Sort a Google Doc by Headings (sections) while preserving contents (paragraphs) of each section

sort_google_doc_by_headings.gs

function onOpen() {
  var ui = DocumentApp.getUi();
  // Create custom menu
  ui.createMenu('Sort and Write')
    .addItem('Sort and Write Ascending', 'sortAndWriteAscending')
    .addItem('Sort and Write Descending', 'sortAndWriteDescending')
    .addToUi();
}

function sortAndWriteAscending() {
  sortAndWrite(true);
}

function sortAndWriteDescending() {
  sortAndWrite(false);
}


function sortAndWrite(ascending=true) {
  var heading2List = extractHeadingStructure();
  var entries = convertToEntryStructure(heading2List);
  var sortedEntries = sortEntries(entries, ascending);
  writeEntriesToDoc(sortedEntries)
}

// Function to extract Heading structure and paragraphs into a list of objects
function extractHeadingStructure() {
  var doc = DocumentApp.getActiveDocument();
  var body = doc.getBody();

  // Create a list to store Heading 2 objects
  var heading2List = [];

  var paragraphs = body.getParagraphs();
  var currentHeading2 = null;
  var currentHeading2Object = null;
  var currentHeading3 = null;

  function parseHeading2Date(text) {
    // Extract the year (YYYY) from Heading 2 text
    var match = text.match(/\b\d{4}\b/);

    var result = match ? parseInt(match[0]) : null
    
    if (result) {
      return result;
    }
    Logger.log("ERROR parsing year: " + text + " into date");

  }

  function parseHeading3Date(text) {
    // Extract the sortable date (DD.MM.) from Heading 3 text
    var match = text.match(/\b(\d{1,2})[^\d]*(\d{1,2})\b/);
    if (match) {
      var day = match[1].padStart(2, '0');
      var month = match[2].padStart(2, '0');
      return day + '.' + month + '.';
    }
    Logger.log("ERROR parsing day.month: " + text + " into date");
    return null;

  }

  paragraphs.forEach(function (paragraph) {
    var textStyle = paragraph.getAttributes();
    var style = textStyle[DocumentApp.Attribute.HEADING];

    // Skip empty headings
    if (style && ((style === DocumentApp.ParagraphHeading.HEADING2) || (style === DocumentApp.ParagraphHeading.HEADING3)) && paragraph.getText().trim() === '') {
      return;
    }

    if (style === DocumentApp.ParagraphHeading.HEADING2) {
        // Handle Heading 2
        currentHeading2 = paragraph.getText();
        currentHeading3 = null;

        currentHeading2Object = { heading2: currentHeading2, heading3List: [], date: parseHeading2Date(currentHeading2) };
        heading2List.push(currentHeading2Object);
      
    } else if (style === DocumentApp.ParagraphHeading.HEADING3) {
        // Handle Heading 3
        currentHeading3 = { heading3: paragraph.getText(), paragraphs: [], date: parseHeading3Date(paragraph.getText()) };

        if (currentHeading2Object) {
          currentHeading2Object.heading3List.push(currentHeading3);
        }
    } else {
      // Handle regular paragraph text
      if (currentHeading3) {
        // If there is a Heading 3, append to its list of paragraphs
        currentHeading3.paragraphs.push(paragraph.copy());
      }
    }
  });

  Logger.log('Heading 2 List:');
  Logger.log(JSON.stringify(heading2List, null, 2));

  // Return the extracted list of objects
  return heading2List;
}

// Function to convert Heading 2, Heading 3 structure to Entry structure
function convertToEntryStructure(heading2List) {
  var entryList = [];

  heading2List.forEach(function (heading2Obj) {
    var heading2 = heading2Obj.heading2;
    var heading3List = heading2Obj.heading3List;

    heading3List.forEach(function (heading3Obj) {
      var heading3 = heading3Obj.heading3;
      var date = heading3Obj.date;
      var paragraphs = heading3Obj.paragraphs;

      // Extract day and month from the date
      var match = date.match(/(\d{1,2})\.(\d{1,2})\./);
      var day = match ? match[1] : '';
      var month = match ? match[2] : '';

      // Create an Entry object
      var entry = {
        year: heading2,
        month: month,
        day: day,
        title: "",
        text: paragraphs
      };

      entryList.push(entry);
    });
  });

  Logger.log('Entry List:');
  Logger.log(JSON.stringify(entryList, null, 2));

  // Return the converted list of Entry objects
  return entryList;
}

// Function to sort entries based on date in ascending or descending order
function sortEntries(entryList, ascending) {
  // Convert year, month, and day to integers for comparison
  entryList.forEach(function (entry) {
    entry.year = parseInt(entry.year);
    entry.month = parseInt(entry.month);
    entry.day = parseInt(entry.day);
  });

  // Default to ascending order if the sort direction is not provided or invalid
  var sortDirection = ascending ? 1 : -1;

  // Custom sorting function based on year, month, and day
  function customSort(a, b) {
    // Compare by year first, then month, and finally day
    if (a.year !== b.year) {
      return sortDirection * (a.year - b.year);
    } else if (a.month !== b.month) {
      return sortDirection * (a.month - b.month);
    } else {
      return sortDirection * (a.day - b.day);
    }
  }

  // Sort the entryList using the custom sort function
  entryList.sort(customSort);

  // Log the sorted entryList
  Logger.log('Sorted Entry List:');
  Logger.log(JSON.stringify(entryList, null, 2));

  // Return the sorted entryList
  return entryList;
}

// Function to create a new Google Doc and write entries in the specified order
function writeEntriesToDoc(sortedEntryList) {
  var doc = DocumentApp.getActiveDocument();
  var currentDocName = doc.getName();

  var newDoc = DocumentApp.create(currentDocName + '_sorted');
  var newDocBody = newDoc.getBody();

  // Czech month names
  var czechMonths = [
    "Leden", "Únor", "Březen", "Duben", "Květen", "Červen", "Červenec",
    "Srpen", "Září", "Říjen", "Listopad", "Prosinec"
  ];

    var czechMonthsGenitiv = [
    "ledna", "února", "března", "dubna", "května", "června", "července",
    "srpna", "září", "října", "listopadu", "prosince"
  ];

  // Initialize variables to keep track of current year, month, and day
  var currentYear = null;
  var currentMonth = null;
  var currentDay = null;

  // Iterate through the sorted entries and write to the new document
  sortedEntryList.forEach(function (entry) {
    // Check if the year has changed
    if (entry.year !== currentYear) {
      newDocBody.appendParagraph(String(entry.year)).setHeading(DocumentApp.ParagraphHeading.HEADING2);
      currentYear = entry.year;
      // Reset month and day when year changes
      currentMonth = null;
      currentDay = null;
    }

    // Check if the month has changed
    if (entry.month !== currentMonth) {
      newDocBody.appendParagraph(`${czechMonths[entry.month - 1]} ${entry.year}`).setHeading(DocumentApp.ParagraphHeading.HEADING3);
      currentMonth = entry.month;
      // Reset day when month changes
      currentDay = null;
    }

    // Check if the day has changed
    if (entry.day !== currentDay) {
      // var formattedDate = Utilities.formatString('%02d. %02d. %d', entry.day, entry.month, entry.year);
      var formattedDate = `${entry.day}. ${czechMonthsGenitiv[entry.month - 1]} ${entry.year}`;

      newDocBody.appendParagraph(formattedDate).setHeading(DocumentApp.ParagraphHeading.HEADING4);
      currentDay = entry.day;
    }

    // Append paragraphs as regular text, images as image, etc.
    entry.text.forEach(function (paragraph) {
      try {
        newDocBody.appendParagraph(paragraph);
      } catch {
        try {
          newDocBody.appendListItem(paragraph);
        } catch {
          try {
            newDocBody.appendImage(paragraph);
          } catch {
            try {
            newDocBody.appendPageBreak(paragraph);
            } catch {
                try {
                  newDocBody.appendTable(paragraph);
                } catch (e) {
                throw e;
              }
            }
          }
        }
      }
    });
  });

  // Log the URL of the new document
  Logger.log('New document created and entries written successfully.');
  Logger.log('New document URL: ' + newDoc.getUrl());
}