running in docker

Dockerfile

@@ -1,4 +1,8 @@
-FROM python
+FROM python:3.12.0
+# RUN apt-get update && apt-get install -y \
+#     sqlite3 \
+#     libsqlite3-dev \
+#     && rm -rf /var/lib/apt/lists/*
 COPY requirements.txt /app/requirements.txt
 WORKDIR /app
 RUN pip install -r requirements.txt
@@ -7,8 +11,8 @@ RUN useradd -m -u 1000 user
 USER user
 ENV HOME=/home/user \
     PATH=/home/user/.local/bin:$PATH
-
-WORKDIR $HOME/app
+    
+WORKDIR /app
 RUN playwright install
 
 COPY --chown=user . $HOME/app
@@ -17,5 +21,10 @@ RUN scrapy crawl pages && \
     cd $HOME/app && \
     python setup.py
 WORKDIR $HOME/app
+RUN curl https://cdn-media.huggingface.co/frpc-gradio-0.2/frpc_linux_amd64 -o frpc_linux_amd64_v0.2
+USER root
+RUN mv frpc_linux_amd64_v0.2 /usr/local/lib/python3.12/site-packages/gradio/
+RUN chmod +x /usr/local/lib/python3.12/site-packages/gradio/frpc_linux_amd64_v0.2
 EXPOSE 7860
-CMD ["python", "main.py"]
+USER user
+CMD ["gradio", "main.py"]

main.py

@@ -67,7 +67,8 @@ Answer:
         )
         self.db = Chroma(persist_directory="./pragetx_chroma", embedding_function=HuggingFaceEmbeddings(), collection_name="pragetx")  
         # self.llm = ChatOllama(model="phi3:3.8b", base_url="http://localhost:11434", num_gpu=16)
-        self.llm = HuggingChat(email = os.getenv("HF_EMAIL") , psw = os.getenv("HF_PASS") )
+        # self.llm = HuggingChat(email = os.getenv("HF_EMAIL") , psw = os.getenv("HF_PASS") )
+        self.llm = HuggingChat(email = "brij1808" , psw = "Brijesh321@R" )
         self.chain = (
             {"chat_history": self.chat_history, "context": self.db.as_retriever(search_kwargs={"k":3}), "question": RunnablePassthrough()} |
             self.prompt |

pragetx_scraper/pragetx_scraper/spiders/pages.py

@@ -20,11 +20,11 @@ def md(html, **options):
 
 class PagesSpider(scrapy.Spider):
     name = "pages"
-    allowed_domains = ["pragetx.com"]
-    start_urls = ["https://pragetx.com"]
+    allowed_domains = ["aadee.app"]
+    start_urls = ["https://aadee.app"]
 
     def start_requests(self):
-        url = "https://pragetx.com"
+        url = "https://aadee.app"
         print("Starting request")
         print("URL", url)
         yield scrapy.Request(url, meta={'playwright': True})

requirements.txt

@@ -47,7 +47,7 @@ h11==0.14.0
 httpcore==1.0.5
 httptools==0.6.1
 httpx==0.27.0
-hugchat==0.4.6
+hugchat==0.4.10
 huggingface-hub==0.23.2
 humanfriendly==10.0
 hyperlink==21.0.0

setup.py

@@ -31,4 +31,5 @@ for file in files:
         # print(docs[idx].metadata)
         docs[idx].metadata['url'] = url
         print(docs[idx].metadata)
-    chroma.add_documents(docs)
+    if docs:
+        chroma.add_documents(docs)

utils.py

@@ -127,6 +127,7 @@ class HuggingChat(LLM):
             return str(resp)
             
         except Exception as e:
+            print('e: ', e)
             raise ValueError("ChatBot failed, please check your parameters. " + str(e))
 
     @property