parse_capitalone_pdf_statement 1.0.0
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.
- checksums.yaml +7 -0
- data/.gitignore +14 -0
- data/.ruby-gemset +1 -0
- data/.ruby-version +1 -0
- data/.travis.yml +3 -0
- data/.yardopts +4 -0
- data/Gemfile +4 -0
- data/LICENSE.txt +22 -0
- data/README.md +70 -0
- data/Rakefile +27 -0
- data/bin/capitalone_pdf_to_json.rb +13 -0
- data/lib/parse_capitalone_pdf_statement.rb +261 -0
- data/lib/parse_capitalone_pdf_statement/version.rb +3 -0
- data/parse_capitalone_pdf_statement.gemspec +35 -0
- data/test/data/bad_statement1.pdf +0 -0
- data/test/data/bad_statement1.txt +161 -0
- data/test/data/test_statement.json +610 -0
- data/test/data/test_statement.pdf +0 -0
- data/test/data/test_statement.txt +161 -0
- data/test/helper.rb +10 -0
- data/test/test_cli.rb +16 -0
- data/test/test_parser.rb +22 -0
- metadata +163 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: 5d442f232d7445dc288b32cd2aa6810d118f0a9d
|
4
|
+
data.tar.gz: 77a644f514d32bc3ab777c5733e32a92261a5f73
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 241704b7f7219fead5c464e9787ecde2d3af95501147efcf844a770a13013eb785aa3d143d5081b10122fe0f8006ed52b0398954e3d3e221d6a8adb793297036
|
7
|
+
data.tar.gz: 23dd29b08081e26b56d29163e5b1f3b56de8708e4e5691848fd845a1938c21082edd0ec98078dea078c0c9c82c1009b4f4bb5c0c169273646703681073a63dd8
|
data/.gitignore
ADDED
data/.ruby-gemset
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
parse_capitalone_pdf_statement
|
data/.ruby-version
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
ruby-2.1.4
|
data/.travis.yml
ADDED
data/.yardopts
ADDED
data/Gemfile
ADDED
data/LICENSE.txt
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
Copyright (c) 2014 Joe Sortelli
|
2
|
+
|
3
|
+
MIT License
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
6
|
+
a copy of this software and associated documentation files (the
|
7
|
+
"Software"), to deal in the Software without restriction, including
|
8
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
9
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
10
|
+
permit persons to whom the Software is furnished to do so, subject to
|
11
|
+
the following conditions:
|
12
|
+
|
13
|
+
The above copyright notice and this permission notice shall be
|
14
|
+
included in all copies or substantial portions of the Software.
|
15
|
+
|
16
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
17
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
18
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
19
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
20
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
21
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
22
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,70 @@
|
|
1
|
+
# parse_capitalone_pdf_statement
|
2
|
+
|
3
|
+
[](http://badge.fury.io/rb/parse_capitalone_pdf_statement)
|
4
|
+
[](https://travis-ci.org/sortelli/parse_capitalone_pdf_statement)
|
5
|
+
[](https://gemnasium.com/sortelli/parse_capitalone_pdf_statement)
|
6
|
+
|
7
|
+
|
8
|
+
The Capital One website only provides a way to download structured
|
9
|
+
data of credit card transaction history for the previous 180 days.
|
10
|
+
However, you are able to download monthly PDF account statements
|
11
|
+
for the previous few years.
|
12
|
+
|
13
|
+
This library allows you to parse a Capital One PDF monthly statement,
|
14
|
+
and access structured transaction history data.
|
15
|
+
|
16
|
+
## Installation
|
17
|
+
|
18
|
+
Add this line to your application's Gemfile:
|
19
|
+
|
20
|
+
```ruby
|
21
|
+
gem 'parse_capitalone_pdf_statement'
|
22
|
+
```
|
23
|
+
|
24
|
+
And then execute:
|
25
|
+
|
26
|
+
```bash
|
27
|
+
% bundle
|
28
|
+
```
|
29
|
+
|
30
|
+
Or install it yourself as:
|
31
|
+
|
32
|
+
```bash
|
33
|
+
% gem install parse_capitalone_pdf_statement
|
34
|
+
```
|
35
|
+
|
36
|
+
## Convert PDF to JSON
|
37
|
+
|
38
|
+
Use the ```capitalone_pdf_to_json.rb``` script to convert a PDF
|
39
|
+
montly statement to JSON.
|
40
|
+
|
41
|
+
```bash
|
42
|
+
% capitalone_pdf_to_json.rb my_monthly_statement.pdf > my_monthly_statement.json
|
43
|
+
```
|
44
|
+
|
45
|
+
## API Example
|
46
|
+
|
47
|
+
Parse a PDF monthly statement and print all payments:
|
48
|
+
|
49
|
+
```ruby
|
50
|
+
require 'parse_capitalone_pdf_statement'
|
51
|
+
|
52
|
+
statement = CapitalOneStatement.new('/path/to/my_monthly_statement.pdf')
|
53
|
+
|
54
|
+
statement.payments.each do |payment|
|
55
|
+
puts 'Transaction ID: %d' % payment.id
|
56
|
+
puts 'Date: %s' % payment.date
|
57
|
+
puts 'Description: %s' % payment.description
|
58
|
+
puts 'Amount: %.2f' % payment.amount
|
59
|
+
end
|
60
|
+
```
|
61
|
+
|
62
|
+
See the [API
|
63
|
+
Documentation](http://sortelli.github.io/parse_capitalone_pdf_statement/frames.html#!CapitalOneStatement.html)
|
64
|
+
for more information.
|
65
|
+
|
66
|
+
## License
|
67
|
+
|
68
|
+
Copyright (c) 2014 Joe Sortelli
|
69
|
+
|
70
|
+
MIT License
|
data/Rakefile
ADDED
@@ -0,0 +1,27 @@
|
|
1
|
+
require 'bundler/gem_tasks'
|
2
|
+
require 'rake/testtask'
|
3
|
+
require 'yard'
|
4
|
+
|
5
|
+
YARD::Rake::YardocTask.new
|
6
|
+
|
7
|
+
Rake::TestTask.new do |t|
|
8
|
+
t.test_files = FileList['test/test_*.rb']
|
9
|
+
end
|
10
|
+
|
11
|
+
task :build => :test
|
12
|
+
|
13
|
+
task :coverage do
|
14
|
+
ENV['COVERAGE'] = 'true'
|
15
|
+
Rake::Task['test'].execute
|
16
|
+
end
|
17
|
+
|
18
|
+
desc "Create test pdf files"
|
19
|
+
task :make_test_pdf do
|
20
|
+
text_to_pdf = 'enscript -B -f "Times-Roman6.0" %s --output=- | ps2pdf - > %s'
|
21
|
+
|
22
|
+
Dir.chdir(File.join(File.dirname(__FILE__), 'test', 'data')) do
|
23
|
+
%w{test_statement bad_statement1}.each do |name|
|
24
|
+
system(text_to_pdf % [name + '.txt', name + '.pdf'])
|
25
|
+
end
|
26
|
+
end
|
27
|
+
end
|
@@ -0,0 +1,13 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
|
3
|
+
require "bundler/setup"
|
4
|
+
require "parse_capitalone_pdf_statement"
|
5
|
+
|
6
|
+
unless ARGV.size == 1
|
7
|
+
$stderr.puts "usage: #{File.basename(__FILE__)} statement.pdf"
|
8
|
+
exit 1
|
9
|
+
end
|
10
|
+
|
11
|
+
pdf_path = ARGV.first
|
12
|
+
|
13
|
+
puts JSON.pretty_generate(CapitalOneStatement.new(pdf_path))
|
@@ -0,0 +1,261 @@
|
|
1
|
+
require 'parse_capitalone_pdf_statement/version'
|
2
|
+
require 'pdf-reader'
|
3
|
+
require 'json'
|
4
|
+
require 'date'
|
5
|
+
|
6
|
+
# CapitalOneStatement object, with data parsed from a PDF monthly statement.
|
7
|
+
#
|
8
|
+
# @!attribute start_date [r]
|
9
|
+
# @return [Date] the first day of the monthly statement
|
10
|
+
#
|
11
|
+
# @!attribute end_date [r]
|
12
|
+
# @return [Date] the final day of the monthly statement
|
13
|
+
#
|
14
|
+
# @!attribute previous_balance [r]
|
15
|
+
# @return [Float] the "Previous Balance" field listed in the statement
|
16
|
+
#
|
17
|
+
# @!attribute total_payments [r]
|
18
|
+
# @return [Float] the "Payments and Credits" field listed in the statement
|
19
|
+
#
|
20
|
+
# @!attribute total_fees [r]
|
21
|
+
# @return [Float] the "Fees and Interest Charged" field listed in the statement
|
22
|
+
#
|
23
|
+
# @!attribute total_transactions [r]
|
24
|
+
# @return [Float] the "Transactions" field listed in the statement
|
25
|
+
#
|
26
|
+
# @!attribute new_balance [r]
|
27
|
+
# @return [Float] the New Balance listed in the statement
|
28
|
+
#
|
29
|
+
# @!attribute payments [r]
|
30
|
+
# @return [Array<CapitalOneStatement::Transaction>] array of payment transactions
|
31
|
+
#
|
32
|
+
# @!attribute transactions [r]
|
33
|
+
# @return [Array<CapitalOneStatement::Transaction>] array of charge transactions
|
34
|
+
#
|
35
|
+
# @!attribute fees [r]
|
36
|
+
# @return [Array<CapitalOneStatement::Transaction>] array of fee transactions
|
37
|
+
|
38
|
+
class CapitalOneStatement
|
39
|
+
DATE_REGEX = /(\w{3})\. (\d\d) - (\w{3})\. (\d\d), (\d{4})/
|
40
|
+
AMOUNT_REGEX = /\(?\$[\d,]+\.\d\d\)?/
|
41
|
+
AMOUNT_ONLY_REGEX = /^ *#{AMOUNT_REGEX.source} *$/
|
42
|
+
FEES_REGEX = /Total Fees This Period +(#{AMOUNT_REGEX.source})/
|
43
|
+
INTEREST_REGEX = /Total Interest This Period +(#{AMOUNT_REGEX.source})/
|
44
|
+
TRANSACTION_REGEX = /^ *(\d+) +(\d\d) ([A-Z][A-Z][A-Z]) (.+[^ ]) +(#{
|
45
|
+
AMOUNT_REGEX.source
|
46
|
+
}) *$/
|
47
|
+
|
48
|
+
attr_reader :start_date,
|
49
|
+
:end_date,
|
50
|
+
:previous_balance,
|
51
|
+
:total_payments,
|
52
|
+
:total_fees,
|
53
|
+
:total_transactions,
|
54
|
+
:new_balance,
|
55
|
+
:payments,
|
56
|
+
:transactions
|
57
|
+
|
58
|
+
def initialize(pdf_path)
|
59
|
+
@dec_from_prev_year = nil
|
60
|
+
@year = nil
|
61
|
+
@start_date = nil
|
62
|
+
@end_date = nil
|
63
|
+
@previous_balance = nil
|
64
|
+
@new_balance = nil
|
65
|
+
@total_payments = nil
|
66
|
+
@total_transactions = nil
|
67
|
+
@total_fees = nil
|
68
|
+
@payments = []
|
69
|
+
@transactions = []
|
70
|
+
@fees = []
|
71
|
+
|
72
|
+
parse_from_pdf pdf_path
|
73
|
+
|
74
|
+
%w{payments transactions fees}.each do |type|
|
75
|
+
trxs = "@#{type}"
|
76
|
+
total = "@total_#{type}"
|
77
|
+
|
78
|
+
instance_variable_set(trxs, instance_variable_get(trxs).sort_by {|trx| trx[:id]})
|
79
|
+
|
80
|
+
check_total(
|
81
|
+
type,
|
82
|
+
instance_variable_get(total),
|
83
|
+
instance_variable_get(trxs).inject(0) {|sum, trx| sum += trx[:amount]}
|
84
|
+
)
|
85
|
+
end
|
86
|
+
end
|
87
|
+
|
88
|
+
def to_json(*args)
|
89
|
+
{
|
90
|
+
:start_date => @start_date,
|
91
|
+
:end_date => @end_date,
|
92
|
+
:previous_balance => @previous_balance,
|
93
|
+
:total_payments => @total_payments,
|
94
|
+
:total_fees => @total_fees,
|
95
|
+
:total_transactions => @total_transactions,
|
96
|
+
:new_balance => @new_balance,
|
97
|
+
:payments => @payments,
|
98
|
+
:transactions => @transactions,
|
99
|
+
:fees => @fees
|
100
|
+
}.to_json(*args)
|
101
|
+
end
|
102
|
+
|
103
|
+
private
|
104
|
+
|
105
|
+
def parse_from_pdf(pdf_path)
|
106
|
+
PDF::Reader.new(pdf_path).pages.each_with_index do |page, page_num|
|
107
|
+
if @year.nil?
|
108
|
+
walker = Struct.new(:year, :offset, :start_date, :end_date) do
|
109
|
+
def respond_to?(_)
|
110
|
+
true
|
111
|
+
end
|
112
|
+
|
113
|
+
def method_missing(name, *args)
|
114
|
+
return unless name =~ /show_text/
|
115
|
+
|
116
|
+
if args.any? {|str| str.to_s =~ DATE_REGEX}
|
117
|
+
self.offset = ($1.upcase == 'DEC' && $3.upcase == 'JAN') ? 1 : 0
|
118
|
+
self.year = $5.to_i
|
119
|
+
self.start_date = Date.parse('%s-%s-%s' % [year - offset, $1, $2])
|
120
|
+
self.end_date = Date.parse('%s-%s-%s' % [year, $3, $4])
|
121
|
+
end
|
122
|
+
end
|
123
|
+
end.new
|
124
|
+
|
125
|
+
page.walk walker
|
126
|
+
|
127
|
+
@dec_from_prev_year = walker.offset == 1
|
128
|
+
@year = walker.year
|
129
|
+
@start_date = walker.start_date
|
130
|
+
@end_date = walker.end_date
|
131
|
+
end
|
132
|
+
|
133
|
+
enum = page.text.split("\n").each
|
134
|
+
|
135
|
+
loop do
|
136
|
+
current_line = enum.next
|
137
|
+
enum.next until (enum.peek rescue nil) != ''
|
138
|
+
next_line = (enum.peek rescue nil)
|
139
|
+
|
140
|
+
parse_pdf_line page_num, current_line, next_line
|
141
|
+
end
|
142
|
+
end
|
143
|
+
end
|
144
|
+
|
145
|
+
def parse_pdf_line(page_num, line, next_line)
|
146
|
+
if @previous_balance.nil?
|
147
|
+
amount_strs = line.scan AMOUNT_REGEX
|
148
|
+
if amount_strs.size == 5
|
149
|
+
@previous_balance,
|
150
|
+
@total_payments,
|
151
|
+
@total_fees,
|
152
|
+
@total_transactions,
|
153
|
+
@new_balance = amount_strs.map {|amount| parse_amount(amount)}
|
154
|
+
|
155
|
+
@total_payments = -@total_payments
|
156
|
+
end
|
157
|
+
end
|
158
|
+
|
159
|
+
if line =~ FEES_REGEX && $1 != '$0.00'
|
160
|
+
check_billing_cycle
|
161
|
+
@fees << Transaction.new(
|
162
|
+
@fees.size + 1,
|
163
|
+
@end_date, "CAPITAL ONE MEMBER FEE",
|
164
|
+
$1,
|
165
|
+
parse_amount($1)
|
166
|
+
)
|
167
|
+
end
|
168
|
+
|
169
|
+
if line =~ INTEREST_REGEX && $1 != '$0.00'
|
170
|
+
check_billing_cycle
|
171
|
+
@fees << Transaction.new(
|
172
|
+
@fees.size + 1,
|
173
|
+
@end_date, "INTEREST CHARGE:PURCHASES",
|
174
|
+
$1,
|
175
|
+
parse_amount($1)
|
176
|
+
)
|
177
|
+
end
|
178
|
+
|
179
|
+
transactions, payments = [(0..78), (80..-1)].map do |index|
|
180
|
+
str = line .to_s[index].to_s
|
181
|
+
next_str = next_line.to_s[index].to_s
|
182
|
+
|
183
|
+
repair_transaction_line str, next_str
|
184
|
+
end.map do |str|
|
185
|
+
parse_transaction(str)
|
186
|
+
end.compact.partition do |trx|
|
187
|
+
trx[:amount] >= 0
|
188
|
+
end
|
189
|
+
|
190
|
+
@transactions += transactions
|
191
|
+
@payments += payments
|
192
|
+
end
|
193
|
+
|
194
|
+
def repair_transaction_line(line, next_line)
|
195
|
+
if next_line =~ AMOUNT_ONLY_REGEX && !(line =~ AMOUNT_REGEX)
|
196
|
+
line += " #{next_line.strip}"
|
197
|
+
else
|
198
|
+
line
|
199
|
+
end
|
200
|
+
end
|
201
|
+
|
202
|
+
def parse_transaction(line)
|
203
|
+
return nil unless line =~ TRANSACTION_REGEX &&
|
204
|
+
$4 != "CAPITAL ONE MEMBER FEE"
|
205
|
+
|
206
|
+
check_billing_cycle
|
207
|
+
|
208
|
+
year = ($3.upcase == 'DEC' && @dec_from_prev_year) ? @year - 1 : @year
|
209
|
+
date = Date.parse('%s-%s-%s' % [year, $3, $2])
|
210
|
+
|
211
|
+
Transaction.new($1.to_i, date, $4, $5, parse_amount($5))
|
212
|
+
end
|
213
|
+
|
214
|
+
def parse_amount(amount)
|
215
|
+
num = amount.gsub(/[^\d.]/, '').to_f
|
216
|
+
amount.start_with?(?() ? -num : num
|
217
|
+
end
|
218
|
+
|
219
|
+
def check_billing_cycle
|
220
|
+
raise "Failed to determine billing cycle dates" if @year.nil?
|
221
|
+
end
|
222
|
+
|
223
|
+
def check_total(type, expected, actual)
|
224
|
+
return if actual.round(2) == expected.round(2)
|
225
|
+
|
226
|
+
raise "Calculated %s mismatch %.2f != %.2f" % [
|
227
|
+
type,
|
228
|
+
actual,
|
229
|
+
expected
|
230
|
+
]
|
231
|
+
end
|
232
|
+
|
233
|
+
# CapitalOneStatement::Transaction represents a single credit transaction
|
234
|
+
class CapitalOneStatement::Transaction < Struct.new(
|
235
|
+
:id,
|
236
|
+
:date,
|
237
|
+
:description,
|
238
|
+
:amount_str,
|
239
|
+
:amount
|
240
|
+
)
|
241
|
+
# @!attribute id
|
242
|
+
# @return [Fixnum] transaction id
|
243
|
+
#
|
244
|
+
# @!attribute date
|
245
|
+
# @return [Date] the date of the transaction
|
246
|
+
#
|
247
|
+
# @!attribute description
|
248
|
+
# @return [String] the description of the transaction
|
249
|
+
#
|
250
|
+
# @!attribute amount_str
|
251
|
+
# @return [String] the dollar amount string of the transaction
|
252
|
+
#
|
253
|
+
# @!attribute amount
|
254
|
+
# @return [Float] the dollar amount parsed into a Float, negative for payments
|
255
|
+
|
256
|
+
# @return [String] JSON representation of Transaction
|
257
|
+
def to_json(*args)
|
258
|
+
to_h.to_json(*args)
|
259
|
+
end
|
260
|
+
end
|
261
|
+
end
|