jimibt]

good suggestion, that would help a lot, tho of course, you'd need some sort of split() algo to create those entries and/or an exception table to store known candidates.

along with this, i'd like to raise again the other option that i mentioned a while back, the use of marketid and selectionid. obviously, only really useful if you have access to the betfair data, but would pin it 100%

viewtopic.php?f=20&t=23204

[sniffer66]

good suggestion, that would help a lot, tho of course, you'd need some sort of split() algo to create those entries and/or an exception table to store known candidates.

along with this, i'd like to raise again the other option that i mentioned a while back, the use of marketid and selectionid. obviously, only really useful if you have access to the betfair data, but would pin it 100%

viewtopic.php?f=20&t=23204

Cheers Jim. That split function is already in my posted footy score script. It just doesn't work properly due to the multiple wildcard limitation in the CSV import

[jimibt]

good suggestion, that would help a lot, tho of course, you'd need some sort of split() algo to create those entries and/or an exception table to store known candidates.

along with this, i'd like to raise again the other option that i mentioned a while back, the use of marketid and selectionid. obviously, only really useful if you have access to the betfair data, but would pin it 100%

viewtopic.php?f=20&t=23204

Cheers Jim. That split function is already in my posted footy score script. It just doesn't work properly due to the multiple wildcard limitation in the CSV import

yeah, i can see how it could get complicated pretty quickly... reminds me of zeno's paradox (Dichotomy paradox): https://en.wikipedia.org/wiki/Zeno%27s_paradoxes