hlsv 1.0.0

data/lib/hlsv/web_app.rb

@@ -0,0 +1,569 @@
+#######
+# Copyright (c) 2026 AdClin
+# Licensed under the GNU General Public License v3.0
+#######
+
+# frozen_string_literal: true
+
+require 'json'
+require 'yaml'
+require 'csv'
+require 'fast_excel'
+require 'zip'
+require 'nokogiri'
+
+module Hlsv
+class WebApp < Sinatra::Base
+
+  set :root, Hlsv::INSTALL_ROOT
+
+  # ---------------------------------------------------------------------------
+  # :section: ROUTES — Views
+  # ---------------------------------------------------------------------------
+
+  # Homepage: loads config
+  get '/' do
+    @config  = load_config
+    erb :index
+  end
+
+  # CSV viewer: displays a CSV file as an HTML table
+  # Params:
+  #   :file           — path to the CSV file
+  #   :last_valid_key — comma-separated list of last tested keys (optional)
+  get '/csv_view' do
+    file = params[:file]
+    halt 400, "Missing file parameter" unless file
+    halt 403, "Access denied"          if file.include?('..') || file.start_with?('/')
+    halt 404, "File not found"         unless File.exist?(file)
+
+    csv_name          = File.basename(file, '.csv')
+    type_dup, @ds_name = csv_name.split('_')
+    @type             = type_dup == 'data' ? 'dataset' : 'define.xml'
+    @last_valid_key   = params[:last_valid_key]&.split(',') || []
+    @rows             = CSV.read(file, headers: true)
+
+    erb :csv_view
+  end
+
+  # ---------------------------------------------------------------------------
+  # :section: ROUTES — Configuration (JSON API)
+  # ---------------------------------------------------------------------------
+
+  # GET /config — Returns the current configuration as JSON
+  get '/config' do
+    content_type :json
+    load_config.to_json
+  end
+
+  # POST /config — Updates editable fields in config.yaml
+  # Body: JSON object with field/value pairs
+  post '/config' do
+    content_type :json
+    begin
+      config_params = JSON.parse(request.body.read)
+      save_config(config_params)
+      { success: true, message: "Configuration updated successfully" }.to_json
+    rescue => e
+      status 500
+      { success: false, erreur: e.message }.to_json
+    end
+  end
+
+  # POST /config/reset — Reloads config.yaml from config.default.yaml
+  post '/config/reset' do
+    content_type :json
+    begin
+      unless File.exist?(Hlsv.default_config_path)
+        return { success: false, erreur: "File config.default.yaml not found in project root" }.to_json
+      end
+
+      config_default = YAML.load_file(Hlsv.default_config_path)
+      File.write(Hlsv.config_path, config_default.to_yaml)
+      { success: true, message: "Default configuration loaded from config.default.yaml" }.to_json
+    rescue => e
+      status 500
+      { success: false, erreur: e.message }.to_json
+    end
+  end
+
+  # POST /config/clear — Resets all config fields to nil (keeps structure)
+  post '/config/clear' do
+    content_type :json
+    begin
+      empty_config = {
+        'study_name'           => nil,
+        'output_type'          => 'csv',
+        'output_directory'     => nil,
+        'data_directory'       => nil,
+        'define_path'          => nil,
+        'excluded_ds'          => nil,
+        'event_key'            => nil,
+        'intervention_key'     => nil,
+        'finding_key'          => nil,
+        'finding_about_key'    => nil,
+        'ds_key'               => nil,
+        'relrec_key'           => nil,
+        'CO_key'               => nil,
+        'TA_key'               => nil,
+        'TE_key'               => nil,
+        'TI_key'               => nil,
+        'TS_key'               => nil,
+        'TV_key'               => nil
+      }
+
+      File.write(Hlsv.config_path, empty_config.to_yaml)
+      { success: true, message: "Configuration cleared" }.to_json
+    rescue => e
+      status 500
+      { success: false, erreur: e.message }.to_json
+    end
+  end
+
+  # ---------------------------------------------------------------------------
+  # :section: ROUTES — Processing
+  # ---------------------------------------------------------------------------
+
+  # POST /traiter — Validates config then runs the main processing script
+  post '/traiter' do
+    content_type :json
+    begin
+      config = load_config
+      errors = validate_config(config)
+
+      if errors.any?
+        return {
+          success: false,
+          erreur:  "Incomplete configuration",
+          details: errors
+        }.to_json
+      end
+
+      result = MonScript.executer(config)
+      { success: true, resultat: result }.to_json
+    rescue => e
+      status 500
+      { success: false, erreur: e.message }.to_json
+    end
+  end
+
+  # ---------------------------------------------------------------------------
+  # :section: ROUTES — Results browsing
+  # ---------------------------------------------------------------------------
+
+  # GET /resultats — Returns the hlsv_results/ directory tree as JSON
+  get '/resultats' do
+    content_type :json
+    results_dir = 'hlsv_results'
+
+    unless Dir.exist?(results_dir)
+      return { success: true, arborescence: {} }.to_json
+    end
+
+    arborescence = build_tree(results_dir)
+    { success: true, arborescence: arborescence }.to_json
+  end
+
+  # GET /telecharger/* — Serves a result file (inline or as download)
+  # Inline for: .html, .htm, .pdf, .png, .jpg, .jpeg, .gif, .svg, .txt, .css, .js
+  # Download for: all other extensions
+  get '/telecharger/*' do
+    relative_file = params['splat'].first
+    halt 403, "Access denied" if relative_file.include?('..') || relative_file.start_with?('/')
+
+    file_path = File.join('hlsv_results', relative_file)
+    halt 404, "File not found: #{relative_file}" unless File.exist?(file_path)
+
+    extension = File.extname(file_path).downcase
+
+    if ['.html', '.htm', '.pdf', '.png', '.jpg', '.jpeg', '.gif', '.svg', '.txt'].include?(extension)
+      send_file file_path, filename: File.basename(file_path), disposition: 'inline'
+    elsif ['.css', '.js'].include?(extension)
+      content_type extension == '.css' ? 'text/css' : 'application/javascript'
+      send_file file_path, disposition: 'inline'
+    else
+      send_file file_path, filename: File.basename(file_path)
+    end
+  end
+
+  # GET /telecharger_zip_dossier/* — Packages a result folder as a ZIP archive
+  get '/telecharger_zip_dossier/*' do
+    relative_folder = params['splat'].first
+    halt 403, "Access denied"    if relative_folder.include?('..') || relative_folder.start_with?('/')
+
+    folder_path = File.join('hlsv_results', relative_folder)
+    halt 404, "Folder not found" unless Dir.exist?(folder_path)
+
+    # Build ZIP in memory, preserving relative paths inside the folder
+    zip_data = Zip::OutputStream.write_buffer do |zip|
+      Dir.glob(File.join(folder_path, '**', '*')).each do |file_path|
+        next if File.directory?(file_path)
+
+        relative_path = file_path.sub("#{folder_path}/", '')
+        zip.put_next_entry(relative_path)
+        zip.write(File.read(file_path))
+      end
+    end
+
+    zip_data.rewind
+    content_type  'application/zip'
+    folder_name = File.basename(relative_folder)
+    attachment    "#{Time.now.strftime('%Y-%m-%d')}-#{folder_name}.zip"
+    zip_data.read
+  end
+
+  # ---------------------------------------------------------------------------
+  # :section: ROUTES — Excel exports
+  # ---------------------------------------------------------------------------
+
+  # GET /excel_export — Exports a single CSV file as a 2-sheet Excel workbook
+  # Sheet 1: README with metadata
+  # Sheet 2: Duplicate data with alternating row colors
+  # Params:
+  #   :file           — path to the CSV file
+  #   :last_valid_key — comma-separated list of last tested keys (optional)
+  get '/excel_export' do
+    file = params[:file]
+    halt 403, "Access denied"    if file.include?('..') || file.start_with?('/')
+    halt 404, "File not found"   unless File.exist?(file)
+
+    last_valid_key         = params[:last_valid_key]&.split(',') || []
+    type_dup, sheet_name   = File.basename(file, '.csv').split('_')
+    csv                    = CSV.read(file, headers: true)
+
+    workbook = FastExcel.open(constant_memory: true)
+
+    # --- Shared formats ---
+    title_format   = workbook.add_format(bold: true, font_size: 16, font_color: "#2c3e50", align: "left")
+    heading_format = workbook.add_format(bold: true, font_size: 12, font_color: "#2c3e50", bg_color: "#e8f4f8", pattern: 1, border: 1)
+    text_format    = workbook.add_format(font_size: 11, align: "left", text_wrap: true)
+    footer_format  = workbook.add_format(font_size: 9,  font_color: "#6c757d", align: "center")
+    header_format  = workbook.add_format(bold: true, bg_color: "#c0c0c0", pattern: 1, border: 1)
+    even_format    = workbook.add_format(pattern: 1, border: 1)
+    odd_format     = workbook.add_format(bg_color: "#E3F2FD", pattern: 1, border: 1)
+
+    # --- Sheet 1: README ---
+    readme_sheet = workbook.add_worksheet("README")
+    readme_sheet.set_column(0, 0, 80)
+    readme_sheet.set_column(1, 1, 20)
+
+    current_row = 0
+    readme_sheet.write_string(current_row, 0, "Duplicates Analysis Report - #{sheet_name}", title_format)
+    current_row += 2
+
+    readme_sheet.write_string(current_row, 0, "About This Report", heading_format)
+    current_row += 1
+
+    [
+      "This Excel file displays the detected duplicates, grouped according to the last key tested.",
+      "The duplicate groups are represented by a number in the 'No' column.",
+      "Alternating colors (white and light blue) are used to distinguish them visually.",
+      "All variables present in the dataset are displayed."
+    ].each { |line| readme_sheet.write_string(current_row, 0, line, text_format); current_row += 1 }
+    current_row += 1
+
+    readme_sheet.write_string(current_row, 0, "Sheet Information", heading_format)
+    current_row += 1
+
+    [
+      "• Dataset: #{sheet_name}",
+      "• Duplicate Type: #{type_dup == 'data' ? 'Duplicates in dataset' : 'Duplicates in define.xml'}",
+      "• Total Records: #{csv.size}",
+      "• Number of Variables: #{csv.headers.size}",
+      "• Generated: #{Time.now.strftime('%Y-%m-%d at %H:%M')}",
+      "• Last key tested: #{last_valid_key.join(' ')}"
+    ].each { |line| readme_sheet.write_string(current_row, 0, line, text_format); current_row += 1 }
+    current_row += 1
+
+    readme_sheet.write_string(current_row, 0, "Next Steps", heading_format)
+    current_row += 1
+
+    [
+      "→ Click on the '#{sheet_name}' tab at the bottom to view the duplicate records.",
+      "→ Identify variables to add in the configuration form to remove this duplicate.",
+      "→ Highlight issues with data cleaning."
+    ].each { |line| readme_sheet.write_string(current_row, 0, line, text_format); current_row += 1 }
+    current_row += 1
+
+    readme_sheet.write_string(current_row, 0,
+      "© #{Time.now.year} AdClin. All rights reserved. | Licensed under AGPL v3",
+      footer_format)
+
+    # --- Sheet 2: Duplicate data ---
+    data_sheet  = workbook.add_worksheet(sheet_name)
+    col_widths  = csv.headers.map { |h| h.to_s.length }
+
+    # Write frozen header row
+    csv.headers.each_with_index { |h, col| data_sheet.write_string(0, col, h, header_format) }
+    data_sheet.freeze_panes(1, 5)
+
+    # Write data rows with alternating colors based on the 'No' column (col 0)
+    csv.each_with_index do |row, row_index|
+      fmt = row[0].to_i.even? ? even_format : odd_format
+      row.fields.each_with_index do |value, col|
+        value_str = value.to_s
+        data_sheet.write_string(row_index + 1, col, value_str, fmt)
+        col_widths[col] = [col_widths[col], value_str.length].max
+      end
+    end
+
+    col_widths.each_with_index { |width, col| data_sheet.set_column(col, col, width + 2) }
+
+    # Send file
+    type      = type_dup == 'data' ? 'duplicate_in_dataset' : 'duplicate_in_define'
+    file_name = "#{Time.now.strftime('%Y-%m-%d')}-#{type}.xlsx"
+    content_type 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet'
+    attachment file_name
+    workbook.read_string
+  end
+
+  # GET /telecharger_dossier_excel/* — Exports all CSV files in a folder as a multi-sheet Excel workbook
+  # Sheet 1: README with folder metadata
+  # One additional sheet per CSV file
+  get '/telecharger_dossier_excel/*' do
+    relative_folder = params['splat'].first
+    halt 403, "Access denied"              if relative_folder.include?('..') || relative_folder.start_with?('/')
+
+    folder_path = File.join('hlsv_results', relative_folder)
+    halt 404, "Folder not found"           unless Dir.exist?(folder_path)
+
+    csv_files = Dir.glob(File.join(folder_path, '*.csv')).sort
+    halt 404, "No CSV files found in this folder" if csv_files.empty?
+
+    workbook    = FastExcel.open(constant_memory: true)
+    folder_name = File.dirname(relative_folder)
+
+    # --- Shared formats ---
+    title_format   = workbook.add_format(bold: true, font_size: 16, font_color: "#2c3e50", align: "left")
+    heading_format = workbook.add_format(bold: true, font_size: 12, font_color: "#2c3e50", bg_color: "#e8f4f8", pattern: 1, border: 1)
+    text_format    = workbook.add_format(font_size: 11, align: "left", text_wrap: true)
+    footer_format  = workbook.add_format(font_size: 9,  font_color: "#6c757d", align: "center")
+    header_format  = workbook.add_format(bold: true, bg_color: "#c0c0c0", pattern: 1, border: 1)
+    even_format    = workbook.add_format(pattern: 1, border: 1)
+    odd_format     = workbook.add_format(bg_color: "#E3F2FD", pattern: 1, border: 1)
+
+    # --- Sheet 1: README ---
+    readme_sheet = workbook.add_worksheet("README")
+    readme_sheet.set_column(0, 0, 80)
+    readme_sheet.set_column(1, 1, 20)
+
+    current_row = 0
+    readme_sheet.write_string(current_row, 0, "Duplicates Analysis Report - #{folder_name}", title_format)
+    current_row += 2
+
+    readme_sheet.write_string(current_row, 0, "About This Report", heading_format)
+    current_row += 1
+
+    [
+      "This Excel file displays the detected duplicates, grouped according to the last key tested.",
+      "The duplicate groups are represented by a number in the 'No' column.",
+      "Alternating colors (white and light blue) are used to distinguish them visually.",
+      "All variables present in the dataset are displayed."
+    ].each { |line| readme_sheet.write_string(current_row, 0, line, text_format); current_row += 1 }
+    current_row += 1
+
+    readme_sheet.write_string(current_row, 0, "Workbook Information", heading_format)
+    current_row += 1
+
+    [
+      "• Folder: #{folder_name}",
+      "• Number of Datasets: #{csv_files.size}",
+      "• Generated: #{Time.now.strftime('%Y-%m-%d at %H:%M')}"
+    ].each { |line| readme_sheet.write_string(current_row, 0, line, text_format); current_row += 1 }
+    current_row += 1
+
+    readme_sheet.write_string(current_row, 0, "Next Steps", heading_format)
+    current_row += 1
+
+    [
+      "→ Each tab corresponds to one CSV file from the folder.",
+      "→ Review duplicate records and identify variables to add in the configuration.",
+      "→ Highlight issues with data cleaning."
+    ].each { |line| readme_sheet.write_string(current_row, 0, line, text_format); current_row += 1 }
+    current_row += 1
+
+    readme_sheet.write_string(current_row, 0,
+      "© #{Time.now.year} AdClin. All rights reserved. | Licensed under AGPL v3",
+      footer_format)
+
+    # --- Data sheets: one per CSV file ---
+    csv_files.each do |csv_file|
+      sheet_name = File.basename(csv_file, '.csv')
+      sheet_name = sheet_name[0..30] if sheet_name.length > 31  # Excel sheet name limit: 31 chars
+
+      csv_data = CSV.read(csv_file, headers: true)
+      next if csv_data.empty?
+
+      data_sheet = workbook.add_worksheet(sheet_name)
+      col_widths  = csv_data.headers.map { |h| h.to_s.length }
+
+      # Write frozen header row
+      csv_data.headers.each_with_index { |header, col| data_sheet.write_string(0, col, header.to_s, header_format) }
+      data_sheet.freeze_panes(1, 5)
+
+      # Write data rows with alternating colors based on the 'No' column (col 0)
+      csv_data.each_with_index do |row, row_index|
+        fmt = row[0].to_i.even? ? even_format : odd_format
+        row.fields.each_with_index do |value, col|
+          value_str = value.to_s
+          data_sheet.write_string(row_index + 1, col, value_str, fmt)
+          col_widths[col] = [col_widths[col], value_str.length].max
+        end
+      end
+
+      col_widths.each_with_index { |width, col| data_sheet.set_column(col, col, width + 2) }
+    end
+
+    # Send file
+    content_type 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet'
+    attachment   "#{Time.now.strftime('%Y-%m-%d')}_#{folder_name}.xlsx"
+    workbook.read_string
+  end
+
+  # ---------------------------------------------------------------------------
+  # :section: ROUTES — HTML to Word conversion
+  # ---------------------------------------------------------------------------
+
+  # GET /telecharger_html_word/* — Converts an HTML result file to a .docx Word document
+  # Returns JSON with success status and generated filename
+  get '/telecharger_html_word/*' do
+    content_type :json
+    chemin_relatif = params['splat'].first
+
+    halt 403, { success: false, erreur: 'Accès refusé' }.to_json \
+      if chemin_relatif.include?('..') || chemin_relatif.start_with?('/')
+
+    chemin_absolu = File.join('hlsv_results', chemin_relatif)
+
+    halt 400, { success: false, erreur: 'Fichier HTML introuvable' }.to_json \
+      unless File.exist?(chemin_absolu) && chemin_absolu.end_with?('.html')
+
+    output_docx = chemin_absolu.sub(/\.html$/, '.docx')
+
+    puts "Parsing #{chemin_absolu}..."
+    blocks = RiReportParser.new(chemin_absolu).parse
+    puts "  -> #{blocks.size} blocks extracted"
+
+    puts "Building #{output_docx}..."
+    DocxWriter.new(output_docx).write(blocks)
+    puts "  -> Done: #{output_docx}"
+
+    { success: true, message: "Word généré : #{File.basename(output_docx)}" }.to_json
+  end
+
+  # ---------------------------------------------------------------------------
+  # :section: HELPERS
+  # ---------------------------------------------------------------------------
+
+  helpers do
+
+    # Recursively builds a directory tree hash for a given base path.
+    # Returns: { fichiers: [...], dossiers: { name => subtree, ... } }
+    def build_tree(base_path, relative_path = '')
+      tree      = { fichiers: [], dossiers: {} }
+      full_path = relative_path.empty? ? base_path : File.join(base_path, relative_path)
+
+      return tree unless Dir.exist?(full_path)
+
+      Dir.foreach(full_path) do |entry|
+        next if entry == '.' || entry == '..'
+
+        entry_path          = File.join(full_path, entry)
+        relative_entry_path = relative_path.empty? ? entry : File.join(relative_path, entry)
+
+        if File.directory?(entry_path)
+          tree[:dossiers][entry] = build_tree(base_path, relative_entry_path)
+        else
+          tree[:fichiers] << {
+            nom:       entry,
+            chemin:    relative_entry_path,
+            taille:    File.size(entry_path),
+            date:      File.mtime(entry_path).strftime('%Y-%m-%d %H:%M:%S'),
+            extension: File.extname(entry)
+          }
+        end
+      end
+
+      tree[:fichiers].sort_by! { |f| f[:nom] }
+      tree
+    end
+
+    # Loads and returns config.yaml as a Hash. Halts with 500 if file is missing.
+    def load_config
+      if File.exist?(Hlsv.config_path)
+        YAML.load_file(Hlsv.config_path) || {}
+      else
+        halt 500, "File config.yaml not found"
+      end
+    end
+
+    # Merges config_params into the existing config.yaml, for editable fields only.
+    # output_type is always forced to 'csv'.
+    def save_config(config_params)
+      current_config = File.exist?(Hlsv.config_path) ? (YAML.load_file(Hlsv.config_path) || {}) : {}
+
+      editable_fields = %w(
+        study_name output_directory data_directory define_path excluded_ds
+        event_key intervention_key finding_key finding_about_key
+        ds_key relrec_key CO_key TA_key TE_key TI_key TS_key TV_key
+      )
+
+      editable_fields.each do |field|
+        current_config[field] = config_params[field] if config_params.key?(field)
+      end
+
+      current_config['output_type'] = 'csv'
+      File.write(Hlsv.config_path, current_config.to_yaml)
+    end
+
+    # Validates the configuration hash.
+    # Checks all required fields are present and validates filesystem paths.
+    # Returns an array of error messages (empty if config is valid).
+    def validate_config(config)
+      errors = []
+
+      required_fields = {
+        'study_name'        => 'Study name',
+        'output_directory'  => 'Output directory',
+        'data_directory'    => 'Data directory',
+        'define_path'       => 'Define.xml path',
+        'event_key'         => 'Event key',
+        'intervention_key'  => 'Intervention key',
+        'finding_key'       => 'Finding key',
+        'finding_about_key' => 'Finding about key',
+        'ds_key'            => 'DS key',
+        'relrec_key'        => 'RELREC key',
+        'CO_key'            => 'CO key',
+        'TA_key'            => 'TA key',
+        'TE_key'            => 'TE key',
+        'TI_key'            => 'TI key',
+        'TS_key'            => 'TS key',
+        'TV_key'            => 'TV key'
+      }
+
+      # Check all required fields are filled
+      required_fields.each do |key, name|
+        value = config[key]
+        errors << "#{name} is empty" if value.nil? || value.to_s.strip.empty?
+      end
+
+      # Check data_directory exists and contains at least one .xpt file
+      if config['data_directory'] && !config['data_directory'].to_s.strip.empty?
+        dir = config['data_directory'].gsub('\\', '/')
+        errors << "Data directory does not exist: #{dir}"       unless Dir.exist?(dir)
+        errors << "Directory is empty, no .xpt files detected"  if Dir["#{dir}/*"].none? { |f| File.extname(f) == '.xpt' }
+      end
+
+      # Check define_path exists (skip if value is '-', meaning no define file)
+      if config['define_path'] && config['define_path'] != '-'
+        errors << "Invalid path: #{config['define_path']}" unless File.exist?(config['define_path'])
+      end
+
+      config['output_type'] = 'csv'
+      errors
+    end
+
+  end # helpers
+
+end # class WebApp
+end # module Hlsv

data/lib/hlsv/xpt/dataset.rb

@@ -0,0 +1,38 @@
+class SAS
+module XPT
+
+class Dataset
+
+  attr_reader :name
+  attr_reader :label
+  attr_reader :type
+  attr_reader :create_date
+  attr_reader :modify_date
+  attr_reader :sas_version
+  attr_reader :sas_os
+
+  # array of Variable
+  attr_reader :variables
+
+  # array of arrays of values, each array has the same number of elements at variables.size
+  attr_reader :observations
+
+  def initialize(name, label, type, create_date, modify_date, sas_version, sas_os)
+    @name = name
+    @label = label
+    @type = type
+    @create_date = create_date
+    @modify_date = modify_date
+    @sas_version = sas_version
+    @sas_os = sas_os
+    @variables = []
+    @observations = []
+  end
+
+  def obs_record_length
+    variables.last.position + variables.last.length
+  end
+
+end
+end
+end

data/lib/hlsv/xpt/library.rb

@@ -0,0 +1,28 @@
+class SAS
+module XPT
+
+##
+# All datasets in a .xpt file.
+
+class Library
+
+  attr_reader :source_file_path
+  attr_reader :create_date
+  attr_reader :modify_date
+  attr_reader :sas_version
+  attr_reader :sas_os
+
+  attr_reader :datasets
+
+  def initialize(source_file_path, create_date, modify_date, sas_version, sas_os)
+    @source_file_path = source_file_path
+    @create_date = create_date
+    @modify_date = modify_date
+    @sas_version = sas_version
+    @sas_os = sas_os
+    @datasets = []
+  end
+
+end
+end
+end