使用python修复列缩进

Question

There is a file format called .xyz that helps visualizing molecular bonds. Basically the format asks for a specific pattern:

At the first line there must be the number of atoms, which in my case is 30. After that there should be the data where the first line is the name of the atom, in my case they are all carbon. The second line is the x information and the third line is the y information and the last line is the z information which are all 0 in my case. Indentation should be correct so that all of the corresponding lines should start at the same place. So something like this:

30
C x1 y1 z1 
C x2 y2 z2
...
...
...

and not:

30 
C x1 y1 z1
C   x2 y2  z2

since this is the wrong indentation.

My generated data is stored like this in a .txt file:

C       2.99996     7.31001e-05     0
C       2.93478     0.623697        0
C       2.74092     1.22011     0
C       2.42702     1.76343     0
C       2.0079      2.22961     0
C       1.50006     2.59812     0
C       0.927076        2.8532      0
C       0.313848        2.98349     0
C       -0.313623       2.9837      0
C       -0.927229       2.85319     0
C       -1.5003     2.5981      0
C       -2.00732        2.22951     0
C       -2.42686        1.76331     0
C       -2.74119        1.22029     0
C       -2.93437        0.623802        0
C       -2.99992        -5.5509e-05     0
C       -2.93416        -0.623574       0
C       -2.7409     -1.22022        0
C       -2.42726        -1.7634     0
C       -2.00723        -2.22941        0
C       -1.49985        -2.59809        0
C       -0.92683        -2.85314        0
C       -0.313899       -2.98358        0
C       0.31363     -2.98356        0
C       0.927096        -2.85308        0
C       1.50005     -2.59792        0
C       2.00734     -2.22953        0
C       2.4273      -1.76339        0
C       2.74031     -1.22035        0
C       2.93441     -0.623647       0

I want to correct the indentation of this by making all of the lines start from the same point. I tried to do this with AWK to no avail. So I turned to Python. So far I have this:

#!/usr/bin/env/python
text_file = open("output.txt","r")
lines = text_file.readlines()
myfile = open("output.xyz","w")
for line in lines:
    atom, x, y, z = line.split()
    x, y, z = map(float(x,y,z))
    myfile.write("{}\t {}\t {}\t {}\t".format(atom,x,y,z))
myfile.close()
text_file.close()

but I don't know currently as to how indentation can be added into this.

tl;dr: I have a data file in . txt , I want to change it into .xyz that's been specified but I am running into problems with indentation.

Answer 1

It appears that I misinterpreted your requirement...

To achieve a fixed width output using awk, you could use printf with a format string like this:

$ awk '{printf "%-4s%12.6f%12.6f%5d\n", $1, $2, $3, $4}' data.txt 
C       2.999960    0.000073    0
C       2.934780    0.623697    0
C       2.740920    1.220110    0
C       2.427020    1.763430    0
C       2.007900    2.229610    0
C       1.500060    2.598120    0
C       0.927076    2.853200    0
C       0.313848    2.983490    0
C      -0.313623    2.983700    0
# etc.

Numbers after the % specify the width of the field. A negative number means that the output should be left aligned (as in the first column). I have specified 6 decimal places for the floating point numbers.

---

Original answer, in case it is useful:

To ensure that there is a tab character between each of the columns of your input, you could use this awk script:

awk '{$1=$1}1' OFS="\t" data.txt > output.xyz

$1=$1 just forces awk to touch each line, which makes sure that the new Output Field Separator ( OFS ) is applied.

awk scripts are built up from a series of condition { action } . If no condition is supplied, the action is performed for every line. If a condition but no action is supplied, the default action is to print the line. 1 is a condition that always evaluates to true, so awk prints the line.

Note that even though the columns are all tab-separated, they are still not lined up because the content of each column is of a variable length.

Answer 2

Your data has already been ill formatted and converted to string. To correctly allign the numeric and non-numeric data, you need to parse the individual fields to respective data types (possibly using duck-typing) before formating using str.format

for line in st.splitlines():
    def convert(st):
        try:
            return int(st)
        except ValueError:
            pass
        try:
            return float(st)
        except ValueError:
            pass
        return st
    print "{:8}{:12.5f}{:12.5f}{:5d}".format(*map(convert,line.split()))


C            2.99996     0.00007    0
C            2.93478     0.62370    0
C            2.74092     1.22011    0
C            2.42702     1.76343    0
C            2.00790     2.22961    0
C            1.50006     2.59812    0
C            0.92708     2.85320    0
C            0.31385     2.98349    0
C           -0.31362     2.98370    0
C           -0.92723     2.85319    0

Answer 3

Using this: awk '{printf "%s\\t%10f\\t%10f\\t%i\\n",$1,$2,$3,$4}' atoms

give this output:

C         2.999960        0.000073      0
C         2.934780        0.623697      0
C         2.740920        1.220110      0
C         2.427020        1.763430      0
C         2.007900        2.229610      0
C         1.500060        2.598120      0
C         0.927076        2.853200      0
C         0.313848        2.983490      0
C        -0.313623        2.983700      0
C        -0.927229        2.853190      0
C        -1.500300        2.598100      0
C        -2.007320        2.229510      0
C        -2.426860        1.763310      0
C        -2.741190        1.220290      0
C        -2.934370        0.623802      0
C        -2.999920       -0.000056      0
C        -2.934160       -0.623574      0
C        -2.740900       -1.220220      0
C        -2.427260       -1.763400      0
C        -2.007230       -2.229410      0
C        -1.499850       -2.598090      0
C        -0.926830       -2.853140      0
C        -0.313899       -2.983580      0
C         0.313630       -2.983560      0
C         0.927096       -2.853080      0
C         1.500050       -2.597920      0
C         2.007340       -2.229530      0
C         2.427300       -1.763390      0
C         2.740310       -1.220350      0
C         2.934410       -0.623647      0

Is it what you're meaning or did I misunderstood ?

Edit for side note: I used tabs \\t for separation, a space could do too and I limited the output to a precision of 10, I didn't verify your input lenght

Answer 4

You can use string formatting to print values with consistent padding. For your case, you might write lines like this to the file:

>>> '%-12s %-12s %-12s %-12s\n' % ('C', '2.99996', '7.31001e-05', '0')
'C            2.99996      7.31001e-05  0           '

"%-12s" means "take the str() of the value and make it take up at least 12 characters left-justified.