Move stuff from main.py to README.md and utils.py

README.md

@@ -0,0 +1,19 @@
+# BNP PDF statement parser
+
+Depends on `pdftotext`.
+
+Assuming file hierarchy like:
+
+```
+.
+├── compte_de_cheques/
+│   ├── 2022/
+│   │   ├── 20221121.pdf
+│   │   └── 20221221.pdf
+│   └── 2023/
+│       ├── 20230123.pdf
+│       └── 20230221.pdf
+livret_a/
+├── 20230721.pdf
+└── 20240122.pdf
+```

bnp_pdf_statement_parser.py

@@ -1,40 +1,15 @@
 #!/usr/bin/env python
 
-# Depends on `pdftotext`.
-
 import os
-import subprocess
-import re
 import matplotlib.pyplot as plt
 import matplotlib.ticker as ticker
 from datetime import datetime
+from utils import getTextFromPdf, getDatetimeFromFileName, getMonthIndexSinceEpoch, getMonthNameFromMonthIndex, FIRST_LINE_OF_PAYMENT_REGEX, END_PAGE_AFTER_THE_FIRST_ONE_REGEX, SOLDE_CREDITEUR_AU_REGEX, TOTAL_DES_OPERATIONS_REGEX
 
 PATH = f'/home/benjamin/Desktop/bens_folder/bazaar/documents/bnp/bank_statements/compte_de_cheques/'
 
 os.chdir(PATH)
 
-'''
-Assuming file hierarchy like:
-
-2022
-├── 20221121.pdf
-└── 20221221.pdf
-2023
-├── 20230123.pdf
-└── 20230221.pdf
-'''
-
-def execute(command):
-    return subprocess.check_output(command).decode('utf-8')
-
-def getTextFromPdf(pdfPath):
-    return execute(['pdftotext', '-raw', pdfPath, '-'])
-
-FIRST_LINE_OF_PAYMENT_REGEX = re.compile('\\d{2}\\.\\d{2} \\d{2}\\.\\d{2} \\d+,\\d{2}')
-END_PAGE_AFTER_THE_FIRST_ONE_REGEX = re.compile('P\\. \\d+/\\d+')
-SOLDE_CREDITEUR_AU_REGEX = re.compile('SOLDE CREDITEUR AU \\d{2}\\.\\d{2}\\.\\d{4}')
-TOTAL_DES_OPERATIONS_REGEX = re.compile('TOTAL\\ DES\\ OPERATIONS\\ ([0-9 ]+,\\d{2})\\ ([0-9 ]+,\\d{2})')
-
 PRINT_TRANSACTIONS = False
 
 totalMonthlyDebits = []
@@ -47,7 +22,7 @@ for folder in sorted(os.listdir()):
     for file in sorted(os.listdir(folder)):
         filePath = f'{folder}/{file}'
         print(filePath)
-        currentDatetime = getDatetime(file)
+        currentDatetime = getDatetimeFromFileName(file)
         if firstDatetime is None:
             firstDatetime = currentDatetime
         content = getTextFromPdf(filePath)
@@ -100,7 +75,7 @@ for folder in sorted(os.listdir()):
                     comment += [line]
         #break
     #break
-lastDatetime = getDatetime(file)
+lastDatetime = getDatetimeFromFileName(file)
 
 fig, ax = plt.subplots()
 plt.title('Monthly debits and credits')
@@ -108,14 +83,7 @@ plt.xlabel('Date')
 plt.ylabel('€')
 ALPHA = 0.5
 
-def getDatetime(aDatetimeStr):
+xTicks = range(getMonthIndexSinceEpoch(firstDatetime), getMonthIndexSinceEpoch(lastDatetime) + 1)
-    return datetime.strptime(aDatetimeStr, '%Y%m%d.pdf')
-
-def getMonthIndex(aDatetime):
-    return aDatetime.year * 12 + aDatetime.month
-
-xTicks = range(getMonthIndex(firstDatetime), getMonthIndex(lastDatetime) + 1)
-# sign does not seem respected for `totalMonthlyDifferences`.
 totalMonthlyAmountAndLabel = (
     #(totalMonthlyDebits, 'Debit'),
     #(totalMonthlyCredits, 'Credit'),
@@ -129,10 +97,7 @@ plt.legend()
 #plt.yscale('symlog')
 ax.yaxis.set_major_formatter(ticker.StrMethodFormatter('{x:,}'))
 
-def getMonthName(monthIndex):
+ticksLabels = [getMonthNameFromMonthIndex(monthIndex) for monthIndex in xTicks]
-    return datetime((monthIndex - 1) // 12, 1 + (monthIndex - 1) % 12, 1).strftime('%b %Y')
-
-ticksLabels = [getMonthName(monthIndex) for monthIndex in xTicks]
 plt.xticks(xTicks, ticksLabels, rotation = 90)
 #plt.tight_layout()
 # How to show the horizontal lines for subticks?

utils.py

@@ -0,0 +1,23 @@
+import subprocess
+from datetime import datetime
+import re
+
+FIRST_LINE_OF_PAYMENT_REGEX = re.compile('\\d{2}\\.\\d{2} \\d{2}\\.\\d{2} \\d+,\\d{2}')
+END_PAGE_AFTER_THE_FIRST_ONE_REGEX = re.compile('P\\. \\d+/\\d+')
+SOLDE_CREDITEUR_AU_REGEX = re.compile('SOLDE CREDITEUR AU \\d{2}\\.\\d{2}\\.\\d{4}')
+TOTAL_DES_OPERATIONS_REGEX = re.compile('TOTAL\\ DES\\ OPERATIONS\\ ([0-9 ]+,\\d{2})\\ ([0-9 ]+,\\d{2})')
+
+def execute(command):
+    return subprocess.check_output(command).decode('utf-8')
+
+def getTextFromPdf(pdfPath):
+    return execute(['pdftotext', '-raw', pdfPath, '-'])
+
+def getDatetimeFromFileName(aDatetimeStr):
+    return datetime.strptime(aDatetimeStr, '%Y%m%d.pdf')
+
+def getMonthIndexSinceEpoch(aDatetime):
+    return aDatetime.year * 12 + aDatetime.month
+
+def getMonthNameFromMonthIndex(monthIndex):
+    return datetime((monthIndex - 1) // 12, 1 + (monthIndex - 1) % 12, 1).strftime('%b %Y')