Added first version of console autocompletion

src/core/sheerka/Sheerka.py

@@ -19,6 +19,7 @@ from sdp.sheerkaDataProvider import SheerkaDataProvider, Event
 
 BASE_NODE_PARSER_CLASS = "parsers.BaseNodeParser.BaseNodeParser"
 CONCEPTS_FILE = "_concepts.txt"
+EXIT_COMMANDS = ("quit", "exit", "bye")
 
 
 class Sheerka(Concept):
@@ -893,4 +894,4 @@ class Sheerka(Concept):
         logging.basicConfig(format=log_format, level=log_level, handlers=[console_handler])
         logging.addLevelName(logging.ERROR, "\033[1;41m%s\033[1;0m" % logging.getLevelName(logging.ERROR))
         # uncomment the following line to enable colors
-        #logging.StreamHandler.emit = add_coloring_to_emit_ansi(logging.StreamHandler.emit)
+        # logging.StreamHandler.emit = add_coloring_to_emit_ansi(logging.StreamHandler.emit)

src/core/sheerka/services/SheerkaAdmin.py

@@ -0,0 +1,31 @@
+from core.builtin_concepts import BuiltinConcepts
+from core.sheerka.services.sheerka_service import BaseService
+
+
+class SheerkaAdmin(BaseService):
+    NAME = "Admin"
+
+    def __init__(self, sheerka):
+        super().__init__(sheerka)
+
+    def initialize(self):
+        self.sheerka.bind_service_method(self.caches_names)
+        self.sheerka.bind_service_method(self.cache)
+
+    def caches_names(self):
+        """
+        Returns the name of all the caches
+        :return:
+        """
+        return list(self.sheerka.cache_manager.caches.keys())
+
+    def cache(self, name):
+        """
+        Returns the content of a cache
+        :param name:
+        :return:
+        """
+        if name not in self.sheerka.cache_manager.caches:
+            return self.sheerka.new(BuiltinConcepts.NOT_FOUND, body={"cache": name})
+
+        return self.sheerka.cache_manager.caches[name].cache.copy()

src/core/sheerka/services/SheerkaComparisonManager.py

@@ -130,6 +130,12 @@ class SheerkaComparisonManager(BaseService):
         return self._inner_add_comparison(comparison_obj)
 
     def get_partition(self, prop_name, comparison_context="#"):
+        """
+        Returns the equivalent classes for the property, using the comparison_context
+        :param prop_name:
+        :param comparison_context:
+        :return:
+        """
         weighted_concept = self.get_concepts_weights(prop_name, comparison_context)
 
         return self._get_partition(weighted_concept)

src/core/sheerka/services/SheerkaExecute.py

@@ -306,9 +306,17 @@ class SheerkaExecute(BaseService):
 
                                         result = evaluator.eval(sub_context, item)
                                         if result is None:
+                                            # match() was successful but nothing was done in eval
+                                            # most of the time, it's because checks made in eval were unsuccessful
                                             debug_result.append({"input": item, "return_value": None})
                                             continue
 
+                                        if id(result) == id(item):
+                                            # eval was successful, but we don't want to alter the processing flow
+                                            debug_result.append({"input": item, "return_value": item})
+                                            continue
+
+                                        # otherwise, item will be removed and replaced by result
                                         to_delete.append(item)
                                         if isinstance(result, list):
                                             evaluated_items.extend(result)
@@ -319,6 +327,8 @@ class SheerkaExecute(BaseService):
                                                                      evaluator=evaluator)
                                             result = self.sheerka.ret("sheerka.process", False, error, parents=[item])
                                             evaluated_items.append(result)
+
+                                        # TODO: Add a validation to make sure that item is somewhere in return_value.parents
                                         debug_result.append({"input": item, "return_value": result})
                                     else:
                                         debug_result.append({"input": item, "return_value": NO_MATCH})

src/core/sheerka/services/SheerkaFunctionsParametersHistory.py

@@ -0,0 +1,78 @@
+from dataclasses import dataclass
+from operator import itemgetter
+from typing import Tuple, Dict, List
+
+from cache.Cache import Cache
+from core.sheerka.services.sheerka_service import BaseService, ServiceObj
+
+
+@dataclass
+class FunctionParametersObj(ServiceObj):
+    name: str
+    params: Dict[int, List[Tuple[str, int]]]
+    #
+    # params = {
+    #    1 : [('value 1', 1), ('value 2', 5)],
+    #    3 : [('value 3', 5)]
+    #
+    # the key is the number of the parameter
+    # the values are a tuple of (value seen, number of time this value is seen)
+
+
+class SheerkaFunctionsParametersHistory(BaseService):
+    """
+    This service remembers the parameters expected by functions
+    """
+
+    NAME = "FunctionsParametersHistory"
+    FUNCTIONS_PARAMETERS_ENTRY = "FunctionsParametersHistory:Functions"
+
+    def __init__(self, sheerka):
+        super().__init__(sheerka)
+        self.cache = Cache(max_size=1024, default=lambda k: self.sheerka.sdp.get(self.FUNCTIONS_PARAMETERS_ENTRY, k))
+
+    def initialize(self):
+        self.sheerka.cache_manager.register_cache(self.FUNCTIONS_PARAMETERS_ENTRY, self.cache, True, True)
+        return self
+
+    def record_function_parameter(self, context, func_name: str, param_number: int, param_value: str):
+        """
+        Declare a new value to the parameter and function
+        :param context:
+        :param func_name:
+        :param param_number:
+        :param param_value:
+        :return:
+        """
+        old = self.cache.get(func_name)
+        if old is not None:
+            if param_number in old.params:
+                lst = old.params[param_number]
+                for i, value in enumerate(lst):  # value is a tuple (param_value, counter)
+                    if value[0] == param_value:
+                        lst[i] = (param_value, value[1] + 1)
+                        break
+                else:
+                    lst.append((param_value, 1))
+            else:
+                old.params[param_number] = [(param_value, 1)]
+            self.cache.put(func_name, old)
+        else:
+            obj = FunctionParametersObj(context.event.get_digest(), func_name, {param_number: [(param_value, 1)]})
+            self.cache.put(func_name, obj)
+
+    def get_function_parameters(self, func_name: str, param_number: int):
+        """
+        Return the list of param
+        :param func_name:
+        :param param_number:
+        :return:
+        """
+        values = self.cache.get(func_name)
+        if values is None:
+            return []
+
+        if param_number not in values.params:
+            return []
+
+        return [item[0] for item in sorted(values.params[param_number], key=itemgetter(1), reverse=True)]

src/core/sheerka/services/sheerka_service.py

@@ -3,7 +3,7 @@ from dataclasses import dataclass
 
 @dataclass
 class ServiceObj:
-    event_id: str  # event where the object is created / modified
+    event_id: str  # digest of the event where the object is created / modified
 
 
 class BaseService:

src/core/tokenizer.py

@@ -49,6 +49,7 @@ class TokenKind(Enum):
     WORD = "word"
     EQUALSEQUALS = "=="
     VAR_DEF = "__var__"
+    REGEX = "r'xxx' or r\"xxx\" or r:xxx: or r|xxx| or r/xxx/"
 
 
 @dataclass()
@@ -322,6 +323,12 @@ class Tokenizer:
                 yield Token(TokenKind.CONCEPT, (name, id), self.i, self.line, self.column)
                 self.i += length + 2
                 self.column += length + 2
+            elif c == "r" and self.i + 1 < self.text_len and self.text[self.i + 1] in "'\":|/":
+                string, newlines, column_index = self.eat_string(self.i + 1, self.line, self.column)
+                yield Token(TokenKind.REGEX, string, self.i, self.line, self.column)  # quotes are kept
+                self.i += len(string) + 1
+                self.column = column_index  # 1 if newlines > 0 else self.column + len(string)
+                self.line += newlines
             elif self.parse_word and (c.isalpha() or c.isdigit()):
                 word = self.eat_word(self.i)
                 yield Token(TokenKind.WORD, word, self.i, self.line, self.column)
@@ -340,10 +347,10 @@ class Tokenizer:
                 self.i += len(number)
                 self.column += len(number)
             elif c == "'" or c == '"':
-                string, newlines = self.eat_string(self.i, self.line, self.column)
+                string, newlines, column_index = self.eat_string(self.i, self.line, self.column)
                 yield Token(TokenKind.STRING, string, self.i, self.line, self.column)  # quotes are kept
                 self.i += len(string)
-                self.column = 1 if newlines > 0 else self.column + len(string)
+                self.column = column_index  # 1 if newlines > 0 else self.column + len(string)
                 self.line += newlines
             elif c == "_":
                 yield Token(TokenKind.UNDERSCORE, "_", self.i, self.line, self.column)
@@ -445,21 +452,20 @@ class Tokenizer:
         quote = self.text[start_index]
         result = self.text[start_index]
         lines_count = 0
+        column_index = start_column + 1
 
         i = start_index + 1
         escape = False
-        newline = None
+        #newline = None
         while i < self.text_len:
             c = self.text[i]
             result += c
             i += 1
+            column_index += 1
 
-            if newline:
+            if c == "\n":
                 lines_count += 1
-                newline = c if c == newline else None
-            else:
-                if c == "\r" or c == "\n":
-                    newline = c
+                column_index = 1
 
             if c == "\\":
                 escape = True
@@ -468,15 +474,16 @@ class Tokenizer:
             else:
                 escape = False
 
-        # add trailing new line if needed
-        if newline:
-            lines_count += 1
+        # # add trailing new line if needed
+        # if newline:
+        #     lines_count += 1
+        #     column_index = 1
 
         if result[-1] != quote:
             raise LexerError("Missing Trailing quote", result, i, start_line + lines_count,
                              1 if lines_count > 0 else start_column + len(result))
 
-        return result, lines_count
+        return result, lines_count, column_index
 
     def eat_word(self, start):
         result = self.text[start]