xtotxt 0.7 → 0.8
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- data/lib/xtotxt.rb +22 -13
- metadata +4 -4
data/lib/xtotxt.rb
CHANGED
@@ -3,8 +3,9 @@ require 'yaml'
|
|
3
3
|
class XtotxtError < StandardError; end
|
4
4
|
|
5
5
|
class Xtotxt
|
6
|
-
VERSION = 0.
|
6
|
+
VERSION = 0.8
|
7
7
|
SUPPORTED_EXTENSIONS = %w{txt pdf doc docx odt rtf html}
|
8
|
+
TMP_DIR = "/tmp"
|
8
9
|
|
9
10
|
@@config_file_name = "xtotxt.yml"
|
10
11
|
@@dirs_to_check = %w{. ~ /etc}
|
@@ -23,14 +24,14 @@ class Xtotxt
|
|
23
24
|
@@ext = @ext_default
|
24
25
|
end
|
25
26
|
|
26
|
-
def
|
27
|
-
dot_ext
|
27
|
+
def convert_file(input_file_name, tmp_dir = TMP_DIR)
|
28
|
+
dot_ext = File.extname(input_file_name)
|
28
29
|
file_ext = dot_ext.slice(1,dot_ext.length)
|
29
30
|
raise XtotxtError.new("not a supported document extension: #{file_ext}") unless SUPPORTED_EXTENSIONS.member?(file_ext)
|
30
31
|
|
31
|
-
file_base = File.basename(input_file_name,
|
32
|
+
file_base = File.basename(input_file_name, dot_ext)
|
32
33
|
|
33
|
-
output_file_name = "#{tmp_dir}/#{file_base}txt"
|
34
|
+
output_file_name = "#{tmp_dir}/#{file_base}.txt"
|
34
35
|
|
35
36
|
command_line = case file_ext
|
36
37
|
when "txt"
|
@@ -55,22 +56,30 @@ class Xtotxt
|
|
55
56
|
#puts "executing: #{command_line}"
|
56
57
|
|
57
58
|
command_output = `#{command_line} 2>/dev/null` if command_line and not command_line.empty?
|
58
|
-
|
59
|
-
File.
|
59
|
+
if $? == 0 && File.exists?(output_file_name)
|
60
|
+
File.new(output_file_name,"r")
|
60
61
|
else
|
61
62
|
raise XtotxtError.new("Failed to convert #{input_file_name}. Exit status: #{$?.exitstatus}. Output: #{command_output}")
|
62
63
|
end
|
64
|
+
end
|
63
65
|
|
64
|
-
File.delete(output_file_name) unless retain_output
|
65
66
|
|
66
|
-
|
67
|
-
|
68
|
-
|
69
|
-
|
70
|
-
|
67
|
+
def convert(input_file_name,tmp_dir = TMP_DIR,retain_output=false)
|
68
|
+
|
69
|
+
file = convert_file(input_file_name, tmp_dir)
|
70
|
+
text = file.read
|
71
|
+
file.close
|
72
|
+
File.delete(file.path) unless retain_output
|
73
|
+
|
74
|
+
case File.extname(input_file_name)
|
75
|
+
when ".rtf"
|
76
|
+
skip_unrtf_header(text)
|
77
|
+
else
|
78
|
+
text
|
71
79
|
end
|
72
80
|
end
|
73
81
|
|
82
|
+
|
74
83
|
def initialize(ext=nil)
|
75
84
|
@ext =
|
76
85
|
case
|
metadata
CHANGED
@@ -1,12 +1,12 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: xtotxt
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
hash:
|
4
|
+
hash: 27
|
5
5
|
prerelease:
|
6
6
|
segments:
|
7
7
|
- 0
|
8
|
-
-
|
9
|
-
version: "0.
|
8
|
+
- 8
|
9
|
+
version: "0.8"
|
10
10
|
platform: ruby
|
11
11
|
authors:
|
12
12
|
- Alexy Khrabrov
|
@@ -14,7 +14,7 @@ autorequire:
|
|
14
14
|
bindir: bin
|
15
15
|
cert_chain: []
|
16
16
|
|
17
|
-
date: 2011-10-
|
17
|
+
date: 2011-10-12 00:00:00 -07:00
|
18
18
|
default_executable:
|
19
19
|
dependencies: []
|
20
20
|
|