Convert Evernote exports to structured data

To convert Evernote exports to structured data, you’ll typically work with ENEX files (Evernote’s XML export format). These files contain metadata and content for each note, structured in XML. The goal is to parse these and convert them into structured formats like JSON, CSV, or a database-friendly format.

Step-by-Step Guide to Convert Evernote Exports to Structured Data

---

1. Export Notes from Evernote

• Open Evernote.

• Select the notes or notebook.

• Go to File > Export Notes….

• Choose Evernote XML Format (.enex).

---

2. Understand the ENEX Structure

Each ENEX file has:

xml
CopyEdit
<en-export>
  <note>
    <title>Note Title</title>
    <content><![CDATA[<?xml version="1.0"?><!DOCTYPE en-note ...>Content Here</en-note>]]></content>
    <created>20220101T120000Z</created>
    <updated>20220102T130000Z</updated>
    <tag>Tag1</tag>
    <tag>Tag2</tag>
    <note-attributes>
      <author>John Doe</author>
    </note-attributes>
    <resource>...</resource>
  </note>
  ...
</en-export>

---

3. Use a Parser to Extract Data

Python is a common tool for this. Here’s a basic example:

python
CopyEdit
import xml.etree.ElementTree as ET
import json

def parse_enex(enex_path):
    tree = ET.parse(enex_path)
    root = tree.getroot()

    namespace = {'en': 'http://xml.evernote.com/pub/enml2.dtd'}

    notes = []
    for note in root.findall('note'):
        title = note.findtext('title')
        content_raw = note.findtext('content')
        created = note.findtext('created')
        updated = note.findtext('updated')
        tags = [tag.text for tag in note.findall('tag')]
        author = note.find('note-attributes/author')
        author = author.text if author is not None else None

        notes.append({
            'title': title,
            'content': content_raw,
            'created': created,
            'updated': updated,
            'tags': tags,
            'author': author
        })

    return notes

parsed_data = parse_enex("your_notes.enex")
with open("notes.json", "w") as f:
    json.dump(parsed_data, f, indent=2)

---

4. Optional: Clean HTML Content

The content field contains ENML (Evernote Markup Language), which is based on XHTML. You may want to extract plain text or HTML.

Use libraries like BeautifulSoup:

python
CopyEdit
from bs4 import BeautifulSoup

def extract_plain_text(enml_content):
    soup = BeautifulSoup(enml_content, 'html.parser')
    return soup.get_text(separator='n')

# Apply to each note
for note in parsed_data:
    note['text'] = extract_plain_text(note['content'])

---

5. Convert to CSV or Database Format

You can use pandas to convert the cleaned data to a CSV:

python
CopyEdit
import pandas as pd

df = pd.DataFrame(parsed_data)
df.to_csv("notes.csv", index=False)

---

6. Storing in a Database (Optional)

You can also insert the data into a SQL database using sqlite3:

python
CopyEdit
import sqlite3

conn = sqlite3.connect('notes.db')
c = conn.cursor()
c.execute('''CREATE TABLE IF NOT EXISTS notes
             (title TEXT, content TEXT, created TEXT, updated TEXT, tags TEXT, author TEXT)''')

for note in parsed_data:
    c.execute("INSERT INTO notes VALUES (?, ?, ?, ?, ?, ?)", (
        note['title'],
        note['content'],
        note['created'],
        note['updated'],
        ','.join(note['tags']),
        note['author']
    ))

conn.commit()
conn.close()

---

7. Advanced: Handling Attachments

The <resource> element contains attachments (like images, PDFs):

xml
CopyEdit
<resource>
  <data encoding="base64">...binary data...</data>
  <mime>image/png</mime>
  <resource-attributes>
    <file-name>image.png</file-name>
  </resource-attributes>
</resource>

You can decode and save these using Python:

python
CopyEdit
import base64

for resource in note.findall('resource'):
    mime = resource.findtext('mime')
    filename = resource.find('resource-attributes/file-name').text
    data = base64.b64decode(resource.findtext('data'))

    with open(filename, 'wb') as f:
        f.write(data)

---

Summary

• Export notes as .enex.

• Parse using Python or another XML tool.

• Extract key fields: title, content, tags, created/updated dates, author.

• Clean and convert ENML to plain text.

• Output to structured format: JSON, CSV, or SQL database.

• Optionally extract attachments.

This approach gives you full control to repurpose Evernote content into any system like Notion, Roam, Obsidian, or custom apps.