masticate 0.1.4 → 0.1.5
Sign up to get free protection for your applications and to get access to all the features.
- data/bin/masticate +18 -14
- data/lib/masticate/max_rows.rb +2 -5
- data/lib/masticate/version.rb +1 -1
- metadata +7 -7
data/bin/masticate
CHANGED
@@ -9,52 +9,56 @@ options = {}
|
|
9
9
|
OptionParser.new do |opts|
|
10
10
|
opts.banner = "Usage: example.rb [options]"
|
11
11
|
|
12
|
-
opts.on("--format FORMAT", "Specify format") do |v|
|
12
|
+
opts.on("--format FORMAT", String, "Specify format") do |v|
|
13
13
|
options[:format] = v
|
14
14
|
end
|
15
15
|
|
16
|
-
opts.on("--delim DELIMITER", "Specify field delimiter (character or TAB; default is ',')") do |v|
|
16
|
+
opts.on("--delim DELIMITER", String, "Specify field delimiter (character or TAB; default is ',')") do |v|
|
17
17
|
options[:col_sep] = v
|
18
18
|
options[:col_sep] = "\t" if options[:col_sep] == "TAB"
|
19
19
|
end
|
20
20
|
|
21
|
-
opts.on("--quote QUOTE-CHAR", "Specify character used for quoting fields (optional; default is no quoting)") do |char|
|
21
|
+
opts.on("--quote QUOTE-CHAR", String, "Specify character used for quoting fields (optional; default is no quoting)") do |char|
|
22
22
|
options[:quote_char] = char
|
23
23
|
end
|
24
24
|
|
25
|
+
opts.on("--stats", "(for *sniff*) collect & display input stats") do
|
26
|
+
options[:stats] = true
|
27
|
+
end
|
28
|
+
|
25
29
|
opts.on("--fields LIST", Array, "Specify fields to select") do |list|
|
26
30
|
options[:fields] = list
|
27
31
|
end
|
28
32
|
|
29
|
-
opts.on("--field FIELD", "Specify field to convert") do |f|
|
33
|
+
opts.on("--field FIELD", String, "Specify field to convert") do |f|
|
30
34
|
options[:field] = f
|
31
35
|
end
|
32
36
|
|
33
|
-
opts.on("--snip DIRECTIVE", "Specify header fields to snip: first N, or by name") do |f|
|
37
|
+
opts.on("--snip DIRECTIVE", String, "Specify header fields to snip: first N, or by name") do |f|
|
34
38
|
options[:snip] = f.to_i
|
35
39
|
end
|
36
40
|
|
37
|
-
opts.on("--from REGEXP", "Regular expression for gsub conversion") do |s|
|
41
|
+
opts.on("--from REGEXP", String, "Regular expression for gsub conversion") do |s|
|
38
42
|
options[:from] = s
|
39
43
|
end
|
40
44
|
|
41
|
-
opts.on("--to STRING", "Result string for gsub conversion") do |s|
|
45
|
+
opts.on("--to STRING", String, "Result string for gsub conversion") do |s|
|
42
46
|
options[:to] = s
|
43
47
|
end
|
44
48
|
|
45
|
-
opts.on("--inlined", "(for *mend* only) Source file has headers inlined on each line") do |
|
46
|
-
options[:inlined] =
|
49
|
+
opts.on("--inlined", "(for *mend* only) Source file has headers inlined on each line") do |b|
|
50
|
+
options[:inlined] = true
|
47
51
|
end
|
48
52
|
|
49
|
-
opts.on("--dejunk", "(for *mend* only) Expunge junk lines from source") do |
|
50
|
-
options[:dejunk] =
|
53
|
+
opts.on("--dejunk", "(for *mend* only) Expunge junk lines from source") do |b|
|
54
|
+
options[:dejunk] = true
|
51
55
|
end
|
52
56
|
|
53
|
-
opts.on("--by FIELD", "(for *maxrows* only) Field to group by") do |f|
|
57
|
+
opts.on("--by FIELD", String, "(for *maxrows* only) Field to group by") do |f|
|
54
58
|
options[:by] = f
|
55
59
|
end
|
56
60
|
|
57
|
-
opts.on("--max FIELD", "(for *maxrows* only) Field to find max value for") do |f|
|
61
|
+
opts.on("--max FIELD", String, "(for *maxrows* only) Field to find max value for") do |f|
|
58
62
|
options[:max] = f
|
59
63
|
end
|
60
64
|
end.parse!
|
@@ -74,7 +78,7 @@ end
|
|
74
78
|
|
75
79
|
case command
|
76
80
|
when 'sniff'
|
77
|
-
results = Masticate.sniff(filename)
|
81
|
+
results = Masticate.sniff(filename, options)
|
78
82
|
col_sep = results[:col_sep]
|
79
83
|
col_sep = "TAB" if col_sep == "\t"
|
80
84
|
quote_char = results[:quote_char] || "NONE"
|
data/lib/masticate/max_rows.rb
CHANGED
@@ -3,10 +3,7 @@ require "csv"
|
|
3
3
|
|
4
4
|
class Masticate::MaxRows < Masticate::Base
|
5
5
|
def maxrows(opts)
|
6
|
-
|
7
|
-
csv_options = {}
|
8
|
-
csv_options[:col_sep] = opts[:col_sep] if opts[:col_sep]
|
9
|
-
csv_options[:quote_char] = opts[:quote_char] || "\0"
|
6
|
+
standard_options(opts)
|
10
7
|
|
11
8
|
groupby = opts[:by] or raise "missing field to group by"
|
12
9
|
maxon = opts[:max] or raise "missing field to max on"
|
@@ -44,7 +41,7 @@ class Masticate::MaxRows < Masticate::Base
|
|
44
41
|
@output.close if opts[:output]
|
45
42
|
|
46
43
|
{
|
47
|
-
:input_count => input_count,
|
44
|
+
:input_count => @input_count,
|
48
45
|
:output_count => @output_count
|
49
46
|
}
|
50
47
|
end
|
data/lib/masticate/version.rb
CHANGED
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: masticate
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.1.
|
4
|
+
version: 0.1.5
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -13,7 +13,7 @@ date: 2012-04-16 00:00:00.000000000 Z
|
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: rspec
|
16
|
-
requirement: &
|
16
|
+
requirement: &2156618420 !ruby/object:Gem::Requirement
|
17
17
|
none: false
|
18
18
|
requirements:
|
19
19
|
- - ~>
|
@@ -21,10 +21,10 @@ dependencies:
|
|
21
21
|
version: 2.9.0
|
22
22
|
type: :development
|
23
23
|
prerelease: false
|
24
|
-
version_requirements: *
|
24
|
+
version_requirements: *2156618420
|
25
25
|
- !ruby/object:Gem::Dependency
|
26
26
|
name: guard-rspec
|
27
|
-
requirement: &
|
27
|
+
requirement: &2156617660 !ruby/object:Gem::Requirement
|
28
28
|
none: false
|
29
29
|
requirements:
|
30
30
|
- - ~>
|
@@ -32,10 +32,10 @@ dependencies:
|
|
32
32
|
version: 0.7.0
|
33
33
|
type: :development
|
34
34
|
prerelease: false
|
35
|
-
version_requirements: *
|
35
|
+
version_requirements: *2156617660
|
36
36
|
- !ruby/object:Gem::Dependency
|
37
37
|
name: ruby_gntp
|
38
|
-
requirement: &
|
38
|
+
requirement: &2156616980 !ruby/object:Gem::Requirement
|
39
39
|
none: false
|
40
40
|
requirements:
|
41
41
|
- - ~>
|
@@ -43,7 +43,7 @@ dependencies:
|
|
43
43
|
version: 0.3.4
|
44
44
|
type: :development
|
45
45
|
prerelease: false
|
46
|
-
version_requirements: *
|
46
|
+
version_requirements: *2156616980
|
47
47
|
description: Data file crunching
|
48
48
|
email:
|
49
49
|
- jmay@pobox.com
|