@postalsys/email-content 0.0.1

package/dist/detect/patterns/disclaimers.json

@@ -0,0 +1,113 @@
+{
+    "description": "Machine appended boilerplate, per the DISCLAIMER convention in docs/annotation-guide.md: text attached by a system, gateway or template independently of what the sender wrote. Matching is on the block as a whole. Deliberately generic: patterns target the shapes that recur across senders, not any one vendor's template, because fitting a single sender's footer would not transfer.",
+    "phrases": [
+        {
+            "id": "confidentiality_en",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "(this (e-?mail|message)[^.]{0,80}(is )?(strictly )?confidential|confidential and (may be )?(legally )?privileged|intended (solely |only )?for the (named |intended )?(addressee|recipient))"
+        },
+        {
+            "id": "unintended_recipient_en",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "(if you (have received|are not) .{0,80}(in error|the intended recipient)|please (notify the sender|delete) .{0,60}(immediately|and delete))"
+        },
+        {
+            "id": "no_liability_en",
+            "strength": 0.7,
+            "flags": "i",
+            "pattern": "(accepts? no (liability|responsibility)|any views or opinions (presented|expressed) .{0,60}are (solely )?those of the (author|sender))"
+        },
+        {
+            "id": "virus_scan_en",
+            "strength": 0.7,
+            "flags": "i",
+            "pattern": "(this (e-?mail|message) has been (scanned|checked) for viruses|virus-free|free from (any )?virus)"
+        },
+        {
+            "id": "confidentiality_de",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "(vertrauliche Informationen|Diese E-?Mail (ist|enthält) .{0,60}vertraulich|Sollten Sie nicht der richtige Adressat sein)"
+        },
+        {
+            "id": "confidentiality_fr",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "(Ce message .{0,60}confidentiel|destin. (exclusivement|uniquement) aux personnes)"
+        },
+        {
+            "id": "confidentiality_es",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "(Este mensaje .{0,60}confidencial|informaci.n confidencial|destinatario)"
+        },
+        {
+            "id": "gdpr_notice",
+            "strength": 0.7,
+            "flags": "i",
+            "pattern": "(privacy (notice|policy)|processes? your personal data|GDPR|isikuandmete töötlemise)"
+        },
+        {
+            "id": "company_registration",
+            "strength": 0.6,
+            "flags": "i",
+            "pattern": "(registered (in|office) .{0,60}(no|number)\\.?\\s*[0-9]|VAT (registration )?(no|number)|Handelsregister|Reg\\.?\\s*nr\\.?\\s*[0-9])"
+        },
+        {
+            "id": "unsubscribe_footer",
+            "strength": 0.8,
+            "flags": "i",
+            "pattern": "(unsubscribe|opt out of these emails|manage your (email )?preferences|loobu (uudiskirjast|teavitustest))"
+        },
+        {
+            "id": "confidentiality_et",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "(Käesolev (e-?kiri|sõnum) .{0,60}(konfidentsiaal|asutusesisene)|kui te ei ole .{0,40}saaja)"
+        },
+        {
+            "id": "automated_do_not_reply",
+            "strength": 0.85,
+            "flags": "i",
+            "comment": "Standard notification preamble or footer.",
+            "pattern": "(this is an? (automated|automatic|system.generated) (e-?mail|message|notification)|please do not reply to this (e-?mail|message)|do not reply directly to this)"
+        },
+        {
+            "id": "receiving_because",
+            "strength": 0.85,
+            "flags": "i",
+            "comment": "The canonical notification footer: why you got this and how to stop it. Recurs across almost every service.",
+            "pattern": "(you (are |have )?receiv(ed|ing) (this|these) .{0,60}(because|since|as (a|an) )|you'?re receiving (this|these) .{0,60}(because|since))"
+        },
+        {
+            "id": "manage_notification_preferences",
+            "strength": 0.8,
+            "flags": "i",
+            "comment": "Opt out and preference management, distinct from a plain unsubscribe link.",
+            "pattern": "(to (stop|change|update|manage|adjust) (receiving |getting )?(these|this|your) .{0,50}(e-?mails?|notifications?|messages?|preferences|settings)|(change|manage|update) your (e-?mail |notification )?(preferences|settings|subscription))"
+        },
+        {
+            "id": "subscribe_invitation_footer",
+            "strength": 0.7,
+            "flags": "i",
+            "comment": "Standing promotional footer on a notification: an invitation to subscribe to something other than what the message is about.",
+            "pattern": "(want to (know|hear|stay|learn) .{0,80}\\?\\s*(would you|subscribe|sign up|join)|would you like to (stay|be) (up.to.date|informed|notified) .{0,60}\\?)"
+        },
+        {
+            "id": "powered_by_footer",
+            "strength": 0.7,
+            "flags": "i",
+            "comment": "Provider branding appended by the sending platform.",
+            "pattern": "(powered by [A-Z]|sent (via|with|using) [A-Z][\\w.]{2,}|delivered by [A-Z][\\w.]{2,})"
+        },
+        {
+            "id": "support_contact_footer",
+            "strength": 0.6,
+            "flags": "i",
+            "comment": "Standing support or help pointer in a service footer.",
+            "pattern": "(if you (have any )?(questions|need help|require assistance)[^.]{0,60}(contact|reach|e-?mail) (us|our|the) (support|team|help)|visit our (help|support) (cent(er|re)|site|page))"
+        }
+    ]
+}

package/dist/detect/patterns/forward-headers.json

@@ -0,0 +1,40 @@
+{
+    "description": "Markers introducing intentionally forwarded content. Forwarded material is preserved as structured, collapsible content and is never dropped, so these patterns open a FORWARDED_CONTENT region rather than a hidden one.",
+    "patterns": [
+        {
+            "id": "begin_forwarded_message",
+            "lang": "*",
+            "strength": 0.97,
+            "flags": "i",
+            "pattern": "^\\s*Begin forwarded message\\s*:?\\s*$"
+        },
+        {
+            "id": "forwarded_message_dashes",
+            "lang": "*",
+            "strength": 0.97,
+            "flags": "i",
+            "pattern": "^\\s*-{2,}\\s*Forwarded message\\s*-{2,}\\s*$"
+        },
+        {
+            "id": "forwarded_message_localised",
+            "lang": "*",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*-{2,}\\s*(Weitergeleitete Nachricht|Message transf.r.|Mensaje reenviado|Messaggio inoltrato|Doorgestuurd bericht|Edastatud kiri|Edasisaadetud (sõnumi|kirja).{0,20}|Пересылаемое сообщение)\\s*-{2,}\\s*$"
+        },
+        {
+            "id": "original_message_forward",
+            "lang": "*",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "^\\s*-{2,}\\s*Forwarded by\\b.{0,120}$"
+        },
+        {
+            "id": "fyi_forward_header_block",
+            "lang": "*",
+            "strength": 0.7,
+            "flags": "i",
+            "pattern": "^\\s*(From|Van|Von|De|Da|Fra|Kellelt|От)\\s*:\\s*.{1,300}$"
+        }
+    ]
+}

package/dist/detect/patterns/html-clients.json

@@ -0,0 +1,157 @@
+{
+    "description": "Client specific HTML structures, expressed as CSS selectors. Prevalence figures are measured on human authored replies in the working corpus and drive the priority order. Each entry declares a role: 'quote' encloses history, 'separator' means everything after it in document order is history, 'signature' and 'disclaimer' mark trailing footers, and 'new' marks containers a client uses for freshly typed text.",
+    "selectors": [
+        {
+            "id": "gmail_quote",
+            "client": "gmail",
+            "role": "quote",
+            "strength": 0.95,
+            "selector": ".gmail_quote",
+            "prevalence": 0.51
+        },
+        {
+            "id": "gmail_quote_container",
+            "client": "gmail",
+            "role": "quote",
+            "strength": 0.95,
+            "selector": ".gmail_quote_container",
+            "prevalence": 0.2
+        },
+        {
+            "id": "gmail_attr",
+            "client": "gmail",
+            "role": "attribution",
+            "strength": 0.92,
+            "selector": ".gmail_attr",
+            "prevalence": 0.47
+        },
+        {
+            "id": "gmail_signature",
+            "client": "gmail",
+            "role": "signature",
+            "strength": 0.9,
+            "selector": ".gmail_signature",
+            "prevalence": 0.27
+        },
+        {
+            "id": "owa_reply_separator",
+            "client": "outlook-web",
+            "role": "separator",
+            "strength": 0.95,
+            "selector": "div#divRplyFwdMsg",
+            "prevalence": 0.26
+        },
+        {
+            "id": "owa_append_on_send",
+            "client": "outlook-web",
+            "role": "separator",
+            "strength": 0.85,
+            "selector": "div#appendonsend",
+            "prevalence": 0.1
+        },
+        {
+            "id": "owa_element_to_proof",
+            "client": "outlook-web",
+            "role": "new",
+            "strength": 0.7,
+            "selector": ".elementToProof",
+            "prevalence": 0.12
+        },
+        {
+            "id": "outlook_mso_normal",
+            "client": "outlook-desktop",
+            "role": "hint",
+            "strength": 0.5,
+            "selector": "p.MsoNormal",
+            "prevalence": 0.16
+        },
+        {
+            "id": "outlook_word_section",
+            "client": "outlook-desktop",
+            "role": "hint",
+            "strength": 0.5,
+            "selector": "div.WordSection1",
+            "prevalence": 0.13
+        },
+        {
+            "id": "apple_mail_cite",
+            "client": "apple-mail",
+            "role": "quote",
+            "strength": 0.9,
+            "selector": "blockquote[type=cite]",
+            "prevalence": 0.04
+        },
+        {
+            "id": "apple_interchange_newline",
+            "client": "apple-mail",
+            "role": "hint",
+            "strength": 0.6,
+            "selector": "br.Apple-interchange-newline",
+            "prevalence": 0.04
+        },
+        {
+            "id": "thunderbird_moz_cite",
+            "client": "thunderbird",
+            "role": "attribution",
+            "strength": 0.95,
+            "selector": ".moz-cite-prefix",
+            "prevalence": 0.008
+        },
+        {
+            "id": "thunderbird_signature",
+            "client": "thunderbird",
+            "role": "signature",
+            "strength": 0.9,
+            "selector": "div.moz-signature",
+            "prevalence": 0.005
+        },
+        {
+            "id": "proton_quote",
+            "client": "proton",
+            "role": "quote",
+            "strength": 0.95,
+            "selector": ".protonmail_quote",
+            "prevalence": 0.002
+        },
+        {
+            "id": "proton_signature",
+            "client": "proton",
+            "role": "signature",
+            "strength": 0.9,
+            "selector": "div.protonmail_signature_block",
+            "prevalence": 0.002
+        },
+        {
+            "id": "yahoo_quoted",
+            "client": "yahoo",
+            "role": "quote",
+            "strength": 0.95,
+            "selector": ".yahoo_quoted",
+            "prevalence": 0.002
+        },
+        {
+            "id": "zimbra_separator",
+            "client": "zimbra",
+            "role": "separator",
+            "strength": 0.9,
+            "selector": "hr#zwchr",
+            "prevalence": 0.001
+        },
+        {
+            "id": "roundcube_quote",
+            "client": "roundcube",
+            "role": "quote",
+            "strength": 0.85,
+            "selector": "blockquote.reply",
+            "prevalence": 0.001
+        },
+        {
+            "id": "generic_blockquote",
+            "client": "*",
+            "role": "quote",
+            "strength": 0.6,
+            "selector": "blockquote",
+            "prevalence": 0.44
+        }
+    ]
+}

package/dist/detect/patterns/quoted-headers.json

@@ -0,0 +1,41 @@
+{
+    "description": "Field names of an embedded copy of a previous message's headers, as Outlook and several webmail clients emit above quoted history. A single line matching one of these is weak evidence on its own, because ordinary prose can start with 'From:'. Two or more of these in consecutive lines, including at least one of the date-like fields, is a strong anchor. The consecutive-line rule lives in the detector, not here.",
+    "fields": [
+        {
+            "id": "from",
+            "role": "originator",
+            "flags": "i",
+            "pattern": "^\\s*(From|De|Von|Van|Da|Fra|Från|Lähettäjä|Kellelt|Saatja|От|Od|Nadawca|Remetente)\\s*:\\s*\\S"
+        },
+        {
+            "id": "sent",
+            "role": "date",
+            "flags": "i",
+            "pattern": "^\\s*(Sent|Date|Datum|Envoy.|Enviado|Inviato|Verzonden|Skickat|Lähetetty|Saadetud|Kuupäev|Отправлено|Дата|Wysłano|Data)\\s*:\\s*\\S"
+        },
+        {
+            "id": "to",
+            "role": "recipient",
+            "flags": "i",
+            "pattern": "^\\s*(To|A|An|Aan|Para|À|Till|Vastaanottaja|Kellele|Saaja|Кому|Do|Adresat)\\s*:\\s*\\S"
+        },
+        {
+            "id": "cc",
+            "role": "recipient",
+            "flags": "i",
+            "pattern": "^\\s*(Cc|CC|Copie|Kopie|Koopia|Kopio|Копия|DW)\\s*:\\s*\\S"
+        },
+        {
+            "id": "subject",
+            "role": "subject",
+            "flags": "i",
+            "pattern": "^\\s*(Subject|Betreff|Objet|Asunto|Oggetto|Onderwerp|Ämne|Aihe|Teema|Teated|Тема|Temat|Assunto)\\s*:\\s*\\S"
+        },
+        {
+            "id": "reply_to",
+            "role": "originator",
+            "flags": "i",
+            "pattern": "^\\s*(Reply-To|Antwort an|Répondre à)\\s*:\\s*\\S"
+        }
+    ]
+}

package/dist/detect/patterns/reply-headers.json

@@ -0,0 +1,139 @@
+{
+    "description": "Attribution lines that a mail client inserts above quoted reply history. Matching one of these is a structural anchor: it is the strongest single piece of evidence that everything following is embedded history rather than new content. Patterns are matched per line against the canonicalised plain text projection.",
+    "patterns": [
+        {
+            "id": "on_x_wrote_en",
+            "lang": "en",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*On\\s+.{3,200}?\\bwrote\\s*:\\s*$"
+        },
+        {
+            "id": "on_x_wrote_en_wrapped",
+            "lang": "en",
+            "comment": "Attribution line whose trailing 'wrote:' was wrapped onto the next line. Requires a date or clock time so that ordinary sentences beginning with 'On ' do not match, and is scored well below an anchor because it cannot be trusted alone. Measured firing on 10.4 percent of the corpus before the date requirement was added.",
+            "strength": 0.5,
+            "flags": "i",
+            "pattern": "^\\s*On\\s+.{3,200}?(\\b\\d{4}\\b|\\b\\d{1,2}:\\d{2}\\b).{0,80}[,:]?\\s*$"
+        },
+        {
+            "id": "x_wrote_trailing",
+            "lang": "en",
+            "strength": 0.8,
+            "flags": "i",
+            "pattern": "^\\s*.{3,200}?\\bwrote\\s*:\\s*$"
+        },
+        {
+            "id": "at_x_wrote",
+            "lang": "en",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "^\\s*At\\s+\\d{1,2}:\\d{2}.{0,200}?\\bwrote\\s*:\\s*$"
+        },
+        {
+            "id": "le_x_a_ecrit_fr",
+            "lang": "fr",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*Le\\s+.{3,200}?a\\s+.crit\\s*:\\s*$"
+        },
+        {
+            "id": "am_x_schrieb_de",
+            "lang": "de",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*(Am|Von)\\s+.{3,200}?schrieb.{0,60}:\\s*$"
+        },
+        {
+            "id": "el_x_escribio_es",
+            "lang": "es",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*El\\s+.{3,200}?escribi.{0,3}\\s*:\\s*$"
+        },
+        {
+            "id": "il_x_ha_scritto_it",
+            "lang": "it",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*Il\\s+.{3,200}?ha\\s+scritto\\s*:\\s*$"
+        },
+        {
+            "id": "em_x_escreveu_pt",
+            "lang": "pt",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*Em\\s+.{3,200}?escreveu\\s*:\\s*$"
+        },
+        {
+            "id": "op_x_schreef_nl",
+            "lang": "nl",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*Op\\s+.{3,200}?schreef\\s*.{0,80}:\\s*$"
+        },
+        {
+            "id": "w_dniu_x_napisal_pl",
+            "lang": "pl",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*W\\s+dniu\\s+.{3,200}?napisa.{1,2}\\s*.{0,40}:\\s*$"
+        },
+        {
+            "id": "x_kirjutas_et",
+            "lang": "et",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*.{3,200}?\\bkirjutas\\s*:\\s*$"
+        },
+        {
+            "id": "x_kirjoitti_fi",
+            "lang": "fi",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*.{3,200}?\\bkirjoitti\\s*:\\s*$"
+        },
+        {
+            "id": "den_x_skrev_sv",
+            "lang": "sv",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*(Den|.{0,60})\\s*.{3,200}?\\bskrev\\s*.{0,80}:\\s*$"
+        },
+        {
+            "id": "x_pisal_ru",
+            "lang": "ru",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*.{3,200}?(писал|написал)\\s*.{0,40}:\\s*$"
+        },
+        {
+            "id": "date_addr_colon",
+            "lang": "*",
+            "strength": 0.85,
+            "flags": "i",
+            "pattern": "^\\s*\\d{4}[-/]\\d{1,2}[-/]\\d{1,2}.{0,120}<[^>]+@[^>]+>.{0,40}:\\s*$"
+        },
+        {
+            "id": "original_message_separator",
+            "lang": "*",
+            "strength": 0.97,
+            "flags": "i",
+            "pattern": "^\\s*-{2,}\\s*(Original Message|Originalnachricht|Message d'origine|Mensaje original|Messaggio originale|Oorspronkelijk bericht|Algne teade|Исходное сообщение)\\s*-{2,}\\s*$"
+        },
+        {
+            "id": "reply_message_separator",
+            "lang": "*",
+            "strength": 0.95,
+            "flags": "i",
+            "pattern": "^\\s*-{2,}\\s*(Reply Message|Antwort|Svar)\\s*-{2,}\\s*$"
+        },
+        {
+            "id": "outlook_underscore_rule",
+            "lang": "*",
+            "strength": 0.6,
+            "flags": "",
+            "pattern": "^_{10,}\\s*$"
+        }
+    ]
+}

package/dist/detect/patterns/signatures.json

@@ -0,0 +1,58 @@
+{
+    "description": "Signature evidence. Signatures are demoted, never deleted, so these patterns only need to be good enough to collapse a trailing block. The RFC 3676 delimiter is the only high confidence rule; everything else is positional evidence that the sequence classifier weighs.",
+    "delimiters": [
+        {
+            "id": "dash_dash_space",
+            "strength": 0.95,
+            "flags": "",
+            "pattern": "^-- ?$"
+        },
+        {
+            "id": "dash_run",
+            "strength": 0.4,
+            "flags": "",
+            "pattern": "^\\s*[-_=*~]{3,}\\s*$"
+        }
+    ],
+    "phrases": [
+        {
+            "id": "sent_from_device",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "^\\s*(Sent from|Get) (my|Outlook|BlueMail|Mail for|Yahoo Mail|Proton Mail|the) .{0,60}$"
+        },
+        {
+            "id": "sent_from_localised",
+            "strength": 0.9,
+            "flags": "i",
+            "pattern": "^\\s*(Von meinem|Envoy. de mon|Enviado desde mi|Inviato da|Verzonden vanaf mijn|Skickat från min|Lähetetty|Saadetud minu|Отправлено с) .{0,60}$"
+        },
+        {
+            "id": "best_regards",
+            "strength": 0.35,
+            "flags": "i",
+            "pattern": "^\\s*(Best regards|Kind regards|Warm regards|Best wishes|Regards|Cheers|Thanks|Thank you|Sincerely|Best|Yours( sincerely| truly)?|Mit freundlichen Grüßen|Viele Grüße|Beste Grüße|Cordialement|Bien à vous|Saludos|Atentamente|Cordiali saluti|Groeten|Met vriendelijke groet|Med vänliga hälsningar|Vänliga hälsningar|Mvh|Ystävällisin terveisin|Terveisin|Parimat|Parimate soovidega|Heade soovidega|Lugupidamisega|Tervitades|Tervitused|Terv|Kõike head|Ilusat päeva|Head aega|Austusega|С уважением|Z poważaniem|Pozdrawiam)\\s*[,.!]?\\s*$",
+            "comment": "Closing greeting. Mined from blocks the reference labelled SIGNATURE that this baseline missed, which is why the Estonian forms are here: the working corpus is roughly ten percent Estonian and those accounted for the largest group of misses."
+        },
+        {
+            "id": "contact_line",
+            "strength": 0.3,
+            "flags": "i",
+            "pattern": "^\\s*(Tel|Phone|Mobile|Mob|Fax|Cell|M|T|E|W)\\s*[:.]?\\s*[+0-9(][-0-9()\\s.]{6,}$"
+        },
+        {
+            "id": "company_registration_line",
+            "comment": "Continental European corporate footers carry a registered office and company number. These lines are boilerplate, never authored prose.",
+            "strength": 0.6,
+            "flags": "i",
+            "pattern": "^\\s*(Geschäftsführer|Handelsregister|Registergericht|Sitz der Gesellschaft|HRB\\s*\\d|USt-IdNr|VAT\\s*(No|ID)|Reg\\.?\\s*(nr|no)\\.?\\s*\\d|Registrikood|Äriregistri)"
+        },
+        {
+            "id": "postal_address_line",
+            "comment": "A street or postal code line in a trailing block. Weak on its own, but it pairs with a name or a closing greeting to identify a corporate signature.",
+            "strength": 0.4,
+            "flags": "i",
+            "pattern": "^\\s*([A-ZÄÖÜÕŠŽ][\\w.-]*\\s+)?(Str\\.|Straße|Strasse|Street|Avenue|Ave\\.|Road|Rd\\.|tee|tänav|pst)\\s*\\.?\\s*\\d|^\\s*[A-Z]{0,2}-?\\d{4,6}\\s+[A-ZÄÖÜÕ]"
+        }
+    ]
+}

package/dist/detect/patterns.d.ts

@@ -0,0 +1,43 @@
+/**
+ * Pattern set loader.
+ *
+ * Patterns live in JSON so they can be extended or overridden without touching code, which matters
+ * because the multilingual and client specific surface keeps growing. A broken pattern is skipped
+ * with a warning rather than crashing the process: a bad regex in a config file must never take
+ * down a corpus run.
+ *
+ * Set EMAIL_CONTENT_PATTERNS to a directory to override the bundled sets. Files are matched by
+ * name, so an override directory may contain only the files it wants to replace.
+ */
+export interface LinePattern {
+    id: string;
+    lang?: string;
+    role?: string;
+    client?: string;
+    strength: number;
+    regex: RegExp;
+}
+export interface SelectorPattern {
+    id: string;
+    client: string;
+    role: 'quote' | 'attribution' | 'separator' | 'signature' | 'disclaimer' | 'new' | 'hint';
+    strength: number;
+    selector: string;
+    prevalence: number;
+}
+export interface PatternSets {
+    replyHeaders: LinePattern[];
+    forwardHeaders: LinePattern[];
+    quotedHeaderFields: LinePattern[];
+    signatureDelimiters: LinePattern[];
+    signaturePhrases: LinePattern[];
+    disclaimerPhrases: LinePattern[];
+    htmlSelectors: SelectorPattern[];
+    /** Ids of patterns that failed to compile, for diagnostics. */
+    invalid: string[];
+}
+/** Load and compile all pattern sets. Cached after the first call. */
+export declare function loadPatterns(overrideDir?: string | null): PatternSets;
+/** Reset the cache. Used by tests that swap pattern directories. */
+export declare function resetPatternCache(): void;
+//# sourceMappingURL=patterns.d.ts.map

package/dist/detect/patterns.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"patterns.d.ts","sourceRoot":"","sources":["../../src/detect/patterns.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAKH,MAAM,WAAW,WAAW;IACxB,EAAE,EAAE,MAAM,CAAC;IACX,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,QAAQ,EAAE,MAAM,CAAC;IACjB,KAAK,EAAE,MAAM,CAAC;CACjB;AAED,MAAM,WAAW,eAAe;IAC5B,EAAE,EAAE,MAAM,CAAC;IACX,MAAM,EAAE,MAAM,CAAC;IACf,IAAI,EAAE,OAAO,GAAG,aAAa,GAAG,WAAW,GAAG,WAAW,GAAG,YAAY,GAAG,KAAK,GAAG,MAAM,CAAC;IAC1F,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,MAAM,CAAC;IACjB,UAAU,EAAE,MAAM,CAAC;CACtB;AAED,MAAM,WAAW,WAAW;IACxB,YAAY,EAAE,WAAW,EAAE,CAAC;IAC5B,cAAc,EAAE,WAAW,EAAE,CAAC;IAC9B,kBAAkB,EAAE,WAAW,EAAE,CAAC;IAClC,mBAAmB,EAAE,WAAW,EAAE,CAAC;IACnC,gBAAgB,EAAE,WAAW,EAAE,CAAC;IAChC,iBAAiB,EAAE,WAAW,EAAE,CAAC;IACjC,aAAa,EAAE,eAAe,EAAE,CAAC;IACjC,+DAA+D;IAC/D,OAAO,EAAE,MAAM,EAAE,CAAC;CACrB;AAwFD,sEAAsE;AACtE,wBAAgB,YAAY,CAAC,WAAW,CAAC,EAAE,MAAM,GAAG,IAAI,GAAG,WAAW,CA6BrE;AAED,oEAAoE;AACpE,wBAAgB,iBAAiB,IAAI,IAAI,CAExC"}