/**
  * Web Site Titles and Descriptions Length Checker for HTML files..
  * @author Robert John Morton
  * @version 22 May 2019 */

/* This program list all html files with: 
   1) titles shorter than 60 characters in title_too_short.txt
   2) titles longer than 70 characters in title_too_long.txt
   3) descriptions shorter than 140 characters in descr_too_short.txt
   4) descriptions longer than 160 characters in descr_too_long.txt */

import java.io.*;

class titles_checker {
  private static Writer 
    ttl,  // for HTML files with titles too long
    tts,  // for HTML files with titles too short
    dtl,  // for HTML files with descriptions too long
    dts;  // for HTML files with descriptions too short

  private static Reader
    aif;  // file object for 'articles_index.html'

  private static String 
    fp = "../../articles-index/",
    fn = "articles_index.html",
    tag = "",    // for capturing tag name text
    DT = "",     // for capturing DT text
    DD = "",     // for capturing DD text
    path = "",   // title entry's href path 
    title = "",  // title entry's title text
    s = "";      // used to form the length of the title or description

  private static int 
    c = 0,   // for input character
    tl = 0,  // title length
    dl = 0,  // description length
    L = 0;   // length of title text

  private static boolean 
    inTag = false,  // True indicates we are inside the <> brackets
    inDT = false,   // True indicates we are reading the title text
    inDD = false;   // True indicates we are reading the title text




  /* Format of the title section of an entry in articles_index.html
  <dt>001 
  <a href=../book/chap03/moon.html>
  7-day Week: Its Relation to Moon Phases
  </a></dt>
  Called only from one place in main(). */

  private static void parseDT() throws Exception {
    DT = DT.trim();
    int
      x = 0,
      ldt = DT.length();

    if(ldt > 22) {  // length of extraneous tags

      /* Chop off the extraneous tags and trim spurious white-space
      characters such as C/Rs, NULLs and EOF from the title entry. */

      DT = DT.substring(13,ldt - 8).trim();

      if((x = DT.indexOf(">")) != -1) {   //'>' char link and title text

        /* Chop off the extraneous tags and trim spurious white-space
        characters such as C/Rs, NULLs and EOF from the path. */

        path = DT.substring(0,x).trim();

        /* Chop off the extraneous tags and trim spurious white-space
        characters such as C/Rs, NULLs and EOF from the title text. */

        title = DT.substring(x + 1,DT.length()).trim();

        /* Form the length of the title text as a 3-digit
        string with leading zeros where necessary. */ 

        int l = title.length();
        s = "" + l;
        if(l < 100)
          s = "0" + l;
        if(l < 10)
          s = "0" + l;

        // Form the output entry of the link + title text
        // s = s + "  " + path + "\n" + title + "\n\n";
        s = path + "  " + l + "\n";

        if(l < 60)
          tts.write(s);  // if title too short
        else if(l > 70)
          ttl.write(s);  // if title too long
      }
    }
    DT = "";
    inDT = false;
  }




  /* Format of the description section of an entry in articles_index.html
  <dd><div align=justify> [23 characters]
  Today is always a little bit longer than yesterday and tomorrow
  will be a little bit longer than today.
  </div></dd>  [11 characters]
  Called from only one place in main(). */

  private static void parseDD() throws Exception {
    DD = DD.trim();
    int x = 0, ldd = DD.length();

    if(ldd > 34) {  // length of extraneous tags

      /* Chop off the extraneous tags and trim spurious white-space
      characters such as C/Rs, NULLs and EOF from description text. */

      DD = DD.substring(20,ldd - 10).trim();

      /* Form the length of the title text as a 3-digit
      string with leading zeros where necessary. */ 

      int l = DD.length();
      s = "" + l;
      if(l < 100)
        s = "0" + l;
      if(l < 10)
        s = "0" + l;

      // Form the output entry of the link + title text
      // s = s + "  " + path + "\n" + DD + "\n\n";
      s = path + "  " + l + "\n";

      if(l < 140)       //if description too short
        dts.write(s);
      else if(l > 160)  //if description too long
        dtl.write(s);
    } 
    DD = "";       // clear description string ready for next pass
    inDD = false;  // and clear the 'doing description' flag
  }




  public static void main(String args[]) throws Exception {

    /* Create a file reader for article_index.html
    and file writers for the four results files. */

    aif = new FileReader("../../articles-index/articles_index.html");
    ttl = new FileWriter("title_too_long.txt");
    tts = new FileWriter("title_too_short.txt");
    dtl = new FileWriter("descr_too_long.txt");
    dts = new FileWriter("descr_too_short.txt");

    /* Read and examine each each character in turn from art-
    icles_index.html until its end-of-file is encountered. */

    try {
      while((c = aif.read()) != -1) {
        if(c == '<') {   // If the new char is an HTML tag start-bracket '<'
          inTag = true;  // set the 'in tag' flag
          tag = "";      // clear the tag string
        }
        else         // else, if we are currently insid
        if(inTag) {  // a tag [between a '<' and a '>'] ...

          if(c == '>') {    // if new character is a tag terminator
            inTag = false;  // baracket '>', clear the 'in tag' flag

            /* If we have just captured a <dt> or a <dd> tag, switch to
            capturing title or description respectively, otherwise, if
            we have just captured a </dt> or a </dd> tag, go and parse
            the respective content. */

            if(tag.equals("dt"))
              inDT = true;
            else if(tag.equals("dd"))
              inDD = true;
            else if(tag.equals("/dt"))
              parseDT();
            else if(tag.equals("/dd"))
              parseDD();
          }
          else
            tag += (char)c;  // else add the new char to tag string
        }

        /* Add the current character to the appropriate capture
        string for the title or description respectively. */ 

        if(inDT)
          DT += (char)c;
        else if(inDD)
          DD += (char)c;
      }
    } catch(Exception e) { }  // catch 'end-of-file' exception

    // Close all files.
    aif.close();
    ttl.close();
    tts.close();
    dtl.close();
    dts.close();
  }
}