init

.gitignore

@@ -0,0 +1 @@
+*.pyc

PdfInfo.py

@@ -0,0 +1,32 @@
+import subprocess
+
+"""
+ideas from https://gist.github.com/godber/7692812
+"""
+
+class PdfInfo:
+    def __init__(self, filepath):
+        self.filepath = filepath
+        self.output = {}
+        self.cmd = "pdfinfo"
+        self.process()
+
+    def process(self):
+        labels = ['Title', 'Author', 'Creator', 'Producer', 'CreationDate', \
+                'ModDate', 'Tagged', 'Pages', 'Encrypted', 'Page size', \
+                'File size', 'Optimized', 'PDF version']        
+        cmdOutput = subprocess.check_output([self.cmd, self.filepath])
+        for line in cmdOutput.splitlines():
+            for label in labels:
+                if label in line:
+                    self.output[label] = self.extract(line)
+
+    def extract(self, row):
+        return row.split(':', 1)[1].strip()
+
+    def getPages(self):
+        return int(self.output['Pages'])
+
+    def getFileSizeInBytes(self):
+        return int(self.output['File size'][:-5].strip())
+

PdfProcessor.py

@@ -0,0 +1,32 @@
+from os import listdir
+import os.path
+from PdfInfo import *
+from PdfToText import *
+
+class PDFProcessor:
+    def __init__(self, filePath, outputDir):
+        self.filePath = filePath
+        self.outputDir = outputDir
+        self.totalPages = 0
+        self.structured = False
+        pass
+
+    def process(self):
+        pdfInfo = PdfInfo(self.filePath)
+        self.totalPages = pdfInfo.getPages()
+        pdfToText = PdfToText(self.filePath, self.totalPages, self.outputDir)
+        pdfToText.extractPages()
+
+    def smellExtractedPages(self):
+        for f in listdir(self.outputDir):
+            if os.path.isfile(os.path.join(self.outputDir, f)):
+                txt = open(os.path.join(self.outputDir, f))
+
+    def isStructure(self):
+        return True if self.structured else False
+
+    def __init__(self, filePath):
+        self.cmd = ''
+
+
+

PdfToText.py

@@ -0,0 +1,25 @@
+import subprocess
+import os.path
+
+"""
+ideas from https://gist.github.com/godber/7692812
+"""
+
+class PdfToText:
+    def __init__(self, filepath, pages, outputDir):
+        self.filepath = filepath
+        self.pages = pages
+        self.output = {}
+        self.outputDir = outputDir
+        self.cmd = "pdftotext"        
+
+    def extractPage(self, page):
+        outputFileName = os.path.join(self.outputDir, str(page) + ".txt")
+        cmdOutput = subprocess.call([self.cmd, "-f", str(page), "-l", str(page), self.filepath, outputFileName])
+
+    def extractPages(self):
+        for page in range(1, self.pages+1):
+            self.extractPage(page)
+
+
+

runtest.sh

@@ -0,0 +1,2 @@
+python -m unittest tests.PdfInfoTest
+python -m unittest tests.PdfToTextTest

tests/PdfInfoTest.py

@@ -0,0 +1,17 @@
+#!/usr/local/bin/python
+
+import unittest
+import sys
+
+from PdfInfo import *
+
+class PdfInfoTest(unittest.TestCase):
+    def setUp(self):
+        pass
+
+    def testPdfPages(self):
+        pdfInfo = PdfInfo('tests/sample.pdf')
+        pdfInfo.process()
+        self.assertEqual(pdfInfo.getPages(), 5)
+        self.assertEqual(pdfInfo.getFileSizeInBytes(), 81691)
+

tests/PdfToTextTest.py

@@ -0,0 +1,22 @@
+#!/usr/local/bin/python
+
+import unittest
+import sys
+
+from PdfToText import *
+
+class PdfToTextTest(unittest.TestCase):
+    def setUp(self):
+        pass
+
+    def testPdfPages(self):
+        pdfToText = PdfToText('tests/sample.pdf', 5, "tests/out")
+        pdfToText.extractPage(1)
+
+    def testPdfAllPages(self):
+        pdfToText = PdfToText('tests/sample.pdf', 5, "tests/out")
+        pdfToText.extractPages()
+
+    def testPdfPageScanned(self):
+        pdfToText = PdfToText('tests/sample-scanned.pdf', 5, "tests/out")
+        pdfToText.extractPage(2)

tests/out/1.txt

@@ -0,0 +1,70 @@
+Table of Contents
+
+UNITED STATES
+SECURITIES AND EXCHANGE COMMISSION
+Washington, D.C. 20549
+
+FORM 10-Q
+(Mark One)
+
+x QUARTERLY REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE
+ACT OF 1934
+For the quarterly period ended September 30, 2013
+
+o TRANSITION REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE
+ACT OF 1934
+For the transition period from
+
+to
+
+Commission file number: 001-35167
+
+Kosmos Energy Ltd.
+(Exact name of registrant as specified in its charter)
+
+Bermuda
+
+98-0686001
+
+(State or other jurisdiction of
+incorporation or organization)
+
+(I.R.S. Employer
+Identification No.)
+
+Clarendon House
+2 Church Street
+Hamilton, Bermuda
+(Address of principal executive offices)
+
+(Zip Code)
+
+HM 11
+
+Registrant’s telephone number, including area code: +1 441 295 5950
+
+Not applicable
+(Former name, former address and former fiscal year, if changed since last report)
+Indicate by check mark whether the registrant: (1) has filed all reports required to be filed by Section 13 or 15(d) of the Securities Exchange Act of 1934
+during the preceding 12 months (or for such shorter period that the registrant was required to file such reports), and (2) has been subject to such filing
+requirements for the past 90 days. Yes x No o
+
+Indicate by check mark whether the registrant has submitted electronically and posted on its corporate Web site, if any, every Interactive Data File
+required to be submitted and posted pursuant to Rule 405 of Regulation S-T (§232.405 of this chapter) during the preceding 12 months (or for such shorter
+period that the registrant was required to submit and post such files). Yes x No o
+Indicate by check mark whether the registrant is a large accelerated filer, an accelerated filer, a non-accelerated filer, or a smaller reporting company. See
+the definitions of “large accelerated filer,” “accelerated filer” and “smaller reporting company” in Rule 12b-2 of the Exchange Act.
+Large accelerated filer x
+
+Accelerated filer o
+
+Non-accelerated filer o
+(Do not check if a smaller reporting company)
+
+Smaller reporting company o
+
+Indicate by check mark whether the registrant is a shell company (as defined in Rule 12b-2 of the Exchange Act). Yes o No x
+
+Indicate the number of shares outstanding of each of the issuer’s classes of common stock, as of the latest practicable date.
+
+

tests/out/2.txt

@@ -0,0 +1 @@
+

tests/out/3.txt

@@ -0,0 +1,63 @@
+Table of Contents
+
+TABLE OF CONTENTS
+
+Unless otherwise stated in this report, references to “Kosmos,” “we,” “us” or “the company” refer to Kosmos Energy Ltd. and its
+subsidiaries. We have provided definitions for some of the industry terms used in this report in the “Glossary and Selected Abbreviations” beginning
+on page 3.
+Page
+
+PART I. FINANCIAL INFORMATION
+Glossary and Select Abbreviations
+
+3
+
+Item 1. Financial Statements
+
+Consolidated Balance Sheets as of September 30, 2013 and December 31, 2012
+Consolidated Statements of Operations for the three and nine months ended September 30, 2013 and 2012
+Consolidated Statements of Comprehensive Loss for the three and nine months ended September 30, 2013 and 2012
+Consolidated Statements of Shareholders’ Equity for the nine months ended September 30, 2013
+Consolidated Statements of Cash Flows for the nine months ended September 30, 2013 and 2012
+Notes to Consolidated Financial Statements
+Item 2. Management’s Discussion and Analysis of Financial Condition and Results of Operations
+Item 3. Quantitative and Qualitative Disclosures about Market Risk
+Item 4. Controls and Procedures
+
+6
+7
+8
+9
+10
+
+11
+22
+31
+33
+
+PART II. OTHER INFORMATION
+Item 1. Legal Proceedings
+Item 1A. Risk Factors
+
+34
+34
+34
+34
+34
+34
+
+Item 2. Unregistered Sales of Equity Securities and Use of Proceeds
+Item 3. Defaults Upon Senior Securities
+Item 4. Mine Safety Disclosures
+Item 5. Other Information
+Item 6. Exhibits
+Signatures
+Index to Exhibits
+
+35
+36
+37
+
+2
+
+

tests/out/4.txt

@@ -0,0 +1,91 @@
+Table of Contents
+
+KOSMOS ENERGY LTD.
+GLOSSARY AND SELECTED ABBREVIATIONS
+The following are abbreviations and definitions of certain terms that may be used in this report. Unless listed below, all defined terms under Rule 410(a) of Regulation S-X shall have their statutorily prescribed meanings.
+
+“2D seismic data”
+
+Two-dimensional seismic data, serving as interpretive data that allows a view of a vertical cross-section beneath
+a prospective area.
+
+“3D seismic data”
+
+Three-dimensional seismic data, serving as geophysical data that depicts the subsurface strata in three
+dimensions. 3D seismic data typically provides a more detailed and accurate interpretation of the subsurface
+strata than 2D seismic data.
+
+“API”
+
+A specific gravity scale, expressed in degrees, that denotes the relative density of various petroleum liquids. The
+scale increases inversely with density. Thus lighter petroleum liquids will have a higher API than heavier ones.
+
+“ASC”
+
+Financial Accounting Standards Board Accounting Standards Codification.
+
+“ASU”
+
+Financial Accounting Standards Board Accounting Standards Update.
+
+“Barrel” or “Bbl”
+
+A standard measure of volume for petroleum corresponding to approximately 42 gallons at 60 degrees
+Fahrenheit.
+
+“BBbl”
+
+Billion barrels of oil.
+
+“BBoe”
+
+Billion barrels of oil equivalent.
+
+“Bcf”
+
+Billion cubic feet.
+
+“Boe”
+
+Barrels of oil equivalent. Volumes of natural gas converted to barrels of oil using a conversion factor of 6,000
+cubic feet of natural gas to one barrel of oil.
+
+“Boepd”
+
+Barrels of oil equivalent per day.
+
+“Bopd”
+
+Barrels of oil per day.
+
+“Bwpd”
+
+Barrels of water per day.
+
+“Debt cover ratio”
+
+The “debt cover ratio” is broadly defined, for each applicable calculation date, as the ratio of (x) total long-term
+debt less cash and cash equivalents and restricted cash, to (y) the aggregate EBITDAX (see below) of the
+Company for the previous twelve months.
+
+“Developed acreage”
+
+The number of acres that are allocated or assignable to productive wells or wells capable of production.
+
+“Development”
+
+The phase in which an oil or natural gas field is brought into production by drilling development wells and
+installing appropriate production systems.
+
+“Dry hole”
+
+A well that has not encountered a hydrocarbon bearing reservoir expected to produce in commercial quantities.
+
+“EBITDAX”
+
+Net income (loss) plus (1) exploration expense, (2) depletion, depreciation and amortization expense, (3) equitybased compensation expense, (4) (gain) loss on commodity derivatives, (5) (gain) loss on sale of oil and gas
+properties, (6) interest (income) expense, (7) income taxes, (8) loss on extinguishment of debt, (9) doubtful
+accounts expense, and (10) similar items.
+3
+
+