package Torello.HTML.Tools.NewsSite;

import Torello.JavaDoc.LinkJavaSource;

import Torello.HTML.TagNode;
import Torello.HTML.HTMLNode;
import Torello.HTML.DotPair;
import Torello.HTML.TC;
import Torello.HTML.URLFilter;

import Torello.HTML.NodeSearch.TagNodeFindL1Inclusive;
import Torello.HTML.NodeSearch.InnerTagFindInclusive;
import Torello.HTML.NodeSearch.TagNodeGet;
import Torello.HTML.NodeSearch.InnerTagGet;
import Torello.HTML.NodeSearch.InnerTagGetInclusive;
import Torello.HTML.NodeSearch.TextComparitor;

import Torello.Java.StrFilter;
import Torello.Java.FileRW;
import Torello.Java.LFEC;
import Torello.Java.Country;
import Torello.Java.C;

import Torello.Languages.LC;

import java.util.Vector;
import java.util.Hashtable;

import java.util.regex.Pattern;

import java.net.URL;

import java.io.IOException;

/**
 * This class is nothing more than an 'Example Class' that contains some foreign-language
 * based news web-pages, from both overseas and from Latin America.
 * 
 * <BR /><BR /><EMBED CLASS='external-html' DATA-FILE-ID=NEWS_SITES>
 */
public class NewsSites
{
    private NewsSites() { }

    @SuppressWarnings("unchecked")
    private static final Hashtable<String, Vector<URL>> newsPaperSections = 
        (Hashtable<String, Vector<URL>>) LFEC.readObjectFromFile_JAR
        (NewsSite.class, "data-files/SectionURLs.htdat", true, Hashtable.class);


    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=NEWS_SITES_RUN_EX>
     * @throws IOException This throws for IO errors that may occur when reading the web-server,
     * or when saving the web-pages or images to the file-system.
     * 
     * @see FileRW#delTree(String, boolean, Appendable)
     * @see NewsSite
     * @see FileRW#writeFile(CharSequence, String)
     * @see C#toHTML(String, boolean, boolean, boolean)
     */
    @LinkJavaSource(handle="RunExample")
    public static void runExample() throws IOException
    {
        // Click on the @LinkJavaSource Curved-Arrow to view the example code in full screen
        RunExample.run();
    }

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=NEWS_SITES_MAIN>
     * @param argv These are the command line arguments passed by the JRE to this method.
     * 
     * @throws IOException If there are any problems while attempting to save the output to the
     * the output file (if one was named / requested).
     */
    public static void main(String[] argv) throws IOException
    {
        // Uncomment this line to run the example code (instead of section-data print)
        // runExample(); System.exit(0);

        // The data-file is loaded into private field "newsPaperSections"
        // This private field is a Hashtable<String, Vector<URL>>.  Convert each of
        // these sections so that they may be printed to terminal and maybe to a text
        // file.

        StringBuilder sb = new StringBuilder();

        for (String newspaper : newsPaperSections.keySet())
        {
            sb.append(newspaper + '\n');
            for (URL section : newsPaperSections.get(newspaper))
                sb.append(section.toString() + '\n');
            sb.append("\n\n***************************************************\n\n");
        }
        
        String s = sb.toString();
        System.out.println(s);
        
        // If there is a command-line parameter, it shall be interpreted a file-name.
        // The contents of the "sections data-file" (as text) will be written a file on the
        // file-system using the String-value of "argv[0]" as the name of the output-filename.

        if (argv.length == 1) FileRW.writeFile(s, argv[0]);
    }


    // URLFilter.regexKEEP(Pattern.compile("^http.+baidu\\.com\\/s\\?id=\\d+$")));
    // ArticleGet.usual(TextComparitor.CN_CI, "article-content"));

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=ABC_ES_LG>
     * @see TagNodeFindL1Inclusive#all(Vector, String)
     * @see TagNodeGet#first(Vector, int, int, TC, String[])
     * @see TagNode#AV(String)
     */
    public static Vector<String> ABC_LINKS_GETTER(URL url, Vector<HTMLNode> page)
    {
        final Vector<String> ret = new Vector<>();

        TagNode tn;
        String urlStr;

        // Links are kept inside <ARTICLE> ... </ARTICLE> on the main / section page.
        for (DotPair article : TagNodeFindL1Inclusive.all(page, "article"))

            // Now find the <A HREF=...> ... </A>
            if ((tn = TagNodeGet.first(page, article.start, article.end, TC.OpeningTags, "a"))
                != null)

                if ((urlStr = tn.AV("href")) != null)
                    ret.add(urlStr);

        return ret;
    }

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=ABC_ES>
     * <EMBED CLASS='external-html' DATA-FILE-ID=NEWS_STE_CHANGE>
     */
    public static final NewsSite ABCES = new NewsSite
    (
        "ABC España",
        Country.Spain,
        "https://www.abc.es/",
        LC.ES,
        "ABC is a Spanish national daily newspaper.  It is the third largest general-interest " +
        "newspaper in Spain, and the oldest newspaper still operating in Madrid.",
        newsPaperSections.get("ABCES"),
        StrFilter.comparitor(TextComparitor.EW_CI, ".html"),
        NewsSites::ABC_LINKS_GETTER,
        ArticleGet.usual("main"),
        null // bannerAndAdFinder
    );

    /** <EMBED CLASS='external-html' DATA-FILE-ID=PULSO> */
    public static final NewsSite Pulso = new NewsSite
    (
        "El Pulso, México",
        Country.Mexico,
        "https://elpulso.mx",
        LC.ES,
        "El Pulso newspaper is Spanish language newspaper in Mexico. It is showing breaking " +
            "news, " +
        "headlines, kids news, tourism news, entertainment news, study news, industrial news, " +
        "economical news, health & beauty news, crime news, career news, Travel news, " +
        "diet & fitness news, Top stories, special news, celebrity news.",
        newsPaperSections.get("PULSO"),
        StrFilter.regExKEEP(Pattern.compile(
            "^https?:\\/{2}.*?\\/\\d{4}\\/\\d{2}\\/\\d{2}\\/[\\w-]{10,}\\/$"
        ), false),
        null, // LinksGet
        ArticleGet.usual(TextComparitor.C, "entry-content"),
        null // bannerAndAddFinder
    );

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=EL_NACIONAL_LG>
     * @see InnerTagFindInclusive#all(Vector, String, String, TextComparitor, String[])
     * @see TagNodeGet#first(Vector, int, int, TC, String[])
     * @see TagNode#AV(String)
     */
    public static Vector<String> EL_NACIONAL_LINKS_GETTER(URL url, Vector<HTMLNode> page)
    {
        Vector<String> ret = new Vector<>();       TagNode tn;     String urlStr;

        // Links are kept inside <DIV CLASS=td-module-thumb> ... </DIV> on the main / section page.
        for (DotPair article : InnerTagFindInclusive.all
            (page, "div", "class", TextComparitor.C, "td-module-thumb"))

            // Now find the <A HREF=...> ... </A>
            if ((tn = TagNodeGet.first
                (page, article.start, article.end, TC.OpeningTags, "a")) != null)

                if ((urlStr = tn.AV("href")) != null)
                    ret.add(urlStr);

        return ret;
    }

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=EL_NACIONAL>
     * <EMBED CLASS='external-html' DATA-FILE-ID=NEWS_STE_CHANGE>
     */
    public static final NewsSite ElNacional = new NewsSite
    (
        "El Nacional",
        Country.Venezuela,
        "https://elnacional.com",
        LC.ES,
        "El Nacional is a Venezuelan publishing company under the name C.A. Editorial " +
            "El Nacional, " +
        "most widely known for its El Nacional newspaper and website. It, along with Últimas " +
        "Noticias and El Universal, are the most widely read and circulated daily national " +
        "newspapers in the country, and it has an average of more than 80,000 papers distributed " +
        "daily and 170,000 copies on weekends.",
        newsPaperSections.get("ElNacional"),
        (URLFilter) null, /* The LinksGetter will only return valid Anchor's */
        NewsSites::EL_NACIONAL_LINKS_GETTER,
        ArticleGet.usual("article"),
        null /* bannerAndAdFinder */
    );

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=EL_ESPECTADOR_LG>
     * @see InnerTagFindInclusive#all(Vector, String, String, TextComparitor, String[])
     * @see InnerTagGet#first(Vector, int, int, String, String, TextComparitor, String[])
     * @see TagNode#AV(String)
     */
    public static Vector<String> EL_ESPECTADOR_LINKS_GETTER(URL url, Vector<HTMLNode> page)
    {
        Vector<String> ret = new Vector<>();

        TagNode tn;
        String  urlStr;

        // Links are kept inside <DIV CLASS="Card ..."> ... </DIV> on the main / section page.
        for (DotPair article : InnerTagFindInclusive.all
            (page, "div", "class", TextComparitor.C, "Card"))

            // Now find the <A CLASS="card-link" HREF=...> ... </A>
            if ((tn = InnerTagGet.first
                (page, article.start, article.end, "a", "class", TextComparitor.C, "card-link"))
                    != null)

                if ((urlStr = tn.AV("href")) != null)
                    ret.add(urlStr);

        return ret;
    }

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=EL_ESPECTADOR>
     * <EMBED CLASS='external-html' DATA-FILE-ID=NEWS_STE_CHANGE>
     */
    public static final NewsSite ElEspectador = new NewsSite
    (
        "El Espectador, Columbia",
        Country.Colombia, 
        "https://elespectador.com",
        LC.ES,
        "El Espectador (meaning \"The Spectator\") is a newspaper with national circulation within "+
        "Colombia, founded by Fidel Cano Gutiérrez on 22 March 1887 in Medellín and published "     +
        "since 1915 in Bogotá. It changed from a daily to a weekly edition in 2001, following a "   +
        "financial crisis, and became a daily again on 11 May 2008, a comeback which had been "     +
        "long rumoured, in tabloid format (28 x 39.5 cm). From 1997 to 2011 its main shareholder "  +
        "was Julio Mario Santo Domingo.",
        newsPaperSections.get("ElEspectador"),
        StrFilter.comparitor(TextComparitor.ENDS_WITH, "/"),
        NewsSites::EL_ESPECTADOR_LINKS_GETTER,
        ArticleGet.usual("article"),
        null /* bannerAndAdFinder */
    );

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=GOV_CN_CAROUSEL_LG>
     * @see InnerTagGetInclusive#first(Vector, String, String, TextComparitor, String[])
     * @see TagNodeGet#all(Vector, TC, String[])
     * @see TagNode#AV(String)
     */
    public static Vector<String> GOVCN_CAROUSEL_LINKS_GETTER(URL url, Vector<HTMLNode> page)
    {
        Vector<String>  ret     = new Vector<>();
        String          urlStr;

        // Find the first <DIV CLASS="slider-carousel"> ... </DIV> section
        Vector<HTMLNode> carouselDIV = InnerTagGetInclusive.first
            (page, "div", "class", TextComparitor.CN_CI, "slider-carousel");


        // Retrieve any HTML Anchor <A HREF=...> ... </A> found within the contents of the
        // Divider.

        for (TagNode tn: TagNodeGet.all(carouselDIV, TC.OpeningTags, "a"))
            if ((urlStr = tn.AV("href")) != null)
                ret.add(urlStr);

        return ret;
    };

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=GOV_CN_CAROUSEL>
     * <EMBED CLASS='external-html' DATA-FILE-ID=NEWS_STE_CHANGE>
     */
    public static final NewsSite GovCNCarousel = new NewsSite
    (
        "Chinese Government Web Portal",
        Country.China,
        "https://gov.cn/",
        LC.ZH_CN,
        "The Chinese Government Sponsored Web-Site",
        newsPaperSections.get("GovCNCarousel"),
        StrFilter.regExKEEP(Pattern.compile(
            "^http://www.gov.cn/(?:.+?/)?\\d{4}-\\d{2}/\\d{2}/(?:.+?/)?" +
                "content_\\d+.htm(?:l)?(#\\d+)?"
        ), false),
        NewsSites::GOVCN_CAROUSEL_LINKS_GETTER,
        ArticleGet.usual(TextComparitor.C, "article"),
        null /* bannerAndAddFinder */
    );

    /**
     * <EMBED CLASS='external-html' DATA-FILE-ID=GOV_CN>
     * <EMBED CLASS='external-html' DATA-FILE-ID=NEWS_STE_CHANGE>
     */
    public static final NewsSite GovCN = new NewsSite
    (
        "Chinese Government Web Portal",
        Country.China,
        "https://gov.cn/",
        LC.ZH_CN,
        "The Chinese Government Sponsored Web-Site",
        newsPaperSections.get("GovCN"),
        StrFilter.regExKEEP(Pattern.compile(
            "^http://www.gov.cn/(?:.+?/)?\\d{4}-\\d{2}/\\d{2}/(?:.+?/)?" +
                "content_\\d+.htm(?:l)?(#\\d+)?"
        ), false),
        null,
        ArticleGet.usual(TextComparitor.C, "article"),
        null /* bannerAndAddFinder */
    );
}