Assignment2-Misra-micheleriva
Shrawani Misra
30/03/2022
Github user data analysis: micheleriva
The aim of this project is to analyze github user data of a particular user on GitHub. For this purpose, I make use of the github API for accessing data and subsequently wrangle and analyze the data in R
micheleriva is the Github user handle of Michele Riva. The user profile can be found on https://github.com/micheleriva
1. Description: Senior Architect @nearform · @google GDE · @microsoft MVP · Book Author · International Speaker. Company : @nearform Location: Italy email:ciao@micheleriva.it webiste: https://www.micheleriva.it Twitter handle:@MicheleRivaCode
2. Followers: 469, Following: 442
3. Number of repositories: 49
Project coding
Loading the libraries required in the project
#install.packages("repurrrsive")
#install.packages("tidyverse")
#install.packages("httr")
#install.packages("gh")
#install.packages("ggplot2")
#install.packages("kableExtra")
#install.packages("ggthemes")
#install.packages("treemapify")
#install.packages("ggplotify")
library(repurrrsive)## Warning: package 'repurrrsive' was built under R version 4.1.3library(tidyverse)## Warning: package 'tidyverse' was built under R version 4.1.3## -- Attaching packages --------------------------------------- tidyverse 1.3.1 --## v ggplot2 3.3.5 v purrr 0.3.4
## v tibble 3.1.6 v dplyr 1.0.7
## v tidyr 1.1.4 v stringr 1.4.0
## v readr 2.1.2 v forcats 0.5.1## Warning: package 'ggplot2' was built under R version 4.1.3## Warning: package 'purrr' was built under R version 4.1.3## -- Conflicts ------------------------------------------ tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(httr)## Warning: package 'httr' was built under R version 4.1.3library(gh)## Warning: package 'gh' was built under R version 4.1.3library(ggplot2)
library(kableExtra)## Warning: package 'kableExtra' was built under R version 4.1.3##
## Attaching package: 'kableExtra'## The following object is masked from 'package:dplyr':
##
## group_rowslibrary(ggthemes)## Warning: package 'ggthemes' was built under R version 4.1.3library(treemapify)## Warning: package 'treemapify' was built under R version 4.1.3library(ggplotify)## Warning: package 'ggplotify' was built under R version 4.1.3micheleriva Profile details
Getting profile details by accessing github through personal token (mapped to Shrawani Misra) and setting env variable. Limit=Inf does not set a limit to the data fetched.
Showing micheleriva user’s login, name, repositories and followers in a kable table.
token <- "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"
Sys.setenv(GITHUB_TOKEN = token)
profile_michele <- gh("GET /users/micheleriva",
username = "micheleriva",.limit=Inf)
profile_michele <- tibble(
login = profile_michele$login,
name = profile_michele$name,
public_repos = profile_michele$public_repos,
followers = profile_michele$followers
)
head(profile_michele, n = 50) %>%
kable() %>%
kable_paper(bootstrap_options = "striped", full_width=F)| login | name | public_repos | followers |
|---|---|---|---|
| micheleriva | Michele Riva | 49 | 469 |
Data Collection
Fetching follower data of micheleriva
Fetching follwers of micheleriva using gh package. The same is then input to a dataframe. Later, we make sure there are no null values by making a null list and comparing it with our data using is not null
#Get followers info and populate in dataframe
followers_michele <- gh("/users/micheleriva/followers", .limit = Inf)## i Running gh query## i Running gh query, got 100 records of about 500## i Running gh query, got 200 records of about 500## i Running gh query, got 300 records of about 500## i Running gh query, got 400 records of about 500df_followers_michele <- data.frame(User=character(),
login=character(),
public_repos=integer(),
followers=integer())
null_list <- function(x){map_chr(x, ~{ifelse(is.null(.x), NA, .x)})}
is.not.null <- function(x) !is.null(x)
n <- length(followers_michele)Next, to effectively iterate through every follower, we run a for loop for the entire 469 observations (that is the total number of followers of micheleriva as per his github profile). Getting the login details of a follower, their name, repositories,followers using gh package. Again, we check for null values and make a dataframe
for (i in 1:n)
{
# Login of i follower to fetch further data
login = followers_michele[[i]]$login
# fetch that follower's profile
f_profile <- gh("GET /users/:login", login = login, .limit = Inf)
name = f_profile$name
public_repos =f_profile$public_repos
followers = f_profile$followers
# Making sure there are no null values. Finally, populate in dataframe
if (is.not.null(name) & is.not.null(login) & is.not.null(public_repos)
& is.not.null(followers))
{
df_followers_michele <- rbind(df_followers_michele, data.frame(User=null_list(name),
login = (login),
public_repos=null_list(public_repos),
followers = null_list(followers)))
}
}
head(df_followers_michele, n = 10) %>%
kable() %>%
kable_paper(bootstrap_options = "striped", full_width=F)| User | login | public_repos | followers |
|---|---|---|---|
| Jonas Galvez | galvez | 62 | 427 |
| =Bill.Barnhill | BillBarnhill | 143 | 45 |
| Jiten (Jits) Bhagat | jits | 18 | 127 |
| Giorgio Pomettini | Pomettini | 51 | 75 |
| Andrey Esin | esin | 33 | 6209 |
| Antonio Caputo | tonycaputome | 27 | 40 |
| Daniel Vinciguerra | dvinciguerra | 97 | 133 |
| Fabrizio Guglielmino | guglielmino | 48 | 30 |
| X Code | singerdmx | 15 | 401 |
| Tristan de Cacqueray | TristanCacqueray | 267 | 83 |
Fetching repositories data of micheleriva
Similarly, we do the same for repositories of micheleriva. Fetch the repositories data using gh package, checking for null values and populating data frame. Here, we iteratively run a loop for 49 observations(the total number of repositories of micheleriva)
# Getting repositories info and poopulate it in dataframe
repos_michele <- gh("GET /users/micheleriva/repos", username = "micheleriva",.limit = Inf)
length(repos_michele)## [1] 49df_repoinfo <- data.frame(Repo_Name=character(),
size=integer(),
forks=integer(),
open_issues_count=integer(),
closed_issues_count=integer())
for (i in 1:length(repos_michele))
{
#Details of i repo to fetch further data
name = repos_michele[[i]]$name
size = repos_michele[[i]]$size
created_year = as.integer(substring(repos_michele[[i]]$created_at,1,4))
forks = repos_michele[[i]]$forks_count
open_issues_count = repos_michele[[i]]$open_issues_count
closed_issues_url <-
paste0(repos_michele[[i]]$url,"/issues?state=closed")
closed_issues = gh(closed_issues_url,username = "micheleriva",.limit = Inf)
closed_issues_count = length(closed_issues)
# Populate data to data frame
if (is.not.null(name) & is.not.null(size) & is.not.null(forks)
& is.not.null(created_year)
& is.not.null(open_issues_count) & is.not.null(closed_issues_count))
{
df_repoinfo<-rbind(df_repoinfo, data.frame(Repo_Name = null_list(name),
size = null_list(size),
forks = null_list(forks),
created_year = null_list(created_year),
open_issues_count = null_list(open_issues_count),
closed_issues_count = null_list(closed_issues_count)))
}
}
head(df_repoinfo, n = 15) %>% kable() %>%
kable_paper(bootstrap_options = "striped", full_width=F)| Repo_Name | size | forks | created_year | open_issues_count | closed_issues_count |
|---|---|---|---|---|---|
| aika | 51 | 0 | 2019 | 0 | 0 |
| Aquarium | 5238 | 1 | 2019 | 6 | 13 |
| c-vs-ts-wasm | 41 | 0 | 2018 | 0 | 1 |
| CadregaLisp | 3041 | 0 | 2018 | 0 | 1 |
| caesar-cipher | 240 | 0 | 2019 | 0 | 0 |
| coronablocker | 527 | 2 | 2020 | 1 | 0 |
| cvTracker | 36 | 0 | 2018 | 0 | 0 |
| dat-theme | 2537 | 0 | 2018 | 0 | 1 |
| DNN-Pose-Estimator | 4587 | 1 | 2018 | 0 | 1 |
| editorjs-go | 243 | 2 | 2020 | 1 | 1 |
| ElixirIdenticon | 717 | 0 | 2018 | 0 | 0 |
| Emotion-Detection-ML-Example | 6 | 10 | 2018 | 2 | 1 |
| face-detection | 1769 | 3 | 2018 | 1 | 0 |
| fastify-vite | 3267 | 0 | 2022 | 0 | 0 |
| gauguin | 1256 | 8 | 2020 | 3 | 3 |
Thus our data collection has been done.
##Data Visualization
As we can see, the user micheleriva has been active for many years, so we see which years have seen the most amount of contirbutions from him. To visualize this, we use treemap and see that 2018 is when he has made the most amount of contributions.
df_repoinfo$size <- as.numeric(as.character(df_repoinfo$size))
df_repoinfo$forks <- as.numeric(as.character(df_repoinfo$forks))
df_repoinfo$created_year <- as.numeric(as.character(df_repoinfo$created_year))
df_repoinfo$open_issues_count <- as.numeric(as.character(df_repoinfo$open_issues_count))
df_repoinfo$closed_issues_count <- as.numeric(as.character(df_repoinfo$closed_issues_count))
df_repo_summary <- df_repoinfo %>% group_by(created_year) %>%
summarise(Repo_Count = n())
ggplot(df_repo_summary, aes(area = Repo_Count, fill = created_year, label=created_year)) +
geom_treemap() +
geom_treemap_text( colour = "grey", place = "centre",grow = TRUE)+
scale_fill_viridis_c()+
ggtitle("michelerivas: Amount of Repository contributions distributed over years")
As a developer, it is very easy to have errors and issues even when checking code to production. The count of open issues will give an idea of the repositories that might have issues like these.
Thus, plotting the names of those repositories that have more than 0 open issue, we can see that a few of them stand out, like “krabs”, “vue-product-spinner”, “Aquarium” that have significant amount of open issues. This visualization also tells us that the user micheleriva ran into significant issues for “krabs” and “std” even in 2021 despite having experience, denoting that every repository and their issue may be unique and prior expertise may not always be a solution.
ggplot(data = df_repoinfo %>% filter(open_issues_count!=0), mapping = aes(
x=Repo_Name, y=open_issues_count)) +
geom_point(size=4, colour='mediumseagreen')+
facet_grid(~created_year)+
theme_solarized()+
ggtitle("michelerivas: Repositories with open issue counts")+
xlab("Repository Name") + ylab("Number of Open issues")+
coord_flip()