Begin work on biergarten data generator pipeline

pipeline/includes/data_downloader.h

@@ -0,0 +1,111 @@
+/**
+ * @file data_downloader.h
+ * @brief Download geographic data from GitHub repositories using libcurl.
+ *
+ * Provides functionality to fetch JSON data from GitHub using libcurl, with
+ * support for commit-based versioning to ensure reproducible builds. Downloads
+ * are cached to avoid repeated network requests.
+ *
+ * Example usage:
+ * @code
+ *   DataDownloader downloader;
+ *   std::string jsonPath = downloader.DownloadCountriesDatabase(
+ *       "/tmp/countries-data.json",  // local cache path
+ *       "c5eb7772"                    // optional commit hash or HEAD
+ *   );
+ *   // Now use jsonPath with JsonLoader::LoadWorldCities(jsonPath, db)
+ * @endcode
+ */
+
+#ifndef DATA_DOWNLOADER_H
+#define DATA_DOWNLOADER_H
+
+#include <stdexcept>
+#include <string>
+
+/**
+ * @class DataDownloader
+ * @brief Manages downloading and caching of geographic data from GitHub.
+ *
+ * This class encapsulates libcurl networking operations for reproducible
+ * data fetching. All methods are non-blocking and synchronous.
+ *
+ * @note Requires libcurl to be available at runtime.
+ * @note GitHub raw content CDN is used for efficient downloads.
+ */
+class DataDownloader {
+public:
+  /**
+   * @brief Default constructor.
+   *
+   * Initializes the downloader without any specific state. The downloader
+   * is ready to use immediately.
+   */
+  DataDownloader();
+
+  /**
+   * @brief Destructor.
+   *
+   * Cleans up any resources. No explicit cleanup needed beyond destruction.
+   */
+  ~DataDownloader();
+
+  /**
+   * @brief Download the countries+states+cities JSON database from GitHub.
+   *
+   * Downloads the geographic data from the
+   * dr5hn/countries-states-cities-database repository. If the file already
+   * exists at cachePath, it is used directly without downloading again.
+   *
+   * The download URL format is:
+   * @verbatim
+   * https://raw.githubusercontent.com/dr5hn/countries-states-cities-database/
+   * {commit}/json/countries+states+cities.json
+   * @endverbatim
+   *
+   * @param cachePath   Local filesystem path where the JSON file should be
+   * stored. If the file already exists, download is skipped.
+   * @param commit      Git commit hash or branch name (default: "c5eb7772").
+   *                    Examples: "HEAD", "main", "c5eb7772",
+   * "c5eb7772225f6b1802a54f39adb8c73464a85be1a"
+   *
+   * @return            The file path where JSON was saved (same as cachePath).
+   *
+   * @throws std::runtime_error if:
+   *         - Network download fails
+   *         - File cannot be written to cachePath
+   *         - Commit hash is invalid (404 on GitHub)
+   *
+   * Example with default commit (stable v2026-03-28):
+   * @code
+   *   std::string path =
+   * downloader.DownloadCountriesDatabase("/tmp/data.json");
+   * @endcode
+   *
+   * Example with custom commit:
+   * @code
+   *   std::string path = downloader.DownloadCountriesDatabase(
+   *       "/tmp/data.json",
+   *       "main"  // Download latest from main branch
+   *   );
+   * @endcode
+   */
+  std::string DownloadCountriesDatabase(
+      const std::string &cachePath,
+      const std::string &commit = "c5eb7772" // Stable commit: 2026-03-28 export
+  );
+
+private:
+  /**
+   * @brief Check if a file already exists at the given path.
+   *
+   * Used internally to implement cache-hit logic. No download occurs if
+   * the file already exists.
+   *
+   * @param filePath  Path to check.
+   * @return          True if file exists and is readable, false otherwise.
+   */
+  bool FileExists(const std::string &filePath) const;
+};
+
+#endif // DATA_DOWNLOADER_H