1. import sys
  3. g = {'#': []} # graph/tree
  4. cur_lex = '' # current continuation lexicon
  5. lines = [] # cache of the lines
  6. nodes = {'#': 0} # node to id mapping
  7. i = 1 # current id
  9. # First read through the lexc file and make a list of the continuation lexica
  10. for line in sys.stdin.readlines():
  11. line = line.strip()
  12. if line == '': continue
  13. if line[0] == '!': continue
  14. row = line.replace('% ', '%<SPACE>').replace(' ;', ';').replace(';', ' ;').split(' ')
  15. lines.append(row)
  16. if row[0] == 'LEXICON':
  17. if row[1] not in g:
  18. g[row[1]] = []
  19. nodes[row[1]] = i
  20. i += 1
  21. continue
  23. # Now read through again and do a map of lexicon -> outgoing lexicons
  24. cur_lex = ''
  25. for row in lines:
  26. s = 0
  27. if row[0] == 'LEXICON':
  28. cur_lex = row[1]
  29. continue
  30. if row[0].count(':') > 0 and row[1] != '' and row[1] != ';':
  31. s = 1
  32. g[cur_lex].append(row[1]);
  33. elif len(row) >= 2 and row[1] == ';' and row[0] not in ['',';'] and row[0].count('<') == 0:
  34. s = 2
  35. g[cur_lex].append(row[0]);
  36. # print('!', s, cur_lex, row, g, file=sys.stderr)
  39. # Print out the graph
  40. print('digraph G { rankdir="LR"')
  41. print('node [fontname="Tahoma",shape=box,fontsize=14,fixedsize=false,fillcolor="grey",style=filled]')
  42. print('edge [fontname="FreeMono",fontsize=14]')
  43. for lex in g:
  44. print('%s [label="%s"];' % (nodes[lex], lex))
  45. g[lex] = list(set(g[lex]))
  46. for cont in g[lex]:
  47. print('@@', lex, cont, file=sys.stderr)
  48. print('%s -> %s ;' % (nodes[lex], nodes[cont]))
  49. print('}')