Merge branch 'json-spells'

merge-requests/29/head
James Miller 2019-08-10 16:17:45 -05:00
commit bc1dc03bdb
4 changed files with 310 additions and 23 deletions

View File

@ -1,9 +1,13 @@
INSERT INTO spelltypes (
spelltypes_id,
name
)
VALUES
(1, 'Spell'),
(2, 'Cantrip'),
(3, 'Focus'),
(3, 'Ritual')
);
(4, 'Ritual');
INSERT INTO spellcomponents (
spellcomponents_id,
@ -41,4 +45,4 @@ VALUES
(5, 1, '298','Evocation', 'TODO'),
(6, 1, '298','Illusion', 'TODO'),
(7, 1, '298','Necromancy', 'TODO'),
(8, 1, '298','Transmutation', 'Transmutation spells make alterations to or transform the physical form of a creature or object. The morph and polymorph traits appear primarily in transmutation spells.'),
(8, 1, '298','Transmutation', 'Transmutation spells make alterations to or transform the physical form of a creature or object. The morph and polymorph traits appear primarily in transmutation spells.');

View File

@ -1,4 +1,5 @@
import json
import sqlite3
def main():
# load json into python
@ -25,12 +26,271 @@ def main():
sorted_dicts.append(x)
# NOW we can go alphabetically spell by spell
for i in sorted_dicts:
do_sql(i)
# TODO write this function after sql schema drafted
def do_sql():
pass
## Get database connection
conn = sqlite3.connect('../../pf2.db')
# load in ids for traits from traits table so we only call this once
# instead of every spell
stmt = "SELECT trait_id, short_name FROM traits"
c = conn.cursor()
c.execute(stmt)
traits = c.fetchall()
# print(traits)
# load in ids for spelltypes from spelltypes table so we only call this once
# instead of every spell
stmt = "SELECT spelltypes_id, name FROM spelltypes"
c = conn.cursor()
c.execute(stmt)
stypes = c.fetchall()
# TODO FIX THIS FOR SPELL COMPONENTS
# CREATE TABLE spellcomponents (
# spellcomponents_id INTEGER PRIMARY KEY,
# name TEXT NOT NULL UNIQUE
# );
# load in ids for spelltypes from spelltypes table so we only call this once
# instead of every spell
stmt = "SELECT spellcomponents_id, name FROM spellcomponents"
c = conn.cursor()
c.execute(stmt)
ctypes = c.fetchall()
# List the various triggers and see if there are any duplicates
# THERE ARE NOT IN THE CRB SO NOT BOTHERING WITH SEPARATE TRIGGERS TABLE YET
### trigs = []
### for i in sorted_dicts:
### if 'trigger' in i:
### trigs.append(i['trigger'])
### print(sorted(trigs))
### print(len(trigs))
### print(len(set(trigs)))
# List the various targets and see if there are any duplicates
## YES, there are MANY duplicates, so we need a separate targets table
targs = []
for i in sorted_dicts:
if 'targets' in i:
targs.append(i['targets'])
dedup_targs = set(targs)
sorted_targs = sorted(dedup_targs)
inp_targs = []
id = 0
for i in sorted_targs:
id += 1
inp_targs.append((id,i))
stmt = "INSERT INTO spelltargets (spelltargets_id, name) VALUES (?,?)"
try:
conn.executemany(stmt,inp_targs)
except:
print("Error creating targets")
else:
conn.commit()
# load in ids for targets so just doing this once
stmt = "SELECT spelltargets_id, name FROM spelltargets"
c = conn.cursor()
c.execute(stmt)
ttypes = c.fetchall()
# print(sorted(targs))
# print(len(targs))
# print(len(set(targs)))
id = 0
for i in sorted_dicts:
id += 1
do_basic_sql(i, id, conn)
do_range_numbers(i,id,conn)
do_sources_pages(i,id,conn)
do_spell_traits(i,id,conn,traits)
do_spell_types(i,id,conn,stypes)
do_spell_components(i,id,conn,ctypes)
do_spell_targets(i,id,conn,ttypes)
def do_spell_components(i,id,conn,ctypes):
res = None
for j in ctypes:
for k in i['components']:
if k.capitalize() == j[1]:
res = j[0]
inp = (res, id)
stmt = "INSERT INTO spells_spellcomponents (spells_id, spellcomponents_id) VALUES (?,?)"
try:
conn.execute(stmt, inp)
except:
print("Error inserting spell components")
else:
conn.commit()
def do_spell_targets(i,id,conn,ttypes):
if 'targets' not in i:
return
res = 0
for j in ttypes:
if i['targets'] == j[1]:
res = j[0]
# print(id , res)
inp = (res, id)
stmt = "UPDATE spells SET spelltargets_id=? WHERE spells_id=?"
try:
conn.execute(stmt, inp)
except:
print("Error updating spelltargets_id")
else:
conn.commit()
def do_spell_types(i,id,conn,stypes):
res = 0
for j in stypes:
if i['type'] == j[1]:
res = j[0]
# print(id , res)
inp = (res, id)
stmt = "UPDATE spells SET spelltypes_id=? WHERE spells_id=?"
try:
conn.execute(stmt, inp)
except:
print("Error updating spell types")
else:
conn.commit()
def do_spell_traits(i, id, conn, traits):
# get list of traits from the json and capitalize first letter
traits_json = []
for item in i['traits']:
traits_json.append(item.capitalize())
trait_ids =[]
for j in traits_json:
for k in traits:
if j == k[1]:
trait_ids.append(k[0])
# print(trait_ids)
inp = []
for j in trait_ids:
inp.append((id,j))
# print(inp)
# insert into sql
stmt = "INSERT OR REPLACE INTO spells_traits (spells_id, traits_id) VALUES (?,?)"
try:
conn.executemany(stmt, inp)
except:
print("Error updating traits")
else:
conn.commit()
def do_sources_pages(i, id, conn):
if 'source' not in i:
return
res = ''
source_id = 0
# Do Core Rulebook branch
if "Core Rulebook" in i['source']:
res = i['source'].replace('Core Rulebook pg.','').strip()
source_id = 1
stmt = "UPDATE spells SET sources_id=?, sources_pages=? WHERE spells_id=?"
inp = (source_id, res, id)
try:
conn.execute(stmt, inp)
except:
print("Error updating sources")
else:
conn.commit()
def do_range_numbers(i, id, conn):
# no need to do range
if 'range' not in i:
return
rg = -1
# convert range_text to an integer representation
if i['range'] == 'touch':
rg = 0
elif i['range'] == 'planetary':
rg = 999999999
# is the only one in CRB with emanation 40' from current scraping
elif i['name'] == 'Repulsion':
rg = 40
else:
# DO SPLITS
splits = i['range'].split(' ')
# print(splits)
rg = splits[0]
inp = (rg, id)
stmt = "UPDATE spells SET range_ft=? WHERE spells_id=?"
try:
conn.execute(stmt, inp)
except:
print("Error updating range_ft")
else:
conn.commit()
# print("Successfully updated range_ft")
def do_basic_sql(i, id, conn):
print("Doing spell id #{}: {}".format(id, i['name']))
stmt = """INSERT INTO spells (
spells_id,
sources_id,
sources_pages,
nethysurl,
name,
level,
descr,
range_text,
trigger,
area_text)
VALUES (?,?,?,?,?,?,?,?,?,?)"""
rge = None
if 'range' in i:
rge = i['range']
dscr = None
if 'description' in i:
dscr = i['description']
trg = None
if 'trigger' in i:
trg = i['trigger']
area = None
if 'area' in i:
area = i['area']
inp = (id, 1, i['source'], i['nethysUrl'], i['name'], i['level'], dscr, rge, trg, area)
try:
conn.execute(stmt, inp)
except:
print("Error inserting row")
else:
conn.commit()
# print("Successfully inserted row")
if __name__ == "__main__":

View File

@ -8,6 +8,7 @@ sqlite3 pf2.db < schema/bulk.sql
sqlite3 pf2.db < schema/sizes.sql
sqlite3 pf2.db < schema/langs.sql
sqlite3 pf2.db < schema/traits.sql
sqlite3 pf2.db < schema/spells.sql
sqlite3 pf2.db < schema/feats.sql
sqlite3 pf2.db < schema/senses.sql
sqlite3 pf2.db < schema/ancestries.sql
@ -22,9 +23,20 @@ sqlite3 pf2.db < data/senses.sql
sqlite3 pf2.db < data/sizes.sql
sqlite3 pf2.db < data/langs.sql
sqlite3 pf2.db < data/traits.sql
sqlite3 pf2.db < data/spells.sql
sqlite3 pf2.db < data/feats.sql
sqlite3 pf2.db < data/ancestries.sql
sqlite3 pf2.db < data/armor.sql
sqlite3 pf2.db < data/heritages.sql
# Comment out the following three lines if you don't want to generate the spell data.
cd data/third_party_json
python3 spells.py
cd ../..
# TODO Eventually we will stop relying on the spells.py script and I will have
# the actual .sql files for the spell data; I am waiting to see if the
# third-party source improves the data in the next few weeks. If not, we'll
# "divorce" from that data, dump to .sql, and manually manipulate going
# forward.

View File

@ -15,6 +15,11 @@ CREATE TABLE spelltraditions (
name TEXT NOT NULL UNIQUE
);
CREATE TABLE spelltargets (
spelltargets_id INTEGER PRIMARY KEY,
name TEXT NOT NULL UNIQUE
);
CREATE TABLE spellschools (
spellschools_id INTEGER PRIMARY KEY,
sources_id INTEGER NOT NULL,
@ -24,27 +29,32 @@ CREATE TABLE spellschools (
FOREIGN KEY (sources_id) REFERENCES sources(sources_id)
);
-- TODO eventually once data is finalized, lock down variables as NOT NULL /
-- UNIQUE as sanity requires :)
-- TODO Area eventually needs its own table
CREATE TABLE spells (
spells_id INTEGER PRIMARY KEY,
sources_id INTEGER NOT NULL,
sources_pages TEXT,
nethysurl TEXT,
name TEXT NOT NULL UNIQUE,
source TEXT,
level INTEGER NOT NULL,
has_trigger BOOLEAN NOT NULL,
trigger TEXT,
descr TEXT NOT NULL,
spelltypes_id INTEGER NOT NULL,
range_text TEXT,
range_ft INTEGER,
targets TEXT,
sources_id INTEGER NOT NULL, -- generated in spells.py from scraped data
sources_pages TEXT, -- generated in spells.py from scraped data
name TEXT NOT NULL UNIQUE, -- scraped from github repo
level INTEGER, -- scraped from github repo
trigger TEXT, -- scraped from spells.py NOTE, there are no duplicate triggers
-- as of CRB, so not bothering with a separate spell triggers
-- table at this time
descr TEXT, -- scraped from github repo
spelltypes_id INTEGER, -- generated from spells.py
range_text TEXT, -- scraped from github repo
range_ft INTEGER, -- generated from text in spells.py
area_text TEXT, -- TODO need to figure out some sort of programmatic representation for this too
spelltargets_id INTEGER,
nethysurl TEXT, -- scraped from github repo
FOREIGN KEY (sources_id) REFERENCES sources(sources_id),
FOREIGN KEY (spelltypes_id) REFERENCES spelltypes(spelltypes_id)
FOREIGN KEY (spelltypes_id) REFERENCES spelltypes(spelltypes_id),
FOREIGN KEY (spelltargets_id) REFERENCES spelltargets(spelltargets_id)
);
CREATE TABLE spells_spellcomponents(
id INTEGER PRIMARY KEY,
spells_id INTEGER NOT NULL,
spellcomponents_id INTEGER NOT NULL,
FOREIGN KEY (spells_id) REFERENCES spells(spells_id),
@ -56,6 +66,7 @@ CREATE TABLE spells_traits (
id INTEGER PRIMARY KEY,
spells_id INTEGER NOT NULL,
traits_id INTEGER NOT NULL,
UNIQUE(spells_id, traits_id),
FOREIGN KEY (spells_id) REFERENCES spells(spells_id),
FOREIGN KEY (traits_id) REFERENCES traits(traits_id)
);