From 5bf2f6f7161bfd64b2de8d86c9020833678f4068 Mon Sep 17 00:00:00 2001
From: Mathieu Loiseau <mathieu.loiseau@liris.cnrs.fr>
Date: Thu, 16 Mar 2023 01:00:04 +0100
Subject: [PATCH] v1 avec identifiants
---
parsers/en_en.py | 4 +---
parsers/fr_en.py | 1 -
wikstraktor.py | 31 ++++++++++++++++---------------
3 files changed, 17 insertions(+), 19 deletions(-)
diff --git a/parsers/en_en.py b/parsers/en_en.py
index 1e1a5de..8761a51 100644
--- a/parsers/en_en.py
+++ b/parsers/en_en.py
@@ -19,7 +19,7 @@ class En_en_straktor(Wikstraktor):
i = 0
pronunciations = []
while i < len(l.fullitems):
- p = Pronunciation(len(pronunciations))
+ p = Pronunciation()
templates = self.wtp.parse(l.fullitems[i]).templates
a = None
for j, t in enumerate(templates):
@@ -49,7 +49,6 @@ class En_en_straktor(Wikstraktor):
return pos
def process_senses(self, sensesContent):
- print("process_senses")
l = sensesContent.get_lists(('\\# ', '\\#:','\\## ', '\\##:' ))
i = 0
senses = []
@@ -95,7 +94,6 @@ class En_en_straktor(Wikstraktor):
if cnt > 0:
i -= 1
i += 1
- print(f"process_senses done {senses[-1].serializable()}")
return senses
if __name__ == "__main__":
diff --git a/parsers/fr_en.py b/parsers/fr_en.py
index a44ff5d..49f9896 100644
--- a/parsers/fr_en.py
+++ b/parsers/fr_en.py
@@ -53,7 +53,6 @@ class Fr_en_straktor(Wikstraktor):
keys = list(self.constants['POS'].keys())
pos = keys[ik]
ik += 1
- print(pos)
return pos
def process_senses(self, sensesContent):
diff --git a/wikstraktor.py b/wikstraktor.py
index 23a81f4..0a38474 100755
--- a/wikstraktor.py
+++ b/wikstraktor.py
@@ -83,7 +83,7 @@ class Pronunciation(SubInfo):
return res
def __str__(self):
- return f"{self.serializable('')}"
+ return json.dumps(self.serializable(''))
def __eq__(self, other):
res = self.ipa == other.ipa and self.accent == other.accent and len(self.sounds)==len(other.sounds)
@@ -167,12 +167,15 @@ class Sense(SubInfo):
self.translations = [] #liste des traductions dans d'autres langues
self.domain = None #domaine d'usage du mot dans ce sens
- def set_id(self, prefix):
+ def set_id(self, prefix=None):
if prefix != None and self.label == None:
- self.label = f"{self.lang}.{prefix}_{self.__class__.next_id}" #l'identifiant du sens
+ self.label = f"{prefix}_{self.__class__.next_id}" #l'identifiant du sens
self.__class__.inc_n_id()
return self.label
+ def get_id(self):
+ return f"{self.lang}.{self.label}"
+
def set_domain(self, d):
self.domain = d
@@ -242,8 +245,11 @@ class Sense(SubInfo):
res["Translations"].append(t.serializable(prefix))
return res
+ def __str__(self):
+ return json.dumps(self.serializable())
+
class SubSense(Sense):
- def set_id(self, prefix):
+ def set_id(self, prefix=None):
if prefix != None and self.label == None:
self.label = f"{prefix}.{self.__class__.next_id}" #l'identifiant du sens
self.__class__.inc_n_id()
@@ -268,8 +274,9 @@ class Entry:
self.pos = pos
def get_id(self, source_id=0):
- if self.pos !=None:
- pos = "_"+self.pos
+ #TODO: remplacer un jour le source id par la bonne source
+ if self.pos != None:
+ pos = self.pos
else:
pos = ""
return f"{self.lang}-{source_id}.{self.lemma}{pos}"
@@ -292,7 +299,7 @@ class Entry:
self.pronunciations.append(p)
def set_senses(self, senses):
- for s in self.senses:
+ for s in senses:
if isinstance(s, Sense):
self.add_sense(s)
else:
@@ -301,7 +308,7 @@ class Entry:
def add_sense(self, s):
if s not in self.senses:
s.set_id(self.get_id())
- self.senses.append(s)##ICITE
+ self.senses.append(s)
def is_valid(self):
return self.lemma != None and len(self.pronunciations) > 0 and self.pos != None and len(self.senses) > 0
@@ -333,7 +340,6 @@ class Entry:
res[self.lemma]["senses"] = {}
for s in self.senses:
res[self.lemma]["senses"][s.get_id()]=s.serializable(id)
- print(f"Entry:{res}")##
return res
def __str__(self):
@@ -387,7 +393,6 @@ class ParserContext:
#Dans le dictionnaire de keys, il n'y a jamais de senses ou de POS
res = Entry(self.lemma, self.lang, self.wiki_lang, self.version_id)
for l in self.context:
- #print(l.keys())
if "pro" in l.keys():
res.set_pronunciations(l['pro'])
if "ety" in l.keys():
@@ -397,7 +402,6 @@ class ParserContext:
if "senses" in l.keys():
res.set_senses(l['senses'])
# TODO: Ajouter les autres types
- print(res)
if res.is_valid() and res not in self.entries:
self.entries.append(res)
else:
@@ -491,6 +495,7 @@ class Wikstraktor:
self.parserContext.set_top_entry_info('POS', pos, False)
self.parserContext.set_top_entry_info('senses', self.process_senses(self.wtp.parse(s.contents))) #cette ligne le prob
# self.parserContext.entries augmente pas même lorsque qu'on entre dans le if au dessus.
+
res = len(self.parserContext.entries)
if res > 0:
for e in self.parserContext.entries:
@@ -502,7 +507,6 @@ class Wikstraktor:
res = title == self.constants['pro']
else:
res = title in self.constants['pro']
- #print(title, res)
return res
def isEty(self, title):
@@ -531,7 +535,6 @@ class Wikstraktor:
res = []
for e in self.entries:
res.append(e.serializable(id))
- print(f"Export{res}")##
if compact:
return json.dumps(res, ensure_ascii=ascii)
else:
@@ -558,8 +561,6 @@ if __name__ == "__main__":
resp = None
if w.fetch(args.mot) > 0:
resp = w.export(not args.no_id, args.force_ascii, args.compact)
- print(w)##
- print(resp)##
if args.destination_file != None:
f = open(args.destination_file, "w")
f.write(resp)
--
GitLab