contrib: intro the sort_downloads script
[nit.git] / contrib / sort_downloads / src / sort_downloads.nit
1 #!/usr/bin/env nit
2
3 # This file is part of NIT ( http://www.nitlanguage.org ).
4 #
5 # Copyright 2013 Alexis Laferrière <alexis.laf@xymus.net>
6 #
7 # Licensed under the Apache License, Version 2.0 (the "License");
8 # you may not use this file except in compliance with the License.
9 # You may obtain a copy of the License at
10 #
11 # http://www.apache.org/licenses/LICENSE-2.0
12 #
13 # Unless required by applicable law or agreed to in writing, software
14 # distributed under the License is distributed on an "AS IS" BASIS,
15 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16 # See the License for the specific language governing permissions and
17 # limitations under the License.
18
19 # Smart file sorting using folder names to sort files. Has built in support for
20 # a distributed or multi-disk setup.
21 module sort_downloads
22
23 # TODO sort patterns by longer first
24 # TODO allow for config overriding using `opts`
25
26 import opts
27
28 `{
29 #include <errno.h>
30 `}
31
32 # Local config
33 # Modify these according to your needs. It is also possible to have alternates
34 # using class refinment and a main calling to super.
35 class Config
36 # Source directory where are the files to be sorted.
37 var source_dir = "~/Downloads/"
38
39 # Destination super directory where classification directories will be
40 # created to hold files.
41 var dest_dir = "~/Videos/"
42
43 # Super directories with wanted folder names, which will be used to sort
44 # the files (only their name are used, the files won't be copied there).
45 var regex_source_dirs: Array[String] = ["~/Videos/"]
46
47 # Will only sort files older than the number of `elapsed_days`.
48 var elapsed_days = 7
49 end
50
51 redef class String
52 fun check_file_existence: Bool
53 do
54 if not file_exists then
55 print "config error: file \"{self}\" does not exists."
56 return true
57 else return false
58 end
59
60 # Returns null on success
61 fun file_rename_to(dest: String): nullable String import String::to_cstring,
62 String::from_cstring, String as nullable `{
63 int res = rename(String_to_cstring(recv), String_to_cstring(dest));
64 if (res == 0) return null_String();
65 return String_as_nullable(new_String_from_cstring(strerror(errno)));
66 `}
67
68 # Replace `~` by the path to the home diretory
69 fun replace_tilde: String
70 do
71 var home_folder = "HOME".environ
72 return replace("~", home_folder)
73 end
74 end
75
76 # Hack to keep track of the real directory name associated to this pattern
77 redef class BM_Pattern
78 var dir: String
79
80 init with_dir(motif, dir: String)
81 do
82 init(motif)
83 self.dir = dir
84 end
85 end
86
87 class XySorter
88 var opts_context = new OptionContext
89 var opt_help = new OptionBool("Print this help message", "-h", "--help")
90 var opt_list_series = new OptionBool("Only list the folders in the regex target directories", "--list-series")
91 var opt_verbose = new OptionBool("Print information about the operations", "-v", "--verbose")
92 var opt_dry_run = new OptionBool("Simulate work without modifying the filesystem", "--dry-run")
93
94 fun fill_opts_context
95 do
96 opts_context.add_option(opt_help, opt_list_series, opt_verbose, opt_dry_run)
97 end
98
99 fun run(source_dir, dest_dir: String, regex_source_dirs: Sequence[String], older_than: TimeT )
100 do
101 # manage command line options
102 fill_opts_context
103 opts_context.parse(args)
104 if not opts_context.rest.is_empty or opt_help.value then
105 print "Usage: {sys.program_name} [Options]"
106 print "Options:"
107 opts_context.usage
108
109 if opt_help.value then exit 0
110 exit 1
111 end
112
113 # replace `~` by home path
114 source_dir = source_dir.replace_tilde
115 dest_dir = dest_dir.replace_tilde
116 for f in [0..regex_source_dirs.length[ do
117 regex_source_dirs[f] = regex_source_dirs[f].replace_tilde
118 end
119
120 # check config validity
121 var failed = false
122 failed = failed or source_dir.check_file_existence
123 failed = failed or dest_dir.check_file_existence
124 for dir in regex_source_dirs do failed = failed or dir.check_file_existence
125 if failed then exit 1
126
127 # collect possible series dir names
128 var dirs_name = new HashSet[String]
129 for dir in regex_source_dirs do for file in dir.files do dirs_name.add(file)
130
131 # if asked only to print ou the list of series, do so and quit
132 if opt_list_series.value then
133 print dirs_name.join(", ")
134 return
135 end
136
137 # build regexxes
138 var patterns = new HashSet[BM_Pattern]
139 for dir in dirs_name do
140 patterns.add new BM_Pattern.with_dir(dir, dir)
141 patterns.add new BM_Pattern.with_dir(dir.replace(' ', "."), dir)
142 patterns.add new BM_Pattern.with_dir(dir.replace(' ', "_"), dir)
143 end
144
145 # compare source files with patterns and sort
146 for file in source_dir.files do
147 var full_source = source_dir + "/" + file
148 var stat = full_source.file_lstat
149
150 # if not a file or dir, skip
151 if not stat.is_reg and not stat.is_dir then continue
152
153 # is it old enough?
154 var time = new TimeT.from_i(stat.mtime)
155 if time.to_i > older_than.to_i then continue
156
157 # does it fit our regexxes?
158 var move_to_dir: nullable String = null
159 for pattern in patterns do if file.search(pattern) != null then
160 move_to_dir = pattern.dir
161 break
162 end
163
164 # lets' move it!
165 if move_to_dir != null then
166 var full_dir_dest = dest_dir + "/" + move_to_dir
167 var full_dest = full_dir_dest + "/" + file
168
169 if opt_verbose.value then print "moving {full_source} -> {full_dest}"
170
171 if not opt_dry_run.value then
172 if not full_dir_dest.file_exists then full_dir_dest.mkdir
173
174 var res = full_source.file_rename_to(full_dest)
175 if res != null then
176 print "Moving error: {res}"
177 abort
178 end
179 end
180 end
181 end
182 end
183 end
184
185 var config = new Config
186 var sorter = new XySorter
187
188 # calculate cut off time using `elapsed_days` compared to now
189 var now = new TimeT
190 var wanted_elapsed_secs = config.elapsed_days * 24 * 60 * 60
191 var from_time = new TimeT.from_i(now.to_i - wanted_elapsed_secs)
192
193 sorter.run(config.source_dir, config.dest_dir, config.regex_source_dirs, from_time)