Added tense correction file for errors coming from pyinflect library database

.gitignore

@@ -1 +1 @@
-__pycache__
+__pycache__

app.py

@@ -23,7 +23,7 @@ with iface:
         output = gr.Markdown()
 
     btn_get.click(
-        fn = execute.get_verbs,
+        fn = execute.get,
         inputs = input_verb,
         outputs = [error, output],
         api_name="get"

data/overrides.csv

@@ -0,0 +1,12 @@
+VB,VBD,VBN
+work,worked,worked
+can,could,-
+wake,woke,woken
+cost,cost,cost
+drink,drank,drunk
+learn,learnt,learnt
+leave,left,left
+pay,paid,paid
+ring,rang,rung
+speak,spoke,spoken
+write,wrote,written

modules/m_parser.py

@@ -1,14 +1,25 @@
 import subprocess
 import spacy
 import pyinflect
-from typing import Tuple, Union
+import pandas as pd
+from typing import Tuple, Union, Any
 
 class Parser:
     def __init__(
         self
     ) -> None:
         
-        self.parser = self.__init_parser("en_core_web_md")
+        self.parser = self.__init_parser("en_core_web_sm")
+        self.__overrides = self.__load_overrides("data/overrides.csv")
+    
+    def __load_overrides(
+        self,
+        file_path: str
+    ) -> Any:
+
+        dic = pd.read_csv(file_path).set_index('VB').to_dict()
+        fun = lambda verb, tag: dic[tag].get(verb, False)
+        return fun
 
     def __init_parser(
         self,
@@ -30,63 +41,86 @@ class Parser:
         
         return parser
     
-    def __v2participle(
+    def __format_error(
         self,
-        tk_verb: spacy.tokens.token.Token
-    ) -> Union[str, None]:
+        error: str
+    ) -> str:
 
-        return tk_verb._.inflect('VBN')
-    
-    def __v2past(
+        template = """
+            <center>
+                <div class="alert alert-warning" role="alert">
+                    <h6><b>{}</b></h6>
+                </div>
+            </center>
+        """
+        return template.format(error)
+
+    def __format_output(
         self,
-        tk_verb: spacy.tokens.token.Token
-    ) -> Union[str, None]:
+        infinitive: str,
+        past: str,
+        participle: str,
+    ) -> str:
 
-        return tk_verb._.inflect('VBD')
+        template = """
+            |Infinitive| Simple Past | Past Participle |
+            | :----: | :----: | :----: |
+            |{} | {}| {}|
+        """
+        return template.format(infinitive, past, participle)
 
-    def __v2infinitive(
+    def is_in_overrides(
         self,
-        tk_verb: spacy.tokens.token.Token
+        verb: str,
+        tense: str
+    ) -> Tuple[str, bool]:
+
+        res = self.__overrides(verb, tense)
+        if isinstance(res, bool):
+            return res, ""
+
+        return True, res
+
+    def __get_inflections(
+        self, 
+        infinitive: spacy.tokens.token.Token,
+        tense: str
     ) -> Union[str, None]:
+        
+        # Check if verb is in overrides file
+        res, inflections = self.is_in_overrides(infinitive.text, tense)
 
-        return tk_verb._.inflect('VB')
+        if not res:
+            form1 = infinitive._.inflect(tense, form_num=0)
+            form2 = infinitive._.inflect(tense, form_num=1)
+            inflections = list(set([form1, form2]))
 
-    def __tokenizer(
-        self,
-        verb: str
-    ) -> spacy.tokens.token.Token:
+            if len(inflections) == 1 and inflections[0] == None:
+                return None
+            
+            inflections = '/'.join(inflections)
 
-        return self.parser(verb)[0]
+        return inflections
 
-    def ___format_output(
-        self,
-        infiniteve: str,
-        past: str,
-        participle: str
-    ) -> str:
-        template = """
-            |infinitive| Simple Past | Past Participle |
-            | :----: | :----: | :----: |
-            |{} | {}| {}|
-        """
-        return template.format(infiniteve, past, participle)
+    def __getAllTenses(
+        self, 
+        tk_verb: spacy.tokens.token.Token
+    ) -> Tuple[str,str,str,str]:
+
+        infinitive  = tk_verb._.inflect('VB')
+        past        = self.__get_inflections(tk_verb, 'VBD')
+        participle  = self.__get_inflections(tk_verb, 'VBN')
 
+        return infinitive, past, participle
 
-    def ___format_error(
+    def __tokenizer(
         self,
-        error: str
-    ) -> str:
+        verb: str
+    ) -> spacy.tokens.token.Token:
 
-        template = """
-            <center>
-                <div class="alert alert-warning" role="alert">
-                    <h6><b>{}</b></h6>
-                </div>
-            </center>
-        """
-        return template.format(error)
+        return self.parser(verb)[0]
 
-    def get_verbs(
+    def get(
         self,
         verb: str
     ) -> Tuple[str,str]:
@@ -95,22 +129,20 @@ class Parser:
         error, output = "", ""
 
         if verb == "":
-            error = self.___format_error(
+            error = self.__format_error(
                 f"Error: The Verb field can not be empty!"
             )
             return error, output
 
         tk_verb = self.__tokenizer(verb)
-        infinitive = self.__v2infinitive(tk_verb)
-        past = self.__v2past(tk_verb)
-        participle = self.__v2participle(tk_verb)
+        infinitive, past, participle = self.__getAllTenses(tk_verb)
 
         if infinitive is None or past is None or participle is None:
-            error = self.___format_error(
+            error = self.__format_error(
                 f"Error: The verb '<b>{verb}</b>' has not been found or not spelled correctly!"
             )
             return error, output
         
-        output = self.___format_output(infinitive, past, participle)
+        output = self.__format_output(infinitive, past, participle)
 
         return error, output