Public Access
Free the terminal from the Hosts screen, and fill the room it left
The WebView sat inside the Hosts grid, so navigating to Files or the keychain hid every open terminal and the strip that named them. A connection you had opened was invisible from four of the five screens. The window now has two surfaces rather than one: a nav rail that says which page you are on, and a terminal strip that is always there and switches the whole content area to a shell. Screen keeps meaning "which page" and never becomes a sixth kind of page, which is why this is two properties instead of one enum with a terminal member in it. Every screen lives inside one wrapper panel that collapses when a terminal is showing. That is not tidiness — the WebView hosts a Win32 child window that composites above everything Avalonia draws, so a screen left visible over its rectangle is a screen sliced in half, and this window has shipped that defect once already. One decision point, IsTerminalShowing, and a nested panel rather than five compound bindings nobody would remember to extend. The focus choreography is the part no test in this repo can see. Every reveal path now focuses in the same turn the WebView appeared, so all three of them post at DispatcherPriority.Loaded and let the native control re-push its bounds first. Going the other way had a real bug: the screen-changed branch called a bare Focus() where it had to release the keyboard from the native child, so switching from a terminal to Files silently ate the first keystrokes. Rare before this commit and the primary gesture after it. The tab strip grew a cross inside each tab, a plus that opens the quick-connect palette, and middle-click close. Nested buttons are correct here: Avalonia handles a left press on the cross and deliberately does not handle other buttons, which is exactly what lets middle-click bubble up from the cross as well as the tab. The test is PointerUpdateKind rather than IsMiddleButtonPressed, because the latter reports button state and is also true for a left press made while the middle button happens to be held. The handler is on the tab and not the strip, so the background closes nothing by construction. Plus opens the palette rather than a flyout, since a menu dropping into the WebView's rectangle may or may not composite above a child HWND and this repo does not make rendering claims it has not photographed. Everything a user reads now says keychain. The wire, the database and the cryptographic spec still say vault, deliberately: renaming those is a migration and a protocol change for a word. That split is written down rather than left to be rediscovered as an inconsistency. Four things that were squeezed into the keychain's category rail, or into nothing at all, now have screens. Pinned host keys get one, with fingerprints never truncated and a filter that matches them, because comparing what you have against what the operator published is the whole workflow; the approved date is read out of the item's UUIDv7 rather than added as a column, and says so, since it means first approval and not last use. Keys can be generated in the client, which needed the openssh-key-v1 container written by hand — there is no BCL or NSec helper, and the PKCS#8 route is unverified in the SSH library this uses. The armour carries no passphrase: encrypting it needs bcrypt_pbkdf, which is Blowfish with a swizzle, in a project whose crypto is otherwise entirely libsodium, for a protection the key's own remarks argue is redundant inside a vault. Generation fills the existing editor and stops, so SAVE stays the one thing that writes. ~/.ssh/config can be imported behind a preview that is ticked per row and writes nothing until the button; IdentityFile records the path and imports the key material only on an explicit opt-in, because reading somebody's private key into a vault is precisely the act this product exists to make deliberate. Match blocks and ProxyJump are reported rather than obeyed — one cannot be evaluated statically and the other has nothing behind it to route with, and a preview that implied otherwise would be worse than one that admits it. Files can be dragged in all four directions that are honestly available. Remote to Explorer does not ship and is not pretended to: the shell wants the bytes during the drop, which needs a virtual file and a native COM data object, outside what Avalonia offers. Note for the next person that Avalonia 12 replaced the drag model outright — DataObject and DataFormats are no-op stubs and IDataObject is not in the reference assembly, so every tutorial written for 11 does not compile here. Hosts can be grouped, flat and never nested. A parent id merged as a scalar lets two offline clients each re-parent A under B and B under A, producing a cycle inside an encrypted payload that no server can police and every reader would have to detect for ever. Membership lives in that payload rather than in the one plaintext concession ADR 0001 allows, whose test is that the relay cannot function without it — nothing on the server reads a group, so what plaintext would hand over is a clustering of the estate for nothing. The plaintext column reserved for it is dropped, provably always null, and the server now refuses a client that sends one; it was never populated, was copied on apply, and was not cleared on delete, so a group id would have outlived the host it described. Snippets insert through xterm rather than through the pump, because xterm is the only thing that knows whether the remote has bracketed paste on, and that is what makes a shell treat embedded newlines as text instead of as execute. The host process moves opaque bytes and never parses output, so it would have to guess, and guessing wrong runs every line. Running is off by default and the copy says the text goes into whatever is there — the terminal has no notion of being at a prompt, and may be in vi or at a password prompt with echo off, so the Enter the user presses themselves is the entire safety property. Connections and keychain changes are recorded as synced encrypted items, which is what makes them auditable by a team later and costs the server knowledge of connection rate and timing from row counts alone. ADR 0001 already concedes it cannot hide that class of metadata; the trade is now written into it rather than left implicit. A connection entry is written once, at close, which is what makes a synced log tractable: nothing to merge, one outbox row, no chance of colliding with itself. Live sessions come from memory, not from the log. The write is void by contract and posts to a bounded channel, because putting an encrypt-and-write on the teardown path of every session is how closing the application comes to take four seconds. A ticket opened before a lock still closes afterwards, since a shell outlives the vault. The activity log hooks the one generic repository every kind writes through, so it cannot miss a caller — which is also why the log kinds themselves declare they are not audited, or the first entry would write an entry about writing an entry. It records the names of the fields that changed and never their values; a log with an old password in it would be a plaintext credential store with no vault around it. Retention is 90 days or 5,000 entries, whichever bites first, pruned on the sync loop rather than on a second timer. That log traffic then broke the status line, which is worth recording because the fix is a shape and not a patch: background sync counted its own log rows as pushed items, so the quiet rule stopped being quiet and every action's message was overwritten a second later by a sync report. The report now separates log rows from user items and the rule reads the latter. S3 buckets appear as a remote in the file browser, behind the same interface an SFTP session implements, so the queue and both panes did not have to learn what they are talking to. Uploads go through a pipe, because the queue wants to write and the SDK wants to read; memory is then bounded by the part size instead of buffering a file to disk twice. Finally, the Windows device key store moved out of the session project, which was the one thing keeping it from being portable — everything else in it is platform-neutral, and a Windows CNG dependency in the middle of the vault code meant a second head could not reference it without dragging Windows along. The seam that made the move free was already there. docs/android-port.md is the audit behind that: what ports, what does not, in order of cost, the four decisions taken, and an inventory of every screen and state the interface has to carry, written so a design can be made from it directly. dotnet build, dotnet test and dotnet format --verify-no-changes are all clean: 1240 tests at zero warnings, including the end-to-end suite against real containers. The manual checks that headless Avalonia cannot make — the drag from Explorer, a generated key against a real host, twelve tabs at the minimum window width — are listed in docs/manual-checks.md and are still outstanding.
This commit is contained in:
@@ -7,6 +7,12 @@
|
||||
|
||||
<ItemGroup>
|
||||
<PackageReference Include="SSH.NET" />
|
||||
<!--
|
||||
For Ed25519 key generation, which .NET has no primitive for at all. The same reason DodoSSH.Crypto
|
||||
takes it; see Directory.Packages.props. Nothing else here touches it, and no key generated with it is
|
||||
part of the DSH1 envelope — this is an SSH file format, not the vault's cryptography.
|
||||
-->
|
||||
<PackageReference Include="NSec.Cryptography" />
|
||||
</ItemGroup>
|
||||
|
||||
<ItemGroup>
|
||||
|
||||
@@ -0,0 +1,211 @@
|
||||
using System.Buffers.Binary;
|
||||
using System.Security.Cryptography;
|
||||
using System.Text;
|
||||
|
||||
namespace DodoSSH.Client.Ssh;
|
||||
|
||||
/// <summary>
|
||||
/// Writes the two files <c>ssh-keygen</c> writes.
|
||||
/// </summary>
|
||||
/// <remarks>
|
||||
/// <para>
|
||||
/// <b>Why this is hand-rolled.</b> Neither the BCL nor NSec can write an OpenSSH private key. .NET has no
|
||||
/// Ed25519 at all — that is why NSec is here in the first place — and the <c>openssh-key-v1</c> container is
|
||||
/// an SSH-specific framing that no general-purpose library emits. The one alternative was PKCS#8 with the
|
||||
/// Ed25519 OID <c>1.3.101.112</c>, which would also have had to be hand-encoded and which SSH.NET 2025.1.0
|
||||
/// is not confirmed to parse — its PKCS#8 path historically switches on RSA, DSA and EC OIDs only. This
|
||||
/// format is the one SSH.NET definitely reads and the one every other tool reads.
|
||||
/// </para>
|
||||
/// <para>
|
||||
/// <b>The container is written unencrypted, on purpose.</b> Encrypting one needs <c>bcrypt_pbkdf</c> —
|
||||
/// Blowfish with a byte-swizzling quirk — plus AES-256-CTR. .NET has no Blowfish, <c>Rfc2898DeriveBytes</c>
|
||||
/// is in <c>BannedSymbols.txt</c> and is the wrong primitive anyway, and the only oracle for a hand-written
|
||||
/// implementation is <c>ssh-keygen</c> itself. That is a standing crypto maintenance cost for a defence the
|
||||
/// product does not need: a passphrase protects a key file sitting on a disk, and a key generated here goes
|
||||
/// straight into an end-to-end encrypted keychain and never touches one. See
|
||||
/// <c>SshKeySecret.Passphrase</c>, which makes the same argument at more length.
|
||||
/// </para>
|
||||
/// <para>
|
||||
/// Everything below is length-prefixed big-endian, which is the whole of the SSH wire format. Getting a
|
||||
/// prefix wrong yields a file that parses far enough to look plausible and then fails authentication with
|
||||
/// an error that says nothing about the encoding.
|
||||
/// </para>
|
||||
/// </remarks>
|
||||
internal static class OpenSshKeyWriter
|
||||
{
|
||||
private const string Magic = "openssh-key-v1\0";
|
||||
|
||||
private const string Ed25519Algorithm = "ssh-ed25519";
|
||||
|
||||
private const string RsaAlgorithm = "ssh-rsa";
|
||||
|
||||
/// <summary>
|
||||
/// How wide the base64 body is wrapped.
|
||||
/// </summary>
|
||||
/// <remarks>
|
||||
/// OpenSSH writes 70. Nothing parses by line length — but a key that diffs against one <c>ssh-keygen</c>
|
||||
/// produced, in a repository or a paste, should differ in its bytes and not in its wrapping.
|
||||
/// </remarks>
|
||||
private const int WrapAt = 70;
|
||||
|
||||
/// <summary>
|
||||
/// The armoured private key for an Ed25519 pair, in <c>openssh-key-v1</c> form.
|
||||
/// </summary>
|
||||
/// <param name="seed">The 32-byte private scalar seed, as NSec exports it.</param>
|
||||
/// <param name="publicKey">The 32-byte public point.</param>
|
||||
/// <param name="comment">The trailing comment, which OpenSSH stores inside the private section.</param>
|
||||
internal static string WriteEd25519PrivateKey(
|
||||
ReadOnlySpan<byte> seed,
|
||||
ReadOnlySpan<byte> publicKey,
|
||||
string comment)
|
||||
{
|
||||
var publicBlob = Ed25519PublicBlob(publicKey);
|
||||
|
||||
using var privateSection = new MemoryStream();
|
||||
|
||||
// Two copies of the same random value. OpenSSH uses them as a decryption check: after decrypting an
|
||||
// encrypted key it compares them, and a mismatch is a wrong passphrase. Nothing here is encrypted,
|
||||
// so nothing checks them — they are written because the format says so, and a parser is entitled to
|
||||
// insist.
|
||||
var check = RandomNumberGenerator.GetBytes(4);
|
||||
privateSection.Write(check);
|
||||
privateSection.Write(check);
|
||||
|
||||
WriteString(privateSection, Ed25519Algorithm);
|
||||
WriteString(privateSection, publicKey);
|
||||
|
||||
// The private field of an Ed25519 OpenSSH key is the seed followed by the public point, 64 bytes,
|
||||
// not the 32-byte seed alone. A file carrying only the seed loads and then signs with a key whose
|
||||
// public half nobody agrees on.
|
||||
Span<byte> expanded = stackalloc byte[64];
|
||||
seed.CopyTo(expanded);
|
||||
publicKey.CopyTo(expanded[32..]);
|
||||
WriteString(privateSection, expanded);
|
||||
|
||||
WriteString(privateSection, comment);
|
||||
|
||||
Pad(privateSection);
|
||||
|
||||
using var container = new MemoryStream();
|
||||
container.Write(Encoding.ASCII.GetBytes(Magic));
|
||||
WriteString(container, "none");
|
||||
WriteString(container, "none");
|
||||
WriteString(container, ReadOnlySpan<byte>.Empty);
|
||||
WriteUInt32(container, 1);
|
||||
WriteString(container, publicBlob);
|
||||
WriteString(container, privateSection.ToArray());
|
||||
|
||||
return Armour("OPENSSH PRIVATE KEY", container.ToArray());
|
||||
}
|
||||
|
||||
/// <summary>The <c>authorized_keys</c> line for an Ed25519 public point.</summary>
|
||||
internal static string WriteEd25519PublicKey(ReadOnlySpan<byte> publicKey, string comment) =>
|
||||
PublicLine(Ed25519Algorithm, Ed25519PublicBlob(publicKey), comment);
|
||||
|
||||
/// <summary>The <c>authorized_keys</c> line for an RSA key.</summary>
|
||||
internal static string WriteRsaPublicKey(RSA rsa, string comment) =>
|
||||
PublicLine(RsaAlgorithm, RsaPublicBlob(rsa), comment);
|
||||
|
||||
/// <summary>The raw public key blob, which is what a fingerprint is taken over.</summary>
|
||||
internal static byte[] Ed25519PublicBlob(ReadOnlySpan<byte> publicKey)
|
||||
{
|
||||
using var blob = new MemoryStream();
|
||||
WriteString(blob, Ed25519Algorithm);
|
||||
WriteString(blob, publicKey);
|
||||
|
||||
return blob.ToArray();
|
||||
}
|
||||
|
||||
/// <inheritdoc cref="Ed25519PublicBlob" />
|
||||
internal static byte[] RsaPublicBlob(RSA rsa)
|
||||
{
|
||||
var parameters = rsa.ExportParameters(includePrivateParameters: false);
|
||||
|
||||
using var blob = new MemoryStream();
|
||||
WriteString(blob, RsaAlgorithm);
|
||||
WriteMpint(blob, parameters.Exponent!);
|
||||
WriteMpint(blob, parameters.Modulus!);
|
||||
|
||||
return blob.ToArray();
|
||||
}
|
||||
|
||||
private static string PublicLine(string algorithm, byte[] blob, string comment)
|
||||
{
|
||||
var line = $"{algorithm} {Convert.ToBase64String(blob)}";
|
||||
|
||||
return string.IsNullOrWhiteSpace(comment) ? line : $"{line} {comment.Trim()}";
|
||||
}
|
||||
|
||||
/// <remarks>
|
||||
/// To a multiple of eight, with the bytes 1, 2, 3… — the block size of the "none" cipher, which OpenSSH
|
||||
/// applies even though nothing is being blocked. This is the classic place to get an
|
||||
/// <c>openssh-key-v1</c> writer wrong, because whether it is wrong depends on the length of the comment:
|
||||
/// a name that happens to land on a boundary produces a file that loads everywhere, and one character
|
||||
/// more produces one that does not.
|
||||
/// </remarks>
|
||||
private static void Pad(Stream destination)
|
||||
{
|
||||
var remainder = (int)(destination.Length % 8);
|
||||
|
||||
if (remainder == 0)
|
||||
{
|
||||
return;
|
||||
}
|
||||
|
||||
for (var i = 1; i <= 8 - remainder; i++)
|
||||
{
|
||||
destination.WriteByte((byte)i);
|
||||
}
|
||||
}
|
||||
|
||||
private static string Armour(string label, byte[] body)
|
||||
{
|
||||
var builder = new StringBuilder();
|
||||
builder.Append("-----BEGIN ").Append(label).Append("-----\n");
|
||||
|
||||
var base64 = Convert.ToBase64String(body);
|
||||
|
||||
for (var offset = 0; offset < base64.Length; offset += WrapAt)
|
||||
{
|
||||
builder.Append(base64.AsSpan(offset, Math.Min(WrapAt, base64.Length - offset))).Append('\n');
|
||||
}
|
||||
|
||||
builder.Append("-----END ").Append(label).Append("-----\n");
|
||||
|
||||
return builder.ToString();
|
||||
}
|
||||
|
||||
private static void WriteString(Stream destination, string value) =>
|
||||
WriteString(destination, Encoding.UTF8.GetBytes(value));
|
||||
|
||||
private static void WriteString(Stream destination, ReadOnlySpan<byte> value)
|
||||
{
|
||||
WriteUInt32(destination, (uint)value.Length);
|
||||
destination.Write(value);
|
||||
}
|
||||
|
||||
private static void WriteUInt32(Stream destination, uint value)
|
||||
{
|
||||
Span<byte> encoded = stackalloc byte[4];
|
||||
BinaryPrimitives.WriteUInt32BigEndian(encoded, value);
|
||||
destination.Write(encoded);
|
||||
}
|
||||
|
||||
/// <remarks>
|
||||
/// Signed big-endian, so a leading byte with its high bit set needs a zero in front of it or it reads as
|
||||
/// a negative number. An RSA modulus has that bit set roughly half the time, which is what makes this
|
||||
/// the kind of bug that ships.
|
||||
/// </remarks>
|
||||
private static void WriteMpint(Stream destination, byte[] value)
|
||||
{
|
||||
if (value.Length > 0 && (value[0] & 0x80) != 0)
|
||||
{
|
||||
var padded = new byte[value.Length + 1];
|
||||
value.CopyTo(padded, 1);
|
||||
WriteString(destination, padded);
|
||||
return;
|
||||
}
|
||||
|
||||
WriteString(destination, value);
|
||||
}
|
||||
}
|
||||
@@ -214,14 +214,40 @@ public static class SftpPath
|
||||
/// transfer runs is fine, and is the point of not opening a session per transfer.
|
||||
/// </para>
|
||||
/// </remarks>
|
||||
public interface ISftpSession : IAsyncDisposable
|
||||
public interface ISftpSession : IRemoteFileStore
|
||||
{
|
||||
/// <summary>The host key that was accepted for this session.</summary>
|
||||
HostKeyPresentation HostKey { get; }
|
||||
}
|
||||
|
||||
/// <summary>
|
||||
/// A remote place with files in it, whatever protocol reaches it.
|
||||
/// </summary>
|
||||
/// <remarks>
|
||||
/// <para>
|
||||
/// Extracted from <see cref="ISftpSession"/> when buckets arrived, and unchanged in shape — the transfer
|
||||
/// queue reads, writes, stats and lists, and never once needed anything SSH-specific. What stayed behind on
|
||||
/// <c>ISftpSession</c> is the one member that could not be answered by a bucket: a host key.
|
||||
/// </para>
|
||||
/// <para>
|
||||
/// <b>It lives in a project called <c>.Ssh</c>, which is a naming debt worth writing down rather than
|
||||
/// paying.</b> <see cref="SftpEntry"/> is here too and is the type every listing is made of, so moving the
|
||||
/// interface without moving that would split the vocabulary in half — and moving both means renaming a
|
||||
/// record that the whole file browser and its tests are written against. The cost of leaving it is a
|
||||
/// reference that reads oddly from the object-store project; the cost of moving it is a rename with no
|
||||
/// behaviour in it.
|
||||
/// </para>
|
||||
/// <para>
|
||||
/// <b>Not every implementation can do everything, and the contract says which.</b> An object store has no
|
||||
/// directories, no rename and no way to resume a half-finished upload; each of those is documented on the
|
||||
/// member and refused with a reason rather than silently approximated. See <c>S3FileStore</c>.
|
||||
/// </para>
|
||||
/// </remarks>
|
||||
public interface IRemoteFileStore : IAsyncDisposable
|
||||
{
|
||||
/// <summary>Whether the transport is still up.</summary>
|
||||
bool IsConnected { get; }
|
||||
|
||||
/// <summary>The host key that was accepted for this session.</summary>
|
||||
HostKeyPresentation HostKey { get; }
|
||||
|
||||
/// <summary>
|
||||
/// Where the session starts, which is the account's home directory.
|
||||
/// </summary>
|
||||
|
||||
@@ -0,0 +1,122 @@
|
||||
using System.Security.Cryptography;
|
||||
using NSec.Cryptography;
|
||||
|
||||
namespace DodoSSH.Client.Ssh;
|
||||
|
||||
/// <summary>Which kind of key pair to make.</summary>
|
||||
public enum SshKeyAlgorithm
|
||||
{
|
||||
/// <summary>Ed25519. Small, fast, and what every current OpenSSH prefers.</summary>
|
||||
Ed25519 = 0,
|
||||
|
||||
/// <summary>RSA at 4096 bits, for servers too old to accept the above.</summary>
|
||||
Rsa4096 = 1,
|
||||
}
|
||||
|
||||
/// <summary>
|
||||
/// A freshly generated key pair, in the two forms anybody needs it in.
|
||||
/// </summary>
|
||||
/// <param name="PrivateKeyArmour">
|
||||
/// The private half, in the armoured form <c>ssh-keygen</c> writes. Goes straight into
|
||||
/// <c>SshKeySecret.PrivateKeyPem</c>, which stores it verbatim.
|
||||
/// </param>
|
||||
/// <param name="PublicKeyLine">
|
||||
/// The public half, as one <c>authorized_keys</c> line. This is what gets installed on a host.
|
||||
/// </param>
|
||||
/// <param name="Fingerprint">
|
||||
/// The <c>SHA256:…</c> fingerprint, in the format <c>ssh-keygen -lf</c> prints, so it can be read out to
|
||||
/// somebody or compared against what a host reports.
|
||||
/// </param>
|
||||
public sealed record GeneratedSshKey(string PrivateKeyArmour, string PublicKeyLine, string Fingerprint);
|
||||
|
||||
/// <summary>
|
||||
/// Makes a new SSH key pair without shelling out to <c>ssh-keygen</c>.
|
||||
/// </summary>
|
||||
/// <remarks>
|
||||
/// <para>
|
||||
/// <b>Why the client can do this at all.</b> Every part is already here: NSec does Ed25519 because .NET
|
||||
/// does not, the BCL does RSA, and the SSH wire encoding is a few length-prefixed strings — see
|
||||
/// <see cref="OpenSshKeyWriter"/>. What it buys is that the private key is never written to a disk. The
|
||||
/// alternative flow is "run ssh-keygen, find the file, open it, copy the text, paste it here, remember to
|
||||
/// delete the file", and the last step is the one nobody does.
|
||||
/// </para>
|
||||
/// <para>
|
||||
/// <b>The armour has no passphrase</b>, and that is a deliberate limitation with its reasoning in
|
||||
/// <see cref="OpenSshKeyWriter"/>. The key is protected by the keychain it lands in.
|
||||
/// </para>
|
||||
/// <para>
|
||||
/// This lives in the SSH project rather than in <c>DodoSSH.Crypto</c>, which is the normative
|
||||
/// implementation of <c>docs/crypto.md</c> and has nothing to say about SSH file formats. It is also where
|
||||
/// <see cref="SshHostKeyFingerprint"/> already lives, and a second <c>SHA256:</c> encoder would be a second
|
||||
/// thing to get wrong.
|
||||
/// </para>
|
||||
/// </remarks>
|
||||
public static class SshKeyGenerator
|
||||
{
|
||||
/// <summary>
|
||||
/// Generates a key pair.
|
||||
/// </summary>
|
||||
/// <param name="algorithm">Which kind.</param>
|
||||
/// <param name="comment">
|
||||
/// The trailing comment, conventionally <c>user@machine</c>. It identifies the key in a host's
|
||||
/// <c>authorized_keys</c> and is the only thing there that will say where it came from.
|
||||
/// </param>
|
||||
/// <remarks>
|
||||
/// Synchronous and CPU-bound. RSA at 4096 bits is seconds of work on an ordinary machine, so a caller on
|
||||
/// a UI thread has to move this to one of its own — the window would otherwise freeze at exactly the
|
||||
/// moment somebody is watching it. Ed25519 is effectively instant, and the caller should not have to
|
||||
/// know which is which.
|
||||
/// </remarks>
|
||||
public static GeneratedSshKey Generate(SshKeyAlgorithm algorithm, string comment) => algorithm switch
|
||||
{
|
||||
SshKeyAlgorithm.Ed25519 => Ed25519(comment),
|
||||
SshKeyAlgorithm.Rsa4096 => Rsa4096(comment),
|
||||
_ => throw new ArgumentOutOfRangeException(nameof(algorithm)),
|
||||
};
|
||||
|
||||
private static GeneratedSshKey Ed25519(string comment)
|
||||
{
|
||||
var parameters = new KeyCreationParameters
|
||||
{
|
||||
// The seed has to come back out to be written into the file. NSec holds key material in
|
||||
// libsodium's guarded memory and refuses to export it unless asked at creation time.
|
||||
ExportPolicy = KeyExportPolicies.AllowPlaintextExport,
|
||||
};
|
||||
|
||||
using var key = Key.Create(SignatureAlgorithm.Ed25519, parameters);
|
||||
|
||||
var seed = key.Export(KeyBlobFormat.RawPrivateKey);
|
||||
|
||||
try
|
||||
{
|
||||
var publicKey = key.PublicKey.Export(KeyBlobFormat.RawPublicKey);
|
||||
|
||||
return new GeneratedSshKey(
|
||||
OpenSshKeyWriter.WriteEd25519PrivateKey(seed, publicKey, comment),
|
||||
OpenSshKeyWriter.WriteEd25519PublicKey(publicKey, comment),
|
||||
SshHostKeyFingerprint.Format(OpenSshKeyWriter.Ed25519PublicBlob(publicKey)));
|
||||
}
|
||||
finally
|
||||
{
|
||||
// The one copy of the private scalar this method makes, and it is an ordinary managed array
|
||||
// outside libsodium's guarded memory. Clearing it does not undo anything the garbage collector
|
||||
// may already have moved, which is why the export happens once and is used immediately.
|
||||
CryptographicOperations.ZeroMemory(seed);
|
||||
}
|
||||
}
|
||||
|
||||
/// <remarks>
|
||||
/// PKCS#1, which is what <c>ExportRSAPrivateKeyPem</c> writes and what SSH.NET's <c>RSA PRIVATE KEY</c>
|
||||
/// branch reads. No hand-encoding is needed on this path at all — only the public line, because there is
|
||||
/// no BCL helper for the SSH wire format.
|
||||
/// </remarks>
|
||||
private static GeneratedSshKey Rsa4096(string comment)
|
||||
{
|
||||
using var rsa = RSA.Create(4096);
|
||||
|
||||
return new GeneratedSshKey(
|
||||
rsa.ExportRSAPrivateKeyPem() + "\n",
|
||||
OpenSshKeyWriter.WriteRsaPublicKey(rsa, comment),
|
||||
SshHostKeyFingerprint.Format(OpenSshKeyWriter.RsaPublicBlob(rsa)));
|
||||
}
|
||||
}
|
||||
@@ -14,6 +14,15 @@
|
||||
"resolved": "5.6.0",
|
||||
"contentHash": "Kcobt3pnOdO0A+6CKiMHZdTEluJpsfxiV20axtZdmfBQnDmiWTKPJADlgAfdTuKNAnVarrkJa0UEGwuOo91muw=="
|
||||
},
|
||||
"NSec.Cryptography": {
|
||||
"type": "Direct",
|
||||
"requested": "[26.4.0, )",
|
||||
"resolved": "26.4.0",
|
||||
"contentHash": "0vsCtY5f+YgQROiWNqzgWp+l2pddfk9FkWoGV/bEo0MuEYPKlJWuoA8aOfO6qp3f+EnObKE3zSJhn1PspJeJVg==",
|
||||
"dependencies": {
|
||||
"libsodium": "[1.0.22, 1.0.23)"
|
||||
}
|
||||
},
|
||||
"SSH.NET": {
|
||||
"type": "Direct",
|
||||
"requested": "[2025.1.0, )",
|
||||
@@ -42,6 +51,12 @@
|
||||
"requested": "[2.6.2, )",
|
||||
"resolved": "2.6.2",
|
||||
"contentHash": "7oWOcvnntmMKNzDLsdxAYqApt+AjpRpP2CShjMfIa3umZ42UQMvH0tl1qAliYPNYO6vTdcGMqnRrCPmsfzTI1w=="
|
||||
},
|
||||
"libsodium": {
|
||||
"type": "CentralTransitive",
|
||||
"requested": "[1.0.22, )",
|
||||
"resolved": "1.0.22",
|
||||
"contentHash": "KPD9SloJFclrsjnhABu7dzWrcyYkwPbvx5l1gRSPAX/0n+OBtSiVCKtGFv4n+ecWUHU0tCG9LSSwoZZx673zBQ=="
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user