README.qmd

---
format: gfm
---


# Software development coordination

This repository is used for coordinating all the software development activities by the ForeSITE group.

```{r}
#| label: load-data
#| echo: false
#| warning: false
library(yaml)
library(data.table)

existing_software <- list.files(
    "data", pattern = "ya?ml$", full.names = TRUE
    ) |> 
    lapply(FUN = yaml.load_file, as.named.list = TRUE) |>
    lapply(as.data.table) |>
    rbindlist()

writexl::write_xlsx(x = existing_software, path = "data/software.xlsx")

# Renaming relevant columns
cnames <- c(
    Tool = "tool_name",
    Description = "tool_description",
    Languages = "languages",
    Repo = "github_link",
    Contact = "contact_name",
    Email = "contact_email",
    Links = "docs_link",
    Type = "tool_type",
    Diseases = "relevant_diseases"
)

setnames(existing_software, cnames, names(cnames))

# For the moment, we are only keeping tools that have a url under Links
# or Repo
total_software <- nrow(existing_software)
existing_software <- existing_software[
    grepl("http", Links, ignore.case = TRUE) |
    grepl("http", Repo, ignore.case = TRUE)
]
```

The list includes `r nrow(existing_software)` tools. A complete list featuring completed and work-in-progress tools holds `r total_software` entries. We will be updating this list as we progress.

```{r}
#| label: some-stats
#| echo: false
existing_software[, table(Type)] |>
    as.data.frame(responseName = "N") |>
    setorder(-N) |>
    knitr::kable(caption = "Number of software packages by type")
```

```{r}
#| label: languages
#| echo: false
allowed_languages <- c(
    "R", "C++", "C", "Java", "Python", "Julia", "Matlab", "JavaScript",
    "cli", "sas", "Stata", "AnyLogic"
    ) |> tolower()

listed_languages <- existing_software[, tolower(Languages)] |>
    strsplit(",") |>
    unlist() |>
    trimws() |>
    gsub(pattern = "[.]$|and ", replacement = "")

listed_languages[!listed_languages %in% allowed_languages] <- "Other"
table(listed_languages) |>
    as.data.table() |>
    setnames(c("Language", "N")) |>
    setorder(-N) |>
    knitr::kable(
        caption = "Number of software packages by programming language."
    )
```


# Existing software packages

List of existing software packages that are being used by the ForeSITE group. This list was last updated on `r Sys.Date()`.

```{r}
#| label: existing_software
#| echo: false
# Adding link to GitHub (if it exists)
existing_software[, Repo := fifelse(
    grepl("github", Repo, ignore.case = TRUE),
    sprintf(
        "**GitHub:** [![GitHub last commit](https://img.shields.io/github/last-commit/%s)](%s)",
        gsub(".+github\\.com/", "", Repo),
        Repo
    ),
    ""
)]

# Creating a slug for the link
existing_software[, slug := gsub(" ", "-", tolower(trimws(Tool)))]
existing_software[, slug := gsub("[^a-z0-9-]+", "-", slug)]
existing_software[, slug := gsub("-+", "-", slug)]
existing_software[, slug := gsub("^-|-$", "", slug)]

existing_software[, Description2 := sprintf(
    "%s<br>([Learn more...](profiles.md#%s))",
    Description,
    slug
)]

dat <- existing_software[, .(
    Tool = sprintf("%s (%s)", Tool, Languages),
    Description = sprintf(
        "%s<br>%s",
        Description2,
        Repo
    )
)]

# Adding a number
dat[, N := 1:.N]
dat <- dat[, c("N", "Tool", "Description")]

knitr::kable(dat)
```

```{r}
#| label: create-profiles
#| echo: false
unlink("profiles.md", force = TRUE)
writer <- \(x) {
    cat(x, file = "profiles.md", append = TRUE)
}

writer("# Profiles\n\n")

# Adding a line that specifies that this file (profiles.md)
# is automatically generated and should not be edited manually
writer(
    sprintf(
        "This file is automatically generated and should not be edited manually. Last updated on %s.\n\n",
        Sys.Date()
    )
)

for (i in 1:nrow(existing_software)) {

    # Header
    writer(
        sprintf(
            "## %s\n\n",
            existing_software$Tool[i]
        ))
    
    # Description
    writer(
        sprintf("%s\n\n", existing_software$Description[i])
        )

    # Contact
    writer(
        sprintf(
            "Contact: %s (email: %s)\n\n",
            existing_software$Contact[i],
            existing_software$Email[i]
        )
    )

    # Link
    writer(
        sprintf(
            "Links:<br> %s\n\n",
            existing_software$Links[i]
        )
    )

}
```

```{r}
#| label: wordcloud
#| echo: false
#| fig-dpi: 400
#| fig-width: 8
#| fig-height: 8

tool_names <- dat$Tool

# Removing the part in parentheses
tool_names <- gsub("\\s*\\([^)]+\\)$", "", tool_names)

tool_names <- strsplit(tool_names, " ")
tool_names <- sapply(tool_names, function(tool) {
    nletters <- nchar(tool) |> cumsum()
    nletters <- nletters %/% 20 |> diff() 
    tool[which(nletters == 1)] <- tool[which(nletters == 1)] |> paste0("\n")
    paste(tool, collapse = " ")
})

plot.new()
set.seed(331)

tool_xy = list(
    x = runif(nrow(dat), 0, 1),
    y = runif(nrow(dat), 0, 1),
    cex = 1.5 - runif(nrow(dat), 0, 1),
    col = sample(colors(), nrow(dat))
)

plot.window(xlim = c(-.25, 1.25), ylim = c(-.25, 1.25))
text(
    x = tool_xy$x,
    y = tool_xy$y,
    labels = tool_names,
    cex = tool_xy$cex,
    col = tool_xy$col,
    font = 2,
)
```