Project

General

Profile

« Previous | Next » 

Revision 18

Added by Aaron Marcuse-Kubitza almost 13 years ago

Changed xml2db and data2xml to help standardize mapping to different XML formats

View differences:

scripts/xml2db/xml_util.py
46 46 
    return True
47 47 

  		




  48
  48
  
    
def value(node):


  		




  49
  
  
    
    if node.firstChild != None: return node.firstChild.nodeValue


  		




  
  49
  
    
    if node.firstChild != None: return node.firstChild.nodeValue.strip()


  		




  50
  50
  
    
    else: return None


  		












  

    
      scripts/data2xml/XpathParser.py
    
  





  63
  63
  
    
    


  		




  64
  64
  
    
    def _name(self): return self._match_re(r'[\w.]+', required=True)


  		




  65
  65
  
    
    


  		




  66
  
  
    
    def _value(self): return self._match_re(r'[\w|]+', required=True)


  		




  
  66
  
    
    def _value(self): return self._match_re(r'[\w.|]+', required=True)


  		












  

    
      scripts/data2xml/data2xml
    
  





  16
  16
  
    
        dest = sys.argv.pop(0)


  		




  17
  17
  
    
        mappings_path = sys.argv.pop(0)


  		




  18
  18
  
    
    except IndexError: raise Exception('Usage: '+prog_name


  		




  19
  
  
    
        +' dest_mappings_column mappings_path <dataset')


  		




  
  19
  
    
        +' dest_mappings_column mappings_path <dataset >output')


  		




  20
  20
  
    
    


  		




  21
  21
  
    
    # Get mappings


  		




  22
  22
  
    
    mappings = {}


  		




  
  23
  
    
    has_type_containers = False # whether outer elements are type containiners


  		




  23
  24
  
    
    stream = open(mappings_path, 'rb')


  		




  24
  25
  
    
    reader = csv.reader(stream, delimiter=',', quotechar='"')


  		




  25
  26
  
    
    fieldnames = reader.next()


  		




  ......




  28
  29
  
    
    for row in reader:


  		




  29
  30
  
    
        name = row[0]


  		




  30
  31
  
    
        path = row[dest_idx]


  		




  31
  
  
    
        if path != '':


  		




  
  32
  
    
        if name != '' and path != '':


  		




  32
  33
  
    
            if path.startswith('?'): path = path[1:]


  		




  33
  34
  
    
            path = path.replace('<name>', name)


  		




  34
  
  
    
            path = re.sub(r'(?<=/)\*(?=s/(\w+))', r'\1', path)


  		




  
  35
  
    
            path, repl_ct = re.subn(r'(?<=/)\*(?=s/(\w+))', r'\1', path)


  		




  
  36
  
    
            if repl_ct > 0: has_type_containers = True # *s used for type elem


  		




  35
  37
  
    
            mappings[name] = XpathParser(path).parse()


  		




  36
  38
  
    
    stream.close()


  		




  
  39
  
    
    if has_type_containers: id_level = 1


  		




  
  40
  
    
    else: id_level = 0  


  		




  37
  41
  
    
    


  		




  38
  42
  
    
    # Process dataset


  		




  39
  43
  
    
    doc = getDOMImplementation().createDocument(None, dest, None)


  		




  ......




  42
  46
  
    
    fieldnames = reader.next()


  		




  43
  47
  
    
    row_idx = 0


  		




  44
  48
  
    
    for row in reader:


  		




  45
  
  
    
        row_id = src+'_'+str(row_idx)


  		




  
  49
  
    
        row_id = str(row_idx)


  		




  46
  50
  
    
        for idx, name in enumerate(fieldnames):


  		




  47
  51
  
    
            value = row[idx]


  		




  48
  
  
    
            if value and name in mappings:


  		




  
  52
  
    
            if value != '' and name in mappings:


  		




  49
  53
  
    
                path = deepcopy(mappings[name]) # don't modify main value!


  		




  50
  
  
    
                path[1].attrs.append([XpathElem('id', row_id, is_attr=True)])


  		




  
  54
  
    
                path[id_level].attrs.append([XpathElem('id', row_id,


  		




  
  55
  
    
                    is_attr=True)])


  		




  51
  56
  
    
                path[-1].value = value


  		




  52
  57
  
    
                by_path(doc, path, True)


  		




  53
  58
  
    
        row_idx += 1


  		












Also available in:  Unified diff