docstring for functions related to involved in

main
Jean-Damien 3 years ago
parent 966be7c4bf
commit ca604c277c

@ -232,22 +232,36 @@ def __grape_indiv(list_person, role: str):
print(person_text, "==", prince_q[0]) print(person_text, "==", prince_q[0])
def _create_involved_in(folder: str): def _create_involved_in(folder: str):
"""create involved in table
NB : relation table containing 3 foreigns keys
"""
princes_actes = [] princes_actes = []
for acte in os.listdir(folder): for acte in os.listdir(folder):
if acte.endswith(".xml"): if acte.endswith(".xml"):
# 1/ Get act's id.
# the file name is the act'id when you remove ".xml"
# so we make a query on Acte to get the primary key.
# the act's id remains the same whether the individual
# is a prince or a signatory, so no need to iterate.
acte_q = [t.id_acte for t in Acte.select().where( acte_q = [t.id_acte for t in Acte.select().where(
Acte.filename == acte.replace(".xml", ""))] Acte.filename == acte.replace(".xml", ""))]
# print(acte, "==", acte_q[0]) # print(acte, "==", acte_q[0])
soup = make_soup(os.path.join(folder, acte)) soup = make_soup(os.path.join(folder, acte))
# 2A/ 1st iteration for princes individuals.
for persons in soup.sourceDesc.find_all("listPerson", {"type": "prince"}): for persons in soup.sourceDesc.find_all("listPerson", {"type": "prince"}):
# persons == //sourceDesc/listPerson/listPerson[@type="prince"]
for person_tag in persons.find_all("person"): for person_tag in persons.find_all("person"):
# person_tag = <person>\n<name>\n?prince's name\n?</name>\n</person>
person_text = person_tag.text.replace("\n", "") person_text = person_tag.text.replace("\n", "")
if person_text != "None": if person_text != "None":
# get the prince's id
prince_q = [t.id_indiv for t in Individual.select().where( prince_q = [t.id_indiv for t in Individual.select().where(
Individual.name_indiv == person_text)] Individual.name_indiv == person_text)]
# a prince is always an act's producer
interv_q = [t.id_intev for t in Intervention_type.select().where( interv_q = [t.id_intev for t in Intervention_type.select().where(
Intervention_type.interv_label == "producer")] Intervention_type.interv_label == "producer")]
# print(person_text, "==", prince_q[0]) # print(person_text, "==", prince_q[0])
# Exception Handling : when we forget to add the name in the csv
try: try:
prince_q[0] prince_q[0]
except IndexError: except IndexError:
@ -256,6 +270,9 @@ def _create_involved_in(folder: str):
princes_actes.append({"involved_in_acte": acte_q[0], princes_actes.append({"involved_in_acte": acte_q[0],
"involved_in_prince": prince_q[0], "involved_in_prince": prince_q[0],
"invol_in_interv": interv_q[0]}) "invol_in_interv": interv_q[0]})
# 2B/ 2th iteration for signatories individuals.
# NB : a prince may sign an act
# the process is basically the same than 2A
for persons in soup.sourceDesc.find_all("listPerson", {"type": "signatory"}): for persons in soup.sourceDesc.find_all("listPerson", {"type": "signatory"}):
for person_tag in persons.find_all("person"): for person_tag in persons.find_all("person"):
person_text = person_tag.text.replace("\n", "") person_text = person_tag.text.replace("\n", "")
@ -273,6 +290,7 @@ def _create_involved_in(folder: str):
princes_actes.append({"involved_in_acte": acte_q[0], princes_actes.append({"involved_in_acte": acte_q[0],
"involved_in_prince": prince_q[0], "involved_in_prince": prince_q[0],
"invol_in_interv": interv_q[0]}) "invol_in_interv": interv_q[0]})
# 3/ Create the table
for data in tqdm(princes_actes, desc="Populating involved_in..."): for data in tqdm(princes_actes, desc="Populating involved_in..."):
Involved_in.create(**data) Involved_in.create(**data)

Loading…
Cancel
Save