From 2e141b5cca2586536fe3f312c35eb9db117e5fb5 Mon Sep 17 00:00:00 2001
From: ajokar <ajokar91@yahoo.com>
Date: Tue, 17 Sep 2024 16:46:00 +0000
Subject: [PATCH] Upload files to "import_data"

---
 import_data/tokenizer.py | 59 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 59 insertions(+)
 create mode 100644 import_data/tokenizer.py

diff --git a/import_data/tokenizer.py b/import_data/tokenizer.py
new file mode 100644
index 0000000..8a6fc99
--- /dev/null
+++ b/import_data/tokenizer.py
@@ -0,0 +1,59 @@
+import re
+
+
+class Tokenizer():
+    def __init__(self):
+        pass
+
+    def tokenize_words(self, doc_string):
+        token_list = doc_string.strip().split()
+        #token_list = [x.strip("\u200c").strip('\u200f') for x in token_list if len(x.strip("\u200c")) != 0]
+        new_token_list = []
+        new_token = ''
+        for index,token in enumerate(token_list):
+            if len(token.strip("\u200c")) != 0:
+                new_token = token.strip("\u200c")
+            if len(token.strip("\u200f")) != 0:
+                new_token = new_token.strip('\u200f')
+            new_token_list.append(new_token)
+        return new_token_list
+
+    def tokenize_sentences(self, doc_string):
+        #finding the numbers
+        pattern = r"[-+]?\d*\.\d+|\d+"
+        nums_list = re.findall(pattern, doc_string)
+        doc_string = re.sub(pattern, 'floatingpointnumber', doc_string)
+
+        pattern = r'([!\.\?؟]+)[\n]*'
+        tmp = re.findall(pattern, doc_string)
+        doc_string = re.sub(pattern, self.add_tab, doc_string)
+
+        pattern = r':\n'
+        tmp = re.findall(pattern, doc_string)
+        doc_string = re.sub(pattern, self.add_tab, doc_string)
+
+        pattern = r';\n'
+        tmp = re.findall(pattern, doc_string)
+        doc_string = re.sub(pattern, self.add_tab, doc_string)
+
+        pattern = r'؛\n'
+        tmp = re.findall(pattern, doc_string)
+        doc_string = re.sub(pattern, self.add_tab, doc_string)
+
+        pattern = r'[\n]+'
+        doc_string = re.sub(pattern, self.add_tab, doc_string)
+
+        for number in nums_list:
+            pattern = 'floatingpointnumber'
+            doc_string = re.sub(pattern, number, doc_string, 1)
+
+        doc_string = doc_string.split('\t\t')
+        doc_string = [x for x in doc_string if len(x) > 0]
+        return doc_string
+
+    def add_tab(self, mystring):
+        mystring = mystring.group()  # this method return the string matched by re
+        mystring = mystring.strip(' ')  # ommiting the whitespace around the pucntuation
+        mystring = mystring.strip('\n') # ommiting the newline around the pucntuation
+        mystring = " " + mystring + "\t\t"  # adding a space after and before punctuation
+        return mystring
\ No newline at end of file