Module: SwissmedicDiff::Diff

Included in:
SwissmedicDiff
Defined in:
lib/swissmedic-diff.rb

Constant Summary collapse

COLUMNS_OLD = 
[:iksnr, :seqnr, :name_base, :company,
:index_therapeuticus, :atc_class, :production_science,
:registration_date, :sequence_date, :expiry_date, :ikscd,
:size, :unit, :ikscat, :substances, :composition,
:indication_registration, :indication_sequence]
COLUMNS_FEBRUARY_2019 = 
{
  iksnr: /Zulassungs-Nummer/i,                  # column-nr: 0
  seqnr: /Dosisstärke-nummer/i,
  name_base: /Bezeichnung des Arzneimittels/i,
  company: /Zulassungsinhaberin/i,
  production_science: /Heilmittelcode/i,
  index_therapeuticus: /IT-Nummer/i,            # column-nr: 5
  atc_class: /ATC-Code/i,
  registration_date: /Erstzul.datum Arzneimittel/i,
  sequence_date: /Zul.datum Dosisstärke/i,
  expiry_date: /Gültigkeitsdauer der Zulassung/i,
  ikscd: /Packungscode/i,                 # column-nr: 10
  size: /Packungsgrösse/i,
  unit: /Einheit/i,
  ikscat: /Abgabekategorie Packung/i,
  ikscat_seq: /Abgabekategorie Dosisstärke/i,
  ikscat_preparation: /Abgabekategorie Arzneimittel/i, # column-nr: 15
  substances: /Wirkstoff/i,
  composition: /Zusammensetzung/i,
  composition_AMZV: /Volldeklaration rev. AMZV umgesetzt/i,
  indication_registration: /Anwendungsgebiet Arzneimittel/i,
  indication_sequence: /Anwendungsgebiet Dosisstärke/i, # column-nr 20
  gen_production: /Gentechnisch hergestellte Wirkstoffe/i,
  insulin_category: /Kategorie bei Insulinen/i,
  # swissmedi corrected in february 2018 the typo  betäubunsmittel to  betäubungsmittel-
  drug_index: /Verz. bei betäubungsmittel-haltigen Arzneimittel/i
}
FLAGS = 
{
  new: "Neues Produkt",
  name_base: "Namensänderung",
  ikscat: "Abgabekategorie",
  index_therapeuticus: "Index Therapeuticus",
  indication_registration: "Anwendungsgebiet Präparate",
  indication_sequence: "Anwendungsgebiet Sequenz",
  company: "Zulassungsinhaber",
  composition: "Zusammensetzung",
  sequence: "Packungen",
  size: "Packungsgrösse",
  expiry_date: "Ablaufdatum der Zulassung",
  registration_date: "Erstzulassungsdatum",
  sequence_date: "Zulassungsdatum Sequenz",
  delete: "Das Produkt wurde gelöscht",
  replaced_package: "Packungs-Nummer",
  substances: "Wirkstoffe",
  production_science: "Heilmittelcode",
  atc_class: "ATC-Code"
}
DATE_FORMAT = 
"%Y:%m:%d"

Instance Method Summary collapse

Instance Method Details

#_comparable(key, row, idx) ⇒ Object 



323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
 
# File 'lib/swissmedic-diff.rb', line 323

def _comparable(key, row, idx)
  if row[idx]
    case key
    when :registration_date, :expiry_date
      if row[idx] && REGEXP_UNLIMITED.match(row[idx].to_s)
        VALUE_UNLIMITED # Date.new(2099,12,31)
      else
        row[idx]
      end
    when :seqnr
      sprintf "%02i", cell(row, idx).to_i
    else
      cell(row, idx).downcase.gsub(/\s+/, "")
    end
  end
end

#_known_data(latest, known_regs, known_seqs, known_pacs, newest_rows) ⇒ Object 



224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
 
# File 'lib/swissmedic-diff.rb', line 224

def _known_data(latest, known_regs, known_seqs, known_pacs, newest_rows)
  sheet = SimpleXlsxReader.open(latest).sheets.first
  rows = sheet.rows.slurp
  @latest_keys = get_column_indices(rows).keys
  idx, prr, prp = nil
  multiples = {}

  each_valid_row(rows) { |row|
    iksnr = cell(row, @latest_keys.index(:iksnr))
    seqnr = cell(row, @latest_keys.index(:seqnr))
    pacnr = cell(row, @latest_keys.index(:ikscd))
    multiples[iksnr] ||= {}
    if prr == iksnr && prp == pacnr
      idx += 1
    elsif (previous = multiples[iksnr][pacnr])
      prr = iksnr
      prp = pacnr
      idx = previous[@latest_keys.size].to_i + 1
    else
      prr = iksnr
      prp = pacnr
      idx = 0
    end
    multiples[iksnr][pacnr] = row
    row[@latest_keys.size] = idx
    known_regs.store [iksnr], row
    known_seqs.store [iksnr, seqnr], row
    known_pacs.store [iksnr, pacnr, idx], row
    (newest_rows[iksnr] ||= {})[pacnr] = row
  }
end

#_sort_by(sort, iksnr, flags) ⇒ Object 



305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
 
# File 'lib/swissmedic-diff.rb', line 305

def _sort_by(sort, iksnr, flags)
  case sort
  when :name
    [name(@diff, iksnr), iksnr]
  when :registration
    iksnr
  else
    weight = if flags.include? :new
      0
    elsif flags.include? :delete
      1
    else
      2
    end
    [weight, iksnr]
  end
end

#capitalize(string) ⇒ Object 



80
81
82
 
# File 'lib/swissmedic-diff.rb', line 80

def capitalize(string)
  string.split(/\s+/).collect { |word| word.capitalize }.join(" ")
end

#cell(row, pos) ⇒ Object 



84
85
86
87
88
 
# File 'lib/swissmedic-diff.rb', line 84

def cell(row, pos)
  if (cell = row[pos])
    cell.to_s
  end
end

#describe(diff, iksnr) ⇒ Object 



90
91
92
 
# File 'lib/swissmedic-diff.rb', line 90

def describe(diff, iksnr)
  sprintf("%s: %s", iksnr, name(diff, iksnr))
end

#describe_flag(diff, iksnr, flag) ⇒ Object 



94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
 
# File 'lib/swissmedic-diff.rb', line 94

def describe_flag(diff, iksnr, flag)
  txt = FLAGS.fetch(flag, flag)
  case flag
  when :sequence
    txt
  when :replaced_package
    pairs = diff.newest_rows[iksnr].collect { |rep, row|
      if (old = diff.replacements[row])
        [old, rep].join(" -> ")
      end
    }.compact
    sprintf "%s (%s)", txt, pairs.join(",")
  when :registration_date, :expiry_date
    row = diff.newest_rows[iksnr].min.last
    if row[COLUMNS_FEBRUARY_2019.keys.index(flag)].to_s.match(REGEXP_UNLIMITED)
      sprintf "%s (%s)", txt, "unbegrenzt"
    else
      sprintf "%s (%s)", txt, row[COLUMNS_FEBRUARY_2019.keys.index(flag)].strftime("%d.%m.%Y")
    end
  else
    row = diff.newest_rows[iksnr].min.last
    sprintf "%s (%s)", txt, cell(row, COLUMNS_FEBRUARY_2019.keys.index(flag))
  end
end

#diff(target, latest, ignore = []) ⇒ Object

Comparison two Excel files

target

new file path (String)

latest

old file path (String)

ignore

columns not to be compared (Symbol)

return

differences (OpenStruct class)



126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
 
# File 'lib/swissmedic-diff.rb', line 126

def diff(target, latest, ignore = [])
  replacements = {}
  known_regs, known_seqs, known_pacs, newest_rows = known_data(latest)
  @diff = OpenStruct.new
  @diff.news = news = []
  @diff.updates = updates = []
  @diff.changes = changes = {}
  @diff.newest_rows = newest_rows
  t_rows = SimpleXlsxReader.open(target).sheets.first.rows.slurp
  idx, prr, prp = nil
  multiples = {}
  l_rows = SimpleXlsxReader.open(latest).sheets.first.rows.slurp
  @latest_keys = get_column_indices(l_rows).keys
  @target_keys = get_column_indices(t_rows).keys
  each_valid_row(t_rows) { |row|
    iksnr = cell(row, @target_keys.index(:iksnr))
    seqnr = cell(row, @target_keys.index(:seqnr))
    pacnr = cell(row, @target_keys.index(:ikscd))
    multiples[iksnr] ||= {}
    if prr == iksnr && prp == pacnr
      idx += 1
    elsif (previous = multiples[iksnr][pacnr])
      prr = iksnr
      prp = pacnr
      idx = previous[@target_keys.size].to_i + 1
    else
      prr = iksnr
      prp = pacnr
      idx = 0
    end
    row[@target_keys.size] = idx
    (newest_rows[iksnr] ||= {})[pacnr] = row
    multiples[iksnr][pacnr] = row
    changes[iksnr] ||= if known_regs.delete([iksnr])
      []
    else
      [:new]
    end
    known_seqs.delete([iksnr, seqnr])
    if (other = known_pacs.delete([iksnr, pacnr, idx]))
      flags = rows_diff(row, other, ignore)
      (changes[iksnr].concat flags).uniq!
      updates.push row unless flags.empty?
    else
      replacements.store [iksnr, seqnr, cell(row, @target_keys.index(:size)),
        cell(row, @target_keys.index(:unit))], row
      flags = changes[iksnr]
      flags.push(:sequence).uniq! unless flags.include? :new
      news.push row
    end
  }
  @diff.replacements = reps = {}
  known_pacs.each { |(iksnr, pacnr), row|
    key = [iksnr, "%02i" % cell(row, @target_keys.index(:seqnr)).to_i,
      cell(row, @target_keys.index(:size)), cell(row, @target_keys.index(:unit))]
    if (rep = replacements[key])
      changes[iksnr].push :replaced_package
      reps.store rep, pacnr
    end
  }
  known_regs.each_key { |(iksnr, _)| changes[iksnr] = [:delete] }
  changes.delete_if { |iksnr, flags| flags.empty? }
  @diff.package_deletions = known_pacs.collect { |key, row|
    ## the keys in known_pacs don't include the sequence number (which
    #  would prevent us from properly recognizing multi-sequence-Packages),
    #  so we need complete the path to the package now
    key[1, 0] = "%02i" % cell(row, @target_keys.index(:seqnr)).to_i
    key
  }
  @diff.sequence_deletions = known_seqs.keys
  @diff.registration_deletions = known_regs.keys
  @@stat = {}
  @@stat[:target] = "#{File.basename(target)} #{File.size(target)} bytes"
  @@stat[:latest] = "#{File.basename(latest)} #{File.size(latest)} bytes"
  @diff.to_h.keys.each { |name| @diff.instance_eval("@@stat[:#{name}] = #{name}.size", __FILE__, __LINE__) }
  @diff
end

#each_valid_row(rows) ⇒ Object

iterate over all valid rows of a swissmedic Packungen.xls

Iterates over all rows, ignoring Tierarzneimittel and lines with not enough data Patches the fields :iksnr, :seqnr, :ikscd to match the old swissmedic convention of a fixed sized string

example:

SwissmedicDiff.new.each_valid_row(Spreadsheet.open('path/to/file')) { |x| puts "iksnr #{x[0]}" }
rows

rows to operate on

return 


368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
 
# File 'lib/swissmedic-diff.rb', line 368

def each_valid_row(rows)
  skip_rows = rows_to_skip(rows)
  column_keys = get_column_indices(rows).keys
  row_nr = 0
  rows.each { |row|
    row_nr += 1
    next if row_nr <= skip_rows
    break unless row
    if row.size < column_keys.size / 2
      $stdout.puts "Data missing in \n(line " + row_nr.to_s + "): " + row.join(", ").to_s + "\n"
      next
    end
    next if cell(row, column_keys.index(:production_science)) == "Tierarzneimittel"
    row[column_keys.index(:iksnr)] = "%05i" % cell(row, column_keys.index(:iksnr)).to_i
    row[column_keys.index(:seqnr)] = "%02i" % cell(row, column_keys.index(:seqnr)).to_i
    row[column_keys.index(:ikscd)] = "%03i" % cell(row, column_keys.index(:ikscd)).to_i
    yield row
  }
end

#format_flags(flags) ⇒ Object 



208
209
210
211
212
213
 
# File 'lib/swissmedic-diff.rb', line 208

def format_flags(flags)
  flags.delete(:revision)
  flags.collect { |flag|
    "- %s\n" % FLAGS.fetch(flag, "Unbekannt (#{flag})")
  }.compact.join
end

#get_column_indices(rows) ⇒ Object 



340
341
342
343
344
345
346
347
348
349
350
351
352
353
 
# File 'lib/swissmedic-diff.rb', line 340

def get_column_indices(rows)
  header_row_id = rows_to_skip(rows) - 1
  rows[header_row_id]
  row = rows[5] # Headers are found at row 5 since February 5
  0.upto(COLUMNS_FEBRUARY_2019.size - 1).each { |idx| puts "#{idx}: #{row[idx]}" } if $VERBOSE
  COLUMNS_FEBRUARY_2019.each { |key, value|
    header_name = row[COLUMNS_FEBRUARY_2019.keys.index(key)].to_s
    unless value.match(header_name)
      puts "#{__LINE__}: #{key} ->  #{COLUMNS_FEBRUARY_2019.keys.index(key)} #{value}\nbut was  #{header_name}" if $VERBOSE
      break
    end
  }
  COLUMNS_FEBRUARY_2019
end

#known_data(latest) ⇒ Object 



215
216
217
218
219
220
221
222
 
# File 'lib/swissmedic-diff.rb', line 215

def known_data(latest)
  known_regs = {}
  known_seqs = {}
  known_pacs = {}
  newest_rows = {}
  _known_data latest, known_regs, known_seqs, known_pacs, newest_rows
  [known_regs, known_seqs, known_pacs, newest_rows]
end

#name(diff, iksnr) ⇒ Object 



256
257
258
259
260
 
# File 'lib/swissmedic-diff.rb', line 256

def name(diff, iksnr)
  rows = diff.newest_rows[iksnr]
  row = rows.min.last
  cell(row, COLUMNS_FEBRUARY_2019.keys.index(:name_base))
end

#rows_diff(row, other, ignore = []) ⇒ Object 



262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
 
# File 'lib/swissmedic-diff.rb', line 262

def rows_diff(row, other, ignore = [])
  flags = []
  COLUMNS_OLD.each_with_index { |key, idx|
    if !ignore.include?(key)
      left = _comparable(key, row, @target_keys.index(key))
      right = _comparable(key, other, @latest_keys.index(key))
      next if left.is_a?(Date) && right.is_a?(Date) && left.strftime(DATE_FORMAT).eql?(right.strftime(DATE_FORMAT))
      next if left.is_a?(String) && left.empty? && !right
      next if right.is_a?(String) && right.empty? && !left
      if left != right
        flags.push key
      end
    end
  }
  flags
end

#rows_to_skip(rows) ⇒ Object 



388
389
390
391
392
393
394
395
396
 
# File 'lib/swissmedic-diff.rb', line 388

def rows_to_skip(rows)
  j = 0
  loop do
    cell = rows[j][0]
    break if cell.respond_to?(:to_i) && cell.to_i != 0
    j += 1
  end
  j
end

#to_s(sort = :group) ⇒ Object

Output the differencies with String

This should be called after diff method.

sort

sort key (:group | :name | :registration)

return

difference (String)



286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
 
# File 'lib/swissmedic-diff.rb', line 286

def to_s(sort = :group)
  @diff ||= nil
  return "" unless @diff
  @diff.changes.sort_by { |iksnr, flags|
    _sort_by(sort, iksnr, flags)
  }.collect { |iksnr, flags|
    if flags.include? :new
      "+ " + describe(@diff, iksnr)
    elsif flags.include? :delete
      "- " + describe(@diff, iksnr)
    else
      "> " + describe(@diff, iksnr) << "; " \
        + flags.collect { |flag|
            describe_flag(@diff, iksnr, flag)
          }.compact.join(", ")
    end
  }.join("\n")
end