logdetective 0.2.5__tar.gz → 0.2.6__tar.gz

{logdetective-0.2.5 → logdetective-0.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: logdetective
-Version: 0.2.5
+Version: 0.2.6
 Summary: Log using LLM AI to search for build/test failures and provide ideas for fixing these.
 License: Apache-2.0
 Author: Jiri Podivin
@@ -83,6 +83,10 @@ Example you want to use a different model:
     logdetective https://example.com/logs.txt --model https://huggingface.co/QuantFactory/Meta-Llama-3-8B-Instruct-GGUF/resolve/main/Meta-Llama-3-8B-Instruct.Q5_K_S.gguf?download=true
     logdetective https://example.com/logs.txt --model QuantFactory/Meta-Llama-3-8B-Instruct-GGUF
 
+Note that streaming with some models (notably Meta-Llama-3 is broken) is broken and can be workarounded by `no-stream` option:
+
+    logdetective https://example.com/logs.txt --model QuantFactory/Meta-Llama-3-8B-Instruct-GGUF --no-stream
+
 
 Real Example
 ------------
@@ -181,7 +185,7 @@ or
 Server
 ------
 
-FastApi based server is implemented in `logdetective/server.py`. In order to run in a development mode,
+FastApi based server is implemented in `logdetective/server.py`. In order to run it in a development mode,
 simply start llama-cpp-python server with your chosen model as described in llama-cpp-python [docs](https://llama-cpp-python.readthedocs.io/en/latest/server/#running-the-server).
 
 Afterwards, start the logdetective server with `fastapi dev logdetective/server.py --port 8080`.
@@ -189,6 +193,17 @@ Requests can then be made with post requests, for example:
 
     curl --header "Content-Type: application/json" --request POST --data '{"url":"<YOUR_URL_HERE>"}' http://localhost:8080/analyze
 
+We also have a Containerfile and composefile to run the logdetective server and llama server in containers.
+
+Before doing `podman-compose up`, make sure to set `MODELS_PATH` environment variable and point to a directory with your local model files:
+```
+$ export MODELS_PATH=/path/to/models/
+$ ll $MODELS_PATH
+-rw-r--r--. 1 tt tt 3.9G apr 10 17:18  mistral-7b-instruct-v0.2.Q4_K_S.gguf
+```
+
+If the variable is not set, `./models` is mounted inside by default.
+
 
 License
 -------

{logdetective-0.2.5 → logdetective-0.2.6}/README.md

@@ -55,6 +55,10 @@ Example you want to use a different model:
     logdetective https://example.com/logs.txt --model https://huggingface.co/QuantFactory/Meta-Llama-3-8B-Instruct-GGUF/resolve/main/Meta-Llama-3-8B-Instruct.Q5_K_S.gguf?download=true
     logdetective https://example.com/logs.txt --model QuantFactory/Meta-Llama-3-8B-Instruct-GGUF
 
+Note that streaming with some models (notably Meta-Llama-3 is broken) is broken and can be workarounded by `no-stream` option:
+
+    logdetective https://example.com/logs.txt --model QuantFactory/Meta-Llama-3-8B-Instruct-GGUF --no-stream
+
 
 Real Example
 ------------
@@ -153,7 +157,7 @@ or
 Server
 ------
 
-FastApi based server is implemented in `logdetective/server.py`. In order to run in a development mode,
+FastApi based server is implemented in `logdetective/server.py`. In order to run it in a development mode,
 simply start llama-cpp-python server with your chosen model as described in llama-cpp-python [docs](https://llama-cpp-python.readthedocs.io/en/latest/server/#running-the-server).
 
 Afterwards, start the logdetective server with `fastapi dev logdetective/server.py --port 8080`.
@@ -161,6 +165,17 @@ Requests can then be made with post requests, for example:
 
     curl --header "Content-Type: application/json" --request POST --data '{"url":"<YOUR_URL_HERE>"}' http://localhost:8080/analyze
 
+We also have a Containerfile and composefile to run the logdetective server and llama server in containers.
+
+Before doing `podman-compose up`, make sure to set `MODELS_PATH` environment variable and point to a directory with your local model files:
+```
+$ export MODELS_PATH=/path/to/models/
+$ ll $MODELS_PATH
+-rw-r--r--. 1 tt tt 3.9G apr 10 17:18  mistral-7b-instruct-v0.2.Q4_K_S.gguf
+```
+
+If the variable is not set, `./models` is mounted inside by default.
+
 
 License
 -------

{logdetective-0.2.5 → logdetective-0.2.6}/logdetective/constants.py

@@ -1,6 +1,6 @@
 
 # pylint: disable=line-too-long
-DEFAULT_ADVISOR = "TheBloke/Mistral-7B-Instruct-v0.2-GGUF"
+DEFAULT_ADVISOR = "fedora-copr/Mistral-7B-Instruct-v0.2-GGUF"
 
 PROMPT_TEMPLATE = """
 Given following log snippets, and nothing else, explain what failure, if any, occured during build of this package.

{logdetective-0.2.5 → logdetective-0.2.6}/logdetective/logdetective.py

@@ -8,9 +8,8 @@ from logdetective.extractors import LLMExtractor, DrainExtractor
 
 LOG = logging.getLogger("logdetective")
 
-
-def main():
-    """Main execution function."""
+def setup_args():
+    """ Setup argument parser and return arguments. """
     parser = argparse.ArgumentParser("logdetective")
     parser.add_argument("file", type=str,
                         default="", help="The URL or path to the log file to be analyzed.")
@@ -21,6 +20,7 @@ def main():
                         help="Suffix of the model file name to be retrieved from Hugging Face.\
                             Makes sense only if the model is specified with Hugging Face name.",
                         default="Q4_K_S.gguf")
+    parser.add_argument("-n", "--no-stream", action='store_true')
     parser.add_argument("-S", "--summarizer", type=str, default="drain",
                         help="Choose between LLM and Drain template miner as the log summarizer.\
                                 LLM must be specified as path to a model, URL or local file.")
@@ -32,7 +32,12 @@ def main():
                             This only makes sense when you are summarizing with Drain")
     parser.add_argument("-v", "--verbose", action='count', default=0)
     parser.add_argument("-q", "--quiet", action='store_true')
-    args = parser.parse_args()
+    return parser.parse_args()
+
+
+def main():
+    """Main execution function."""
+    args = setup_args()
 
     if args.verbose and args.quiet:
         sys.stderr.write("Error: --quiet and --verbose is mutually exclusive.\n")
@@ -83,7 +88,19 @@ def main():
     log_summary = format_snippets(log_summary)
     LOG.info("Log summary: \n %s", log_summary)
 
-    print(f"Explanation: \n{process_log(log_summary, model)}")
+    stream = True
+    if args.no_stream:
+        stream = False
+    response = process_log(log_summary, model, stream)
+    print("Explanation:")
+    if args.no_stream:
+        print(response["choices"][0]["text"])
+    else:
+        # Stream the output
+        for chunk in response:
+            delta = chunk['choices'][0]['text']
+            print(delta, end='', flush=True)
+        print()
 
 
 if __name__ == "__main__":

logdetective-0.2.6/logdetective/server.py

@@ -0,0 +1,87 @@
+import logging
+import os
+import json
+
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+
+import requests
+
+from logdetective.constants import PROMPT_TEMPLATE
+from logdetective.extractors import DrainExtractor
+from logdetective.utils import validate_url
+
+class BuildLog(BaseModel):
+    """Model of data submitted to API.
+    """
+    url: str
+
+LOG = logging.getLogger("logdetective")
+
+app = FastAPI()
+
+LLM_CPP_HOST = os.environ.get("LLAMA_CPP_HOST", "localhost")
+LLM_CPP_SERVER_ADDRESS = f"http://{LLM_CPP_HOST}"
+LLM_CPP_SERVER_PORT = os.environ.get("LLAMA_CPP_SERVER_PORT", 8000)
+LLM_CPP_SERVER_TIMEOUT = os.environ.get("LLAMA_CPP_SERVER_TIMEOUT", 600)
+LOG_SOURCE_REQUEST_TIMEOUT = os.environ.get("LOG_SOURCE_REQUEST_TIMEOUT", 60)
+
+@app.post("/analyze", )
+async def analyze_log(build_log: BuildLog):
+    """Provide endpoint for log file submission and analysis.
+    Request must be in form {"url":"<YOUR_URL_HERE>"}.
+    URL must be valid for the request to be passed to the LLM server.
+    Meaning that it must contain appropriate scheme, path and netloc,
+    while lacking  result, params or query fields.
+    """
+    extractor = DrainExtractor(verbose=True, context=True, max_clusters=8)
+
+    LOG.info("Getting summary")
+    # Perform basic validation of the URL
+    if validate_url(url=build_log.url):
+        try:
+            log_request = requests.get(build_log.url, timeout=int(LOG_SOURCE_REQUEST_TIMEOUT))
+        except requests.RequestException as ex:
+            raise HTTPException(
+                status_code=400,
+                detail=f"We couldn't obtain the logs: {ex}") from ex
+
+        if not log_request.ok:
+            raise HTTPException(status_code=400,
+                                detail="Something went wrong while getting the logs: "
+                                    f"[{log_request.status_code}] {log_request.text}")
+    else:
+        LOG.error("Invalid URL received ")
+        raise HTTPException(status_code=400,
+                            detail=f"Invalid log URL: {build_log.url}")
+
+    log = log_request.text
+    log_summary = extractor(log)
+
+    ratio = len(log_summary) / len(log.split('\n'))
+    LOG.debug("Log summary: \n %s", log_summary)
+    LOG.info("Compression ratio: %s", ratio)
+
+    LOG.info("Analyzing the text")
+    data = {
+            "prompt": PROMPT_TEMPLATE.format(log_summary),
+            "max_tokens": "0"}
+
+    try:
+        # Expects llama-cpp server to run on LLM_CPP_SERVER_ADDRESS:LLM_CPP_SERVER_PORT
+        response = requests.post(
+            f"{LLM_CPP_SERVER_ADDRESS}:{LLM_CPP_SERVER_PORT}/v1/completions",
+            headers={"Content-Type":"application/json"},
+            data=json.dumps(data),
+            timeout=int(LLM_CPP_SERVER_TIMEOUT))
+    except requests.RequestException as ex:
+        raise HTTPException(
+            status_code=400,
+            detail=f"Llama-cpp query failed: {ex}") from ex
+
+    if not log_request.ok:
+        raise HTTPException(
+            status_code=400,
+            detail="Something went wrong while getting a response from the llama server: "
+                   f"[{log_request.status_code}] {log_request.text}")
+    return response.text

{logdetective-0.2.5 → logdetective-0.2.6}/logdetective/utils.py

@@ -53,6 +53,9 @@ def initialize_model(model_pth: str, filename_suffix: str = ".gguf", verbose: bo
         filename_suffix (str): suffix of the model file name to be pulled from Hugging Face
         verbose (bool): level of verbosity for llamacpp
     """
+
+    LOG.info("Loading model from %s", model_pth)
+
     if os.path.isfile(model_pth):
         model = Llama(
             model_path=model_pth,
@@ -68,7 +71,7 @@ def initialize_model(model_pth: str, filename_suffix: str = ".gguf", verbose: bo
     return model
 
 
-def process_log(log: str, model: Llama) -> str:
+def process_log(log: str, model: Llama, stream: bool) -> str:
     """
     Processes a given log using the provided language model and returns its summary.
 
@@ -79,11 +82,16 @@ def process_log(log: str, model: Llama) -> str:
     Returns:
         str: The summary of the given log generated by the language model.
     """
-    return model(PROMPT_TEMPLATE.format(log), max_tokens=0)["choices"][0]["text"]
+    return model(
+        prompt=PROMPT_TEMPLATE.format(log),
+        stream=stream,
+        max_tokens=0)
 
 
 def retrieve_log_content(log_path: str) -> str:
-    """Get content of the file on the log_path path."""
+    """Get content of the file on the log_path path.
+    Path is assumed to be valid URL if it has a scheme.
+    Otherwise it attempts to pull it from local filesystem."""
     parsed_url = urlparse(log_path)
     log = ""
 
@@ -113,3 +121,18 @@ def format_snippets(snippets: list[str]) -> str:
         ================
         """
     return summary
+
+
+def validate_url(url: str) -> bool:
+    """Validate incoming URL to be at least somewhat sensible for log files
+    Only http and https protocols permitted. No result, params or query fields allowed.
+    Either netloc or path must have non-zero length.
+    """
+    result = urlparse(url)
+    if result.scheme not in ['http', 'https']:
+        return False
+    if any([result.params, result.query, result.fragment]):
+        return False
+    if not (result.path or result.netloc):
+        return False
+    return True

{logdetective-0.2.5 → logdetective-0.2.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "logdetective"
-version = "0.2.5"
+version = "0.2.6"
 description = "Log using LLM AI to search for build/test failures and provide ideas for fixing these."
 authors = ["Jiri Podivin <jpodivin@gmail.com>"]
 license = "Apache-2.0"

logdetective-0.2.5/logdetective/server.py

@@ -1,56 +0,0 @@
-import logging
-import os
-import json
-
-from fastapi import FastAPI
-from pydantic import BaseModel
-
-import requests
-
-from logdetective.constants import PROMPT_TEMPLATE
-from logdetective.extractors import DrainExtractor
-
-
-class BuildLog(BaseModel):
-    """Model of data submitted to API.
-    """
-    url: str
-
-LOG = logging.getLogger("logdetective")
-
-app = FastAPI()
-
-LLM_CPP_SERVER_ADDRESS = os.environ.get("LLAMA_CPP_SERVER", " http://localhost")
-LLM_CPP_SERVER_PORT = os.environ.get("LLAMA_CPP_SERVER_PORT", 8000)
-LLM_CPP_SERVER_TIMEOUT = os.environ.get("LLAMA_CPP_SERVER_TIMEOUT", 600)
-LOG_SOURCE_REQUEST_TIMEOUT = os.environ.get("LOG_SOURCE_REQUEST_TIMEOUT", 60)
-
-@app.post("/analyze", )
-async def analyze_log(build_log: BuildLog):
-    """Provide endpoint for log file submission and analysis.
-    Request must be in form {"url":"<YOUR_URL_HERE>"}.
-    """
-    extractor = DrainExtractor(verbose=True, context=True, max_clusters=8)
-
-    LOG.info("Getting summary")
-
-    log = requests.get(build_log.url, timeout=int(LOG_SOURCE_REQUEST_TIMEOUT)).text
-    log_summary = extractor(log)
-
-    ratio = len(log_summary) / len(log.split('\n'))
-    LOG.debug("Log summary: \n %s", log_summary)
-    LOG.info("Compression ratio: %s", ratio)
-
-    LOG.info("Analyzing the text")
-    data = {
-            "prompt": PROMPT_TEMPLATE.format(log_summary),
-            "max_tokens": "0"}
-
-    # Expects llama-cpp server to run on LLM_CPP_SERVER_ADDRESS:LLM_CPP_SERVER_PORT
-    response = requests.post(
-        f"{LLM_CPP_SERVER_ADDRESS}:{LLM_CPP_SERVER_PORT}/v1/completions",
-        headers={"Content-Type":"application/json"},
-        data=json.dumps(data),
-        timeout=int(LLM_CPP_SERVER_TIMEOUT))
-
-    return response.text