属性错误列表对象没有属性条

Question

I am writing some code to tag a file, which looks at the previous line to see if there is a SCI tag, and if so, tag the current line with SCI_NXT in a fifth column (in a tab delimited file).

However, I get the attribute error that I am trying to strip a list (at line previous_line = split_line(previous_line) when the first line which is not a one item line is the object of the variable. This I understand is because it is writing the lines as lists, not as strings, but I do not understand how I might go about rectifying this. I have tried using "extend" but this resulted in the first line being written as each character being a different element, which is also not what I am looking to do.

Here is the test text I am working on:

</s>
<s>
Diptera NP  Diptera-n        SCI
was VBD be-v
the DT  the-x
most    RBS most-a
common  JJ  common-j
prey    NN  prey-n
among   IN  among-i
the DT  the-x
insects NNS insect-n
potentially RB  potentially-a
available   JJ  available-j
to  IN  to-i

Here is the code:

"""Tags a file with NEXT_SCI in extra feature column. Reads and writes vert files.
"""
import json

#from pip._vendor.pyparsing import line


VFILE = 'test_next.vert'

def split_line(line):
    """Split a line into five parts, word, tag, lempos, ti, sci"""
    # TODO: Speak to Diana about the spaces in the vert file - do they mean
    # anything?
    line = line.strip().split()
    if len(line) == 1:
        word = line[0]
        pos, lempos, tag = None, None, None
    elif len(line) == 3:
        word, pos, lempos = line
        tag = None
    elif len(line) == 4:
        word, pos, lempos, tag = line
    return [word, pos, lempos, tag]

def tag_next_sci(lines):
    """Loops through lines of original document to add to new file (tagged)
    """
    taggedlines = []
    for line in lines:
        taggedlines.append(tagline_next_sci(line, taggedlines))
    return taggedlines


def tagline_next_sci(line, taggedlines):
    """Assigns an indicator tag to a line
    """
    #<> are structural and do not need to be considered for feature tags so can be committed directly
    if line.startswith('<'):
        return line
    #look back at previous line to see if SCI, if so tag current line
    previous_line  = taggedlines[-1]
    previous_line = split_line(previous_line)
    line = split_line(line)
    #look at last column. if SCI, print line, go to next line and add tag in final column ("\t\t\tNXT_SCI\n")
    if previous_line[-1] == "SCI":
        if len(line) == 3:
            print(line + "\t\t\tSCI_MOD\n")
            return(line + "\t\t\tSCI_MOD\n")
        if len(line) == 4:
            print(line + "\t\tSCI_MOD\n")
            return(line + "\t\tSCI_MOD\n")
    return line

def read_vfile(fname):
    """Reads a vert file
    """
    with open(fname, 'r') as vfile:
        lines = vfile.readlines()
        return lines

def write_vfile(fname, taggedlines):
    """Writes a vert file
    """
    # write to file
    with open(fname, 'w') as outfile:
        outfile.writelines(taggedlines)

def tag_vert_sci_next(fname, fname_out):
    """Creates a new file with tags
    """
    # read vertical file
    lines = read_vfile(fname)
    # tag file    
    taggedlines = tag_next_sci(lines)
    # call write file
    write_vfile(fname_out, taggedlines)

def main(fname, fname_out):
    #call sci_next tagging
    tag_vert_sci_next('test_next.vert', fname_out)

if __name__ == "__main__":
    main('test_next.vert', 'zenodo_tagged_SCI_MOD.vert')

And the trackback error:

Traceback (most recent call last):
  File "/home/sandra/git/trophic/tagging/tagging_NEXT.py", line 123, in <module>
    main('test_next.vert', 'zenodo_tagged_SCI_MOD.vert')
  File "/home/sandra/git/trophic/tagging/tagging_NEXT.py", line 120, in main
    tag_vert_sci_next('test_next.vert', fname_out)
  File "/home/sandra/git/trophic/tagging/tagging_NEXT.py", line 78, in tag_vert_sci_next
    taggedlines = tag_next_sci(lines)
  File "/home/sandra/git/trophic/tagging/tagging_NEXT.py", line 31, in tag_next_sci
    taggedlines.append(tagline_next_sci(line, taggedlines))
  File "/home/sandra/git/trophic/tagging/tagging_NEXT.py", line 43, in tagline_next_sci
    previous_line = split_line(previous_line)
  File "/home/sandra/git/trophic/tagging/tagging_NEXT.py", line 14, in split_line
    line = line.strip().split()
AttributeError: 'list' object has no attribute 'strip'

Answer 1

Your issue seems to be that tagline_next_sci sometimes returns a list and not a string. For example, I tried putting a print inside the function to see what was going on;

...
def tagline_next_sci(line, taggedlines):
    print('taggedlines', taggedlines)
    """Assigns an indicator tag to a line
    """
...

and got the output

taggedlines []
taggedlines ['</s>\n']
taggedlines ['</s>\n', '<s>\n']
taggedlines ['</s>\n', '<s>\n', ['Diptera', 'NP', 'Diptera-n', 'SCI']]

So you should check at the bottom of the function to make sure you always return a string, and maybe do a "\\t".join(line) if you need to puzzle together your list to a string, with something like

return line if isinstance(line, str) else "\t".join(line)

Answer 2

Thank you all for your help. Here is the code I ended up with:

    """Tags a file with SCI_MOD in extra feature column. Reads and writes vert files.
"""
import json


VFILE = 'zenodotaggedWS_ALL.vert'

def split_line(line):
    """Split a line into its parts"""
    line = line.strip().split()
    if len(line) == 1:
        word = line[0]
        pos, lempos, tag ="", "", ""
    elif len(line) == 3:
        word, pos, lempos = line
        tag = ""
    elif len(line) == 4:
        word, pos, lempos, tag = line
    return [word, pos, lempos, tag]

def tag_next_sci(lines):
    """Loops through lines of original document to add to new file (tagged)
    """
    taggedlines = []
    for line in lines:
        taggedlines.append(tagline_next_sci(line, taggedlines))
    return taggedlines


def tagline_next_sci(line, taggedlines):
    """Assigns an indicator tag to a line
    """
    #<> are structural and do not need to be considered for feature tags so can be committed directly
    if line.startswith('<'):
        return line
    #look back at previous line to see if SCI, if so tag current line
    previous_line  = taggedlines[-1]
    previous_line = split_line(previous_line)
    line = split_line(line)
    if previous_line[2] == "SCI-n":
            print("\t".join(line) + "\tSCI_MOD\n")
            return "\t".join(line) + "\tSCI_MOD\n"
    return line + "\n" if isinstance(line, str) else "\t".join(line) + "\n"

def read_vfile(fname):
    """Reads a vert file
    """
    with open(fname, 'r') as vfile:
        lines = vfile.readlines()
        return lines

def write_vfile(fname, taggedlines):
    """Writes a vert file
    """
    # write to file
    with open(fname, 'w') as outfile:
        outfile.writelines(taggedlines)

def tag_vert_sci_next(fname, fname_out):
    """Creates a new file with tags
    """
    # vertical file location
    # make list of species names
    # read vertical file
    lines = read_vfile(fname)
    # tag file    
    taggedlines = tag_next_sci(lines)
    # call write file
    write_vfile(fname_out, taggedlines)

def main(fname, fname_out):
    #call sci_next tagging
    tag_vert_sci_next('zenodotaggedWS_ALL.vert', fname_out)

if __name__ == "__main__":
    main('zenodotaggedWS_ALL.vert', 'zenodo_tagged_SCIMOD2.vert')