From 3f325408dfd666e43e50ac11dd48c99787a5a5b1 Mon Sep 17 00:00:00 2001 From: Brian Coca Date: Sun, 29 Mar 2015 02:25:12 -0400 Subject: [PATCH] find module updated to be able to get checksums and stat data, also now it only stats a file once so it should be quite faster on large file sets --- lib/ansible/modules/files/find.py | 299 ++++++++++++++++++++++++++++++ 1 file changed, 299 insertions(+) create mode 100644 lib/ansible/modules/files/find.py diff --git a/lib/ansible/modules/files/find.py b/lib/ansible/modules/files/find.py new file mode 100644 index 00000000000..d5441aad273 --- /dev/null +++ b/lib/ansible/modules/files/find.py @@ -0,0 +1,299 @@ +#!/usr/bin/python +# -*- coding: utf-8 -*- + +# (c) 2014, Ruggero Marchei +# (c) 2015, Brian Coca +# +# This file is part of Ansible +# +# Ansible is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# Ansible is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with Ansible. If not, see + + +import os +import stat +import fnmatch +import time +import re +import shutil + + +DOCUMENTATION = ''' +--- +module: find +author: Brian Coca (based on Ruggero Marchei's Tidy) +version_added: "2.0" +short_description: return a list of files based on specific criteria +requirements: [] +description: + - Return a list files based on specific criteria. Multiple criteria are AND'd together. +options: + age: + required: false + default: null + description: + - Select files whose age is equal to or greater than the specified time. + Use a negative age to find files equal to or less than the specified time. + You can choose seconds, minutes, hours, days, or weeks by specifying the + first letter of any of those words (e.g., "1w"). + patterns: + required: false + default: '*' + description: + - One or more (shell type) file glob patterns, which restrict the list of files to be returned to + those whose basenames match at least one of the patterns specified. Multiple patterns can be + specified using a list. + paths: + required: true + aliases: [ "name" ] + description: + - List of paths to the file or directory to search. All paths must be fully qualified. + file_type: + required: false + description: + - Type of file to select + choices: [ "file", "directory" ] + default: "file" + recurse: + required: false + default: "no" + choices: [ "yes", "no" ] + description: + - If target is a directory, recursively descend into the directory looking for files. + size: + required: false + default: null + description: + - Select files whose size is equal to or greater than the specified size. + Use a negative size to find files equal to or less than the specified size. + Unqualified values are in bytes, but b, k, m, g, and t can be appended to specify + bytes, kilobytes, megabytes, gigabytes, and terabytes, respectively. + Size is not evaluated for directories. + age_stamp: + required: false + default: "mtime" + choices: [ "atime", "mtime", "ctime" ] + description: + - Choose the file property against which we compare age. Default is mtime. + hidden: + required: false + default: "False" + choices: [ True, False ] + description: + - Set this to true to include hidden files, otherwise they'll be ignored. + follow: + required: false + default: "False" + choices: [ True, False ] + description: + - Set this to true to follow symlinks in path. + get_checksum: + required: false + default: "False" + choices: [ True, False ] + description: + - Set this to true to retrieve a file's sha1 checksum +''' + + +EXAMPLES = ''' +# Recursively find /tmp files older than 2 days +- find: paths="/tmp" age="2d" recurse=yes + +# Recursively find /tmp files older than 4 weeks and equal or greater than 1 megabyte +- find: paths="/tmp" age="4w" size="1m" recurse=yes + +# Recursively find /var/tmp files with last access time greater than 3600 seconds +- find: paths="/var/tmp" age="3600" age_stamp=atime recurse=yes + +# find /var/log files equal or greater than 10 megabytes ending with .log or .log.gz +- find: paths="/var/tmp" patterns="*.log","*.log.gz" size="10m" +''' + +RETURN = ''' +files: + description: all matches found with the specified criteria (see stat module for full output of each dictionary) + returned: success + type: list of dictionaries + sample: [ + { path="/var/tmp/test1", + mode=0644, + ..., + checksum=16fac7be61a6e4591a33ef4b729c5c3302307523 + }, + { path="/var/tmp/test2", + ... + }, + ] +matched: + description: number of matches + returned: success + type: string + sample: 14 +examined: + description: number of filesystem objects looked at + returned: success + type: string + sample: 34 +''' + +def pfilter(f, patterns=None): + '''filter using glob patterns''' + if patterns is None: + return True + for p in patterns: + if fnmatch.fnmatch(f, p): + return True + return False + + +def agefilter(st, now, age, timestamp): + '''filter files older than age''' + if age is None or \ + (age >= 0 and now - st.__getattribute__("st_%s" % timestamp) >= abs(age)) or \ + (age < 0 and now - st.__getattribute__("st_%s" % timestamp) <= abs(age)): + + return True + return False + + +def sizefilter(st, size): + '''filter files greater than size''' + if size is None or \ + (size >= 0 and st.st_size >= abs(size)) or \ + (size < 0 and st.st_size <= abs(size)): + + return True + + return False + + +def statinfo(st): + return { + 'mode' : "%04o" % stat.S_IMODE(st.st_mode), + 'isdir' : stat.S_ISDIR(st.st_mode), + 'ischr' : stat.S_ISCHR(st.st_mode), + 'isblk' : stat.S_ISBLK(st.st_mode), + 'isreg' : stat.S_ISREG(st.st_mode), + 'isfifo' : stat.S_ISFIFO(st.st_mode), + 'islnk' : stat.S_ISLNK(st.st_mode), + 'issock' : stat.S_ISSOCK(st.st_mode), + 'uid' : st.st_uid, + 'gid' : st.st_gid, + 'size' : st.st_size, + 'inode' : st.st_ino, + 'dev' : st.st_dev, + 'nlink' : st.st_nlink, + 'atime' : st.st_atime, + 'mtime' : st.st_mtime, + 'ctime' : st.st_ctime, + 'wusr' : bool(st.st_mode & stat.S_IWUSR), + 'rusr' : bool(st.st_mode & stat.S_IRUSR), + 'xusr' : bool(st.st_mode & stat.S_IXUSR), + 'wgrp' : bool(st.st_mode & stat.S_IWGRP), + 'rgrp' : bool(st.st_mode & stat.S_IRGRP), + 'xgrp' : bool(st.st_mode & stat.S_IXGRP), + 'woth' : bool(st.st_mode & stat.S_IWOTH), + 'roth' : bool(st.st_mode & stat.S_IROTH), + 'xoth' : bool(st.st_mode & stat.S_IXOTH), + 'isuid' : bool(st.st_mode & stat.S_ISUID), + 'isgid' : bool(st.st_mode & stat.S_ISGID), + } + + +def main(): + module = AnsibleModule( + argument_spec = dict( + paths = dict(required=True, aliases=['name'], type='list'), + patterns = dict(default=['*'], type='list'), + file_type = dict(default="file", choices=['file', 'directory'], type='str'), + age = dict(default=None, type='str'), + age_stamp = dict(default="mtime", choices=['atime','mtime','ctime'], type='str'), + size = dict(default=None, type='str'), + recurse = dict(default='no', type='bool'), + hidden = dict(default="False", type='bool'), + follow = dict(default="False", type='bool'), + get_checksum = dict(default="False", type='bool'), + ), + ) + + params = module.params + + filelist = [] + + if params['age'] is None: + age = None + else: + # convert age to seconds: + m = re.match("^(-?\d+)(s|m|h|d|w)?$", params['age'].lower()) + seconds_per_unit = {"s": 1, "m": 60, "h": 3600, "d": 86400, "w": 604800} + if m: + age = int(m.group(1)) * seconds_per_unit.get(m.group(2), 1) + else: + module.fail_json(age=params['age'], msg="failed to process age") + + if params['size'] is None: + size = None + else: + # convert size to bytes: + m = re.match("^(-?\d+)(b|k|m|g|t)?$", params['size'].lower()) + bytes_per_unit = {"b": 1, "k": 1024, "m": 1024**2, "g": 1024**3, "t": 1024**4} + if m: + size = int(m.group(1)) * bytes_per_unit.get(m.group(2), 1) + else: + module.fail_json(size=params['size'], msg="failed to process size") + + now = time.time() + msg = '' + looked = 0 + for npath in params['paths']: + if os.path.isdir(npath): + for root,dirs,files in os.walk( npath, followlinks=params['follow'] ): + + looked = looked + len(files) + len(dirs) + for fsobj in (files + dirs): + fsname=os.path.normpath(os.path.join(root, fsobj)) + + if os.path.basename(fsname).startswith('.') and not params['hidden']: + continue + + st = os.stat(fsname) + r = {'path': fsname} + if stat.S_ISDIR(st.st_mode) and params['file_type'] == 'directory': + if pfilter(fsobj, params['patterns']) and agefilter(st, now, age, params['age_stamp']): + + r.update(statinfo(st)) + filelist.append(r) + + elif stat.S_ISREG(st.st_mode) and params['file_type'] == 'file': + if pfilter(fsobj, params['patterns']) and \ + agefilter(st, now, age, params['age_stamp']) and \ + sizefilter(st, size): + + r.update(statinfo(st)) + if params['get_checksum']: + r['checksum'] = module.sha1(fsname) + filelist.append(r) + + if not params['recurse']: + break + else: + msg+="%s was skipped as it does not seem to be a valid directory or it cannot be accessed\n" + + matched = len(filelist) + module.exit_json(files=filelist, changed=False, msg=msg, matched=matched, examined=looked) + +# import module snippets +from ansible.module_utils.basic import * +main() +