chg: Migrate Cookies names index to new format.

pull/952/head
Raphaël Vinot 2024-10-07 20:45:08 +02:00
parent 393a598d77
commit f88f664307
3 changed files with 125 additions and 99 deletions

View File

@ -170,21 +170,24 @@ class Indexing():
# ###### Cookies ######
def _reindex_cookies(self, cookie_name: str) -> None:
# We changed the format of the indexes, so we need to make sure they're re-triggered.
pipeline = self.redis.pipeline()
if self.redis.type(f'cn|{cookie_name}|captures') == 'set': # type: ignore[no-untyped-call]
pipeline.srem('indexed_cookies', *[entry.split('|')[0] for entry in self.redis.smembers(f'cn|{cookie_name}|captures')])
pipeline.delete(f'cn|{cookie_name}|captures')
if self.redis.type(f'cn|{cookie_name}') == 'zset': # type: ignore[no-untyped-call]
for domain in self.redis.zrevrangebyscore(f'cn|{cookie_name}', '+inf', '-inf'):
pipeline.delete(f'cn|{cookie_name}|{domain}')
pipeline.delete(domain)
pipeline.delete(f'cn|{cookie_name}')
if self.redis.type('cookies_names') == 'zset': # type: ignore[no-untyped-call]
pipeline.delete('cookies_names')
pipeline.execute()
@property
def cookies_names(self) -> list[tuple[str, float]]:
return self.redis.zrevrange('cookies_names', 0, -1, withscores=True)
def cookies_names_number_domains(self, cookie_name: str) -> int:
return self.redis.zcard(f'cn|{cookie_name}')
def cookies_names_domains_values(self, cookie_name: str, domain: str) -> list[tuple[str, float]]:
return self.redis.zrevrange(f'cn|{cookie_name}|{domain}', 0, -1, withscores=True)
def get_cookie_domains(self, cookie_name: str) -> list[tuple[str, float]]:
return self.redis.zrevrange(f'cn|{cookie_name}', 0, -1, withscores=True)
def get_cookies_names_captures(self, cookie_name: str) -> list[tuple[str, str]]:
return [uuids.split('|') for uuids in self.redis.smembers(f'cn|{cookie_name}|captures')] # type: ignore[misc]
def cookies_names(self) -> set[str]:
return self.redis.smembers('cookies_names')
def index_cookies_capture(self, crawled_tree: CrawledTree) -> None:
if self.redis.sismember('indexed_cookies', crawled_tree.uuid):
@ -192,39 +195,62 @@ class Indexing():
return
self.logger.debug(f'Indexing cookies for {crawled_tree.uuid} ... ')
self.redis.sadd('indexed_cookies', crawled_tree.uuid)
pipeline = self.redis.pipeline()
already_loaded: set[tuple[str, str]] = set()
# used if we need to reindex a capture
already_cleaned_up: set[str] = set()
is_reindex = False
# Add the cookies_names key in internal indexes set
internal_index = f'capture_indexes|{crawled_tree.uuid}'
pipeline.sadd(internal_index, 'cookies_names')
already_indexed_global: set[str] = set()
for urlnode in crawled_tree.root_hartree.url_tree.traverse():
if 'cookies_received' not in urlnode.features:
continue
for domain, cookie, _ in urlnode.cookies_received:
name, value = cookie.split('=', 1)
if (name, domain) in already_loaded:
# Only add cookie name once / capture
continue
already_loaded.add((name, domain))
if name not in already_cleaned_up:
# We only run this srem once per name for a capture,
# before adding it for the first time
to_remove = [key for key in self.redis.sscan_iter(f'cn|{name}|captures', f'{crawled_tree.uuid}|*')]
if to_remove:
pipeline.srem(f'cn|{name}|captures', *to_remove)
is_reindex = True
self.logger.debug(f'reindexing cookies for {crawled_tree.uuid} ... ')
already_cleaned_up.add(name)
pipeline.sadd(f'cn|{name}|captures', f'{crawled_tree.uuid}|{urlnode.uuid}')
if not is_reindex:
pipeline.zincrby('cookies_names', 1, name)
pipeline.zincrby(f'cn|{name}', 1, domain)
pipeline.zincrby(f'cn|{name}|{domain}', 1, value)
pipeline.sadd(domain, name)
self._reindex_cookies(name)
if name not in already_indexed_global:
# The cookie hasn't been indexed in that run yet
already_indexed_global.add(name)
pipeline.sadd(f'{internal_index}|cookies_names', name)
pipeline.sadd('cookies_names', name)
pipeline.zadd(f'cookies_names|{name}|captures',
mapping={crawled_tree.uuid: crawled_tree.start_time.timestamp()})
# Add hostnode UUID in internal index
pipeline.sadd(f'{internal_index}|cookies_names|{name}', urlnode.uuid)
pipeline.execute()
self.logger.debug(f'done with cookies for {crawled_tree.uuid}.')
def get_captures_cookies_name(self, cookie_name: str, most_recent_capture: datetime | None = None,
oldest_capture: datetime | None= None) -> list[tuple[str, float]]:
"""Get all the captures for a specific cookie name, on a time interval starting from the most recent one.
:param cookie_name: The cookie name
:param most_recent_capture: The capture time of the most recent capture to consider
:param oldest_capture: The capture time of the oldest capture to consider, defaults to 15 days ago.
"""
max_score: str | float = most_recent_capture.timestamp() if most_recent_capture else '+Inf'
min_score: str | float = oldest_capture.timestamp() if oldest_capture else (datetime.now() - timedelta(days=15)).timestamp()
if self.redis.type(f'cookies_names|{cookie_name}|captures') == 'set': # type: ignore[no-untyped-call]
# triggers the re-index soon.
self.redis.srem('indexed_cookies', *[entry.split('|')[0] for entry in self.redis.smembers(f'cn|{cookie_name}|captures')])
return []
return self.redis.zrevrangebyscore(f'cookies_names|{cookie_name}|captures', max_score, min_score, withscores=True)
def get_captures_cookie_name_count(self, cookie_name: str) -> int:
return self.redis.zcard(f'cookies_names|{cookie_name}|captures')
def get_capture_cookie_name_nodes(self, capture_uuid: str, cookie_name: str) -> set[str]:
if url_nodes := self.redis.smembers(f'capture_indexes|{capture_uuid}|cookies_names|{cookie_name}'):
return set(url_nodes)
return set()
def cookies_names_domains_values(self, cookie_name: str, domain: str) -> list[tuple[str, float]]:
return self.redis.zrevrange(f'cn|{cookie_name}|{domain}', 0, -1, withscores=True)
def get_cookie_domains(self, cookie_name: str) -> list[tuple[str, float]]:
return self.redis.zrevrange(f'cn|{cookie_name}', 0, -1, withscores=True)
# ###### Body hashes ######
def _reindex_ressources(self, h: str) -> None:
@ -381,7 +407,7 @@ class Indexing():
min_score: str | float = oldest_capture.timestamp() if oldest_capture else (datetime.now() - timedelta(days=15)).timestamp()
if self.redis.type(f'hhhashes|{hhh}|captures') == 'set': # type: ignore[no-untyped-call]
# triggers the re-index soon.
self.redis.srem('indexed_urls', *self.redis.smembers(f'hhhashes|{hhh}|captures'))
self.redis.srem('indexed_hhhashes', *self.redis.smembers(f'hhhashes|{hhh}|captures'))
return []
return self.redis.zrevrangebyscore(f'hhhashes|{hhh}|captures', max_score, min_score, withscores=True)

View File

@ -417,13 +417,13 @@ def get_url_investigator(url: str) -> list[tuple[str, str, str, datetime, set[st
) for cache in cached_captures]
def get_cookie_name_investigator(cookie_name: str, /) -> tuple[list[tuple[str, str]], list[tuple[str, float, list[tuple[str, float]]]]]:
def get_cookie_name_investigator(cookie_name: str, /) -> list[tuple[str, str, datetime, set[str]]]:
'''Returns all the captures related to a cookie name entry, used in the web interface.'''
cached_captures = lookyloo.sorted_capture_cache([entry[0] for entry in get_indexing(flask_login.current_user).get_cookies_names_captures(cookie_name)])
captures = [(cache.uuid, cache.title) for cache in cached_captures]
domains = [(domain, freq, get_indexing(flask_login.current_user).cookies_names_domains_values(cookie_name, domain))
for domain, freq in get_indexing(flask_login.current_user).get_cookie_domains(cookie_name)]
return captures, domains
cached_captures = lookyloo.sorted_capture_cache(
[uuid for uuid, _ in get_indexing(flask_login.current_user).get_captures_cookies_name(cookie_name=cookie_name)],
cached_captures_only=True)
captures = [(cache.uuid, cache.title, cache.timestamp, get_indexing(flask_login.current_user).get_capture_cookie_name_nodes(cache.uuid, cookie_name)) for cache in cached_captures]
return captures
def get_identifier_investigator(identifier_type: str, identifier: str) -> list[tuple[str, str, str, datetime]]:
@ -1420,8 +1420,9 @@ def index_hidden() -> str:
@app.route('/cookies', methods=['GET'])
def cookies_lookup() -> str:
cookies_names = [(name, freq, get_indexing(flask_login.current_user).cookies_names_number_domains(name))
for name, freq in get_indexing(flask_login.current_user).cookies_names]
cookies_names = []
for name in get_indexing(flask_login.current_user).cookies_names:
cookies_names.append((name, get_indexing(flask_login.current_user).get_captures_cookie_name_count(name)))
return render_template('cookies.html', cookies_names=cookies_names)
@ -1737,8 +1738,8 @@ def simple_capture() -> str | Response | WerkzeugResponse:
@app.route('/cookies/<string:cookie_name>', methods=['GET'])
def cookies_name_detail(cookie_name: str) -> str:
captures, domains = get_cookie_name_investigator(cookie_name.strip())
return render_template('cookie_name.html', cookie_name=cookie_name, domains=domains, captures=captures)
captures = get_cookie_name_investigator(cookie_name.strip())
return render_template('cookie_name.html', cookie_name=cookie_name, captures=captures)
@app.route('/hhhdetails/<string:hhh>', methods=['GET'])

View File

@ -1,67 +1,66 @@
{% extends "main.html" %}
{% from 'bootstrap5/utils.html' import render_messages %}
{% block title %}{{ cookie_name }}{% endblock %}
{% from "macros.html" import shorten_string %}
{% block scripts %}
{{ super() }}
<script type="text/javascript" nonce="{{ csp_nonce() }}">
$('#table').DataTable( {
"order": [[ 1, "desc" ]],
"pageLength": 500
});
new DataTable('#cookieNameTable', {
order: [[ 0, "desc" ]],
columnDefs: [{ width: '30%', targets: 0,
render: (data) => {
const date = new Date(data);
return date.getFullYear() + '-' + (date.getMonth() + 1).toString().padStart(2, "0") + '-' + date.getDate().toString().padStart(2, "0") + ' ' + date.toTimeString();
}
},
{ width: '70%', targets: 1 }]
});
</script>
<script type="text/javascript" nonce="{{ csp_nonce() }}">
<script type="text/javascript" nonce="{{ csp_nonce() }}">
const openNewTabButtons = document.querySelectorAll('.openNewTab');
if (openNewTabButtons) {
openNewTabButtons.forEach(el => el.addEventListener('click', event => {
window.opener.openTreeInNewTab(el.dataset.capture);
window.opener.openTreeInNewTab(el.dataset.capture, el.dataset.hostnode);
}));
}
</script>
{% endblock %}
{% block content %}
<center>
<h2>{{ cookie_name }}</h2>
<button class="btn btn-primary goBack" type="button">Go Back</button>
</center>
<div class="table-responsive">
<table id="table" class="table" style="width:96%">
<thead>
<tr>
<th>Domain name</th>
<th>Frequency</th>
<th>Value</th>
</tr>
</thead>
<tbody>
{% for domain, freq, values in domains %}
<tr>
<td>
{{ domain }}
</td>
<td>{{ freq }}</td>
<td>
<ul>
{% for value, freq in values %}
<li>{{ value }} - {{ freq }}</li>
{% endfor %}
</ul>
</td>
</tr>
{% endfor %}
</tbody>
</table>
</div>
<p>A cookie with that name was seen in these captures:</p>
<ul>
{% for capture_uuid, title in captures %}
<li>
<button type="button" class="btn btn-link openNewTab" data-capture="{{capture_uuid}}">{{ title }}</button>
</li>
{% endfor %}
</ul>
<center>
<h4>{{ cookie_name }}</h4>
<h6>Only the most recent captures are listed below, this will change soon.</h6>
</center>
<table id="cookieNameTable" class="table table-striped" style="width:100%">
<thead>
<tr>
<th>Capture Time</th>
<th>Capture Title</th>
</tr>
</thead>
<tbody>
{% for capture_uuid, title, capture_time, nodes in captures %}
<tr>
<td>
{{capture_time}}
</td>
<td>
<a href="{{ url_for('tree', tree_uuid=capture_uuid) }}">
{{ title }}
</a>
<br>
The capture contains this URL in {{ nodes|length }} nodes, click below to see them on the tree:
<ul>
{% for node in nodes %}
<li>
<button type="button" class="btn btn-link openNewTab" data-capture="{{capture_uuid}}" data-hostnode="{{node}}">{{ node }}</button>
</li>
{% endfor %}
</ul>
</td>
</tr>
{% endfor %}
</tbody>
</table>
{% endblock %}