remoll/coss_archiving
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Bug fixes, makefile for launch

Browse Source
This commit is contained in:
Remy Moll 2022-12-09 11:19:45 +01:00
parent 24b3bc3b51
commit 647944d23c
25 changed files with 321 additions and 300 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

87
Makefile Normal file
View File

@ -0,0 +1,87 @@
include config/local.env
export
build:
@echo "Building..."
docker compose build $(flags)
down:
@echo "Stopping containers..."
docker compose down -t 0 --volumes
# Variables specific to debug
debug: export DEBUG=true
debug: export HEADFULL=true
debug: export ENTRYPOINT=/bin/bash
debug: export CODE=./
debug:
@echo "Running in debug mode..."
docker compose up -d geckodriver
docker compose run -it --service-ports $(target) $(flags) || true
make down
production: export DEBUG=false
production:
@echo "Running in production mode..."
docker compose run -it --service-ports $(target) $(flags) || true
make down
nas_sync:
@echo "Syncing NAS..."
SYNC_FOLDER=$(folder) docker compose run -it nas_sync $(flags) || true
docker compose down
docker container prune -f
make down
## Misc:
edit_profile: export CODE=./
edit_profile: export HEADFULL=true
edit_profile:
@echo "Editing profile..."
docker compose up -d geckodriver
sleep 5
docker compose exec geckodriver /bin/bash /code/geckodriver/edit_profile.sh || true
# runs inside the container
make down
db_interface:
docker create \
--name pgadmin \
-p 8080:80 \
-e 'PGADMIN_DEFAULT_EMAIL=${UNAME}@test.com' \
-e 'PGADMIN_DEFAULT_PASSWORD=password' \
-e 'PGADMIN_CONFIG_ENHANCED_COOKIE_PROTECTION=True' \
-e 'PGADMIN_CONFIG_LOGIN_BANNER="Authorised users only!"' \
dpage/pgadmin4
docker start pgadmin
sleep 5
# TODO auto add the server to the list displayed in the browser
# docker exec pgadmin sh -c "echo ${SERVER_DATA} > /tmp/servers.json"
# docker exec pgadmin sh -c "/venv/bin/python setup.py --load-servers /tmp/servers.json --user remy@test.com"
@echo "Go to http://localhost:8080 to access the database interface"
@echo "Username: ${UNAME}@test.com"
@echo "Password: password"
@echo "Hit any key to stop (not ctrl+c)"
read STOP
docker stop pgadmin
docker rm pgadmin
logs:
docker compose logs -f $(target) $(flags)
make down

2
README.md
View File

@ -124,4 +124,4 @@ I use `rsync`. Mounting the NAS locally, I navigate to the location of the local
`rsync -Razq --no-perms --no-owner --no-group --temp-dir=/tmp --progress --log-file=rsync.log <local folder>/ "<remote>"`
where `<remote>` is the location where the NAS is mounted. (options:`R` - relative paths , `a` - archive mode (multiple actions), `z` - ??, `q` - quiet. We also don't copy most of the metadata and we keep a log of the transfers.)
You can also use your OS' native copy option and select *de not overwrite*. This should only copy the missing files, significantly speeding up the operation.
You can also use your OS' native copy option and select *do not overwrite*. This should only copy the missing files, significantly speeding up the operation.

8
config/README.md
View File

@ -1,8 +0,0 @@
## Configuration: example
The files inside this directory (not the ones in `env/`) are a sample of the required configuration.
Please create a copy of these files under `<location of downloads>/config/...`.
> Note:
>
> Some of the fields are blank, please fill them in as needed.

37
config/container.yaml Normal file
View File

@ -0,0 +1,37 @@
mail:
smtp_server: smtp.ethz.ch
port: 587
sender: "****************"
recipient: "****************"
uname: "****************"
password: "************"
slack:
bot_id: U02MR1R8UJH
archive_id: C02MM7YG1V4
debug_id: C02NM2H9J5Q
api_wait_time: 90
auth_token: "****************"
app_token: "****************"
database:
debug_db: /app/containerdata/debug/downloads.db
db_printout: /app/containerdata/backups
production_db_name: coss_archiving
production_user_name: "ca_rw"
production_password: "****************"
## user_name: ca_ro
## password: "****************"
downloads:
local_storage_path: /app/containerdata/files
debug_storage_path: /app/containerdata/debug/
default_download_path: /app/containerdata/tmp
remote_storage_path: /helbing_support/Archiving-Pipeline
browser_profile_path: /app/containerdata/dependencies/news_fetch.profile
# please keep this exact name
browser_print_delay: 3

7
config/db.config.ini
View File

@ -1,7 +0,0 @@
[DATABASE]
db_name: coss_archiving
user_name: ****************
password: ****************
## user_name: ca_ro
## password: #TK5cLxA^YyoxWjR6

18
config/local.env Normal file
View File

@ -0,0 +1,18 @@
CONTAINER_DATA=***********
UNAME=***********
U_ID=***********
DB_HOST=***********
OPENCONNECT_URL=***********
OPENCONNECT_USER=***********
OPENCONNECT_PASSWORD=***********
OPENCONNECT_OPTIONS=--authgroup student-net
NAS_HOST=***********
NAS_PATH=/gess_coss_1/helbing_support/Archiving-Pipeline
NAS_USERNAME=***********
NAS_PASSWORD=***********
# Special characters like # need to be escaped (write: \#)

3
config/nas_login.config
View File

@ -1,3 +0,0 @@
user=remoll
domain=D
password=****************

12
config/nas_sync.config
View File

@ -1,12 +0,0 @@
settings {
logfile = "/tmp/lsyncd.log",
statusFile = "/tmp/lsyncd.status",
nodaemon = true,
}
sync {
default.rsync,
source = "/sync/local_files",
target = "/sync/remote_files",
init = false,
}

31
config/news_fetch.config.ini
View File

@ -1,31 +0,0 @@
[MAIL]
smtp_server: smtp.ethz.ch
port: 587
sender: ****************
recipient: ****************
uname: ****************
password: ****************
[SLACK]
bot_id: U02MR1R8UJH
archive_id: C02MM7YG1V4
debug_id: C02NM2H9J5Q
api_wait_time: 90
auth_token: ****************
app_token: ****************
[DATABASE]
download_db_debug: /app/containerdata/debug/downloads.db
db_printout: /app/containerdata/backups
[DOWNLOADS]
local_storage_path: /app/containerdata/files
debug_storage_path: /app/containerdata/debug/
default_download_path: /app/containerdata/tmp
remote_storage_path: /helbing_support/Archiving-Pipeline
browser_profile_path: /app/containerdata/dependencies/news_fetch.profile
# please keep this exact name
browser_print_delay: 3

4
config/vpn.config
View File

@ -1,4 +0,0 @@
OPENCONNECT_URL=sslvpn.ethz.ch/student-net
OPENCONNECT_USER=****************
OPENCONNECT_PASSWORD=****************
OPENCONNECT_OPTIONS=--authgroup student-net

58
docker-compose.yaml
View File

@ -4,33 +4,17 @@ services:
vpn: # Creates a connection behind the ETH Firewall to access NAS and Postgres
image: wazum/openconnect-proxy:latest
env_file:
- ${CONTAINER_DATA}/config/vpn.config
environment:
- OPENCONNECT_URL=${OPENCONNECT_URL}
- OPENCONNECT_USER=${OPENCONNECT_USER}
- OPENCONNECT_PASSWORD=${OPENCONNECT_PASSWORD}
- OPENCONNECT_OPTIONS=${OPENCONNECT_OPTIONS}
cap_add:
- NET_ADMIN
volumes:
- /dev/net/tun:/dev/net/tun
# alternative to cap_add & volumes: specify privileged: true
expose: ["5432"] # exposed here because db_passhtrough uses this network. See below for more details
nas_sync: # Syncs locally downloaded files with the NAS-share on nas22.ethz.ch/...
depends_on:
- vpn
network_mode: "service:vpn" # used to establish a connection to the SMB server from inside ETH network
build: nas_sync # local folder to build
image: nas_sync:latest
cap_add: # capabilities needed for mounting the SMB share
- SYS_ADMIN
- DAC_READ_SEARCH
volumes:
- ${CONTAINER_DATA}/files:/sync/local_files
- ${CONTAINER_DATA}/config/nas_sync.config:/sync/nas_sync.config
- ${CONTAINER_DATA}/config/nas_login.config:/sync/nas_login.config
command:
- nas22.ethz.ch/gess_coss_1/helbing_support/Archiving-Pipeline # first command is the target mount path
- lsyncd
- /sync/nas_sync.config
geckodriver: # separate docker container for pdf-download. This hugely improves stability (and creates shorter build times for the containers)
@ -40,7 +24,6 @@ services:
- START_VNC=${HEADFULL-false} # as opposed to headless, used when requiring supervision (eg. for websites that crash)
- START_XVFB=${HEADFULL-false}
- SE_VNC_NO_PASSWORD=1
# - SE_OPTS="--profile /user_data/news_fetch.profile.firefox"
volumes:
- ${CONTAINER_DATA}/dependencies:/firefox_profile/
- ${CODE:-/dev/null}:/code
@ -53,7 +36,7 @@ services:
db_passthrough: # Allows a container on the local network to connect to a service (here postgres) through the vpn
network_mode: "service:vpn"
image: alpine/socat:latest
command: ["tcp-listen:5432,reuseaddr,fork", "tcp-connect:id-hdb-psgr-cp48.ethz.ch:5432"]
command: ["tcp-listen:5432,reuseaddr,fork", "tcp-connect:${DB_HOST}:5432"]
# expose: ["5432"] We would want this passthrough to expose its ports to the other containers
# BUT since it uses the same network as the vpn-service, it can't expose ports on its own. 5432 is therefore exposed under service.vpn.expose
@ -62,14 +45,14 @@ services:
build: news_fetch
image: news_fetch:latest
depends_on: # when using docker compose run news_fetch, the dependencies are started as well
- nas_sync
- geckodriver
- db_passthrough
volumes:
- ${CONTAINER_DATA}:/app/containerdata # always set
- ./config/container.yaml:/app/config.yaml
- ${CODE:-/dev/null}:/code # not set in prod, defaults to /dev/null
environment:
- CONFIG_FILE=/app/config.yaml
- DEBUG=${DEBUG}
- UNAME=${UNAME}
user: ${U_ID}:${U_ID} # since the app writes files to the local filesystem, it must be run as the current user
@ -86,10 +69,33 @@ services:
- db_passthrough
volumes:
- ${CONTAINER_DATA}:/app/containerdata # always set
- ./config/container.yaml:/app/config.yaml
- ${CODE:-/dev/null}:/code # not set in prod, defaults to /dev/null
environment:
- CONFIG_FILE=/app/config.yaml
- UNAME=${UNAME}
ports:
- "8080:80" # 80 inside container
entrypoint: ${ENTRYPOINT:-python app.py} # by default launch workers as defined in the Dockerfile
tty: true
nas_sync:
image: alpine:latest
volumes:
- ${CONTAINER_DATA}/files:/sync/local_files
- coss_smb_share:/sync/remote_files
command:
- /bin/sh
- -c
- |
apk add rsync
rsync -av --no-perms --no-owner --no-group --progress /sync/local_files/${SYNC_FOLDER}/ /sync/remote_files/${SYNC_FOLDER} -n
volumes:
coss_smb_share:
driver: local
driver_opts:
type: cifs
o: "addr=${NAS_HOST},nounix,file_mode=0777,dir_mode=0777,domain=D,username=${NAS_USERNAME},password=${NAS_PASSWORD}"
device: //${NAS_HOST}${NAS_PATH}

70
launch
View File

@ -1,70 +0,0 @@
#!/bin/bash
set -e
set -o ignoreeof
echo "Bash script launching COSS_ARCHIVING..."
# CHANGE ME ONCE!
export CONTAINER_DATA=/mnt/media/@Bulk/COSS/Downloads/coss_archiving
export UNAME=remy
export U_ID=1000
### Main use cases ###
if [[ $1 == "debug" ]]
then
export DEBUG=true
export HEADFULL=true
export CODE=./
export ENTRYPOINT=/bin/bash
# since service ports does not open ports on implicitly started containers, also start geckodriver:
docker compose up -d geckodriver
elif [[ $1 == "production" ]]
then
export DEBUG=false
elif [[ $1 == "build" ]]
then
export DEBUG=false
shift
docker compose build "$@"
exit 0
### Manual Shutdown ###
elif [[ $1 == "down" ]]
then
docker compose down -t 0
exit 0
### Edge cases -> for firefox ###
elif [[ $1 == "edit_profile" ]]
then
export CODE=./
export HEADFULL=true
docker compose up -d geckodriver
sleep 5
docker compose exec geckodriver /bin/bash /code/geckodriver/edit_profile.sh # inside the container
docker compose down -t 0
### Fallback ####
else
echo "Please specify the execution mode (debug/production/build/edit_profile/down) as the first argument"
exit 1
fi
shift # consumes the variable set in $1 so that $@ only contains the remaining arguments
docker compose run -it --service-ports "$@"
echo "Docker run finished, shutting down containers..."
docker compose down -t 0
echo "Bye!"

10
manual/gather_media_files.py
View File

@ -46,7 +46,13 @@ def fetch():
def show():
for a in runner.models.ArticleDownload.select():
print(f"URL: {a.article_url} \nARCHIVE_URL: {a.archive_url} \nFILE_NAME: {a.file_name}")
print(f"""
URL: {a.article_url}
ARCHIVE_URL: {a.archive_url}
ARTICLE_SOURCE: {a.source_name}
FILE_NAME: {a.file_name}
""")
if __name__ == "__main__":
logger.info("Overwriting production values for single time media-fetch")
@ -55,7 +61,7 @@ if __name__ == "__main__":
runner.configuration.models.set_db(
runner.configuration.SqliteDatabase("../.dev/media_downloads.db")
)
runner.configuration.main_config["DOWNLOADS"]["local_storage_path"] = "../.dev/"
runner.configuration.main_config["downloads"]["local_storage_path"] = "../.dev/"
if len(sys.argv) == 1: # no additional arguments

9
nas_sync/Dockerfile
View File

@ -1,9 +0,0 @@
FROM bash:latest
# alpine with bash instead of sh
ENV TZ=Europe/Berlin
RUN apk add lsyncd cifs-utils rsync
RUN mkdir -p /sync/remote_files
COPY entrypoint.sh /sync/entrypoint.sh
ENTRYPOINT ["bash", "/sync/entrypoint.sh"]

10
nas_sync/entrypoint.sh
View File

@ -1,10 +0,0 @@
#!/bin/bash
set -e
sleep 5 # waits for the vpn to have an established connection
echo "Starting NAS sync"
mount -t cifs "//$1" -o credentials=/sync/nas_login.config /sync/remote_files
echo "Successfully mounted SAMBA remote: $1 --> /sync/remote_files"
shift # consumes the variable set in $1 so tat $@ only contains the remaining arguments
exec "$@"

3
news_check/requirements.txt
View File

@ -1,4 +1,5 @@
flask
peewee
markdown
psycopg2
psycopg2
pyyaml

15
news_check/server/configuration.py
View File

@ -1,17 +1,16 @@
from peewee import PostgresqlDatabase
import configparser
import time
import yaml
import os
main_config = configparser.ConfigParser()
main_config.read("/app/containerdata/config/news_fetch.config.ini")
config_location = os.getenv("CONFIG_FILE")
with open(config_location, "r") as f:
config = yaml.safe_load(f)
db_config = configparser.ConfigParser()
db_config.read("/app/containerdata/config/db.config.ini")
cred = db_config["DATABASE"]
cred = config["database"]
time.sleep(10) # wait for the vpn to connect (can't use a healthcheck because there is no depends_on)
db = PostgresqlDatabase(
cred["db_name"], user=cred["user_name"], password=cred["password"], host="vpn", port=5432
cred["production_db_name"], user=cred["production_user_name"], password=cred["production_password"], host="vpn", port=5432
)
import models

8
news_check/server/models.py
View File

@ -6,7 +6,7 @@ import os
import datetime
import configuration
config = configuration.main_config["DOWNLOADS"]
downloads_config = configuration.config["downloads"]
# set the nature of the db at runtime
download_db = DatabaseProxy()
@ -34,14 +34,14 @@ class ArticleDownload(DownloadBaseModel):
file_name = TextField(default = '')
@property
def save_path(self):
return f"{config['local_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/"
return f"{downloads_config['local_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/"
@property
def fname_nas(self, file_name=""):
if self.download_date:
if file_name:
return f"NAS: {config['remote_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/{file_name}"
return f"NAS: {downloads_config['remote_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/{file_name}"
else: # return the self. name
return f"NAS: {config['remote_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/{self.file_name}"
return f"NAS: {downloads_config['remote_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/{self.file_name}"
else:
return None

27
news_fetch/configuration.py
View File

@ -1,9 +1,7 @@
import os
import configparser
import logging
import time
# import shutil
# from datetime import datetime
import os
import logging
import yaml
from peewee import SqliteDatabase, PostgresqlDatabase
from rich.logging import RichHandler
@ -19,22 +17,21 @@ logger = logging.getLogger(__name__)
# load config file containing constants and secrets
main_config = configparser.ConfigParser()
main_config.read("/app/containerdata/config/news_fetch.config.ini")
db_config = configparser.ConfigParser()
db_config.read("/app/containerdata/config/db.config.ini")
config_location = os.getenv("CONFIG_FILE")
with open(config_location, "r") as f:
config = yaml.safe_load(f)
# DEBUG MODE:
if os.getenv("DEBUG", "false") == "true":
logger.warning("Found 'DEBUG=true', setting up dummy databases")
main_config["SLACK"]["archive_id"] = main_config["SLACK"]["debug_id"]
main_config["MAIL"]["recipient"] = main_config["MAIL"]["sender"]
main_config["DOWNLOADS"]["local_storage_path"] = main_config["DOWNLOADS"]["debug_storage_path"]
config["slack"]["archive_id"] = config["slack"]["debug_id"]
config["mail"]["recipient"] = config["mail"]["sender"]
config["downloads"]["local_storage_path"] = config["downloads"]["debug_storage_path"]
download_db = SqliteDatabase(
main_config["DATABASE"]["download_db_debug"],
config["database"]["debug_db"],
pragmas = {'journal_mode': 'wal'} # mutliple threads can read at once
)
@ -43,9 +40,9 @@ else:
logger.warning("Found 'DEBUG=false' and running on production databases, I hope you know what you're doing...")
time.sleep(10) # wait for the vpn to connect (can't use a healthcheck because there is no depends_on)
cred = db_config["DATABASE"]
cred = config["database"]
download_db = PostgresqlDatabase(
cred["db_name"], user=cred["user_name"], password=cred["password"], host="vpn", port=5432
cred["production_db_name"], user=cred["production_user_name"], password=cred["production_password"], host="vpn", port=5432
)
# TODO Reimplement backup/printout
# logger.info("Backing up databases")

1
news_fetch/requirements.txt
View File

@ -10,3 +10,4 @@ markdown
rich
psycopg2
unidecode
pyyaml

20
news_fetch/utils_mail/runner.py
View File

@ -7,16 +7,20 @@ import logging
import configuration
logger = logging.getLogger(__name__)
config = configuration.main_config["MAIL"]
mail_config = configuration.config["mail"]
def send(article_model):
mail = MIMEMultipart()
mail['Subject'] = "{} -- {}".format(article_model.source_name, article_model.title)
mail['From'] = config["sender"]
mail['To'] = config["recipient"]
msg, files = article_model.mail_info() # this is html
mail['From'] = mail_config["sender"]
mail['To'] = mail_config["recipient"]
try:
msg, files = article_model.mail_info() # this is html
except: # Raised by model if article has no associated file
logger.info("Skipping mail sending")
return
content = MIMEText(msg, "html")
mail.attach(content)
@ -29,14 +33,14 @@ def send(article_model):
try:
try:
smtp = smtplib.SMTP(config["smtp_server"], config["port"])
smtp = smtplib.SMTP(mail_config["smtp_server"], mail_config["port"])
except ConnectionRefusedError:
logger.error("Server refused connection. Is this an error on your side?")
return False
smtp.starttls()
smtp.login(config["uname"], config["password"])
smtp.sendmail(config["sender"], config["recipient"], mail.as_string())
smtp.login(mail_config["uname"], mail_config["password"])
smtp.sendmail(mail_config["sender"], mail_config["recipient"], mail.as_string())
smtp.quit()
logger.info("Mail successfully sent.")
except smtplib.SMTPException as e:

20
news_fetch/utils_slack/runner.py
View File

@ -7,7 +7,7 @@ import re
import time
import configuration
config = configuration.main_config["SLACK"]
slack_config = configuration.config["slack"]
models = configuration.models
class MessageIsUnwanted(Exception):
@ -61,7 +61,7 @@ class Message:
@property
def is_by_human(self):
return self.user.user_id != config["bot_id"]
return self.user.user_id != slack_config["bot_id"]
@property
@ -87,7 +87,7 @@ class BotApp(App):
def say_substitute(self, *args, **kwargs):
self.client.chat_postMessage(
channel=config["archive_id"],
channel=slack_config["archive_id"],
text=" - ".join(args),
**kwargs
)
@ -101,7 +101,7 @@ class BotApp(App):
last_ts = presaved.slack_ts_full
result = self.client.conversations_history(
channel=config["archive_id"],
channel=slack_config["archive_id"],
oldest=last_ts
)
@ -116,7 +116,7 @@ class BotApp(App):
while refetch: # we have not actually fetched them all
try:
result = self.client.conversations_history(
channel = config["archive_id"],
channel = slack_config["archive_id"],
cursor = result["response_metadata"]["next_cursor"],
oldest = last_ts
) # fetches 100 messages, older than the [-1](=oldest) element of new_fetches
@ -126,8 +126,8 @@ class BotApp(App):
for m in new_messages:
return_messages.append(Message(m))
except SlackApiError: # Most likely a rate-limit
self.logger.error("Error while fetching channel messages. (likely rate limit) Retrying in {} seconds...".format(config["api_wait_time"]))
time.sleep(config["api_wait_time"])
self.logger.error("Error while fetching channel messages. (likely rate limit) Retrying in {} seconds...".format(slack_config["api_wait_time"]))
time.sleep(slack_config["api_wait_time"])
refetch = True
self.logger.info(f"Fetched {len(return_messages)} new channel messages.")
@ -181,7 +181,7 @@ class BotRunner():
"""Stupid encapsulation so that we can apply the slack decorators to the BotApp"""
def __init__(self, callback, *args, **kwargs) -> None:
self.bot_worker = BotApp(callback, token=config["auth_token"])
self.bot_worker = BotApp(callback, token=slack_config["auth_token"])
@self.bot_worker.event(event="message", matchers=[is_message_in_archiving])
def handle_incoming_message(message, say):
@ -195,7 +195,7 @@ class BotRunner():
def handle_all_other_reactions(event, say):
self.logger.log("Ignoring slack event that isn't a message")
self.handler = SocketModeHandler(self.bot_worker, config["app_token"])
self.handler = SocketModeHandler(self.bot_worker, slack_config["app_token"])
def start(self):
@ -215,5 +215,5 @@ class BotRunner():
def is_message_in_archiving(message) -> bool:
return message["channel"] == config["archive_id"]
return message["channel"] == slack_config["archive_id"]

47
news_fetch/utils_storage/models.py
View File

@ -8,8 +8,7 @@ import configuration
import datetime
from . import helpers
config = configuration.main_config["DOWNLOADS"]
slack_config = configuration.main_config["SLACK"]
downloads_config = configuration.config["downloads"]
FILE_SIZE_THRESHOLD = 15 * 1024 * 1024 # 15MB
@ -34,7 +33,8 @@ class ArticleDownload(DownloadBaseModel):
def is_title_bad(self): # add incrementally
return "PUR-Abo" in self.title \
or "Redirecting" in self.title \
or "Error while running fetch" in self.title
or "Error while running fetch" in self.title \
or self.title == ""
summary = TextField(default = '')
source_name = CharField(default = '')
@ -44,14 +44,14 @@ class ArticleDownload(DownloadBaseModel):
file_name = TextField(default = '')
@property
def save_path(self):
return f"{config['local_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/"
return f"{downloads_config['local_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/"
@property
def fname_nas(self, file_name=""):
if self.download_date:
if file_name:
return f"NAS: {config['remote_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/{file_name}"
return f"NAS: {downloads_config['remote_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/{file_name}"
else: # return the self. name
return f"NAS: {config['remote_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/{self.file_name}"
return f"NAS: {downloads_config['remote_storage_path']}/{self.download_date.year}/{self.download_date.strftime('%B')}/{self.file_name}"
else:
return None
@property
@ -102,18 +102,22 @@ class ArticleDownload(DownloadBaseModel):
answer_files = []
# displays the summary in a blockquote
status = self.file_status
if status == 1: # file_name was empty
return None # there has been an error do not send any message
elif status == 2: # no file found at specified location
answer_text += f"*{self.title}*\n{summary}\nFilename: {self.file_name}"
elif status == 3: # file found but deemed too big
location = f"File not sent directly. Location on NAS:\n`{self.fname_nas}`"
answer_text += f"*{self.title}*\n{summary}\n{location}"
else: # everything nominal
try:
self.ensure_file_present()
answer_text += f"*{self.title}*\n{summary}"
answer_files.append(self.save_path + self.file_name)
except Exception as e:
msg = e.args[0]
logger.error(f"Article {self} has file-issues: {msg}")
if "file too big" in msg:
location = f"File too big to send directly. Location on NAS:\n`{self.fname_nas}`"
answer_text += f"*{self.title}*\n{summary}\n{location}"
else: # file not found, or filename not set
raise e
# reraise the exception, so that the caller can handle it
# then the related files
if self.related:
rel_text = "Related files on NAS:"
@ -144,19 +148,14 @@ class ArticleDownload(DownloadBaseModel):
related_file_name = r
)
@property
def file_status(self):
"""0 = file exists, 1 = no file name!, 2 = file does not exit,3 = file exists but is too large"""
def ensure_file_present(self):
if not self.file_name:
logger.error(f"Article {self} has no filename!")
return 2
raise Exception("no filename")
file_path_abs = self.save_path + self.file_name
if not os.path.exists(file_path_abs):
logger.error(f"Article {self} has a filename, but the file does not exist at that location!")
return 2
raise Exception("file not found")
if (os.path.splitext(file_path_abs)[1] != ".pdf") or (os.path.getsize(file_path_abs) > FILE_SIZE_THRESHOLD):
logger.warning(f"Article {self} has a file that exceeds the file size limit.")
return 3
raise Exception("file too big")

100
news_fetch/utils_worker/download/browser.py
View File

@ -11,7 +11,7 @@ from selenium import webdriver
import configuration
config = configuration.main_config["DOWNLOADS"]
download_config = configuration.config["downloads"]
def driver_running(f):
def wrapper(*args, **kwargs):
@ -66,74 +66,88 @@ class PDFDownloader:
@driver_running
def download(self, article_object):
sleep_time = int(config["browser_print_delay"])
url = article_object.article_url
if url[-4:] == ".pdf": # calling the ususal pdf generation would not yield a nice pdf, just download it directly
self.logger.info("Downloading existing pdf")
success = self.get_exisiting_pdf(article_object)
# get a page title if required
if article_object.is_title_bad:
article_object.title = self.driver.title.replace(".pdf", "") # some titles end with .pdf
# will be propagated to the saved file (dst) as well
else:
success = self.get_new_pdf(article_object)
if not success:
self.logger.error("Download failed")
# TODO: need to reset the file name to empty?
return article_object # changes to this are saved later by the external caller
def get_exisiting_pdf(self, article_object):
# get a better page title if required
if article_object.is_title_bad:
article_object.title = article_object.article_url.split("/")[-1].split(".pdf")[0]
try:
self.driver.get(url)
r = requests.get(article_object.article_url)
bytes = r.content
except:
return False
return self.write_pdf(bytes, article_object)
def get_new_pdf(self, article_object):
sleep_time = int(download_config["browser_print_delay"])
try:
self.driver.get(article_object.article_url)
except Exception as e:
self.logger.critical("Selenium .get(url) failed with error {}".format(e))
self.finish()
return article_object # without changes
return False
time.sleep(sleep_time)
# leave the page time to do any funky business
# in the mean time, get a page title if required
if article_object.is_title_bad:
article_object.title = self.driver.title.replace(".pdf", "") # some titles end with .pdf
# will be propagated to the saved file (dst) as well
article_object.title = self.driver.title
try:
result = self.driver.print_page()
bytes = base64.b64decode(result, validate=True)
except:
self.logger.error("Failed, probably because the driver went extinct.")
return False
return self.write_pdf(bytes, article_object)
def get_file_destination(self, article_object):
fname = article_object.fname_template
fname = ensure_unique(article_object.save_path, fname)
dst = os.path.join(article_object.save_path, fname)
return dst, fname
if url[-4:] == ".pdf": # calling the ususal pdf generation would not yield a nice pdf, just download it directly
success = self.get_exisiting_pdf(url, dst)
else:
success = self.get_new_pdf(dst)
if success:
article_object.file_name = fname
else:
article_object.file_name = ""
return article_object # this change is saved later by the external caller
def get_exisiting_pdf(self, url, dst):
try:
r = requests.get(url)
bytes = r.content
except:
return False
return self.get_new_pdf(dst, other_bytes=bytes)
def get_new_pdf(self, dst, other_bytes=None):
def write_pdf(self, content, article_object):
dst, fname = self.get_file_destination(article_object)
os.makedirs(os.path.dirname(dst), exist_ok=True)
if other_bytes is None:
try:
result = self.driver.print_page()
bytes = base64.b64decode(result, validate=True)
except:
self.logger.error("Failed, probably because the driver went extinct.")
return False
else:
bytes = other_bytes
try:
with open(dst, "wb+") as f:
f.write(bytes)
f.write(content)
article_object.file_name = fname
return True
except Exception as e:
self.logger.error(f"Failed, because of FS-operation: {e}")
return False
def create_tmp_profile(self, full_profile_path: Path = Path(config["browser_profile_path"])) -> Path:
def create_tmp_profile(self, full_profile_path: Path = Path(download_config["browser_profile_path"])) -> Path:
reduced_profile_path = Path(f"/tmp/firefox_profile_{uuid.uuid4()}")
os.mkdir(reduced_profile_path)
# copy needed directories

14
news_fetch/utils_worker/download/youtube.py
View File

@ -1,10 +1,11 @@
import youtube_dl
import os
import logging
import configuration
download_config = configuration.config["downloads"]
logger = logging.getLogger(__name__)
class MyLogger(object):
def debug(self, msg): pass
def warning(self, msg): pass
@ -19,7 +20,6 @@ class YouTubeDownloader:
def post_download_hook(self, ret_code):
# print(ret_code)
if ret_code['status'] == 'finished':
file_loc = ret_code["filename"]
fname = os.path.basename(file_loc)
@ -35,9 +35,11 @@ class YouTubeDownloader:
ydl_opts = {
'format': 'best[height<=720]',
'outtmpl': f"{file_path}.%(ext)s", # basically the filename from the object, but with a custom extension depending on the download
'logger': MyLogger(),
'logger': MyLogger(), # supress verbosity
'progress_hooks': [self.post_download_hook],
'updatetime': False
'updatetime': False,
# File is also used by firefox so make sure to not write to it!
# youtube dl apparenlty does not support cookies.sqlite and the documentation is not clear on how to use cookies.txt
}
try:
with youtube_dl.YoutubeDL(ydl_opts) as ydl:
@ -46,5 +48,9 @@ class YouTubeDownloader:
except Exception as e:
logger.error(f"Youtube download crashed: {e}")
article_object.file_name = ""
logfile = os.path.join(download_config["local_storage_path"], "failed_downloads.csv")
logger.info(f"Logging youtube errors seperately to {logfile}")
with open(logfile, "a+") as f:
f.write(f"{url}\n")
return article_object