[V3 Setup] Overhaul backend conversion process through setup scripts (#2579)

* swap to click for setup

* Initial changes

* expose some stuff to allow for per-driver optimizations

* overwrite base config

* add red log

* add one print juuuust in case

* fix this

* thanks kowlin

* damn

* oops

* fix thing

* partial commit

* Working mongo -> json conversion, it sucks tho

* remove unused line

* Wrote initial optimized json importer

* optimized json importer

* remove useless line

* update mongo to json converter

* lets try writing the correct entry

* oops

* style fix

* add some garbage data filters going from old mongo to json

* ignore garbage data in mongov2 conversions

* simplify code a bit and add a completion message

* missed one

* Update pipfile lock

* Lock click version
This commit is contained in:
Will 2019-04-20 20:10:44 -04:00 committed by Michael H
parent ad06b0e723
commit 6c296a9a17
8 changed files with 363 additions and 97 deletions

19
Pipfile.lock generated
View File

@ -76,6 +76,13 @@
],
"version": "==3.0.4"
},
"click": {
"hashes": [
"sha256:2335065e6395b9e67ca716de5f7526736bfa6ceead690adf616d925bdc622b13",
"sha256:5b94b49521f6456670fdb30cd82a4eca9412788a93fa6dd6df72c94d5a8ff2d7"
],
"version": "==7.0"
},
"colorama": {
"hashes": [
"sha256:05eed71e2e327246ad6b38c540c4a3117230b19679b875190486ddd2d721422d",
@ -747,11 +754,11 @@
},
"tox": {
"hashes": [
"sha256:04f8f1aa05de8e76d7a266ccd14e0d665d429977cd42123bc38efa9b59964e9e",
"sha256:25ef928babe88c71e3ed3af0c464d1160b01fca2dd1870a5bb26c2dea61a17fc"
"sha256:1b166b93d2ce66bb7b253ba944d2be89e0c9d432d49eeb9da2988b4902a4684e",
"sha256:665cbdd99f5c196dd80d1d8db8c8cf5d48b1ae1f778bccd1bdf14d5aaf4ca0fc"
],
"index": "pypi",
"version": "==3.7.0"
"version": "==3.9.0"
},
"urllib3": {
"hashes": [
@ -762,10 +769,10 @@
},
"virtualenv": {
"hashes": [
"sha256:5a3ecdfbde67a4a3b3111301c4d64a5b71cf862c8c42958d30cf3253df1f29dd",
"sha256:dffd40d19ab0168c02cf936de59590a3c0f2c8c4a36f363fcf3dae18728dc94e"
"sha256:6aebaf4dd2568a0094225ebbca987859e369e3e5c22dc7d52e5406d504890417",
"sha256:984d7e607b0a5d1329425dd8845bd971b957424b5ba664729fab51ab8c11bc39"
],
"version": "==16.4.1"
"version": "==16.4.3"
},
"websockets": {
"hashes": [

View File

@ -1,6 +1,14 @@
import enum
from .red_base import IdentifierData
__all__ = ["get_driver", "IdentifierData"]
__all__ = ["get_driver", "IdentifierData", "BackendType"]
class BackendType(enum.Enum):
JSON = "JSON"
MONGO = "MongoDBV2"
MONGOV1 = "MongoDB"
def get_driver(type, *args, **kwargs):

View File

@ -1,8 +1,18 @@
import enum
from typing import Tuple
__all__ = ["BaseDriver", "IdentifierData"]
class ConfigCategory(enum.Enum):
GLOBAL = "GLOBAL"
GUILD = "GUILD"
CHANNEL = "TEXTCHANNEL"
ROLE = "ROLE"
USER = "USER"
MEMBER = "MEMBER"
class IdentifierData:
def __init__(
self,
@ -72,6 +82,9 @@ class BaseDriver:
self.cog_name = cog_name
self.unique_cog_identifier = identifier
async def has_valid_connection(self) -> bool:
raise NotImplementedError
async def get(self, identifier_data: IdentifierData):
"""
Finds the value indicate by the given identifiers.
@ -121,3 +134,75 @@ class BaseDriver:
identifier_data
"""
raise NotImplementedError
def _get_levels(self, category, custom_group_data):
if category == ConfigCategory.GLOBAL.value:
return 0
elif category in (
ConfigCategory.USER.value,
ConfigCategory.GUILD.value,
ConfigCategory.CHANNEL.value,
ConfigCategory.ROLE.value,
):
return 1
elif category == ConfigCategory.MEMBER.value:
return 2
elif category in custom_group_data:
return custom_group_data[category]
else:
raise RuntimeError(f"Cannot convert due to group: {category}")
def _split_primary_key(self, category, custom_group_data, data):
levels = self._get_levels(category, custom_group_data)
if levels == 0:
return (((), data),)
def flatten(levels_remaining, currdata, parent_key=()):
items = []
for k, v in currdata.items():
new_key = parent_key + (k,)
if levels_remaining > 1:
items.extend(flatten(levels_remaining - 1, v, new_key).items())
else:
items.append((new_key, v))
return dict(items)
ret = []
for k, v in flatten(levels, data).items():
ret.append((k, v))
return tuple(ret)
async def export_data(self, custom_group_data):
categories = [c.value for c in ConfigCategory]
categories.extend(custom_group_data.keys())
ret = []
for c in categories:
ident_data = IdentifierData(
self.unique_cog_identifier,
c,
(),
(),
custom_group_data.get(c, {}),
is_custom=c in custom_group_data,
)
try:
data = await self.get(ident_data)
except KeyError:
continue
ret.append((c, data))
return ret
async def import_data(self, cog_data, custom_group_data):
for category, all_data in cog_data:
splitted_pkey = self._split_primary_key(category, custom_group_data, all_data)
for pkey, data in splitted_pkey:
ident_data = IdentifierData(
self.unique_cog_identifier,
category,
pkey,
(),
custom_group_data.get(category, {}),
is_custom=category in custom_group_data,
)
await self.set(ident_data, data)

View File

@ -69,6 +69,9 @@ class JSON(BaseDriver):
self._load_data()
async def has_valid_connection(self) -> bool:
return True
@property
def data(self):
return _shared_datastore.get(self.cog_name)
@ -123,5 +126,29 @@ class JSON(BaseDriver):
else:
await self.jsonIO._threadsafe_save_json(self.data)
async def import_data(self, cog_data, custom_group_data):
def update_write_data(identifier_data: IdentifierData, data):
partial = self.data
idents = identifier_data.to_tuple()
for ident in idents[:-1]:
if ident not in partial:
partial[ident] = {}
partial = partial[ident]
partial[idents[-1]] = data
for category, all_data in cog_data:
splitted_pkey = self._split_primary_key(category, custom_group_data, all_data)
for pkey, data in splitted_pkey:
ident_data = IdentifierData(
self.unique_cog_identifier,
category,
pkey,
(),
custom_group_data.get(category, {}),
is_custom=category in custom_group_data,
)
update_write_data(ident_data, data)
await self.jsonIO._threadsafe_save_json(self.data)
def get_config_details(self):
return

View File

@ -49,6 +49,10 @@ class Mongo(BaseDriver):
if _conn is None:
_initialize(**kwargs)
async def has_valid_connection(self) -> bool:
# Maybe fix this?
return True
@property
def db(self) -> motor.core.Database:
"""

View File

@ -267,11 +267,11 @@ async def reset_red():
if confirm("\nDo you want to create a backup for an instance? (y/n) "):
for index, instance in instances.items():
print("\nRemoving {}...".format(index))
await create_backup(index, instance)
await remove_instance(index, instance)
await create_backup(index)
await remove_instance(index)
else:
for index, instance in instances.items():
await remove_instance(index, instance)
await remove_instance(index)
print("All instances have been removed.")

View File

@ -1,5 +1,4 @@
#!/usr/bin/env python3
import argparse
import asyncio
import json
import os
@ -8,19 +7,30 @@ import tarfile
from copy import deepcopy
from datetime import datetime as dt
from pathlib import Path
import logging
import appdirs
import click
import redbot.logging
from redbot.core.cli import confirm
from redbot.core.data_manager import (
basic_config_default,
load_basic_configuration,
instance_name,
basic_config,
cog_data_path,
core_data_path,
storage_details,
)
from redbot.core.json_io import JsonIO
from redbot.core.utils import safe_delete
from redbot.core import Config
from redbot.core.drivers import BackendType, IdentifierData
from redbot.core.drivers.red_json import JSON
conversion_log = logging.getLogger("red.converter")
config_dir = None
appdir = appdirs.AppDirs("Red-DiscordBot")
if sys.platform == "linux":
@ -36,15 +46,6 @@ except PermissionError:
config_file = config_dir / "config.json"
def parse_cli_args():
parser = argparse.ArgumentParser(description="Red - Discord Bot's instance manager (V3)")
parser.add_argument(
"--delete", "-d", help="Interactively delete an instance", action="store_true"
)
parser.add_argument("--edit", "-e", help="Interactively edit an instance", action="store_true")
return parser.parse_known_args()
def load_existing_config():
if not config_file.exists():
return {}
@ -52,6 +53,13 @@ def load_existing_config():
return JsonIO(config_file)._load_json()
instance_data = load_existing_config()
if instance_data is None:
instance_list = []
else:
instance_list = list(instance_data.keys())
def save_config(name, data, remove=False):
config = load_existing_config()
if remove and name in config:
@ -173,37 +181,138 @@ def basic_setup():
)
async def json_to_mongo(current_data_dir: Path, storage_details: dict):
from redbot.core.drivers.red_mongo import Mongo
def get_current_backend(instance) -> BackendType:
return BackendType(instance_data[instance]["STORAGE_TYPE"])
def get_target_backend(backend) -> BackendType:
if backend == "json":
return BackendType.JSON
elif backend == "mongo":
return BackendType.MONGO
async def json_to_mongov2(instance):
instance_vals = instance_data[instance]
current_data_dir = Path(instance_vals["DATA_PATH"])
load_basic_configuration(instance)
from redbot.core.drivers import red_mongo
storage_details = red_mongo.get_config_details()
core_conf = Config.get_core_conf()
new_driver = red_mongo.Mongo(cog_name="Core", identifier="0", **storage_details)
core_conf.init_custom("CUSTOM_GROUPS", 2)
custom_group_data = await core_conf.custom("CUSTOM_GROUPS").all()
curr_custom_data = custom_group_data.get("Core", {}).get("0", {})
exported_data = await core_conf.driver.export_data(curr_custom_data)
conversion_log.info("Starting Core conversion...")
await new_driver.import_data(exported_data, curr_custom_data)
conversion_log.info("Core conversion complete.")
core_data_file = current_data_dir / "core" / "settings.json"
driver = Mongo(cog_name="Core", identifier="0", **storage_details)
with core_data_file.open(mode="r") as f:
core_data = json.loads(f.read())
data = core_data.get("0", {})
for key, value in data.items():
await driver.set(key, value=value)
for p in current_data_dir.glob("cogs/**/settings.json"):
cog_name = p.parent.stem
if "." in cog_name:
# Garbage handler
continue
with p.open(mode="r") as f:
cog_data = json.load(f)
for identifier, data in cog_data.items():
driver = Mongo(cog_name, identifier, **storage_details)
for key, value in data.items():
await driver.set(key, value=value)
for identifier, all_data in cog_data.items():
try:
conf = Config.get_conf(None, int(identifier), cog_name=cog_name)
except ValueError:
continue
new_driver = red_mongo.Mongo(
cog_name=cog_name, identifier=conf.driver.unique_cog_identifier, **storage_details
)
curr_custom_data = custom_group_data.get(cog_name, {}).get(identifier, {})
exported_data = await conf.driver.export_data(curr_custom_data)
conversion_log.info(f"Converting {cog_name} with identifier {identifier}...")
await new_driver.import_data(exported_data, curr_custom_data)
conversion_log.info("Cog conversion complete.")
return storage_details
async def mongo_to_json(current_data_dir: Path, storage_details: dict):
async def mongov2_to_json(instance):
load_basic_configuration(instance)
core_path = core_data_path()
from redbot.core.drivers import red_json
core_conf = Config.get_core_conf()
new_driver = red_json.JSON(cog_name="Core", identifier="0", data_path_override=core_path)
core_conf.init_custom("CUSTOM_GROUPS", 2)
custom_group_data = await core_conf.custom("CUSTOM_GROUPS").all()
curr_custom_data = custom_group_data.get("Core", {}).get("0", {})
exported_data = await core_conf.driver.export_data(curr_custom_data)
conversion_log.info("Starting Core conversion...")
await new_driver.import_data(exported_data, curr_custom_data)
conversion_log.info("Core conversion complete.")
collection_names = await core_conf.driver.db.list_collection_names()
splitted_names = list(
filter(
lambda elem: elem[1] != "" and elem[0] != "Core",
[n.split(".") for n in collection_names],
)
)
ident_map = {} # Cogname: idents list
for cog_name, category in splitted_names:
if cog_name not in ident_map:
ident_map[cog_name] = set()
idents = await core_conf.driver.db[cog_name][category].distinct("_id.RED_uuid")
ident_map[cog_name].update(set(idents))
for cog_name, idents in ident_map.items():
for identifier in idents:
curr_custom_data = custom_group_data.get(cog_name, {}).get(identifier, {})
try:
conf = Config.get_conf(None, int(identifier), cog_name=cog_name)
except ValueError:
continue
exported_data = await conf.driver.export_data(curr_custom_data)
new_path = cog_data_path(raw_name=cog_name)
new_driver = red_json.JSON(cog_name, identifier, data_path_override=new_path)
conversion_log.info(f"Converting {cog_name} with identifier {identifier}...")
await new_driver.import_data(exported_data, curr_custom_data)
# cog_data_path(raw_name=cog_name)
conversion_log.info("Cog conversion complete.")
return {}
async def mongo_to_json(instance):
load_basic_configuration(instance)
from redbot.core.drivers.red_mongo import Mongo
m = Mongo("Core", "0", **storage_details)
m = Mongo("Core", "0", **storage_details())
db = m.db
collection_names = await db.list_collection_names()
for collection_name in collection_names:
if collection_name == "Core":
c_data_path = current_data_dir / "core"
if "." in collection_name:
# Fix for one of Zeph's problems
continue
elif collection_name == "Core":
c_data_path = core_data_path()
else:
c_data_path = current_data_dir / "cogs" / collection_name
c_data_path = cog_data_path(raw_name=collection_name)
c_data_path.mkdir(parents=True, exist_ok=True)
# Every cog name has its own collection
collection = db[collection_name]
@ -212,9 +321,16 @@ async def mongo_to_json(current_data_dir: Path, storage_details: dict):
# This means if two cogs have the same name but different identifiers, they will
# be two separate documents in the same collection
cog_id = document.pop("_id")
if not isinstance(cog_id, str):
# Another garbage data check
continue
elif not str(cog_id).isdigit():
continue
driver = JSON(collection_name, cog_id, data_path_override=c_data_path)
for key, value in document.items():
await driver.set(key, value=value)
for category, value in document.items():
ident_data = IdentifierData(str(cog_id), category, (), (), {})
await driver.set(ident_data, value=value)
return {}
async def edit_instance():
@ -257,32 +373,6 @@ async def edit_instance():
else:
default_dirs["DATA_PATH"] = str(current_data_dir.resolve())
if confirm("Would you like to change the storage type? (y/n):"):
storage = get_storage_type()
storage_dict = {1: "JSON", 2: "MongoDBV2"}
default_dirs["STORAGE_TYPE"] = storage_dict[storage]
if storage_dict.get(storage, 1) == "MongoDBV2":
from redbot.core.drivers.red_mongo import get_config_details
storage_details = get_config_details()
default_dirs["STORAGE_DETAILS"] = storage_details
if instance_data["STORAGE_TYPE"] == "JSON":
raise NotImplementedError("We cannot convert from JSON to MongoDB at this time.")
# if confirm("Would you like to import your data? (y/n) "):
# await json_to_mongo(current_data_dir, storage_details)
elif storage_dict.get(storage, 1) == "JSON":
storage_details = instance_data["STORAGE_DETAILS"]
default_dirs["STORAGE_DETAILS"] = {}
if instance_data["STORAGE_TYPE"] == "MongoDB":
if confirm("Would you like to import your data? (y/n) "):
await mongo_to_json(current_data_dir, storage_details)
elif instance_data["STORAGE_TYPE"] == "MongoDBV2":
raise NotImplementedError(
"We cannot convert from this version of MongoDB to JSON at this time."
)
if name != selected:
save_config(selected, {}, remove=True)
save_config(name, default_dirs)
@ -290,16 +380,17 @@ async def edit_instance():
print("Your basic configuration has been edited")
async def create_backup(selected, instance_data):
async def create_backup(instance):
instance_vals = instance_data[instance]
if confirm("Would you like to make a backup of the data for this instance? (y/n)"):
load_basic_configuration(selected)
if instance_data["STORAGE_TYPE"] == "MongoDB":
await mongo_to_json(instance_data["DATA_PATH"], instance_data["STORAGE_DETAILS"])
load_basic_configuration(instance)
if instance_vals["STORAGE_TYPE"] == "MongoDB":
await mongo_to_json(instance)
print("Backing up the instance's data...")
backup_filename = "redv3-{}-{}.tar.gz".format(
selected, dt.utcnow().strftime("%Y-%m-%d %H-%M-%S")
instance, dt.utcnow().strftime("%Y-%m-%d %H-%M-%S")
)
pth = Path(instance_data["DATA_PATH"])
pth = Path(instance_vals["DATA_PATH"])
if pth.exists():
backup_pth = pth.home()
backup_file = backup_pth / backup_filename
@ -321,24 +412,27 @@ async def create_backup(selected, instance_data):
repo_filename = pth / "cogs" / "RepoManager" / "repos.json"
with open(str(repo_filename), "w") as f:
f.write(json.dumps(repo_output, indent=4))
instance_data = {instance_name: basic_config}
instance_vals = {instance_name: basic_config}
instance_file = pth / "instance.json"
with open(str(instance_file), "w") as instance_out:
instance_out.write(json.dumps(instance_data, indent=4))
instance_out.write(json.dumps(instance_vals, indent=4))
for f in pth.glob("**/*"):
if not any(ex in str(f) for ex in exclusions):
to_backup.append(f)
with tarfile.open(str(backup_file), "w:gz") as tar:
for f in to_backup:
tar.add(str(f), recursive=False)
print("A backup of {} has been made. It is at {}".format(selected, backup_file))
print("A backup of {} has been made. It is at {}".format(instance, backup_file))
async def remove_instance(selected, instance_data):
if instance_data["STORAGE_TYPE"] == "MongoDB":
async def remove_instance(instance):
await create_backup(instance)
instance_vals = instance_data[instance]
if instance_vals["STORAGE_TYPE"] == "MongoDB":
from redbot.core.drivers.red_mongo import Mongo
m = Mongo("Core", **instance_data["STORAGE_DETAILS"])
m = Mongo("Core", **instance_vals["STORAGE_DETAILS"])
db = m.db
collections = await db.collection_names(include_system_collections=False)
for name in collections:
@ -347,12 +441,11 @@ async def remove_instance(selected, instance_data):
else:
pth = Path(instance_data["DATA_PATH"])
safe_delete(pth)
save_config(selected, {}, remove=True)
print("The instance {} has been removed\n".format(selected))
save_config(instance, {}, remove=True)
print("The instance {} has been removed\n".format(instance))
async def remove_instance_interaction():
instance_list = load_existing_config()
if not instance_list:
print("No instances have been set up!")
return
@ -361,35 +454,76 @@ async def remove_instance_interaction():
"You have chosen to remove an instance. The following "
"is a list of instances that currently exist:\n"
)
for instance in instance_list.keys():
for instance in instance_data.keys():
print("{}\n".format(instance))
print("Please select one of the above by entering its name")
selected = input("> ")
if selected not in instance_list.keys():
if selected not in instance_data.keys():
print("That isn't a valid instance!")
return
instance_data = instance_list[selected]
await create_backup(selected, instance_data)
await remove_instance(selected, instance_data)
await create_backup(selected)
await remove_instance(selected)
def main():
args, _ = parse_cli_args()
if args.delete:
loop = asyncio.get_event_loop()
loop.run_until_complete(remove_instance_interaction())
elif args.edit:
loop = asyncio.get_event_loop()
loop.run_until_complete(edit_instance())
else:
@click.group(invoke_without_command=True)
@click.option("--debug", type=bool)
@click.pass_context
def cli(ctx, debug):
level = logging.DEBUG if debug else logging.INFO
redbot.logging.init_logging(level=level, location=Path.cwd() / "red_setup_logs")
if ctx.invoked_subcommand is None:
basic_setup()
@cli.command()
@click.argument("instance", type=click.Choice(instance_list))
def delete(instance):
loop = asyncio.get_event_loop()
loop.run_until_complete(remove_instance(instance))
@cli.command()
@click.argument("instance", type=click.Choice(instance_list))
@click.argument("backend", type=click.Choice(["json", "mongo"]))
def convert(instance, backend):
current_backend = get_current_backend(instance)
target = get_target_backend(backend)
default_dirs = deepcopy(basic_config_default)
default_dirs["DATA_PATH"] = str(Path(instance_data[instance]["DATA_PATH"]))
loop = asyncio.get_event_loop()
new_storage_details = None
if current_backend == BackendType.MONGOV1:
if target == BackendType.MONGO:
raise RuntimeError(
"Please see conversion docs for updating to the latest mongo version."
)
elif target == BackendType.JSON:
new_storage_details = loop.run_until_complete(mongo_to_json(instance))
elif current_backend == BackendType.JSON:
if target == BackendType.MONGO:
new_storage_details = loop.run_until_complete(json_to_mongov2(instance))
elif current_backend == BackendType.MONGO:
if target == BackendType.JSON:
new_storage_details = loop.run_until_complete(mongov2_to_json(instance))
if new_storage_details is not None:
default_dirs["STORAGE_TYPE"] = target.value
default_dirs["STORAGE_DETAILS"] = new_storage_details
save_config(instance, default_dirs)
conversion_log.info(f"Conversion to {target} complete.")
else:
conversion_log.info(f"Cannot convert {current_backend} to {target} at this time.")
if __name__ == "__main__":
try:
main()
cli()
except KeyboardInterrupt:
print("Exiting...")
else:

View File

@ -32,6 +32,7 @@ install_requires =
async-timeout==3.0.1
attrs==18.2.0
chardet==3.0.4
click==7.0
colorama==0.4.1
distro==1.4.0; sys_platform == "linux"
fuzzywuzzy==0.17.0
@ -86,7 +87,7 @@ test =
[options.entry_points]
console_scripts =
redbot=redbot.__main__:main
redbot-setup=redbot.setup:main
redbot-setup=redbot.setup:cli
redbot-launcher=redbot.launcher:main
pytest11 =
red-discordbot=redbot.pytest