迭代字符串元组和列表列表并将值导出到 Python 中的 csv

Question

I have a function called 'tableextract' which returns a list of lists.

Example of 'tableextract' function output for 1 file called ABC.txt:

[[' ', 'CITRIC ACID, ANHYDROUS', '5 mg', ''], [' ', 'NITROGEN', '5 mg', '21%'], [' ', 'PURIFIED WATER', '5 mg', '']]

Now, I have a directory of files, on which I want to run this function and export all the values in a single csv. And for each file in the directory, based on the number of list of lists generated by the 'tableextract' function for that file, I want a new row in the csv with the filename and each list item inside the list of lists as a separate column.

Expected output(0,1,2,3 column values are generated by pandas, when splitting on comma.. sep=',')

Filename       0               1                         2             3

ABC.txt                 CITRIC ACID, ANHYDROUS         5mg                   

ABC.txt                   NITROGEN                     5 mg          21%

ABC.txt                   PURIFIED WATER               5 mg 

Right now, the code I have written the code:

data = []
    dir ='C:\\Users\\'
    allfiles = os.listdir(dir)
    files =[fnmatch for fnmatch in allfiles if fnmatch.endswith(".txt")]
    for x in files:
        output = tableextract(dir+x)
        newout = x,output
        data.append((newout))
        df = pd.DataFrame(data)

        df.to_csv('./Desktop/newgoofs7.csv', index=False, sep=',')

is giving output in this format in the csv:

    0903882.txt     [[' ', 'HYPROMELLOSE', '1.765 mg', '1.765 %'], [' ', 'PE', '10.000 mg', '10.000 %'], [' ', 'RAMIPRIL', '10 mg', ''], [' ', 'RAMIPRIL', '10.000 mg', '10.000 %'], [' ', 'SODIUM STEARYL FUMARATE', '0.250 mg', '0.250 %']]

    0903777.txt     [[' ', 'HYPROMELLOSE', '0.441 mg', '0.441 %'], [' ', 'PE', '2.500 mg', '2.500 %'], [' ', 'RAMIPRIL', '2.5 mg', ''], [' ', 'RAMIPRIL', '2.500 mg', '2.500 %'], [' ', 'YELLOW FERRIC OXIDE', '0.100 mg', '0.100 %']]

where one column has the filename and the entire output of 'tablextract' function is in one column.

I want the output of my code to provide the output mentioned in the 'Expected output :' format.

Any help on this?

Answer 1

You must put "df = pd.DataFrame(data)" and "df.to_csv(...)" after the "for" loop. Here is my solution:

# testdata
data=[
    ("ABC.txt", [[' ', 'CITRIC ACID, ANHYDROUS', '5 mg', ''], [' ', 'NITROGEN', '5 mg', '21%'], [' ', 'PURIFIED WATER', '5 mg', '']]),
    ("DEF.txt", [[' ', 'citric acid, anhydrous', '4 mg', ''], [' ', 'nitrogen', '8 mg', '1%'], [' ', 'purified water', '9 mg', '']])
  ]

#-----------------------
df= pd.DataFrame(data)
df2= df.explode(1)[1].str.join("|").str.split("|",expand=True)
cols= df2.columns.tolist()
df2["Filename"]= df[0]
df2= df2.reindex(["Filename"]+cols, axis=1)

df2:

  Filename  0                       1     2    3
0  ABC.txt     CITRIC ACID, ANHYDROUS  5 mg     
0  ABC.txt                   NITROGEN  5 mg  21%
0  ABC.txt             PURIFIED WATER  5 mg     
1  DEF.txt     citric acid, anhydrous  4 mg     
1  DEF.txt                   nitrogen  8 mg   1%
1  DEF.txt             purified water  9 mg  

EDIT: Another solution:

for x in files:
    data.append( tableextract(dir+x) )

dd={"Filename": [ fn for fn,dat in data for i in range(len(dat)) ] }
z= zip(*[d for fn,dat in data for d in dat])
dd.update({i:a for i,a in enumerate(z)})
df2=pd.DataFrame(dd)