#ifndef _STREAMS_H
#define _STREAMS_H

#include <iostream>
#include <vector>

#include "extensions.h"

using namespace std;

//CONVERSIONS
//used eveywhere in the system and separated also for other projects
//UTF-8, ISO8859, ascii conversions
//URLEncoding, HTML entity encoding / decoding

//STRING ANALYSIS
//character, word and sentence

//all encoding is done by:
//scanning the string for any charcters that need changing and recording their position
//IF ANY CHARACTERS FOUND: malloc using the number of characters that need changing to esimate the new size
//convert and return

//one character only (static in-line for speed and simplicity)
//often inherited but also used statically
class Character {
public:
	static bool is(                const unsigned char c, const char *chars) {if (!chars) return false; else return strchr(chars, c) != 0;}
	static bool isWhiteSpace(      const unsigned char c)        {return (c <= ' ');}
	static bool isUpper(           const unsigned char c)        {return (c >= 'A' && c <= 'Z');}
	static bool isLower(           const unsigned char c)        {return (c >= 'a' && c <= 'z');}
	static bool isNumeric(         const unsigned char c)        {return (c >= '0' && c <= '9');}
	static bool isAlpha(           const unsigned char c)        {return (isUpper(c) || isLower(c));}
	static bool isAlphaNumeric(    const unsigned char c)        {return (isAlpha(c) || isNumeric(c));}
	static bool isUTF8AlphaNumeric(const unsigned char c)        {return (isAlpha(c) || isNumeric(c) || !isASCII(c));}
	static bool isASCII(           const unsigned char c)        {return (c & 128) == 0;}
	static bool isGrammar(         const unsigned char c)        {return !isWhiteSpace(c) && !isUTF8AlphaNumeric(c);}
	static bool isSentenceEnder(   const unsigned char c)        {return is(c, ".!?:>");}
};

//more than one character for short combinations (static in-line for speed and simplicity)
//these functions only look forward if the byte is non-zero to avoid page faults
class Phrase: public Character {
public:
	static bool isSentenceEnd(const char current, const char next    ) {return (isSentenceEnder(current)      && !isUTF8AlphaNumeric(next));}
	static bool isWordStart(  const char current, const char previous) {return (!isUTF8AlphaNumeric(previous) &&  isUTF8AlphaNumeric(current));}
	static bool isWordEnd(    const char current, const char previous) {return ( isUTF8AlphaNumeric(previous) && !isUTF8AlphaNumeric(current));}
};

//entire stream with conversions (all static)
//can be inherited
class Stream: public Phrase {
public:
	static bool isURLEncoded(const char *string);

	//these 2 go hand-in-hand: calculate the length, malloc and then call the 2nd to write into the space created
	static size_t URLEncodedLength(const char *string);
	static size_t URLEncode(const char *in, char *out);

	//alternative way: mallocs (where necessary) the space and writes the string
	static const char *URLEncode(char *string, const bool reuseSpaces = false, const bool checkNotEncodedAlready = true);

	//saving HTML to and from database (raw UTF-8, not encoded)
	static const char *HTMLEntityDecodeToUTF8(char *string);
	static const char *UTF8ToHTMLEntityEncode(char *string);
};

#endif