This retrieves per-project pageview counts from January 2008 to July 2016. These counts are calculated using the 'legacy' (read: old) model, which overcounts due to its inclusion of web-crawlers and similar automata.
old_pageviews(
project = "en.wikipedia",
platform = "all",
granularity = "daily",
start = "2013100100",
end = "2015100100",
reformat = TRUE,
...
)the name of the project, structured as [language_code].[project]
(see the default).
The platform the pageviews came from; one or more of "all", "desktop" or "mobile". Set to "all" by default.
the granularity of data to return; do you want hourly, daily or monthly counts? Set to "daily" by default.
the start YYYYMMDDHH of the range you want to cover. This can be
easily grabbed from R date/time objects using pageview_timestamps
the end YYYYMMDDHH of the range you want to cover. NULL by default, meaning
that it returns 1 day/hour of data (depending on the value passed to granularity).
Whether to reformat the results as a data.frame or not. TRUE by default.
further arguments to pass to httr's GET.
top_articles for the top articles per project in a given date range,
project_pageviews for per-project pageviews under the new definition,
and article_pageviews for per-article pageviews.
# Basic call
enwiki_2013_2015_old <- old_pageviews()
# Break it down to hourly
old_enwiki_hourly <- old_pageviews(granularity = "hourly", end = "2013110100")
Run the code above in your browser using DataLab