import glob, collections, shelve
from utils_wb import workbook

class icol(object):
  def __init__(self):
    self.cc = collections.defaultdict( list )

class loadSpatial(object):
  def __init__(self):
    wb = workbook( '../workbook/inputs/spatialShape_ref.xls' )
    s1 = wb.book.sheet_by_name(u'ss_ref')
    self.ss = {}
    self.labByUid = {}
    self.uidByLab = {}
    uu = set()
    for i in range(s1.nrows):
      rr = [x.value for x in s1.row(i)]
      if rr[0] != 'label':
        assert rr[2] not in self.ss, 'Duplicate dimension set in spatial shape sheet: %s' % rr[2]
        a,b,k,lf = tuple( [x.strip() for x in rr[:4] ] )
        li = str( int( rr[4] ) )
        u = str( rr[6] )
        assert u not in uu, 'Duplicate uid in spatialShape_ref.xls: %s' % u
        self.labByUid[u] = rr[0]
        self.uidByLab[rr[0]] = u
        self.ss[ k ] = (a,b,lf,li,u)

class struct(object):
  def __init__(self):
    self.force = ('str-a02',)
    self.f = '../workbook/inputs/structRef.xls'
    self.ih = ['Label', 'Title', 'Spatial shape', 'Other dims', 'coords', 'cell methods', 'cell measures', 'flag meanings', 'flag values', 'Spatial shape (title)', 'Spatial dimensions']

    old = False
    if old:
      ii = open( '../workbook/inputs/strUidMap.txt' )
      self.uByLab = {}
      for l in ii.readlines():
        l,u = l.strip().split()
        self.uByLab[ l ] = u
      ii.close()
    else:
      self.refStr()
    self.ldsp = loadSpatial()
    self.summary = {}
    ii = open( 'strRef2.txt' )
    self.labs = set()
    for l in ii.readlines():
      bits = l.strip().split( '\t' )
      description = ''
      if len(bits) == 8:
        u, lab, spid, tmid, title, description, flag_values, tlab = bits
      else:
        print 'SEVERE: can not parse ',l
        raise
      description = description[1:-1]
      flag_values = flag_values[1:-1]
      self.summary[u] = (lab, tmid, spid, title, description, flag_values, tlab)
      self.labs.add(lab)
    ii.close()

## (i.tmid,i.spid,i.odims,i.coords,i.cell_methods,i.cell_measures,i.flag_meanings)
## Label	Title	Spatial shape	Other dims	coords	cell methods	cell measures	flag meanings	flag values	Spatial shape (title)	Spatial dimensions


  def refStr(self):
    wb = workbook( '../workbook/inputs/refStructure.xlsx' )
    sht = wb.book.sheet_by_name( 's1' )
    self.uByLab = {}
    for i in range(1,sht.nrows):
      r = [x.value for x in sht.row(i)]
      if len(r[0]) == 0 or r[0][0] != '#':
        uid = str( r[-1] )
        lab = r[2]
        self.uByLab[lab] = uid
    ii = open( '../workbook/inputs/strUidMap.txt' )
    for l in ii.readlines():
      l,u = l.strip().split()
      if l not in self.uByLab:
        self.uByLab[ l ] = u
      ii.close()



  def read(self):
    self.map = dict()
    wb=workbook( self.f )
    print wb.sns
    for sn in sorted( wb.sns ):
      if sn != '__mods__':
       sht = wb.book.sheet_by_name( sn )
       nf = 0
       for i in range( sht.nrows ):
         r = [x.value for x in sht.row(i)]
         print '>>> ',r
         if r[1] != 'Label' and not( len(r[0]) > 0 and r[0][0] == '#' ):
           if r[1] not in self.uByLab:
             nf += 1
             print 'LABEL NOT FOUND: ',r

       assert nf == 0,'nf= %s [%s]' % (nf,sn)
       for i in range( sht.nrows ):
         r = [x.value for x in sht.row(i)]
         if r[1] != 'Label' and not( len(r[0]) > 0 and r[0][0] == '#' ):
           print '>>> ',r
           u = self.uByLab[ r[1] ]
           if u in self.summary:
             lab, tmid, spid, title, description, fv, tlab = self.summary[u]
             if title != r[2]:
               print 'SEVERE: title mismatch :',title,r[2],r[1],lab,u
             t = (tmid,spid,r[4],r[5],r[6],r[7],r[8])
             if t in self.map:
                print 'Duplicate lookup ... %s' % str(t)
                print '--------',(u,lab,title,description,fv)
                print '--------',self.map[t]
             self.map[t] = (u,lab,title,description,fv,tlab,True)
           else:
             if len(r[0]) > 8 and r[0][0] == '*':
                if r[3] in self.ldsp.uidByLab:
                  spid = self.ldsp.uidByLab[r[3]]
                  t = (sn,spid,r[4],r[5],r[6],r[7],r[8])
                  if t in self.map:
                     print 'Duplicate lookup [2] ... %s' % str(t)
                     print '--------',(u,lab,title,description,fv)
                     print '--------',self.map[t]
                     self.map[t] = (u,r[1],'','','',sn,False)
                  print 'INFO.str.0042: generating incomplete record %s: %s' % (r[1],r[2])
                else:
                  print 'INFO.str.0041: ignoring %s: %s' % (r[1],r[2])
             elif r[1] not in self.labs and r[1] not in self.force:
               print 'INFO.str.0040: ignoring %s: %s' % (r[1],r[2])
             else:
               print 'ERROR: no uid found for ..:',sn,r

  def save(self):
    sh = shelve.open( 'inSh/structRef', 'n' )
    sh['__info__'] = 'Created by ifiles.struct .. information about structures keyed on tmid, spid, odims, coords, cell_methods, cell_measures, flag_meanings'
    sh['__cols__'] = ['tmid', 'spid', 'odims', 'coords', 'cell_methods', 'cell_measures', 'flag_meanings', 'uid','label','title','description','flag_values', 'time_label']
    s1 = set()
    ubl = dict()
    for k in self.map:
      u = self.map[k][0]
      sh[u] = list(k) + list(self.map[k])
      u,l = sh[u][7:9]
      ubl[l] = u
      rr = sh[u][:7]
      rr[0] = self.map[k][-1]
      tt = tuple(rr)
      if tt in s1:
        print 'SEVERE: duplicate of prospective lookup .... ',tt
      s1.add(tt)
      if tt[5] == '@OPT':
         print tt, u, sh[u]
    sh.close()
    sh = shelve.open( 'inSh/structRef02', 'n' )
    sh['__info__'] = 'Created by ifiles.struct .. label to uid'
    sh['__cols__'] = ['uid','label']
    for l in self.uByLab:
      if l not in ubl:
        sh[str(l)] = str( self.uByLab[l] )
        print 'INFO.structure.50080: adding uid-by-lab record: %s  %s' % (self.uByLab[l],l)
    sh.close()

           
class probDup(object):
  def __init__(self):
    self.fl = sorted( glob.glob( 'ingest_files/cmv_probDuplication*.xls' ) )
    self.cc = collections.defaultdict( list )
    self.cch = dict()
    self.records = dict()

  def read(self):
    for f in self.fl:
      wb=workbook( f )
      sht = wb.book.sheet_by_name( 'Sheet1' )
      key = None
      for i in range( sht.nrows ):
        rr = [x.value for x in sht.row(i)]
        if rr[0] == 'NEXT':
          key = tuple( rr[1:] )
        else:
          u = rr[1]
          assert u not in self.records, 'Duplicate mappings for %s' % u
          self.records[u] = rr
          self.cc[key].append( rr[:2] )

  def review(self):
    print 'INFO:number of records: %s' % len(self.cc.keys())
    nsk = 0
    nch = 0
    for k in self.cc:
      c1 = collections.defaultdict( set )
      kb = set()
      for r in self.cc[k]:
        c1[r[0]].add( r[1] )
        if r[0] not in ['+','-']:
          kb.add( r[0] )
      if len(kb) > 0:
        print 'WARN: unrecognised directives: %s: %s' % (str(k),str(kb))
      elif len( c1['-'] )  == 0:
        nsk += 1
      elif len( c1['+'] ) == 1:
        nch += 1
        assert k not in self.cch, 'Duplicate change record key: %s' % k
        self.cch[k] = (list( c1['+'] )[0],sorted( list( c1['-'] ) ) )
      else:
        print 'WARN: cannot scan directives: %s: n+: %s; n-:%s' % (str(k),len( c1['+'] ),len( c1['-'] ))
    print 'INFO: records with no actions: %s' % nsk
    print 'INFO: records with change instruction: %s' % nch

if __name__ == '__main__':
  #pd = probDup()
  #pd.read()
  #pd.review()

  s = struct()
  s.read()
  s.save()