Initial importv0.1.0

1 files changed, 72 insertions, 0 deletions

diff --git a/src/csv.cpp b/src/csv.cpp
new file mode 100644
index 0000000..32b9591
--- /dev/null
+++ b/src/csv.cpp
@@ -0,0 +1,72 @@
+#include "csv.hpp"
+
+#include <cstddef>
+
+namespace cepis::detail {
+
+std::vector<std::vector<std::string>> parse_csv(std::string_view text) {
+  if (text.compare(0, 3, "\xEF\xBB\xBF") == 0) {
+    text.remove_prefix(3);
+  }
+
+  std::vector<std::vector<std::string>> rows;
+  std::vector<std::string> row;
+  std::string field;
+  bool in_quotes = false;
+  bool row_has_content = false;
+
+  const auto end_row = [&] {
+    if (!row_has_content) {
+      field.clear();
+      row.clear();
+      return;
+    }
+    row.push_back(std::move(field));
+    field.clear();
+    rows.push_back(std::move(row));
+    row.clear();
+    row_has_content = false;
+  };
+
+  for (std::size_t i = 0; i < text.size(); ++i) {
+    const char c = text[i];
+
+    if (in_quotes) {
+      row_has_content = true;
+      if (c != '"') {
+        field.push_back(c);
+      } else if (i + 1 < text.size() && text[i + 1] == '"') {
+        field.push_back('"');
+        ++i;
+      } else {
+        in_quotes = false;
+      }
+      continue;
+    }
+
+    switch (c) {
+      case '"':
+        row_has_content = true;
+        in_quotes = true;
+        break;
+      case ',':
+        row_has_content = true;
+        row.push_back(std::move(field));
+        field.clear();
+        break;
+      case '\r':
+        break;
+      case '\n':
+        end_row();
+        break;
+      default:
+        row_has_content = true;
+        field.push_back(c);
+    }
+  }
+
+  end_row();
+  return rows;
+}
+
+}  // namespace cepis::detail