git2r_post.md

Install package

install_github("ropensci/git2r")

Load libary

library("git2r")
library("plyr")

Repositories with activity during hackathon

repos <- c("ropensci/git2r", "ropensci/docs", "ropensci/testdat", "ropensci/apistatus", 
    "ropensci/reproducibility-guide", "ropensci/EML", "ropensci/togeojson", 
    "ropensci/rfigshare", "ropensci/dependencies", "ropensci/plotly", "ropensci/paleobioDB", 
    "ropensci/taxize", "ropensci/ecoretriever", "ropensci/hackathon", "ropensci/elife", 
    "ramnathv/rMaps", "ScienceToolbox/code_citations")

Clone repositories

hackathon <- lapply(repos, function(repo) {
    clone(paste0("https://github.com/", repo, ".git"), paste0("hackathon/", 
        repo))
})

Get commit data

df <- lapply(hackathon, function(repo) {
    cbind(basename(workdir(repo)), contributions(repo, breaks = "day", by = "author"))
})
df <- do.call("rbind", df)
df <- df[df$when > as.Date("2014-03-30") & df$when < as.Date("2014-04-02"), ]
names(df)[1] <- "repo"

Synonymise authors

res <- gsub("EDiLD", "Eduard Szöcs", df$author)
res <- gsub("davharris", "David J. Harris", res)
res <- gsub("Hilary|Hilary Parker", "Hilary Parker", res)
res <- gsub("iamciera", "Ciera Martinez", res)
res <- gsub("Jeff Hollister", "Jeffrey W. Hollister", res)
res <- gsub("jmollica", "Jay Mollica", res)
res <- gsub("Jeff Hollister|Jeffrey W Hollister", "Jeffrey W. Hollister", res)
res <- gsub("anyab", "Anya Bida", res)
res <- gsub("LucianoFSgarbi", "Luciano F. Sgarbi", res)
res <- gsub("mfenner", "Martin Fenner", res)
res <- gsub("Sara_Varela", "Sara Varela", res)
df$author <- res

df <- ddply(df, .(repo, author), summarise, n = sum(n)) # summarise across dates

Plot

library(ggplot2)
ggplot(df, aes(repo, author)) + geom_tile(aes(fill = n), colour = "white", binwidth = 3) + 
    scale_fill_gradient(low = "white", high = "steelblue") + theme_grey(base_size = 16) + 
    labs(x = "", y = "") + scale_x_discrete(expand = c(0, 0)) + scale_y_discrete(expand = c(0, 
    0)) + theme(axis.ticks = element_blank(), axis.text.x = element_text(size = 12, 
    hjust = 1, colour = "grey50", angle = 45), panel.grid.major = element_blank(), 
    panel.grid.minor = element_blank(), panel.border = element_blank(), panel.background = element_rect(fill = "#F8F8F8", 
        colour = NA))

git2r_post.Rmd

Install package

```{r eval=FALSE}
install_github("ropensci/git2r")
```

Load libary

```{r}
library("git2r")
library("plyr")
```

Repositories with activity during hackathon

```{r}
repos <- c('ropensci/git2r', 'ropensci/docs', 'ropensci/testdat', 'ropensci/apistatus','ropensci/reproducibility-guide', 'ropensci/EML', 'ropensci/togeojson','ropensci/rfigshare', 'ropensci/dependencies', 'ropensci/plotly','ropensci/paleobioDB','ropensci/taxize', 'ropensci/ecoretriever','ropensci/hackathon','ropensci/elife','ramnathv/rMaps','ScienceToolbox/code_citations')
```

Clone repositories

```{r cache=TRUE}
hackathon <- lapply(repos, function(repo) {
  clone(paste0('https://github.com/', repo, '.git'),
        paste0('hackathon/', repo))
})
```

Get commit data

```{r cache=TRUE}
df <- lapply(hackathon, function(repo) {
  cbind(basename(workdir(repo)), contributions(repo, breaks='day', by='author'))
})
df <- do.call('rbind', df)
df <- df[df$when > as.Date('2014-03-30') & df$when < as.Date('2014-04-02'),]
names(df)[1] <- 'repo'
```

Synonymise authors

```{r}
res <- gsub("EDiLD", "Eduard Szöcs", df$author)
res <- gsub("davharris", "David J. Harris", res)
res <- gsub("Hilary|Hilary Parker", "Hilary Parker", res)
res <- gsub("iamciera", "Ciera Martinez", res)
res <- gsub("Jeff Hollister", "Jeffrey W. Hollister", res)
res <- gsub("jmollica", "Jay Mollica", res)
res <- gsub("Jeff Hollister|Jeffrey W Hollister", "Jeffrey W. Hollister", res)
res <- gsub("anyab", "Anya Bida", res)
res <- gsub("LucianoFSgarbi", 'Luciano F. Sgarbi', res)
res <- gsub("mfenner", 'Martin Fenner', res)
res <- gsub("Sara_Varela", "Sara Varela", res)
df$author <- res

df <- ddply(df, .(repo, author), summarise, n = sum(n)) # summarise across dates
```

Plot

```{r}
library(ggplot2)
ggplot(df, aes(repo, author)) + 
  geom_tile(aes(fill = n), colour = "white", binwidth = 3) + 
  scale_fill_gradient(low = "white", high = "steelblue") + 
  theme_grey(base_size = 16) + 
  labs(x = "", y = "") + 
  scale_x_discrete(expand = c(0,0)) + 
  scale_y_discrete(expand = c(0, 0)) + 
  theme(axis.ticks = element_blank(), 
        axis.text.x = element_text(size = 12, hjust = 1, colour = "grey50", angle = 45), 
        panel.grid.major = element_blank(), 
        panel.grid.minor = element_blank(), 
        panel.border = element_blank(),
        panel.background = element_rect(fill = "#F8F8F8", 
            colour = NA))
```

Nice work! Where is this post headed?