解析数据库中的文件并将信息添加到字典中

Question

I have text file as follows:

HEADER INFO

Last1, First1       Movie1 (1991) random stuff
                        Movie2 (1992) random stuff
                        Movie3 (1995) random stuff
                        Movie4 (3455) random stuff

Last2, First2       Movie1 (1998) random stuff
                        Movie2 (4568) random stuff
                        Movie3 (2466) random stuff
                        Movie4 (4325) random stuff
                        Movie5 (4875) random stuff
                        Movie6 (3525) random stuff
                        Movie7 (4567) random stuff

FOOTER INFO

It also contains some header/footer info that I can skip. The spaces between the name and movie are not constant. I want to add this data into a dictionary using while loops (no for loops for the whole process). Basically the name will act as the key and the list of following movies will be the values (both are strings). So far I can achieve either obtaining the lines which contain the names OR the lines which contain the movies. I tried an using an if statement to get it to work but to no avail.

Basically I was thinking of using an if statement to say if the line contains the name by some characteristic of the line, then splice out the name and splice out the movie and add to the dictionary. And if the name is not in the line, then associate that movie with the same name(multiple entries). But I think this is where Im lost. This part and maybe how Im iterating with the while loop.

I didn't use any readline(). Instead I used readlines() and I used that to toggle through the lines to pick out the information. I'm just wondering if anyone has any tips/hints they could offer.

If anyone wants the actual data I'm using then please pm me.

Ill rephrase it:

CRC: 0xDE308B96  File: actors.list  Date: Fri Aug 12 00:00:00 2011

Copyright 1990-2007 The Internet Movie Database, Inc.  All rights reserved.

COPYING POLICY: Internet Movie Database (IMDb)
==============================================

CUTTING COPYRIGHT NOTICE

THE ACTORS LIST
===============

Name                    Titles
----                    ------
ActA, A                 m1 (2011)
                            m2 (2011)

ActB, B                 m1 (2011)
                            m2 (2011)
                            m3 (2001)

ActC, C                 m1 (2011)

ActD, D                 m3 (2003)
                            m6 (2006)

ActE, E                 m6 (2006)

ActF, F                 m4 (2004)

ActG, G                 m4 (2004)

ActH, H                 m5 (2005)

Bacon, Kevin            m2 (2011)
                        m5 (2005)

-----------------------------------------------------------------------------
SUBMITTING UPDATES
==================

CUTTING UPDATES

For further info visit http://www.imdb.com/licensing/contact

And basically I want the output to be a dictionary:

{'E Acte': ['m6 (2006)'],
'A Acta': ['m1 (2011)', 'm2 (2011)'],
'G Actg': ['m4 (2004)'],
'B Actb': ['m1 (2011)', 'm2 (2011)', 'm3 (2001)'],
'D Actd': ['m3 (2003)', 'm6 (2006)'],
'F Actf': ['m4 (2004)'],
'Kevin Bacon': ['m2 (2011)', 'm5 (2005)'],
'H Acth': ['m5 (2005)'],
'C Actc': ['m1 (2011)']}

I'm suggested to use while loops since it'll make the process easier, but not restricted solely to it.

Answer 1

Here is a solution with a for loop which is much more natural in Python. It assumes the input file is formatted with spaces, like the code posted in the question above. I have posted an alternative answer now for the case when the list is formatted with tabs instead of spaces.

Of course you could rewrite it as a while loop, but it would not make much sense. You can also simplify it a bit by using a defaultdict(list) for the output in newer Python versions.

output = {}

pos = -1 # char position of title column
current_name = None

for line in open('actors.list'):
    if pos < 0:
        if line.startswith('-'):
            pos = line.find(' ')
            if pos > 0:
                pos = line.find('-', pos)
    else:
        if line.startswith('-'):
            break
        name = line[:pos].strip()
        title = line[pos:].strip()
        if name:
            if ',' in name:
                name = name.split(',', 1)
                name[0] = name[0].rstrip()
                name[1] = name[1].lstrip()
                name.reverse()
                name = ' '.join(name)
            current_name = name
        if title:
            output.setdefault(
                current_name, []).append(title)

print output

Answer 2

Here is another solution for the case when the list is formatted with tab chars instead of spaces:

output = {}

in_list = False
current_name = None

for line in open('actors.list'):
    if in_list:
        if line.startswith('-'):
            break
        if '\t' not in line:
            continue
        name, title = line.split('\t', 1)
        name = name.strip()
        title = title.strip()
        if name:
            if ',' in name:
                name = name.split(',', 1)
                name[0] = name[0].rstrip()
                name[1] = name[1].lstrip()
                name.reverse()
                name = ' '.join(name)
            current_name = name
        if title:
            output.setdefault(
                current_name, []).append(title)
    else:
        if line.startswith('-'):
            in_list = True