flatten 20260225

tjp/core/csv/Csv.cpp

@@ -0,0 +1,168 @@
+// TJP COPYRIGHT HEADER
+
+#include "Csv.hpp"
+
+// Adapted from
+// https://stackoverflow.com/questions/1120140/how-can-i-read-and-parse-csv-files-in-c
+
+#include <tjp/core/iterators/enumerate.hpp>
+#include <sstream>
+
+namespace tjp {
+namespace core {
+
+CSV::CSV(std::istream& str_, bool readHeader_) :
+	str(str_)
+{
+	if (readHeader_)
+		readHeader();
+}
+
+void CSV::readHeader ()
+{
+	readNextRow();
+	int i=0;
+	for (auto &s : m_data)
+	{
+		m_header[s] = i;
+		++i;
+	}
+}
+	
+std::string const& CSV::operator[](std::size_t index) const
+{
+	return m_data[index];
+}
+
+std::string const& CSV::operator[](const std::string &index) const
+{
+	auto i = m_header.find(index);
+	return m_data[i->second];
+}
+
+std::size_t CSV::size() const
+{
+	return m_data.size();
+}
+
+bool CSV::readNextRow()
+{
+	std::string         line;
+	std::getline(str, line);
+	if (!line.empty() && line.back() == '\r')
+		line.resize(line.size()-1);
+
+	std::stringstream   lineStream(line);
+	
+	std::string         cell;
+
+	m_data.clear();
+	while(std::getline(lineStream, cell, ','))
+	{
+		m_data.push_back(cell);
+	}
+	// This checks for a trailing comma with no data after it.
+	if (!lineStream && cell.empty())
+	{
+		// If there was a trailing comma then add an empty element.
+		m_data.push_back("");
+	}
+	
+	return str.good();
+}
+
+FastCSV::FastCSV(std::istream& str_, bool readHeader_) :
+	str(str_),
+	lineTokens(Tokenizer::Tokens("\n")),
+	csvTokens(Tokenizer::Tokens(","))
+{
+	const auto bufferSize = 1024;
+	char buffer[bufferSize];
+	while (str_)
+	{
+		str_.read(buffer, bufferSize);
+		data.insert(data.end(), buffer, buffer + str_.gcount());
+	}
+	
+	tokenizer = Tokenizer(data.data(), data.size(), lineTokens);
+	
+	if (readHeader_)
+		readHeader();
+}
+
+void FastCSV::readHeader ()
+{
+	readNextRow();
+	int i=0;
+	for (auto &s : m_data)
+	{
+		m_header[s] = i;
+		++i;
+	}
+}
+
+FastCSV::Result FastCSV::operator[](std::size_t index) const
+{
+	return Result(m_data[index]);
+}
+
+FastCSV::Segment FastCSV::segment(std::size_t index) const
+{
+	return m_data[index];
+};
+
+std::size_t FastCSV::index(const Segment &index) const
+{
+	auto i = m_header.find(index);
+	if (i == m_header.end())
+		return -1;
+	
+	return i->second;
+}
+
+FastCSV::Result FastCSV::operator[](const Segment &index) const
+{
+	auto i = m_header.find(index);
+	return Result(m_data[i->second]);
+}
+
+std::size_t FastCSV::size() const
+{
+	return m_data.size();
+}
+
+bool FastCSV::readNextRow()
+{
+	Segment line;
+	char token;
+	bool result = tokenizer.next(line, token);
+	
+	if (!result)
+		return false;
+	
+	if (!line.empty() && line.back() == '\r')
+		line.remove_suffix(1);
+
+	Tokenizer lineTokenizer(line.data(), line.size(), csvTokens);
+
+	Segment cell;
+	char cellToken;
+	m_data.clear();
+	
+	while (lineTokenizer.next(cell, cellToken))
+	{
+		m_data.push_back(cell);
+	}
+	// This checks for a trailing comma with no data after it.
+	if (cell.empty())
+	{
+		// If there was a trailing comma then add an empty element.
+		m_data.push_back(Segment());
+	}
+	
+	return true;
+}
+
+} // namespace
+} // namespace
+

tjp/core/csv/Csv.h

@@ -0,0 +1,208 @@
+// TJP COPYRIGHT HEADER
+
+#pragma once
+
+#include <iterator>
+#include <iostream>
+#include <fstream>
+#include <sstream>
+#include <vector>
+#include <string>
+#include <map>
+#include <tjp/core/string/Tokenizer.hpp>
+
+// Adapted from
+// https://stackoverflow.com/questions/1120140/how-can-i-read-and-parse-csv-files-in-c
+
+#include <tjp/core/iterators/enumerate.hpp>
+
+namespace tjp {
+namespace core {
+
+class CSV
+{
+public:
+	CSV(std::istream& str_, bool readHeader_ = true) :
+		str(str_)
+	{
+		if (readHeader_)
+			readHeader();
+	}
+
+	void readHeader ()
+	{
+		readNextRow();
+		int i=0;
+		for (auto &s : m_data)
+		{
+			m_header[s] = i;
+			++i;
+		}
+	}
+	
+public:
+	std::string const& operator[](std::size_t index) const
+	{
+		return m_data[index];
+	}
+	
+	std::string const& operator[](const std::string &index) const
+	{
+		auto i = m_header.find(index);
+		return m_data[i->second];
+	}
+	
+	std::size_t size() const
+	{
+		return m_data.size();
+	}
+	
+	bool readNextRow()
+	{
+		std::string         line;
+		std::getline(str, line);
+		if (!line.empty() && line.back() == '\r')
+			line.resize(line.size()-1);
+
+		std::stringstream   lineStream(line);
+		
+		std::string         cell;
+
+		m_data.clear();
+		while(std::getline(lineStream, cell, ','))
+		{
+			m_data.push_back(cell);
+		}
+		// This checks for a trailing comma with no data after it.
+		if (!lineStream && cell.empty())
+		{
+			// If there was a trailing comma then add an empty element.
+			m_data.push_back("");
+		}
+		
+		return str.good();
+	}
+
+protected:
+	std::istream& str;
+	std::map<std::string, size_t> m_header;
+	std::vector<std::string> m_data;
+};
+
+class FastCSV
+{
+protected:
+	typedef std::string_view Segment;
+	typedef std::string Result;
+	
+	std::vector<char> data;
+	
+	typedef Tokenizer<Segment, TokenVector> Tokenizer;
+	
+	Tokenizer::Tokens lineTokens;
+	Tokenizer::Tokens csvTokens;
+	Tokenizer tokenizer;
+	
+public:
+	FastCSV(std::istream& str_, bool readHeader_ = true) :
+		str(str_),
+		lineTokens(Tokenizer::Tokens("\n")),
+		csvTokens(Tokenizer::Tokens(","))
+	{
+		const auto bufferSize = 1024;
+		char buffer[bufferSize];
+		while (str_)
+		{
+			str_.read(buffer, bufferSize);
+			data.insert(data.end(), buffer, buffer + str_.gcount());
+		}
+		
+		tokenizer = Tokenizer(data.data(), data.size(), lineTokens);
+		
+		if (readHeader_)
+			readHeader();
+	}
+
+	void readHeader ()
+	{
+		readNextRow();
+		int i=0;
+		for (auto &s : m_data)
+		{
+			m_header[s] = i;
+			++i;
+		}
+	}
+	
+public:
+	Result operator[](std::size_t index) const
+	{
+		return Result(m_data[index]);
+	}
+	
+	Segment segment(std::size_t index) const
+	{
+		return m_data[index];
+	};
+	
+	std::size_t index(const Segment &index) const
+	{
+		auto i = m_header.find(index);
+		if (i == m_header.end())
+			return -1;
+		
+		return i->second;
+	}
+	
+	Result operator[](const Segment &index) const
+	{
+		auto i = m_header.find(index);
+		return Result(m_data[i->second]);
+	}
+	
+	std::size_t size() const
+	{
+		return m_data.size();
+	}
+	
+	bool readNextRow()
+	{
+		Segment line;
+		char token;
+		bool result = tokenizer.next(line, token);
+		
+		if (!result)
+			return false;
+		
+		if (!line.empty() && line.back() == '\r')
+			line.remove_suffix(1);
+
+		Tokenizer lineTokenizer(line.data(), line.size(), csvTokens);
+
+		Segment cell;
+		char cellToken;
+		m_data.clear();
+		
+		while (lineTokenizer.next(cell, cellToken))
+		{
+			m_data.push_back(cell);
+		}
+		// This checks for a trailing comma with no data after it.
+		if (cell.empty())
+		{
+			// If there was a trailing comma then add an empty element.
+			m_data.push_back(Segment());
+		}
+		
+		return true;
+	}
+
+protected:
+	std::istream& str;
+	std::map<Segment, size_t> m_header;
+	std::vector<Segment> m_data;
+};
+
+} // namespace
+} // namespace
+

tjp/core/csv/Csv.hpp

@@ -0,0 +1,82 @@
+// TJP COPYRIGHT HEADER
+
+#pragma once
+
+#include <iosfwd>
+#include <vector>
+#include <string>
+#include <map>
+
+#include <tjp/core/string/Tokenizer.hpp>
+
+// Adapted from
+// https://stackoverflow.com/questions/1120140/how-can-i-read-and-parse-csv-files-in-c
+
+#include <tjp/core/iterators/enumerate.hpp>
+
+namespace tjp {
+namespace core {
+
+class CSV
+{
+public:
+	CSV(std::istream& str_, bool readHeader_ = true);
+
+	void readHeader ();
+	
+public:
+	std::string const& operator[](std::size_t index) const;
+	
+	std::string const& operator[](const std::string &index) const;
+	
+	std::size_t size() const;
+	
+	bool readNextRow();
+
+protected:
+	std::istream& str;
+	std::map<std::string, size_t> m_header;
+	std::vector<std::string> m_data;
+};
+
+class FastCSV
+{
+protected:
+	typedef std::string_view Segment;
+	typedef std::string Result;
+	
+	std::vector<char> data;
+	
+	typedef core::Tokenizer<Segment, TokenVector> Tokenizer;
+	
+	Tokenizer::Tokens lineTokens;
+	Tokenizer::Tokens csvTokens;
+	Tokenizer tokenizer;
+	
+public:
+	FastCSV(std::istream& str_, bool readHeader_ = true);
+
+	void readHeader ();
+	
+public:
+	Result operator[](std::size_t index) const;
+	
+	Segment segment(std::size_t index) const;
+	
+	std::size_t index(const Segment &index) const;
+	
+	Result operator[](const Segment &index) const;
+	
+	std::size_t size() const;
+	
+	bool readNextRow();
+
+protected:
+	std::istream& str;
+	std::map<Segment, size_t> m_header;
+	std::vector<Segment> m_data;
+};
+
+} // namespace
+} // namespace
+

tjp/core/csv/_tests/Csv.cpp

@@ -0,0 +1,16 @@
+// TJP COPYRIGHT HEADER
+
+#include <tjp/core/csv/Csv.h>
+
+using namespace tjp::core;
+
+void compile()
+{
+    std::ifstream file("plop.csv");
+
+    CSV csv(file);
+    while(csv.readNextRow())
+    {
+        std::cout << "4th Element(" << csv[3] << ")\n";
+    }
+}