/*
 * [TidyTscribe.java]
 *
 * Summary: tidies tscribe.csv list of transcription software.
 *
 * Copyright: (c) 2016-2017 Roedy Green, Canadian Mind Products, http://mindprod.com
 *
 * Licence: This software may be copied and used freely for any purpose but military.
 *          http://mindprod.com/contact/nonmil.html
 *
 * Requires: JDK 1.8+
 *
 * Created with: JetBrains IntelliJ IDEA IDE http://www.jetbrains.com/idea/
 *
 * Version History:
 *  1.0 2016-05-21 initial version
 */
package com.mindprod.repair;

import com.mindprod.common18.EIO;
import com.mindprod.common18.ST;
import com.mindprod.csv.CSV;
import com.mindprod.csv.CSVCondense;
import com.mindprod.csv.CSVReader;
import com.mindprod.csv.CSVSort;
import com.mindprod.csv.CSVWriter;
import com.mindprod.hunkio.HunkIO;

import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.io.EOFException;
import java.io.File;
import java.io.FileReader;
import java.io.FileWriter;
import java.io.IOException;
import java.io.PrintWriter;
import java.util.Arrays;
import java.util.HashSet;

import static java.lang.System.*;

/**
 * tidies tscribe.csv list of transcription software.
 * <p/>
 * E:\com\mindprod\repair\tscribe.csv
 * has format
 * package name
 * url
 * cost in usd
 * list of file formats supported
 *
 * @author Roedy Green, Canadian Mind Products
 * @version 1.0 2016-05-21 initial version
 * @since 2016-05-21
 */
public class TidyTscribe
    {
    /**
     * footer for the HTML table form of the data
     *
     * @param prw where to emit
     */
    private static void emitHTMLFooter( final PrintWriter prw )
        {
        prw.println( "</tbody>" );
        prw.println( "</table>" );
        }

    /**
     * header for the HTML table form of the data
     *
     * @param prw where to emit
     */
    private static void emitHTMLHeader( final PrintWriter prw )
        {
        prw.println( "<!-- D O  N O T  E D I T.  Generated by com.mindprod.repair.TidyTsribe from data in tscribe.csv. -->" );
        prw.println( "<table class=\"standard\">" );
        prw.println( "<thead>" );
        prw.println( "<tr><th colspan=\"" + ( ScribeFormat.values().length + 2 ) + "\">" + "Transcription Software vendors" + "</th></tr>" );
        prw.print( "<tr><th style=\"width:150px;min-width:150px\">Package</th><th>Cost<br><!-- macro CurrCon cN --></th>" );
        for ( ScribeFormat f : ScribeFormat.values() )
            {
            prw.print( "<th>" + f.getDecorativeName() + "</th>" );
            }
        prw.println( "</tr>" );
        prw.println( "</thead>" );
        prw.println( "<tbody>" );
        }

    /**
     * emit item about one package to the html version
     *
     * @param prw         where to print
     * @param packageName name of program
     * @param url         url of produc pace
     * @param cost        cost US$
     * @param formats     Which formats it supports.
     */
    private static void emitHTMLItem( final PrintWriter prw, final String packageName, final String url, final String cost, final HashSet<String> formats )
        {
        prw.print( "<tr><td><a class=\"com\" href=\"" + url + "\">" + packageName + "</a></td>" );
        if ( cost.equals( "0" ) || cost.equals( "" ) || cost.equals( "?" ) )
            {
            prw.print( "<td>" + cost + "</td>" );
            }
        else
            {
            prw.print( "<td><!-- macro CurrCon $A USD " + cost + " --></td>" );
            }
        for ( ScribeFormat format : ScribeFormat.values() )
            {
            final String cssclass = ( format.ordinal() & 1 ) != 0 ? "stripe2" : "stripe1";
            if ( formats.contains( format.name() ) )
                {
                prw.print( "<td class=\"" + cssclass + " tick\"></td>" );
                }
            else
                {
                prw.print( "<td class=\"" + cssclass + "\"></td>" );
                }
            }
        prw.println( "</tr>" );
        }

    private static void writeTidiedCSVItem( final CSVWriter w, final String packageName, final String url, final String cost, final HashSet<String> formats )
        {
        // sort list of formats.
        String[] formatArray = formats.toArray( new String[ formats.size() ] );
        Arrays.sort( formatArray );
        w.put( packageName );
        w.put( url );
        w.put( cost );
        w.put( formatArray ); // dups mean this list might be shorter than it was
        w.nl();
        }

    /**
     * prepare java source files by replacing the first header comment with a standard.
     * When proofing, logs to old.txt and new.txt in the current directory.
     *
     * @param args word proof or tidy followed by a list of files and directories to process.
     *             -s mean means process subdirs of subsequently mentioned dirs, comes first.
     *
     * @throws IOException if cannot read or write source files.
     */
    public static void main( final String[] args ) throws IOException
        {
        File tscribe = new File( "E:/com/mindprod/repair/tscribe.csv" );
        // C O N D E N S E
        try
            {
            // file,  separatorChar,  quoteChar,  commentChar, cols that can be different
            new CSVCondense( tscribe, ',', '\"', '#', CSV.UTF8 );
            }
        catch ( IOException e )
            {
            err.println();
            e.printStackTrace( err );
            err.println( "CSVCondense failed to condense " + EIO.getCanOrAbsPath( tscribe ) );
            System.exit( 2 );
            }
        // S O R T
        try
            {
            // file, cols, types, directions,  separatorChar,  quoteChar,  commentChar
            new CSVSort( tscribe, new int[] { 0 }, new char[] { 'i' }, new boolean[] { true }, ',', '\"', '#', CSV.UTF8 );
            }
        catch ( IOException e )
            {
            err.println();
            e.printStackTrace( err );
            err.println( "CSVSort failed to sort " + EIO.getCanOrAbsPath( tscribe ) );
            System.exit( 2 );
            }
        // R E A D
        // trimUnquoted allowMultipleLineFields
        CSVReader r = new CSVReader( new BufferedReader( new FileReader( tscribe ) ),
                ',',  /* separatorChar between fields */
                '\"', /* quoteChar to surround fields containing commas */
                "#",  /* char that starts comments */
                true /* hide comments */,
                true /* trimQuoted */,
                true /* trimUnquoted */,
                true /* allow multiline fields  */ );
        // tidied file
        final File tempFile = HunkIO.createTempFile( "temp_", ".tmp", tscribe );
        CSVWriter w =
                new CSVWriter( EIO.getPrintWriter( tempFile, 10 * 1024, EIO.UTF8 ),
                        -1, /* quotelevel
                 * -1 = like 0, but add an extra space after each separator/comma,
                 *  0 = minimal quotes, only around fields containing quotes or separators.
                 *  1 = quotes also around fields containing spaces.
                 *  2 = quotes around all fields, whether or not they contain commas, quotes or spaces.
                 */
                        ',',  /* separatorChar between fields */
                        '\"', /* quoteChar to surround fields containing commas */
                        '#',  /* char that starts comments */
                        true /*  trim fields of lead and trailing blanks */ );
        // HTML version
        final FileWriter fw = new FileWriter( "E:/mindprod/jgloss/include/tscribe.htmlfrag", false /* append */ );
        final BufferedWriter bw = new BufferedWriter( fw, 16_384 /* 32K bytes/16K chars, 50% of 64K byte allocation is optimal */ );
        final PrintWriter prw = new PrintWriter( bw, false /* auto flush on println */ );
        int errors = 0;
        try
            {
            try
                {
                // nl will prepend a #
                w.nl( " Transcription Software" );
                w.nl( "# package,url,cost,formats" );
                emitHTMLHeader( prw );
                while ( true )
                    {
                    // read the existing tscribe.csv file
                    final String[] fields = r.getAllFieldsInLine();
                    final String packageName = fields[ 0 ];
                    if ( fields.length == 2 )
                        {
                        err.println( "missing cost for " + packageName + " line: " + r.lineCount() );
                        errors++;
                        }
                    final String url;
                    if ( fields.length > 1 )
                        {
                        url = fields[ 1 ];
                        }
                    else
                        {
                        url = "";
                        }
                    final String cost;
                    if ( fields.length > 2 )
                        {
                        cost = fields[ 2 ];
                        }
                    else
                        {
                        cost = "";
                        }
                    if ( fields.length > 3 )
                        {
                        // sort and dedup formats
                        HashSet<String> formats = new HashSet<>( 200 );
                        for ( int i = 3; i < fields.length; i++ )
                            {
                            final String format = fields[ i ].toUpperCase();
                            if ( ST.isEmpty( format ) )
                                {
                                // just ignore empty fields
                                continue;
                                }
                            try
                                {
                                ScribeFormat.valueOf( format );
                                }
                            catch ( IllegalArgumentException e )
                                {
                                err.println( "unknown file format: " + format + " for " + packageName + " line: " + r.lineCount() );
                                errors++;
                                }
                            formats.add( format );
                            }
                        writeTidiedCSVItem( w, packageName, url, cost, formats );
                        // only put items with complete info into HTML
                        emitHTMLItem( prw, packageName, url, cost, formats );
                        }
                    else
                        {
                        // end shart line
                        w.nl( fields, false );
                        }
                    }  // end while
                }
            catch ( EOFException e )
                {
                // normal termination.
                r.close();
                w.close();
                if ( errors == 0 )
                    {
                    HunkIO.deleteAndRename( tempFile, tscribe );
                    emitHTMLFooter( prw );
                    prw.close();
                    }
                else
                    {
                    err.println( errors + " errors.  tscript.csv left as is." );
                    }
                }
            }
        catch ( IOException e )
            {
            err.println();
            e.printStackTrace( err );
            System.exit( 2 );
            }
        }
    }