一次诡异的 python 错误排查

输入表格示例：

library	reads_num	reads_len	bases_num	GC	avgQ	Q20	Q30
A-Ctrl	46,875,170	150.0	7.00 Gb	49.90%	36.6	93.24%	85.97%
B-Ctrl	154,351,882	150.0	23.00 Gb	45.40%	36.9	93.80%	86.93%
C-Ctrl	58,126,912	150.0	8.00 Gb	60.05%	36.05	92.19%	83.96%
D-Ctrl	143,761,494	150.0	21.00 Gb	42.00%	37.25	94.46%	88.15%

初次尝试

# 错误的结果
table = open('qc.stat', 'r')
header = [[],] * len(table.readline().strip().split('\t'))
for i in table:
    for x,y in enumerate(i.strip().split('\t')):
        header[x].append(y)
header
# table.close()

结果输出

[['A',
  '46,875,170',
  '150.0',
  '7.00 Gb',
  '49.90%',
  '36.6',
  '93.24%',
  '85.97%',
  'B',
  '154,351,882',
  '150.0',
  '23.00 Gb',
  '45.40%',
  '36.9',
  '93.80%',
  '86.93%',
  'C',
  '58,126,912',
  '150.0',
  '8.00 Gb',
  '60.05%',
  '36.05',
  '92.19%',
  '83.96%',
  'D',
  '143,761,494',
  '150.0',
  '21.00 Gb',
  '42.00%',
  '37.25',
  '94.46%',
  '88.15%'],
 ['A',
  '46,875,170',
  '150.0',
  '7.00 Gb',
  '49.90%',
  '36.6',
  '93.24%',
  '85.97%',
  'B',
  '154,351,882',
  '150.0',
  '23.00 Gb',
  '45.40%',
  '36.9',
  '93.80%',
  '86.93%',
  'C',
  '58,126,912',
  '150.0',
  '8.00 Gb',
  '60.05%',
  '36.05',
  '92.19%',
  '83.96%',
  'D',
  '143,761,494',
  '150.0',
  '21.00 Gb',
  '42.00%',
  '37.25',
  '94.46%',
  '88.15%'],
 ...]

正确解析

# 正确代码
table = open('qc.stat', 'r')
header = [[i,] for i in table.readline().strip().split('\t')]
for i in table:
    for x,y in enumerate(i.strip().split('\t')):
        header[x].append(y)
header
# table.close()

结果输出

[['library', 'A', 'B', 'C', 'D'],
 ['reads_num', '46,875,170', '154,351,882', '58,126,912', '143,761,494'],
 ['reads_len', '150.0', '150.0', '150.0', '150.0'],
 ['bases_num', '7.00 Gb', '23.00 Gb', '8.00 Gb', '21.00 Gb'],
 ['GC', '49.90%', '45.40%', '60.05%', '42.00%'],
 ['avgQ', '36.6', '36.9', '36.05', '37.25'],
 ['Q20', '93.24%', '93.80%', '92.19%', '94.46%'],
 ['Q30', '85.97%', '86.93%', '83.96%', '88.15%']]

原因分析

# 错误原因查找

head = '#library\treads_num\treads_len\tbases_num\tGC\tavgQ\tQ20\tQ30'

header = [[], ] * len(head.strip().split('\t'))
print 'error method...'
for i in header:
    print id(i)

header = [[i,] for i in head.strip().split('\t')]
print '\nright method...'
for i in header:
    print id(i)

error method...
4435237272
4435237272
4435237272
4435237272
4435237272
4435237272
4435237272
4435237272

right method...
4435236984
4435354904
4435318688
4435237416
4435318040
4435319768
4435318472
4435319624

python独特的变量命名方式：变量名（>= 1个）指向储存数据物理地址，使用其中任何一个名称都可以对数据进行操作