bellingcat/auto-archiver
1
0
Fork 0
mirror of https://github.com/bellingcat/auto-archiver.git synced 2026-06-07 19:08:30 +03:00
Code Issues Packages Projects Releases Wiki Activity

Ruff format with defaults.

Browse Source
This commit is contained in:
erinhmclark
2025-03-10 18:44:54 +00:00
parent cbb0414e5f
commit 85abe1837a
155 changed files with 2539 additions and 1908 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
.github/workflows/ruff.yaml vendored Normal file
View File

@@ -0,0 +1,22 @@
name: Ruff Formatting & Linting
on: [push, pull_request]
jobs:
ruff:
name: Run Ruff Checks
runs-on: ubuntu-latest
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Run Ruff (Lint & Format Check)
uses: astral-sh/ruff-action@v1
with:
args: "check . --output-format=concise"
- name: Run Ruff Format Check
uses: astral-sh/ruff-action@v1
with:
args: "format --check ."

2
docs/scripts/__init__.py
View File

@@ -1 +1 @@
from scripts import generate_module_docs
from scripts import generate_module_docs

67
docs/scripts/scripts.py
View File

@@ -10,12 +10,12 @@ MODULES_FOLDER = Path(__file__).parent.parent.parent.parent / "src" / "auto_arch
SAVE_FOLDER = Path(__file__).parent.parent / "source" / "modules" / "autogen"
type_color = {
'feeder': "<span style='color: #FFA500'>[feeder](/core_modules.md#feeder-modules)</a></span>",
'extractor': "<span style='color: #00FF00'>[extractor](/core_modules.md#extractor-modules)</a></span>",
'enricher': "<span style='color: #0000FF'>[enricher](/core_modules.md#enricher-modules)</a></span>",
'database': "<span style='color: #FF00FF'>[database](/core_modules.md#database-modules)</a></span>",
'storage': "<span style='color: #FFFF00'>[storage](/core_modules.md#storage-modules)</a></span>",
'formatter': "<span style='color: #00FFFF'>[formatter](/core_modules.md#formatter-modules)</a></span>",
"feeder": "<span style='color: #FFA500'>[feeder](/core_modules.md#feeder-modules)</a></span>",
"extractor": "<span style='color: #00FF00'>[extractor](/core_modules.md#extractor-modules)</a></span>",
"enricher": "<span style='color: #0000FF'>[enricher](/core_modules.md#enricher-modules)</a></span>",
"database": "<span style='color: #FF00FF'>[database](/core_modules.md#database-modules)</a></span>",
"storage": "<span style='color: #FFFF00'>[storage](/core_modules.md#storage-modules)</a></span>",
"formatter": "<span style='color: #00FFFF'>[formatter](/core_modules.md#formatter-modules)</a></span>",
}
TABLE_HEADER = ("Option", "Description", "Default", "Type")
@@ -34,6 +34,7 @@ steps:
"""
def generate_module_docs():
yaml = YAML()
SAVE_FOLDER.mkdir(exist_ok=True)
@@ -48,49 +49,49 @@ def generate_module_docs():
# generate the markdown file from the __manifest__.py file.
manifest = module.manifest
for type in manifest['type']:
for type in manifest["type"]:
modules_by_type.setdefault(type, []).append(module)
description = "\n".join(l.lstrip() for l in manifest['description'].split("\n"))
types = ", ".join(type_color[t] for t in manifest['type'])
description = "\n".join(l.lstrip() for l in manifest["description"].split("\n"))
types = ", ".join(type_color[t] for t in manifest["type"])
readme_str = f"""
# {manifest['name']}
# {manifest["name"]}
```{{admonition}} Module type
{types}
```
{description}
"""
steps_str = "\n".join(f" {t}s:\n - {module.name}" for t in manifest['type'])
"""
steps_str = "\n".join(f" {t}s:\n - {module.name}" for t in manifest["type"])
if not manifest['configs']:
if not manifest["configs"]:
config_string = f"# No configuration options for {module.name}.*\n"
else:
config_table = header_row
config_yaml = {}
global_yaml[module.name] = CommentedMap()
global_yaml.yaml_set_comment_before_after_key(module.name, f"\n\n{module.display_name} configuration options")
global_yaml.yaml_set_comment_before_after_key(
module.name, f"\n\n{module.display_name} configuration options"
)
for key, value in manifest['configs'].items():
type = value.get('type', 'string')
if type == 'json_loader':
value['type'] = 'json'
elif type == 'str':
for key, value in manifest["configs"].items():
type = value.get("type", "string")
if type == "json_loader":
value["type"] = "json"
elif type == "str":
type = "string"
default = value.get('default', '')
default = value.get("default", "")
config_yaml[key] = default
global_yaml[module.name][key] = default
if value.get('help', ''):
global_yaml[module.name].yaml_add_eol_comment(value.get('help', ''), key)
if value.get("help", ""):
global_yaml[module.name].yaml_add_eol_comment(value.get("help", ""), key)
help = "**Required**. " if value.get('required', False) else "Optional. "
help += value.get('help', '')
help = "**Required**. " if value.get("required", False) else "Optional. "
help += value.get("help", "")
config_table += f"| `{module.name}.{key}` | {help} | {value.get('default', '')} | {type} |\n"
global_table += f"| `{module.name}.{key}` | {help} | {default} | {type} |\n"
readme_str += "\n## Configuration Options\n"
@@ -98,18 +99,18 @@ def generate_module_docs():
config_string = io.BytesIO()
yaml.dump({module.name: config_yaml}, config_string)
config_string = config_string.getvalue().decode('utf-8')
config_string = config_string.getvalue().decode("utf-8")
yaml_string = EXAMPLE_YAML.format(steps_str=steps_str, config_string=config_string)
readme_str += f"```{{code}} yaml\n{yaml_string}\n```\n"
if manifest['configs']:
if manifest["configs"]:
readme_str += "\n### Command Line:\n"
readme_str += config_table
# add a link to the autodoc refs
readme_str += f"\n[API Reference](../../../autoapi/{module.name}/index)\n"
# create the module.type folder, use the first type just for where to store the file
for type in manifest['type']:
for type in manifest["type"]:
type_folder = SAVE_FOLDER / type
type_folder.mkdir(exist_ok=True)
with open(type_folder / f"{module.name}.md", "w") as f:
@@ -117,10 +118,10 @@ def generate_module_docs():
f.write(readme_str)
generate_index(modules_by_type)
del global_yaml['placeholder']
del global_yaml["placeholder"]
global_string = io.BytesIO()
global_yaml = yaml.dump(global_yaml, global_string)
global_string = global_string.getvalue().decode('utf-8')
global_string = global_string.getvalue().decode("utf-8")
global_yaml = f"```yaml\n{global_string}\n```"
with open(SAVE_FOLDER / "configs_cheatsheet.md", "w") as f:
f.write("### Configuration File\n" + global_yaml + "\n### Command Line\n" + global_table)
@@ -144,4 +145,4 @@ def generate_index(modules_by_type):
if __name__ == "__main__":
generate_module_docs()
generate_module_docs()

47
docs/source/conf.py
View File

@@ -5,7 +5,7 @@ import os
from importlib.metadata import metadata
from datetime import datetime
sys.path.append(os.path.abspath('../scripts'))
sys.path.append(os.path.abspath("../scripts"))
from scripts import generate_module_docs
from auto_archiver.version import __version__
@@ -20,33 +20,35 @@ project = package_metadata["name"]
copyright = str(datetime.now().year)
author = "Bellingcat"
release = package_metadata["version"]
language = 'en'
language = "en"
# -- General configuration ---------------------------------------------------
extensions = [
"myst_parser", # Markdown support
"autoapi.extension", # Generate API documentation from docstrings
"sphinxcontrib.mermaid", # Mermaid diagrams
"sphinx.ext.viewcode", # Source code links
"myst_parser", # Markdown support
"autoapi.extension", # Generate API documentation from docstrings
"sphinxcontrib.mermaid", # Mermaid diagrams
"sphinx.ext.viewcode", # Source code links
"sphinx_copybutton",
"sphinx.ext.napoleon", # Google-style and NumPy-style docstrings
"sphinx.ext.napoleon", # Google-style and NumPy-style docstrings
"sphinx.ext.autosectionlabel",
# 'sphinx.ext.autosummary', # Summarize module/class/function docs
]
templates_path = ['_templates']
templates_path = ["_templates"]
exclude_patterns = ["_build", "Thumbs.db", ".DS_Store", ""]
# -- AutoAPI Configuration ---------------------------------------------------
autoapi_type = 'python'
autoapi_type = "python"
autoapi_dirs = ["../../src/auto_archiver/core/", "../../src/auto_archiver/utils/"]
# get all the modules and add them to the autoapi_dirs
autoapi_dirs.extend([f"../../src/auto_archiver/modules/{m}" for m in os.listdir("../../src/auto_archiver/modules")])
autodoc_typehints = "signature" # Include type hints in the signature
autoapi_ignore = ["*/version.py", ] # Ignore specific modules
autoapi_keep_files = True # Option to retain intermediate JSON files for debugging
autoapi_add_toctree_entry = True # Include API docs in the TOC
autodoc_typehints = "signature" # Include type hints in the signature
autoapi_ignore = [
"*/version.py",
] # Ignore specific modules
autoapi_keep_files = True # Option to retain intermediate JSON files for debugging
autoapi_add_toctree_entry = True # Include API docs in the TOC
autoapi_python_use_implicit_namespaces = True
autoapi_template_dir = "../_templates/autoapi"
autoapi_options = [
@@ -59,13 +61,13 @@ autoapi_options = [
# -- Markdown Support --------------------------------------------------------
myst_enable_extensions = [
"deflist", # Definition lists
"html_admonition", # HTML-style admonitions
"html_image", # Inline HTML images
"replacements", # Substitutions like (C)
"smartquotes", # Smart quotes
"linkify", # Auto-detect links
"substitution", # Text substitutions
"deflist", # Definition lists
"html_admonition", # HTML-style admonitions
"html_image", # Inline HTML images
"replacements", # Substitutions like (C)
"smartquotes", # Smart quotes
"linkify", # Auto-detect links
"substitution", # Text substitutions
]
myst_heading_anchors = 2
myst_fence_as_directive = ["mermaid"]
@@ -76,7 +78,7 @@ source_suffix = {
}
# -- Options for HTML output -------------------------------------------------
html_theme = 'sphinx_book_theme'
html_theme = "sphinx_book_theme"
html_static_path = ["../_static"]
html_css_files = ["custom.css"]
html_title = f"Auto Archiver v{__version__}"
@@ -87,7 +89,6 @@ html_theme_options = {
}
copybutton_prompt_text = r">>> |\.\.\."
copybutton_prompt_is_regexp = True
copybutton_only_copy_prompt_lines = False
copybutton_only_copy_prompt_lines = False

3
pyproject.toml
View File

@@ -98,6 +98,9 @@ line-length = 120
[tool.ruff.lint]
#add bugbear?
# I : isort
# UP : upgrade, e.g. use fstrings
# ANN : annotations
#extend-select = ["B"]
# E701 - multiple statements on one line (I vote to keep this but I notice it's used quite a lot!)

6
scripts/create_update_gdrive_oauth_token.py
View File

@@ -70,11 +70,7 @@ def main(credentials, token):
print(emailAddress)
# Call the Drive v3 API and return some files
results = (
service.files()
.list(pageSize=10, fields="nextPageToken, files(id, name)")
.execute()
)
results = service.files().list(pageSize=10, fields="nextPageToken, files(id, name)").execute()
items = results.get("files", [])
if not items:

46
scripts/generate_settings_schema.py
View File

@@ -8,12 +8,14 @@ from auto_archiver.core.module import ModuleFactory
from auto_archiver.core.consts import MODULE_TYPES
from auto_archiver.core.config import EMPTY_CONFIG
class SchemaEncoder(json.JSONEncoder):
def default(self, obj):
if isinstance(obj, set):
return list(obj)
return json.JSONEncoder.default(self, obj)
# Get available modules
module_factory = ModuleFactory()
available_modules = module_factory.available_modules()
@@ -21,32 +23,40 @@ available_modules = module_factory.available_modules()
modules_by_type = {}
# Categorize modules by type
for module in available_modules:
for type in module.manifest.get('type', []):
for type in module.manifest.get("type", []):
modules_by_type.setdefault(type, []).append(module)
all_modules_ordered_by_type = sorted(available_modules, key=lambda x: (MODULE_TYPES.index(x.type[0]), not x.requires_setup))
all_modules_ordered_by_type = sorted(
available_modules, key=lambda x: (MODULE_TYPES.index(x.type[0]), not x.requires_setup)
)
yaml: YAML = YAML()
config_string = io.BytesIO()
yaml.dump(EMPTY_CONFIG, config_string)
config_string = config_string.getvalue().decode('utf-8')
config_string = config_string.getvalue().decode("utf-8")
output_schema = {
'modules': dict((module.name,
{
'name': module.name,
'display_name': module.display_name,
'manifest': module.manifest,
'configs': module.configs or None
}
) for module in all_modules_ordered_by_type),
'steps': dict((f"{module_type}s", [module.name for module in modules_by_type[module_type]]) for module_type in MODULE_TYPES),
'configs': [m.name for m in all_modules_ordered_by_type if m.configs],
'module_types': MODULE_TYPES,
'empty_config': config_string
"modules": dict(
(
module.name,
{
"name": module.name,
"display_name": module.display_name,
"manifest": module.manifest,
"configs": module.configs or None,
},
)
for module in all_modules_ordered_by_type
),
"steps": dict(
(f"{module_type}s", [module.name for module in modules_by_type[module_type]]) for module_type in MODULE_TYPES
),
"configs": [m.name for m in all_modules_ordered_by_type if m.configs],
"module_types": MODULE_TYPES,
"empty_config": config_string,
}
current_file_dir = os.path.dirname(os.path.abspath(__file__))
output_file = os.path.join(current_file_dir, 'settings/src/schema.json')
with open(output_file, 'w') as file:
json.dump(output_schema, file, indent=4, cls=SchemaEncoder)
output_file = os.path.join(current_file_dir, "settings/src/schema.json")
with open(output_file, "w") as file:
json.dump(output_schema, file, indent=4, cls=SchemaEncoder)

2
scripts/telegram_setup.py
View File

@@ -12,7 +12,6 @@ Then run this script to create a new session file.
You will need to provide your phone number and a 2FA code the first time you run this script.
"""
import os
from telethon.sync import TelegramClient
from loguru import logger
@@ -26,4 +25,3 @@ SESSION_FILE = "secrets/anon-insta"
os.makedirs("secrets", exist_ok=True)
with TelegramClient(SESSION_FILE, API_ID, API_HASH) as client:
logger.success(f"New session file created: {SESSION_FILE}.session")

8
src/auto_archiver/__main__.py
View File

@@ -1,9 +1,13 @@
""" Entry point for the auto_archiver package. """
"""Entry point for the auto_archiver package."""
from auto_archiver.core.orchestrator import ArchivingOrchestrator
import sys
def main():
for _ in ArchivingOrchestrator()._command_line_run(sys.argv[1:]): pass
for _ in ArchivingOrchestrator()._command_line_run(sys.argv[1:]):
pass
if __name__ == "__main__":
main()

5
src/auto_archiver/core/__init__.py
View File

@@ -1,6 +1,5 @@
""" Core modules to handle things such as orchestration, metadata and configs..
"""Core modules to handle things such as orchestration, metadata and configs.."""
"""
from .metadata import Metadata
from .media import Media
from .base_module import BaseModule
@@ -14,4 +13,4 @@ from .enricher import Enricher
from .feeder import Feeder
from .storage import Storage
from .extractor import Extractor
from .formatter import Formatter
from .formatter import Formatter

49
src/auto_archiver/core/base_module.py
View File

@@ -1,7 +1,6 @@
from __future__ import annotations
from typing import Mapping, Any, Type, TYPE_CHECKING
from typing import Mapping, Any, Type, TYPE_CHECKING
from abc import ABC
from copy import deepcopy, copy
from tempfile import TemporaryDirectory
@@ -13,8 +12,8 @@ from loguru import logger
if TYPE_CHECKING:
from .module import ModuleFactory
class BaseModule(ABC):
class BaseModule(ABC):
"""
Base module class. All modules should inherit from this class.
@@ -46,14 +45,13 @@ class BaseModule(ABC):
@property
def storages(self) -> list:
return self.config.get('storages', [])
return self.config.get("storages", [])
def config_setup(self, config: dict):
# this is important. Each instance is given its own deepcopied config, so modules cannot
# change values to affect other modules
config = deepcopy(config)
authentication = deepcopy(config.pop('authentication', {}))
authentication = deepcopy(config.pop("authentication", {}))
self.authentication = authentication
self.config = config
@@ -68,7 +66,7 @@ class BaseModule(ABC):
"""
Returns the authentication information for a given site. This is used to authenticate
with a site before extracting data. The site should be the domain of the site, e.g. 'twitter.com'
:param site: the domain of the site to get authentication information for
:param extract_cookies: whether or not to extract cookies from the given browser/file and return the cookie jar (disabling can speed up processing if you don't actually need the cookies jar).
@@ -94,7 +92,6 @@ class BaseModule(ABC):
# add the 'www' version of the site to the list of sites to check
authdict = {}
for to_try in [site, f"www.{site}"]:
if to_try in self.authentication:
authdict.update(self.authentication[to_try])
@@ -104,17 +101,20 @@ class BaseModule(ABC):
if not authdict:
for key in self.authentication.keys():
if key in site or site in key:
logger.debug(f"Could not find exact authentication information for site '{site}'. \
logger.debug(
f"Could not find exact authentication information for site '{site}'. \
did find information for '{key}' which is close, is this what you meant? \
If so, edit your authentication settings to make sure it exactly matches.")
If so, edit your authentication settings to make sure it exactly matches."
)
def get_ytdlp_cookiejar(args):
import yt_dlp
from yt_dlp import parse_options
logger.debug(f"Extracting cookies from settings: {args[1]}")
# parse_options returns a named tuple as follows, we only need the ydl_options part
# collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
ytdlp_opts = getattr(parse_options(args), 'ydl_opts')
ytdlp_opts = getattr(parse_options(args), "ydl_opts")
return yt_dlp.YoutubeDL(ytdlp_opts).cookiejar
get_cookiejar_options = None
@@ -125,22 +125,21 @@ If so, edit your authentication settings to make sure it exactly matches.")
# 3. cookies_from_browser setting in global config
# 4. cookies_file setting in global config
if 'cookies_from_browser' in authdict:
get_cookiejar_options = ['--cookies-from-browser', authdict['cookies_from_browser']]
elif 'cookies_file' in authdict:
get_cookiejar_options = ['--cookies', authdict['cookies_file']]
elif 'cookies_from_browser' in self.authentication:
authdict['cookies_from_browser'] = self.authentication['cookies_from_browser']
get_cookiejar_options = ['--cookies-from-browser', self.authentication['cookies_from_browser']]
elif 'cookies_file' in self.authentication:
authdict['cookies_file'] = self.authentication['cookies_file']
get_cookiejar_options = ['--cookies', self.authentication['cookies_file']]
if "cookies_from_browser" in authdict:
get_cookiejar_options = ["--cookies-from-browser", authdict["cookies_from_browser"]]
elif "cookies_file" in authdict:
get_cookiejar_options = ["--cookies", authdict["cookies_file"]]
elif "cookies_from_browser" in self.authentication:
authdict["cookies_from_browser"] = self.authentication["cookies_from_browser"]
get_cookiejar_options = ["--cookies-from-browser", self.authentication["cookies_from_browser"]]
elif "cookies_file" in self.authentication:
authdict["cookies_file"] = self.authentication["cookies_file"]
get_cookiejar_options = ["--cookies", self.authentication["cookies_file"]]
if get_cookiejar_options:
authdict['cookies_jar'] = get_ytdlp_cookiejar(get_cookiejar_options)
authdict["cookies_jar"] = get_ytdlp_cookiejar(get_cookiejar_options)
return authdict
def repr(self):
return f"Module<'{self.display_name}' (config: {self.config[self.name]})>"
return f"Module<'{self.display_name}' (config: {self.config[self.name]})>"

54
src/auto_archiver/core/config.py
View File

@@ -20,12 +20,14 @@ _yaml: YAML = YAML()
DEFAULT_CONFIG_FILE = "secrets/orchestration.yaml"
EMPTY_CONFIG = _yaml.load("""
EMPTY_CONFIG = _yaml.load(
"""
# Auto Archiver Configuration
# Steps are the modules that will be run in the order they are defined
steps:""" + "".join([f"\n {module}s: []" for module in MODULE_TYPES]) + \
"""
steps:"""
+ "".join([f"\n {module}s: []" for module in MODULE_TYPES])
+ """
# Global configuration
@@ -52,14 +54,14 @@ authentication: {}
logging:
level: INFO
""")
"""
)
# note: 'logging' is explicitly added above in order to better format the config file
# Arg Parse Actions/Classes
class AuthenticationJsonParseAction(argparse.Action):
def __call__(self, parser, namespace, values, option_string=None):
try:
auth_dict = json.loads(values)
setattr(namespace, self.dest, auth_dict)
@@ -68,34 +70,38 @@ class AuthenticationJsonParseAction(argparse.Action):
def load_from_file(path):
try:
with open(path, 'r') as f:
with open(path, "r") as f:
try:
auth_dict = json.load(f)
except json.JSONDecodeError:
f.seek(0)
# maybe it's yaml, try that
auth_dict = _yaml.load(f)
if auth_dict.get('authentication'):
auth_dict = auth_dict['authentication']
auth_dict['load_from_file'] = path
if auth_dict.get("authentication"):
auth_dict = auth_dict["authentication"]
auth_dict["load_from_file"] = path
return auth_dict
except:
return None
if isinstance(auth_dict, dict) and auth_dict.get('from_file'):
auth_dict = load_from_file(auth_dict['from_file'])
if isinstance(auth_dict, dict) and auth_dict.get("from_file"):
auth_dict = load_from_file(auth_dict["from_file"])
elif isinstance(auth_dict, str):
# if it's a string
auth_dict = load_from_file(auth_dict)
if not isinstance(auth_dict, dict):
raise argparse.ArgumentTypeError("Authentication must be a dictionary of site names and their authentication methods")
global_options = ['cookies_from_browser', 'cookies_file', 'load_from_file']
raise argparse.ArgumentTypeError(
"Authentication must be a dictionary of site names and their authentication methods"
)
global_options = ["cookies_from_browser", "cookies_file", "load_from_file"]
for key, auth in auth_dict.items():
if key in global_options:
continue
if not isinstance(key, str) or not isinstance(auth, dict):
raise argparse.ArgumentTypeError(f"Authentication must be a dictionary of site names and their authentication methods. Valid global configs are {global_options}")
raise argparse.ArgumentTypeError(
f"Authentication must be a dictionary of site names and their authentication methods. Valid global configs are {global_options}"
)
setattr(namespace, self.dest, auth_dict)
@@ -106,8 +112,8 @@ class UniqueAppendAction(argparse.Action):
if value not in getattr(namespace, self.dest):
getattr(namespace, self.dest).append(value)
class DefaultValidatingParser(argparse.ArgumentParser):
class DefaultValidatingParser(argparse.ArgumentParser):
def error(self, message):
"""
Override of error to format a nicer looking error message using logger
@@ -136,8 +142,10 @@ class DefaultValidatingParser(argparse.ArgumentParser):
return super().parse_known_args(args, namespace)
# Config Utils
def to_dot_notation(yaml_conf: CommentedMap | dict) -> dict:
dotdict = {}
@@ -151,6 +159,7 @@ def to_dot_notation(yaml_conf: CommentedMap | dict) -> dict:
process_subdict(yaml_conf)
return dotdict
def from_dot_notation(dotdict: dict) -> dict:
normal_dict = {}
@@ -171,9 +180,11 @@ def from_dot_notation(dotdict: dict) -> dict:
def is_list_type(value):
return isinstance(value, list) or isinstance(value, tuple) or isinstance(value, set)
def is_dict_type(value):
return isinstance(value, dict) or isinstance(value, CommentedMap)
def merge_dicts(dotdict: dict, yaml_dict: CommentedMap) -> CommentedMap:
yaml_dict: CommentedMap = deepcopy(yaml_dict)
@@ -184,7 +195,7 @@ def merge_dicts(dotdict: dict, yaml_dict: CommentedMap) -> CommentedMap:
yaml_subdict[key] = value
continue
if key == 'steps':
if key == "steps":
for module_type, modules in value.items():
# overwrite the 'steps' from the config file with the ones from the CLI
yaml_subdict[key][module_type] = modules
@@ -199,6 +210,7 @@ def merge_dicts(dotdict: dict, yaml_dict: CommentedMap) -> CommentedMap:
update_dict(from_dot_notation(dotdict), yaml_dict)
return yaml_dict
def read_yaml(yaml_filename: str) -> CommentedMap:
config = None
try:
@@ -212,6 +224,7 @@ def read_yaml(yaml_filename: str) -> CommentedMap:
return config
# TODO: make this tidier/find a way to notify of which keys should not be stored
@@ -219,13 +232,14 @@ def store_yaml(config: CommentedMap, yaml_filename: str) -> None:
config_to_save = deepcopy(config)
auth_dict = config_to_save.get("authentication", {})
if auth_dict and auth_dict.get('load_from_file'):
if auth_dict and auth_dict.get("load_from_file"):
# remove all other values from the config, don't want to store it in the config file
auth_dict = {"load_from_file": auth_dict["load_from_file"]}
config_to_save.pop('urls', None)
config_to_save.pop("urls", None)
with open(yaml_filename, "w", encoding="utf-8") as outf:
_yaml.dump(config_to_save, outf)
def is_valid_config(config: CommentedMap) -> bool:
return config and config != EMPTY_CONFIG
return config and config != EMPTY_CONFIG

30
src/auto_archiver/core/consts.py
View File

@@ -1,23 +1,15 @@
MODULE_TYPES = [
'feeder',
'extractor',
'enricher',
'database',
'storage',
'formatter'
]
MODULE_TYPES = ["feeder", "extractor", "enricher", "database", "storage", "formatter"]
MANIFEST_FILE = "__manifest__.py"
DEFAULT_MANIFEST = {
'name': '', # the display name of the module
'author': 'Bellingcat', # creator of the module, leave this as Bellingcat or set your own name!
'type': [], # the type of the module, can be one or more of MODULE_TYPES
'requires_setup': True, # whether or not this module requires additional setup such as setting API Keys or installing additional software
'description': '', # a description of the module
'dependencies': {}, # external dependencies, e.g. python packages or binaries, in dictionary format
'entry_point': '', # the entry point for the module, in the format 'module_name::ClassName'. This can be left blank to use the default entry point of module_name::ModuleName
'version': '1.0', # the version of the module
'configs': {} # any configuration options this module has, these will be exposed to the user in the config file or via the command line
}
"name": "", # the display name of the module
"author": "Bellingcat", # creator of the module, leave this as Bellingcat or set your own name!
"type": [], # the type of the module, can be one or more of MODULE_TYPES
"requires_setup": True, # whether or not this module requires additional setup such as setting API Keys or installing additional software
"description": "", # a description of the module
"dependencies": {}, # external dependencies, e.g. python packages or binaries, in dictionary format
"entry_point": "", # the entry point for the module, in the format 'module_name::ClassName'. This can be left blank to use the default entry point of module_name::ModuleName
"version": "1.0", # the version of the module
"configs": {}, # any configuration options this module has, these will be exposed to the user in the config file or via the command line
}

7
src/auto_archiver/core/database.py
View File

@@ -1,6 +1,6 @@
"""
Database module for the auto-archiver that defines the interface for implementing database modules
in the media archiving framework.
in the media archiving framework.
"""
from __future__ import annotations
@@ -9,6 +9,7 @@ from typing import Union
from auto_archiver.core import Metadata, BaseModule
class Database(BaseModule):
"""
Base class for implementing database modules in the media archiving framework.
@@ -20,7 +21,7 @@ class Database(BaseModule):
"""signals the DB that the given item archival has started"""
pass
def failed(self, item: Metadata, reason:str) -> None:
def failed(self, item: Metadata, reason: str) -> None:
"""update DB accordingly for failure"""
pass
@@ -34,6 +35,6 @@ class Database(BaseModule):
return False
@abstractmethod
def done(self, item: Metadata, cached: bool=False) -> None:
def done(self, item: Metadata, cached: bool = False) -> None:
"""archival result ready - should be saved to DB"""
pass

4
src/auto_archiver/core/enricher.py
View File

@@ -8,13 +8,15 @@ the archiving step and before storage or formatting.
Enrichers are optional but highly useful for making the archived data more powerful.
"""
from __future__ import annotations
from abc import abstractmethod
from auto_archiver.core import Metadata, BaseModule
class Enricher(BaseModule):
"""Base classes and utilities for enrichers in the Auto Archiver system.
Enricher modules must implement the `enrich` method to define their behavior.
"""

28
src/auto_archiver/core/extractor.py
View File

@@ -1,10 +1,11 @@
""" The `extractor` module defines the base functionality for implementing extractors in the media archiving framework.
This class provides common utility methods and a standard interface for extractors.
"""The `extractor` module defines the base functionality for implementing extractors in the media archiving framework.
This class provides common utility methods and a standard interface for extractors.
Factory method to initialize an extractor instance based on its name.
Factory method to initialize an extractor instance based on its name.
"""
from __future__ import annotations
from pathlib import Path
from abc import abstractmethod
@@ -39,7 +40,7 @@ class Extractor(BaseModule):
Used to clean unnecessary URL parameters OR unfurl redirect links
"""
return url
def match_link(self, url: str) -> re.Match:
"""
Returns a match object if the given URL matches the valid_url pattern or False/None if not.
@@ -58,7 +59,7 @@ class Extractor(BaseModule):
"""
if self.valid_url:
return self.match_link(url) is not None
return True
def _guess_file_type(self, path: str) -> str:
@@ -74,16 +75,17 @@ class Extractor(BaseModule):
@retry(wait_random_min=500, wait_random_max=3500, stop_max_attempt_number=5)
def download_from_url(self, url: str, to_filename: str = None, verbose=True) -> str:
"""
downloads a URL to provided filename, or inferred from URL, returns local filename
downloads a URL to provided filename, or inferred from URL, returns local filename
"""
if not to_filename:
to_filename = url.split('/')[-1].split('?')[0]
to_filename = url.split("/")[-1].split("?")[0]
if len(to_filename) > 64:
to_filename = to_filename[-64:]
to_filename = os.path.join(self.tmp_dir, to_filename)
if verbose: logger.debug(f"downloading {url[0:50]=} {to_filename=}")
if verbose:
logger.debug(f"downloading {url[0:50]=} {to_filename=}")
headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.138 Safari/537.36'
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.138 Safari/537.36"
}
try:
d = requests.get(url, stream=True, headers=headers, timeout=30)
@@ -91,12 +93,12 @@ class Extractor(BaseModule):
# get mimetype from the response headers
if not mimetypes.guess_type(to_filename)[0]:
content_type = d.headers.get('Content-Type') or self._guess_file_type(url)
content_type = d.headers.get("Content-Type") or self._guess_file_type(url)
extension = mimetypes.guess_extension(content_type)
if extension:
to_filename += extension
with open(to_filename, 'wb') as f:
with open(to_filename, "wb") as f:
for chunk in d.iter_content(chunk_size=8192):
f.write(chunk)
return to_filename
@@ -108,8 +110,8 @@ class Extractor(BaseModule):
def download(self, item: Metadata) -> Metadata | False:
"""
Downloads the media from the given URL and returns a Metadata object with the downloaded media.
If the URL is not supported or the download fails, this method should return False.
"""
pass
pass

8
src/auto_archiver/core/feeder.py
View File

@@ -1,5 +1,5 @@
"""
The feeder base module defines the interface for implementing feeders in the media archiving framework.
The feeder base module defines the interface for implementing feeders in the media archiving framework.
"""
from __future__ import annotations
@@ -7,8 +7,8 @@ from abc import abstractmethod
from auto_archiver.core import Metadata
from auto_archiver.core import BaseModule
class Feeder(BaseModule):
class Feeder(BaseModule):
"""
Base class for implementing feeders in the media archiving framework.
@@ -19,7 +19,7 @@ class Feeder(BaseModule):
def __iter__(self) -> Metadata:
"""
Returns an iterator (use `yield`) over the items to be archived.
These should be instances of Metadata, typically created with Metadata().set_url(url).
"""
return None
return None

4
src/auto_archiver/core/formatter.py
View File

@@ -12,7 +12,7 @@ from auto_archiver.core import Metadata, Media, BaseModule
class Formatter(BaseModule):
"""
Base class for implementing formatters in the media archiving framework.
Subclasses must implement the `format` method to define their behavior.
"""
@@ -21,4 +21,4 @@ class Formatter(BaseModule):
"""
Formats a Metadata object into a user-viewable format (e.g. HTML) and stores it if needed.
"""
return None
return None

16
src/auto_archiver/core/media.py
View File

@@ -27,6 +27,7 @@ class Media:
- properties: Additional metadata or transformations for the media.
- _mimetype: The media's mimetype (e.g., image/jpeg, video/mp4).
"""
filename: str
key: str = None
urls: List[str] = field(default_factory=list)
@@ -52,14 +53,15 @@ class Media:
This function returns a generator for all the inner media.
"""
if include_self: yield self
if include_self:
yield self
for prop in self.properties.values():
if isinstance(prop, Media):
if isinstance(prop, Media):
for inner_media in prop.all_inner_media(include_self=True):
yield inner_media
if isinstance(prop, list):
for prop_media in prop:
if isinstance(prop_media, Media):
if isinstance(prop_media, Media):
for inner_media in prop_media.all_inner_media(include_self=True):
yield inner_media
@@ -110,15 +112,17 @@ class Media:
# checks for video streams with ffmpeg, or min file size for a video
# self.is_video() should be used together with this method
try:
streams = ffmpeg.probe(self.filename, select_streams='v')['streams']
streams = ffmpeg.probe(self.filename, select_streams="v")["streams"]
logger.warning(f"STREAMS FOR {self.filename} {streams}")
return any(s.get("duration_ts", 0) > 0 for s in streams)
except Error: return False # ffmpeg errors when reading bad files
except Error:
return False # ffmpeg errors when reading bad files
except Exception as e:
logger.error(e)
logger.error(traceback.format_exc())
try:
fsize = os.path.getsize(self.filename)
return fsize > 20_000
except: pass
except:
pass
return True

80
src/auto_archiver/core/metadata.py
View File

@@ -21,6 +21,7 @@ from loguru import logger
from .media import Media
@dataclass_json # annotation order matters
@dataclass
class Metadata:
@@ -40,7 +41,8 @@ class Metadata:
- If `True`, this instance's values are overwritten by `right`.
- If `False`, the inverse applies.
"""
if not right: return self
if not right:
return self
if overwrite_left:
if right.status and len(right.status):
self.status = right.status
@@ -50,8 +52,10 @@ class Metadata:
if type(v) not in [dict, list, set] or k not in self.metadata:
self.set(k, v)
else: # key conflict
if type(v) in [dict, set]: self.set(k, self.get(k) | v)
elif type(v) == list: self.set(k, self.get(k) + v)
if type(v) in [dict, set]:
self.set(k, self.get(k) | v)
elif type(v) == list:
self.set(k, self.get(k) + v)
self.media.extend(right.media)
else: # invert and do same logic
return right.merge(self)
@@ -69,7 +73,7 @@ class Metadata:
def append(self, key: str, val: Any) -> Metadata:
if key not in self.metadata:
self.metadata[key] = []
self.metadata[key] = []
self.metadata[key] = val
return self
@@ -80,24 +84,26 @@ class Metadata:
return self.metadata.get(key, default)
def success(self, context: str = None) -> Metadata:
if context: self.status = f"{context}: success"
else: self.status = "success"
if context:
self.status = f"{context}: success"
else:
self.status = "success"
return self
def is_success(self) -> bool:
return "success" in self.status
def is_empty(self) -> bool:
meaningfull_ids = set(self.metadata.keys()) - set(["_processed_at", "url", "total_bytes", "total_size", "archive_duration_seconds"])
meaningfull_ids = set(self.metadata.keys()) - set(
["_processed_at", "url", "total_bytes", "total_size", "archive_duration_seconds"]
)
return not self.is_success() and len(self.media) == 0 and len(meaningfull_ids) == 0
@property # getter .netloc
def netloc(self) -> str:
return urlparse(self.get_url()).netloc
# custom getter/setters
# custom getter/setters
def set_url(self, url: str) -> Metadata:
assert type(url) is str and len(url) > 0, "invalid URL"
@@ -127,12 +133,17 @@ class Metadata:
def get_timestamp(self, utc=True, iso=True) -> datetime.datetime:
ts = self.get("timestamp")
if not ts: return
if not ts:
return
try:
if type(ts) == str: ts = datetime.datetime.fromisoformat(ts)
if type(ts) == float: ts = datetime.datetime.fromtimestamp(ts)
if utc: ts = ts.replace(tzinfo=datetime.timezone.utc)
if iso: return ts.isoformat()
if type(ts) == str:
ts = datetime.datetime.fromisoformat(ts)
if type(ts) == float:
ts = datetime.datetime.fromtimestamp(ts)
if utc:
ts = ts.replace(tzinfo=datetime.timezone.utc)
if iso:
return ts.isoformat()
return ts
except Exception as e:
logger.error(f"Unable to parse timestamp {ts}: {e}")
@@ -140,16 +151,20 @@ class Metadata:
def add_media(self, media: Media, id: str = None) -> Metadata:
# adds a new media, optionally including an id
if media is None: return
if media is None:
return
if id is not None:
assert not len([1 for m in self.media if m.get("id") == id]), f"cannot add 2 pieces of media with the same id {id}"
assert not len([1 for m in self.media if m.get("id") == id]), (
f"cannot add 2 pieces of media with the same id {id}"
)
media.set("id", id)
self.media.append(media)
return media
def get_media_by_id(self, id: str, default=None) -> Media:
for m in self.media:
if m.get("id") == id: return m
if m.get("id") == id:
return m
return default
def remove_duplicate_media_by_hash(self) -> None:
@@ -159,7 +174,8 @@ class Metadata:
with open(filename, "rb") as f:
while True:
buf = f.read(chunksize)
if not buf: break
if not buf:
break
hash_algo.update(buf)
return hash_algo.hexdigest()
@@ -167,15 +183,18 @@ class Metadata:
new_media = []
for m in self.media:
h = m.get("hash")
if not h: h = calculate_hash_in_chunks(hashlib.sha256(), int(1.6e7), m.filename)
if len(h) and h in media_hashes: continue
if not h:
h = calculate_hash_in_chunks(hashlib.sha256(), int(1.6e7), m.filename)
if len(h) and h in media_hashes:
continue
media_hashes.add(h)
new_media.append(m)
self.media = new_media
def get_first_image(self, default=None) -> Media:
for m in self.media:
if "image" in m.mimetype: return m
if "image" in m.mimetype:
return m
return default
def set_final_media(self, final: Media) -> Metadata:
@@ -193,22 +212,25 @@ class Metadata:
def __str__(self) -> str:
return self.__repr__()
@staticmethod
def choose_most_complete(results: List[Metadata]) -> Metadata:
# returns the most complete result from a list of results
# prioritizes results with more media, then more metadata
if len(results) == 0: return None
if len(results) == 1: return results[0]
if len(results) == 0:
return None
if len(results) == 1:
return results[0]
most_complete = results[0]
for r in results[1:]:
if len(r.media) > len(most_complete.media): most_complete = r
elif len(r.media) == len(most_complete.media) and len(r.metadata) > len(most_complete.metadata): most_complete = r
if len(r.media) > len(most_complete.media):
most_complete = r
elif len(r.media) == len(most_complete.media) and len(r.metadata) > len(most_complete.metadata):
most_complete = r
return most_complete
def set_context(self, key: str, val: Any) -> Metadata:
self._context[key] = val
return self
def get_context(self, key: str, default: Any = None) -> Any:
return self._context.get(key, default)
return self._context.get(key, default)

97
src/auto_archiver/core/module.py
View File

@@ -3,6 +3,7 @@ Defines the Step abstract base class, which acts as a blueprint for steps in the
by handling user configuration, validating the steps properties, and implementing dynamic instantiation.
"""
from __future__ import annotations
from dataclasses import dataclass
@@ -24,17 +25,17 @@ if TYPE_CHECKING:
HAS_SETUP_PATHS = False
class ModuleFactory:
class ModuleFactory:
def __init__(self):
self._lazy_modules = {}
def setup_paths(self, paths: list[str]) -> None:
"""
Sets up the paths for the modules to be loaded from
This is necessary for the modules to be imported correctly
"""
global HAS_SETUP_PATHS
@@ -46,11 +47,13 @@ class ModuleFactory:
# see odoo/module/module.py -> initialize_sys_path
if path not in auto_archiver.modules.__path__:
if HAS_SETUP_PATHS == True:
logger.warning(f"You are attempting to re-initialise the module paths with: '{path}' for a 2nd time. \
if HAS_SETUP_PATHS == True:
logger.warning(
f"You are attempting to re-initialise the module paths with: '{path}' for a 2nd time. \
This could lead to unexpected behaviour. It is recommended to only use a single modules path. \
If you wish to load modules from different paths then load a 2nd python interpreter (e.g. using multiprocessing).")
auto_archiver.modules.__path__.append(path)
If you wish to load modules from different paths then load a 2nd python interpreter (e.g. using multiprocessing)."
)
auto_archiver.modules.__path__.append(path)
# sort based on the length of the path, so that the longest path is last in the list
auto_archiver.modules.__path__ = sorted(auto_archiver.modules.__path__, key=len, reverse=True)
@@ -60,20 +63,20 @@ class ModuleFactory:
def get_module(self, module_name: str, config: dict) -> BaseModule:
"""
Gets and sets up a module using the provided config
This will actually load and instantiate the module, and load all its dependencies (i.e. not lazy)
"""
return self.get_module_lazy(module_name).load(config)
def get_module_lazy(self, module_name: str, suppress_warnings: bool = False) -> LazyBaseModule:
"""
Lazily loads a module, returning a LazyBaseModule
This has all the information about the module, but does not load the module itself or its dependencies
To load an actual module, call .setup() on a lazy module
"""
if module_name in self._lazy_modules:
return self._lazy_modules[module_name]
@@ -81,13 +84,14 @@ class ModuleFactory:
available = self.available_modules(limit_to_modules=[module_name], suppress_warnings=suppress_warnings)
if not available:
message = f"Module '{module_name}' not found. Are you sure it's installed/exists?"
if 'archiver' in module_name:
if "archiver" in module_name:
message += f" Did you mean {module_name.replace('archiver', 'extractor')}?"
raise IndexError(message)
return available[0]
def available_modules(self, limit_to_modules: List[str]= [], suppress_warnings: bool = False) -> List[LazyBaseModule]:
def available_modules(
self, limit_to_modules: List[str] = [], suppress_warnings: bool = False
) -> List[LazyBaseModule]:
# search through all valid 'modules' paths. Default is 'modules' in the current directory
# see odoo/modules/module.py -> get_modules
@@ -119,7 +123,7 @@ class ModuleFactory:
self._lazy_modules[possible_module] = lazy_module
all_modules.append(lazy_module)
if not suppress_warnings:
for module in limit_to_modules:
if not any(module == m.name for m in all_modules):
@@ -127,15 +131,16 @@ class ModuleFactory:
return all_modules
@dataclass
class LazyBaseModule:
"""
A lazy module class, which only loads the manifest and does not load the module itself.
This is useful for getting information about a module without actually loading it.
"""
name: str
description: str
path: str
@@ -152,30 +157,30 @@ class LazyBaseModule:
@property
def type(self):
return self.manifest['type']
return self.manifest["type"]
@property
def entry_point(self):
if not self._entry_point and not self.manifest['entry_point']:
if not self._entry_point and not self.manifest["entry_point"]:
# try to create the entry point from the module name
self._entry_point = f"{self.name}::{self.name.replace('_', ' ').title().replace(' ', '')}"
return self._entry_point
@property
def dependencies(self) -> dict:
return self.manifest['dependencies']
return self.manifest["dependencies"]
@property
def configs(self) -> dict:
return self.manifest['configs']
return self.manifest["configs"]
@property
def requires_setup(self) -> bool:
return self.manifest['requires_setup']
return self.manifest["requires_setup"]
@property
def display_name(self) -> str:
return self.manifest['name']
return self.manifest["name"]
@property
def manifest(self) -> dict:
@@ -190,16 +195,15 @@ class LazyBaseModule:
manifest.update(ast.literal_eval(f.read()))
except (ValueError, TypeError, SyntaxError, MemoryError, RecursionError) as e:
raise ValueError(f"Error loading manifest from file {self.path}/{MANIFEST_FILE}: {e}")
self._manifest = manifest
self._entry_point = manifest['entry_point']
self.description = manifest['description']
self.version = manifest['version']
self._entry_point = manifest["entry_point"]
self.description = manifest["description"]
self.version = manifest["version"]
return manifest
def load(self, config) -> BaseModule:
if self._instance:
return self._instance
@@ -210,8 +214,10 @@ class LazyBaseModule:
# clear out any empty strings that a user may have erroneously added
continue
if not check(dep):
logger.error(f"Module '{self.name}' requires external dependency '{dep}' which is not available/setup. \
Have you installed the required dependencies for the '{self.name}' module? See the README for more information.")
logger.error(
f"Module '{self.name}' requires external dependency '{dep}' which is not available/setup. \
Have you installed the required dependencies for the '{self.name}' module? See the README for more information."
)
exit(1)
def check_python_dep(dep):
@@ -219,7 +225,7 @@ class LazyBaseModule:
try:
m = self.module_factory.get_module_lazy(dep, suppress_warnings=True)
try:
# we must now load this module and set it up with the config
# we must now load this module and set it up with the config
m.load(config)
return True
except:
@@ -231,13 +237,12 @@ class LazyBaseModule:
return find_spec(dep)
check_deps(self.dependencies.get('python', []), check_python_dep)
check_deps(self.dependencies.get('bin', []), lambda dep: shutil.which(dep))
check_deps(self.dependencies.get("python", []), check_python_dep)
check_deps(self.dependencies.get("bin", []), lambda dep: shutil.which(dep))
logger.debug(f"Loading module '{self.display_name}'...")
for qualname in [self.name, f'auto_archiver.modules.{self.name}']:
for qualname in [self.name, f"auto_archiver.modules.{self.name}"]:
try:
# first import the whole module, to make sure it's working properly
__import__(qualname)
@@ -246,10 +251,10 @@ class LazyBaseModule:
pass
# then import the file for the entry point
file_name, class_name = self.entry_point.split('::')
sub_qualname = f'{qualname}.{file_name}'
file_name, class_name = self.entry_point.split("::")
sub_qualname = f"{qualname}.{file_name}"
__import__(f'{qualname}.{file_name}', fromlist=[self.entry_point])
__import__(f"{qualname}.{file_name}", fromlist=[self.entry_point])
# finally, get the class instance
instance: BaseModule = getattr(sys.modules[sub_qualname], class_name)()
@@ -257,11 +262,11 @@ class LazyBaseModule:
instance.name = self.name
instance.display_name = self.display_name
instance.module_factory = self.module_factory
# merge the default config with the user config
default_config = dict((k, v['default']) for k, v in self.configs.items() if 'default' in v)
config[self.name] = default_config | config.get(self.name, {})
# merge the default config with the user config
default_config = dict((k, v["default"]) for k, v in self.configs.items() if "default" in v)
config[self.name] = default_config | config.get(self.name, {})
instance.config_setup(config)
instance.setup()
@@ -270,4 +275,4 @@ class LazyBaseModule:
return instance
def __repr__(self):
return f"Module<'{self.display_name}' ({self.name})>"
return f"Module<'{self.display_name}' ({self.name})>"

303
src/auto_archiver/core/orchestrator.py
View File

@@ -1,6 +1,6 @@
""" Orchestrates all archiving steps, including feeding items,
archiving them with specific archivers, enrichment, storage,
formatting, database operations and clean up.
"""Orchestrates all archiving steps, including feeding items,
archiving them with specific archivers, enrichment, storage,
formatting, database operations and clean up.
"""
@@ -19,8 +19,17 @@ import requests
from .metadata import Metadata, Media
from auto_archiver.version import __version__
from .config import read_yaml, store_yaml, to_dot_notation, merge_dicts, is_valid_config, \
DefaultValidatingParser, UniqueAppendAction, AuthenticationJsonParseAction, DEFAULT_CONFIG_FILE
from .config import (
read_yaml,
store_yaml,
to_dot_notation,
merge_dicts,
is_valid_config,
DefaultValidatingParser,
UniqueAppendAction,
AuthenticationJsonParseAction,
DEFAULT_CONFIG_FILE,
)
from .module import ModuleFactory, LazyBaseModule
from . import validators, Feeder, Extractor, Database, Storage, Formatter, Enricher
from .consts import MODULE_TYPES
@@ -30,10 +39,12 @@ if TYPE_CHECKING:
from .base_module import BaseModule
from .module import LazyBaseModule
class SetupError(ValueError):
pass
class ArchivingOrchestrator:
class ArchivingOrchestrator:
# instance variables
module_factory: ModuleFactory
setup_finished: bool
@@ -63,30 +74,63 @@ class ArchivingOrchestrator:
epilog="Check the code at https://github.com/bellingcat/auto-archiver",
formatter_class=RichHelpFormatter,
)
parser.add_argument('--help', '-h', action='store_true', dest='help', help='show a full help message and exit')
parser.add_argument('--version', action='version', version=__version__)
parser.add_argument('--config', action='store', dest="config_file", help='the filename of the YAML configuration file (defaults to \'config.yaml\')', default=DEFAULT_CONFIG_FILE)
parser.add_argument('--mode', action='store', dest='mode', type=str, choices=['simple', 'full'], help='the mode to run the archiver in', default='simple')
parser.add_argument("--help", "-h", action="store_true", dest="help", help="show a full help message and exit")
parser.add_argument("--version", action="version", version=__version__)
parser.add_argument(
"--config",
action="store",
dest="config_file",
help="the filename of the YAML configuration file (defaults to 'config.yaml')",
default=DEFAULT_CONFIG_FILE,
)
parser.add_argument(
"--mode",
action="store",
dest="mode",
type=str,
choices=["simple", "full"],
help="the mode to run the archiver in",
default="simple",
)
# override the default 'help' so we can inject all the configs and show those
parser.add_argument('-s', '--store', dest='store', default=False, help='Store the created config in the config file', action=argparse.BooleanOptionalAction)
parser.add_argument('--module_paths', dest='module_paths', nargs='+', default=[], help='additional paths to search for modules', action=UniqueAppendAction)
parser.add_argument(
"-s",
"--store",
dest="store",
default=False,
help="Store the created config in the config file",
action=argparse.BooleanOptionalAction,
)
parser.add_argument(
"--module_paths",
dest="module_paths",
nargs="+",
default=[],
help="additional paths to search for modules",
action=UniqueAppendAction,
)
self.basic_parser = parser
return parser
def check_steps(self, config):
for module_type in MODULE_TYPES:
if not config['steps'].get(f"{module_type}s", []):
if module_type == 'feeder' or module_type == 'formatter' and config['steps'].get(f"{module_type}"):
raise SetupError(f"It appears you have '{module_type}' set under 'steps' in your configuration file, but as of version 0.13.0 of Auto Archiver, you must use '{module_type}s'. Change this in your configuration file and try again. \
Here's how that would look: \n\nsteps:\n {module_type}s:\n - [your_{module_type}_name_here]\n {'extractors:...' if module_type == 'feeder' else '...'}\n")
if module_type == 'extractor' and config['steps'].get('archivers'):
raise SetupError(f"As of version 0.13.0 of Auto Archiver, the 'archivers' step name has been changed to 'extractors'. Change this in your configuration file and try again. \
Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_here]\n enrichers:...\n")
raise SetupError(f"No {module_type}s were configured. Make sure to set at least one {module_type} in your configuration file or on the command line (using --{module_type}s)")
if not config["steps"].get(f"{module_type}s", []):
if module_type == "feeder" or module_type == "formatter" and config["steps"].get(f"{module_type}"):
raise SetupError(
f"It appears you have '{module_type}' set under 'steps' in your configuration file, but as of version 0.13.0 of Auto Archiver, you must use '{module_type}s'. Change this in your configuration file and try again. \
Here's how that would look: \n\nsteps:\n {module_type}s:\n - [your_{module_type}_name_here]\n {'extractors:...' if module_type == 'feeder' else '...'}\n"
)
if module_type == "extractor" and config["steps"].get("archivers"):
raise SetupError(
f"As of version 0.13.0 of Auto Archiver, the 'archivers' step name has been changed to 'extractors'. Change this in your configuration file and try again. \
Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_here]\n enrichers:...\n"
)
raise SetupError(
f"No {module_type}s were configured. Make sure to set at least one {module_type} in your configuration file or on the command line (using --{module_type}s)"
)
def setup_complete_parser(self, basic_config: dict, yaml_config: dict, unused_args: list[str]) -> None:
# modules parser to get the overridden 'steps' values
modules_parser = argparse.ArgumentParser(
add_help=False,
@@ -94,7 +138,9 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
self.add_modules_args(modules_parser)
cli_modules, unused_args = modules_parser.parse_known_args(unused_args)
for module_type in MODULE_TYPES:
yaml_config['steps'][f"{module_type}s"] = getattr(cli_modules, f"{module_type}s", []) or yaml_config['steps'].get(f"{module_type}s", [])
yaml_config["steps"][f"{module_type}s"] = getattr(cli_modules, f"{module_type}s", []) or yaml_config[
"steps"
].get(f"{module_type}s", [])
parser = DefaultValidatingParser(
add_help=False,
@@ -117,30 +163,32 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
enabled_modules = []
# first loads the modules from the config file, then from the command line
for module_type in MODULE_TYPES:
enabled_modules.extend(yaml_config['steps'].get(f"{module_type}s", []))
enabled_modules.extend(yaml_config["steps"].get(f"{module_type}s", []))
# clear out duplicates, but keep the order
enabled_modules = list(dict.fromkeys(enabled_modules))
avail_modules = self.module_factory.available_modules(limit_to_modules=enabled_modules, suppress_warnings=True)
avail_modules = self.module_factory.available_modules(
limit_to_modules=enabled_modules, suppress_warnings=True
)
self.add_individual_module_args(avail_modules, parser)
elif basic_config.mode == 'simple':
elif basic_config.mode == "simple":
simple_modules = [module for module in self.module_factory.available_modules() if not module.requires_setup]
self.add_individual_module_args(simple_modules, parser)
# add them to the config
for module in simple_modules:
for module_type in module.type:
yaml_config['steps'].setdefault(f"{module_type}s", []).append(module.name)
yaml_config["steps"].setdefault(f"{module_type}s", []).append(module.name)
else:
# load all modules, they're not using the 'simple' mode
all_modules = self.module_factory.available_modules()
# add all the modules to the steps
for module in all_modules:
for module_type in module.type:
yaml_config['steps'].setdefault(f"{module_type}s", []).append(module.name)
yaml_config["steps"].setdefault(f"{module_type}s", []).append(module.name)
self.add_individual_module_args(all_modules, parser)
parser.set_defaults(**to_dot_notation(yaml_config))
# reload the parser with the new arguments, now that we have them
@@ -166,43 +214,76 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
store_yaml(config, basic_config.config_file)
return config
def add_modules_args(self, parser: argparse.ArgumentParser = None):
if not parser:
parser = self.parser
# Module loading from the command line
for module_type in MODULE_TYPES:
parser.add_argument(f'--{module_type}s', dest=f'{module_type}s', nargs='+', help=f'the {module_type}s to use', default=[], action=UniqueAppendAction)
parser.add_argument(
f"--{module_type}s",
dest=f"{module_type}s",
nargs="+",
help=f"the {module_type}s to use",
default=[],
action=UniqueAppendAction,
)
def add_additional_args(self, parser: argparse.ArgumentParser = None):
if not parser:
parser = self.parser
parser.add_argument('--authentication', dest='authentication', help='A dictionary of sites and their authentication methods \
parser.add_argument(
"--authentication",
dest="authentication",
help="A dictionary of sites and their authentication methods \
(token, username etc.) that extractors can use to log into \
a website. If passing this on the command line, use a JSON string. \
You may also pass a path to a valid JSON/YAML file which will be parsed.',
default={},
nargs="?",
action=AuthenticationJsonParseAction)
You may also pass a path to a valid JSON/YAML file which will be parsed.",
default={},
nargs="?",
action=AuthenticationJsonParseAction,
)
# logging arguments
parser.add_argument('--logging.level', action='store', dest='logging.level', choices=['INFO', 'DEBUG', 'ERROR', 'WARNING'], help='the logging level to use', default='INFO', type=str.upper)
parser.add_argument('--logging.file', action='store', dest='logging.file', help='the logging file to write to', default=None)
parser.add_argument('--logging.rotation', action='store', dest='logging.rotation', help='the logging rotation to use', default=None)
def add_individual_module_args(self, modules: list[LazyBaseModule] = None, parser: argparse.ArgumentParser = None) -> None:
parser.add_argument(
"--logging.level",
action="store",
dest="logging.level",
choices=["INFO", "DEBUG", "ERROR", "WARNING"],
help="the logging level to use",
default="INFO",
type=str.upper,
)
parser.add_argument(
"--logging.file", action="store", dest="logging.file", help="the logging file to write to", default=None
)
parser.add_argument(
"--logging.rotation",
action="store",
dest="logging.rotation",
help="the logging rotation to use",
default=None,
)
def add_individual_module_args(
self, modules: list[LazyBaseModule] = None, parser: argparse.ArgumentParser = None
) -> None:
if not modules:
modules = self.module_factory.available_modules()
for module in modules:
if module.name == 'cli_feeder':
if module.name == "cli_feeder":
# special case. For the CLI feeder, allow passing URLs directly on the command line without setting --cli_feeder.urls=
parser.add_argument('urls', nargs='*', default=[], help='URL(s) to archive, either a single URL or a list of urls, should not come from config.yaml')
parser.add_argument(
"urls",
nargs="*",
default=[],
help="URL(s) to archive, either a single URL or a list of urls, should not come from config.yaml",
)
continue
if not module.configs:
# this module has no configs, don't show anything in the help
# (TODO: do we want to show something about this module though, like a description?)
@@ -211,21 +292,21 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
group = parser.add_argument_group(module.display_name or module.name, f"{module.description[:100]}...")
for name, kwargs in module.configs.items():
if not kwargs.get('metavar', None):
if not kwargs.get("metavar", None):
# make a nicer metavar, metavar is what's used in the help, e.g. --cli_feeder.urls [METAVAR]
kwargs['metavar'] = name.upper()
kwargs["metavar"] = name.upper()
if kwargs.get('required', False):
if kwargs.get("required", False):
# required args shouldn't have a 'default' value, remove it
kwargs.pop('default', None)
kwargs.pop("default", None)
kwargs.pop('cli_set', None)
should_store = kwargs.pop('should_store', False)
kwargs['dest'] = f"{module.name}.{kwargs.pop('dest', name)}"
kwargs.pop("cli_set", None)
should_store = kwargs.pop("should_store", False)
kwargs["dest"] = f"{module.name}.{kwargs.pop('dest', name)}"
try:
kwargs['type'] = getattr(validators, kwargs.get('type', '__invalid__'))
kwargs["type"] = getattr(validators, kwargs.get("type", "__invalid__"))
except AttributeError:
kwargs['type'] = __builtins__.get(kwargs.get('type'), str)
kwargs["type"] = __builtins__.get(kwargs.get("type"), str)
arg = group.add_argument(f"--{module.name}.{name}", **kwargs)
arg.should_store = should_store
@@ -240,12 +321,11 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
self.basic_parser.exit()
def setup_logging(self, config):
logging_config = config["logging"]
logging_config = config['logging']
if logging_config.get('enabled', True) is False:
if logging_config.get("enabled", True) is False:
# disabled logging settings, they're set on a higher level
logger.disable('auto_archiver')
logger.disable("auto_archiver")
return
# setup loguru logging
@@ -255,38 +335,45 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
pass
# add other logging info
if self.logger_id is None: # note - need direct comparison to None since need to consider falsy value 0
self.logger_id = logger.add(sys.stderr, level=logging_config['level'])
if log_file := logging_config['file']:
logger.add(log_file) if not logging_config['rotation'] else logger.add(log_file, rotation=logging_config['rotation'])
if self.logger_id is None: # note - need direct comparison to None since need to consider falsy value 0
self.logger_id = logger.add(sys.stderr, level=logging_config["level"])
if log_file := logging_config["file"]:
logger.add(log_file) if not logging_config["rotation"] else logger.add(
log_file, rotation=logging_config["rotation"]
)
def install_modules(self, modules_by_type):
"""
Traverses all modules in 'steps' and loads them into the orchestrator, storing them in the
Traverses all modules in 'steps' and loads them into the orchestrator, storing them in the
orchestrator's attributes (self.feeders, self.extractors etc.). If no modules of a certain type
are loaded, the program will exit with an error message.
"""
invalid_modules = []
for module_type in MODULE_TYPES:
step_items = []
modules_to_load = modules_by_type[f"{module_type}s"]
if not modules_to_load:
raise SetupError(f"No {module_type}s were configured. Make sure to set at least one {module_type} in your configuration file or on the command line (using --{module_type}s)")
raise SetupError(
f"No {module_type}s were configured. Make sure to set at least one {module_type} in your configuration file or on the command line (using --{module_type}s)"
)
def check_steps_ok():
if not len(step_items):
if len(modules_to_load):
logger.error(f"Unable to load any {module_type}s. Tried the following, but none were available: {modules_to_load}")
raise SetupError(f"NO {module_type.upper()}S LOADED. Please check your configuration and try again.")
logger.error(
f"Unable to load any {module_type}s. Tried the following, but none were available: {modules_to_load}"
)
raise SetupError(
f"NO {module_type.upper()}S LOADED. Please check your configuration and try again."
)
if (module_type == 'feeder' or module_type == 'formatter') and len(step_items) > 1:
raise SetupError(f"Only one {module_type} is allowed, found {len(step_items)} {module_type}s. Please remove one of the following from your configuration file: {modules_to_load}")
if (module_type == "feeder" or module_type == "formatter") and len(step_items) > 1:
raise SetupError(
f"Only one {module_type} is allowed, found {len(step_items)} {module_type}s. Please remove one of the following from your configuration file: {modules_to_load}"
)
for module in modules_to_load:
if module in invalid_modules:
continue
@@ -295,7 +382,7 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
loaded_module: BaseModule = self.module_factory.get_module(module, self.config)
except (KeyboardInterrupt, Exception) as e:
logger.error(f"Error during setup of modules: {e}\n{traceback.format_exc()}")
if loaded_module and module_type == 'extractor':
if loaded_module and module_type == "extractor":
loaded_module.cleanup()
raise e
@@ -310,11 +397,13 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
def load_config(self, config_file: str) -> dict:
if not os.path.exists(config_file) and config_file != DEFAULT_CONFIG_FILE:
logger.error(f"The configuration file {config_file} was not found. Make sure the file exists and try again, or run without the --config file to use the default settings.")
logger.error(
f"The configuration file {config_file} was not found. Make sure the file exists and try again, or run without the --config file to use the default settings."
)
raise FileNotFoundError(f"Configuration file {config_file} not found")
return read_yaml(config_file)
def setup_config(self, args: list) -> dict:
"""
Sets up the configuration file, merging the default config with the user's config
@@ -337,13 +426,13 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
yaml_config = self.load_config(basic_config.config_file)
return self.setup_complete_parser(basic_config, yaml_config, unused_args)
def check_for_updates(self):
response = requests.get("https://pypi.org/pypi/auto-archiver/json").json()
latest_version = response['info']['version']
latest_version = response["info"]["version"]
# check version compared to current version
if latest_version != __version__:
if os.environ.get('RUNNING_IN_DOCKER'):
if os.environ.get("RUNNING_IN_DOCKER"):
update_cmd = "`docker pull bellingcat/auto-archiver:latest`"
else:
update_cmd = "`pip install --upgrade auto-archiver`"
@@ -353,33 +442,36 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
logger.warning(f"Make sure to update to the latest version using: {update_cmd}")
logger.warning("")
def setup(self, args: list):
"""
Function to configure all setup of the orchestrator: setup configs and load modules.
This method should only ever be called once
"""
self.check_for_updates()
if self.setup_finished:
logger.warning("The `setup_config()` function should only ever be run once. \
logger.warning(
"The `setup_config()` function should only ever be run once. \
If you need to re-run the setup, please re-instantiate a new instance of the orchestrator. \
For code implementatations, you should call .setup_config() once then you may call .feed() \
multiple times to archive multiple URLs.")
multiple times to archive multiple URLs."
)
return
self.setup_basic_parser()
self.config = self.setup_config(args)
logger.info(f"======== Welcome to the AUTO ARCHIVER ({__version__}) ==========")
self.install_modules(self.config['steps'])
self.install_modules(self.config["steps"])
# log out the modules that were loaded
for module_type in MODULE_TYPES:
logger.info(f"{module_type.upper()}S: " + ", ".join(m.display_name for m in getattr(self, f"{module_type}s")))
logger.info(
f"{module_type.upper()}S: " + ", ".join(m.display_name for m in getattr(self, f"{module_type}s"))
)
self.setup_finished = True
def _command_line_run(self, args: list) -> Generator[Metadata]:
@@ -387,9 +479,9 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
This is the main entry point for the orchestrator, when run from the command line.
:param args: list of arguments to pass to the orchestrator - these are the command line args
You should not call this method from code implementations.
This method sets up the configuration, loads the modules, and runs the feed.
If you wish to make code invocations yourself, you should use the 'setup' and 'feed' methods separately.
To test configurations, without loading any modules you can also first call 'setup_configs'
@@ -407,7 +499,6 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
e.cleanup()
def feed(self) -> Generator[Metadata]:
url_count = 0
for feeder in self.feeders:
for item in feeder:
@@ -438,7 +529,7 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
self.cleanup()
exit()
except Exception as e:
logger.error(f'Got unexpected error on item {item}: {e}\n{traceback.format_exc()}')
logger.error(f"Got unexpected error on item {item}: {e}\n{traceback.format_exc()}")
for d in self.databases:
if type(e) == AssertionError:
d.failed(item, str(e))
@@ -453,13 +544,13 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
def archive(self, result: Metadata) -> Union[Metadata, None]:
"""
Runs the archiving process for a single URL
1. Each archiver can sanitize its own URLs
2. Check for cached results in Databases, and signal start to the databases
3. Call Archivers until one succeeds
4. Call Enrichers
5. Store all downloaded/generated media
6. Call selected Formatter and store formatted if needed
Runs the archiving process for a single URL
1. Each archiver can sanitize its own URLs
2. Check for cached results in Databases, and signal start to the databases
3. Call Archivers until one succeeds
4. Call Enrichers
5. Store all downloaded/generated media
6. Call selected Formatter and store formatted if needed
"""
original_url = result.get_url().strip()
@@ -475,7 +566,8 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
url = a.sanitize_url(url)
result.set_url(url)
if original_url != url: result.set("original_url", original_url)
if original_url != url:
result.set("original_url", original_url)
# 2 - notify start to DBs, propagate already archived if feature enabled in DBs
cached_result = None
@@ -486,7 +578,8 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
if cached_result:
logger.debug("Found previously archived entry")
for d in self.databases:
try: d.done(cached_result, cached=True)
try:
d.done(cached_result, cached=True)
except Exception as e:
logger.error(f"ERROR database {d.name}: {e}: {traceback.format_exc()}")
return cached_result
@@ -496,13 +589,15 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
logger.info(f"Trying extractor {a.name} for {url}")
try:
result.merge(a.download(result))
if result.is_success(): break
if result.is_success():
break
except Exception as e:
logger.error(f"ERROR archiver {a.name}: {e}: {traceback.format_exc()}")
# 4 - call enrichers to work with archived content
for e in self.enrichers:
try: e.enrich(result)
try:
e.enrich(result)
except Exception as exc:
logger.error(f"ERROR enricher {e.name}: {exc}: {traceback.format_exc()}")
@@ -520,12 +615,12 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
# signal completion to databases and archivers
for d in self.databases:
try: d.done(result)
try:
d.done(result)
except Exception as e:
logger.error(f"ERROR database {d.name}: {e}: {traceback.format_exc()}")
return result
def setup_authentication(self, config: dict) -> dict:
"""
@@ -534,7 +629,7 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
Split up strings into multiple sites if they are comma separated
"""
authentication = config.get('authentication', {})
authentication = config.get("authentication", {})
# extract out concatenated sites
for key, val in copy(authentication).items():
@@ -543,8 +638,8 @@ Here's how that would look: \n\nsteps:\n extractors:\n - [your_extractor_name_
site = site.strip()
authentication[site] = val
del authentication[key]
config['authentication'] = authentication
config["authentication"] = authentication
return config
# Helper Properties

15
src/auto_archiver/core/storage.py
View File

@@ -15,16 +15,16 @@ from auto_archiver.utils.misc import random_str
from auto_archiver.core import Media, BaseModule, Metadata
from auto_archiver.modules.hash_enricher.hash_enricher import HashEnricher
class Storage(BaseModule):
"""
Base class for implementing storage modules in the media archiving framework.
Subclasses must implement the `get_cdn_url` and `uploadf` methods to define their behavior.
"""
def store(self, media: Media, url: str, metadata: Metadata=None) -> None:
if media.is_stored(in_storage=self):
def store(self, media: Media, url: str, metadata: Metadata = None) -> None:
if media.is_stored(in_storage=self):
logger.debug(f"{media.key} already stored, skipping")
return
self.set_key(media, url, metadata)
@@ -46,14 +46,15 @@ class Storage(BaseModule):
pass
def upload(self, media: Media, **kwargs) -> bool:
logger.debug(f'[{self.__class__.__name__}] storing file {media.filename} with key {media.key}')
with open(media.filename, 'rb') as f:
logger.debug(f"[{self.__class__.__name__}] storing file {media.filename} with key {media.key}")
with open(media.filename, "rb") as f:
return self.uploadf(f, media, **kwargs)
def set_key(self, media: Media, url, metadata: Metadata) -> None:
"""takes the media and optionally item info and generates a key"""
if media.key is not None and len(media.key) > 0: return
folder = metadata.get_context('folder', '')
if media.key is not None and len(media.key) > 0:
return
folder = metadata.get_context("folder", "")
filename, ext = os.path.splitext(media.filename)
# Handle path_generator logic

5
src/auto_archiver/core/validators.py
View File

@@ -3,11 +3,13 @@ from pathlib import Path
import argparse
import json
def example_validator(value):
if "example" not in value:
raise argparse.ArgumentTypeError(f"{value} is not a valid value for this argument")
return value
def positive_number(value):
if value < 0:
raise argparse.ArgumentTypeError(f"{value} is not a positive number")
@@ -19,5 +21,6 @@ def valid_file(value):
raise argparse.ArgumentTypeError(f"File '{value}' does not exist.")
return value
def json_loader(cli_val):
return json.loads(cli_val)
return json.loads(cli_val)

2
src/auto_archiver/modules/api_db/__init__.py
View File

@@ -1 +1 @@
from .api_db import AAApiDb
from .api_db import AAApiDb

3
src/auto_archiver/modules/api_db/__manifest__.py
View File

@@ -11,8 +11,7 @@
"required": True,
"help": "API endpoint where calls are made to",
},
"api_token": {"default": None,
"help": "API Bearer token."},
"api_token": {"default": None, "help": "API Bearer token."},
"public": {
"default": False,
"type": "bool",

26
src/auto_archiver/modules/api_db/api_db.py
View File

@@ -12,10 +12,11 @@ class AAApiDb(Database):
"""Connects to auto-archiver-api instance"""
def fetch(self, item: Metadata) -> Union[Metadata, bool]:
""" query the database for the existence of this item.
Helps avoid re-archiving the same URL multiple times.
"""query the database for the existence of this item.
Helps avoid re-archiving the same URL multiple times.
"""
if not self.use_api_cache: return
if not self.use_api_cache:
return
params = {"url": item.get_url(), "limit": 15}
headers = {"Authorization": f"Bearer {self.api_token}", "accept": "application/json"}
@@ -32,22 +33,25 @@ class AAApiDb(Database):
def done(self, item: Metadata, cached: bool = False) -> None:
"""archival result ready - should be saved to DB"""
if not self.store_results: return
if not self.store_results:
return
if cached:
logger.debug(f"skipping saving archive of {item.get_url()} to the AA API because it was cached")
return
logger.debug(f"saving archive of {item.get_url()} to the AA API.")
payload = {
'author_id': self.author_id,
'url': item.get_url(),
'public': self.public,
'group_id': self.group_id,
'tags': list(self.tags),
'result': item.to_json(),
"author_id": self.author_id,
"url": item.get_url(),
"public": self.public,
"group_id": self.group_id,
"tags": list(self.tags),
"result": item.to_json(),
}
headers = {"Authorization": f"Bearer {self.api_token}"}
response = requests.post(os.path.join(self.api_endpoint, "interop/submit-archive"), json=payload, headers=headers)
response = requests.post(
os.path.join(self.api_endpoint, "interop/submit-archive"), json=payload, headers=headers
)
if response.status_code == 201:
logger.success(f"AA API: {response.json()}")

2
src/auto_archiver/modules/atlos_feeder_db_storage/__init__.py
View File

@@ -1 +1 @@
from .atlos_feeder_db_storage import AtlosFeederDbStorage
from .atlos_feeder_db_storage import AtlosFeederDbStorage

6
src/auto_archiver/modules/atlos_feeder_db_storage/__manifest__.py
View File

@@ -1,7 +1,7 @@
{
"name": "Atlos Feeder Database Storage",
"type": ["feeder", "database", "storage"],
"entry_point": "atlos_feeder_db_storage::AtlosFeederDbStorage",
"entry_point": "atlos_feeder_db_storage::AtlosFeederDbStorage",
"requires_setup": True,
"dependencies": {
"python": ["loguru", "requests"],
@@ -15,7 +15,7 @@
"atlos_url": {
"default": "https://platform.atlos.org",
"help": "The URL of your Atlos instance (e.g., https://platform.atlos.org), without a trailing slash.",
"type": "str"
"type": "str",
},
},
"description": """
@@ -42,5 +42,5 @@
- Requires an Atlos account with a project and a valid API token for authentication.
- Ensures only unprocessed, visible, and ready-to-archive URLs are returned.
- Feches any media items within an Atlos project, regardless of separation into incidents.
"""
""",
}

16
src/auto_archiver/modules/atlos_feeder_db_storage/atlos_feeder_db_storage.py
View File

@@ -10,7 +10,6 @@ from auto_archiver.utils import calculate_file_hash
class AtlosFeederDbStorage(Feeder, Database, Storage):
def setup(self) -> requests.Session:
"""create and return a persistent session."""
self.session = requests.Session()
@@ -18,9 +17,7 @@ class AtlosFeederDbStorage(Feeder, Database, Storage):
def _get(self, endpoint: str, params: Optional[dict] = None) -> dict:
"""Wrapper for GET requests to the Atlos API."""
url = f"{self.atlos_url}{endpoint}"
response = self.session.get(
url, headers={"Authorization": f"Bearer {self.api_token}"}, params=params
)
response = self.session.get(url, headers={"Authorization": f"Bearer {self.api_token}"}, params=params)
response.raise_for_status()
return response.json()
@@ -85,10 +82,7 @@ class AtlosFeederDbStorage(Feeder, Database, Storage):
def _process_metadata(self, item: Metadata) -> dict:
"""Process metadata for storage on Atlos. Will convert any datetime
objects to ISO format."""
return {
k: v.isoformat() if hasattr(v, "isoformat") else v
for k, v in item.metadata.items()
}
return {k: v.isoformat() if hasattr(v, "isoformat") else v for k, v in item.metadata.items()}
def done(self, item: Metadata, cached: bool = False) -> None:
"""Mark an item as successfully archived in Atlos."""
@@ -129,10 +123,7 @@ class AtlosFeederDbStorage(Feeder, Database, Storage):
# Check whether the media has already been uploaded
source_material = self._get(f"/api/v2/source_material/{atlos_id}")["result"]
existing_media = [
artifact.get("file_hash_sha256")
for artifact in source_material.get("artifacts", [])
]
existing_media = [artifact.get("file_hash_sha256") for artifact in source_material.get("artifacts", [])]
if media_hash in existing_media:
logger.info(f"{media.filename} with SHA256 {media_hash} already uploaded to Atlos")
return True
@@ -150,4 +141,3 @@ class AtlosFeederDbStorage(Feeder, Database, Storage):
def uploadf(self, file: IO[bytes], key: str, **kwargs: dict) -> bool:
"""Upload a file-like object; not implemented."""
pass

22
src/auto_archiver/modules/cli_feeder/__manifest__.py
View File

@@ -1,16 +1,16 @@
{
'name': 'Command Line Feeder',
'type': ['feeder'],
'entry_point': 'cli_feeder::CLIFeeder',
'requires_setup': False,
'description': 'Feeds URLs to orchestrator from the command line',
'configs': {
'urls': {
'default': None,
'help': 'URL(s) to archive, either a single URL or a list of urls, should not come from config.yaml',
"name": "Command Line Feeder",
"type": ["feeder"],
"entry_point": "cli_feeder::CLIFeeder",
"requires_setup": False,
"description": "Feeds URLs to orchestrator from the command line",
"configs": {
"urls": {
"default": None,
"help": "URL(s) to archive, either a single URL or a list of urls, should not come from config.yaml",
},
},
'description': """
"description": """
The Command Line Feeder is the default enabled feeder for the Auto Archiver. It allows you to pass URLs directly to the orchestrator from the command line
without the need to specify any additional configuration or command line arguments:
@@ -20,4 +20,4 @@ You can pass multiple URLs by separating them with a space. The URLs will be pro
`auto-archiver --feeder cli_feeder -- https://example.com/1/ https://example.com/2/`
""",
}
}

12
src/auto_archiver/modules/cli_feeder/cli_feeder.py
View File

@@ -3,19 +3,21 @@ from loguru import logger
from auto_archiver.core.feeder import Feeder
from auto_archiver.core.metadata import Metadata
class CLIFeeder(Feeder):
class CLIFeeder(Feeder):
def setup(self) -> None:
self.urls = self.config['urls']
self.urls = self.config["urls"]
if not self.urls:
raise ValueError("No URLs provided. Please provide at least one URL via the command line, or set up an alternative feeder. Use --help for more information.")
raise ValueError(
"No URLs provided. Please provide at least one URL via the command line, or set up an alternative feeder. Use --help for more information."
)
def __iter__(self) -> Metadata:
urls = self.config['urls']
urls = self.config["urls"]
for url in urls:
logger.debug(f"Processing {url}")
m = Metadata().set_url(url)
m.set_context("folder", "cli")
yield m
logger.success(f"Processed {len(urls)} URL(s)")
logger.success(f"Processed {len(urls)} URL(s)")

2
src/auto_archiver/modules/console_db/__init__.py
View File

@@ -1 +1 @@
from .console_db import ConsoleDb
from .console_db import ConsoleDb

8
src/auto_archiver/modules/console_db/console_db.py
View File

@@ -6,18 +6,18 @@ from auto_archiver.core import Metadata
class ConsoleDb(Database):
"""
Outputs results to the console
Outputs results to the console
"""
def started(self, item: Metadata) -> None:
logger.info(f"STARTED {item}")
def failed(self, item: Metadata, reason:str) -> None:
def failed(self, item: Metadata, reason: str) -> None:
logger.error(f"FAILED {item}: {reason}")
def aborted(self, item: Metadata) -> None:
logger.warning(f"ABORTED {item}")
def done(self, item: Metadata, cached: bool=False) -> None:
def done(self, item: Metadata, cached: bool = False) -> None:
"""archival result ready - should be saved to DB"""
logger.success(f"DONE {item}")
logger.success(f"DONE {item}")

2
src/auto_archiver/modules/csv_db/__init__.py
View File

@@ -1 +1 @@
from .csv_db import CSVDb
from .csv_db import CSVDb

9
src/auto_archiver/modules/csv_db/__manifest__.py
View File

@@ -2,12 +2,11 @@
"name": "CSV Database",
"type": ["database"],
"requires_setup": False,
"dependencies": {"python": ["loguru"]
},
'entry_point': 'csv_db::CSVDb',
"dependencies": {"python": ["loguru"]},
"entry_point": "csv_db::CSVDb",
"configs": {
"csv_file": {"default": "db.csv", "help": "CSV file name to save metadata to"},
},
"csv_file": {"default": "db.csv", "help": "CSV file name to save metadata to"},
},
"description": """
Handles exporting archival results to a CSV file.

7
src/auto_archiver/modules/csv_db/csv_db.py
View File

@@ -9,14 +9,15 @@ from auto_archiver.core import Metadata
class CSVDb(Database):
"""
Outputs results to a CSV file
Outputs results to a CSV file
"""
def done(self, item: Metadata, cached: bool=False) -> None:
def done(self, item: Metadata, cached: bool = False) -> None:
"""archival result ready - should be saved to DB"""
logger.success(f"DONE {item}")
is_empty = not os.path.isfile(self.csv_file) or os.path.getsize(self.csv_file) == 0
with open(self.csv_file, "a", encoding="utf-8") as outf:
writer = DictWriter(outf, fieldnames=asdict(Metadata()))
if is_empty: writer.writeheader()
if is_empty:
writer.writeheader()
writer.writerow(asdict(item))

2
src/auto_archiver/modules/csv_feeder/__init__.py
View File

@@ -1 +1 @@
from .csv_feeder import CSVFeeder
from .csv_feeder import CSVFeeder

33
src/auto_archiver/modules/csv_feeder/__manifest__.py
View File

@@ -2,26 +2,23 @@
"name": "CSV Feeder",
"type": ["feeder"],
"requires_setup": False,
"dependencies": {
"python": ["loguru"],
"bin": [""]
},
'requires_setup': True,
'entry_point': "csv_feeder::CSVFeeder",
"dependencies": {"python": ["loguru"], "bin": [""]},
"requires_setup": True,
"entry_point": "csv_feeder::CSVFeeder",
"configs": {
"files": {
"default": None,
"help": "Path to the input file(s) to read the URLs from, comma separated. \
"files": {
"default": None,
"help": "Path to the input file(s) to read the URLs from, comma separated. \
Input files should be formatted with one URL per line",
"required": True,
"type": "valid_file",
"nargs": "+",
},
"column": {
"default": None,
"help": "Column number or name to read the URLs from, 0-indexed",
}
"required": True,
"type": "valid_file",
"nargs": "+",
},
"column": {
"default": None,
"help": "Column number or name to read the URLs from, 0-indexed",
},
},
"description": """
Reads URLs from CSV files and feeds them into the archiving process.
@@ -33,5 +30,5 @@
### Setup
- Input files should be formatted with one URL per line, with or without a header row.
- If you have a header row, you can specify the column number or name to read URLs from using the 'column' config option.
"""
""",
}

11
src/auto_archiver/modules/csv_feeder/csv_feeder.py
View File

@@ -5,11 +5,10 @@ from auto_archiver.core import Feeder
from auto_archiver.core import Metadata
from auto_archiver.utils import url_or_none
class CSVFeeder(Feeder):
column = None
def __iter__(self) -> Metadata:
for file in self.files:
with open(file, "r") as f:
@@ -20,9 +19,11 @@ class CSVFeeder(Feeder):
try:
url_column = first_row.index(url_column)
except ValueError:
logger.error(f"Column {url_column} not found in header row: {first_row}. Did you set the 'column' config correctly?")
logger.error(
f"Column {url_column} not found in header row: {first_row}. Did you set the 'column' config correctly?"
)
return
elif not(url_or_none(first_row[url_column])):
elif not (url_or_none(first_row[url_column])):
# it's a header row, but we've been given a column number already
logger.debug(f"Skipping header row: {first_row}")
else:
@@ -35,4 +36,4 @@ class CSVFeeder(Feeder):
continue
url = row[url_column]
logger.debug(f"Processing {url}")
yield Metadata().set_url(url)
yield Metadata().set_url(url)

2
src/auto_archiver/modules/gdrive_storage/__init__.py
View File

@@ -1 +1 @@
from .gdrive_storage import GDriveStorage
from .gdrive_storage import GDriveStorage

19
src/auto_archiver/modules/gdrive_storage/__manifest__.py
View File

@@ -22,11 +22,18 @@
"help": "how to name stored files: 'random' creates a random string; 'static' uses a replicable strategy such as a hash.",
"choices": ["random", "static"],
},
"root_folder_id": {"required": True,
"help": "root google drive folder ID to use as storage, found in URL: 'https://drive.google.com/drive/folders/FOLDER_ID'"},
"oauth_token": {"default": None,
"help": "JSON filename with Google Drive OAuth token: check auto-archiver repository scripts folder for create_update_gdrive_oauth_token.py. NOTE: storage used will count towards owner of GDrive folder, therefore it is best to use oauth_token_filename over service_account."},
"service_account": {"default": "secrets/service_account.json", "help": "service account JSON file path, same as used for Google Sheets. NOTE: storage used will count towards the developer account."},
"root_folder_id": {
"required": True,
"help": "root google drive folder ID to use as storage, found in URL: 'https://drive.google.com/drive/folders/FOLDER_ID'",
},
"oauth_token": {
"default": None,
"help": "JSON filename with Google Drive OAuth token: check auto-archiver repository scripts folder for create_update_gdrive_oauth_token.py. NOTE: storage used will count towards owner of GDrive folder, therefore it is best to use oauth_token_filename over service_account.",
},
"service_account": {
"default": "secrets/service_account.json",
"help": "service account JSON file path, same as used for Google Sheets. NOTE: storage used will count towards the developer account.",
},
},
"description": """
@@ -94,5 +101,5 @@ This module integrates Google Drive as a storage backend, enabling automatic fol
https://davemateer.com/2022/04/28/google-drive-with-python#tokens
"""
""",
}

103
src/auto_archiver/modules/gdrive_storage/gdrive_storage.py
View File

@@ -1,4 +1,3 @@
import json
import os
import time
@@ -15,12 +14,9 @@ from auto_archiver.core import Media
from auto_archiver.core import Storage
class GDriveStorage(Storage):
def setup(self) -> None:
self.scopes = ['https://www.googleapis.com/auth/drive']
self.scopes = ["https://www.googleapis.com/auth/drive"]
# Initialize Google Drive service
self._setup_google_drive_service()
@@ -37,25 +33,25 @@ class GDriveStorage(Storage):
def _initialize_with_oauth_token(self):
"""Initialize Google Drive service with OAuth token."""
with open(self.oauth_token, 'r') as stream:
with open(self.oauth_token, "r") as stream:
creds_json = json.load(stream)
creds_json['refresh_token'] = creds_json.get("refresh_token", "")
creds_json["refresh_token"] = creds_json.get("refresh_token", "")
creds = Credentials.from_authorized_user_info(creds_json, self.scopes)
if not creds.valid and creds.expired and creds.refresh_token:
creds.refresh(Request())
with open(self.oauth_token, 'w') as token_file:
with open(self.oauth_token, "w") as token_file:
logger.debug("Saving refreshed OAuth token.")
token_file.write(creds.to_json())
elif not creds.valid:
raise ValueError("Invalid OAuth token. Please regenerate the token.")
return build('drive', 'v3', credentials=creds)
return build("drive", "v3", credentials=creds)
def _initialize_with_service_account(self):
"""Initialize Google Drive service with service account."""
creds = service_account.Credentials.from_service_account_file(self.service_account, scopes=self.scopes)
return build('drive', 'v3', credentials=creds)
return build("drive", "v3", credentials=creds)
def get_cdn_url(self, media: Media) -> str:
"""
@@ -79,7 +75,7 @@ class GDriveStorage(Storage):
return f"https://drive.google.com/file/d/{file_id}/view?usp=sharing"
def upload(self, media: Media, **kwargs) -> bool:
logger.debug(f'[{self.__class__.__name__}] storing file {media.filename} with key {media.key}')
logger.debug(f"[{self.__class__.__name__}] storing file {media.filename} with key {media.key}")
"""
1. for each sub-folder in the path check if exists or create
2. upload file to root_id/other_paths.../filename
@@ -95,25 +91,30 @@ class GDriveStorage(Storage):
parent_id = upload_to
# upload file to gd
logger.debug(f'uploading {filename=} to folder id {upload_to}')
file_metadata = {
'name': [filename],
'parents': [upload_to]
}
logger.debug(f"uploading {filename=} to folder id {upload_to}")
file_metadata = {"name": [filename], "parents": [upload_to]}
media = MediaFileUpload(media.filename, resumable=True)
gd_file = self.service.files().create(supportsAllDrives=True, body=file_metadata, media_body=media, fields='id').execute()
logger.debug(f'uploadf: uploaded file {gd_file["id"]} successfully in folder={upload_to}')
gd_file = (
self.service.files()
.create(supportsAllDrives=True, body=file_metadata, media_body=media, fields="id")
.execute()
)
logger.debug(f"uploadf: uploaded file {gd_file['id']} successfully in folder={upload_to}")
# must be implemented even if unused
def uploadf(self, file: IO[bytes], key: str, **kwargs: dict) -> bool: pass
def uploadf(self, file: IO[bytes], key: str, **kwargs: dict) -> bool:
pass
def _get_id_from_parent_and_name(self, parent_id: str,
name: str,
retries: int = 1,
sleep_seconds: int = 10,
use_mime_type: bool = False,
raise_on_missing: bool = True,
use_cache=False):
def _get_id_from_parent_and_name(
self,
parent_id: str,
name: str,
retries: int = 1,
sleep_seconds: int = 10,
use_mime_type: bool = False,
raise_on_missing: bool = True,
use_cache=False,
):
"""
Retrieves the id of a folder or file from its @name and the @parent_id folder
Optionally does multiple @retries and sleeps @sleep_seconds between them
@@ -137,29 +138,36 @@ class GDriveStorage(Storage):
query_string += f" and mimeType='application/vnd.google-apps.folder' "
for attempt in range(retries):
results = self.service.files().list(
# both below for Google Shared Drives
supportsAllDrives=True,
includeItemsFromAllDrives=True,
q=query_string,
spaces='drive', # ie not appDataFolder or photos
fields='files(id, name)'
).execute()
items = results.get('files', [])
results = (
self.service.files()
.list(
# both below for Google Shared Drives
supportsAllDrives=True,
includeItemsFromAllDrives=True,
q=query_string,
spaces="drive", # ie not appDataFolder or photos
fields="files(id, name)",
)
.execute()
)
items = results.get("files", [])
if len(items) > 0:
logger.debug(f"{debug_header} found {len(items)} matches, returning last of {','.join([i['id'] for i in items])}")
_id = items[-1]['id']
if use_cache: self.api_cache[cache_key] = _id
logger.debug(
f"{debug_header} found {len(items)} matches, returning last of {','.join([i['id'] for i in items])}"
)
_id = items[-1]["id"]
if use_cache:
self.api_cache[cache_key] = _id
return _id
else:
logger.debug(f'{debug_header} not found, attempt {attempt+1}/{retries}.')
logger.debug(f"{debug_header} not found, attempt {attempt + 1}/{retries}.")
if attempt < retries - 1:
logger.debug(f'sleeping for {sleep_seconds} second(s)')
logger.debug(f"sleeping for {sleep_seconds} second(s)")
time.sleep(sleep_seconds)
if raise_on_missing:
raise ValueError(f'{debug_header} not found after {retries} attempt(s)')
raise ValueError(f"{debug_header} not found after {retries} attempt(s)")
return None
def _mkdir(self, name: str, parent_id: str):
@@ -167,12 +175,7 @@ class GDriveStorage(Storage):
Creates a new GDrive folder @name inside folder @parent_id
Returns id of the created folder
"""
logger.debug(f'Creating new folder with {name=} inside {parent_id=}')
file_metadata = {
'name': [name],
'mimeType': 'application/vnd.google-apps.folder',
'parents': [parent_id]
}
gd_folder = self.service.files().create(supportsAllDrives=True, body=file_metadata, fields='id').execute()
return gd_folder.get('id')
logger.debug(f"Creating new folder with {name=} inside {parent_id=}")
file_metadata = {"name": [name], "mimeType": "application/vnd.google-apps.folder", "parents": [parent_id]}
gd_folder = self.service.files().create(supportsAllDrives=True, body=file_metadata, fields="id").execute()
return gd_folder.get("id")

2
src/auto_archiver/modules/generic_extractor/__init__.py
View File

@@ -1 +1 @@
from .generic_extractor import GenericExtractor
from .generic_extractor import GenericExtractor

16
src/auto_archiver/modules/generic_extractor/bluesky.py
View File

@@ -4,15 +4,16 @@ from auto_archiver.core.extractor import Extractor
from auto_archiver.core.metadata import Metadata, Media
from .dropin import GenericDropin, InfoExtractor
class Bluesky(GenericDropin):
class Bluesky(GenericDropin):
def create_metadata(self, post: dict, ie_instance: InfoExtractor, archiver: Extractor, url: str) -> Metadata:
result = Metadata()
result.set_url(url)
result.set_title(post["record"]["text"])
result.set_timestamp(post["record"]["createdAt"])
for k, v in self._get_post_data(post).items():
if v: result.set(k, v)
if v:
result.set(k, v)
# download if embeds present (1 video XOR >=1 images)
for media in self._download_bsky_embeds(post, archiver):
@@ -23,12 +24,12 @@ class Bluesky(GenericDropin):
def extract_post(self, url: str, ie_instance: InfoExtractor) -> dict:
# TODO: If/when this PR (https://github.com/yt-dlp/yt-dlp/pull/12098) is merged on ytdlp, remove the comments and delete the code below
handle, video_id = ie_instance._match_valid_url(url).group('handle', 'id')
handle, video_id = ie_instance._match_valid_url(url).group("handle", "id")
return ie_instance._extract_post(handle=handle, post_id=video_id)
def _download_bsky_embeds(self, post: dict, archiver: Extractor) -> list[Media]:
"""
Iterates over image(s) or video in a Bluesky post and downloads them
Iterates over image(s) or video in a Bluesky post and downloads them
"""
media = []
embed = post.get("record", {}).get("embed", {})
@@ -37,16 +38,15 @@ class Bluesky(GenericDropin):
media_url = "https://bsky.social/xrpc/com.atproto.sync.getBlob?cid={}&did={}"
for image_media in image_medias:
url = media_url.format(image_media['image']['ref']['$link'], post['author']['did'])
url = media_url.format(image_media["image"]["ref"]["$link"], post["author"]["did"])
image_media = archiver.download_from_url(url)
media.append(Media(image_media))
for video_media in video_medias:
url = media_url.format(video_media['ref']['$link'], post['author']['did'])
url = media_url.format(video_media["ref"]["$link"], post["author"]["did"])
video_media = archiver.download_from_url(url)
media.append(Media(video_media))
return media
def _get_post_data(self, post: dict) -> dict:
"""
Extracts relevant information returned by the .getPostThread api call (excluding text/created_at): author, mentions, tags, links.
@@ -74,4 +74,4 @@ class Bluesky(GenericDropin):
res["tags"] = tags
if links:
res["links"] = links
return res
return res

13
src/auto_archiver/modules/generic_extractor/dropin.py
View File

@@ -2,11 +2,12 @@ from yt_dlp.extractor.common import InfoExtractor
from auto_archiver.core.metadata import Metadata
from auto_archiver.core.extractor import Extractor
class GenericDropin:
"""Base class for dropins for the generic extractor.
In many instances, an extractor will exist in ytdlp, but it will only process videos.
Dropins can be created and used to make use of the already-written private code of a
Dropins can be created and used to make use of the already-written private code of a
specific extractor from ytdlp.
The dropin should be able to handle the following methods:
@@ -28,21 +29,19 @@ class GenericDropin:
This method should return the post data from the url.
"""
raise NotImplementedError("This method should be implemented in the subclass")
def create_metadata(self, post: dict, ie_instance: InfoExtractor, archiver: Extractor, url: str) -> Metadata:
"""
This method should create a Metadata object from the post data.
"""
raise NotImplementedError("This method should be implemented in the subclass")
def skip_ytdlp_download(self, url: str, ie_instance: InfoExtractor):
"""
This method should return True if you want to skip the ytdlp download method.
"""
return False
def keys_to_clean(self, video_data: dict, info_extractor: InfoExtractor):
"""
This method should return a list of strings (keys) to clean from the video_data dict.
@@ -50,9 +49,9 @@ class GenericDropin:
E.g. ["uploader", "uploader_id", "tiktok_specific_field"]
"""
return []
def download_additional_media(self, video_data: dict, info_extractor: InfoExtractor, metadata: Metadata):
"""
This method should download any additional media from the post.
"""
return metadata
return metadata

13
src/auto_archiver/modules/generic_extractor/facebook.py
View File

@@ -3,16 +3,15 @@ from .dropin import GenericDropin
class Facebook(GenericDropin):
def extract_post(self, url: str, ie_instance):
video_id = ie_instance._match_valid_url(url).group('id')
ie_instance._download_webpage(
url.replace('://m.facebook.com/', '://www.facebook.com/'), video_id)
webpage = ie_instance._download_webpage(url, ie_instance._match_valid_url(url).group('id'))
video_id = ie_instance._match_valid_url(url).group("id")
ie_instance._download_webpage(url.replace("://m.facebook.com/", "://www.facebook.com/"), video_id)
webpage = ie_instance._download_webpage(url, ie_instance._match_valid_url(url).group("id"))
# TODO: fix once https://github.com/yt-dlp/yt-dlp/pull/12275 is merged
post_data = ie_instance._extract_metadata(webpage)
return post_data
def create_metadata(self, post: dict, ie_instance, archiver, url):
metadata = archiver.create_metadata(url)
metadata.set_title(post.get('title')).set_content(post.get('description')).set_post_data(post)
return metadata
metadata.set_title(post.get("title")).set_content(post.get("description")).set_post_data(post)
return metadata

228
src/auto_archiver/modules/generic_extractor/generic_extractor.py
View File

@@ -12,6 +12,7 @@ from loguru import logger
from auto_archiver.core.extractor import Extractor
from auto_archiver.core import Metadata, Media
class GenericExtractor(Extractor):
_dropins = {}
@@ -19,14 +20,14 @@ class GenericExtractor(Extractor):
# check for file .ytdlp-update in the secrets folder
if self.ytdlp_update_interval < 0:
return
use_secrets = os.path.exists('secrets')
path = os.path.join('secrets' if use_secrets else '', '.ytdlp-update')
use_secrets = os.path.exists("secrets")
path = os.path.join("secrets" if use_secrets else "", ".ytdlp-update")
next_update_check = None
if os.path.exists(path):
with open(path, "r") as f:
next_update_check = datetime.datetime.fromisoformat(f.read())
if not next_update_check or next_update_check < datetime.datetime.now():
self.update_ytdlp()
@@ -36,8 +37,11 @@ class GenericExtractor(Extractor):
def update_ytdlp(self):
logger.info("Checking and updating yt-dlp...")
logger.info(f"Tip: change the 'ytdlp_update_interval' setting to control how often yt-dlp is updated. Set to -1 to disable or 0 to enable on every run. Current setting: {self.ytdlp_update_interval}")
logger.info(
f"Tip: change the 'ytdlp_update_interval' setting to control how often yt-dlp is updated. Set to -1 to disable or 0 to enable on every run. Current setting: {self.ytdlp_update_interval}"
)
from importlib.metadata import version as get_version
old_version = get_version("yt-dlp")
try:
# try and update with pip (this works inside poetry environment and in a normal virtualenv)
@@ -59,15 +63,17 @@ class GenericExtractor(Extractor):
for info_extractor in yt_dlp.YoutubeDL()._ies.values():
if info_extractor.suitable(url) and info_extractor.working():
yield info_extractor
def suitable(self, url: str) -> bool:
"""
Checks for valid URLs out of all ytdlp extractors.
Returns False for the GenericIE, which as labelled by yt-dlp: 'Generic downloader that works on some sites'
"""
return any(self.suitable_extractors(url))
def download_additional_media(self, video_data: dict, info_extractor: InfoExtractor, metadata: Metadata) -> Metadata:
def download_additional_media(
self, video_data: dict, info_extractor: InfoExtractor, metadata: Metadata
) -> Metadata:
"""
Downloads additional media like images, comments, subtitles, etc.
@@ -76,7 +82,7 @@ class GenericExtractor(Extractor):
# Just get the main thumbnail. More thumbnails are available in
# video_data['thumbnails'] should they be required
thumbnail_url = video_data.get('thumbnail')
thumbnail_url = video_data.get("thumbnail")
if thumbnail_url:
try:
cover_image_path = self.download_from_url(thumbnail_url)
@@ -99,15 +105,65 @@ class GenericExtractor(Extractor):
Clean up the ytdlp generic video data to make it more readable and remove unnecessary keys that ytdlp adds
"""
base_keys = ['formats', 'thumbnail', 'display_id', 'epoch', 'requested_downloads',
'duration_string', 'thumbnails', 'http_headers', 'webpage_url_basename', 'webpage_url_domain',
'extractor', 'extractor_key', 'playlist', 'playlist_index', 'duration_string', 'protocol', 'requested_subtitles',
'format_id', 'acodec', 'vcodec', 'ext', 'epoch', '_has_drm', 'filesize', 'audio_ext', 'video_ext', 'vbr', 'abr',
'resolution', 'dynamic_range', 'aspect_ratio', 'cookies', 'format', 'quality', 'preference', 'artists',
'channel_id', 'subtitles', 'tbr', 'url', 'original_url', 'automatic_captions', 'playable_in_embed', 'live_status',
'_format_sort_fields', 'chapters', 'requested_formats', 'format_note',
'audio_channels', 'asr', 'fps', 'was_live', 'is_live', 'heatmap', 'age_limit', 'stretched_ratio']
base_keys = [
"formats",
"thumbnail",
"display_id",
"epoch",
"requested_downloads",
"duration_string",
"thumbnails",
"http_headers",
"webpage_url_basename",
"webpage_url_domain",
"extractor",
"extractor_key",
"playlist",
"playlist_index",
"duration_string",
"protocol",
"requested_subtitles",
"format_id",
"acodec",
"vcodec",
"ext",
"epoch",
"_has_drm",
"filesize",
"audio_ext",
"video_ext",
"vbr",
"abr",
"resolution",
"dynamic_range",
"aspect_ratio",
"cookies",
"format",
"quality",
"preference",
"artists",
"channel_id",
"subtitles",
"tbr",
"url",
"original_url",
"automatic_captions",
"playable_in_embed",
"live_status",
"_format_sort_fields",
"chapters",
"requested_formats",
"format_note",
"audio_channels",
"asr",
"fps",
"was_live",
"is_live",
"heatmap",
"age_limit",
"stretched_ratio",
]
dropin = self.dropin_for_name(info_extractor.ie_key())
if dropin:
try:
@@ -116,8 +172,8 @@ class GenericExtractor(Extractor):
pass
return base_keys
def add_metadata(self, video_data: dict, info_extractor: InfoExtractor, url:str, result: Metadata) -> Metadata:
def add_metadata(self, video_data: dict, info_extractor: InfoExtractor, url: str, result: Metadata) -> Metadata:
"""
Creates a Metadata object from the given video_data
"""
@@ -126,29 +182,36 @@ class GenericExtractor(Extractor):
result = self.download_additional_media(video_data, info_extractor, result)
# keep both 'title' and 'fulltitle', but prefer 'title', falling back to 'fulltitle' if it doesn't exist
result.set_title(video_data.pop('title', video_data.pop('fulltitle', "")))
result.set_title(video_data.pop("title", video_data.pop("fulltitle", "")))
result.set_url(url)
if "description" in video_data: result.set_content(video_data["description"])
if "description" in video_data:
result.set_content(video_data["description"])
# extract comments if enabled
if self.comments:
result.set("comments", [{
"text": c["text"],
"author": c["author"],
"timestamp": datetime.datetime.fromtimestamp(c.get("timestamp"), tz = datetime.timezone.utc)
} for c in video_data.get("comments", [])])
result.set(
"comments",
[
{
"text": c["text"],
"author": c["author"],
"timestamp": datetime.datetime.fromtimestamp(c.get("timestamp"), tz=datetime.timezone.utc),
}
for c in video_data.get("comments", [])
],
)
# then add the common metadata
if timestamp := video_data.pop("timestamp", None):
timestamp = datetime.datetime.fromtimestamp(timestamp, tz = datetime.timezone.utc).isoformat()
timestamp = datetime.datetime.fromtimestamp(timestamp, tz=datetime.timezone.utc).isoformat()
result.set_timestamp(timestamp)
if upload_date := video_data.pop("upload_date", None):
upload_date = datetime.datetime.strptime(upload_date, '%Y%m%d').replace(tzinfo=datetime.timezone.utc)
upload_date = datetime.datetime.strptime(upload_date, "%Y%m%d").replace(tzinfo=datetime.timezone.utc)
result.set("upload_date", upload_date)
# then clean away any keys we don't want
for clean_key in self.keys_to_clean(info_extractor, video_data):
video_data.pop(clean_key, None)
# then add the rest of the video data
for k, v in video_data.items():
if v:
@@ -169,22 +232,24 @@ class GenericExtractor(Extractor):
logger.debug(f"""Could not find valid dropin for {info_extractor.IE_NAME}.
Why not try creating your own, and make sure it has a valid function called 'create_metadata'. Learn more: https://auto-archiver.readthedocs.io/en/latest/user_guidelines.html#""")
return False
post_data = dropin.extract_post(url, ie_instance)
return dropin.create_metadata(post_data, ie_instance, self, url)
def get_metadata_for_video(self, data: dict, info_extractor: Type[InfoExtractor], url: str, ydl: yt_dlp.YoutubeDL) -> Metadata:
def get_metadata_for_video(
self, data: dict, info_extractor: Type[InfoExtractor], url: str, ydl: yt_dlp.YoutubeDL
) -> Metadata:
# this time download
ydl.params['getcomments'] = self.comments
#TODO: for playlist or long lists of videos, how to download one at a time so they can be stored before the next one is downloaded?
ydl.params["getcomments"] = self.comments
# TODO: for playlist or long lists of videos, how to download one at a time so they can be stored before the next one is downloaded?
data = ydl.extract_info(url, ie_key=info_extractor.ie_key(), download=True)
if "entries" in data:
entries = data.get("entries", [])
if not len(entries):
logger.warning('YoutubeDLArchiver could not find any video')
logger.warning("YoutubeDLArchiver could not find any video")
return False
else: entries = [data]
else:
entries = [data]
result = Metadata()
@@ -192,17 +257,18 @@ class GenericExtractor(Extractor):
try:
filename = ydl.prepare_filename(entry)
if not os.path.exists(filename):
filename = filename.split('.')[0] + '.mkv'
filename = filename.split(".")[0] + ".mkv"
new_media = Media(filename)
for x in ["duration", "original_url", "fulltitle", "description", "upload_date"]:
if x in entry: new_media.set(x, entry[x])
if x in entry:
new_media.set(x, entry[x])
# read text from subtitles if enabled
if self.subtitles:
for lang, val in (data.get('requested_subtitles') or {}).items():
try:
subs = pysubs2.load(val.get('filepath'), encoding="utf-8")
for lang, val in (data.get("requested_subtitles") or {}).items():
try:
subs = pysubs2.load(val.get("filepath"), encoding="utf-8")
text = " ".join([line.text for line in subs])
new_media.set(f"subtitles_{lang}", text)
except Exception as e:
@@ -212,8 +278,8 @@ class GenericExtractor(Extractor):
logger.error(f"Error processing entry {entry}: {e}")
return self.add_metadata(data, info_extractor, url, result)
def dropin_for_name(self, dropin_name: str, additional_paths = [], package=__package__) -> Type[InfoExtractor]:
def dropin_for_name(self, dropin_name: str, additional_paths=[], package=__package__) -> Type[InfoExtractor]:
dropin_name = dropin_name.lower()
if dropin_name == "generic":
@@ -221,6 +287,7 @@ class GenericExtractor(Extractor):
return None
dropin_class_name = dropin_name.title()
def _load_dropin(dropin):
dropin_class = getattr(dropin, dropin_class_name)()
return self._dropins.setdefault(dropin_name, dropin_class)
@@ -244,7 +311,7 @@ class GenericExtractor(Extractor):
return _load_dropin(dropin)
except (FileNotFoundError, ModuleNotFoundError):
pass
# fallback to loading the dropins within auto-archiver
try:
return _load_dropin(importlib.import_module(f".{dropin_name}", package=package))
@@ -256,12 +323,12 @@ class GenericExtractor(Extractor):
def download_for_extractor(self, info_extractor: InfoExtractor, url: str, ydl: yt_dlp.YoutubeDL) -> Metadata:
"""
Tries to download the given url using the specified extractor
It first tries to use ytdlp directly to download the video. If the post is not a video, it will then try to
use the extractor's _extract_post method to get the post metadata if possible.
"""
# when getting info without download, we also don't need the comments
ydl.params['getcomments'] = False
ydl.params["getcomments"] = False
result = False
dropin_submodule = self.dropin_for_name(info_extractor.ie_key())
@@ -272,7 +339,7 @@ class GenericExtractor(Extractor):
# don't download since it can be a live stream
data = ydl.extract_info(url, ie_key=info_extractor.ie_key(), download=False)
if data.get('is_live', False) and not self.livestreams:
if data.get("is_live", False) and not self.livestreams:
logger.warning("Livestream detected, skipping due to 'livestreams' configuration setting")
return False
# it's a valid video, that the youtubdedl can download out of the box
@@ -283,16 +350,21 @@ class GenericExtractor(Extractor):
# don't clutter the logs with issues about the 'generic' extractor not having a dropin
return False
logger.debug(f'Issue using "{info_extractor.IE_NAME}" extractor to download video (error: {repr(e)}), attempting to use extractor to get post data instead')
logger.debug(
f'Issue using "{info_extractor.IE_NAME}" extractor to download video (error: {repr(e)}), attempting to use extractor to get post data instead'
)
try:
result = self.get_metadata_for_post(info_extractor, url, ydl)
except (yt_dlp.utils.DownloadError, yt_dlp.utils.ExtractorError) as post_e:
logger.error(f'Error downloading metadata for post: {post_e}')
logger.error(f"Error downloading metadata for post: {post_e}")
return False
except Exception as generic_e:
logger.debug(f'Attempt to extract using ytdlp extractor "{info_extractor.IE_NAME}" failed: \n {repr(generic_e)}', exc_info=True)
logger.debug(
f'Attempt to extract using ytdlp extractor "{info_extractor.IE_NAME}" failed: \n {repr(generic_e)}',
exc_info=True,
)
return False
if result:
extractor_name = "yt-dlp"
if info_extractor:
@@ -308,43 +380,49 @@ class GenericExtractor(Extractor):
def download(self, item: Metadata) -> Metadata:
url = item.get_url()
#TODO: this is a temporary hack until this issue is closed: https://github.com/yt-dlp/yt-dlp/issues/11025
# TODO: this is a temporary hack until this issue is closed: https://github.com/yt-dlp/yt-dlp/issues/11025
if url.startswith("https://ya.ru"):
url = url.replace("https://ya.ru", "https://yandex.ru")
item.set("replaced_url", url)
ydl_options = {
"outtmpl": os.path.join(self.tmp_dir, f"%(id)s.%(ext)s"),
"quiet": False,
"noplaylist": not self.allow_playlist,
"writesubtitles": self.subtitles,
"writeautomaticsub": self.subtitles,
"live_from_start": self.live_from_start,
"proxy": self.proxy,
"max_downloads": self.max_downloads,
"playlistend": self.max_downloads,
}
ydl_options = {'outtmpl': os.path.join(self.tmp_dir, f'%(id)s.%(ext)s'),
'quiet': False, 'noplaylist': not self.allow_playlist ,
'writesubtitles': self.subtitles,'writeautomaticsub': self.subtitles,
"live_from_start": self.live_from_start, "proxy": self.proxy,
"max_downloads": self.max_downloads, "playlistend": self.max_downloads}
# set up auth
auth = self.auth_for_site(url, extract_cookies=False)
# order of importance: username/pasword -> api_key -> cookie -> cookies_from_browser -> cookies_file
if auth:
if 'username' in auth and 'password' in auth:
logger.debug(f'Using provided auth username and password for {url}')
ydl_options['username'] = auth['username']
ydl_options['password'] = auth['password']
elif 'cookie' in auth:
logger.debug(f'Using provided auth cookie for {url}')
yt_dlp.utils.std_headers['cookie'] = auth['cookie']
elif 'cookies_from_browser' in auth:
logger.debug(f'Using extracted cookies from browser {auth["cookies_from_browser"]} for {url}')
ydl_options['cookiesfrombrowser'] = auth['cookies_from_browser']
elif 'cookies_file' in auth:
logger.debug(f'Using cookies from file {auth["cookies_file"]} for {url}')
ydl_options['cookiefile'] = auth['cookies_file']
if "username" in auth and "password" in auth:
logger.debug(f"Using provided auth username and password for {url}")
ydl_options["username"] = auth["username"]
ydl_options["password"] = auth["password"]
elif "cookie" in auth:
logger.debug(f"Using provided auth cookie for {url}")
yt_dlp.utils.std_headers["cookie"] = auth["cookie"]
elif "cookies_from_browser" in auth:
logger.debug(f"Using extracted cookies from browser {auth['cookies_from_browser']} for {url}")
ydl_options["cookiesfrombrowser"] = auth["cookies_from_browser"]
elif "cookies_file" in auth:
logger.debug(f"Using cookies from file {auth['cookies_file']} for {url}")
ydl_options["cookiefile"] = auth["cookies_file"]
ydl = yt_dlp.YoutubeDL(ydl_options) # allsubtitles and subtitleslangs not working as expected, so default lang is always "en"
ydl = yt_dlp.YoutubeDL(
ydl_options
) # allsubtitles and subtitleslangs not working as expected, so default lang is always "en"
for info_extractor in self.suitable_extractors(url):
result = self.download_for_extractor(info_extractor, url, ydl)
if result:
return result
return False

41
src/auto_archiver/modules/generic_extractor/truth.py
View File

@@ -9,11 +9,11 @@ from dateutil.parser import parse as parse_dt
from .dropin import GenericDropin
class Truth(GenericDropin):
class Truth(GenericDropin):
def extract_post(self, url, ie_instance: InfoExtractor) -> dict:
video_id = ie_instance._match_id(url)
truthsocial_url = f'https://truthsocial.com/api/v1/statuses/{video_id}'
truthsocial_url = f"https://truthsocial.com/api/v1/statuses/{video_id}"
return ie_instance._download_json(truthsocial_url, video_id)
def skip_ytdlp_download(self, url, ie_instance: Type[InfoExtractor]) -> bool:
@@ -22,31 +22,42 @@ class Truth(GenericDropin):
def create_metadata(self, post: dict, ie_instance: InfoExtractor, archiver: Extractor, url: str) -> Metadata:
"""
Creates metadata from a truth social post
Only used for posts that contain no media. ytdlp.TruthIE extractor can handle posts with media
Format is:
{'id': '109598702184774628', 'created_at': '2022-12-29T19:51:18.161Z', 'in_reply_to_id': None, 'quote_id': None, 'in_reply_to_account_id': None, 'sensitive': False, 'spoiler_text': '', 'visibility': 'public', 'language': 'en', 'uri': 'https://truthsocial.com/@bbcnewa/109598702184774628', 'url': 'https://truthsocial.com/@bbcnewa/109598702184774628', 'content': '<p>Pele, regarded by many as football\'s greatest ever player, has died in Brazil at the age of 82. <a href="https://www.bbc.com/sport/football/42751517" rel="nofollow noopener noreferrer" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">bbc.com/sport/football/4275151</span><span class="invisible">7</span></a></p>', 'account': {'id': '107905163010312793', 'username': 'bbcnewa', 'acct': 'bbcnewa', 'display_name': 'BBC News', 'locked': False, 'bot': False, 'discoverable': True, 'group': False, 'created_at': '2022-03-05T17:42:01.159Z', 'note': '<p>News, features and analysis by the BBC</p>', 'url': 'https://truthsocial.com/@bbcnewa', 'avatar': 'https://static-assets-1.truthsocial.com/tmtg:prime-ts-assets/accounts/avatars/107/905/163/010/312/793/original/e7c07550dc22c23a.jpeg', 'avatar_static': 'https://static-assets-1.truthsocial.com/tmtg:prime-ts-assets/accounts/avatars/107/905/163/010/312/793/original/e7c07550dc22c23a.jpeg', 'header': 'https://static-assets-1.truthsocial.com/tmtg:prime-ts-assets/accounts/headers/107/905/163/010/312/793/original/a00eeec2b57206c7.jpeg', 'header_static': 'https://static-assets-1.truthsocial.com/tmtg:prime-ts-assets/accounts/headers/107/905/163/010/312/793/original/a00eeec2b57206c7.jpeg', 'followers_count': 1131, 'following_count': 3, 'statuses_count': 9, 'last_status_at': '2024-11-12', 'verified': False, 'location': '', 'website': 'https://www.bbc.com/news', 'unauth_visibility': True, 'chats_onboarded': True, 'feeds_onboarded': True, 'accepting_messages': False, 'show_nonmember_group_statuses': None, 'emojis': [], 'fields': [], 'tv_onboarded': True, 'tv_account': False}, 'media_attachments': [], 'mentions': [], 'tags': [], 'card': None, 'group': None, 'quote': None, 'in_reply_to': None, 'reblog': None, 'sponsored': False, 'replies_count': 1, 'reblogs_count': 0, 'favourites_count': 2, 'favourited': False, 'reblogged': False, 'muted': False, 'pinned': False, 'bookmarked': False, 'poll': None, 'emojis': []}
"""
result = Metadata()
result.set_url(url)
timestamp = post['created_at'] # format is 2022-12-29T19:51:18.161Z
timestamp = post["created_at"] # format is 2022-12-29T19:51:18.161Z
result.set_timestamp(parse_dt(timestamp))
result.set('description', post['content'])
result.set('author', post['account']['username'])
result.set("description", post["content"])
result.set("author", post["account"]["username"])
for key in ['replies_count', 'reblogs_count', 'favourites_count', ('account', 'followers_count'), ('account', 'following_count'), ('account', 'statuses_count'), ('account', 'display_name'), 'language', 'in_reply_to_account', 'replies_count']:
for key in [
"replies_count",
"reblogs_count",
"favourites_count",
("account", "followers_count"),
("account", "following_count"),
("account", "statuses_count"),
("account", "display_name"),
"language",
"in_reply_to_account",
"replies_count",
]:
if isinstance(key, tuple):
store_key = " ".join(key)
else:
store_key = key
result.set(store_key, traverse_obj(post, key))
# add the media
for media in post.get('media_attachments', []):
filename = archiver.download_from_url(media['url'])
result.add_media(Media(filename), id=media.get('id'))
return result
# add the media
for media in post.get("media_attachments", []):
filename = archiver.download_from_url(media["url"])
result.add_media(Media(filename), id=media.get("id"))
return result

36
src/auto_archiver/modules/generic_extractor/twitter.py
View File

@@ -10,9 +10,8 @@ from auto_archiver.core.extractor import Extractor
from .dropin import GenericDropin, InfoExtractor
class Twitter(GenericDropin):
def choose_variant(self, variants):
# choosing the highest quality possible
variant, width, height = None, 0, 0
@@ -27,9 +26,9 @@ class Twitter(GenericDropin):
else:
variant = var if not variant else variant
return variant
def extract_post(self, url: str, ie_instance: InfoExtractor):
twid = ie_instance._match_valid_url(url).group('id')
twid = ie_instance._match_valid_url(url).group("id")
return ie_instance._extract_status(twid=twid)
def create_metadata(self, tweet: dict, ie_instance: InfoExtractor, archiver: Extractor, url: str) -> Metadata:
@@ -41,30 +40,29 @@ class Twitter(GenericDropin):
except (ValueError, KeyError) as ex:
logger.warning(f"Unable to parse tweet: {str(ex)}\nRetreived tweet data: {tweet}")
return False
result\
.set_title(tweet.get('full_text', ''))\
.set_content(json.dumps(tweet, ensure_ascii=False))\
.set_timestamp(timestamp)
result.set_title(tweet.get("full_text", "")).set_content(json.dumps(tweet, ensure_ascii=False)).set_timestamp(
timestamp
)
if not tweet.get("entities", {}).get("media"):
logger.debug('No media found, archiving tweet text only')
logger.debug("No media found, archiving tweet text only")
result.status = "twitter-ytdl"
return result
for i, tw_media in enumerate(tweet["entities"]["media"]):
media = Media(filename="")
mimetype = ""
if tw_media["type"] == "photo":
media.set("src", UrlUtil.twitter_best_quality_url(tw_media['media_url_https']))
media.set("src", UrlUtil.twitter_best_quality_url(tw_media["media_url_https"]))
mimetype = "image/jpeg"
elif tw_media["type"] == "video":
variant = self.choose_variant(tw_media['video_info']['variants'])
media.set("src", variant['url'])
mimetype = variant['content_type']
variant = self.choose_variant(tw_media["video_info"]["variants"])
media.set("src", variant["url"])
mimetype = variant["content_type"]
elif tw_media["type"] == "animated_gif":
variant = tw_media['video_info']['variants'][0]
media.set("src", variant['url'])
mimetype = variant['content_type']
variant = tw_media["video_info"]["variants"][0]
media.set("src", variant["url"])
mimetype = variant["content_type"]
ext = mimetypes.guess_extension(mimetype)
media.filename = archiver.download_from_url(media.get("src"), f'{slugify(url)}_{i}{ext}')
media.filename = archiver.download_from_url(media.get("src"), f"{slugify(url)}_{i}{ext}")
result.add_media(media)
return result
return result

2
src/auto_archiver/modules/gsheet_feeder_db/__init__.py
View File

@@ -1,2 +1,2 @@
from .gworksheet import GWorksheet
from .gsheet_feeder_db import GsheetsFeederDB
from .gsheet_feeder_db import GsheetsFeederDB

6
src/auto_archiver/modules/gsheet_feeder_db/__manifest__.py
View File

@@ -12,9 +12,7 @@
"default": None,
"help": "the id of the sheet to archive (alternative to 'sheet' config)",
},
"header": {"default": 1,
"type": "int",
"help": "index of the header row (starts at 1)", "type": "int"},
"header": {"default": 1, "type": "int", "help": "index of the header row (starts at 1)", "type": "int"},
"service_account": {
"default": "secrets/service_account.json",
"help": "service account JSON file path. Learn how to create one: https://gspread.readthedocs.io/en/latest/oauth2.html",
@@ -65,7 +63,7 @@
"default": True,
"type": "bool",
"help": "if True the stored files path will include 'workbook_name/worksheet_name/...'",
}
},
},
"description": """
GsheetsFeederDatabase

40
src/auto_archiver/modules/gsheet_feeder_db/gsheet_feeder_db.py
View File

@@ -8,6 +8,7 @@ The filtered rows are processed into `Metadata` objects.
- validates the sheet's structure and filters rows based on input configurations.
- Ensures only rows with valid URLs and unprocessed statuses are included.
"""
import os
from typing import Tuple, Union
from urllib.parse import quote
@@ -23,7 +24,6 @@ from auto_archiver.utils.misc import calculate_file_hash, get_current_timestamp
class GsheetsFeederDB(Feeder, Database):
def setup(self) -> None:
self.gsheets_client = gspread.service_account(filename=self.service_account)
# TODO mv to validators
@@ -42,24 +42,28 @@ class GsheetsFeederDB(Feeder, Database):
if not self.should_process_sheet(worksheet.title):
logger.debug(f"SKIPPED worksheet '{worksheet.title}' due to allow/block rules")
continue
logger.info(f'Opening worksheet {ii=}: {worksheet.title=} header={self.header}')
logger.info(f"Opening worksheet {ii=}: {worksheet.title=} header={self.header}")
gw = GWorksheet(worksheet, header_row=self.header, columns=self.columns)
if len(missing_cols := self.missing_required_columns(gw)):
logger.warning(f"SKIPPED worksheet '{worksheet.title}' due to missing required column(s) for {missing_cols}")
logger.warning(
f"SKIPPED worksheet '{worksheet.title}' due to missing required column(s) for {missing_cols}"
)
continue
# process and yield metadata here:
yield from self._process_rows(gw)
logger.success(f'Finished worksheet {worksheet.title}')
logger.success(f"Finished worksheet {worksheet.title}")
def _process_rows(self, gw: GWorksheet):
for row in range(1 + self.header, gw.count_rows() + 1):
url = gw.get_cell(row, 'url').strip()
if not len(url): continue
original_status = gw.get_cell(row, 'status')
status = gw.get_cell(row, 'status', fresh=original_status in ['', None])
url = gw.get_cell(row, "url").strip()
if not len(url):
continue
original_status = gw.get_cell(row, "status")
status = gw.get_cell(row, "status", fresh=original_status in ["", None])
# TODO: custom status parser(?) aka should_retry_from_status
if status not in ['', None]: continue
if status not in ["", None]:
continue
# All checks done - archival process starts here
m = Metadata().set_url(url)
@@ -70,10 +74,10 @@ class GsheetsFeederDB(Feeder, Database):
# TODO: Check folder value not being recognised
m.set_context("gsheet", {"row": row, "worksheet": gw})
if gw.get_cell_or_default(row, 'folder', "") is None:
folder = ''
if gw.get_cell_or_default(row, "folder", "") is None:
folder = ""
else:
folder = slugify(gw.get_cell_or_default(row, 'folder', "").strip())
folder = slugify(gw.get_cell_or_default(row, "folder", "").strip())
if len(folder):
if self.use_sheet_names_in_stored_paths:
m.set_context("folder", os.path.join(folder, slugify(self.sheet), slugify(gw.wks.title)))
@@ -91,12 +95,11 @@ class GsheetsFeederDB(Feeder, Database):
def missing_required_columns(self, gw: GWorksheet) -> list:
missing = []
for required_col in ['url', 'status']:
for required_col in ["url", "status"]:
if not gw.col_exists(required_col):
missing.append(required_col)
return missing
def started(self, item: Metadata) -> None:
logger.warning(f"STARTED {item}")
gw, row = self._retrieve_gsheet(item)
@@ -155,9 +158,7 @@ class GsheetsFeederDB(Feeder, Database):
if len(pdq_hashes):
batch_if_valid("pdq_hash", ",".join(pdq_hashes))
if (screenshot := item.get_media_by_id("screenshot")) and hasattr(
screenshot, "urls"
):
if (screenshot := item.get_media_by_id("screenshot")) and hasattr(screenshot, "urls"):
batch_if_valid("screenshot", "\n".join(screenshot.urls))
if thumbnail := item.get_first_image("thumbnail"):
@@ -186,11 +187,12 @@ class GsheetsFeederDB(Feeder, Database):
logger.debug(f"Unable to update sheet: {e}")
def _retrieve_gsheet(self, item: Metadata) -> Tuple[GWorksheet, int]:
if gsheet := item.get_context("gsheet"):
gw: GWorksheet = gsheet.get("worksheet")
row: int = gsheet.get("row")
elif self.sheet_id:
logger.error(f"Unable to retrieve Gsheet for {item.get_url()}, GsheetDB must be used alongside GsheetFeeder.")
logger.error(
f"Unable to retrieve Gsheet for {item.get_url()}, GsheetDB must be used alongside GsheetFeeder."
)
return gw, row

47
src/auto_archiver/modules/gsheet_feeder_db/gworksheet.py
View File

@@ -5,24 +5,25 @@ class GWorksheet:
"""
This class makes read/write operations to the a worksheet easier.
It can read the headers from a custom row number, but the row references
should always include the offset of the header.
eg: if header=4, row 5 will be the first with data.
should always include the offset of the header.
eg: if header=4, row 5 will be the first with data.
"""
COLUMN_NAMES = {
'url': 'link',
'status': 'archive status',
'folder': 'destination folder',
'archive': 'archive location',
'date': 'archive date',
'thumbnail': 'thumbnail',
'timestamp': 'upload timestamp',
'title': 'upload title',
'text': 'text content',
'screenshot': 'screenshot',
'hash': 'hash',
'pdq_hash': 'perceptual hashes',
'wacz': 'wacz',
'replaywebpage': 'replaywebpage',
"url": "link",
"status": "archive status",
"folder": "destination folder",
"archive": "archive location",
"date": "archive date",
"thumbnail": "thumbnail",
"timestamp": "upload timestamp",
"title": "upload title",
"text": "text content",
"screenshot": "screenshot",
"hash": "hash",
"pdq_hash": "perceptual hashes",
"wacz": "wacz",
"replaywebpage": "replaywebpage",
}
def __init__(self, worksheet, columns=COLUMN_NAMES, header_row=1):
@@ -36,7 +37,7 @@ class GWorksheet:
def _check_col_exists(self, col: str):
if col not in self.columns:
raise Exception(f'Column {col} is not in the configured column names: {self.columns.keys()}')
raise Exception(f"Column {col} is not in the configured column names: {self.columns.keys()}")
def _col_index(self, col: str):
self._check_col_exists(col)
@@ -58,7 +59,7 @@ class GWorksheet:
def get_cell(self, row, col: str, fresh=False):
"""
returns the cell value from (row, col),
returns the cell value from (row, col),
where row can be an index (1-based) OR list of values
as received from self.get_row(row)
if fresh=True, the sheet is queried again for this cell
@@ -71,7 +72,7 @@ class GWorksheet:
row = self.get_row(row)
if col_index >= len(row):
return ''
return ""
return row[col_index]
def get_cell_or_default(self, row, col: str, default: str = None, fresh=False, when_empty_use_default=True):
@@ -96,13 +97,9 @@ class GWorksheet:
receives a list of [(row:int, col:str, val)] and batch updates it, the parameters are the same as in the self.set_cell() method
"""
cell_updates = [
{
'range': self.to_a1(row, col),
'values': [[str(val)[0:49999]]]
}
for row, col, val in cell_updates
{"range": self.to_a1(row, col), "values": [[str(val)[0:49999]]]} for row, col, val in cell_updates
]
self.wks.batch_update(cell_updates, value_input_option='USER_ENTERED')
self.wks.batch_update(cell_updates, value_input_option="USER_ENTERED")
def to_a1(self, row: int, col: str):
# row is 1-based

2
src/auto_archiver/modules/hash_enricher/__init__.py
View File

@@ -1 +1 @@
from .hash_enricher import HashEnricher
from .hash_enricher import HashEnricher

15
src/auto_archiver/modules/hash_enricher/__manifest__.py
View File

@@ -3,16 +3,17 @@
"type": ["enricher"],
"requires_setup": False,
"dependencies": {
"python": ["loguru"],
"python": ["loguru"],
},
"configs": {
"algorithm": {"default": "SHA-256", "help": "hash algorithm to use", "choices": ["SHA-256", "SHA3-512"]},
# TODO add non-negative requirement to match previous implementation?
"chunksize": {"default": 16000000,
"help": "number of bytes to use when reading files in chunks (if this value is too large you will run out of RAM), default is 16MB",
'type': 'int',
},
"algorithm": {"default": "SHA-256", "help": "hash algorithm to use", "choices": ["SHA-256", "SHA3-512"]},
# TODO add non-negative requirement to match previous implementation?
"chunksize": {
"default": 16000000,
"help": "number of bytes to use when reading files in chunks (if this value is too large you will run out of RAM), default is 16MB",
"type": "int",
},
},
"description": """
Generates cryptographic hashes for media files to ensure data integrity and authenticity.

7
src/auto_archiver/modules/hash_enricher/hash_enricher.py
View File

@@ -1,4 +1,4 @@
""" Hash Enricher for generating cryptographic hashes of media files.
"""Hash Enricher for generating cryptographic hashes of media files.
The `HashEnricher` calculates cryptographic hashes (e.g., SHA-256, SHA3-512)
for media files stored in `Metadata` objects. These hashes are used for
@@ -7,6 +7,7 @@ exact duplicates. The hash is computed by reading the file's bytes in chunks,
making it suitable for handling large files efficiently.
"""
import hashlib
from loguru import logger
@@ -20,7 +21,6 @@ class HashEnricher(Enricher):
Calculates hashes for Media instances
"""
def enrich(self, to_enrich: Metadata) -> None:
url = to_enrich.get_url()
logger.debug(f"calculating media hashes for {url=} (using {self.algorithm})")
@@ -35,5 +35,6 @@ class HashEnricher(Enricher):
hash_algo = hashlib.sha256
elif self.algorithm == "SHA3-512":
hash_algo = hashlib.sha3_512
else: return ""
else:
return ""
return calculate_file_hash(filename, hash_algo, self.chunksize)

2
src/auto_archiver/modules/html_formatter/__init__.py
View File

@@ -1 +1 @@
from .html_formatter import HtmlFormatter
from .html_formatter import HtmlFormatter

13
src/auto_archiver/modules/html_formatter/__manifest__.py
View File

@@ -2,14 +2,13 @@
"name": "HTML Formatter",
"type": ["formatter"],
"requires_setup": False,
"dependencies": {
"python": ["hash_enricher", "loguru", "jinja2"],
"bin": [""]
},
"dependencies": {"python": ["hash_enricher", "loguru", "jinja2"], "bin": [""]},
"configs": {
"detect_thumbnails": {"default": True,
"help": "if true will group by thumbnails generated by thumbnail enricher by id 'thumbnail_00'",
"type": "bool"},
"detect_thumbnails": {
"default": True,
"help": "if true will group by thumbnails generated by thumbnail enricher by id 'thumbnail_00'",
"type": "bool",
},
},
"description": """ """,
}

15
src/auto_archiver/modules/html_formatter/html_formatter.py
View File

@@ -11,6 +11,7 @@ from auto_archiver.core import Metadata, Media
from auto_archiver.core import Formatter
from auto_archiver.utils.misc import random_str
class HtmlFormatter(Formatter):
environment: Environment = None
template: any = None
@@ -21,9 +22,9 @@ class HtmlFormatter(Formatter):
self.environment = Environment(loader=FileSystemLoader(template_dir), autoescape=True)
# JinjaHelper class static methods are added as filters
self.environment.filters.update({
k: v.__func__ for k, v in JinjaHelpers.__dict__.items() if isinstance(v, staticmethod)
})
self.environment.filters.update(
{k: v.__func__ for k, v in JinjaHelpers.__dict__.items() if isinstance(v, staticmethod)}
)
# Load a specific template or default to "html_template.html"
template_name = self.config.get("template_name", "html_template.html")
@@ -36,11 +37,7 @@ class HtmlFormatter(Formatter):
return
content = self.template.render(
url=url,
title=item.get_title(),
media=item.media,
metadata=item.metadata,
version=__version__
url=url, title=item.get_title(), media=item.media, metadata=item.metadata, version=__version__
)
html_path = os.path.join(self.tmp_dir, f"formatted{random_str(24)}.html")
@@ -49,7 +46,7 @@ class HtmlFormatter(Formatter):
final_media = Media(filename=html_path, _mimetype="text/html")
# get the already instantiated hash_enricher module
he = self.module_factory.get_module('hash_enricher', self.config)
he = self.module_factory.get_module("hash_enricher", self.config)
if len(hd := he.calculate_hash(final_media.filename)):
final_media.set("hash", f"{he.algorithm}:{hd}")

20
src/auto_archiver/modules/instagram_api_extractor/__manifest__.py
View File

@@ -2,18 +2,18 @@
"name": "Instagram API Extractor",
"type": ["extractor"],
"entry_point": "instagram_api_extractor::InstagramAPIExtractor",
"dependencies":
{"python": ["requests",
"loguru",
"retrying",
"tqdm",],
},
"dependencies": {
"python": [
"requests",
"loguru",
"retrying",
"tqdm",
],
},
"requires_setup": True,
"configs": {
"access_token": {"default": None,
"help": "a valid instagrapi-api token"},
"api_endpoint": {"required": True,
"help": "API endpoint to use"},
"access_token": {"default": None, "help": "a valid instagrapi-api token"},
"api_endpoint": {"required": True, "help": "API endpoint to use"},
"full_profile": {
"default": False,
"type": "bool",

85
src/auto_archiver/modules/instagram_api_extractor/instagram_api_extractor.py
View File

@@ -36,21 +36,16 @@ class InstagramAPIExtractor(Extractor):
if self.api_endpoint[-1] == "/":
self.api_endpoint = self.api_endpoint[:-1]
def download(self, item: Metadata) -> Metadata:
url = item.get_url()
url.replace("instagr.com", "instagram.com").replace(
"instagr.am", "instagram.com"
)
url.replace("instagr.com", "instagram.com").replace("instagr.am", "instagram.com")
insta_matches = self.valid_url.findall(url)
logger.info(f"{insta_matches=}")
if not len(insta_matches) or len(insta_matches[0]) != 3:
return
if len(insta_matches) > 1:
logger.warning(
f"Multiple instagram matches found in {url=}, using the first one"
)
logger.warning(f"Multiple instagram matches found in {url=}, using the first one")
return
g1, g2, g3 = insta_matches[0][0], insta_matches[0][1], insta_matches[0][2]
if g1 == "":
@@ -73,9 +68,7 @@ class InstagramAPIExtractor(Extractor):
def call_api(self, path: str, params: dict) -> dict:
headers = {"accept": "application/json", "x-access-key": self.access_token}
logger.debug(f"calling {self.api_endpoint}/{path} with {params=}")
return requests.get(
f"{self.api_endpoint}/{path}", headers=headers, params=params
).json()
return requests.get(f"{self.api_endpoint}/{path}", headers=headers, params=params).json()
def cleanup_dict(self, d: dict | list) -> dict:
# repeats 3 times to remove nested empty values
@@ -88,8 +81,7 @@ class InstagramAPIExtractor(Extractor):
return {
k: clean_v
for k, v in d.items()
if (clean_v := self.cleanup_dict(v))
not in [0.0, 0, [], {}, "", None, "null"]
if (clean_v := self.cleanup_dict(v)) not in [0.0, 0, [], {}, "", None, "null"]
and k not in ["x", "y", "width", "height"]
}
@@ -126,9 +118,7 @@ class InstagramAPIExtractor(Extractor):
try:
self.download_all_tagged(result, user_id)
except Exception as e:
result.append(
"errors", f"Error downloading tagged posts for {username}"
)
result.append("errors", f"Error downloading tagged posts for {username}")
logger.error(f"Error downloading tagged posts for {username}: {e}")
# download all highlights
@@ -153,22 +143,13 @@ class InstagramAPIExtractor(Extractor):
"errors",
f"Error downloading highlight id{h.get('pk')} for {username}",
)
logger.error(
f"Error downloading highlight id{h.get('pk')} for {username}: {e}"
)
if (
self.full_profile_max_posts
and count_highlights >= self.full_profile_max_posts
):
logger.info(
f"HIGHLIGHTS reached full_profile_max_posts={self.full_profile_max_posts}"
)
logger.error(f"Error downloading highlight id{h.get('pk')} for {username}: {e}")
if self.full_profile_max_posts and count_highlights >= self.full_profile_max_posts:
logger.info(f"HIGHLIGHTS reached full_profile_max_posts={self.full_profile_max_posts}")
break
result.set("#highlights", count_highlights)
def download_post(
self, result: Metadata, code: str = None, id: str = None, context: str = None
) -> Metadata:
def download_post(self, result: Metadata, code: str = None, id: str = None, context: str = None) -> Metadata:
if id:
post = self.call_api(f"v1/media/by/id", {"id": id})
else:
@@ -196,11 +177,7 @@ class InstagramAPIExtractor(Extractor):
h_info = full_h.get("response", {}).get("reels", {}).get(f"highlight:{id}")
assert h_info, f"Highlight {id} not found: {full_h=}"
if (
cover_media := h_info.get("cover_media", {})
.get("cropped_image_version", {})
.get("url")
):
if cover_media := h_info.get("cover_media", {}).get("cropped_image_version", {}).get("url"):
filename = self.download_from_url(cover_media)
result.add_media(Media(filename=filename), id=f"cover_media highlight {id}")
@@ -210,9 +187,7 @@ class InstagramAPIExtractor(Extractor):
self.scrape_item(result, h, "highlight")
except Exception as e:
result.append("errors", f"Error downloading highlight {h.get('id')}")
logger.error(
f"Error downloading highlight, skipping {h.get('id')}: {e}"
)
logger.error(f"Error downloading highlight, skipping {h.get('id')}: {e}")
return h_info
@@ -244,9 +219,7 @@ class InstagramAPIExtractor(Extractor):
post_count = 0
while end_cursor != "":
posts = self.call_api(
f"v1/user/medias/chunk", {"user_id": user_id, "end_cursor": end_cursor}
)
posts = self.call_api(f"v1/user/medias/chunk", {"user_id": user_id, "end_cursor": end_cursor})
if not len(posts) or not type(posts) == list or len(posts) != 2:
break
posts, end_cursor = posts[0], posts[1]
@@ -260,13 +233,8 @@ class InstagramAPIExtractor(Extractor):
logger.error(f"Error downloading post, skipping {p.get('id')}: {e}")
pbar.update(1)
post_count += 1
if (
self.full_profile_max_posts
and post_count >= self.full_profile_max_posts
):
logger.info(
f"POSTS reached full_profile_max_posts={self.full_profile_max_posts}"
)
if self.full_profile_max_posts and post_count >= self.full_profile_max_posts:
logger.info(f"POSTS reached full_profile_max_posts={self.full_profile_max_posts}")
break
result.set("#posts", post_count)
@@ -276,9 +244,7 @@ class InstagramAPIExtractor(Extractor):
tagged_count = 0
while next_page_id != None:
resp = self.call_api(
f"v2/user/tag/medias", {"user_id": user_id, "page_id": next_page_id}
)
resp = self.call_api(f"v2/user/tag/medias", {"user_id": user_id, "page_id": next_page_id})
posts = resp.get("response", {}).get("items", [])
if not len(posts):
break
@@ -290,21 +256,12 @@ class InstagramAPIExtractor(Extractor):
try:
self.scrape_item(result, p, "tagged")
except Exception as e:
result.append(
"errors", f"Error downloading tagged post {p.get('id')}"
)
logger.error(
f"Error downloading tagged post, skipping {p.get('id')}: {e}"
)
result.append("errors", f"Error downloading tagged post {p.get('id')}")
logger.error(f"Error downloading tagged post, skipping {p.get('id')}: {e}")
pbar.update(1)
tagged_count += 1
if (
self.full_profile_max_posts
and tagged_count >= self.full_profile_max_posts
):
logger.info(
f"TAGS reached full_profile_max_posts={self.full_profile_max_posts}"
)
if self.full_profile_max_posts and tagged_count >= self.full_profile_max_posts:
logger.info(f"TAGS reached full_profile_max_posts={self.full_profile_max_posts}")
break
result.set("#tagged", tagged_count)
@@ -318,9 +275,7 @@ class InstagramAPIExtractor(Extractor):
context can be used to give specific id prefixes to media
"""
if "clips_metadata" in item:
if reusable_text := item.get("clips_metadata", {}).get(
"reusable_text_attribute_string"
):
if reusable_text := item.get("clips_metadata", {}).get("reusable_text_attribute_string"):
item["clips_metadata_text"] = reusable_text
if self.minimize_json_output:
del item["clips_metadata"]

2
src/auto_archiver/modules/instagram_extractor/__init__.py
View File

@@ -1 +1 @@
from .instagram_extractor import InstagramExtractor
from .instagram_extractor import InstagramExtractor

3
src/auto_archiver/modules/instagram_extractor/__manifest__.py
View File

@@ -9,8 +9,7 @@
},
"requires_setup": True,
"configs": {
"username": {"required": True,
"help": "A valid Instagram username."},
"username": {"required": True, "help": "A valid Instagram username."},
"password": {
"required": True,
"help": "The corresponding Instagram account password.",

67
src/auto_archiver/modules/instagram_extractor/instagram_extractor.py
View File

@@ -1,8 +1,9 @@
""" Uses the Instaloader library to download content from Instagram. This class handles both individual posts
and user profiles, downloading as much information as possible, including images, videos, text, stories,
highlights, and tagged posts. Authentication is required via username/password or a session file.
"""Uses the Instaloader library to download content from Instagram. This class handles both individual posts
and user profiles, downloading as much information as possible, including images, videos, text, stories,
highlights, and tagged posts. Authentication is required via username/password or a session file.
"""
import re, os, shutil
import instaloader
from loguru import logger
@@ -11,6 +12,7 @@ from auto_archiver.core import Extractor
from auto_archiver.core import Metadata
from auto_archiver.core import Media
class InstagramExtractor(Extractor):
"""
Uses Instaloader to download either a post (inc images, videos, text) or as much as possible from a profile (posts, stories, highlights, ...)
@@ -25,13 +27,12 @@ class InstagramExtractor(Extractor):
# TODO: links to stories
def setup(self) -> None:
self.insta = instaloader.Instaloader(
download_geotags=True,
download_comments=True,
compress_json=False,
dirname_pattern=self.download_folder,
filename_pattern="{date_utc}_UTC_{target}__{typename}"
filename_pattern="{date_utc}_UTC_{target}__{typename}",
)
try:
self.insta.load_session_from_file(self.username, self.session_file)
@@ -44,7 +45,6 @@ class InstagramExtractor(Extractor):
except Exception as e:
logger.error(f"Failed to setup Instagram Extractor with Instagrapi. {e}")
def download(self, item: Metadata) -> Metadata:
url = item.get_url()
@@ -53,7 +53,8 @@ class InstagramExtractor(Extractor):
profile_matches = self.profile_pattern.findall(url)
# return if not a valid instagram link
if not len(post_matches) and not len(profile_matches): return
if not len(post_matches) and not len(profile_matches):
return
result = None
try:
@@ -65,7 +66,9 @@ class InstagramExtractor(Extractor):
elif len(profile_matches):
result = self.download_profile(url, profile_matches[0])
except Exception as e:
logger.error(f"Failed to download with instagram extractor due to: {e}, make sure your account credentials are valid.")
logger.error(
f"Failed to download with instagram extractor due to: {e}, make sure your account credentials are valid."
)
finally:
shutil.rmtree(self.download_folder, ignore_errors=True)
return result
@@ -84,35 +87,50 @@ class InstagramExtractor(Extractor):
profile = instaloader.Profile.from_username(self.insta.context, username)
try:
for post in profile.get_posts():
try: self.insta.download_post(post, target=f"profile_post_{post.owner_username}")
except Exception as e: logger.error(f"Failed to download post: {post.shortcode}: {e}")
except Exception as e: logger.error(f"Failed profile.get_posts: {e}")
try:
self.insta.download_post(post, target=f"profile_post_{post.owner_username}")
except Exception as e:
logger.error(f"Failed to download post: {post.shortcode}: {e}")
except Exception as e:
logger.error(f"Failed profile.get_posts: {e}")
try:
for post in profile.get_tagged_posts():
try: self.insta.download_post(post, target=f"tagged_post_{post.owner_username}")
except Exception as e: logger.error(f"Failed to download tagged post: {post.shortcode}: {e}")
except Exception as e: logger.error(f"Failed profile.get_tagged_posts: {e}")
try:
self.insta.download_post(post, target=f"tagged_post_{post.owner_username}")
except Exception as e:
logger.error(f"Failed to download tagged post: {post.shortcode}: {e}")
except Exception as e:
logger.error(f"Failed profile.get_tagged_posts: {e}")
try:
for post in profile.get_igtv_posts():
try: self.insta.download_post(post, target=f"igtv_post_{post.owner_username}")
except Exception as e: logger.error(f"Failed to download igtv post: {post.shortcode}: {e}")
except Exception as e: logger.error(f"Failed profile.get_igtv_posts: {e}")
try:
self.insta.download_post(post, target=f"igtv_post_{post.owner_username}")
except Exception as e:
logger.error(f"Failed to download igtv post: {post.shortcode}: {e}")
except Exception as e:
logger.error(f"Failed profile.get_igtv_posts: {e}")
try:
for story in self.insta.get_stories([profile.userid]):
for item in story.get_items():
try: self.insta.download_storyitem(item, target=f"story_item_{story.owner_username}")
except Exception as e: logger.error(f"Failed to download story item: {item}: {e}")
except Exception as e: logger.error(f"Failed get_stories: {e}")
try:
self.insta.download_storyitem(item, target=f"story_item_{story.owner_username}")
except Exception as e:
logger.error(f"Failed to download story item: {item}: {e}")
except Exception as e:
logger.error(f"Failed get_stories: {e}")
try:
for highlight in self.insta.get_highlights(profile.userid):
for item in highlight.get_items():
try: self.insta.download_storyitem(item, target=f"highlight_item_{highlight.owner_username}")
except Exception as e: logger.error(f"Failed to download highlight item: {item}: {e}")
except Exception as e: logger.error(f"Failed get_highlights: {e}")
try:
self.insta.download_storyitem(item, target=f"highlight_item_{highlight.owner_username}")
except Exception as e:
logger.error(f"Failed to download highlight item: {item}: {e}")
except Exception as e:
logger.error(f"Failed get_highlights: {e}")
return self.process_downloads(url, f"@{username}", profile._asdict(), None)
@@ -124,7 +142,8 @@ class InstagramExtractor(Extractor):
all_media = []
for f in os.listdir(self.download_folder):
if os.path.isfile((filename := os.path.join(self.download_folder, f))):
if filename[-4:] == ".txt": continue
if filename[-4:] == ".txt":
continue
all_media.append(Media(filename))
assert len(all_media) > 1, "No uploaded media found"

21
src/auto_archiver/modules/instagram_tbot_extractor/__manifest__.py
View File

@@ -1,16 +1,21 @@
{
"name": "Instagram Telegram Bot Extractor",
"type": ["extractor"],
"dependencies": {"python": ["loguru", "telethon",],
},
"dependencies": {
"python": [
"loguru",
"telethon",
],
},
"requires_setup": True,
"configs": {
"api_id": {"default": None, "help": "telegram API_ID value, go to https://my.telegram.org/apps"},
"api_hash": {"default": None, "help": "telegram API_HASH value, go to https://my.telegram.org/apps"},
"session_file": {"default": "secrets/anon-insta", "help": "optional, records the telegram login session for future usage, '.session' will be appended to the provided value."},
"timeout": {"default": 45,
"type": "int",
"help": "timeout to fetch the instagram content in seconds."},
"api_id": {"default": None, "help": "telegram API_ID value, go to https://my.telegram.org/apps"},
"api_hash": {"default": None, "help": "telegram API_HASH value, go to https://my.telegram.org/apps"},
"session_file": {
"default": "secrets/anon-insta",
"help": "optional, records the telegram login session for future usage, '.session' will be appended to the provided value.",
},
"timeout": {"default": 45, "type": "int", "help": "timeout to fetch the instagram content in seconds."},
},
"description": """
The `InstagramTbotExtractor` module uses a Telegram bot (`instagram_load_bot`) to fetch and archive Instagram content,

15
src/auto_archiver/modules/instagram_tbot_extractor/instagram_tbot_extractor.py
View File

@@ -65,15 +65,15 @@ class InstagramTbotExtractor(Extractor):
session_file_name = self.session_file + ".session"
if os.path.exists(session_file_name):
os.remove(session_file_name)
def download(self, item: Metadata) -> Metadata:
url = item.get_url()
if not "instagram.com" in url: return False
if not "instagram.com" in url:
return False
result = Metadata()
tmp_dir = self.tmp_dir
with self.client.start():
chat, since_id = self._send_url_to_bot(url)
message = self._process_messages(chat, since_id, tmp_dir, result)
@@ -110,13 +110,14 @@ class InstagramTbotExtractor(Extractor):
for post in self.client.iter_messages(chat, min_id=since_id):
since_id = max(since_id, post.id)
# Skip known filler message:
if post.message == 'The bot receives information through https://hikerapi.com/p/hJqpppqi':
if post.message == "The bot receives information through https://hikerapi.com/p/hJqpppqi":
continue
if post.media and post.id not in seen_media:
filename_dest = os.path.join(tmp_dir, f'{chat.id}_{post.id}')
filename_dest = os.path.join(tmp_dir, f"{chat.id}_{post.id}")
media = self.client.download_media(post.media, filename_dest)
if media:
result.add_media(Media(media))
seen_media.append(post.id)
if post.message: message += post.message
return message.strip()
if post.message:
message += post.message
return message.strip()

2
src/auto_archiver/modules/local_storage/__init__.py
View File

@@ -1 +1 @@
from .local_storage import LocalStorage
from .local_storage import LocalStorage

10
src/auto_archiver/modules/local_storage/__manifest__.py
View File

@@ -17,9 +17,11 @@
"choices": ["random", "static"],
},
"save_to": {"default": "./local_archive", "help": "folder where to save archived content"},
"save_absolute": {"default": False,
"type": "bool",
"help": "whether the path to the stored file is absolute or relative in the output result inc. formatters (WARN: leaks the file structure)"},
"save_absolute": {
"default": False,
"type": "bool",
"help": "whether the path to the stored file is absolute or relative in the output result inc. formatters (WARN: leaks the file structure)",
},
},
"description": """
LocalStorage: A storage module for saving archived content locally on the filesystem.
@@ -33,5 +35,5 @@
### Notes
- Default storage folder is `./archived`, but this can be changed via the `save_to` configuration.
- The `save_absolute` option can reveal the file structure in output formats; use with caution.
"""
""",
}

7
src/auto_archiver/modules/local_storage/local_storage.py
View File

@@ -1,4 +1,3 @@
import shutil
from typing import IO
import os
@@ -9,7 +8,6 @@ from auto_archiver.core import Storage
class LocalStorage(Storage):
def get_cdn_url(self, media: Media) -> str:
# TODO: is this viable with Storage.configs on path/filename?
dest = os.path.join(self.save_to, media.key)
@@ -21,10 +19,11 @@ class LocalStorage(Storage):
# override parent so that we can use shutil.copy2 and keep metadata
dest = os.path.join(self.save_to, media.key)
os.makedirs(os.path.dirname(dest), exist_ok=True)
logger.debug(f'[{self.__class__.__name__}] storing file {media.filename} with key {media.key} to {dest}')
logger.debug(f"[{self.__class__.__name__}] storing file {media.filename} with key {media.key} to {dest}")
res = shutil.copy2(media.filename, dest)
logger.info(res)
return True
# must be implemented even if unused
def uploadf(self, file: IO[bytes], key: str, **kwargs: dict) -> bool: pass
def uploadf(self, file: IO[bytes], key: str, **kwargs: dict) -> bool:
pass

2
src/auto_archiver/modules/meta_enricher/__manifest__.py
View File

@@ -3,7 +3,7 @@
"type": ["enricher"],
"requires_setup": False,
"dependencies": {
"python": ["loguru"],
"python": ["loguru"],
},
"description": """
Adds metadata information about the archive operations, Adds metadata about archive operations, including file sizes and archive duration./

7
src/auto_archiver/modules/meta_enricher/meta_enricher.py
View File

@@ -23,7 +23,9 @@ class MetaEnricher(Enricher):
self.enrich_archive_duration(to_enrich)
def enrich_file_sizes(self, to_enrich: Metadata):
logger.debug(f"calculating archive file sizes for url={to_enrich.get_url()} ({len(to_enrich.media)} media files)")
logger.debug(
f"calculating archive file sizes for url={to_enrich.get_url()} ({len(to_enrich.media)} media files)"
)
total_size = 0
for media in to_enrich.get_all_media():
file_stats = os.stat(media.filename)
@@ -34,7 +36,6 @@ class MetaEnricher(Enricher):
to_enrich.set("total_bytes", total_size)
to_enrich.set("total_size", self.human_readable_bytes(total_size))
def human_readable_bytes(self, size: int) -> str:
# receives number of bytes and returns human readble size
for unit in ["bytes", "KB", "MB", "GB", "TB"]:
@@ -46,4 +47,4 @@ class MetaEnricher(Enricher):
logger.debug(f"calculating archive duration for url={to_enrich.get_url()} ")
archive_duration = datetime.datetime.now(datetime.timezone.utc) - to_enrich.get("_processed_at")
to_enrich.set("archive_duration_seconds", archive_duration.seconds)
to_enrich.set("archive_duration_seconds", archive_duration.seconds)

2
src/auto_archiver/modules/metadata_enricher/__init__.py
View File

@@ -1 +1 @@
from .metadata_enricher import MetadataEnricher
from .metadata_enricher import MetadataEnricher

7
src/auto_archiver/modules/metadata_enricher/__manifest__.py
View File

@@ -2,10 +2,7 @@
"name": "Media Metadata Enricher",
"type": ["enricher"],
"requires_setup": True,
"dependencies": {
"python": ["loguru"],
"bin": ["exiftool"]
},
"dependencies": {"python": ["loguru"], "bin": ["exiftool"]},
"description": """
Extracts metadata information from files using ExifTool.
@@ -17,5 +14,5 @@
### Notes
- Requires ExifTool to be installed and accessible via the system's PATH.
- Skips enrichment for files where metadata extraction fails.
"""
""",
}

5
src/auto_archiver/modules/metadata_enricher/metadata_enricher.py
View File

@@ -11,7 +11,6 @@ class MetadataEnricher(Enricher):
Extracts metadata information from files using exiftool.
"""
def enrich(self, to_enrich: Metadata) -> None:
url = to_enrich.get_url()
logger.debug(f"extracting EXIF metadata for {url=}")
@@ -23,13 +22,13 @@ class MetadataEnricher(Enricher):
def get_metadata(self, filename: str) -> dict:
try:
# Run ExifTool command to extract metadata from the file
cmd = ['exiftool', filename]
cmd = ["exiftool", filename]
result = subprocess.run(cmd, capture_output=True, text=True)
# Process the output to extract individual metadata fields
metadata = {}
for line in result.stdout.splitlines():
field, value = line.strip().split(':', 1)
field, value = line.strip().split(":", 1)
metadata[field.strip()] = value.strip()
return metadata
except FileNotFoundError:

3
src/auto_archiver/modules/mute_formatter/__manifest__.py
View File

@@ -2,8 +2,7 @@
"name": "Mute Formatter",
"type": ["formatter"],
"requires_setup": True,
"dependencies": {
},
"dependencies": {},
"description": """ Default formatter.
""",
}

4
src/auto_archiver/modules/mute_formatter/mute_formatter.py
View File

@@ -5,5 +5,5 @@ from auto_archiver.core import Formatter
class MuteFormatter(Formatter):
def format(self, item: Metadata) -> Media: return None
def format(self, item: Metadata) -> Media:
return None

2
src/auto_archiver/modules/pdq_hash_enricher/__init__.py
View File

@@ -1 +1 @@
from .pdq_hash_enricher import PdqHashEnricher
from .pdq_hash_enricher import PdqHashEnricher

2
src/auto_archiver/modules/pdq_hash_enricher/__manifest__.py
View File

@@ -17,5 +17,5 @@
### Notes
- Best used after enrichers like `thumbnail_enricher` or `screenshot_enricher` to ensure images are available.
- Uses the `pdqhash` library to compute 256-bit perceptual hashes, which are stored as hexadecimal strings.
"""
""",
}

12
src/auto_archiver/modules/pdq_hash_enricher/pdq_hash_enricher.py
View File

@@ -10,6 +10,7 @@ This enricher is typically used after thumbnail or screenshot enrichers
to ensure images are available for hashing.
"""
import traceback
import pdqhash
import numpy as np
@@ -34,7 +35,12 @@ class PdqHashEnricher(Enricher):
for m in to_enrich.media:
for media in m.all_inner_media(True):
media_id = media.get("id", "")
if media.is_image() and "screenshot" not in media_id and "warc-file-" not in media_id and len(hd := self.calculate_pdq_hash(media.filename)):
if (
media.is_image()
and "screenshot" not in media_id
and "warc-file-" not in media_id
and len(hd := self.calculate_pdq_hash(media.filename))
):
media.set("pdq_hash", hd)
media_with_hashes.append(media.filename)
@@ -51,5 +57,7 @@ class PdqHashEnricher(Enricher):
hash = "".join(str(b) for b in hash_array)
return hex(int(hash, 2))[2:]
except UnidentifiedImageError as e:
logger.error(f"Image {filename=} is likely corrupted or in unsupported format {e}: {traceback.format_exc()}")
logger.error(
f"Image {filename=} is likely corrupted or in unsupported format {e}: {traceback.format_exc()}"
)
return ""

2
src/auto_archiver/modules/s3_storage/__init__.py
View File

@@ -1 +1 @@
from .s3_storage import S3Storage
from .s3_storage import S3Storage

22
src/auto_archiver/modules/s3_storage/__manifest__.py
View File

@@ -20,20 +20,20 @@
"region": {"default": None, "help": "S3 region name"},
"key": {"default": None, "help": "S3 API key"},
"secret": {"default": None, "help": "S3 API secret"},
"random_no_duplicate": {"default": False,
"type": "bool",
"help": "if set, it will override `path_generator`, `filename_generator` and `folder`. It will check if the file already exists and if so it will not upload it again. Creates a new root folder path `no-dups/`"},
"random_no_duplicate": {
"default": False,
"type": "bool",
"help": "if set, it will override `path_generator`, `filename_generator` and `folder`. It will check if the file already exists and if so it will not upload it again. Creates a new root folder path `no-dups/`",
},
"endpoint_url": {
"default": 'https://{region}.digitaloceanspaces.com',
"help": "S3 bucket endpoint, {region} are inserted at runtime"
"default": "https://{region}.digitaloceanspaces.com",
"help": "S3 bucket endpoint, {region} are inserted at runtime",
},
"cdn_url": {
"default": 'https://{bucket}.{region}.cdn.digitaloceanspaces.com/{key}',
"help": "S3 CDN url, {bucket}, {region} and {key} are inserted at runtime"
"default": "https://{bucket}.{region}.cdn.digitaloceanspaces.com/{key}",
"help": "S3 CDN url, {bucket}, {region} and {key} are inserted at runtime",
},
"private": {"default": False,
"type": "bool",
"help": "if true S3 files will not be readable online"},
"private": {"default": False, "type": "bool", "help": "if true S3 files will not be readable online"},
},
"description": """
S3Storage: A storage module for saving media files to an S3-compatible object storage.
@@ -50,5 +50,5 @@
- The `random_no_duplicate` option ensures no duplicate uploads by leveraging hash-based folder structures.
- Uses `boto3` for interaction with the S3 API.
- Depends on the `HashEnricher` module for hash calculation.
"""
""",
}

41
src/auto_archiver/modules/s3_storage/s3_storage.py
View File

@@ -1,4 +1,3 @@
from typing import IO
import boto3
@@ -11,60 +10,62 @@ from auto_archiver.utils.misc import calculate_file_hash, random_str
NO_DUPLICATES_FOLDER = "no-dups/"
class S3Storage(Storage):
class S3Storage(Storage):
def setup(self) -> None:
self.s3 = boto3.client(
's3',
"s3",
region_name=self.region,
endpoint_url=self.endpoint_url.format(region=self.region),
aws_access_key_id=self.key,
aws_secret_access_key=self.secret
aws_secret_access_key=self.secret,
)
if self.random_no_duplicate:
logger.warning("random_no_duplicate is set to True, this will override `path_generator`, `filename_generator` and `folder`.")
logger.warning(
"random_no_duplicate is set to True, this will override `path_generator`, `filename_generator` and `folder`."
)
def get_cdn_url(self, media: Media) -> str:
return self.cdn_url.format(bucket=self.bucket, region=self.region, key=media.key)
def uploadf(self, file: IO[bytes], media: Media, **kwargs: dict) -> None:
if not self.is_upload_needed(media): return True
if not self.is_upload_needed(media):
return True
extra_args = kwargs.get("extra_args", {})
if not self.private and 'ACL' not in extra_args:
extra_args['ACL'] = 'public-read'
if not self.private and "ACL" not in extra_args:
extra_args["ACL"] = "public-read"
if 'ContentType' not in extra_args:
if "ContentType" not in extra_args:
try:
if media.mimetype:
extra_args['ContentType'] = media.mimetype
extra_args["ContentType"] = media.mimetype
except Exception as e:
logger.warning(f"Unable to get mimetype for {media.key=}, error: {e}")
self.s3.upload_fileobj(file, Bucket=self.bucket, Key=media.key, ExtraArgs=extra_args)
return True
def is_upload_needed(self, media: Media) -> bool:
if self.random_no_duplicate:
# checks if a folder with the hash already exists, if so it skips the upload
hd = calculate_file_hash(media.filename)
path = os.path.join(NO_DUPLICATES_FOLDER, hd[:24])
if existing_key:=self.file_in_folder(path):
if existing_key := self.file_in_folder(path):
media.key = existing_key
media.set("previously archived", True)
logger.debug(f"skipping upload of {media.filename} because it already exists in {media.key}")
return False
_, ext = os.path.splitext(media.key)
media.key = os.path.join(path, f"{random_str(24)}{ext}")
return True
def file_in_folder(self, path:str) -> str:
def file_in_folder(self, path: str) -> str:
# checks if path exists and is not an empty folder
if not path.endswith('/'):
path = path + '/'
resp = self.s3.list_objects(Bucket=self.bucket, Prefix=path, Delimiter='/', MaxKeys=1)
if 'Contents' in resp:
return resp['Contents'][0]['Key']
if not path.endswith("/"):
path = path + "/"
resp = self.s3.list_objects(Bucket=self.bucket, Prefix=path, Delimiter="/", MaxKeys=1)
if "Contents" in resp:
return resp["Contents"][0]["Key"]
return False

43
src/auto_archiver/modules/screenshot_enricher/__manifest__.py
View File

@@ -6,26 +6,29 @@
"python": ["loguru", "selenium"],
},
"configs": {
"width": {"default": 1280,
"type": "int",
"help": "width of the screenshots"},
"height": {"default": 1024,
"type": "int",
"help": "height of the screenshots"},
"timeout": {"default": 60,
"type": "int",
"help": "timeout for taking the screenshot"},
"sleep_before_screenshot": {"default": 4,
"type": "int",
"help": "seconds to wait for the pages to load before taking screenshot"},
"http_proxy": {"default": "", "help": "http proxy to use for the webdriver, eg http://proxy-user:password@proxy-ip:port"},
"save_to_pdf": {"default": False,
"type": "bool",
"help": "save the page as pdf along with the screenshot. PDF saving options can be adjusted with the 'print_options' parameter"},
"print_options": {"default": {},
"help": "options to pass to the pdf printer, in JSON format. See https://www.selenium.dev/documentation/webdriver/interactions/print_page/ for more information",
"type": "json_loader"},
"width": {"default": 1280, "type": "int", "help": "width of the screenshots"},
"height": {"default": 1024, "type": "int", "help": "height of the screenshots"},
"timeout": {"default": 60, "type": "int", "help": "timeout for taking the screenshot"},
"sleep_before_screenshot": {
"default": 4,
"type": "int",
"help": "seconds to wait for the pages to load before taking screenshot",
},
"http_proxy": {
"default": "",
"help": "http proxy to use for the webdriver, eg http://proxy-user:password@proxy-ip:port",
},
"save_to_pdf": {
"default": False,
"type": "bool",
"help": "save the page as pdf along with the screenshot. PDF saving options can be adjusted with the 'print_options' parameter",
},
"print_options": {
"default": {},
"help": "options to pass to the pdf printer, in JSON format. See https://www.selenium.dev/documentation/webdriver/interactions/print_page/ for more information",
"type": "json_loader",
},
},
"description": """
Captures screenshots and optionally saves web pages as PDFs using a WebDriver.
@@ -37,5 +40,5 @@
### Notes
- Requires a WebDriver (e.g., ChromeDriver) installed and accessible via the system's PATH.
"""
""",
}

13
src/auto_archiver/modules/screenshot_enricher/screenshot_enricher.py
View File

@@ -9,8 +9,8 @@ from auto_archiver.core import Enricher
from auto_archiver.utils import Webdriver, url as UrlUtil, random_str
from auto_archiver.core import Media, Metadata
class ScreenshotEnricher(Enricher):
class ScreenshotEnricher(Enricher):
def __init__(self, webdriver_factory=None):
super().__init__()
self.webdriver_factory = webdriver_factory or Webdriver
@@ -25,8 +25,14 @@ class ScreenshotEnricher(Enricher):
logger.debug(f"Enriching screenshot for {url=}")
auth = self.auth_for_site(url)
with self.webdriver_factory(
self.width, self.height, self.timeout, facebook_accept_cookies='facebook.com' in url,
http_proxy=self.http_proxy, print_options=self.print_options, auth=auth) as driver:
self.width,
self.height,
self.timeout,
facebook_accept_cookies="facebook.com" in url,
http_proxy=self.http_proxy,
print_options=self.print_options,
auth=auth,
) as driver:
try:
driver.get(url)
time.sleep(int(self.sleep_before_screenshot))
@@ -43,4 +49,3 @@ class ScreenshotEnricher(Enricher):
logger.info("TimeoutException loading page for screenshot")
except Exception as e:
logger.error(f"Got error while loading webdriver for screenshot enricher: {e}")

2
src/auto_archiver/modules/ssl_enricher/__init__.py
View File

@@ -1 +1 @@
from .ssl_enricher import SSLEnricher
from .ssl_enricher import SSLEnricher

12
src/auto_archiver/modules/ssl_enricher/__manifest__.py
View File

@@ -5,11 +5,13 @@
"dependencies": {
"python": ["loguru", "slugify"],
},
'entry_point': 'ssl_enricher::SSLEnricher',
"entry_point": "ssl_enricher::SSLEnricher",
"configs": {
"skip_when_nothing_archived": {"default": True,
"type": 'bool',
"help": "if true, will skip enriching when no media is archived"},
"skip_when_nothing_archived": {
"default": True,
"type": "bool",
"help": "if true, will skip enriching when no media is archived",
},
},
"description": """
Retrieves SSL certificate information for a domain and stores it as a file.
@@ -21,5 +23,5 @@
### Notes
- Requires the target URL to use the HTTPS scheme; other schemes are not supported.
"""
""",
}

10
src/auto_archiver/modules/ssl_enricher/ssl_enricher.py
View File

@@ -13,16 +13,18 @@ class SSLEnricher(Enricher):
"""
def enrich(self, to_enrich: Metadata) -> None:
if not to_enrich.media and self.skip_when_nothing_archived: return
if not to_enrich.media and self.skip_when_nothing_archived:
return
url = to_enrich.get_url()
parsed = urlparse(url)
assert parsed.scheme in ["https"], f"Invalid URL scheme {url=}"
domain = parsed.netloc
logger.debug(f"fetching SSL certificate for {domain=} in {url=}")
cert = ssl.get_server_certificate((domain, 443))
cert_fn = os.path.join(self.tmp_dir, f"{slugify(domain)}.pem")
with open(cert_fn, "w") as f: f.write(cert)
with open(cert_fn, "w") as f:
f.write(cert)
to_enrich.add_media(Media(filename=cert_fn), id="ssl_certificate")

2
src/auto_archiver/modules/telegram_extractor/__init__.py
View File

@@ -1 +1 @@
from .telegram_extractor import TelegramExtractor
from .telegram_extractor import TelegramExtractor

25
src/auto_archiver/modules/telegram_extractor/telegram_extractor.py
View File

@@ -15,11 +15,11 @@ class TelegramExtractor(Extractor):
def download(self, item: Metadata) -> Metadata:
url = item.get_url()
# detect URLs that we definitely cannot handle
if 't.me' != item.netloc:
if "t.me" != item.netloc:
return False
headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.138 Safari/537.36'
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.138 Safari/537.36"
}
# TODO: check if we can do this more resilient to variable URLs
@@ -27,11 +27,11 @@ class TelegramExtractor(Extractor):
url += "?embed=1"
t = requests.get(url, headers=headers)
s = BeautifulSoup(t.content, 'html.parser')
s = BeautifulSoup(t.content, "html.parser")
result = Metadata()
result.set_content(html.escape(str(t.content)))
if (timestamp := (s.find_all('time') or [{}])[0].get('datetime')):
if timestamp := (s.find_all("time") or [{}])[0].get("datetime"):
result.set_timestamp(timestamp)
video = s.find("video")
@@ -41,25 +41,26 @@ class TelegramExtractor(Extractor):
image_urls = []
for im in image_tags:
urls = [u.replace("'", "") for u in re.findall(r'url\((.*?)\)', im['style'])]
urls = [u.replace("'", "") for u in re.findall(r"url\((.*?)\)", im["style"])]
image_urls += urls
if not len(image_urls): return False
if not len(image_urls):
return False
for img_url in image_urls:
result.add_media(Media(self.download_from_url(img_url)))
else:
video_url = video.get('src')
video_url = video.get("src")
m_video = Media(self.download_from_url(video_url))
# extract duration from HTML
try:
duration = s.find_all('time')[0].contents[0]
if ':' in duration:
duration = float(duration.split(
':')[0]) * 60 + float(duration.split(':')[1])
duration = s.find_all("time")[0].contents[0]
if ":" in duration:
duration = float(duration.split(":")[0]) * 60 + float(duration.split(":")[1])
else:
duration = float(duration)
m_video.set("duration", duration)
except: pass
except:
pass
result.add_media(m_video)
return result.success("telegram")

2
src/auto_archiver/modules/telethon_extractor/__init__.py
View File

@@ -1 +1 @@
from .telethon_extractor import TelethonExtractor
from .telethon_extractor import TelethonExtractor

47
src/auto_archiver/modules/telethon_extractor/__manifest__.py
View File

@@ -3,26 +3,35 @@
"type": ["extractor"],
"requires_setup": True,
"dependencies": {
"python": ["telethon",
"loguru",
"tqdm",
],
"bin": [""]
"python": [
"telethon",
"loguru",
"tqdm",
],
"bin": [""],
},
"configs": {
"api_id": {"default": None, "help": "telegram API_ID value, go to https://my.telegram.org/apps"},
"api_hash": {"default": None, "help": "telegram API_HASH value, go to https://my.telegram.org/apps"},
"bot_token": {"default": None, "help": "optional, but allows access to more content such as large videos, talk to @botfather"},
"session_file": {"default": "secrets/anon", "help": "optional, records the telegram login session for future usage, '.session' will be appended to the provided value."},
"join_channels": {"default": True,
"type": "bool",
"help": "disables the initial setup with channel_invites config, useful if you have a lot and get stuck"},
"channel_invites": {
"default": {},
"help": "(JSON string) private channel invite links (format: t.me/joinchat/HASH OR t.me/+HASH) and (optional but important to avoid hanging for minutes on startup) channel id (format: CHANNEL_ID taken from a post url like https://t.me/c/CHANNEL_ID/1), the telegram account will join any new channels on setup",
"type": "json_loader",
}
"api_id": {"default": None, "help": "telegram API_ID value, go to https://my.telegram.org/apps"},
"api_hash": {"default": None, "help": "telegram API_HASH value, go to https://my.telegram.org/apps"},
"bot_token": {
"default": None,
"help": "optional, but allows access to more content such as large videos, talk to @botfather",
},
"session_file": {
"default": "secrets/anon",
"help": "optional, records the telegram login session for future usage, '.session' will be appended to the provided value.",
},
"join_channels": {
"default": True,
"type": "bool",
"help": "disables the initial setup with channel_invites config, useful if you have a lot and get stuck",
},
"channel_invites": {
"default": {},
"help": "(JSON string) private channel invite links (format: t.me/joinchat/HASH OR t.me/+HASH) and (optional but important to avoid hanging for minutes on startup) channel id (format: CHANNEL_ID taken from a post url like https://t.me/c/CHANNEL_ID/1), the telegram account will join any new channels on setup",
"type": "json_loader",
},
},
"description": """
The `TelethonExtractor` uses the Telethon library to archive posts and media from Telegram channels and groups.
It supports private and public channels, downloading grouped posts with media, and can join channels using invite links
@@ -46,5 +55,5 @@ To use the `TelethonExtractor`, you must configure the following:
The first time you run, you will be prompted to do a authentication with the phone number associated, alternatively you can put your `anon.session` in the root.
"""
}
""",
}

47
src/auto_archiver/modules/telethon_extractor/telethon_extractor.py
View File

@@ -1,9 +1,13 @@
import shutil
from telethon.sync import TelegramClient
from telethon.errors import ChannelInvalidError
from telethon.tl.functions.messages import ImportChatInviteRequest
from telethon.errors.rpcerrorlist import UserAlreadyParticipantError, FloodWaitError, InviteRequestSentError, InviteHashExpiredError
from telethon.errors.rpcerrorlist import (
UserAlreadyParticipantError,
FloodWaitError,
InviteRequestSentError,
InviteHashExpiredError,
)
from loguru import logger
from tqdm import tqdm
import re, time, os
@@ -17,9 +21,7 @@ class TelethonExtractor(Extractor):
valid_url = re.compile(r"https:\/\/t\.me(\/c){0,1}\/(.+)\/(\d+)")
invite_pattern = re.compile(r"t.me(\/joinchat){0,1}\/\+?(.+)")
def setup(self) -> None:
"""
1. makes a copy of session_file that is removed in cleanup
2. trigger login process for telegram or proceed if already saved in a session file
@@ -34,7 +36,7 @@ class TelethonExtractor(Extractor):
# initiate the client
self.client = TelegramClient(self.session_file, self.api_id, self.api_hash)
with self.client.start():
logger.success(f"SETUP {self.name} login works.")
@@ -52,13 +54,15 @@ class TelethonExtractor(Extractor):
channel_invite = self.channel_invites[i]
channel_id = channel_invite.get("id", False)
invite = channel_invite["invite"]
if (match := self.invite_pattern.search(invite)):
if match := self.invite_pattern.search(invite):
try:
if channel_id:
ent = self.client.get_entity(int(channel_id)) # fails if not a member
else:
ent = self.client.get_entity(invite) # fails if not a member
logger.warning(f"please add the property id='{ent.id}' to the 'channel_invites' configuration where {invite=}, not doing so can lead to a minutes-long setup time due to telegram's rate limiting.")
logger.warning(
f"please add the property id='{ent.id}' to the 'channel_invites' configuration where {invite=}, not doing so can lead to a minutes-long setup time due to telegram's rate limiting."
)
except ValueError as e:
logger.info(f"joining new channel {invite=}")
try:
@@ -95,7 +99,8 @@ class TelethonExtractor(Extractor):
# detect URLs that we definitely cannot handle
match = self.valid_url.search(url)
logger.debug(f"TELETHON: {match=}")
if not match: return False
if not match:
return False
is_private = match.group(1) == "/c"
chat = int(match.group(2)) if is_private else match.group(2)
@@ -105,45 +110,53 @@ class TelethonExtractor(Extractor):
# NB: not using bot_token since then private channels cannot be archived: self.client.start(bot_token=self.bot_token)
with self.client.start():
# with self.client.start(bot_token=self.bot_token):
# with self.client.start(bot_token=self.bot_token):
try:
post = self.client.get_messages(chat, ids=post_id)
except ValueError as e:
logger.error(f"Could not fetch telegram {url} possibly it's private: {e}")
return False
except ChannelInvalidError as e:
logger.error(f"Could not fetch telegram {url}. This error may be fixed if you setup a bot_token in addition to api_id and api_hash (but then private channels will not be archived, we need to update this logic to handle both): {e}")
logger.error(
f"Could not fetch telegram {url}. This error may be fixed if you setup a bot_token in addition to api_id and api_hash (but then private channels will not be archived, we need to update this logic to handle both): {e}"
)
return False
logger.debug(f"TELETHON GOT POST {post=}")
if post is None: return False
if post is None:
return False
media_posts = self._get_media_posts_in_group(chat, post)
logger.debug(f'got {len(media_posts)=} for {url=}')
logger.debug(f"got {len(media_posts)=} for {url=}")
tmp_dir = self.tmp_dir
group_id = post.grouped_id if post.grouped_id is not None else post.id
title = post.message
for mp in media_posts:
if len(mp.message) > len(title): title = mp.message # save the longest text found (usually only 1)
if len(mp.message) > len(title):
title = mp.message # save the longest text found (usually only 1)
# media can also be in entities
if mp.entities:
other_media_urls = [e.url for e in mp.entities if hasattr(e, "url") and e.url and self._guess_file_type(e.url) in ["video", "image", "audio"]]
other_media_urls = [
e.url
for e in mp.entities
if hasattr(e, "url") and e.url and self._guess_file_type(e.url) in ["video", "image", "audio"]
]
if len(other_media_urls):
logger.debug(f"Got {len(other_media_urls)} other media urls from {mp.id=}: {other_media_urls}")
for i, om_url in enumerate(other_media_urls):
filename = self.download_from_url(om_url, f'{chat}_{group_id}_{i}')
filename = self.download_from_url(om_url, f"{chat}_{group_id}_{i}")
result.add_media(Media(filename=filename), id=f"{group_id}_{i}")
filename_dest = os.path.join(tmp_dir, f'{chat}_{group_id}', str(mp.id))
filename_dest = os.path.join(tmp_dir, f"{chat}_{group_id}", str(mp.id))
filename = self.client.download_media(mp.media, filename_dest)
if not filename:
logger.debug(f"Empty media found, skipping {str(mp)=}")
continue
result.add_media(Media(filename))
result.set_title(title).set_timestamp(post.date).set("api_data", post.to_dict())
if post.message != title:
result.set_content(post.message)

23
src/auto_archiver/modules/thumbnail_enricher/__manifest__.py
View File

@@ -2,18 +2,19 @@
"name": "Thumbnail Enricher",
"type": ["enricher"],
"requires_setup": False,
"dependencies": {
"python": ["loguru", "ffmpeg"],
"bin": ["ffmpeg"]
},
"dependencies": {"python": ["loguru", "ffmpeg"], "bin": ["ffmpeg"]},
"configs": {
"thumbnails_per_minute": {"default": 60,
"type": "int",
"help": "how many thumbnails to generate per minute of video, can be limited by max_thumbnails"},
"max_thumbnails": {"default": 16,
"type": "int",
"help": "limit the number of thumbnails to generate per video, 0 means no limit"},
"thumbnails_per_minute": {
"default": 60,
"type": "int",
"help": "how many thumbnails to generate per minute of video, can be limited by max_thumbnails",
},
"max_thumbnails": {
"default": 16,
"type": "int",
"help": "limit the number of thumbnails to generate per video, 0 means no limit",
},
},
"description": """
Generates thumbnails for video files to provide visual previews.
@@ -27,5 +28,5 @@
- Requires `ffmpeg` to be installed and accessible via the system's PATH.
- Handles videos without pre-existing duration metadata by probing with `ffmpeg`.
- Skips enrichment for non-video media files.
"""
""",
}

15
src/auto_archiver/modules/thumbnail_enricher/thumbnail_enricher.py
View File

@@ -6,6 +6,7 @@ visual snapshots of the video's keyframes, helping users preview content
and identify important moments without watching the entire video.
"""
import ffmpeg, os
from loguru import logger
@@ -18,7 +19,7 @@ class ThumbnailEnricher(Enricher):
"""
Generates thumbnails for all the media
"""
def enrich(self, to_enrich: Metadata) -> None:
"""
Uses or reads the video duration to generate thumbnails
@@ -36,7 +37,9 @@ class ThumbnailEnricher(Enricher):
if duration is None:
try:
probe = ffmpeg.probe(m.filename)
duration = float(next(stream for stream in probe['streams'] if stream['codec_type'] == 'video')['duration'])
duration = float(
next(stream for stream in probe["streams"] if stream["codec_type"] == "video")["duration"]
)
to_enrich.media[m_id].set("duration", duration)
except Exception as e:
logger.error(f"error getting duration of video {m.filename}: {e}")
@@ -48,11 +51,13 @@ class ThumbnailEnricher(Enricher):
thumbnails_media = []
for index, timestamp in enumerate(timestamps):
output_path = os.path.join(folder, f"out{index}.jpg")
ffmpeg.input(m.filename, ss=timestamp).filter('scale', 512, -1).output(output_path, vframes=1, loglevel="quiet").run()
ffmpeg.input(m.filename, ss=timestamp).filter("scale", 512, -1).output(
output_path, vframes=1, loglevel="quiet"
).run()
try:
thumbnails_media.append(Media(
filename=output_path)
thumbnails_media.append(
Media(filename=output_path)
.set("id", f"thumbnail_{index}")
.set("timestamp", "%.3fs" % timestamp)
)

49
src/auto_archiver/modules/timestamping_enricher/__manifest__.py
View File

@@ -3,38 +3,29 @@
"type": ["enricher"],
"requires_setup": True,
"dependencies": {
"python": [
"loguru",
"slugify",
"tsp_client",
"asn1crypto",
"certvalidator",
"certifi"
],
"python": ["loguru", "slugify", "tsp_client", "asn1crypto", "certvalidator", "certifi"],
},
"configs": {
"tsa_urls": {
"default": [
# [Adobe Approved Trust List] and [Windows Cert Store]
"http://timestamp.digicert.com",
"http://timestamp.identrust.com",
# "https://timestamp.entrust.net/TSS/RFC3161sha2TS", # not valid for timestamping
# "https://timestamp.sectigo.com", # wait 15 seconds between each request.
# [Adobe: European Union Trusted Lists].
# "https://timestamp.sectigo.com/qualified", # wait 15 seconds between each request.
# [Windows Cert Store]
"http://timestamp.globalsign.com/tsa/r6advanced1",
# [Adobe: European Union Trusted Lists] and [Windows Cert Store]
# "http://ts.quovadisglobal.com/eu", # not valid for timestamping
# "http://tsa.belgium.be/connect", # self-signed certificate in certificate chain
# "https://timestamp.aped.gov.gr/qtss", # self-signed certificate in certificate chain
# "http://tsa.sep.bg", # self-signed certificate in certificate chain
# "http://tsa.izenpe.com", #unable to get local issuer certificate
# "http://kstamp.keynectis.com/KSign", # unable to get local issuer certificate
"http://tss.accv.es:8318/tsa",
],
# [Adobe Approved Trust List] and [Windows Cert Store]
"http://timestamp.digicert.com",
"http://timestamp.identrust.com",
# "https://timestamp.entrust.net/TSS/RFC3161sha2TS", # not valid for timestamping
# "https://timestamp.sectigo.com", # wait 15 seconds between each request.
# [Adobe: European Union Trusted Lists].
# "https://timestamp.sectigo.com/qualified", # wait 15 seconds between each request.
# [Windows Cert Store]
"http://timestamp.globalsign.com/tsa/r6advanced1",
# [Adobe: European Union Trusted Lists] and [Windows Cert Store]
# "http://ts.quovadisglobal.com/eu", # not valid for timestamping
# "http://tsa.belgium.be/connect", # self-signed certificate in certificate chain
# "https://timestamp.aped.gov.gr/qtss", # self-signed certificate in certificate chain
# "http://tsa.sep.bg", # self-signed certificate in certificate chain
# "http://tsa.izenpe.com", #unable to get local issuer certificate
# "http://kstamp.keynectis.com/KSign", # unable to get local issuer certificate
"http://tss.accv.es:8318/tsa",
],
"help": "List of RFC3161 Time Stamp Authorities to use, separate with commas if passed via the command line.",
}
},
@@ -50,5 +41,5 @@
### Notes
- Should be run after the `hash_enricher` to ensure file hashes are available.
- Requires internet access to interact with the configured TSAs.
"""
""",
}

23
src/auto_archiver/modules/timestamping_enricher/timestamping_enricher.py
View File

@@ -11,6 +11,7 @@ import certifi
from auto_archiver.core import Enricher
from auto_archiver.core import Metadata, Media
class TimestampingEnricher(Enricher):
"""
Uses several RFC3161 Time Stamp Authorities to generate a timestamp token that will be preserved. This can be used to prove that a certain file existed at a certain time, useful for legal purposes, for example, to prove that a certain file was not tampered with after a certain date.
@@ -25,27 +26,30 @@ class TimestampingEnricher(Enricher):
logger.debug(f"RFC3161 timestamping existing files for {url=}")
# create a new text file with the existing media hashes
hashes = [m.get("hash").replace("SHA-256:", "").replace("SHA3-512:", "") for m in to_enrich.media if m.get("hash")]
hashes = [
m.get("hash").replace("SHA-256:", "").replace("SHA3-512:", "") for m in to_enrich.media if m.get("hash")
]
if not len(hashes):
logger.warning(f"No hashes found in {url=}")
return
tmp_dir = self.tmp_dir
hashes_fn = os.path.join(tmp_dir, "hashes.txt")
data_to_sign = "\n".join(hashes)
with open(hashes_fn, "w") as f:
with open(hashes_fn, "w") as f:
f.write(data_to_sign)
hashes_media = Media(filename=hashes_fn)
timestamp_tokens = []
from slugify import slugify
for tsa_url in self.tsa_urls:
try:
signing_settings = SigningSettings(tsp_server=tsa_url, digest_algorithm=DigestAlgorithm.SHA256)
signer = TSPSigner()
message = bytes(data_to_sign, encoding='utf8')
message = bytes(data_to_sign, encoding="utf8")
# send TSQ and get TSR from the TSA server
signed = signer.sign(message=message, signing_settings=signing_settings)
# fail if there's any issue with the certificates, uses certifi list of trusted CAs
@@ -54,7 +58,8 @@ class TimestampingEnricher(Enricher):
cert_chain = self.download_and_verify_certificate(signed)
# continue with saving the timestamp token
tst_fn = os.path.join(tmp_dir, f"timestamp_token_{slugify(tsa_url)}")
with open(tst_fn, "wb") as f: f.write(signed)
with open(tst_fn, "wb") as f:
f.write(signed)
timestamp_tokens.append(Media(filename=tst_fn).set("tsa", tsa_url).set("cert_chain", cert_chain))
except Exception as e:
logger.warning(f"Error while timestamping {url=} with {tsa_url=}: {e}")
@@ -75,7 +80,7 @@ class TimestampingEnricher(Enricher):
tst = ContentInfo.load(signed)
trust_roots = []
with open(certifi.where(), 'rb') as f:
with open(certifi.where(), "rb") as f:
for _, _, der_bytes in pem.unarmor(f.read(), multiple=True):
trust_roots.append(der_bytes)
context = ValidationContext(trust_roots=trust_roots)
@@ -83,11 +88,11 @@ class TimestampingEnricher(Enricher):
certificates = tst["content"]["certificates"]
first_cert = certificates[0].dump()
intermediate_certs = []
for i in range(1, len(certificates)): # cannot use list comprehension [1:]
for i in range(1, len(certificates)): # cannot use list comprehension [1:]
intermediate_certs.append(certificates[i].dump())
validator = CertificateValidator(first_cert, intermediate_certs=intermediate_certs, validation_context=context)
path = validator.validate_usage({'digital_signature'}, extended_key_usage={'time_stamping'})
path = validator.validate_usage({"digital_signature"}, extended_key_usage={"time_stamping"})
cert_chain = []
for cert in path:
@@ -96,4 +101,4 @@ class TimestampingEnricher(Enricher):
f.write(cert.dump())
cert_chain.append(Media(filename=cert_fn).set("subject", cert.subject.native["common_name"]))
return cert_chain
return cert_chain

2
src/auto_archiver/modules/twitter_api_extractor/__init__.py
View File

@@ -1 +1 @@
from .twitter_api_extractor import TwitterApiExtractor
from .twitter_api_extractor import TwitterApiExtractor

34
src/auto_archiver/modules/twitter_api_extractor/__manifest__.py
View File

@@ -3,21 +3,28 @@
"type": ["extractor"],
"requires_setup": True,
"dependencies": {
"python": ["requests",
"loguru",
"pytwitter",
"slugify",],
"bin": [""]
"python": [
"requests",
"loguru",
"pytwitter",
"slugify",
],
"bin": [""],
},
"configs": {
"bearer_token": {"default": None, "help": "[deprecated: see bearer_tokens] twitter API bearer_token which is enough for archiving, if not provided you will need consumer_key, consumer_secret, access_token, access_secret"},
"bearer_tokens": {"default": [], "help": " a list of twitter API bearer_token which is enough for archiving, if not provided you will need consumer_key, consumer_secret, access_token, access_secret, if provided you can still add those for better rate limits. CSV of bearer tokens if provided via the command line",
},
"consumer_key": {"default": None, "help": "twitter API consumer_key"},
"consumer_secret": {"default": None, "help": "twitter API consumer_secret"},
"access_token": {"default": None, "help": "twitter API access_token"},
"access_secret": {"default": None, "help": "twitter API access_secret"},
"bearer_token": {
"default": None,
"help": "[deprecated: see bearer_tokens] twitter API bearer_token which is enough for archiving, if not provided you will need consumer_key, consumer_secret, access_token, access_secret",
},
"bearer_tokens": {
"default": [],
"help": " a list of twitter API bearer_token which is enough for archiving, if not provided you will need consumer_key, consumer_secret, access_token, access_secret, if provided you can still add those for better rate limits. CSV of bearer tokens if provided via the command line",
},
"consumer_key": {"default": None, "help": "twitter API consumer_key"},
"consumer_secret": {"default": None, "help": "twitter API consumer_secret"},
"access_token": {"default": None, "help": "twitter API access_token"},
"access_secret": {"default": None, "help": "twitter API access_secret"},
},
"description": """
The `TwitterApiExtractor` fetches tweets and associated media using the Twitter API.
It supports multiple API configurations for extended rate limits and reliable access.
@@ -39,6 +46,5 @@
- **Access Token and Secret**: Complements the consumer key for enhanced API capabilities.
Credentials can be obtained by creating a Twitter developer account at [Twitter Developer Platform](https://developer.twitter.com/en).
"""
,
""",
}

71
src/auto_archiver/modules/twitter_api_extractor/twitter_api_extractor.py
View File

@@ -11,8 +11,8 @@ from slugify import slugify
from auto_archiver.core import Extractor
from auto_archiver.core import Metadata, Media
class TwitterApiExtractor(Extractor):
class TwitterApiExtractor(Extractor):
valid_url: re.Pattern = re.compile(r"(?:twitter|x).com\/(?:\#!\/)?(\w+)\/status(?:es)?\/(\d+)")
def setup(self) -> None:
@@ -23,30 +23,38 @@ class TwitterApiExtractor(Extractor):
if self.bearer_token:
self.apis.append(Api(bearer_token=self.bearer_token))
if self.consumer_key and self.consumer_secret and self.access_token and self.access_secret:
self.apis.append(Api(consumer_key=self.consumer_key, consumer_secret=self.consumer_secret,
access_token=self.access_token, access_secret=self.access_secret))
assert self.api_client is not None, "Missing Twitter API configurations, please provide either AND/OR (consumer_key, consumer_secret, access_token, access_secret) to use this archiver, you can provide both for better rate-limit results."
self.apis.append(
Api(
consumer_key=self.consumer_key,
consumer_secret=self.consumer_secret,
access_token=self.access_token,
access_secret=self.access_secret,
)
)
assert self.api_client is not None, (
"Missing Twitter API configurations, please provide either AND/OR (consumer_key, consumer_secret, access_token, access_secret) to use this archiver, you can provide both for better rate-limit results."
)
@property # getter .mimetype
def api_client(self) -> str:
return self.apis[self.api_index]
def sanitize_url(self, url: str) -> str:
# expand URL if t.co and clean tracker GET params
if 'https://t.co/' in url:
if "https://t.co/" in url:
try:
r = requests.get(url, timeout=30)
logger.debug(f'Expanded url {url} to {r.url}')
logger.debug(f"Expanded url {url} to {r.url}")
url = r.url
except:
logger.error(f'Failed to expand url {url}')
logger.error(f"Failed to expand url {url}")
return url
def download(self, item: Metadata) -> Metadata:
# call download retry until success or no more apis
while self.api_index < len(self.apis):
if res := self.download_retry(item): return res
if res := self.download_retry(item):
return res
self.api_index += 1
self.api_index = 0
return False
@@ -54,7 +62,8 @@ class TwitterApiExtractor(Extractor):
def get_username_tweet_id(self, url):
# detect URLs that we definitely cannot handle
matches = self.valid_url.findall(url)
if not len(matches): return False, False
if not len(matches):
return False, False
username, tweet_id = matches[0] # only one URL supported
logger.debug(f"Found {username=} and {tweet_id=} in {url=}")
@@ -65,10 +74,16 @@ class TwitterApiExtractor(Extractor):
url = item.get_url()
# detect URLs that we definitely cannot handle
username, tweet_id = self.get_username_tweet_id(url)
if not username: return False
if not username:
return False
try:
tweet = self.api_client.get_tweet(tweet_id, expansions=["attachments.media_keys"], media_fields=["type", "duration_ms", "url", "variants"], tweet_fields=["attachments", "author_id", "created_at", "entities", "id", "text", "possibly_sensitive"])
tweet = self.api_client.get_tweet(
tweet_id,
expansions=["attachments.media_keys"],
media_fields=["type", "duration_ms", "url", "variants"],
tweet_fields=["attachments", "author_id", "created_at", "entities", "id", "text", "possibly_sensitive"],
)
logger.debug(tweet)
except Exception as e:
logger.error(f"Could not get tweet: {e}")
@@ -88,29 +103,35 @@ class TwitterApiExtractor(Extractor):
mimetype = "image/jpeg"
elif hasattr(m, "variants"):
variant = self.choose_variant(m.variants)
if not variant: continue
if not variant:
continue
media.set("src", variant.url)
mimetype = variant.content_type
else:
continue
logger.info(f"Found media {media}")
ext = mimetypes.guess_extension(mimetype)
media.filename = self.download_from_url(media.get("src"), f'{slugify(url)}_{i}{ext}')
media.filename = self.download_from_url(media.get("src"), f"{slugify(url)}_{i}{ext}")
result.add_media(media)
result.set_content(json.dumps({
"id": tweet.data.id,
"text": tweet.data.text,
"created_at": tweet.data.created_at,
"author_id": tweet.data.author_id,
"geo": tweet.data.geo,
"lang": tweet.data.lang,
"media": urls
}, ensure_ascii=False, indent=4))
result.set_content(
json.dumps(
{
"id": tweet.data.id,
"text": tweet.data.text,
"created_at": tweet.data.created_at,
"author_id": tweet.data.author_id,
"geo": tweet.data.geo,
"lang": tweet.data.lang,
"media": urls,
},
ensure_ascii=False,
indent=4,
)
)
return result.success("twitter-api")
def choose_variant(self, variants):
"""
Chooses the highest quality variable possible out of a list of variants
"""

Some files were not shown because too many files have changed in this diff Show More