docs.qmd

---
title: "Data documentation"
---

Click on each dataset's name to see its description, license, citation, and link to its original source. Code used to create these data can be found [here](https://github.com/ben-domingue/irw/tree/main/data).

**Every dataset here has been modified.** IRW does not redistribute source files: each one is reshaped into the [IRW data standard](standard.qmd) — long format, one row per person-item response, with columns renamed and non-response material dropped. The processing script for a given dataset is the record of exactly what was changed, and is linked above. The license shown beside each description is the **source's** license, which continues to govern reuse of that dataset: where it carries a non-commercial restriction, that restriction travels with IRW's copy.

```{=html}
<div class="panel-tabset-container">
  <button onclick="toggleTabset()">To cite using IRW's R package</button>
  <div class="panel-tabset" style="display: none;">
    <div class="r-command">
      <p>&gt; <code>irw_save_bibtex("gilbert_meta_48")</code></p>
    </div>
    <div class="bibtex-entry">
      <pre>
@misc{gilbert_meta_48,
  doi = {10.7910/DVN/3SF36J},
  url = {https://dataverse.harvard.edu/citation?persistentId=doi:10.7910/DVN/3SF36J},
  author = {Cárdenas, Sergio and Evans, David K.},
  keywords = {Social Sciences},
  title = {Replication Data for: Parent Training and Child Development at Low Cost? Evidence from a Randomized Field Experiment in Mexico},
  publisher = {Harvard Dataverse},
  year = {2023}
}
      </pre>
    </div>
  </div>
</div>

<style>
/* Style for the toggle button */
.panel-tabset-container button {
  margin-bottom: 10px;
  padding: 5px 10px;
  background-color: #007BFF;
  color: white;
  border: none;
  border-radius: 4px;
  cursor: pointer;
}

.panel-tabset-container button:hover {
  background-color: #0056b3;
}

/* Style for the panel content */
.panel-tabset {
  display: none; /* Hide by default */
  border: 1px solid #ddd;
  padding: 10px;
  border-radius: 4px;
  background: #f9f9f9;
  font-family: 'Courier New', Courier, monospace;
}

/* Style for the R command line */
.r-command {
  margin-bottom: 15px;
  color: #333;
  font-size: 14px;
}

.r-command p {
  margin: 0;
  padding: 5px;
  background-color: #f3f3f3;
  border-left: 3px solid #007BFF;
  font-family: monospace;
}

/* Style for the BibTeX block */
.bibtex-entry pre {
  background-color: #f3f3f3;
  padding: 10px;
  border-radius: 5px;
  border: 1px solid #ccc;
  font-family: monospace;
  font-size: 14px;
  overflow-x: auto;
}
</style>

<script>
  function toggleTabset() {
    const tabset = document.querySelector('.panel-tabset');
    tabset.style.display = (tabset.style.display === 'none' || tabset.style.display === '') ? 'block' : 'none';
  }
</script>

```

```{r}
library(dplyr)
library(purrr)
library(readr)
library(stringr)
library(glue)
library(redivis)

biblio <- redivis$user("datapages")$dataset("irw_meta:bdxt:latest")$table("biblio")$to_tibble()

# Load datasets
data_index <- biblio |>
  rename(dataset = `table`, url = URL__for_data_, license = `Derived_License`,
         description = Description, reference = Reference_x, doi = DOI__for_paper_) |>
  mutate(dataset = str_remove(dataset, ".R.ata$")) |>
  arrange(dataset)
```

```{=html}
<input type="text" id="searchBar" placeholder="Search datasets..." style="width: 100%; padding: 10px; margin-bottom: 20px; font-size: 16px;">
<p id="search-placeholder" style="color: #888;">Start typing to search across all datasets by name, description, or reference.</p>
<script>
function searchDatasets() {
  var input = document.getElementById('searchBar');
  var filter = input.value.toLowerCase().trim();
  var items = document.getElementsByClassName('dataset-item');
  var table = document.getElementById('dataset-table-wrapper');
  var placeholder = document.getElementById('search-placeholder');

  if (filter === '') {
    table.style.display = 'none';
    placeholder.style.display = '';
    return;
  }
  table.style.display = '';
  placeholder.style.display = 'none';

  for (var i = 0; i < items.length; i++) {
    var text = items[i].textContent || items[i].innerText;
    items[i].style.display = text.toLowerCase().indexOf(filter) > -1 ? '' : 'none';
  }
}

function sleep(ms) {
  return new Promise(resolve => setTimeout(resolve, ms));
}

async function getCitation(bibtex, buttonId, targetId) {
  const button = document.getElementById(buttonId);
  button.disabled = true;
  button.textContent = 'Loading...';

  try {
    await sleep(100); // Small delay for UI feedback
    
    const target = document.getElementById(targetId);
    const container = document.createElement('div');
    container.className = 'citation-container';

    const citationText = document.createElement('pre');
    citationText.className = 'citation-pre';
    citationText.textContent = bibtex;

    const copyButton = document.createElement('button');
    copyButton.className = 'copy-button';
    copyButton.innerHTML = '📋 Copy';
    copyButton.onclick = () => {
      navigator.clipboard
        .writeText(bibtex)
        .then(() => {
          copyButton.innerHTML = '✓ Copied!';
          setTimeout(() => (copyButton.innerHTML = '📋 Copy'), 2000);
        })
        .catch(() => (copyButton.innerHTML = '❌ Error'));
    };

    target.innerHTML = '';
    container.appendChild(citationText);
    container.appendChild(copyButton);
    target.appendChild(container);
    target.style.display = 'block';
  } catch (error) {
    console.error('Error displaying citation:', error);
    const target = document.getElementById(targetId);
    target.textContent = 'Error displaying citation. Please try again.';
    target.style.display = 'block';
  } finally {
    button.disabled = false;
    button.textContent = 'Get BibTex';
  }
}

document.getElementById('searchBar').addEventListener('input', searchDatasets);
</script>

<style>
.citation-container {
  position: relative;
  margin-top: 10px;
  padding: 10px;
  background-color: #f5f5f5;
  border-radius: 4px;
  min-height: 50px;
}
.citation-pre {
  font-family: monospace;
  white-space: pre-wrap;
  word-wrap: break-word;
  margin: 0;
  padding: 20px 10px 10px 10px;
  max-width: 100%;
  overflow-x: auto;
}
.copy-button {
  position: absolute;
  top: 5px;
  right: 5px;
  background-color: #ffffff;
  border: 1px solid #dddddd;
  border-radius: 4px;
  padding: 4px 8px;
  cursor: pointer;
  font-size: 12px;
  z-index: 1;
}
.copy-button:hover {
  background-color: #f0f0f0;
}
.citation-button {
  background-color: #007bff;
  color: white;
  border: none;
  padding: 5px 10px;
  border-radius: 4px;
  cursor: pointer;
  margin-top: 10px;
}
.citation-button:disabled {
  background-color: #cccccc;
  cursor: not-allowed;
}

.dataset-table {
  width: 100%;
  table-layout: fixed;
  border-collapse: collapse;
  font-size: 15px;
}
.dataset-table th, .dataset-table td {
  padding: 10px 12px;
  border-bottom: 1px solid #e2e2e2;
  text-align: left;
  vertical-align: top;
  overflow-wrap: break-word;
}
.dataset-table th {
  border-bottom: 2px solid #ccc;
  font-weight: 600;
}
.dataset-table tr:hover {
  background-color: #f7f7f7;
}
.dataset-table th:nth-child(1), .dataset-table td:nth-child(1) { width: 16%; }
.dataset-table th:nth-child(2), .dataset-table td:nth-child(2) { width: 58%; }
.dataset-table th:nth-child(3), .dataset-table td:nth-child(3) { width: 8%; }
.dataset-table th:nth-child(4), .dataset-table td:nth-child(4) { width: 18%; }
</style>
```

```{r}
#| results: asis

# A dataset's license is recorded as a name; the license itself lives elsewhere.
# Reusers need the terms, not the label -- and where a source is offered under a
# non-commercial or share-alike license, its own conditions ask that anyone
# redistributing it link to that license rather than merely name it. So the
# name is rendered as a link wherever there is a canonical text to point at.
# Values with no such text -- "Custom", "Permission via Email" -- render as
# plain text, which is the honest rendering: there is nothing to link to. So
# does "CC BY-NC-ND", deliberately: it is recorded without a version, and
# guessing 4.0 would put a wrong link on the one license whose terms most need
# reading.
license_urls <- c(
  "CC0 1.0"         = "https://creativecommons.org/publicdomain/zero/1.0/",
  "CC BY 4.0"       = "https://creativecommons.org/licenses/by/4.0/",
  "CC BY-SA 4.0"    = "https://creativecommons.org/licenses/by-sa/4.0/",
  "CC BY-NC 4.0"    = "https://creativecommons.org/licenses/by-nc/4.0/",
  "CC BY-NC 3.0"    = "https://creativecommons.org/licenses/by-nc/3.0/",
  "CC BY-NC-SA 4.0" = "https://creativecommons.org/licenses/by-nc-sa/4.0/",
  "CC BY-NC-SA 3.0" = "https://creativecommons.org/licenses/by-nc-sa/3.0/",
  "ODbL 1.0"        = "https://opendatacommons.org/licenses/odbl/1-0/",
  "ODC-By"          = "https://opendatacommons.org/licenses/by/1-0/",
  "MIT"             = "https://opensource.org/license/mit",
  "GPL-2.0"         = "https://www.gnu.org/licenses/old-licenses/gpl-2.0.html",
  "GPL-3.0"         = "https://www.gnu.org/licenses/gpl-3.0.en.html",
  "LGPL-3.0"        = "https://www.gnu.org/licenses/lgpl-3.0.en.html"
)

# A few rows carry two comma-separated licenses, so link each part separately.
license_html <- function(x) {
  if (is.null(x) || length(x) == 0 || is.na(x) || !nzchar(trimws(x))) return("")
  parts <- trimws(strsplit(x, ",")[[1]])
  parts <- parts[nzchar(parts)]
  linked <- vapply(parts, function(part) {
    url <- unname(license_urls[part])
    if (is.na(url)) part
    else glue('<a href="{url}" target="_blank" rel="license noopener">{part}</a>')
  }, character(1))
  glue("({paste(linked, collapse = ', ')})")
}

row <- function(ds) {
  lic <- license_html(ds$license)
  bibtex <- ds$BibTex  # Use BibTex column directly
  escaped_bibtex <- if (!is.na(bibtex) && bibtex != "") gsub("'", "\\\\'", bibtex) else ""

  citation_button <- if (escaped_bibtex != "") {
    button_id <- paste0("cite-button-", make.names(ds$dataset))
    target_id <- paste0("cite-text-", make.names(ds$dataset))
    glue(
      '<button id="{button_id}" class="citation-button" onclick="getCitation(`{escaped_bibtex}`, \'{button_id}\', \'{target_id}\')">
        Get BibTex
      </button>
      <div id="{target_id}" class="citation-text"></div>'
    )
  } else ""

  glue(
    '
    <tr class="dataset-item">
      <td><strong>{ds$dataset}</strong></td>
      <td>{ds$description} {lic} <i>{ds$reference}</i></td>
      <td><a href="{ds$url}" target="_blank">link</a></td>
      <td>{citation_button}</td>
    </tr>
    '
  )
}

rows <- transpose(data_index) |> map(row) |> paste(collapse = "\n")
cat(glue(
  '
  <div id="dataset-table-wrapper" style="display: none;">
  <table class="dataset-table">
    <thead>
      <tr><th>Dataset</th><th>Description</th><th>Source</th><th>Citation</th></tr>
    </thead>
    <tbody>
    {rows}
    </tbody>
  </table>
  </div>
  '
))
```