Ricompilato versione 0.4b

.gitignore

@@ -1,3 +1,4 @@
 /.DS_Store
 /.project
 /.pydevproject
+bananaSPLIT/build

bananaSPLIT/bananaSPLIT.exe.spec

@@ -0,0 +1,32 @@
+# -*- mode: python -*-
+
+block_cipher = None
+
+
+a = Analysis(['main.py'],
+             pathex=['D:\\Emanuele\\Documenti\\workspace\\bananaSPLIT\\bananaSPLIT'],
+             binaries=[],
+             datas=[('C:\\Users\\Emanuele Trabattoni\\AppData\\Roaming\\Python\\Python37\\site-packages\\text_unidecode\\data.bin', '.\\text_unidecode')],
+             hiddenimports=[],
+             hookspath=[],
+             runtime_hooks=[],
+             excludes=[],
+             win_no_prefer_redirects=False,
+             win_private_assemblies=False,
+             cipher=block_cipher,
+             noarchive=False)
+pyz = PYZ(a.pure, a.zipped_data,
+             cipher=block_cipher)
+exe = EXE(pyz,
+          a.scripts,
+          a.binaries,
+          a.zipfiles,
+          a.datas,
+          [],
+          name='bananaSPLIT.exe',
+          debug=False,
+          bootloader_ignore_signals=False,
+          strip=False,
+          upx=True,
+          runtime_tmpdir=None,
+          console=True , icon='banana.ico')

bananaSPLIT/dist/bananaconfITA.json

@@ -37,8 +37,7 @@
 			"DOCUMENT-TYPE:",
 			"CHARTS:"
 		]
-	},
+	}, 
-	"settings": 
 	"settings": {
 		"encoding": "utf-8",
 		"monthPosition": 0,

bananaSPLIT/main.py

@@ -15,26 +15,24 @@ from pprint import pprint
 from glob import glob
 from copy import deepcopy
 from slugify import slugify
-
-     
 ####### VAR GLOBALI #######
 
 ####### FUNZIONI GLOBALI #######        
 def printTitle():
     print(".  .   .                 ,     \n|  | _ | _. _ ._ _  _   -+- _  *    \n|/\|(/,|(_.(_)[ | )(/,   | (_) *")
 
-    print(Fore.LIGHTYELLOW_EX,\
+    print(Fore.LIGHTYELLOW_EX,'    _                                   ___________ _     _____ _____  \n\
-    '    _                                   ___________ _     _____ _____     \n\
     | |                                 /  ___| ___ \ |   |_   _|_   _|        \n\
     | |__   __ _ _ __   __ _ _ __   __ _\ `--.| |_/ / |     | |   | |          \n\
     | \'_ \ / _` | \'_ \ / _` | \'_ \ / _` |`--. \  __/| |     | |   | |       \n\
     | |_) | (_| | | | | (_| | | | | (_| /\__/ / |   | |_____| |_  | |          \n\
     |_.__/ \__,_|_| |_|\__,_|_| |_|\__,_\____/\_|   \_____/\___/  \_/')
     
-    print ('\t\t\t\t\t\t\tVersione 0.4a\n\n', Style.RESET_ALL)
+    print ('\t\t\t\t\t\t\tVersione 0.4b\n\n', Style.RESET_ALL)
     print('Iniziamo!!')
     pass
     
+    
 #############################################################
 ####################### MAIN ################################
 #############################################################
@@ -111,14 +109,18 @@ try:
         else:
             print('Non ho capito la risposta, sii un po\' piu\' specifico... [y/n]\n')
             pass
-    if settings['removeOldFiles']:
-            print("Rimuovo i vecchi file dalla cartella di destinazione..")
-            for x in glob(cfg['OUTworkPath']+'*.txt'):
-                os.remove(x)
 except:
     print(Fore.LIGHTRED_EX, 'OOPS! Qualcosa e\' andato storto, non riesco a trovare nessun file da leggere :(', Style.RESET_ALL)
     input()
     sys.exit()
+
+try:
+    if settings['removeOldFiles']:
+        print("Rimuovo i vecchi file dalla cartella di destinazione..")
+        for x in glob(cfg['OUTworkPath']+'*.txt'):
+            os.remove(x)
+except:
+    print(Fore.LIGHTRED_EX, 'OOPS! Qualcosa e\' andato storto, non riesco a rimuovere i vecchi file :(', Style.RESET_ALL)
     
 lastTime=time.time()
 fileCounter = 1

bananaconfEN.json

@@ -0,0 +1,64 @@
+{
+	"INworkPath": "D:\\Test\\",
+	"OUTworkPath": "D:\\Test\\Separati\\",
+	"OUTnameFormat":"TEST_{docnum}_{year:04d}{month:02d}{day:02d}_{title}.txt",
+	"docStruct": {
+		"docSep": "\\s*Copyright [(0-9)]+",
+		"dateFormat":"{month} {day:d}, {year:d}{}",
+		"dateWords": [
+			"January",
+			"February",
+			"March",
+			"April",
+			"May",
+			"June",
+			"July",
+			"August",
+			"September",
+			"October",
+			"November",
+			"December"
+		],
+		"headWords": [
+			"BYLINE:",
+			"SECTION:",
+			"LENGTH:",
+			"DATELINE:",
+			"HIGHLIGHT:",
+			"Email:"
+		],
+		"tailWords": [
+			"Newstex ID" ,
+			"NOTES",
+			"LANGUAGE:",
+			"GRAPHIC:",
+			"TYPE:",
+			"URL:",
+			"LOAD-DATE:",
+			"PUBLICATION-TYPE:",
+			"DOCUMENT-TYPE:",
+			"CHARTS:",
+			"JOURNAL-CODE:"
+		]
+	},
+	"settings": {
+		"encoding": "utf-8",
+		"monthPosition": 0,
+		"getNewsPaperName": true,
+		"nameNotFoundStr" : "--ND--",
+		"includeTitle" : true,
+		"removeDuplicates": true,
+		"showSkipped": false,
+		"showRemovedDuplicates": true,
+		"maxTitleLen": 32,
+		"loadTXT": true,
+		"loadDOCX": false,
+		"removeOldFiles":true,
+		"saveSeparateFiles": true,
+		"saveBodyFile": true,
+		"saveBodyNumber":true,
+		"delLF": false,
+		"delWordBreak": true,
+		"delChars": "'|@|#"
+	}
+}

bananaconfITA.json

@@ -0,0 +1,61 @@
+{
+	"INworkPath": "C:\\Test\\",
+	"OUTworkPath": "C:\\Test\\Separati\\",
+	"OUTnameFormat":"TEST_{docnum}_{year:04d}{month:02d}{day:02d}_{papername}_{title}.txt",
+	"docStruct": {
+		"docSep": "Copyright [(0-9)]+",
+		"dateFormat":"{day:d} {month} {year:d} {}",
+		"dateWords": [
+			"Gennaio",
+			"Febbraio",
+			"Marzo",
+			"Aprile",
+			"Maggio",
+			"Giugno",
+			"Luglio",
+			"Agosto",
+			"Settembre",
+			"Ottobre",
+			"Novembre",
+			"Dicembre"
+		],
+		"headWords": [
+			"BYLINE:",
+			"SECTION:",
+			"LENGTH:",
+			"DATELINE:",
+			"HIGHLIGHT:",
+			"Email:"
+		],
+		"tailWords": [
+			"LANGUAGE:",
+			"GRAPHIC:",
+			"TYPE:",
+			"URL:",
+			"LOAD-DATE:",
+			"PUBLICATION-TYPE:",
+			"DOCUMENT-TYPE:",
+			"CHARTS:"
+		]
+	}, 
+	"settings": {
+		"encoding": "utf-8",
+		"monthPosition": 0,
+		"getNewsPaperName": true,
+		"nameNotFoundStr" : "--ND--",
+		"includeTitle" : true,
+		"removeDuplicates": true,
+		"showSkipped": false,
+		"showRemovedDuplicates": true,
+		"maxTitleLen": 32,
+		"loadTXT": true,
+		"loadDOCX": false,
+		"removeOldFiles":true,
+		"saveSeparateFiles": true,
+		"saveBodyFile": true,
+		"saveBodyNumber":true,
+		"delLF": false,
+		"delWordBreak": true,
+		"delChars": "'|@|#"
+	}
+}