improving pron id

wikstraktor.py

@@ -5,6 +5,7 @@ import importlib
 import json
 from wikstraktor_version import version as the_version
 from wikstraklog import Wikstraklog
+from copy import deepcopy as dc
 import re
 
 def get_list_string_level(wikitext):
@@ -37,12 +38,17 @@ class SubInfo:
 			self.label = f"{prefix}_{self.__class__.prfx}{self.id}"
 		return self.label
 
-	def replace_src_in_id(self, former_src, new_src):
+	def replace_src_in_id(self, former_src, new_src, copy=False):
 		##Attention si on nettoie en mettant des sources partout, il faudra changer
 		res = None
 		if self.label != None and former_src != None and new_src != None :
-			self.label = re.sub(r'^([\w\.]+)-('+str(former_src)+')',r"\1-"+str(new_src), self.label)
-			res = self.label
+			if not copy:
+				self.label = re.sub(r'^([\w\.]+)-('+str(former_src)+')',r"\1-"+str(new_src), self.label)
+				res = self.label
+			else:
+				new_obj = dc(self)
+				new_obj.label = re.sub(r'^([\w\.]+)-('+str(former_src)+')',r"\1-"+str(new_src), self.label)
+				res = new_obj
 		return res
 
 	def get_src_from_id(self):
@@ -242,7 +248,7 @@ class Sense(SubInfo):
 		return key in self.metadata.keys()
 
 	def get_id(self):
-		return self.id
+		return self.label
 
 	def set_domain(self, d):
 		self.domain = d
@@ -272,13 +278,13 @@ class Sense(SubInfo):
 			self.translations.append(theTranslation)
 
 	def add_subsense(self, subsense):
-		if self.id!=None:
+		if self.label!=None:
 			subsense.set_id(self.set_id())
 		if subsense not in self.subsenses:
 			self.subsenses.append(subsense)
 
 	def __eq__(self, other):
-		res = isinstance(other, self.__class__) and self.id == other.id and len(self.definitions) == len(other.definitions) and len(self.examples) == len(other.examples) and len(self.translations) == len(other.translations) and self.domain == other.domain and len(other.metadata) == len(self.metadata) and other.regions == self.regions
+		res = isinstance(other, self.__class__) and len(self.definitions) == len(other.definitions) and len(self.examples) == len(other.examples) and len(self.translations) == len(other.translations) and self.domain == other.domain and len(other.metadata) == len(self.metadata) and other.regions == self.regions #and self.id == other.id
 		i = 0
 		while res and i < len(self.examples):
 			res = self.examples[i] in other.examples
@@ -334,18 +340,17 @@ class Sense(SubInfo):
 		if len(self.subsenses) > 0:
 			res["Subsenses"] = {}
 			for t in self.subsenses:
-				res["Subsenses"][t.set_id(self.id)]= t.serializable(prefix)
+				res["Subsenses"][t.set_id(self.label)]= t.serializable(prefix)
 		return res
 
 	def __str__(self):
 		return json.dumps(self.serializable())
 
 class SubSense(Sense):
-	def set_id(self, prefix=None):
-		if prefix != None and self.id == None:
-			self.id = f"{prefix}.{self.__class__.next_id}"  #l'identifiant du sens
-			self.__class__.inc_n_id()
-		return self.id
+	def set_id(self, prefix, force = False):
+		if (self.label == None or force) and prefix != None:
+			self.label = f"{prefix}.{self.id}"
+		return self.label
 
 class Entry:
 	#version_id : l'identifiant unique de la vesion de la page du wiktionnaire (pywikibot.Page.latest_revision_id)
@@ -433,10 +438,10 @@ class Entry:
 				i += 1
 			for p in other.pronunciations:
 				src = p.get_src_from_id()
-				if src != None and src <= max_id and src_map[src] != src:
+				if src != None and src <= max_id and src < len(src_map) and src_map[src] != src:
 					#max_id, c'est parce qu'un même objet peut être
 					#à plusieurs endroits et avoir déjà été modifié
-					p.replace_src_in_id(src, src_map[src])
+					p=p.replace_src_in_id(src, src_map[src], True)
 				self.add_pronunciation(p)
 			for s in other.senses:
 				src = s.get_src_from_id()
@@ -509,6 +514,10 @@ class ParserContext:
 		self.wikstraktor_version = version_id
 		self.context = []
 		self.entries = []
+		#reset counters
+		Sense.reset()
+		Sense.reset_sub_counters()
+		Pronunciation.reset()
 
 	def get_level(self):
 		if len(self.context) == 0: