[Bug 48270] New: Rebelbetting: Logging into Betfair results in incorrect large numbers

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=48270

            Bug ID: 48270
           Summary: Rebelbetting: Logging into Betfair results in
                    incorrect large numbers
           Product: Wine
           Version: 4.21
          Hardware: x86-64
               URL: http://rebelbetting.com/files/RebelBettingSetup.exe
                OS: Linux
            Status: NEW
          Keywords: download
          Severity: normal
          Priority: P2
         Component: -unknown
          Assignee: wine-bugs at winehq.org
          Reporter: imwellcushtymelike at gmail.com
      Distribution: Ubuntu

Created attachment 65937
  --> https://bugs.winehq.org/attachment.cgi?id=65937
Screenshot

Running Rebelbetting BEFORE (or without) logging into Betfair gives normal,
sensible results.

After logging into Betfair some (but not all) of the calculated numbers are
insanely high, either positive or negative. Screenshot attached.

The cause seems to be from a stake of 2^31 which appears most of the time, but
not always. It also seems to be avoided if two or none of the participants is
Betfair.

Betfair's API uses JSON.

I wasn't sure if this was a bug in the software so I've been running it for a
while on Linux and Windows 10 ― it definitely only occurs in Linux.
WARNING: running Rebelbetting twice with the same log-in details will get your
account suspended for at least one hour.

I don't think this is a regression as this has happened for as long as I can
remember.

Account for both Rebelbetting and Betfair required to test.

To get this far:
1. "winetricks dotnet45" (Bug 48014)
Note there are other bugs which may be encountered (see the AppDB).

Running in 32-bit wineprefix. I haven't yet tested in a 64-bit wineprefix to
see if the numbers change, but I think the app is 32-bit anyway.

Note that this is similar, but subtly different to Bug 38944. The fix there may
offer a clue following a very long, very tedious regression test, but I'm
saving that as a very last resort (and may not even be related).

-- 
Do not reply to this email, post in Bugzilla using the
above URL to reply.
You are receiving this mail because:
You are watching all bug changes.