Skip to content

refine chat image reading #537

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jul 11, 2024
Merged

refine chat image reading #537

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
86779b0
refine chat image reading
Jul 11, 2024
5aa4595
minor change
Jul 11, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion src/Infrastructure/BotSharp.Core/Using.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,6 @@
global using BotSharp.Abstraction.Files.Enums;
global using BotSharp.Abstraction.Translation.Attributes;
global using BotSharp.Abstraction.Messaging.Enums;
global using BotSharp.Abstraction.Http.Settings;
global using BotSharp.Core.Repository;
global using BotSharp.Core.Routing;
global using BotSharp.Core.Agents.Services;
Expand Down
2 changes: 1 addition & 1 deletion src/Infrastructure/BotSharp.OpenAPI/Controllers/InstructModeController.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -85,7 +85,7 @@ public async Task<string> MultiModalCompletion([FromBody] IncomingMessageModel i
try
{
var completion = CompletionProvider.GetChatCompletion(_services, provider: input.Provider ?? "openai",
modelId: input.ModelId ?? "gpt-4", multiModal: true);
model: input.Model ?? "gpt-4o", multiModal: true);
var message = await completion.GetChatCompletions(new Agent()
{
Id = Guid.Empty.ToString(),
Expand Down
50 changes: 21 additions & 29 deletions src/Plugins/BotSharp.Plugin.AzureOpenAI/Providers/Chat/ChatCompletionProvider.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -193,7 +193,6 @@ public async Task<bool> GetChatCompletionsStreamingAsync(Agent agent, List<RoleD
return true;
}


protected (string, IEnumerable<ChatMessage>, ChatCompletionOptions) PrepareOptions(Agent agent, List<RoleDialogModel> conversations)
{
var agentService = _services.GetRequiredService<IAgentService>();
Expand Down Expand Up @@ -257,40 +256,34 @@ public async Task<bool> GetChatCompletionsStreamingAsync(Agent agent, List<RoleD
{
var text = !string.IsNullOrWhiteSpace(message.Payload) ? message.Payload : message.Content;
var textPart = ChatMessageContentPart.CreateTextMessageContentPart(text);
var chat = new UserChatMessage(textPart)
{
ParticipantName = message.FunctionName
};
var contentParts = new List<ChatMessageContentPart> { textPart };

if (allowMultiModal)
if (allowMultiModal && !message.Files.IsNullOrEmpty())
{
if (!message.Files.IsNullOrEmpty())
foreach (var file in message.Files)
{
foreach (var file in message.Files)
if (!string.IsNullOrEmpty(file.FileUrl))
{
var uri = new Uri(file.FileUrl);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(uri, ImageChatMessageContentPartDetail.Low);
contentParts.Add(contentPart);
}
else if (!string.IsNullOrEmpty(file.FileData))
{
if (!string.IsNullOrEmpty(file.FileUrl))
{
var uri = new Uri(file.FileUrl);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(uri, ImageChatMessageContentPartDetail.Low);
chat = new UserChatMessage(textPart, contentPart) { ParticipantName = message.FunctionName };
}
else if (!string.IsNullOrEmpty(file.FileData))
{
var (contentType, bytes) = fileService.GetFileInfoFromData(file.FileData);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(BinaryData.FromBytes(bytes), contentType, ImageChatMessageContentPartDetail.Low);
chat = new UserChatMessage(textPart, contentPart) { ParticipantName = message.FunctionName };
}
else if (!string.IsNullOrEmpty(file.FileStorageUrl))
{
var contentType = fileService.GetFileContentType(file.FileStorageUrl);
using var stream = File.OpenRead(file.FileStorageUrl);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(BinaryData.FromStream(stream), contentType, ImageChatMessageContentPartDetail.Low);
chat = new UserChatMessage(textPart, contentPart) { ParticipantName = message.FunctionName };
}
var (contentType, bytes) = fileService.GetFileInfoFromData(file.FileData);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(BinaryData.FromBytes(bytes), contentType, ImageChatMessageContentPartDetail.Low);
contentParts.Add(contentPart);
}
else if (!string.IsNullOrEmpty(file.FileStorageUrl))
{
var contentType = fileService.GetFileContentType(file.FileStorageUrl);
using var stream = File.OpenRead(file.FileStorageUrl);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(BinaryData.FromStream(stream), contentType, ImageChatMessageContentPartDetail.Low);
contentParts.Add(contentPart);
}
}
}
messages.Add(chat);
messages.Add(new UserChatMessage(contentParts) { ParticipantName = message.FunctionName });
}
else if (message.Role == AgentRole.Assistant)
{
Expand All @@ -302,7 +295,6 @@ public async Task<bool> GetChatCompletionsStreamingAsync(Agent agent, List<RoleD
return (prompt, messages, options);
}


private string GetPrompt(IEnumerable<ChatMessage> messages, ChatCompletionOptions options)
{
var prompt = string.Empty;
Expand Down
5 changes: 2 additions & 3 deletions src/Plugins/BotSharp.Plugin.HttpHandler/Functions/HandleHttpRequestFn.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,4 @@
using System.Net.Http;
using BotSharp.Plugin.HttpHandler.LlmContexts;
using Microsoft.AspNetCore.Http;
using Microsoft.Extensions.Logging;

Expand Down Expand Up @@ -58,7 +57,7 @@ public async Task<bool> Execute(RoleDialogModel message)
{
if (string.IsNullOrEmpty(url)) return null;

var settings = _services.GetRequiredService<HttpSettings>();
var settings = _services.GetRequiredService<HttpHandlerSettings>();
using var client = _httpClientFactory.CreateClient();
AddRequestHeaders(client);

Expand All @@ -82,7 +81,7 @@ private void AddRequestHeaders(HttpClient client)
{
client.DefaultRequestHeaders.Add("Authorization", $"{_context.HttpContext.Request.Headers["Authorization"]}");

var settings = _services.GetRequiredService<HttpSettings>();
var settings = _services.GetRequiredService<HttpHandlerSettings>();
var origin = !string.IsNullOrEmpty(settings.Origin) ? settings.Origin : $"{_context.HttpContext.Request.Headers["Origin"]}";
if (!string.IsNullOrEmpty(origin))
{
Expand Down
1 change: 0 additions & 1 deletion src/Plugins/BotSharp.Plugin.HttpHandler/Hooks/HttpHandlerHook.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,6 @@
using BotSharp.Abstraction.Agents.Settings;
using BotSharp.Abstraction.Functions.Models;
using BotSharp.Abstraction.Repositories;
using BotSharp.Plugin.HttpHandler.Enums;

namespace BotSharp.Plugin.HttpHandler.Hooks;

Expand Down
1 change: 0 additions & 1 deletion src/Plugins/BotSharp.Plugin.HttpHandler/Hooks/HttpHandlerUtilityHook.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,4 @@
using BotSharp.Abstraction.Agents;
using BotSharp.Plugin.HttpHandler.Enums;

namespace BotSharp.Plugin.HttpHandler.Hooks;

Expand Down
3 changes: 1 addition & 2 deletions src/Plugins/BotSharp.Plugin.HttpHandler/HttpHandlerPlugin.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,5 @@
using BotSharp.Abstraction.Agents;
using BotSharp.Abstraction.Settings;
using BotSharp.Plugin.HttpHandler.Hooks;
using Microsoft.Extensions.Configuration;

namespace BotSharp.Plugin.HttpHandler;
Expand All @@ -18,7 +17,7 @@ public void RegisterDI(IServiceCollection services, IConfiguration config)
services.AddScoped(provider =>
{
var settingService = provider.GetRequiredService<ISettingService>();
return settingService.Bind<HttpSettings>("Http");
return settingService.Bind<HttpHandlerSettings>("HttpHandler");
});

services.AddScoped<IAgentHook, HttpHandlerHook>();
Expand Down
4 changes: 2 additions & 2 deletions ...Abstraction/Http/Settings/HttpSettings.cs → ...tpHandler/Settings/HttpHandlerSettings.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
namespace BotSharp.Abstraction.Http.Settings;
namespace BotSharp.Plugin.HttpHandler.Settings;

public class HttpSettings
public class HttpHandlerSettings
{
public string BaseAddress { get; set; } = string.Empty;
public string Origin { get; set; } = string.Empty;
Expand Down
7 changes: 5 additions & 2 deletions src/Plugins/BotSharp.Plugin.HttpHandler/Using.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,5 +15,8 @@
global using BotSharp.Abstraction.Messaging;
global using BotSharp.Abstraction.Messaging.Models.RichContent;
global using BotSharp.Abstraction.Options;
global using BotSharp.Abstraction.Http.Settings;
global using BotSharp.Abstraction.Messaging.Enums;
global using BotSharp.Abstraction.Messaging.Enums;
global using BotSharp.Plugin.HttpHandler.Hooks;
global using BotSharp.Plugin.HttpHandler.Settings;
global using BotSharp.Plugin.HttpHandler.LlmContexts;
global using BotSharp.Plugin.HttpHandler.Enums;
48 changes: 21 additions & 27 deletions src/Plugins/BotSharp.Plugin.OpenAI/Providers/Chat/ChatCompletionProvider.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -257,40 +257,34 @@ public async Task<bool> GetChatCompletionsStreamingAsync(Agent agent, List<RoleD
{
var text = !string.IsNullOrWhiteSpace(message.Payload) ? message.Payload : message.Content;
var textPart = ChatMessageContentPart.CreateTextMessageContentPart(text);
var chat = new UserChatMessage(textPart)
{
ParticipantName = message.FunctionName
};
var contentParts = new List<ChatMessageContentPart> { textPart };

if (allowMultiModal)
if (allowMultiModal && !message.Files.IsNullOrEmpty())
{
if (!message.Files.IsNullOrEmpty())
foreach (var file in message.Files)
{
foreach (var file in message.Files)
if (!string.IsNullOrEmpty(file.FileUrl))
{
var uri = new Uri(file.FileUrl);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(uri, ImageChatMessageContentPartDetail.Low);
contentParts.Add(contentPart);
}
else if (!string.IsNullOrEmpty(file.FileData))
{
var (contentType, bytes) = fileService.GetFileInfoFromData(file.FileData);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(BinaryData.FromBytes(bytes), contentType, ImageChatMessageContentPartDetail.Low);
contentParts.Add(contentPart);
}
else if (!string.IsNullOrEmpty(file.FileStorageUrl))
{
if (!string.IsNullOrEmpty(file.FileUrl))
{
var uri = new Uri(file.FileUrl);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(uri, ImageChatMessageContentPartDetail.Low);
chat = new UserChatMessage(textPart, contentPart) { ParticipantName = message.FunctionName };
}
else if (!string.IsNullOrEmpty(file.FileData))
{
var (contentType, bytes) = fileService.GetFileInfoFromData(file.FileData);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(BinaryData.FromBytes(bytes), contentType, ImageChatMessageContentPartDetail.Low);
chat = new UserChatMessage(textPart, contentPart) { ParticipantName = message.FunctionName };
}
else if (!string.IsNullOrEmpty(file.FileStorageUrl))
{
var contentType = fileService.GetFileContentType(file.FileStorageUrl);
using var stream = File.OpenRead(file.FileStorageUrl);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(BinaryData.FromStream(stream), contentType, ImageChatMessageContentPartDetail.Low);
chat = new UserChatMessage(textPart, contentPart) { ParticipantName = message.FunctionName };
}
var contentType = fileService.GetFileContentType(file.FileStorageUrl);
using var stream = File.OpenRead(file.FileStorageUrl);
var contentPart = ChatMessageContentPart.CreateImageMessageContentPart(BinaryData.FromStream(stream), contentType, ImageChatMessageContentPartDetail.Low);
contentParts.Add(contentPart);
}
}
}
messages.Add(chat);
messages.Add(new UserChatMessage(contentParts) { ParticipantName = message.FunctionName });
}
else if (message.Role == AgentRole.Assistant)
{
Expand Down
2 changes: 1 addition & 1 deletion src/WebStarter/appsettings.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -154,7 +154,7 @@
"Driver": "Playwright"
},

"Http": {
"HttpHandler": {
"BaseAddress": "",
"Origin": ""
},
Expand Down