From 2827aaf28c8886e20d78f7ce41573e5383d58764 Mon Sep 17 00:00:00 2001
From: Mathieu Loiseau <mathieu.loiseau@liris.cnrs.fr>
Date: Thu, 16 Mar 2023 09:50:57 +0100
Subject: [PATCH] Create Sense with definition
---
.gitignore | 4 ++--
parsers/en_en.py | 9 ++++----
parsers/fr_en.py | 6 ++----
wikstraktor.py | 53 ++++++++++++++++++++++++++++++++----------------
4 files changed, 44 insertions(+), 28 deletions(-)
diff --git a/.gitignore b/.gitignore
index a240499..69e9892 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,6 +1,5 @@
user-config.py
user-password.py
-pywikibot.lwp
__pycache__
apicache-py3
logs
@@ -8,5 +7,6 @@ throttle.ctrl
user_list.py
KNM.csv
.~lock*
-test.json
+*.json
+*.lwp
wikstraktorenv
diff --git a/parsers/en_en.py b/parsers/en_en.py
index 8761a51..cf93078 100644
--- a/parsers/en_en.py
+++ b/parsers/en_en.py
@@ -54,8 +54,10 @@ class En_en_straktor(Wikstraktor):
senses = []
while i < len(l):
if l[i].pattern == '\\# ':
- newSense = Sense(lang=self.entry_language)
- newSense.add_def(self.wiki_language, self.wtp.parse(l[i].items[0]).plain_text().strip())
+ theDef = self.wtp.parse(l[i].items[0]).plain_text().strip()
+ if theDef != "":
+ print(theDef)# DEBUG:
+ newSense = Sense(self.entry_language, theDef, self.wiki_language)
#newSence.add_translation()
elif l[i].pattern == '\\#:':
for j in l[i].items:
@@ -74,8 +76,7 @@ class En_en_straktor(Wikstraktor):
while i < len(l) and l[i].level == 3 :
cnt +=1
if l[i].pattern == '\\## ':
- newSense2 = SubSense(lang=self.entry_language)
- newSense2.add_def(self.wiki_language, self.wtp.parse(l[i].items[0]).plain_text().strip())
+ 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:
diff --git a/parsers/fr_en.py b/parsers/fr_en.py
index 49f9896..fcaa931 100644
--- a/parsers/fr_en.py
+++ b/parsers/fr_en.py
@@ -61,8 +61,7 @@ class Fr_en_straktor(Wikstraktor):
senses = []
while i < len(l):
if l[i].pattern == '\\# ':
- newSense = Sense(lang=self.entry_language)
- newSense.add_def(self.wiki_language, self.wtp.parse(l[i].items[0]).plain_text().strip())
+ newSense = Sense(self.entry_language, self.wtp.parse(l[i].items[0]).plain_text().strip(), self.wiki_language)
#la version d'enzo ajoute +ieurs defs (for i in l[i].items)
elif l[i].pattern == '\\#:':
for j in l[i].items:
@@ -81,8 +80,7 @@ class Fr_en_straktor(Wikstraktor):
while i < len(l) and l[i].level == 3 :
cnt +=1
if l[i].pattern == '\\## ':
- newSense2 = SubSense(lang=self.entry_language)
- newSense2.add_def(self.wiki_language, self.wtp.parse(l[i].items[0]).plain_text().strip())
+ newSense2 = SubSense(self.entry_language, self.wtp.parse(l[i].items[0]).plain_text().strip(), self.wiki_language)
elif l[i].pattern == '\\##:':
for j in l[i].items:
k = 0
diff --git a/wikstraktor.py b/wikstraktor.py
index 0a38474..c1e98fb 100755
--- a/wikstraktor.py
+++ b/wikstraktor.py
@@ -4,6 +4,8 @@ import wikitextparser
import importlib
import json
+#ICITE : fr marche pas, en prend des trucs vides à virer (cf. yellow… def & example)
+
class SubInfo:
next_id = 1
@@ -44,7 +46,7 @@ class Sound:
self.accent = accent
def __eq__(self, other):
- return self.url == other.url and self.accent == other.accent
+ return isinstance(other, self.__class__) and self.url == other.url and self.accent == other.accent
def serializable(self):
if self.accent == None:
@@ -86,7 +88,7 @@ class Pronunciation(SubInfo):
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)
+ res = isinstance(other, self.__class__) and self.ipa == other.ipa and self.accent == other.accent and len(self.sounds)==len(other.sounds)
i = 0
while res and i<len(self.sounds):
res = self.sounds[i] == other.sounds[i]
@@ -110,11 +112,14 @@ class Definition(SubInfo):
def __init__(self, lang, text, prefix=None):
super().__init__(prefix)
- self.lang = lang
- self.text = text
+ if text != "":
+ self.lang = lang
+ self.text = text
+ else:
+ raise ValueError(f"Definition.__init__: “{text}†empty definition.")
def __eq__(self, other):
- return self.lang == other.lang and self.text == other.text
+ return isinstance(other, self.__class__) and self.lang == other.lang and self.text == other.text
def serializable(self, prefix = None):
res = super().serializable(prefix)
@@ -131,12 +136,16 @@ class Example(SubInfo):
def __init__(self, transcript, source=None, url=None, prefix=None):
super().__init__(prefix)
- self.text = transcript
- self.source = source
- self.url = url
+ if transcript != "":
+ self.text = transcript
+ self.source = source
+ self.url = url
+ else:
+ raise ValueError(f"Example.__init__: “{transcript}†empty example.")
+
def __eq__(self, other):
- return self.text==other.text and self.source==other.source and self.url==other.url
+ return isinstance(other, self.__class__) and self.text==other.text and self.source==other.source and self.url==other.url
def serializable(self, prefix = None):
res = super().serializable(prefix)
@@ -150,7 +159,7 @@ class Example(SubInfo):
class Sense(SubInfo):
prfx = ""
- def __init__(self, prefix=None, lang=None):
+ def __init__(self, lang=None, definition=None, wiki_lang=None, prefix=None):
self.lang = lang
self.label = None
self.set_id(prefix)
@@ -166,6 +175,11 @@ class Sense(SubInfo):
self.examples = [] #liste des exemples (un texte obligatoire, source et url sont optionnels)
self.translations = [] #liste des traductions dans d'autres langues
self.domain = None #domaine d'usage du mot dans ce sens
+ if definition != None:
+ try:
+ self.add_def(wiki_lang, definition)
+ except ValueError as err:
+ raise ValueError(f"Sense.__init__() with empty definition\n{err}")
def set_id(self, prefix=None):
if prefix != None and self.label == None:
@@ -181,19 +195,22 @@ class Sense(SubInfo):
def add_def(self, lang, definition):
theDef = Definition(lang, definition)
- if theDef not in self.definitions:
+ if theDef != None and theDef not in self.definitions:
theDef.set_id(self.set_id())
self.definitions.append(theDef)
def add_example(self, transcript, src=None, url=None, prefix=None):
- theEx = Example(transcript, src, url, prefix)
- if theEx not in self.examples:
- theEx.set_id(self.set_id())
- self.examples.append(theEx)
+ try:
+ theEx = Example(transcript, src, url, prefix)
+ if theEx != None and theEx not in self.examples:
+ theEx.set_id(self.set_id())
+ self.examples.append(theEx)
+ except ValueError as e:
+ print(f"Skipped empty example")
def add_translation(self, lang=None, translation=None):
theTranslation = Translation(lang, translation)
- if theTranslation not in self.translations:
+ if theTranslation != None and theTranslation not in self.translations:
theTranslation.set_id(self.set_id())
self.translations.append(theTranslation)
@@ -204,7 +221,7 @@ class Sense(SubInfo):
self.subsenses.append(subsense)
def __eq__(self, other):
- res = self.label == other.label 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
+ res = isinstance(other, self.__class__) and self.label == other.label 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
i = 0
while res and i < len(self.examples):
res = self.examples[i] in other.examples
@@ -314,7 +331,7 @@ class Entry:
return self.lemma != None and len(self.pronunciations) > 0 and self.pos != None and len(self.senses) > 0
def __eq__(self, other):
- res = self.lemma == other.lemma and self.lang == other.lang and self.pos ==other.pos and len(self.pronunciations) == len(other.pronunciations) and len(self.senses) == len(other.senses)
+ res = isinstance(other, self.__class__) and self.lemma == other.lemma and self.lang == other.lang and self.pos ==other.pos and len(self.pronunciations) == len(other.pronunciations) and len(self.senses) == len(other.senses)
i = 0
while res and i < len(self.senses):
res = self.senses[i] == other.senses[i]
--
GitLab