From 9d862c6a6268d379ffbc18c1437414665c9229ff Mon Sep 17 00:00:00 2001
From: Bruno BELANYI <bruno@belanyi.fr>
Date: Fri, 11 Mar 2022 23:57:27 +0100
Subject: [PATCH] kraken: add 'csv' library

A very basic, naive CSV parser.
---
 src/CMakeLists.txt     |  6 ++++++
 src/csv/CMakeLists.txt |  5 +++++
 src/csv/read-csv.cc    | 44 ++++++++++++++++++++++++++++++++++++++++++
 src/csv/read-csv.hh    | 29 ++++++++++++++++++++++++++++
 4 files changed, 84 insertions(+)
 create mode 100644 src/csv/CMakeLists.txt
 create mode 100644 src/csv/read-csv.cc
 create mode 100644 src/csv/read-csv.hh

diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
index 368dfd6..ab55c25 100644
--- a/src/CMakeLists.txt
+++ b/src/CMakeLists.txt
@@ -1,4 +1,10 @@
 add_executable(kraken kraken.cc)
 target_link_libraries(kraken PRIVATE common_options)
 
+add_subdirectory(csv)
+
+target_link_libraries(kraken PRIVATE
+  csv
+)
+
 install(TARGETS kraken)
diff --git a/src/csv/CMakeLists.txt b/src/csv/CMakeLists.txt
new file mode 100644
index 0000000..c682f51
--- /dev/null
+++ b/src/csv/CMakeLists.txt
@@ -0,0 +1,5 @@
+add_library(csv STATIC
+  read-csv.cc
+  read-csv.hh
+)
+target_link_libraries(csv PRIVATE common_options)
diff --git a/src/csv/read-csv.cc b/src/csv/read-csv.cc
new file mode 100644
index 0000000..5274d12
--- /dev/null
+++ b/src/csv/read-csv.cc
@@ -0,0 +1,44 @@
+#include "read-csv.hh"
+
+#include <sstream>
+
+namespace kraken::csv {
+
+namespace {
+
+// for convenience, use a stringstream which does not accept string_view inputs
+csv_line_type parse_line(std::string const& line) {
+    auto parsed = csv_line_type{};
+
+    auto input = std::istringstream(line);
+    for (std::string atom; std::getline(input, atom, ',');) {
+        parsed.emplace_back(std::move(atom));
+    }
+
+    return parsed;
+}
+
+} // namespace
+
+csv_type read_csv(std::istream& input, CsvHeader header) {
+    auto parsed = std::vector<csv_line_type>{};
+
+    bool first = true;
+    for (std::string line; std::getline(input, line);) {
+        if (first && header == CsvHeader::SKIP) {
+            first = false;
+            continue;
+        }
+
+        parsed.emplace_back(parse_line(std::move(line)));
+    }
+
+    return parsed;
+}
+
+csv_type read_csv(std::string const& input, CsvHeader header) {
+    auto input_stream = std::istringstream(input);
+    return read_csv(input_stream, header);
+}
+
+} // namespace kraken::csv
diff --git a/src/csv/read-csv.hh b/src/csv/read-csv.hh
new file mode 100644
index 0000000..a1ceab8
--- /dev/null
+++ b/src/csv/read-csv.hh
@@ -0,0 +1,29 @@
+#pragma once
+
+#include <iosfwd>
+#include <string>
+#include <vector>
+
+namespace kraken::csv {
+
+/// Should the first line of the CSV file be kept or ignored.
+enum class CsvHeader {
+    /// Ignored.
+    SKIP,
+    /// Kepts.
+    KEEP,
+};
+
+/// Represent a raw CSV line as a vector of strings.
+using csv_line_type = std::vector<std::string>;
+
+/// Represent a raw CSV file as a vector of raw CSV lines
+using csv_type = std::vector<csv_line_type>;
+
+/// Parse a CSV file from an input-stream, return a vector of parsed lines.
+csv_type read_csv(std::istream& input, CsvHeader header = CsvHeader::SKIP);
+
+/// Convenience function which reads CSV from a string.
+csv_type read_csv(std::string const& input, CsvHeader header = CsvHeader::SKIP);
+
+} // namespace kraken::csv