Project

General

Profile

« Previous | Next » 

Revision 5439

Added by Aaron Marcuse-Kubitza about 12 years ago

csvs.py: sniff(): Support multi-char delims using \t, such as \t|\t used by NCBI. Support custom line suffixes, such as \t| used by NCBI.

View differences:

lib/csvs.py
8 8 
import util
9 9 

  		




  10
  10
  
    
delims = ',;\t|`'


  		




  
  11
  
    
tab_padded_delims = ['\t|\t']


  		




  11
  12
  
    
tsv_delim = '\t'


  		




  12
  13
  
    
escape = '\\'


  		




  13
  14
  
    

  		




  ......




  20
  21
  
    
    line, ending = strings.extract_line_ending(line)


  		




  21
  22
  
    
    dialect = csv.Sniffer().sniff(line, delims)


  		




  22
  23
  
    
    


  		




  23
  
  
    
    # TSVs usually don't quote fields (nor doublequote embedded quotes)


  		




  24
  
  
    
    if is_tsv(dialect): dialect.quoting = csv.QUOTE_NONE


  		




  
  24
  
    
    if is_tsv(dialect):


  		




  
  25
  
    
        # TSVs usually don't quote fields (nor doublequote embedded quotes)


  		




  
  26
  
    
        dialect.quoting = csv.QUOTE_NONE


  		




  
  27
  
    
        


  		




  
  28
  
    
        # Check multi-char delims using \t


  		




  
  29
  
    
        delim = strings.find_any(line, tab_padded_delims)


  		




  
  30
  
    
        if delim:


  		




  
  31
  
    
            dialect.delimiter = delim


  		




  
  32
  
    
            line_suffix = delim.rstrip('\t')


  		




  
  33
  
    
            if line.endswith(line_suffix): ending = line_suffix+ending


  		




  25
  34
  
    
    else: dialect.doublequote = True # Sniffer doesn't turn this on by default


  		




  26
  35
  
    
    dialect.lineterminator = ending


  		




  27
  36
  
    
    


  		












Also available in:  Unified diff