631 lines
17 KiB
TOML
631 lines
17 KiB
TOML
# warda-lists — the two taxonomies.
|
|
#
|
|
# 1. Blocking categories ([[category]]): what a Warda user switches on or
|
|
# off. Each one gives a file dist/<id>.txt. Sources and extra/ files put
|
|
# domains in them.
|
|
# 2. Site types ([[type]]): what kind of site a domain is (a shop, a blog,
|
|
# a forum…). For classification and statistics only: a site type is
|
|
# never blocked by itself. Each one gives a file dist/types/<id>.txt.
|
|
#
|
|
# Fields of a category:
|
|
# id the name used in sources.toml, extra/<id>.txt and dist/
|
|
# label_fr/en the name shown to people
|
|
# reason why it is blocked: an id of [[reason]] below
|
|
# description_fr what it holds, in the owner's words
|
|
# warda the Warda category id this list feeds ("base" is the
|
|
# list every Warda box blocks by default)
|
|
# includes other categories whose domains are also put in this one
|
|
# public_sources false when no public list exists (the build never fills
|
|
# it from public sources)
|
|
# max_domains the most domains its file may hold (default below)
|
|
#
|
|
# Only ids made of a-z, 0-9 and "-" are allowed. Changing an id changes the
|
|
# address of its file: the Warda boxes that download it must follow.
|
|
#
|
|
# This file also holds the licences a source may carry (licences, below),
|
|
# the budget of the files written by warda-analyst ([analyst]) and the
|
|
# groups that file the sources ([[source_group]], at the end).
|
|
|
|
# Budget of a category or bundle file: the lists live in the memory of the
|
|
# Warda boxes, a Raspberry Pi among them. A file over its budget fails the
|
|
# build (--force does not skip it): disable or replace a big source, or
|
|
# raise max_domains of that category knowingly. Site types have no budget
|
|
# (never loaded to block).
|
|
default_max_domains = 1500000
|
|
|
|
# The licences accepted for a source (SPDX ids): the only place where they
|
|
# are written. scripts/build.py reads them here and publishes them in
|
|
# manifest.json (taxonomy.licences), where warda-analyst reads them. Each
|
|
# one has its text in licenses/<id>.txt, copied to dist/licenses/. Anything
|
|
# else is refused: unknown or proprietary licences, and those that forbid
|
|
# what the build does (NC: Warda has a paid offer; ND: the lists are
|
|
# modified), which the build refuses here too. A new licence must also be
|
|
# known to combine_licences of scripts/build.py when it cannot be mixed
|
|
# with the GPL.
|
|
licences = [
|
|
"GPL-3.0-only",
|
|
"GPL-3.0-or-later",
|
|
"GPL-2.0-only",
|
|
"GPL-2.0-or-later",
|
|
"CC-BY-SA-4.0",
|
|
"CC-BY-SA-3.0",
|
|
"CC-BY-4.0",
|
|
"CC-BY-3.0",
|
|
"CC0-1.0",
|
|
"Unlicense",
|
|
"MIT",
|
|
"ISC",
|
|
"0BSD",
|
|
"BSD-2-Clause",
|
|
"BSD-3-Clause",
|
|
"Apache-2.0",
|
|
]
|
|
|
|
# --- Why a category is blocked -------------------------------------------
|
|
|
|
[[reason]]
|
|
id = "privacy"
|
|
label_fr = "Publicité et pistage"
|
|
label_en = "Advertising and tracking"
|
|
|
|
[[reason]]
|
|
id = "cybersecurity"
|
|
label_fr = "Cybersécurité"
|
|
label_en = "Cybersecurity"
|
|
description_fr = "Malwares, phishing, command & control, cryptojacking, proxies/VPN."
|
|
|
|
[[reason]]
|
|
id = "law"
|
|
label_fr = "Juridique"
|
|
label_en = "Law"
|
|
description_fr = "Apologie du terrorisme, pédopornographie, piratage, contrefaçon et arnaques, jeux d'argent illégaux."
|
|
|
|
[[reason]]
|
|
id = "productivity"
|
|
label_fr = "Productivité au travail"
|
|
label_en = "Workplace productivity"
|
|
description_fr = "Réseaux sociaux, streaming/vidéo, jeux en ligne, sites d'emploi concurrents."
|
|
|
|
[[reason]]
|
|
id = "ethics"
|
|
label_fr = "Éthique"
|
|
label_en = "Ethics"
|
|
description_fr = "Pornographie/contenu adulte, armes/violence, drogues/alcool, haine/discrimination."
|
|
|
|
[[reason]]
|
|
id = "age"
|
|
label_fr = "Ancienneté du domaine"
|
|
label_en = "Domain age"
|
|
|
|
# --- Blocking categories ---------------------------------------------------
|
|
|
|
[[category]]
|
|
id = "ads"
|
|
label_fr = "Publicité"
|
|
label_en = "Advertising"
|
|
reason = "privacy"
|
|
description_fr = "Régies et serveurs publicitaires."
|
|
warda = "base"
|
|
|
|
[[category]]
|
|
id = "tracking"
|
|
label_fr = "Pistage"
|
|
label_en = "Tracking"
|
|
reason = "privacy"
|
|
description_fr = "Traceurs, mesure d'audience, télémétrie."
|
|
warda = "base"
|
|
|
|
[[category]]
|
|
id = "phishing"
|
|
label_fr = "Hameçonnage"
|
|
label_en = "Phishing"
|
|
reason = "cybersecurity"
|
|
description_fr = "Sites d'hameçonnage (phishing) qui volent identifiants et données bancaires."
|
|
# Warda has no phishing category: this list is merged into security.
|
|
warda = "security"
|
|
|
|
[[category]]
|
|
id = "security"
|
|
label_fr = "Sécurité"
|
|
label_en = "Security"
|
|
reason = "cybersecurity"
|
|
description_fr = "Malwares, phishing, command & control, cryptojacking, contrefaçon et arnaques."
|
|
warda = "security"
|
|
includes = ["phishing"]
|
|
|
|
[[category]]
|
|
id = "stalkerware"
|
|
label_fr = "Logiciels espions"
|
|
label_en = "Stalkerware"
|
|
reason = "cybersecurity"
|
|
description_fr = "Serveurs des logiciels espions du commerce (stalkerware) installés en cachette sur un téléphone ou un ordinateur."
|
|
# Not a category of the profiles: Warda blocks it for every device and
|
|
# tells the person of the device (package threats).
|
|
warda = "stalkerware"
|
|
max_domains = 50000
|
|
|
|
[[category]]
|
|
id = "bypass"
|
|
label_fr = "Contournement du filtre"
|
|
label_en = "Filter bypass"
|
|
reason = "cybersecurity"
|
|
description_fr = "Proxies, VPN, Tor et DNS chiffrés publics utilisés pour contourner le filtre."
|
|
warda = "bypass"
|
|
|
|
[[category]]
|
|
id = "csam"
|
|
label_fr = "Pédopornographie"
|
|
label_en = "Child sexual abuse material"
|
|
reason = "law"
|
|
description_fr = "Pédopornographie. Aucune liste publique : ces listes ne sont remises qu'aux organismes et FAI habilités."
|
|
warda = "csam"
|
|
public_sources = false
|
|
|
|
[[category]]
|
|
id = "hate"
|
|
label_fr = "Haine et apologie du terrorisme"
|
|
label_en = "Hate and terrorism apology"
|
|
reason = "law"
|
|
description_fr = "Apologie du terrorisme, haine et discrimination."
|
|
warda = "hate"
|
|
|
|
[[category]]
|
|
id = "piracy"
|
|
label_fr = "Piratage"
|
|
label_en = "Piracy"
|
|
reason = "law"
|
|
description_fr = "Piratage : streaming, téléchargement direct et torrents d'œuvres protégées."
|
|
warda = "piracy"
|
|
|
|
[[category]]
|
|
id = "gambling"
|
|
label_fr = "Jeux d'argent"
|
|
label_en = "Gambling"
|
|
reason = "law"
|
|
description_fr = "Jeux d'argent illégaux (non validés par l'ANJ)."
|
|
warda = "gambling"
|
|
|
|
[[category]]
|
|
id = "social"
|
|
label_fr = "Réseaux sociaux"
|
|
label_en = "Social networks"
|
|
reason = "productivity"
|
|
description_fr = "Réseaux sociaux."
|
|
warda = "social"
|
|
|
|
[[category]]
|
|
id = "video"
|
|
label_fr = "Streaming et vidéo"
|
|
label_en = "Streaming and video"
|
|
reason = "productivity"
|
|
description_fr = "Plateformes de streaming et de vidéo."
|
|
warda = "video"
|
|
|
|
[[category]]
|
|
id = "games"
|
|
label_fr = "Jeux vidéo en ligne"
|
|
label_en = "Online games"
|
|
reason = "productivity"
|
|
description_fr = "Jeux vidéo en ligne."
|
|
warda = "games"
|
|
|
|
[[category]]
|
|
id = "jobsearch"
|
|
label_fr = "Sites de recrutement"
|
|
label_en = "Job search sites"
|
|
reason = "productivity"
|
|
description_fr = "Sites de recrutement et d'offres d'emploi (sites d'emploi concurrents : politique d'entreprise)."
|
|
warda = "jobsearch"
|
|
|
|
[[category]]
|
|
id = "adult"
|
|
label_fr = "Pornographie et contenu adulte"
|
|
label_en = "Pornography and adult content"
|
|
reason = "ethics"
|
|
description_fr = "Pornographie et contenu adulte."
|
|
warda = "adult"
|
|
|
|
[[category]]
|
|
id = "violence"
|
|
label_fr = "Armes et violence"
|
|
label_en = "Weapons and violence"
|
|
reason = "ethics"
|
|
description_fr = "Armes et violence."
|
|
warda = "violence"
|
|
|
|
[[category]]
|
|
id = "drugs"
|
|
label_fr = "Drogues et alcool"
|
|
label_en = "Drugs and alcohol"
|
|
reason = "ethics"
|
|
description_fr = "Drogues et alcool."
|
|
warda = "drugs"
|
|
|
|
[[category]]
|
|
id = "newdomains"
|
|
label_fr = "Domaines récents"
|
|
label_en = "Newly registered domains"
|
|
reason = "age"
|
|
description_fr = "Domaines enregistrés ces derniers jours, souvent utilisés pour le phishing (faux positifs possibles, désactivé par défaut dans Warda)."
|
|
warda = "newdomains"
|
|
# Newly registered domains: about 3 million a week.
|
|
max_domains = 3500000
|
|
|
|
# --- Combined lists --------------------------------------------------------
|
|
# A bundle has no source of its own: it is the union of categories. Sources
|
|
# cannot name it.
|
|
|
|
[[bundle]]
|
|
id = "base"
|
|
label_fr = "Liste de base (publicité et pistage)"
|
|
label_en = "Base list (advertising and tracking)"
|
|
description_fr = "Publicité et pistage : la liste que chaque boîtier Warda bloque par défaut."
|
|
warda = "base"
|
|
includes = ["ads", "tracking"]
|
|
|
|
# --- Files written by warda-analyst -----------------------------------------
|
|
# analyst/add/<category>.txt and analyst/remove.txt are written by the
|
|
# service warda-analyst after a decision of the team, never by hand. Their
|
|
# budgets, in names: over one of them, --check fails on develop and the
|
|
# commit is not promoted to main.
|
|
# max_add the names of all the analyst/add/ files together
|
|
# max_remove the names of analyst/remove.txt
|
|
# And a limit applied by the build, which alone knows what the sources hold:
|
|
# max_effect the most names of the lists one line of analyst/remove.txt
|
|
# may take out, itself and its subdomains. A line over it is
|
|
# not applied (the build says it, manifest.json lists it): it
|
|
# is a suffix that many sites share (co.uk, github.io), not a
|
|
# name wrongly blocked.
|
|
# And where the removals apply:
|
|
# remove_from the categories analyst/remove.txt takes its names out of
|
|
# (with their subdomains): the protection lists only. The
|
|
# team clears a name because it is harmless (no malware, no
|
|
# tracker); that says nothing of what the site is about, so
|
|
# the name stays in the content categories (adult, gambling…)
|
|
# and in the site types. The bundles built from these
|
|
# categories (base) follow, and so does what "includes"
|
|
# brings into them (security includes phishing). allow.txt
|
|
# keeps its own meaning: every list.
|
|
|
|
[analyst]
|
|
max_add = 50000
|
|
max_remove = 20000
|
|
max_effect = 100
|
|
remove_from = ["ads", "tracking", "phishing", "security"]
|
|
|
|
# --- Site types (classification only, never blocked by themselves) --------
|
|
|
|
[[group]]
|
|
id = "business"
|
|
label_fr = "Commerciaux et Business"
|
|
label_en = "Commercial and business"
|
|
|
|
[[group]]
|
|
id = "content"
|
|
label_fr = "Contenu et Information"
|
|
label_en = "Content and information"
|
|
|
|
[[group]]
|
|
id = "community"
|
|
label_fr = "Communautaires et Collaboratifs"
|
|
label_en = "Community and collaborative"
|
|
|
|
[[group]]
|
|
id = "services"
|
|
label_fr = "Annuaires et Plateformes de Services"
|
|
label_en = "Directories and service platforms"
|
|
|
|
[[group]]
|
|
id = "personal"
|
|
label_fr = "Personnels et Divertissement"
|
|
label_en = "Personal and entertainment"
|
|
|
|
[[type]]
|
|
id = "e-commerce"
|
|
number = 1
|
|
group = "business"
|
|
label_fr = "E-commerce"
|
|
label_en = "Online shop"
|
|
description_fr = "Boutique en ligne : vente de produits ou de services."
|
|
|
|
[[type]]
|
|
id = "corporate"
|
|
number = 2
|
|
group = "business"
|
|
label_fr = "Site vitrine d'entreprise"
|
|
label_en = "Corporate showcase site"
|
|
description_fr = "Présentation d'une entreprise, de ses activités et de ses coordonnées."
|
|
|
|
[[type]]
|
|
id = "landing-page"
|
|
number = 3
|
|
group = "business"
|
|
label_fr = "Landing page"
|
|
label_en = "Landing page"
|
|
description_fr = "Page d'atterrissage d'une campagne ou d'une offre."
|
|
|
|
[[type]]
|
|
id = "startup"
|
|
number = 4
|
|
group = "business"
|
|
label_fr = "Start-up"
|
|
label_en = "Start-up"
|
|
description_fr = "Site d'une jeune entreprise et de son produit."
|
|
|
|
[[type]]
|
|
id = "professional-services"
|
|
number = 5
|
|
group = "business"
|
|
label_fr = "Conseil / professionnel"
|
|
label_en = "Consulting / professional services"
|
|
description_fr = "Cabinet de conseil, profession libérale, services aux entreprises."
|
|
|
|
[[type]]
|
|
id = "affiliate"
|
|
number = 6
|
|
group = "business"
|
|
label_fr = "Affiliation"
|
|
label_en = "Affiliate site"
|
|
description_fr = "Site qui recommande des produits contre commission."
|
|
|
|
[[type]]
|
|
id = "blog"
|
|
number = 7
|
|
group = "content"
|
|
label_fr = "Blog"
|
|
label_en = "Blog"
|
|
description_fr = "Blog personnel ou professionnel, plateforme de blogs."
|
|
|
|
[[type]]
|
|
id = "news-media"
|
|
number = 8
|
|
group = "content"
|
|
label_fr = "Actualités et médias"
|
|
label_en = "News and media"
|
|
description_fr = "Presse, actualités, magazines."
|
|
|
|
[[type]]
|
|
id = "government"
|
|
number = 9
|
|
group = "content"
|
|
label_fr = "Institutionnel / gouvernemental"
|
|
label_en = "Institutional / government"
|
|
description_fr = "Administration, collectivité, institution publique."
|
|
|
|
[[type]]
|
|
id = "non-profit"
|
|
number = 10
|
|
group = "content"
|
|
label_fr = "Associatif / non-lucratif"
|
|
label_en = "Non-profit"
|
|
description_fr = "Association, fondation, organisation à but non lucratif."
|
|
|
|
[[type]]
|
|
id = "knowledge-base"
|
|
number = 11
|
|
group = "content"
|
|
label_fr = "Base de connaissances (wiki)"
|
|
label_en = "Knowledge base (wiki)"
|
|
description_fr = "Wiki, documentation, encyclopédie."
|
|
|
|
[[type]]
|
|
id = "education"
|
|
number = 12
|
|
group = "content"
|
|
label_fr = "Éducatif / scolaire"
|
|
label_en = "Educational / school"
|
|
description_fr = "École, université, ressources pédagogiques."
|
|
|
|
[[type]]
|
|
id = "social-network"
|
|
number = 13
|
|
group = "community"
|
|
label_fr = "Réseau social"
|
|
label_en = "Social network"
|
|
description_fr = "Réseau social, messagerie communautaire."
|
|
|
|
[[type]]
|
|
id = "forum"
|
|
number = 14
|
|
group = "community"
|
|
label_fr = "Forum"
|
|
label_en = "Forum"
|
|
description_fr = "Forum de discussion."
|
|
|
|
[[type]]
|
|
id = "petition"
|
|
number = 15
|
|
group = "community"
|
|
label_fr = "Pétitions"
|
|
label_en = "Petitions"
|
|
description_fr = "Plateforme de pétitions en ligne."
|
|
|
|
[[type]]
|
|
id = "crowdfunding"
|
|
number = 16
|
|
group = "community"
|
|
label_fr = "Financement participatif"
|
|
label_en = "Crowdfunding"
|
|
description_fr = "Plateforme de financement participatif."
|
|
|
|
[[type]]
|
|
id = "directory"
|
|
number = 17
|
|
group = "services"
|
|
label_fr = "Annuaire"
|
|
label_en = "Directory"
|
|
description_fr = "Annuaire de sites, d'entreprises ou de personnes."
|
|
|
|
[[type]]
|
|
id = "job-board"
|
|
number = 18
|
|
group = "services"
|
|
label_fr = "Offres d'emploi"
|
|
label_en = "Job board"
|
|
description_fr = "Offres d'emploi et recrutement."
|
|
|
|
[[type]]
|
|
id = "real-estate"
|
|
number = 19
|
|
group = "services"
|
|
label_fr = "Immobilier"
|
|
label_en = "Real estate"
|
|
description_fr = "Annonces et agences immobilières."
|
|
|
|
[[type]]
|
|
id = "reviews"
|
|
number = 20
|
|
group = "services"
|
|
label_fr = "Avis et critiques"
|
|
label_en = "Reviews"
|
|
description_fr = "Avis de consommateurs, critiques."
|
|
|
|
[[type]]
|
|
id = "events"
|
|
number = 21
|
|
group = "services"
|
|
label_fr = "Événements"
|
|
label_en = "Events"
|
|
description_fr = "Billetterie, agenda, organisation d'événements."
|
|
|
|
[[type]]
|
|
id = "personal"
|
|
number = 22
|
|
group = "personal"
|
|
label_fr = "Site personnel"
|
|
label_en = "Personal site"
|
|
description_fr = "Site d'une personne."
|
|
|
|
[[type]]
|
|
id = "portfolio"
|
|
number = 23
|
|
group = "personal"
|
|
label_fr = "Portfolio"
|
|
label_en = "Portfolio"
|
|
description_fr = "Présentation des travaux d'un créatif ou d'un professionnel."
|
|
|
|
[[type]]
|
|
id = "entertainment"
|
|
number = 24
|
|
group = "personal"
|
|
label_fr = "Divertissement"
|
|
label_en = "Entertainment"
|
|
description_fr = "Divertissement, humour, célébrités."
|
|
|
|
[[type]]
|
|
id = "streaming"
|
|
number = 25
|
|
group = "personal"
|
|
label_fr = "Streaming"
|
|
label_en = "Streaming"
|
|
description_fr = "Vidéo, musique ou radio en streaming."
|
|
|
|
[[type]]
|
|
id = "video-games"
|
|
number = 26
|
|
group = "personal"
|
|
label_fr = "Jeux vidéo"
|
|
label_en = "Video games"
|
|
description_fr = "Jeux vidéo, jeux en ligne."
|
|
|
|
[[type]]
|
|
id = "wedding"
|
|
number = 27
|
|
group = "personal"
|
|
label_fr = "Mariage"
|
|
label_en = "Wedding"
|
|
description_fr = "Site de mariage, liste de mariage."
|
|
|
|
[[type]]
|
|
id = "travel"
|
|
number = 28
|
|
group = "personal"
|
|
label_fr = "Voyage"
|
|
label_en = "Travel"
|
|
description_fr = "Voyage, tourisme, réservation."
|
|
|
|
[[type]]
|
|
id = "memorial"
|
|
number = 29
|
|
group = "personal"
|
|
label_fr = "Commémoration"
|
|
label_en = "Memorial"
|
|
description_fr = "Site commémoratif, hommage."
|
|
|
|
[[type]]
|
|
id = "kids"
|
|
number = 30
|
|
group = "personal"
|
|
label_fr = "Jeunesse / enfants"
|
|
label_en = "Youth / children"
|
|
description_fr = "Sites pour la jeunesse et les enfants."
|
|
|
|
# --- Groups of sources (filing only) --------------------------------------
|
|
# Every [[source]] of sources.toml names one of these groups, exactly as
|
|
# written (field group). A group only files and names the lists: it never
|
|
# decides where their domains go (categories or [source.map] do).
|
|
|
|
[[source_group]]
|
|
name = "Hosts list"
|
|
description = "General hosts files that mix several kinds of domains (ads, trackers, malware)."
|
|
|
|
[[source_group]]
|
|
name = "Abuse List"
|
|
description = "Sites known for abusive or deceptive practices."
|
|
|
|
[[source_group]]
|
|
name = "Drugs List"
|
|
description = "Drugs and alcohol."
|
|
|
|
[[source_group]]
|
|
name = "Fraud List"
|
|
description = "Fraud: fake shops, financial and payment fraud."
|
|
|
|
[[source_group]]
|
|
name = "Malware List"
|
|
description = "Malware distribution and command & control servers."
|
|
|
|
[[source_group]]
|
|
name = "Phishing List"
|
|
description = "Phishing: pages that steal credentials or bank data."
|
|
|
|
[[source_group]]
|
|
name = "Ransomware List"
|
|
description = "Ransomware: distribution, payment and command servers."
|
|
|
|
[[source_group]]
|
|
name = "Scam List"
|
|
description = "Scams: fake support, fake giveaways, deceptive offers."
|
|
|
|
[[source_group]]
|
|
name = "Tracking & Telemetry List"
|
|
description = "Trackers, analytics and telemetry of devices and applications."
|
|
|
|
[[source_group]]
|
|
name = "Redirect List"
|
|
description = "URL shorteners and redirectors that hide the real destination."
|
|
|
|
[[source_group]]
|
|
name = "Gambling List"
|
|
description = "Gambling and betting sites."
|
|
|
|
[[source_group]]
|
|
name = "Porn List"
|
|
description = "Pornography and adult content."
|
|
|
|
[[source_group]]
|
|
name = "Suspicious Lists"
|
|
description = "Not proven bad but risky: newly registered domains, parked or disposable domains."
|
|
|
|
[[source_group]]
|
|
name = "Advertising Lists"
|
|
description = "Advertising, and ad-and-tracker lists where ads dominate."
|
|
|
|
[[source_group]]
|
|
name = "Malicious Lists"
|
|
description = "Threat feeds that mix several threats (malware, phishing, C&C, cryptojacking)."
|
|
|
|
[[source_group]]
|
|
name = "Other Lists"
|
|
description = "Everything else: bypass, piracy, social networks, categorized archives such as UT1."
|