NITFr

gem 'nitfr'

require 'nitfr'

# Parse from a string
xml = File.read('article.xml')
doc = NITFr.parse(xml)

# Or parse directly from a file
doc = NITFr.parse_file('article.xml')

# With explicit encoding
doc = NITFr.parse_file('article.xml', encoding: 'ISO-8859-1')

# Get the headline
doc.headline          # => "Revolutionary Technology Changes Industry"
doc.headlines.primary # => "Revolutionary Technology Changes Industry"
doc.headlines.secondary # => "Experts predict widespread adoption"

# Get byline information
doc.byline.text       # => "By Jane Smith, Senior Reporter"
doc.byline.person     # => "Jane Smith"
doc.byline.title      # => "Senior Reporter"

# Get the article text
doc.paragraphs.each do |para|
  puts para.text
end

# Or get all text at once
puts doc.text

# Document metadata
doc.title           # => "Sample News Article Title"
doc.doc_id          # => "article-2024-001"
doc.issue_date      # => #<Date: 2024-12-15>

# Copyright info
doc.docdata.copyright_holder  # => "Example News Corp"
doc.docdata.copyright_year    # => "2024"

# Urgency (1-8, 1 being most urgent)
doc.docdata.urgency           # => 4

# Identified content
doc.docdata.subjects      # => ["Technology", "Business"]
doc.docdata.organizations # => ["TechCorp Inc"]
doc.docdata.people        # => ["John Doe"]
doc.docdata.locations     # => ["San Francisco"]

# Access the body section
body = doc.body

# Dateline and abstract
body.dateline   # => "SAN FRANCISCO, Dec 15"
body.abstract   # => "A new technology platform..."

# Block quotes
body.block_quotes  # => ["Innovation distinguishes..."]

# Tagline from body.end
body.tagline    # => "Contact: press@example.com"

doc.paragraphs.each do |para|
  # Check if it's the lead paragraph
  puts "LEAD: " if para.lead?

  # Get plain text
  puts para.text

  # Get entities mentioned in this paragraph
  puts "People: #{para.people.join(', ')}"
  puts "Organizations: #{para.organizations.join(', ')}"
  puts "Locations: #{para.locations.join(', ')}"

  # Get emphasized text
  puts "Emphasized: #{para.emphasis.join(', ')}"

  # Get links
  para.links.each do |link|
    puts "Link: #{link[:text]} -> #{link[:href]}"
  end

  # Word count
  puts "Words: #{para.word_count}"
end

doc.media.each do |media|
  puts "Caption: #{media.caption}"
  puts "Credit: #{media.credit}"
  puts "MIME type: #{media.mime_type}"

  if media.image?
    puts "Image: #{media.source}"
    puts "Size: #{media.width}x#{media.height}"
    puts "Alt text: #{media.alt_text}"
  elsif media.video?
    puts "Video: #{media.source}"
  elsif media.audio?
    puts "Audio: #{media.source}"
  end

  # Access all references (different sizes/formats)
  media.references.each do |ref|
    puts "  #{ref[:source]} (#{ref[:mime_type]})"
  end
end

begin
  doc = NITFr.parse(xml)
rescue NITFr::ParseError => e
  puts "Invalid XML: #{e.message}"
rescue NITFr::InvalidDocumentError => e
  puts "Not a valid NITF document: #{e.message}"
end

# NITF version and change information
doc.version      # => "-//IPTC//DTD NITF 3.5//EN"
doc.change_date  # => "October 18, 2007"
doc.change_time  # => "19:30"

# Check validity
doc.valid?       # => true

# Get raw XML
doc.to_xml       # => "<?xml version..."

head = doc.head

# Meta tags as a hash
head.meta        # => {"keywords" => "tech, news", "author" => "Jane"}
head.keywords    # => ["tech, news"]

# Publication data
head.pubdata[:type]        # => "print"
head.pubdata[:name]        # => "Example Times"
head.pubdata[:edition]     # => "Morning"
head.pubdata[:volume]      # => "42"

# Revision history
head.revision_history.each do |rev|
  puts "#{rev[:name]} (#{rev[:function]}): #{rev[:comment]}"
end

docdata = doc.docdata

# Additional dates
docdata.release_date  # => #<Date: 2024-12-15>
docdata.expire_date   # => #<Date: 2024-12-31>

# Document scope and fixture
docdata.doc_scope     # => "national"
docdata.fixture       # => "fixture-123"

# Series information
docdata.series[:name]   # => "Investigation"
docdata.series[:part]   # => 2
docdata.series[:total]  # => 5

# Editorial status
docdata.management_status[:info]         # => "Approved"
docdata.management_status[:message_type] # => "advisory"

body = doc.body

# Distributor and series
body.distributor  # => "Wire Service"
body.series[:name]      # => "Special Report"
body.series[:part]      # => "1"
body.series[:totalpart] # => "3"

# Lists in the content
body.lists.each do |list|
  puts "#{list[:type]}: #{list[:items].join(', ')}"
end

# Tables (returns raw REXML elements)
body.tables.each do |table|
  # Process table XML as needed
end

# Notes from body.end
body.notes  # => ["Editor's note: ...", "Correction: ..."]

# Bibliography
body.body_end_content[:bibliography]  # => ["Source 1", "Source 2"]