Wallet Manager

Description

Downloads wallet balance from multiple sources and calculate value in chosen currency based on actual exchange rates. Function is limited to dictionary api.dict plus fiat-fiat exchange rates.

Usage

wallet_manager(market.sources = NULL, blockchain.sources = NULL,
  manual.sources = NULL, min_amount = 1e-04, antispam_interval = 10,
  api.dict = NULL, verbose = getOption("Rbitcoin.verbose", 0),
  value_calc = TRUE, value_currency = "USD", value_currency_type = NULL,
  rate_priority, transfer_currency_pair = c(crypto = "BTC", fiat = "USD"),
  archive_write = FALSE, archive_read = FALSE)

Arguments

Value

data.table object with wallet information in denormilized structure. Number of columns depends on value_calc param, when FALSE then columns related to the value will not be returned. When launch with wallet_read=TRUE then all historical archived wallet statuses will be returned. Field wallet_id is a processing batch id and also the timestamp of single wallet manager processing as integer in Unix time format.

Wallet archive

To be able to track wallet assets value over time user needs to use archive_write=TRUE. It will archive wallet manager result data.table to wallet_archive.rds file in not encrypted format (not a plain text also), sensitive data like amount and value will be available from R by readRDS("wallet_archive.rds"). This can be used to correct/manipulate archived data or union the results of the wallet manager performed on different machines by readRDS(); rbindlist(); saveRDS(). Setting archive_write=FALSE and archive_read=TRUE will skip processing and just load the archive, same as readRDS(). You should be aware the archive file will be growing over time, unless you have tons of sources defined or you scheduled wallet_manager every hour or less you should not experience any issues because of that. In case of the big size of archived rds file you can move data to database, wrap function into database archiver function and query full archive from database only for for plotting.

Exchange rates

Exchange rates will be downloaded from different sources. Fiat-fiat rates will be sourced from yahoo finance, if yahoo would not be available then also fiat-fiat rate cannot be calculated. Rates for cryptocurrencies will be downloaded from market's tickers according to rate_priority and currency pairs available in api.dict. Currency type (crypto or fiat) is already defined in getOption("Rbitcoin.ct.dict"), can be edited for support other/new currency.
Markets used for crypto rates are defined by rate_priority as vector of market names in order of its priority from highest to lowest. User need to chose own trusted exchange rates providers and keep in mind to update rate_priority parameter when necessary. As we recently seen the mtgox after death was still spreading the public API data and any system which sources data from them would be affected, so the control over the source for exchange rates needs to be maintained by user. In case of calculation crypto rate for a currency pair which is not available in api.dict then transfer_currency_pair will be used to get indirect exchange rate. Example: exchange rate for NMC-GBP will be computed as NMC-BTC-USD-GBP using the default transfer_currency_pair and current api.dict. The process was not exhaustively tested, you can track all the exchange rates used by setting options(Rbitcoin.archive_exchange_rate=0) for saveRDS(), options(Rbitcoin.archive_exchange_rate=1) for write.table(sep=",", dec=".") or options(Rbitcoin.archive_exchange_rate=2) for write.table(sep=";", dec=","). This option will append the data to exchange_rate_archive rds/csv file in working directory.

NA measures

In case of missing exchange path (direct and indirect through transfer_currency_pair based on api.dict) between the currency in the wallet and the value_currency the NA will be provided to value for that currency. Any errors while downloading wallet data or exchange rates data will also result NA measure. Be sure to avoid NA measures: for unavailable sources you can provide amounts as manual source, for not supported alt cryptocurrencies precalculate its value to supported currency and provide as manual source. While plotting wallet_manager data any wallet batches which contain at least one NA measure will be omitted from plot.

Schedule wallet tracking

User may consider to schedule execution of the function with archive_write=TRUE for better wallet assets tracking over time. Schedule can be setup on OS by run prepared R script with wallet_manager function execution. In case of scheduling also plot of wallet manager use archive_read=TRUE and add Rbitcoin.plot function execution.

Troubleshooting

In case of the issues with this function verify if all of the sources are returning correct data, use blockchain.api.process and market.api.process functions. Possible sources for wallet data: market api, blockchain api, manually provided. Possible sources for exchange rate data: market tickers, yahoo (see references). If all sources works and issue still occurs please report. Additionally you can always use verbose argument to print processing informations.

References

https://code.google.com/p/yahoo-finance-managed/wiki/csvQuotesDownload

See Also

Rbitcoin.plot, blockchain.api.process, market.api.process, antiddos

Examples

## Not run: 
## define source
# define wallets on markets
market.sources <- list(
  list(market = 'bitstamp', currency_pair = c('BTC', 'USD'),
       client_id = '', key = '', secret = ''),
  list(market = 'btce', currency_pair = c('LTC', 'USD'),
       key = '', secret = ''),
  list(market = 'btce', currency_pair = c('LTC', 'USD'),
       key = '', secret = ''), #multiple accounts on same market possible
  list(market = 'kraken', currency_pair = c('BTC', 'EUR'),
       key = '', secret = '')
)
# define wallets on blockchain
blockchain.sources <- list(
  list(address = ''),
  list(address = '')
)
# define wallets manually
manual.sources <- list(
  list(location = 'while transferring',
       currency = c('BTC','LTC'),
       amount = c(0.08, 0)),
  # manually provided value as workaround for bitstamp api unavailability captcha bug
  list(location = 'bitstamp',
       location_type = 'market'
       currency = c('USD','BTC'),
       amount = c(50,0.012))
)

## launch wallet manager with no value calculation
wallet_dt <- wallet_manager(market.sources,
                            blockchain.sources,
                            manual.sources,
                            value_calc = FALSE)
print(wallet_dt)

## launch wallet manager
wallet_dt <- wallet_manager(
  market.sources = market.sources,
  blockchain.sources = blockchain.sources,
  manual.sources = manual.sources,
  value_currency = 'GBP',
  rate_priority = c('bitstamp','kraken','bitmarket','btce')
  archive_write = TRUE
)
print(wallet_dt)

# export to excel/google spreadsheet
setkey(wallet_dt,wallet_id,currency) #sort
write.table(wallet_dt, "clipboard", sep="\t", row.names=FALSE, na = "")
# now go to excel or google spreadsheet and use "paste" from clipboard

# aggregate measures by currency and type
wallet_dt[,list(amount = sum(amount, na.rm=T),
                value = sum(value, na.rm=T)),
           by = c('wallet_id','currency','value_currency')
           ][order(wallet_id,currency,value_currency)]
# aggregate value by location and type
wallet_dt[,list(value = sum(value, na.rm=T)),
           by = c('wallet_id','location_type','location')
           ][order(wallet_id,location_type,location)]

# send to plot
wallet_dt <- wallet_manager(archive_write=F, archive_read=T)
Rbitcoin.plot(wallet_dt)

# discard processing batch, by id, from wallet archive (will omit on plot)
dt <- readRDS("wallet_archive.rds")
dt[wallet_id==1390000000,`:=`(amount = NA_real_, value = NA_real_)]
saveRDS(dt, "wallet_archive.rds")

# To track exchange rates used set option Rbitcoin.archive_exchange_rate
options(Rbitcoin.archive_exchange_rate=0)
wallet_dt <- wallet_manager(market.sources,
                            blockchain.sources,
                            manual.sources = manual.sources,
                            rate_priority = c('bitstamp','kraken','bitmarket','btce')
                            archive_write = TRUE)
# all exchange rate data as dt
dt <- readRDS("exchange_rate_archive.rds")
# last exchange rate table as dt
dt <- readRDS("exchange_rate_archive.rds")[value_rate_id==max(value_rate_id)]
# save to csv
write.table(dt, "exchange_rate_archive.csv",
            row.names=FALSE,quote=FALSE,append=FALSE,col.names=TRUE,
            sep=";", dec=",")

## End(Not run)

Results