Python中文本文件的字典

Question

Problem:

I have a txt file with this format:

Intestinal infectious diseases (001-003)  
001 Cholera  
002 Fever  
003 Salmonella   
Zoonotic bacterial diseases (020-022)  
020 Plague  
021 Tularemia  
022 Anthrax  
External Cause Status (E000)  
E000 External cause status  
Activity (E001-E002)  
E001 Activities involving x and y  
E002 Other activities

where each line that begins with the 3-integer code/E+3-integer code/V+3-integer code is a value for the preceding header, which are the keys for my dictionary. In other questions I've seen, the use of columns or colons can be used to parse each line to make a key/value pair, but the format of my txt file doesn't allow me to do that.

Is the a way to make a txt file like this into a dictionary where the keys are the group names and the values are the code+disease names?

I also need to parse the code and disease names into a second dictionary, so I end up with a dictionary that contains the group names as keys, with the values being a second dictionary with the codes as keys and the disease names as values.

Script:

def process_file(filename):
    myDict={}
        f = open(filename, 'r')
        for line in f:
            if line[0] is not int:
                if line.startswith("E"):
                    if line[1] is int:
                        line = dictionary1_values
                    else:
                        break
                else:
                    line = dictionary1_key
            myDict[dictionary1_key].append[line]

Desired output format is:
{"Intestinal infectious diseases (001-003)": {"001": "Cholera", "002": "Fever", "003": "Salmonella"}, "Zoonotic bacterial diseases (020-022)": {"020": "Plague", "021": "Tularemia", "022": "Anthrax"}, "External Cause Status (E000)": {"E000": "External cause status"}, "Activity (E001-E002)": {"E001": "Activities involving x and y", "E002": "Other activities"}}

Answer 1

try using regular expressions to determine if it is a header or a disease

import re
mydict = {}
with open(filename, "r") as f:
    header = None
    for line in f:
        match_desease = re.match(r"(E?\d\d\d) (.*)", line)
        if not match_desease:
            header = line
        else:
            code = match_desease.group(1)
            desease = match_desease.group(2)
            mydict[header][code] = desease

Answer 2

One solution would be to use regular expressions to help you characterize and parse the two types of lines you might encounter in this file:

import re
header_re = re.compile(r'([\w\s]+) \(([\w\s\-]+)\)')
entry_re = re.compile(r'([EV]?\d{3}) (.+)')

This allows you to very easily check which type of line you're encountering, and break it apart as desired:

# Check if a line is a header:
header = header_re.match(line)
if header:
    header_name, header_codes = header.groups()  # e.g. ('Intestinal infectious diseases', '001-009')
    # Do whatever you need to do when you encounter a new group
    # ...
else:
    entry = entry_re.match(line)
    # If the line wasn't a header, it ought to be an entry,
    # otherwise we've encountered something we didn't expect
    assert entry is not None
    entry_number, entry_name = entry.groups()  # e.g. ('001', 'Cholera')
    # Do whatever you need to do when you encounter an entry in a group
    # ...

Using that to re-work your function, we could write the following:

import re

def process_file(filename):
    header_re = re.compile(r'([\w\s]+) \(([\w\s\-]+)\)')
    entry_re = re.compile(r'([EV]?\d{3}) (.+)')

    all_groups = {}
    current_group = None

    with open(filename, 'r') as f:
        for line in f:
            # Check if a line is a header:
            header = header_re.match(line)
            if header:
                current_group = {}
                all_groups[header.group(0)] = current_group
            else:
                entry = entry_re.match(line)
                # If the line wasn't a header, it ought to be an entry,
                # otherwise we've encountered something we didn't expect
                assert entry is not None
                entry_number, entry_name = entry.groups()  # e.g. ('001', 'Cholera')

                current_group[entry_number] = entry_name

    return all_groups

Answer 3

def process_file(filename):
    myDict = {}
    rootkey = None
    f = open(filename, 'r')
    for line in f:
        if line[1:3].isdigit():           # if the second and third character from the checked string (line) is the ASCII Code in range 0x30..0x39 ("0".."9"), i.e.: str.isdigit()
            subkey, data = line.rstrip().split(" ",1)     # split into two parts... the first one is the number with or without "E" at begin
            myDict[rootkey][subkey] = data
        else:
            rootkey = line.rstrip()       # str.rstrip() is used to delete newlines (or another so called "empty spaces")
            myDict[rootkey] = {}          # prepare a new empty rootkey into your myDict
    f.close()
    return myDict

Testing in Python console:

>>> d = process_file('/tmp/file.txt')
>>>
>>> d['Intestinal infectious diseases (001-003)']
{'003': 'Salmonella', '002': 'Fever', '001': 'Cholera'}
>>> d['Intestinal infectious diseases (001-003)']['002']
'Fever'
>>> d['Activity (E001-E002)']
{'E001': 'Activities involving x and y', 'E002': 'Other activities'}
>>> d['Activity (E001-E002)']['E001']
'Activities involving x and y'
>>>
>>> d
{'Activity (E001-E002)': {'E001': 'Activities involving x and y', 'E002': 'Other activities'}, 'External Cause Status (E000)': {'E000': 'External cause status'}, 'Intestinal infectious diseases (001-003)': {'003': 'Salmonella', '002': 'Fever', '001': 'Cholera'}, 'Zoonotic bacterial diseases (020-022)': {'021': 'Tularemia', '020': 'Plague', '022': 'Anthrax'}}

Warning: First one line in the file must be just a "rootkey" ! Not "subkey" or data ! Otherwise cause may to be the raise error :-)

A note: Maybe you should remove the first "E" character. Or can not it be done? Do you need to leave this "E" character somewhere?