coffey-labs/ihasmail
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 2 Wiki Activity

Warn about outside senders, large sends and links that mislead

Browse Source 
Four warnings, in Privacy & safety, and all of them start switched off.

That is not timidity. A client that begins by interrupting is one people
learn to click through, and a warning clicked through without reading
costs the same attention and buys nothing. The outside-sender warning
could not be on by default in any case: it measures against the domains
that count as yours, and with nothing configured every message in the
mailbox is from outside.

Your own identity domains are always internal and are not configuration.
An account signed in as [email protected] warning that example.com is
external would be absurd, and making it be typed in first is a foot-gun
that leaves the feature useless the moment it is enabled. Configured
domains are additional, and cover their subdomains -- matched on a dot
boundary, so example.com covers mail.example.com and not notexample.com,
which is exactly the domain somebody registers on purpose.

The four:

A banner names the sender's domain on a message from outside.

Sending outside names the outside recipients and asks, rather than
refusing. "This is going outside" is a rule and not something the sender
can check; a list of addresses is. It reads the full identity list rather
than the visible one, since hiding an identity from the From menu does not
make its domain somebody else's.

Sending to a large group asks once the count crosses a threshold, which is
what catches a reply-all onto a long thread. It counts people rather than
headers, so one address in To and nine in Cc is a message to ten.

Opening a link asks when the destination is not trusted, and always when
the link's text names one domain while its destination is another -- even
where that destination is trusted, because being trusted is not the same
as being the place the text claimed. On that mismatch the offer to trust
the domain is withheld: what would be trusted is the destination, and the
destination is not the thing in question. Anything that is not http or
https is left alone, since warning about a mailto: is noise and noise is
how a warning stops being read. Both bodies are covered, because a link in
a plain-text mail is linkified by us and points wherever it likes just as
readily as one the sender marked up.

The click is cancelled and the navigation re-issued after the answer,
since there is no way to hold a real navigation open across a dialog. The
reopen runs in the continuation of the dialog's own click, which is still
the gesture a popup blocker wants to see.
This commit is contained in:
jcoffey-dev
2026-09-01 22:45:22 -07:00
parent 4c54eb74f9
commit 5b18f1d5d7
9 changed files with 665 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files
web/src/views/settings/PrivacySettings.tsx
+134
View File
@@ -1,4 +1,7 @@
import { useState } from "react";
import { useSettings, type ReadReceiptPolicy } from "@/store/settings";
import { useMail } from "@/store/mail";
import { domainOf } from "@/lib/address";
import { Switch } from "@/ui/misc";
import { X } from "lucide-react";
import { t } from "@/lib/i18n";
@@ -22,6 +25,8 @@ export function PrivacySettings() {
const s = useSettings((st) => st.settings);
const update = useSettings((st) => st.update);
const trusted = s.trustedImageSenders;
const identities = useMail((st) => st.identities);
const ownDomains = [...new Set(identities.map((i) => domainOf(i.email)).filter(Boolean))];
return (
<div>
@@ -74,6 +79,60 @@ export function PrivacySettings() {
</p>
</div>
<h2>{t("Warnings")}</h2>
<p className="hint" style={{ marginTop: -8 }}>
{t("All three start switched off. A client that begins by interrupting is one people learn to click through, and a warning clicked through without reading costs the same attention and buys nothing.")}
</p>
<Switch
checked={s.externalSenderBanner}
onChange={(v) => update({ externalSenderBanner: v })}
label={t("Mark messages from outside")}
hint={t("A banner on any message whose sender is not on one of your own domains.")}
/>
<Switch
checked={s.externalRecipientConfirm}
onChange={(v) => update({ externalRecipientConfirm: v })}
label={t("Ask before sending outside")}
hint={t("Names the outside recipients and asks, rather than refusing.")}
/>
{(s.externalSenderBanner || s.externalRecipientConfirm) && (
<DomainList
label={t("Also count these domains as inside")}
hint={t("Your own identity domains are always inside and do not need listing. A domain here also covers its subdomains.")}
value={s.internalDomains}
onChange={(internalDomains) => update({ internalDomains })}
suggestions={ownDomains}
/>
)}
<div className="field">
<label>{t("Ask before sending to a large group")}</label>
<select className="select" value={String(s.replyAllThreshold)} onChange={(e) => update({ replyAllThreshold: Number(e.target.value) })}>
<option value="0">{t("Never ask")}</option>
<option value="5">{t("5 people or more")}</option>
<option value="10">{t("10 people or more")}</option>
<option value="20">{t("20 people or more")}</option>
<option value="50">{t("50 people or more")}</option>
</select>
<p className="hint">{t("Counts people rather than headers, so one address in To and nine in Cc is a message to ten. Catches a reply-all onto a long thread.")}</p>
</div>
<Switch
checked={s.externalLinkWarning}
onChange={(v) => update({ externalLinkWarning: v })}
label={t("Ask before opening a link in a message")}
hint={t("A link whose text names one domain and whose destination is another is always flagged, even where the destination is trusted — being trusted is not the same as being the place the text claimed.")}
/>
{s.externalLinkWarning && (
<DomainList
label={t("Open links to these domains without asking")}
hint={t("Added here, or from the dialog when a link is opened. A domain also covers its subdomains.")}
value={s.trustedLinkDomains}
onChange={(trustedLinkDomains) => update({ trustedLinkDomains })}
/>
)}
<h2>{t("Before it happens")}</h2>
<div className="field">
<label>{t("Undo send window")}</label>
@@ -91,3 +150,78 @@ export function PrivacySettings() {
</div>
);
}
/**
* A list of domains, added one at a time and removed by their chip.
*
* Typed entries are normalised on the way in -- a leading `@`, stray case, a
* whole address pasted instead of a domain -- because the thing being compared
* against is a hostname, and a list holding "@Example.com " silently matches
* nothing at all.
*/
function DomainList({
label,
hint,
value,
onChange,
suggestions = [],
}: {
label: string;
hint: string;
value: string[];
onChange: (next: string[]) => void;
suggestions?: string[];
}) {
const [draft, setDraft] = useState("");
const add = (raw: string) => {
const d = raw.trim().toLowerCase().replace(/^@/, "").replace(/^.*@/, "").replace(/^https?:\/\//, "").split("/")[0] ?? "";
if (!d || value.includes(d)) {
setDraft("");
return;
}
onChange([...value, d]);
setDraft("");
};
const missing = suggestions.filter((d) => !value.includes(d));
return (
<div className="field">
<label>{label}</label>
{value.length > 0 && (
<div className="trusted-senders">
{value.map((d) => (
<span key={d} className="chip">
<span className="notranslate" translate="no">{d}</span>
<button className="chip-x" aria-label={t("Remove {domain}", { domain: d })} onClick={() => onChange(value.filter((x) => x !== d))}>
<X size={13} />
</button>
</span>
))}
</div>
)}
<div className="row gap-4">
<input
className="input"
value={draft}
placeholder={t("example.com")}
onChange={(e) => setDraft(e.target.value)}
onKeyDown={(e) => {
if (e.key === "Enter") {
e.preventDefault();
add(draft);
}
}}
/>
<button className="btn btn-sm" disabled={!draft.trim()} onClick={() => add(draft)}>{t("Add")}</button>
</div>
{missing.length > 0 && (
<p className="hint">
{t("Your own:")}{" "}
{missing.map((d) => (
<button key={d} className="link-btn notranslate" translate="no" onClick={() => add(d)}>{d}</button>
))}
</p>
)}
<p className="hint">{hint}</p>
</div>
);
}