将所有具有相同名称的文件放在文件夹中

Question

I'm new to Python and I need a program that copies files from the same day into a new folder.

Example files:

20120807_first_day_pic.jpg
20120807_first_day_sheet.jpg
20120807_first_day_sheet2.jpg
20120907_second_day_pic.jpg
20120907_second_day_sheet.jpg
20120907_second_day_sheet2.jpg

This is what I have so far, but every file gets a folder and not the whole day.

import os, re, shutil

tfolder = 'D:/Testing/src/'

os.chdir(tfolder)

re_year19xxxxxx = re.compile('(19[0-9][0-9][0-9][0-9])')
re_year20xxxxxx = re.compile('(20[0-9][0-9][0-9][0-9])')

re_ed = re.compile('(ED[0-9])')
destPath = 'D:/Testing/Dest/'

def analyse_file_name(fname):
    filePath, coords = os.path.split(fname) #the new folders will be named according to the first 4 characters of the original file name
    coordsFolder = coords[:53]
    coordsFname = coords[:53]
    coordsExt = os.path.splitext(fname)
    year = 'year' #create variable year
    ed = 'ed' #create variable ed to store the edition number if necessary
    bname = fname #the original file name
    for re_year in (re_year19xxxxxx, re_year20xxxxxx):
        rx = re_year.search(fname) #search for regex in the file name and store it in rx
        if rx:
            year = rx.group(1) #if the regex is found, store the year
            bname.replace(year, ' ')
            res = re_ed.search(fname)
            if res:
                ed = res.group(1)
                bname.replace(ed, ' ')
        os.chdir(destPath)  
        if year is 'year':
            fname2 = os.path.join(destPath, coordsFolder) + '\\' + coordsFname + coordsExt[1]
        else:
            fname2 = os.path.join(destPath, coordsFolder,year,ed) + '\\' + coordsFname + coordsExt[1]
        print('%s -> %s' % (fname, fname2)) #debug print
        dirn, _ = os.path.split(fname2)
        if not os.path.exists(dirn):
            os.makedirs(dirn)
        shutil.copy(fname, fname2)

for root, dirs, files in os.walk(tfolder):
    for name in files:
        fn = os.path.join(root, name)
        analyse_file_name(fn)

Answer 1

If you just want to copy files that start with a known date string format, how about something like this?

def copyfile(filepath, target_dir):
    p, filename = os.path.split(filepath)

    # get date component of name
    date_component = filename.split("_", 1)[0]

    # try to parse out the date
    try:
        d = datetime.datetime.strptime(date_component, "%Y%m%d")
    except ValueError:  
        print "Could not place: ", filename
        return
    target_date_dir = os.path.join(target_dir, str(d.year), str(d.month), str(d.day))
    os.makedirs(target_date_dir)
    shutil.copy(filepath, target_date_dir)

Answer 2

First, create a dict (a defaultdict was even more convenient here) that will gather the files for a date (it's good to use re , but given the names of your files using split was easier):

>>> import os
>>> import re
>>> pat = r'(\d+)(?:_\d+)?_(\w+?)[\._].*'
>>> from collections import defaultdict
>>> dict_date = defaultdict(lambda : defaultdict(list))
>>> for fil in os.listdir(path):
    if os.path.isfile(os.path.join(path, fil)):
        date, animal = re.match(pat, fil).groups()
        dict_date[date][animal].append(fil)


>>> dict_date['20120807']
defaultdict(<type 'list'>, {'first': ['20120807_first_day_pic.jpg', '20120807_first_day_sheet.jpg', '20120807_first_day_sheet2.jpg']})

Then for each date, create a subfolder and copy the corresponding files there:

>>> from shutil import copyfile
>>> for date in dict_date:
        for animal in dict_date[date]:
        try:
            os.makedirs(os.path.join(path, date, animal))
        except os.error:
            pass
        for fil in dict_date[date][animal]:
            copyfile(os.path.join(path, fil), os.path.join(path, date, animal, fil))

EDIT : took into account OP's new requirements, and Khalid's remark.

Answer 3

import os, shutil

src_path = "D:\\Testing\\Src\\"
dest_path = "D:\\Testing\\Dest\\"

for file in os.listdir(src_path):
    if not os.path.isdir(dest_path + file.split("-")[0]):
       os.mkdir(dest_path + file.split("-")[0])
shutil.copy(src_path + file, dest_path + file.split("-")[0])

Answer 4

Regex day :) What about trying to match the filename with

pattern=r'(?P<filedate>(?P<year>\d{4})(?P<month>\d{2})(?P<day>\d{2}))\_(?P<bloodyrestofname>.*)'

Complete date, year, etc. may be retrieved from the respective named groups in the match.