ValueError: could not convert string to float: '&' about rpscrape HOT 4 CLOSED

I got the same issue this morning when scraping 28/11/2020 gb results, but not when scraping 2/12/2020 gb results. Looking at the csv, the error seems to happen at approximately the 2.25 Newbury, which appears to have a database bug. If you check the result of that race, there are a number of horses which have '&' as their name so it seems to be an issue on RP's side annoyingly (but if there's a way round on our side that would be great).

from rpscrape.

Python's not my language, I'm only using this to scrape and export the csv to use in R, so this may not be the correct approach - however it seems to have worked.

On a branch, I've changed the fraction_to_decimal function. It now has or fraction == '&' after the no odds statement.

def fraction_to_decimal(fractions):
    decimal = []
    for fraction in fractions:
        if fraction == '' or fraction == 'No Odds' or fraction == '&':
            decimal.append('')
        elif 'evens' in fraction.lower() or fraction.lower() == 'evs':
            decimal.append('2.00')
        else:
            decimal.append('{0:.2f}'.format(float(fraction.split('/')[0]) / float(fraction.split('/')[1]) + 1.00))

    return decimal

from rpscrape.

Yep, I've just checked this on the aforementioned 28/11/20 data and it works a treat. I was expecting to have a bunch of horses with '&' as their name, but having checked the csv file that doesn't happen. It seems to just skip those horses where the errors have occurred. Thanks for uploading your changes.

from rpscrape.

Assuming this was a one off edge case and fixed.

from rpscrape.