diff --git a/parsers/en_constants.py b/parsers/en_constants.py
index c933d6ab9b796b86e53f6b4ad1c9b9e96ea8eeb7..42c73cbf041d0f7da4efc20101badfac995a2301 100644
--- a/parsers/en_constants.py
+++ b/parsers/en_constants.py
@@ -8,6 +8,9 @@ string_values = {
"t_acc":"a", #template for accents
"t_deflabel":"lb",
"t_ex":["ux", "usex"],
+ "sense_pattern":[ ## structure(s) for sense patterns add_subdef is to be added to def patterns
+ {"def":"\\#", "ex":"\\#:", "add_subdef":"\\#"}
+ ],
"POS": { #https://en.wiktionary.org/wiki/Wiktionary:POS
"Adjective":"Adj",
"Adverb":"Adv",
diff --git a/parsers/en_en.py b/parsers/en_en.py
index d840524fe795ffeb2d0dbfa26866b46ad7d931d7..1a2fab1bad8094859a53f166b88e337f9d18e9a1 100644
--- a/parsers/en_en.py
+++ b/parsers/en_en.py
@@ -48,54 +48,6 @@ class En_en_straktor(Wikstraktor):
pos = self.constants['POS'][parsedwikitext]
return pos
- def process_senses(self, sensesContent):
- l = sensesContent.get_lists(('\\# ', '\\#:','\\## ', '\\##:' ))
- i = 0
- senses = []
- while i < len(l):
- if l[i].pattern == '\\# ':
- theDef = self.wtp.parse(l[i].items[0]).plain_text().strip()
- if theDef != "":
- newSense = Sense(self.entry_language, theDef, self.wiki_language)
- #newSence.add_translation()
- elif l[i].pattern == '\\#:':
- for j in l[i].items:
- k = 0
- isEx = 0
- while k < len(self.wtp.parse(j).templates) and isEx == 0 :
- if (self.wtp.parse(j).templates[k].normal_name() in self.constants['t_ex']):
- newSense.add_example(self.wtp.parse(j).templates[0].arguments[-1].value)
- isEx = 1
- k += 1
- if isEx == 0:
- newSense.add_example(self.wtp.parse(j).plain_text().strip())
- if i == len(l)-1 or l[i+1].pattern == '\\# ' or l[i+1].pattern == '\\## ':
- senses.append(newSense)
- cnt = 0
- while i < len(l) and l[i].level == 3 :
- cnt +=1
- if l[i].pattern == '\\## ':
- newSense2 = SubSense(self.entry_language, self.wtp.parse(l[i].items[0]).plain_text().strip(), self.wiki_language)
- #newSense2.add_translation()
- elif l[i].pattern == '\\##:':
- for j in l[i].items:
- k = 0
- isEx = 0
- while k < len(self.wtp.parse(j).templates) and isEx == 0 :
- if (self.wtp.parse(j).templates[k].normal_name() in self.constants['t_ex']):
- newSense2.add_example(self.wtp.parse(j).templates[0].arguments[-1].value)
- isEx = 1
- k += 1
- if isEx == 0:
- newSense2.add_example(self.wtp.parse(j).plain_text().strip())
- if i == len(l)-1 or l[i+1].pattern == '\\# ' or l[i+1].pattern == '\\## ':
- newSense.add_subsense(newSense2)
- i += 1
- if cnt > 0:
- i -= 1
- i += 1
- return senses
-
if __name__ == "__main__":
ensk = En_en_straktor()
print(ensk.fetch("test"), "entries added")
diff --git a/parsers/fr_en.py b/parsers/fr_en.py
index 6a18ac5750b64d4ff58ee52bb1d7d4ec3ccc1c6d..91c43743ef70931cdf9db22e27afa6ce7c9bfdd3 100644
--- a/parsers/fr_en.py
+++ b/parsers/fr_en.py
@@ -55,57 +55,6 @@ class Fr_en_straktor(Wikstraktor):
ik += 1
return pos
- def process_example(self, example_wiki_text):
- k = 0
- isEx = 0
- res = None
- #process templates
- while k < len(self.wtp.parse(example_wiki_text).templates) and isEx == 0 :
- if (self.wtp.parse(example_wiki_text).templates[k].normal_name() in self.constants['t_ex']):
- res = self.wtp.parse(example_wiki_text).templates[0].arguments[-1].value
- isEx = 1
- k += 1
- if isEx == 0:
- res = self.wtp.parse(example_wiki_text).plain_text().strip()
- return res
-
- def process_definition(self, definition, sub_items, def_level = True):
- if def_level:
- newSense = Sense(self.entry_language, self.wtp.parse(definition).plain_text().strip(), self.wiki_language)
- pattern_ex = self.constants['sense_pattern'][0]["ex"]
- pattern_subdef = self.constants['sense_pattern'][0]["add_subdef"] + self.constants['sense_pattern'][0]["def"]
- else:
- newSense = SubSense(self.entry_language, self.wtp.parse(item).plain_text().strip(), self.wiki_language)
- pattern_subdef = None
- pattern_ex = self.constants['sense_pattern'][0]["add_subdef"] + self.constants['sense_pattern'][0]["ex"]
- #Process examples
- a = 0
- #print(newSense, sub_items)# DEBUG:
- for item_list in sub_items:
- if item_list.pattern == pattern_ex:
- for item in item_list.items:
- newSense.add_example(self.process_example(item))
- #Si on veut traiter les sous items (ex traductions), on peut utiliser
- #item_list.sublists(a)
- if def_level and item_list.pattern == pattern_subdef:
- for item in item_list.items:
- newSense.add_subsense(self.process_definition(item, item_list.sublists(a), False))
- a += 1
- return newSense
-
- def process_senses(self, sensesContent):
- l = sensesContent.get_lists((self.constants['sense_pattern'][0]["def"]))
- senses = []
- if len(l) > 1:
- self.log.add_log("Wikstraktor.process_senses", f"===== WARNING ======\nmore than one sense list, make sure we don't forget anything\nignored lists : \n{l[1:]}\n===================")
- l = l[0] #l now contains a list of list items
- if l.pattern == self.constants['sense_pattern'][0]["def"]:
- i = 0
- for item in l.items:
- senses.append(self.process_definition(item, l.sublists(i)))
- i += 1
- return senses
-
if __name__ == "__main__":
ensk = Fr_en_straktor()
print(ensk.fetch("test"), "entries added")
diff --git a/wikstraktor.py b/wikstraktor.py
index bf2317c9a39f23446dbaa834409c5450d0d6f8db..fec1efb131518ec026be156fa3cdd4f42538cf62 100755
--- a/wikstraktor.py
+++ b/wikstraktor.py
@@ -572,14 +572,58 @@ class Wikstraktor:
def process_etymology(self, parsedwikitext):
pass#in subclass
+ #can be overloaded
def process_example(self, example_wiki_text):
- pass#in subclass
+ k = 0
+ isEx = 0
+ res = None
+ #process templates
+ while k < len(self.wtp.parse(example_wiki_text).templates) and isEx == 0 :
+ if (self.wtp.parse(example_wiki_text).templates[k].normal_name() in self.constants['t_ex']):
+ res = self.wtp.parse(example_wiki_text).templates[0].arguments[-1].value
+ isEx = 1
+ k += 1
+ if isEx == 0:
+ res = self.wtp.parse(example_wiki_text).plain_text().strip()
+ return res
+ #can be overloaded
def process_definition(self, definition, sub_items, def_level = True):
- pass#in subclass
-
- def process_senses(self, parsedwikitext):
- pass#in subclass
+ if def_level:
+ newSense = Sense(self.entry_language, self.wtp.parse(definition).plain_text().strip(), self.wiki_language)
+ pattern_ex = self.constants['sense_pattern'][0]["ex"]
+ pattern_subdef = self.constants['sense_pattern'][0]["add_subdef"] + self.constants['sense_pattern'][0]["def"]
+ else:
+ newSense = SubSense(self.entry_language, self.wtp.parse(definition).plain_text().strip(), self.wiki_language)
+ pattern_subdef = None
+ pattern_ex = self.constants['sense_pattern'][0]["add_subdef"] + self.constants['sense_pattern'][0]["ex"]
+ #Process examples
+ a = 0
+ #print(newSense, sub_items)# DEBUG:
+ for item_list in sub_items:
+ if item_list.pattern == pattern_ex:
+ for item in item_list.items:
+ newSense.add_example(self.process_example(item))
+ #Si on veut traiter les sous items (ex traductions), on peut utiliser
+ #item_list.sublists(a)
+ if def_level and item_list.pattern == pattern_subdef:
+ for item in item_list.items:
+ newSense.add_subsense(self.process_definition(item, item_list.sublists(a), False))
+ a += 1
+ return newSense
+
+ def process_senses(self, sensesContent):
+ l = sensesContent.get_lists((self.constants['sense_pattern'][0]["def"]))
+ senses = []
+ if len(l) > 1:
+ self.log.add_log("Wikstraktor.process_senses", f"===== WARNING ======\nmore than one sense list, make sure we don't forget anything\nignored lists : \n{l[1:]}\n===================")
+ l = l[0] #l now contains a list of list items
+ if l.pattern == self.constants['sense_pattern'][0]["def"]:
+ i = 0
+ for item in l.items:
+ senses.append(self.process_definition(item, l.sublists(i)))
+ i += 1
+ return senses
def __str__(self):
return self.export()