fares.amer/ScoDoc
1
0
Fork 0
forked from ScoDoc/ScoDoc
Code Issues Pull Requests Projects Releases Wiki Activity

Améliore anonymisation (users) + lien contact + cosmetic

Browse Source
This commit is contained in:
Emmanuel Viennet 2024-03-01 11:12:36 +01:00
parent e56cbfc5a2
commit ab39454a0d
4 changed files with 114 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
app/but/cursus_but.py
View File

@ -563,7 +563,8 @@ def formation_semestre_niveaux_warning(formation: Formation, semestre_idx: int)
if nb_niveaux_tc != nb_ues_tc:
H.append(
f"""<li>{nb_niveaux_tc} niveaux de compétences de tronc commun,
mais {nb_ues_tc} UEs de tronc commun !</li>"""
mais {nb_ues_tc} UEs de tronc commun ! (c'est normal si
vous avez des UEs différenciées par parcours)</li>"""
)
if H:

2
app/scodoc/sco_dept.py
View File

@ -208,6 +208,8 @@ def index_html(showcodes=0, showsemtable=0):
"""<hr>
<h3>Assistance</h3>
<ul>
<li><a class="stdlink" href="https://scodoc.org/Contact" target="_blank"
rel="noopener noreferrer">Contact (Discord)</a></li>
<li><a class="stdlink" href="sco_dump_and_send_db">Envoyer données</a></li>
</ul>
"""

2
sco_version.py
View File

@ -1,7 +1,7 @@
# -*- mode: python -*-
# -*- coding: utf-8 -*-
SCOVERSION = "9.6.949"
SCOVERSION = "9.6.950"
SCONAME = "ScoDoc"

139
tools/anonymize_db.py
View File

@ -30,12 +30,15 @@
Runned as user "scodoc" with scodoc and postgresql up.
E. Viennet, Jan 2019
"""
Travaille entièrement au niveau SQL, n'utilise aucun modèle SQLAlchemy.
import psycopg2
E. Viennet, Jan 2019, Fev 2024
"""
import random
import sys
import traceback
import psycopg2
from psycopg2 import extras
def log(msg):
@ -59,9 +62,21 @@ anonymize_false = "FALSE"
anonymize_question_str = "'?'"
anonymize_null = "NULL"
# --- Listes de noms et prénoms pour remplacer les identités
NOMS = [
x.strip()
for x in open("/opt/scodoc/tools/fakeportal/nomsprenoms/noms.txt", encoding="utf8")
]
PRENOMS = [
x.strip()
for x in open(
"/opt/scodoc/tools/fakeportal/nomsprenoms/prenoms.txt", encoding="utf8"
)
]
# --- Champs à anonymiser (cette configuration pourrait être placé dans
# un fichier séparé et le code serait alors générique pour toute base
# posgresql.
# postgresql.
#
# On essaie de retirer les données personnelles des étudiants et des entreprises
#
@ -111,6 +126,26 @@ def anonymize_column(cursor, tablecolumn):
cursor.execute(f"UPDATE {table} SET {column} = {anonymized};")
def rename_students(cursor):
"""Remet des noms/prenoms fictifs aux étuduiants"""
# Change les noms/prenoms
cursor.execute("""SELECT * FROM "identite";""")
etuds = cursor.fetchall()
for etud in etuds:
nom, prenom = random.choice(NOMS), random.choice(PRENOMS)
cursor.execute(
"""UPDATE "identite"
SET nom=%(nom)s, prenom=%(prenom)s
WHERE id=%(id)s
""",
{
"id": etud["id"],
"nom": nom,
"prenom": prenom,
},
)
def anonymize_users(cursor):
"""Anonymise la table utilisateurs"""
log("processing user table")
@ -121,8 +156,51 @@ def anonymize_users(cursor):
cursor.execute("""UPDATE "user" SET date_expiration = '2201-12-31';""")
cursor.execute("""UPDATE "user" SET token = NULL;""")
cursor.execute("""UPDATE "user" SET token_expiration = NULL;""")
cursor.execute("""UPDATE "user" SET nom=CONCAT('nom_', id);""")
cursor.execute("""UPDATE "user" SET prenom=CONCAT('nom_', id);""")
# Change les noms/prenoms/mail
cursor.execute("""SELECT * FROM "user";""")
users = cursor.fetchall() # fetch tout car modifie cette table ds la boucle
used_user_names = {u["user_name"] for u in users}
for user in users:
user_name = user["user_name"]
nom, prenom = random.choice(NOMS), random.choice(PRENOMS)
new_name = (prenom[0] + nom).lower()
# unique ?
while new_name in used_user_names:
new_name += "x"
used_user_names.add(new_name)
print(f"{user_name} > {new_name}")
cursor.execute(
"""UPDATE "user"
SET nom=%(nom)s, prenom=%(prenom)s, email=%(email)s, user_name=%(new_name)s
WHERE id=%(id)s
""",
{
"email": f"{prenom}.{nom}@ano.nyme",
"id": user["id"],
"nom": nom,
"prenom": prenom,
"new_name": new_name,
},
)
# Change les username: utilisés en référence externe
# dans diverses tables:
for table, field in (
("etud_annotations", "author"),
("scolog", "authenticated_user"),
("scolar_news", "authenticated_user"),
("notes_appreciations", "author"),
("are_historique", "authenticated_user"),
):
cursor.execute(
f"""UPDATE "{table}"
SET {field}=%(new_name)s
WHERE {field}=%(user_name)s
""",
{
"new_name": new_name,
"user_name": user_name,
},
)
def anonymize_db(cursor):
@ -131,32 +209,33 @@ def anonymize_db(cursor):
anonymize_column(cursor, tablecolumn)
process_users = False
if len(sys.argv) < 2 or len(sys.argv) > 3:
usage()
if len(sys.argv) > 2:
if sys.argv[1] != "--users":
if __name__ == "__main__":
PROCESS_USERS = False
if len(sys.argv) < 2 or len(sys.argv) > 3:
usage()
dbname = sys.argv[2]
process_users = True
else:
dbname = sys.argv[1]
if len(sys.argv) > 2:
if sys.argv[1] != "--users":
usage()
dbname = sys.argv[2]
PROCESS_USERS = True
else:
dbname = sys.argv[1]
log(f"\nAnonymizing database {dbname}")
cnx_string = "dbname=" + dbname
try:
cnx = psycopg2.connect(cnx_string)
except Exception as e:
log(f"\n*** Error: can't connect to database {dbname} ***\n")
log(f"""connexion string was "{cnx_string}" """)
traceback.print_exc()
log(f"\nAnonymizing database {dbname}")
cnx_string = "dbname=" + dbname
try:
cnx = psycopg2.connect(cnx_string)
except Exception as e:
log(f"\n*** Error: can't connect to database {dbname} ***\n")
log(f"""connexion string was "{cnx_string}" """)
traceback.print_exc()
cnx.set_session(autocommit=False)
cursor = cnx.cursor()
cnx.set_session(autocommit=False)
cursor = cnx.cursor(cursor_factory=psycopg2.extras.DictCursor)
anonymize_db(cursor)
if process_users:
anonymize_users(cursor)
anonymize_db(cursor)
if PROCESS_USERS:
anonymize_users(cursor)
cnx.commit()
cnx.close()
cnx.commit()
cnx.close()