Ver código fonte

feat: add gemini

Byaidu 1 ano atrás
pai
commit
3a23b4fbef
6 arquivos alterados com 37 adições e 13 exclusões
  1. 1 0
      README.md
  2. 13 12
      README_ja-JP.md
  3. 1 0
      README_zh-CN.md
  4. 4 1
      pdf2zh/converter.py
  5. 2 0
      pdf2zh/gui.py
  6. 16 0
      pdf2zh/translator.py

+ 1 - 0
README.md

@@ -201,6 +201,7 @@ The table below outlines the required [environment variables](https://chatgpt.co
 |**AzureOpenAI**|`azure-openai`|`AZURE_OPENAI_BASE_URL`, `AZURE_OPENAI_API_KEY`, `AZURE_OPENAI_MODEL`|`[Your Endpoint]`, `[Your Key]`, `gpt-4o-mini`|See [Azure OpenAI](https://learn.microsoft.com/zh-cn/azure/ai-services/openai/chatgpt-quickstart?tabs=command-line%2Cjavascript-keyless%2Ctypescript-keyless%2Cpython&pivots=programming-language-python)|
 |**Zhipu**|`zhipu`|`ZHIPU_API_KEY`, `ZHIPU_MODEL`|`[Your Key]`, `glm-4-flash`|See [Zhipu](https://open.bigmodel.cn/dev/api/thirdparty-frame/openai-sdk)|
 |**Silicon**|`silicon`|`SILICON_API_KEY`, `SILICON_MODEL`|`[Your Key]`, `Qwen/Qwen2.5-7B-Instruct`|See [SiliconCloud](https://docs.siliconflow.cn/quickstart)|
+|**Gemini**|`gemini`|`GEMINI_API_KEY`, `GEMINI_MODEL`|`[Your Key]`, `gemini-1.5-flash`|See [Gemini](https://ai.google.dev/gemini-api/docs/openai)|
 |**Azure**|`azure`|`AZURE_ENDPOINT`, `AZURE_API_KEY`|`https://api.translator.azure.cn`, `[Your Key]`|See [Azure](https://docs.azure.cn/en-us/ai-services/translator/text-translation-overview)|
 |**Tencent**|`tencent`|`TENCENTCLOUD_SECRET_ID`, `TENCENTCLOUD_SECRET_KEY`|`[Your ID]`, `[Your Key]`|See [Tencent](https://www.tencentcloud.com/products/tmt?from_qcintl=122110104)|
 

+ 13 - 12
README_ja-JP.md

@@ -190,19 +190,20 @@ pdf2zh example.pdf -li en -lo ja
 
 以下の表は、各翻訳サービスに必要な [環境変数](https://chatgpt.com/share/6734a83d-9d48-800e-8a46-f57ca6e8bcb4) を示しています。各サービスを使用する前に、これらの変数を設定してください。
 
-|**翻訳者**|**サービス**|**環境変数**|**デフォルト値**|**備考**|
+|**Translator**|**Service**|**Environment Variables**|**Default Values**|**Notes**|
 |-|-|-|-|-|
-|**Google (デフォルト)**|`google`|なし|N/A|なし|
-|**Bing**|`bing`|なし|N/A|なし|
-|**DeepL**|`deepl`|`DEEPL_SERVER_URL`,`DEEPL_AUTH_KEY`|`https://api.deepl.com`, `[Your Key]`|[DeepL](https://support.deepl.com/hc/en-us/articles/360020695820-API-Key-for-DeepL-s-API) を参照|
-|**DeepLX**|`deeplx`|`DEEPLX_ENDPOINT`|`https://api.deepl.com/translate`|[DeepLX](https://github.com/OwO-Network/DeepLX) を参照|
-|**Ollama**|`ollama`|`OLLAMA_HOST`, `OLLAMA_MODEL`|`http://127.0.0.1:11434`, `gemma2`|[Ollama](https://github.com/ollama/ollama) を参照|
-|**OpenAI**|`openai`|`OPENAI_BASE_URL`, `OPENAI_API_KEY`, `OPENAI_MODEL`|`https://api.openai.com/v1`, `[Your Key]`, `gpt-4o-mini`|[OpenAI](https://platform.openai.com/docs/overview) を参照|
-|**AzureOpenAI**|`azure-openai`|`AZURE_OPENAI_BASE_URL`, `AZURE_OPENAI_API_KEY`, `AZURE_OPENAI_MODEL`|`[Your Endpoint]`, `[Your Key]`, `gpt-4o-mini`|[Azure OpenAI](https://learn.microsoft.com/zh-cn/azure/ai-services/openai/chatgpt-quickstart?tabs=command-line%2Cjavascript-keyless%2Ctypescript-keyless%2Cpython&pivots=programming-language-python) を参照|
-|**Zhipu**|`zhipu`|`ZHIPU_API_KEY`, `ZHIPU_MODEL`|`[Your Key]`, `glm-4-flash`|[Zhipu](https://open.bigmodel.cn/dev/api/thirdparty-frame/openai-sdk) を参照|
-|**Silicon**|`silicon`|`SILICON_API_KEY`, `SILICON_MODEL`|`[Your Key]`, `Qwen/Qwen2.5-7B-Instruct`|[SiliconCloud](https://docs.siliconflow.cn/quickstart) を参照|
-|**Azure**|`azure`|`AZURE_ENDPOINT`, `AZURE_API_KEY`|`https://api.translator.azure.cn`, `[Your Key]`|[Azure](https://docs.azure.cn/en-us/ai-services/translator/text-translation-overview) を参照|
-|**Tencent**|`tencent`|`TENCENTCLOUD_SECRET_ID`, `TENCENTCLOUD_SECRET_KEY`|`[Your ID]`, `[Your Key]`|[Tencent](https://www.tencentcloud.com/products/tmt?from_qcintl=122110104) を参照|
+|**Google (Default)**|`google`|None|N/A|None|
+|**Bing**|`bing`|None|N/A|None|
+|**DeepL**|`deepl`|`DEEPL_SERVER_URL`,`DEEPL_AUTH_KEY`|`https://api.deepl.com`, `[Your Key]`|See [DeepL](https://support.deepl.com/hc/en-us/articles/360020695820-API-Key-for-DeepL-s-API)|
+|**DeepLX**|`deeplx`|`DEEPLX_ENDPOINT`|`https://api.deepl.com/translate`|See [DeepLX](https://github.com/OwO-Network/DeepLX)|
+|**Ollama**|`ollama`|`OLLAMA_HOST`, `OLLAMA_MODEL`|`http://127.0.0.1:11434`, `gemma2`|See [Ollama](https://github.com/ollama/ollama)|
+|**OpenAI**|`openai`|`OPENAI_BASE_URL`, `OPENAI_API_KEY`, `OPENAI_MODEL`|`https://api.openai.com/v1`, `[Your Key]`, `gpt-4o-mini`|See [OpenAI](https://platform.openai.com/docs/overview)|
+|**AzureOpenAI**|`azure-openai`|`AZURE_OPENAI_BASE_URL`, `AZURE_OPENAI_API_KEY`, `AZURE_OPENAI_MODEL`|`[Your Endpoint]`, `[Your Key]`, `gpt-4o-mini`|See [Azure OpenAI](https://learn.microsoft.com/zh-cn/azure/ai-services/openai/chatgpt-quickstart?tabs=command-line%2Cjavascript-keyless%2Ctypescript-keyless%2Cpython&pivots=programming-language-python)|
+|**Zhipu**|`zhipu`|`ZHIPU_API_KEY`, `ZHIPU_MODEL`|`[Your Key]`, `glm-4-flash`|See [Zhipu](https://open.bigmodel.cn/dev/api/thirdparty-frame/openai-sdk)|
+|**Silicon**|`silicon`|`SILICON_API_KEY`, `SILICON_MODEL`|`[Your Key]`, `Qwen/Qwen2.5-7B-Instruct`|See [SiliconCloud](https://docs.siliconflow.cn/quickstart)|
+|**Gemini**|`gemini`|`GEMINI_API_KEY`, `GEMINI_MODEL`|`[Your Key]`, `gemini-1.5-flash`|See [Gemini](https://ai.google.dev/gemini-api/docs/openai)|
+|**Azure**|`azure`|`AZURE_ENDPOINT`, `AZURE_API_KEY`|`https://api.translator.azure.cn`, `[Your Key]`|See [Azure](https://docs.azure.cn/en-us/ai-services/translator/text-translation-overview)|
+|**Tencent**|`tencent`|`TENCENTCLOUD_SECRET_ID`, `TENCENTCLOUD_SECRET_KEY`|`[Your ID]`, `[Your Key]`|See [Tencent](https://www.tencentcloud.com/products/tmt?from_qcintl=122110104)|
 
 `-s service` または `-s service:model` を使用してサービスを指定します:
 

+ 1 - 0
README_zh-CN.md

@@ -201,6 +201,7 @@ pdf2zh example.pdf -li en -lo ja
 |**AzureOpenAI**|`azure-openai`|`AZURE_OPENAI_BASE_URL`, `AZURE_OPENAI_API_KEY`, `AZURE_OPENAI_MODEL`|`[Your Endpoint]`, `[Your Key]`, `gpt-4o-mini`|See [Azure OpenAI](https://learn.microsoft.com/zh-cn/azure/ai-services/openai/chatgpt-quickstart?tabs=command-line%2Cjavascript-keyless%2Ctypescript-keyless%2Cpython&pivots=programming-language-python)|
 |**Zhipu**|`zhipu`|`ZHIPU_API_KEY`, `ZHIPU_MODEL`|`[Your Key]`, `glm-4-flash`|See [Zhipu](https://open.bigmodel.cn/dev/api/thirdparty-frame/openai-sdk)|
 |**Silicon**|`silicon`|`SILICON_API_KEY`, `SILICON_MODEL`|`[Your Key]`, `Qwen/Qwen2.5-7B-Instruct`|See [SiliconCloud](https://docs.siliconflow.cn/quickstart)|
+|**Gemini**|`gemini`|`GEMINI_API_KEY`, `GEMINI_MODEL`|`[Your Key]`, `gemini-1.5-flash`|See [Gemini](https://ai.google.dev/gemini-api/docs/openai)|
 |**Azure**|`azure`|`AZURE_ENDPOINT`, `AZURE_API_KEY`|`https://api.translator.azure.cn`, `[Your Key]`|See [Azure](https://docs.azure.cn/en-us/ai-services/translator/text-translation-overview)|
 |**Tencent**|`tencent`|`TENCENTCLOUD_SECRET_ID`, `TENCENTCLOUD_SECRET_KEY`|`[Your ID]`, `[Your Key]`|See [Tencent](https://www.tencentcloud.com/products/tmt?from_qcintl=122110104)|
 

+ 4 - 1
pdf2zh/converter.py

@@ -27,6 +27,7 @@ from pdf2zh.translator import (
     OpenAITranslator,
     ZhipuTranslator,
     SiliconTranslator,
+    GeminiTranslator,
     AzureTranslator,
     TencentTranslator,
 )
@@ -142,7 +143,7 @@ class TranslateConverter(PDFConverterEx):
         service_name = param[0]
         service_model = param[1] if len(param) > 1 else None
         for translator in [GoogleTranslator, BingTranslator, DeepLTranslator, DeepLXTranslator, OllamaTranslator, AzureOpenAITranslator,
-                           OpenAITranslator, ZhipuTranslator, SiliconTranslator, AzureTranslator, TencentTranslator]:
+                           OpenAITranslator, ZhipuTranslator, SiliconTranslator, GeminiTranslator, AzureTranslator, TencentTranslator]:
             if service_name == translator.name:
                 self.translator = translator(lang_in, lang_out, service_model)
         if not self.translator:
@@ -170,6 +171,8 @@ class TranslateConverter(PDFConverterEx):
         ops: str = ""                   # 渲染结果
 
         def vflag(font: str, char: str):    # 匹配公式(和角标)字体
+            if isinstance(font, bytes):     # hack 嵌入的 china-ss 会变成 b'Song'
+                font = font.decode()
             font = font.split("+")[-1]      # 字体名截断
             if re.match(r"\(cid:", char):
                 return True

+ 2 - 0
pdf2zh/gui.py

@@ -14,6 +14,7 @@ from pdf2zh.translator import (
     OpenAITranslator,
     ZhipuTranslator,
     SiliconTranslator,
+    GeminiTranslator,
     AzureTranslator,
     TencentTranslator,
 )
@@ -34,6 +35,7 @@ service_map: dict[str, BaseTranslator] = {
     "OpenAI": OpenAITranslator,
     "Zhipu": ZhipuTranslator,
     "Silicon": SiliconTranslator,
+    "Gemini": GeminiTranslator,
     "Azure": AzureTranslator,
     "Tencent": TencentTranslator,
 }

+ 16 - 0
pdf2zh/translator.py

@@ -284,6 +284,22 @@ class SiliconTranslator(OpenAITranslator):
         super().__init__(lang_in, lang_out, model, base_url=base_url, api_key=api_key)
 
 
+class GeminiTranslator(OpenAITranslator):
+    # https://ai.google.dev/gemini-api/docs/openai
+    name = "gemini"
+    envs = {
+        "GEMINI_API_KEY": None,
+        "GEMINI_MODEL": "gemini-1.5-flash",
+    }
+
+    def __init__(self, lang_in, lang_out, model):
+        base_url = "https://generativelanguage.googleapis.com/v1beta/openai/"
+        api_key = os.getenv("GEMINI_API_KEY")
+        if not model:
+            model = os.getenv("GEMINI_MODEL", self.envs["GEMINI_MODEL"])
+        super().__init__(lang_in, lang_out, model, base_url=base_url, api_key=api_key)
+
+
 class AzureTranslator(BaseTranslator):
     # https://github.com/Azure/azure-sdk-for-python
     name = "azure"