Merge branch 'main' into chunk-data

Resolved 29 conflicting files. The notable decisions:

Confidence: main's tool-calling gate (RequiredProviderConfidence) and this
branch's local-RAG gate (DataConfidenceLevel) turned out to be the same rule
on the same axis, so they are now one field. Both tool results and data
sources raise it through RequireProviderConfidence(). The gate checks the
level strictly and no longer exempts providers trusted by configuration:
TrustedProviderIds is documented as applying to data-source security checks
only, and organizations set confidence through DataConfidence
.CustomConfidenceScheme instead. The security axis (DataSecurity, ERI,
IsTrustedForDataSourceSecurityChecks) is unchanged.

Provider creation: main's CreateProvider signature won (hfEndpointKind,
capabilityOverrides, no model parameter); tokenizerPath was added to it and
is set for every provider, including the new Hetzner, IONOS and LiteLLM.
Provider and EmbeddingProvider combine the record parameters, Lua parsing and
Lua serialization of both sides.

File types: main's hierarchy (ODT leaf, WORD parent, PowerPoint without the
legacy .ppt, TABULAR instead of DELIMITED_TABLE) plus this branch's
SPREADSHEET parent with ODS and the xlsm/xlsb/xla/xlam extensions, which the
runtime already reads. Both sides had added a conflicting HTML filter; the
reading family keeps the name, and the export path uses a narrow
HTML_DOCUMENT, following the existing LATEX/TEX split.

Runtime: main's file_data.rs is the base, including the prompt-injection
sanitizer and the extraction routes. Token counting and chunk segmentation
moved into take_released, so they act on the text the filter has released
rather than on text it is still holding. A failed count is logged and left
out instead of ending the extraction, because the app counts such a segment
itself.

Data sources: the participating-provider checks of this branch are kept, and
main's GetAllowedDataSources overload now builds on them. DirectChatService
resolves the launched chat's data source options before the check, so filter
and chat see the same options.

.NET and Rust both build clean; I18N regenerated to 4060 keys.
This commit is contained in:
Thorsten Sommer committed 2026-09-05 21:20:33 +02:00
commit fe35630eff
640 files changed
+45059 -4952

No files matched your search

+30 -4
View File
@@ -21,6 +21,8 @@ namespace AIStudio.Settings;
/// <param name="IsSelfHosted">Whether the provider is self-hosted.</param>
/// <param name="Hostname">The hostname of the provider. Useful for self-hosted providers.</param>
/// <param name="Model">The LLM model to use for chat.</param>
/// <param name="AllowUserProvidedAPIKey">When set by a configuration plugin, the user may set their own API key for this otherwise locked, enterprise-managed provider.</param>
/// <param name="CustomIconDataUrl">The validated custom SVG icon supplied by a configuration plugin.</param>
public sealed record Provider(
uint Num,
string Id,
@@ -35,7 +37,9 @@ public sealed record Provider(
HFInferenceProvider HFInferenceProvider = HFInferenceProvider.NONE,
string AdditionalJsonApiParameters = "",
string TokenizerPath = "",
ProviderCapabilityOverrides? CapabilityOverrides = null) : ConfigurationBaseObject, ISecretId
ProviderCapabilityOverrides? CapabilityOverrides = null,
bool AllowUserProvidedAPIKey = false,
string CustomIconDataUrl = "") : ConfigurationBaseObject, ISecretId, IUserProvidedAPIKey
{
private static readonly ILogger<Provider> LOGGER = Program.LOGGER_FACTORY.CreateLogger<Provider>();
@@ -84,7 +88,7 @@ public sealed record Provider(
#endregion
public static bool TryParseProviderTable(int idx, LuaTable table, Guid configPluginId, out ConfigurationBaseObject provider)
public static bool TryParseProviderTable(int idx, LuaTable table, Guid configPluginId, string pluginPath, out ConfigurationBaseObject provider)
{
provider = NONE;
if (!table.TryGetValue("Id", out var idValue) || !idValue.TryRead<string>(out var idText) || !Guid.TryParse(idText, out var id))
@@ -154,6 +158,19 @@ public sealed record Provider(
}
var capabilityOverrides = ProviderCapabilityOverrides.TryParseFromLuaTable(idx, table, configPluginId, LOGGER);
var allowUserProvidedApiKey = false;
if (table.TryGetValue("AllowUserProvidedAPIKey", out var allowUserProvidedApiKeyValue) && allowUserProvidedApiKeyValue.TryRead<bool>(out var allowUserProvidedApiKeyBool))
allowUserProvidedApiKey = allowUserProvidedApiKeyBool;
var customIconDataUrl = string.Empty;
if (table.TryGetValue("IconPath", out var iconPathValue))
{
if (!iconPathValue.TryRead<string>(out var iconPath))
LOGGER.LogWarning($"The configured provider {idx} does not contain a valid icon path. Falling back to the built-in provider icon. (Plugin ID: {configPluginId})");
else if (!PluginIconFile.TryLoadDataUrl(iconPath, pluginPath, out customIconDataUrl, out var iconIssue))
LOGGER.LogWarning($"The configured provider {idx} contains an invalid icon path. Falling back to the built-in provider icon. Issue: {iconIssue} (Plugin ID: {configPluginId})");
}
provider = new Provider
{
Num = 0, // will be set later by the PluginConfigurationObject
@@ -170,10 +187,19 @@ public sealed record Provider(
AdditionalJsonApiParameters = additionalJsonApiParameters,
TokenizerPath = tokenizerPath,
CapabilityOverrides = capabilityOverrides,
AllowUserProvidedAPIKey = allowUserProvidedApiKey,
CustomIconDataUrl = customIconDataUrl,
};
// Handle encrypted API key if present:
if (table.TryGetValue("APIKey", out var apiKeyValue) && apiKeyValue.TryRead<string>(out var apiKeyText) && !string.IsNullOrWhiteSpace(apiKeyText))
// Handle an encrypted API key if present. When the user manages their own key for this
// provider, we must never enqueue an embedded key: doing so would overwrite the user's
// key in the OS keyring on every configuration reload.
if (allowUserProvidedApiKey)
{
if (table.TryGetValue("APIKey", out var ignoredApiKeyValue) && ignoredApiKeyValue.TryRead<string>(out var ignoredApiKeyText) && !string.IsNullOrWhiteSpace(ignoredApiKeyText))
LOGGER.LogWarning($"The configured provider {idx} sets both AllowUserProvidedAPIKey and an embedded APIKey. Ignoring the embedded key: the user manages their own key for this provider. (Plugin ID: {configPluginId})");
}
else if (table.TryGetValue("APIKey", out var apiKeyValue) && apiKeyValue.TryRead<string>(out var apiKeyText) && !string.IsNullOrWhiteSpace(apiKeyText))
{
if (!EnterpriseEncryption.IsEncrypted(apiKeyText))
LOGGER.LogWarning($"The configured provider {idx} contains a plaintext API key. Only encrypted API keys (starting with 'ENC:v1:') are supported. (Plugin ID: {configPluginId})");