Module: SwissmedicDiff::Diff

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

Constant Summary collapse

COLUMNS = 
[ :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 ]
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',
}
GALFORM_P = 
%r{excipiens\s+(ad|pro)\s+(?<galform>((?!\bpro\b)[^.])+)}

Instance Method Summary collapse

Instance Method Details

#_comparable(key, row, idx) ⇒ Object 



267
268
269
270
271
272
273
274
275
276
277
278
 
# File 'lib/swissmedic-diff.rb', line 267

def _comparable(key, row, idx)
  if cell = row[idx]
    case key
    when :registration_date, :expiry_date
      Spreadsheet.date_cell(row, idx)
    when :seqnr
      sprintf "%02i", cell.to_i
    else
      cell(row, idx).downcase.gsub(/\s+/, "")
    end
  end
end

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



180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
 
# File 'lib/swissmedic-diff.rb', line 180

def _known_data(latest, known_regs, known_seqs, known_pacs, newest_rows)
  lbook = Spreadsheet.open(latest)
  idx, prr, prp = nil
  multiples = {}
  each_valid_row(lbook) { |row|
    iksnr = cell(row, column(:iksnr))
    seqnr = cell(row, column(:seqnr))
    pacnr = cell(row, column(:ikscd))
    multiples[iksnr] ||= {}
    if prr == iksnr && prp == pacnr
      idx += 1
    elsif previous = multiples[iksnr][pacnr]
      prr = iksnr
      prp = pacnr
      idx = previous[COLUMNS.size].to_i + 1
    else
      prr = iksnr
      prp = pacnr
      idx = 0
    end
    multiples[iksnr][pacnr] = row
    row[COLUMNS.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 



250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
 
# File 'lib/swissmedic-diff.rb', line 250

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 



53
54
55
 
# File 'lib/swissmedic-diff.rb', line 53

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

#cell(row, pos) ⇒ Object 



56
57
58
59
60
 
# File 'lib/swissmedic-diff.rb', line 56

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

#column(key) ⇒ Object 



61
62
63
 
# File 'lib/swissmedic-diff.rb', line 61

def column(key)
  COLUMNS.index(key)
end

#describe(diff, iksnr) ⇒ Object 



64
65
66
 
# File 'lib/swissmedic-diff.rb', line 64

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

#describe_flag(diff, iksnr, flag) ⇒ Object 



67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
 
# File 'lib/swissmedic-diff.rb', line 67

def describe_flag(diff, iksnr, flag)
  txt = FLAGS.fetch(flag, flag)
  case flag
  when :sequence
  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].sort.first.last
    sprintf "%s (%s)", txt, row[column(flag)].strftime('%d.%m.%Y')
  else
    row = diff.newest_rows[iksnr].sort.first.last
    sprintf "%s (%s)", txt, cell(row, column(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)



94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
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
 
# File 'lib/swissmedic-diff.rb', line 94

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
  Spreadsheet.client_encoding = 'UTF-8'
  tbook = Spreadsheet.open(target)
  sheet = tbook.worksheet(0)
  idx, prr, prp = nil
  multiples = {}
  each_valid_row(tbook) { |row|
    iksnr = cell(row, column(:iksnr))
    seqnr = cell(row, column(:seqnr))
    pacnr = cell(row, column(:ikscd))
    (multiples[iksnr] ||= {})
    if prr == iksnr && prp == pacnr
      idx += 1
    elsif previous = multiples[iksnr][pacnr]
      prr = iksnr
      prp = pacnr
      idx = previous[COLUMNS.size].to_i + 1
    else
      prr = iksnr
      prp = pacnr
      idx = 0
    end
    row[COLUMNS.size] = idx
    (newest_rows[iksnr] ||= {})[pacnr] = row
    multiples[iksnr][pacnr] = row
    if(other = known_regs.delete([iksnr]))
      changes[iksnr] ||= []
    else
      changes[iksnr] ||= [: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, column(:size)),
                            cell(row, column(: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, column(:seqnr)).to_i,
                  cell(row, column(:size)), cell(row, column(: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, column(:seqnr)).to_i
    key
  }
  @diff.sequence_deletions = known_seqs.keys
  @diff.registration_deletions = known_regs.keys
  @diff
end

#each_valid_row(spreadsheet) ⇒ 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]}" }
spreadsheet

spreadsheet to operate on

return 


293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
 
# File 'lib/swissmedic-diff.rb', line 293

def each_valid_row(spreadsheet)
  skipRows = rows_to_skip(spreadsheet)
  worksheet = spreadsheet.worksheet(0)
  row_nr = 0
  worksheet.each() {
    |row|
    row_nr += 1
    next if row_nr <= skipRows
    if row.size < COLUMNS.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(:production_science)) == 'Tierarzneimittel')
    row[column(:iksnr)] = "%05i" % cell(row, column(:iksnr)).to_i
    row[column(:seqnr)] = "%02i" % cell(row, column(:seqnr)).to_i
    row[column(:ikscd)] = "%03i" % cell(row, column(:ikscd)).to_i
    yield row
  }
end

#format_flags(flags) ⇒ Object 



166
167
168
169
170
171
 
# File 'lib/swissmedic-diff.rb', line 166

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

#known_data(latest) ⇒ Object 



172
173
174
175
176
177
178
179
 
# File 'lib/swissmedic-diff.rb', line 172

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 



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

def name(diff, iksnr)
  rows = diff.newest_rows[iksnr]
  row = rows.sort.first.last
  cell(row, column(:name_base))
end

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



213
214
215
216
217
218
219
220
221
222
223
224
 
# File 'lib/swissmedic-diff.rb', line 213

def rows_diff(row, other, ignore = [])
  flags = []
  COLUMNS.each_with_index { |key, idx|
    if(!ignore.include?(key) \
       && _comparable(key, row, idx) != _comparable(key, other, idx))
      # binding.pry if key == :expiry_date

      flags.push key
    end
  }
  flags
end

#rows_to_skip(spreadsheet) ⇒ Object 



313
314
315
316
317
318
319
320
321
322
323
324
 
# File 'lib/swissmedic-diff.rb', line 313

def rows_to_skip(spreadsheet)
  # Packungen.xls of swissmedic before October 2013 had  3 leading rows
  # Packungen.xls of swissmedic after  October 2013 have 4 leading rows
  j = 0
  while true
    cell = spreadsheet.worksheet(0).row(j)[0]
    cell = cell.value if cell.is_a?(RubyXL::Cell)
    break if cell.respond_to?(:to_i) and 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)



233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
 
# File 'lib/swissmedic-diff.rb', line 233

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