Parse Your Reputation Page From Any of the Stack Exchange Sites

It is very easy to view a detailed account of your reputation at any of the Stack Exchange sites by visiting http://"sitename"/reputation (obviously substituting "sitename" for the actual site of interest, for example, http://stackoverflow.com/reputation). However, that format is not very user-friendly if you want to do any analysis with it. This function parses that page into an R data.frame.

mySOreputation(rep_file)

Arguments

rep_file	The path to a text version of your reputation page. Windows and Linux users can copy the text on the page with select all + copy, and simply use "clipboard" instead of saving the contents to a local file.

References

Values for the "actions" variable determined after visiting http://meta.stackexchange.com/a/43005/214964. There is one value not mentioned at that page, coded as action_id == 99 and action == Bonus that corresponds to the bonus that a user gets when they have above a certain reputation and are active on multiple Stack Exchange sites.

Author

Paul Hiemstra provided the base parser. Built upon by Ananda Mahto.

Examples

## This is a real reputation file,
##    but the "question_id" variable is
##    made up.
rep_file <- system.file("soreputation.txt", package = "SOfun")
readLines(rep_file, 15)
#>  [1] "total votes: 3092"             "   2  9170090  (10)"          
#>  [3] "   2  8533006  (10)"           "   2  8783724  (10)"          
#>  [5] "   2  5173844  (10)"           "-- 2012-03-15 rep +40   = 41" 
#>  [7] "   2  9199717  (10)"           "   2  7256057  (10)"          
#>  [9] "-- 2012-03-16 rep +20   = 61"  "   2  6145733  (10)"          
#> [11] "   1  7509549  (15)"           "   2  3788812  (10)"          
#> [13] "   2  1347644  (10)"           "-- 2012-03-17 rep +45   = 106"
#> [15] "-- 2012-03-25 rep 0     = 106"
mydf <- mySOreputation(rep_file = rep_file)
head(mydf, 15)
#>    action_id question_id rep_change       date             actions
#> 1          2     9170090         10 2012-03-15              Upvote
#> 2          2     8533006         10 2012-03-15              Upvote
#> 3          2     8783724         10 2012-03-15              Upvote
#> 4          2     5173844         10 2012-03-15              Upvote
#> 5          2     9199717         10 2012-03-16              Upvote
#> 6          2     7256057         10 2012-03-16              Upvote
#> 7          2     6145733         10 2012-03-17              Upvote
#> 8          1     7509549         15 2012-03-17  YourAnswerAccepted
#> 9          2     3788812         10 2012-03-17              Upvote
#> 10         2     1347644         10 2012-03-17              Upvote
#> 11         1     6964418         15 2012-03-28  YourAnswerAccepted
#> 12        16     5000073          2 2012-04-07        EditApproved
#> 13         1     1560102         15 2012-04-10  YourAnswerAccepted
#> 14         1     1016692          2 2012-04-11 AnswerAcceptedByYou
#> 15         2     1654001          5 2012-04-11              Upvote
str(mydf)
#> 'data.frame':	2088 obs. of  5 variables:
#>  $ action_id  : Factor w/ 9 levels "1","2","3","4",..: 2 2 2 2 2 2 2 1 2 2 ...
#>  $ question_id: int  9170090 8533006 8783724 5173844 9199717 7256057 6145733 7509549 3788812 1347644 ...
#>  $ rep_change : num  10 10 10 10 10 10 10 15 10 10 ...
#>  $ date       : Date, format: "2012-03-15" "2012-03-15" ...
#>  $ actions    : Factor w/ 11 levels "AnswerAcceptedByYou",..: 3 3 3 3 3 3 3 2 3 3 ...
plot(mydf$date, cumsum(mydf$rep_change))

if (FALSE) {
library(xts)
mydfx <- xts(mydf$rep_change, mydf$date)
apply.monthly(mydfx, sum)
plot(apply.monthly(mydfx, sum))
}