logdetective 0.5.9__py3-none-any.whl → 0.5.11__py3-none-any.whl

logdetective/constants.py

@@ -16,6 +16,8 @@ Snippets are delimited with '================'.
 
 Finally, drawing on information from all snippets, provide complete explanation of the issue and recommend solution.
 
+Explanation of the issue, and recommended solution, should take handful of sentences.
+
 Snippets:
 
 {}
@@ -38,6 +40,8 @@ Answer:
 SNIPPET_PROMPT_TEMPLATE = """
 Analyse following RPM build log snippet. Describe contents accurately, without speculation or suggestions for resolution.
 
+Your analysis must be as concise as possible, while keeping relevant information intact.
+
 Snippet:
 
 {}
@@ -55,6 +59,8 @@ Snippets are delimited with '================'.
 
 Drawing on information from all snippets, provide complete explanation of the issue and recommend solution.
 
+Explanation of the issue, and recommended solution, should take handful of sentences.
+
 Snippets:
 
 {}
@@ -64,3 +70,5 @@ Analysis:
 """
 
 SNIPPET_DELIMITER = "================"
+
+DEFAULT_TEMPERATURE = 0.8

logdetective/logdetective.py

@@ -1,8 +1,9 @@
 import argparse
 import logging
 import sys
+import os
 
-from logdetective.constants import DEFAULT_ADVISOR
+from logdetective.constants import DEFAULT_ADVISOR, DEFAULT_TEMPERATURE
 from logdetective.utils import (
     process_log,
     initialize_model,
@@ -67,7 +68,16 @@ def setup_args():
     parser.add_argument("-v", "--verbose", action="count", default=0)
     parser.add_argument("-q", "--quiet", action="store_true")
     parser.add_argument(
-        "--prompts", type=str, default="", help="Path to prompt configuration file."
+        "--prompts",
+        type=str,
+        default=f"{os.path.dirname(__file__)}/prompts.yml",
+        help="Path to prompt configuration file."
+    )
+    parser.add_argument(
+        "--temperature",
+        type=float,
+        default=DEFAULT_TEMPERATURE,
+        help="Temperature for inference."
     )
     return parser.parse_args()
 
@@ -143,6 +153,7 @@ def main():  # pylint: disable=too-many-statements,too-many-locals
         model,
         stream,
         prompt_template=prompts_configuration.prompt_template,
+        temperature=args.temperature,
     )
     probs = []
     print("Explanation:")

logdetective/prompts.yml

@@ -1,3 +1,8 @@
+# This file is intended for customization of prompts
+# It is used only in server mode.
+# On command line you have to load it using --prompts
+# The defaults are stored in constants.py
+
 prompt_template: |
   Given following log snippets, and nothing else, explain what failure, if any, occured during build of this package.
 
@@ -8,6 +13,8 @@ prompt_template: |
 
   Finally, drawing on information from all snippets, provide complete explanation of the issue and recommend solution.
 
+  Explanation of the issue, and recommended solution, should take handful of sentences.
+
   Snippets:
 
   {}
@@ -28,6 +35,8 @@ summarization_prompt_template: |
 snippet_prompt_template: |
   Analyse following RPM build log snippet. Describe contents accurately, without speculation or suggestions for resolution.
 
+  Your analysis must be as concise as possible, while keeping relevant information intact.
+
   Snippet:
 
   {}
@@ -43,6 +52,8 @@ prompt_template_staged: |
 
   Drawing on information from all snippets, provide complete explanation of the issue and recommend solution.
 
+  Explanation of the issue, and recommended solution, should take handful of sentences.
+
   Snippets:
 
   {}

logdetective/server/database/models.py

@@ -97,17 +97,35 @@ class AnalyzeRequestMetrics(Base):
             metrics.response_certainty = response_certainty
             session.add(metrics)
 
+    @classmethod
+    def get_postgres_time_format(cls, time_format):
+        """Map python time format in the PostgreSQL format."""
+        if time_format == "%Y-%m-%d":
+            pgsql_time_format = "YYYY-MM-DD"
+        else:
+            pgsql_time_format = "YYYY-MM-DD HH24"
+        return pgsql_time_format
+
+    @classmethod
+    def get_dictionary_with_datetime_keys(
+        cls, time_format: str, values_dict: dict[str, any]
+    ) -> dict[datetime.datetime, any]:
+        """Convert from a dictionary with str keys to a dictionary with datetime keys"""
+        new_dict = {
+            datetime.datetime.strptime(r[0], time_format): r[1] for r in values_dict
+        }
+        return new_dict
+
     @classmethod
     def _get_requests_by_time_for_postgres(
         cls, start_time, end_time, time_format, endpoint
     ):
-        """func.to_char is PostgreSQL specific.
+        """Get total requests number in time period.
+
+        func.to_char is PostgreSQL specific.
         Let's unit tests replace this function with the SQLite version.
         """
-        if time_format == "%Y-%m-%d":
-            pgsql_time_format = "YYYY-MM-DD"
-        else:
-            pgsql_time_format = "YYYY-MM-DD HH24"
+        pgsql_time_format = cls.get_postgres_time_format(time_format)
 
         requests_by_time_format = (
             select(
@@ -123,10 +141,12 @@ class AnalyzeRequestMetrics(Base):
         return requests_by_time_format
 
     @classmethod
-    def _get_requests_by_time_for_sqllite(
+    def _get_requests_by_time_for_sqlite(
         cls, start_time, end_time, time_format, endpoint
     ):
-        """func.strftime is SQLite specific.
+        """Get total requests number in time period.
+
+        func.strftime is SQLite specific.
         Use this function in unit test using flexmock:
 
         flexmock(AnalyzeRequestMetrics).should_receive("_get_requests_by_time_for_postgres")
@@ -178,9 +198,193 @@ class AnalyzeRequestMetrics(Base):
             counts = session.execute(count_requests_by_time_format)
             results = counts.fetchall()
 
-            # Convert results to a dictionary with proper datetime keys
-            counts_dict = {
-                datetime.datetime.strptime(r[0], time_format): r[1] for r in results
-            }
+            return cls.get_dictionary_with_datetime_keys(time_format, results)
+
+    @classmethod
+    def _get_average_responses_times_for_postgres(
+        cls, start_time, end_time, time_format, endpoint
+    ):
+        """Get average responses time.
+
+        func.to_char is PostgreSQL specific.
+        Let's unit tests replace this function with the SQLite version.
+        """
+        with transaction(commit=False) as session:
+            pgsql_time_format = cls.get_postgres_time_format(time_format)
+
+            average_responses_times = (
+                select(
+                    func.to_char(cls.request_received_at, pgsql_time_format).label(
+                        "time_range"
+                    ),
+                    (
+                        func.avg(
+                            func.extract(  # pylint: disable=not-callable
+                                "epoch", cls.response_sent_at - cls.request_received_at
+                            )
+                        )
+                    ).label("average_response_seconds"),
+                )
+                .filter(cls.request_received_at.between(start_time, end_time))
+                .filter(cls.endpoint == endpoint)
+                .group_by("time_range")
+                .order_by("time_range")
+            )
+
+            results = session.execute(average_responses_times).fetchall()
+            return results
+
+    @classmethod
+    def _get_average_responses_times_for_sqlite(
+        cls, start_time, end_time, time_format, endpoint
+    ):
+        """Get average responses time.
+
+        func.strftime is SQLite specific.
+        Use this function in unit test using flexmock:
+
+        flexmock(AnalyzeRequestMetrics).should_receive("_get_average_responses_times_for_postgres")
+        .replace_with(AnalyzeRequestMetrics._get_average_responses_times_for_sqlite)
+        """
+        with transaction(commit=False) as session:
+            average_responses_times = (
+                select(
+                    func.strftime(time_format, cls.request_received_at).label(
+                        "time_range"
+                    ),
+                    (
+                        func.avg(
+                            func.julianday(cls.response_sent_at)
+                            - func.julianday(cls.request_received_at)  # noqa: W503 flake8 vs ruff
+                        )
+                        * 86400  # noqa: W503 flake8 vs ruff
+                    ).label("average_response_seconds"),
+                )
+                .filter(cls.request_received_at.between(start_time, end_time))
+                .filter(cls.endpoint == endpoint)
+                .group_by("time_range")
+                .order_by("time_range")
+            )
+
+            results = session.execute(average_responses_times).fetchall()
+            return results
+
+    @classmethod
+    def get_responses_average_time_in_period(
+        cls,
+        start_time: datetime.datetime,
+        end_time: datetime.datetime,
+        time_format: str,
+        endpoint: Optional[EndpointType] = EndpointType.ANALYZE,
+    ) -> dict[datetime.datetime, int]:
+        """
+        Get a dictionary with average responses times
+        grouped by time units within a specified period.
+
+        Args:
+            start_time (datetime): The start of the time period to query
+            end_time (datetime): The end of the time period to query
+            time_format (str): The strftime format string to format timestamps (e.g., '%Y-%m-%d')
+            endpoint (EndpointType): The analyze API endpoint to query
 
-            return counts_dict
+        Returns:
+            dict[datetime, int]: A dictionary mapping datetime objects
+            to average responses times
+        """
+        with transaction(commit=False) as _:
+            average_responses_times = cls._get_average_responses_times_for_postgres(
+                start_time, end_time, time_format, endpoint
+            )
+
+            return cls.get_dictionary_with_datetime_keys(
+                time_format, average_responses_times
+            )
+
+    @classmethod
+    def _get_average_responses_lengths_for_postgres(
+        cls, start_time, end_time, time_format, endpoint
+    ):
+        """Get average responses length.
+
+        func.to_char is PostgreSQL specific.
+        Let's unit tests replace this function with the SQLite version.
+        """
+        with transaction(commit=False) as session:
+            pgsql_time_format = cls.get_postgres_time_format(time_format)
+
+            average_responses_lengths = (
+                select(
+                    func.to_char(cls.request_received_at, pgsql_time_format).label(
+                        "time_range"
+                    ),
+                    (func.avg(cls.response_length)).label("average_responses_length"),
+                )
+                .filter(cls.request_received_at.between(start_time, end_time))
+                .filter(cls.endpoint == endpoint)
+                .group_by("time_range")
+                .order_by("time_range")
+            )
+
+            results = session.execute(average_responses_lengths).fetchall()
+            return results
+
+    @classmethod
+    def _get_average_responses_lengths_for_sqlite(
+        cls, start_time, end_time, time_format, endpoint
+    ):
+        """Get average responses length.
+
+        func.strftime is SQLite specific.
+        Use this function in unit test using flexmock:
+
+        flexmock(AnalyzeRequestMetrics)
+        .should_receive("_get_average_responses_lengths_for_postgres")
+        .replace_with(AnalyzeRequestMetrics._get_average_responses_lengths_for_sqlite)
+        """
+        with transaction(commit=False) as session:
+            average_responses_lengths = (
+                select(
+                    func.strftime(time_format, cls.request_received_at).label(
+                        "time_range"
+                    ),
+                    (func.avg(cls.response_length)).label("average_responses_length"),
+                )
+                .filter(cls.request_received_at.between(start_time, end_time))
+                .filter(cls.endpoint == endpoint)
+                .group_by("time_range")
+                .order_by("time_range")
+            )
+
+            results = session.execute(average_responses_lengths).fetchall()
+            return results
+
+    @classmethod
+    def get_responses_average_length_in_period(
+        cls,
+        start_time: datetime.datetime,
+        end_time: datetime.datetime,
+        time_format: str,
+        endpoint: Optional[EndpointType] = EndpointType.ANALYZE,
+    ) -> dict[datetime.datetime, int]:
+        """
+        Get a dictionary with average responses length
+        grouped by time units within a specified period.
+
+        Args:
+            start_time (datetime): The start of the time period to query
+            end_time (datetime): The end of the time period to query
+            time_format (str): The strftime format string to format timestamps (e.g., '%Y-%m-%d')
+            endpoint (EndpointType): The analyze API endpoint to query
+
+        Returns:
+            dict[datetime, int]: A dictionary mapping datetime objects
+            to average responses lengths
+        """
+        with transaction(commit=False) as _:
+            average_responses_lengths = cls._get_average_responses_lengths_for_postgres(
+                start_time, end_time, time_format, endpoint
+            )
+
+            return cls.get_dictionary_with_datetime_keys(
+                time_format, average_responses_lengths
+            )

logdetective/server/metric.py

@@ -41,12 +41,10 @@ def update_metrics(
         sent_at if sent_at else datetime.datetime.now(datetime.timezone.utc)
     )
     response_length = None
-    if hasattr(response, "explanation") and "choices" in response.explanation:
-        response_length = sum(
-            len(choice["text"])
-            for choice in response.explanation["choices"]
-            if "text" in choice
-        )
+    if hasattr(response, "explanation") and isinstance(
+        response.explanation, models.Explanation
+    ):
+        response_length = len(response.explanation.text)
     response_certainty = (
         response.response_certainty if hasattr(response, "response_certainty") else None
     )

logdetective/server/models.py

@@ -2,7 +2,9 @@ import datetime
 from logging import BASIC_FORMAT
 from typing import List, Dict, Optional, Literal
 
-from pydantic import BaseModel, Field, model_validator, field_validator
+from pydantic import BaseModel, Field, model_validator, field_validator, NonNegativeFloat
+
+from logdetective.constants import DEFAULT_TEMPERATURE
 
 
 class BuildLog(BaseModel):
@@ -95,6 +97,8 @@ class InferenceConfig(BaseModel):
     )
     url: str = ""
     api_token: str = ""
+    model: str = ""
+    temperature: NonNegativeFloat = DEFAULT_TEMPERATURE
 
     def __init__(self, data: Optional[dict] = None):
         super().__init__()
@@ -106,6 +110,8 @@ class InferenceConfig(BaseModel):
         self.api_endpoint = data.get("api_endpoint", "/chat/completions")
         self.url = data.get("url", "")
         self.api_token = data.get("api_token", "")
+        self.model = data.get("model", "default-model")
+        self.temperature = data.get("temperature", DEFAULT_TEMPERATURE)
 
 
 class ExtractorConfig(BaseModel):
@@ -150,7 +156,8 @@ class LogConfig(BaseModel):
     """Logging configuration"""
 
     name: str = "logdetective"
-    level: str | int = "INFO"
+    level_stream: str | int = "INFO"
+    level_file: str | int = "INFO"
     path: str | None = None
     format: str = BASIC_FORMAT
 
@@ -160,7 +167,8 @@ class LogConfig(BaseModel):
             return
 
         self.name = data.get("name", "logdetective")
-        self.level = data.get("level", "INFO").upper()
+        self.level_stream = data.get("level_stream", "INFO").upper()
+        self.level_file = data.get("level_file", "INFO").upper()
         self.path = data.get("path")
         self.format = data.get("format", BASIC_FORMAT)
 

logdetective/server/plot.py

@@ -1,5 +1,5 @@
 import datetime
-from typing import Optional
+from typing import Optional, Union
 
 import numpy
 import matplotlib
@@ -62,24 +62,24 @@ class Definition:
 
 
 def create_time_series_arrays(
-    counts_dict: dict[datetime.datetime, int],
+    values_dict: dict[datetime.datetime, int],
+    plot_def: Definition,
     start_time: datetime.datetime,
     end_time: datetime.datetime,
-    time_delta: datetime.timedelta,
-    time_format: str,
+    value_type: Optional[Union[int, float]] = int,
 ) -> tuple[numpy.ndarray, numpy.ndarray]:
-    """Create time series arrays from a dictionary of counts.
+    """Create time series arrays from a dictionary of values.
 
     This function generates two aligned numpy arrays:
     1. An array of timestamps from start_time to end_time
-    2. A corresponding array of counts for each timestamp
+    2. A corresponding array of valuesfor each timestamp
 
     The timestamps are truncated to the precision specified by time_format.
-    If a timestamp in counts_dict matches a generated timestamp, its count is used;
-    otherwise, the count defaults to zero.
+    If a timestamp in values_dict matches a generated timestamp, its values is used;
+    otherwise, the value defaults to zero.
 
     Args:
-        counts_dict: Dictionary mapping timestamps to their respective counts
+        values_dict: Dictionary mapping timestamps to their respective values
         start_time: The starting timestamp of the time series
         end_time: The ending timestamp of the time series
         time_delta: The time interval between consecutive timestamps
@@ -88,67 +88,70 @@ def create_time_series_arrays(
     Returns:
         A tuple containing:
             - numpy.ndarray: Array of timestamps
-            - numpy.ndarray: Array of corresponding counts
+            - numpy.ndarray: Array of corresponding values
     """
-    num_intervals = int((end_time - start_time) / time_delta) + 1
+    num_intervals = int((end_time - start_time) / plot_def.time_delta) + 1
 
     timestamps = numpy.array(
         [
             datetime.datetime.strptime(
-                (start_time + i * time_delta).strftime(format=time_format), time_format
+                (start_time + i * plot_def.time_delta).strftime(
+                    format=plot_def.time_format
+                ),
+                plot_def.time_format,
             )
             for i in range(num_intervals)
         ]
     )
-    counts = numpy.zeros(num_intervals, dtype=int)
+    values = numpy.zeros(num_intervals, dtype=value_type)
 
     timestamp_to_index = {timestamp: i for i, timestamp in enumerate(timestamps)}
 
-    for timestamp, count in counts_dict.items():
+    for timestamp, count in values_dict.items():
         if timestamp in timestamp_to_index:
-            counts[timestamp_to_index[timestamp]] = count
+            values[timestamp_to_index[timestamp]] = count
 
-    return timestamps, counts
+    return timestamps, values
 
 
-def _add_bar_chart_for_requests_count(
-    ax1: matplotlib.figure.Axes,
+def _add_bar_chart(
+    ax: matplotlib.figure.Axes,
     plot_def: Definition,
     timestamps: numpy.array,
-    counts: numpy.array,
+    values: numpy.array,
+    label: str,
 ) -> None:
-    """Add a bar chart for requests count (axes 1)"""
+    """Add a blue bar chart"""
     bar_width = (
         0.8 * plot_def.time_delta.total_seconds() / 86400
     )  # Convert to days for matplotlib
-    ax1.bar(
+    ax.bar(
         timestamps,
-        counts,
+        values,
         width=bar_width,
         alpha=0.7,
         color="skyblue",
-        label="Requests",
+        label=label,
     )
-    ax1.set_xlabel("Time")
-    ax1.set_ylabel("Requests", color="blue")
-    ax1.tick_params(axis="y", labelcolor="blue")
+    ax.set_xlabel("Time")
+    ax.set_ylabel(label, color="blue")
+    ax.tick_params(axis="y", labelcolor="blue")
 
-    ax1.xaxis.set_major_formatter(matplotlib.dates.DateFormatter(plot_def.time_format))
-    ax1.xaxis.set_major_locator(plot_def.locator)
+    ax.xaxis.set_major_formatter(matplotlib.dates.DateFormatter(plot_def.time_format))
+    ax.xaxis.set_major_locator(plot_def.locator)
 
     matplotlib.pyplot.xticks(rotation=45)
 
-    ax1.grid(True, alpha=0.3)
+    ax.grid(True, alpha=0.3)
 
 
-def _add_cumulative_line_for_requests_count(
-    ax2: matplotlib.figure.Axes, timestamps: numpy.array, counts: numpy.array
+def _add_line_chart(
+    ax: matplotlib.figure.Axes, timestamps: numpy.array, values: numpy.array, label: str
 ) -> None:
-    """Add cumulative line on secondary y-axis"""
-    cumulative = numpy.cumsum(counts)
-    ax2.plot(timestamps, cumulative, "r-", linewidth=2, label="Cumulative")
-    ax2.set_ylabel("Cumulative Requests", color="red")
-    ax2.tick_params(axis="y", labelcolor="red")
+    """Add a red line chart"""
+    ax.plot(timestamps, values, "r-", linewidth=2, label=label)
+    ax.set_ylabel(label, color="red")
+    ax.tick_params(axis="y", labelcolor="red")
 
 
 def requests_per_time(
@@ -183,14 +186,14 @@ def requests_per_time(
         start_time, end_time, plot_def.time_format, endpoint
     )
     timestamps, counts = create_time_series_arrays(
-        requests_counts, start_time, end_time, plot_def.time_delta, plot_def.time_format
+        requests_counts, plot_def, start_time, end_time
     )
 
     fig, ax1 = matplotlib.pyplot.subplots(figsize=(12, 6))
-    _add_bar_chart_for_requests_count(ax1, plot_def, timestamps, counts)
+    _add_bar_chart(ax1, plot_def, timestamps, counts, "Requests")
 
     ax2 = ax1.twinx()
-    _add_cumulative_line_for_requests_count(ax2, timestamps, counts)
+    _add_line_chart(ax2, timestamps, numpy.cumsum(counts), "Cumulative Requests")
 
     matplotlib.pyplot.title(
         f"Requests received for API {endpoint} ({start_time.strftime(plot_def.time_format)} "
@@ -204,3 +207,75 @@ def requests_per_time(
     matplotlib.pyplot.tight_layout()
 
     return fig
+
+
+def average_time_per_responses(  # pylint: disable=too-many-locals
+    period_of_time: models.TimePeriod,
+    endpoint: EndpointType = EndpointType.ANALYZE,
+    end_time: Optional[datetime.datetime] = None,
+) -> matplotlib.figure.Figure:
+    """
+    Generate a visualization of average response time and length over a specified time period.
+
+    This function creates a dual-axis plot showing:
+    1. A bar chart of average response time per time interval
+    1. A line chart of average response length per time interval
+
+    The time intervals are determined by the provided TimePeriod object, which defines
+    the granularity and formatting of the time axis.
+
+    Args:
+        period_of_time: A TimePeriod object that defines the time period and interval
+                        for the analysis (e.g., hourly, daily, weekly)
+        endpoint: One of the API endpoints
+        end_time: The end time for the analysis period. If None, defaults to the current
+                  UTC time
+
+    Returns:
+        A matplotlib Figure object containing the generated visualization
+    """
+    end_time = end_time or datetime.datetime.now(datetime.timezone.utc)
+    start_time = period_of_time.get_period_start_time(end_time)
+    plot_def = Definition(period_of_time)
+    responses_average_time = AnalyzeRequestMetrics.get_responses_average_time_in_period(
+        start_time, end_time, plot_def.time_format, endpoint
+    )
+    timestamps, average_time = create_time_series_arrays(
+        responses_average_time,
+        plot_def,
+        start_time,
+        end_time,
+        float,
+    )
+
+    fig, ax1 = matplotlib.pyplot.subplots(figsize=(12, 6))
+    _add_bar_chart(ax1, plot_def, timestamps, average_time, "average response time (seconds)")
+
+    responses_average_length = (
+        AnalyzeRequestMetrics.get_responses_average_length_in_period(
+            start_time, end_time, plot_def.time_format, endpoint
+        )
+    )
+    timestamps, average_length = create_time_series_arrays(
+        responses_average_length,
+        plot_def,
+        start_time,
+        end_time,
+        float,
+    )
+
+    ax2 = ax1.twinx()
+    _add_line_chart(ax2, timestamps, average_length, "average response length (chars)")
+
+    matplotlib.pyplot.title(
+        f"average response time for API {endpoint} ({start_time.strftime(plot_def.time_format)} "
+        f"to {end_time.strftime(plot_def.time_format)})"
+    )
+
+    lines1, labels1 = ax1.get_legend_handles_labels()
+    lines2, labels2 = ax2.get_legend_handles_labels()
+    ax1.legend(lines1 + lines2, labels1 + labels2, loc="center")
+
+    matplotlib.pyplot.tight_layout()
+
+    return fig

logdetective/server/server.py

@@ -26,10 +26,13 @@ from logdetective.utils import (
     validate_url,
     compute_certainty,
     format_snippets,
-    format_analyzed_snippets,
     load_prompts,
 )
-from logdetective.server.utils import load_server_config, get_log
+from logdetective.server.utils import (
+    load_server_config,
+    get_log,
+    format_analyzed_snippets,
+)
 from logdetective.server.metric import track_request
 from logdetective.server.models import (
     BuildLog,
@@ -183,7 +186,6 @@ async def submit_text(  # pylint: disable=R0913,R0917
     log_probs: int = 1,
     stream: bool = False,
     model: str = "default-model",
-    api_endpoint: str = "/chat/completions",
 ) -> Explanation:
     """Submit prompt to LLM using a selected endpoint.
     max_tokens: number of tokens to be produces, 0 indicates run until encountering EOS
@@ -196,7 +198,7 @@ async def submit_text(  # pylint: disable=R0913,R0917
     if SERVER_CONFIG.inference.api_token:
         headers["Authorization"] = f"Bearer {SERVER_CONFIG.inference.api_token}"
 
-    if api_endpoint == "/chat/completions":
+    if SERVER_CONFIG.inference.api_endpoint == "/chat/completions":
         return await submit_text_chat_completions(
             text, headers, max_tokens, log_probs > 0, stream, model
         )
@@ -224,6 +226,7 @@ async def submit_text_completions(  # pylint: disable=R0913,R0917
         "logprobs": log_probs,
         "stream": stream,
         "model": model,
+        "temperature": SERVER_CONFIG.inference.temperature,
     }
 
     response = await submit_to_llm_endpoint(
@@ -263,6 +266,7 @@ async def submit_text_chat_completions(  # pylint: disable=R0913,R0917
         "logprobs": log_probs,
         "stream": stream,
         "model": model,
+        "temperature": SERVER_CONFIG.inference.temperature,
     }
 
     response = await submit_to_llm_endpoint(
@@ -297,7 +301,8 @@ async def analyze_log(build_log: BuildLog):
     log_summary = format_snippets(log_summary)
     response = await submit_text(
         PROMPT_CONFIG.prompt_template.format(log_summary),
-        api_endpoint=SERVER_CONFIG.inference.api_endpoint,
+        model=SERVER_CONFIG.inference.model,
+        max_tokens=SERVER_CONFIG.inference.max_tokens,
     )
     certainty = 0
 
@@ -337,7 +342,8 @@ async def perform_staged_analysis(log_text: str) -> StagedResponse:
         *[
             submit_text(
                 PROMPT_CONFIG.snippet_prompt_template.format(s),
-                api_endpoint=SERVER_CONFIG.inference.api_endpoint,
+                model=SERVER_CONFIG.inference.model,
+                max_tokens=SERVER_CONFIG.inference.max_tokens,
             )
             for s in log_summary
         ]
@@ -352,7 +358,9 @@ async def perform_staged_analysis(log_text: str) -> StagedResponse:
     )
 
     final_analysis = await submit_text(
-        final_prompt, api_endpoint=SERVER_CONFIG.inference.api_endpoint
+        final_prompt,
+        model=SERVER_CONFIG.inference.model,
+        max_tokens=SERVER_CONFIG.inference.max_tokens,
     )
 
     certainty = 0
@@ -393,7 +401,9 @@ async def analyze_log_stream(build_log: BuildLog):
         headers["Authorization"] = f"Bearer {SERVER_CONFIG.inference.api_token}"
 
     stream = await submit_text_chat_completions(
-        PROMPT_CONFIG.prompt_template.format(log_summary), stream=True, headers=headers
+        PROMPT_CONFIG.prompt_template.format(log_summary), stream=True, headers=headers,
+        model=SERVER_CONFIG.inference.model,
+        max_tokens=SERVER_CONFIG.inference.max_tokens,
     )
 
     return StreamingResponse(stream)
@@ -613,8 +623,8 @@ async def comment_on_mr(
         response.explanation.text,
     )
 
-    # Get the formatted comment.
-    comment = await generate_mr_comment(job, log_url, response)
+    # Get the formatted short comment.
+    short_comment = await generate_mr_comment(job, log_url, response, full=False)
 
     # Look up the merge request
     merge_request = await asyncio.to_thread(
@@ -622,11 +632,33 @@ async def comment_on_mr(
     )
 
     # Submit a new comment to the Merge Request using the Gitlab API
-    await asyncio.to_thread(merge_request.discussions.create, {"body": comment})
+    discussion = await asyncio.to_thread(
+        merge_request.discussions.create, {"body": short_comment}
+    )
+
+    # Get the ID of the first note
+    note_id = discussion.attributes["notes"][0]["id"]
+    note = discussion.notes.get(note_id)
+
+    # Update the comment with the full details
+    # We do this in a second step so we don't bombard the user's email
+    # notifications with a massive message. Gitlab doesn't send email for
+    # comment edits.
+    full_comment = await generate_mr_comment(job, log_url, response, full=True)
+    note.body = full_comment
+
+    # Pause for five seconds before sending the snippet data, otherwise
+    # Gitlab may bundle the edited message together with the creation
+    # message in email.
+    await asyncio.sleep(5)
+    await asyncio.to_thread(note.save)
 
 
 async def generate_mr_comment(
-    job: gitlab.v4.objects.ProjectJob, log_url: str, response: StagedResponse
+    job: gitlab.v4.objects.ProjectJob,
+    log_url: str,
+    response: StagedResponse,
+    full: bool = True,
 ) -> str:
     """Use a template to generate a comment string to submit to Gitlab"""
 
@@ -634,7 +666,11 @@ async def generate_mr_comment(
     script_path = Path(__file__).resolve().parent
     template_path = Path(script_path, "templates")
     jinja_env = jinja2.Environment(loader=jinja2.FileSystemLoader(template_path))
-    tpl = jinja_env.get_template("gitlab_comment.md.j2")
+
+    if full:
+        tpl = jinja_env.get_template("gitlab_full_comment.md.j2")
+    else:
+        tpl = jinja_env.get_template("gitlab_short_comment.md.j2")
 
     artifacts_url = f"{job.project_url}/-/jobs/{job.id}/artifacts/download"
 
@@ -673,6 +709,35 @@ def _svg_figure_response(fig: matplotlib.figure.Figure):
     )
 
 
+def _multiple_svg_figures_response(figures: list[matplotlib.figure.Figure]):
+    """Create a response with multiple svg figures."""
+    svg_contents = []
+    for i, fig in enumerate(figures):
+        buf = BytesIO()
+        fig.savefig(buf, format="svg", bbox_inches="tight")
+        matplotlib.pyplot.close(fig)
+        buf.seek(0)
+        svg_contents.append(buf.read().decode("utf-8"))
+
+    html_content = "<html><body>\n"
+    for i, svg in enumerate(svg_contents):
+        html_content += f"<div id='figure-{i}'>\n{svg}\n</div>\n"
+    html_content += "</body></html>"
+
+    return BasicResponse(content=html_content, media_type="text/html")
+
+
+@app.get("/metrics/analyze", response_class=StreamingResponse)
+async def show_analyze_metrics(period_since_now: TimePeriod = Depends(TimePeriod)):
+    """Show statistics for requests and responses in the given period of time
+    for the /analyze API endpoint."""
+    fig_requests = plot.requests_per_time(period_since_now, EndpointType.ANALYZE)
+    fig_responses = plot.average_time_per_responses(
+        period_since_now, EndpointType.ANALYZE
+    )
+    return _multiple_svg_figures_response([fig_requests, fig_responses])
+
+
 @app.get("/metrics/analyze/requests", response_class=StreamingResponse)
 async def show_analyze_requests(period_since_now: TimePeriod = Depends(TimePeriod)):
     """Show statistics for the requests received in the given period of time
@@ -681,6 +746,27 @@ async def show_analyze_requests(period_since_now: TimePeriod = Depends(TimePerio
     return _svg_figure_response(fig)
 
 
+@app.get("/metrics/analyze/responses", response_class=StreamingResponse)
+async def show_analyze_responses(period_since_now: TimePeriod = Depends(TimePeriod)):
+    """Show statistics for responses given in the specified period of time
+    for the /analyze API endpoint."""
+    fig = plot.average_time_per_responses(period_since_now, EndpointType.ANALYZE)
+    return _svg_figure_response(fig)
+
+
+@app.get("/metrics/analyze/staged", response_class=StreamingResponse)
+async def show_analyze_staged_metrics(
+    period_since_now: TimePeriod = Depends(TimePeriod),
+):
+    """Show statistics for requests and responses in the given period of time
+    for the /analyze/staged API endpoint."""
+    fig_requests = plot.requests_per_time(period_since_now, EndpointType.ANALYZE_STAGED)
+    fig_responses = plot.average_time_per_responses(
+        period_since_now, EndpointType.ANALYZE_STAGED
+    )
+    return _multiple_svg_figures_response([fig_requests, fig_responses])
+
+
 @app.get("/metrics/analyze/staged/requests", response_class=StreamingResponse)
 async def show_analyze_staged_requests(
     period_since_now: TimePeriod = Depends(TimePeriod),
@@ -689,3 +775,13 @@ async def show_analyze_staged_requests(
     for the /analyze/staged API endpoint."""
     fig = plot.requests_per_time(period_since_now, EndpointType.ANALYZE_STAGED)
     return _svg_figure_response(fig)
+
+
+@app.get("/metrics/analyze/staged/responses", response_class=StreamingResponse)
+async def show_analyze_staged_responses(
+    period_since_now: TimePeriod = Depends(TimePeriod),
+):
+    """Show statistics for responses given in the specified period of time
+    for the /analyze/staged API endpoint."""
+    fig = plot.average_time_per_responses(period_since_now, EndpointType.ANALYZE_STAGED)
+    return _svg_figure_response(fig)

logdetective/server/templates/{gitlab_comment.md.j2 → gitlab_full_comment.md.j2}

@@ -9,9 +9,7 @@ In this case, we are {{ certainty }}% certain of the response {{ emoji_face }}.
 <ul>
 {% for snippet in snippets %}
 <li>
-<code>
-Line {{ snippet.line_number }}: {{ snippet.text }}
-</code>
+<b>Line {{ snippet.line_number }}:</b> <code>{{ snippet.text }}</code>
 {{ snippet.explanation }}
 </li>
 {% endfor %}

logdetective/server/templates/gitlab_short_comment.md.j2

@@ -0,0 +1,53 @@
+The package {{ package }} failed to build, here is a possible explanation why.
+
+Please know that the explanation was provided by AI and may be incorrect.
+In this case, we are {{ certainty }}% certain of the response {{ emoji_face }}.
+
+{{ explanation }}
+
+<details>
+  <summary>Logs</summary>
+  <p>
+    Log Detective analyzed the following logs files to provide an explanation:
+  </p>
+
+  <ul>
+    <li><a href="{{ log_url }}">{{ log_url }}</a></li>
+  </ul>
+
+  <p>
+    Additional logs are available from:
+    <ul>
+    <li><a href="{{ artifacts_url }}">artifacts.zip</a></li>
+  </ul>
+  </p>
+
+  <p>
+    Please know that these log files are automatically removed after some
+    time, so you might need a backup.
+  </p>
+</details>
+
+<details>
+  <summary>Help</summary>
+  <p>Don't hesitate to reach out.</p>
+
+  <ul>
+    <li><a href="https://github.com/fedora-copr/logdetective">Upstream</a></li>
+    <li><a href="https://github.com/fedora-copr/logdetective/issues">Issue tracker</a></li>
+    <li><a href="https://redhat.enterprise.slack.com/archives/C06DWNVKKDE">Slack</a></li>
+    <li><a href="https://log-detective.com/documentation">Documentation</a></li>
+  </ul>
+</details>
+
+
+---
+This comment was created by [Log Detective][log-detective].
+
+Was the provided feedback accurate and helpful? <br>Please vote with :thumbsup:
+or :thumbsdown: to help us improve.<br>
+
+
+
+[log-detective]: https://log-detective.com/
+[contact]: https://github.com/fedora-copr

logdetective/server/utils.py

@@ -1,6 +1,18 @@
 import logging
 import yaml
-from logdetective.server.models import Config
+from logdetective.constants import SNIPPET_DELIMITER
+from logdetective.server.models import Config, AnalyzedSnippet
+
+
+def format_analyzed_snippets(snippets: list[AnalyzedSnippet]) -> str:
+    """Format snippets for submission into staged prompt."""
+    summary = f"\n{SNIPPET_DELIMITER}\n".join(
+        [
+            f"[{e.text}] at line [{e.line_number}]: [{e.explanation.text}]"
+            for e in snippets
+        ]
+    )
+    return summary
 
 
 def load_server_config(path: str | None) -> Config:
@@ -25,7 +37,7 @@ def get_log(config: Config):
     if getattr(log, "initialized", False):
         return log
 
-    log.setLevel(config.log.level)
+    log.setLevel("DEBUG")
 
     # Drop the default handler, we will create it ourselves
     log.handlers = []
@@ -33,12 +45,14 @@ def get_log(config: Config):
     # STDOUT
     stream_handler = logging.StreamHandler()
     stream_handler.setFormatter(logging.Formatter(config.log.format))
+    stream_handler.setLevel(config.log.level_stream)
     log.addHandler(stream_handler)
 
     # Log to file
     if config.log.path:
         file_handler = logging.FileHandler(config.log.path)
         file_handler.setFormatter(logging.Formatter(config.log.format))
+        file_handler.setLevel(config.log.level_file)
         log.addHandler(file_handler)
 
     log.initialized = True

logdetective/utils.py

@@ -7,8 +7,6 @@ import requests
 import yaml
 
 from llama_cpp import Llama, CreateCompletionResponse, CreateCompletionStreamResponse
-from logdetective.constants import SNIPPET_DELIMITER
-from logdetective.server.models import AnalyzedSnippet
 from logdetective.models import PromptConfig
 
 
@@ -113,19 +111,23 @@ def compute_certainty(probs: List[Dict]) -> float:
 
 
 def process_log(
-    log: str, model: Llama, stream: bool, prompt_template: str
+    log: str, model: Llama, stream: bool, prompt_template: str,
+    temperature: float
 ) -> CreateCompletionResponse | Iterator[CreateCompletionStreamResponse]:
     """Processes a given log using the provided language model and returns its summary.
 
     Args:
         log (str): The input log to be processed.
         model (Llama): The language model used for processing the log.
-
+        stream (bool): Return output as Iterator.
+        prompt_template (str): Which prompt template to use.
+        temperature (float): Temperature parameter for model runtime.
     Returns:
         str: The summary of the given log generated by the language model.
     """
     response = model(
-        prompt=prompt_template.format(log), stream=stream, max_tokens=0, logprobs=1
+        prompt=prompt_template.format(log), stream=stream, max_tokens=0, logprobs=1,
+        temperature=temperature
     )
 
     return response
@@ -178,17 +180,6 @@ def format_snippets(snippets: list[str] | list[Tuple[int, str]]) -> str:
     return summary
 
 
-def format_analyzed_snippets(snippets: list[AnalyzedSnippet]) -> str:
-    """Format snippets for submission into staged prompt."""
-    summary = f"\n{SNIPPET_DELIMITER}\n".join(
-        [
-            f"[{e.text}] at line [{e.line_number}]: [{e.explanation.text}]"
-            for e in snippets
-        ]
-    )
-    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.

{logdetective-0.5.9.dist-info → logdetective-0.5.11.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: logdetective
-Version: 0.5.9
+Version: 0.5.11
 Summary: Log using LLM AI to search for build/test failures and provide ideas for fixing these.
 License: Apache-2.0
 Author: Jiri Podivin
@@ -47,6 +47,8 @@ Log Detective
 
 A Python tool to analyze logs using a Language Model (LLM) and Drain template miner.
 
+Note: if you are looking for code of website logdetective.com it is in [github.com/fedora-copr/logdetective-website](https://github.com/fedora-copr/logdetective-website).
+
 Installation
 ------------
 
@@ -95,6 +97,17 @@ 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
 
+Example of different suffix (useful for models that were quantized)
+
+    logdetective https://kojipkgs.fedoraproject.org//work/tasks/3367/131313367/build.log --model 'fedora-copr/granite-3.2-8b-instruct-GGUF' -F Q4_K.gguf
+
+Example of altered prompts:
+
+     cp ~/.local/lib/python3.13/site-packages/logdetective/prompts.yml ~/my-prompts.yml
+     vi ~/my-prompts.yml # edit the prompts there to better fit your needs
+     logdetective https://kojipkgs.fedoraproject.org//work/tasks/3367/131313367/build.log --prompts ~/my-prompts.yml
+
+
 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
@@ -337,11 +350,23 @@ certbot certonly --standalone -d logdetective01.fedorainfracloud.org
 Querying statistics
 -------------------
 
-You can retrieve statistics about server requests over a specified time period
-using either the `curl` command or the `http` command (provided by the `httpie` package).
+You can retrieve statistics about server requests and responses over a specified time period
+using either a browser, the `curl` or the `http` command (provided by the `httpie` package).
 
 When no time period is specified, the query defaults to the last 2 days:
 
+You can view requests and responses statistics
+ - for the `/analyze` endpoint at http://localhost:8080/metrics/analyze
+ - for the `/analyze/staged` endpoint at http://localhost:8080/metrics/analyze/staged.
+
+You can retrieve single svg images at the following endpoints:
+ - `/metrics/analyze/requests`
+ - `/metrics/analyze/responses`
+ - `/metrics/analyze/staged/requests`
+ - `/metrics/analyze/stages/responses`
+
+Examples:
+
 ```
 http GET "localhost:8080/metrics/analyze/requests" > /tmp/plot.svg
 curl "localhost:8080/metrics/analyze/staged/requests" > /tmp/plot.svg
@@ -349,7 +374,6 @@ curl "localhost:8080/metrics/analyze/staged/requests" > /tmp/plot.svg
 
 You can specify the time period in hours, days, or weeks.
 The time period:
-
  - cannot be less than one hour
  - cannot be negative
  - ends at the current time (when the query is made)

logdetective-0.5.11.dist-info/RECORD

@@ -0,0 +1,24 @@
+logdetective/__init__.py,sha256=VqRngDcuFT7JWms8Qc_MsOvajoXVOKPr-S1kqY3Pqhc,59
+logdetective/constants.py,sha256=A5PzeqlQqDbBS_kzP2hl-lhJ0lCEqdbvW3CaQUYVxjw,1849
+logdetective/drain3.ini,sha256=ni91eCT1TwTznZwcqWoOVMQcGEnWhEDNCoTPF7cfGfY,1360
+logdetective/extractors.py,sha256=7ahzWbTtU9MveG1Q7wU9LO8OJgs85X-cHmWltUhCe9M,3491
+logdetective/logdetective.py,sha256=Q1SfQ9sWR5sIvHJag61-F-8edwf7p1SV7QZRg9VaWcc,5604
+logdetective/models.py,sha256=nrGBmMRu8i6UhFflQKAp81Y3Sd_Aaoor0i_yqSJoLT0,1115
+logdetective/prompts.yml,sha256=dMW2-bdTIqv7LF_owqRD4xinMK5ZWcNhDynnX1zoKns,1722
+logdetective/server/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+logdetective/server/database/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+logdetective/server/database/base.py,sha256=oMJUvbWeapIUP-8Cf_DR9ptFg8CsYeaBAIjOVEzx8SM,1668
+logdetective/server/database/models.py,sha256=m_3qNBWJwLSwjJn0AmwSxXMJk75Gu1bXFtGAP_4zps4,14088
+logdetective/server/metric.py,sha256=-uM_-yqxNA-EZTCnNRdQ8g1MicmE5eC6jRFI_mBBYUg,2606
+logdetective/server/models.py,sha256=URqZcfx5yUsifZ1pOwZ_uU3Tyjcdvuq6qEnAvTexl4A,8475
+logdetective/server/plot.py,sha256=B2rOngqx7g-Z3NfttboTip3frkypdF1H7FhK8vh45mE,9655
+logdetective/server/server.py,sha256=4NylBojHm9E3gjByVWs870T204ls39EbZmUfU0Kyq4U,28395
+logdetective/server/templates/gitlab_full_comment.md.j2,sha256=DQZ2WVFedpuXI6znbHIW4wpF9BmFS8FaUkowh8AnGhE,1627
+logdetective/server/templates/gitlab_short_comment.md.j2,sha256=fzScpayv2vpRLczP_0O0YxtA8rsKvR6gSv4ntNdWb98,1443
+logdetective/server/utils.py,sha256=QO0H1q55YLCLKxkViqex4Uu31LnakpYUKJfZHysonSc,1838
+logdetective/utils.py,sha256=nklnTipAet9P9aEiuHcnK62WT0DmNHbvO1TvNlrxlik,6463
+logdetective-0.5.11.dist-info/LICENSE,sha256=z8d0m5b2O9McPEK1xHG_dWgUBT6EfBDz6wA0F7xSPTA,11358
+logdetective-0.5.11.dist-info/METADATA,sha256=LOOzu99kJaP02U2OaFQciPdWKhlgr4Vm4tVKijTY7NM,15882
+logdetective-0.5.11.dist-info/WHEEL,sha256=fGIA9gx4Qxk2KDKeNJCbOEwSrmLtjWCwzBz351GyrPQ,88
+logdetective-0.5.11.dist-info/entry_points.txt,sha256=3K_vXja6PmcA8sNdUi63WdImeiNhVZcEGPTaoJmltfA,63
+logdetective-0.5.11.dist-info/RECORD,,

logdetective-0.5.9.dist-info/RECORD

@@ -1,23 +0,0 @@
-logdetective/__init__.py,sha256=VqRngDcuFT7JWms8Qc_MsOvajoXVOKPr-S1kqY3Pqhc,59
-logdetective/constants.py,sha256=eiS6eYhEgl_Rlyi_B9j00DDp9A-UDhuFz3ACWtKf_SU,1558
-logdetective/drain3.ini,sha256=ni91eCT1TwTznZwcqWoOVMQcGEnWhEDNCoTPF7cfGfY,1360
-logdetective/extractors.py,sha256=7ahzWbTtU9MveG1Q7wU9LO8OJgs85X-cHmWltUhCe9M,3491
-logdetective/logdetective.py,sha256=1EFrml_gHdyKEZX4iXBxhGgmU7R7_S26-Fr0WUDaA7E,5316
-logdetective/models.py,sha256=nrGBmMRu8i6UhFflQKAp81Y3Sd_Aaoor0i_yqSJoLT0,1115
-logdetective/prompts.yml,sha256=OBOWDErlbigbLrStcCY5HKPReNb0g-SNlCnD4QawF7k,1268
-logdetective/server/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-logdetective/server/database/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-logdetective/server/database/base.py,sha256=oMJUvbWeapIUP-8Cf_DR9ptFg8CsYeaBAIjOVEzx8SM,1668
-logdetective/server/database/models.py,sha256=arIahOCT-hTmh904DXrWSkH7rlo13Ppu-OO80huX5Dc,6118
-logdetective/server/metric.py,sha256=VYMifrfIhcqgyu6YYN0c1nt8fC1iJ2_LCB7Bh2AheoE,2679
-logdetective/server/models.py,sha256=cf1ngu_-19rP_i49s5cEwIzh6SfL_ZpVy4EykCpfWck,8076
-logdetective/server/plot.py,sha256=3o-CNHjel04ekpwSB4ckV7dbiF663cfPkimQ0aP9U_8,7073
-logdetective/server/server.py,sha256=VGfBgbjUcyBd8hop-ea-O_Mo-FoGLDyP-elAWzRu51g,24605
-logdetective/server/templates/gitlab_comment.md.j2,sha256=kheTkhQ-LfuFkr8av-Mw2a-9VYEUbDTLwaa-CKI6OkI,1622
-logdetective/server/utils.py,sha256=OFvhttjv3yp8kfim5_s4mNG8ly21qyILxE0o3DcVVKg,1340
-logdetective/utils.py,sha256=yTEjfTTaCS8lreKRkwKzLo6Po8cOYzInjSEx4CwpyqA,6665
-logdetective-0.5.9.dist-info/LICENSE,sha256=z8d0m5b2O9McPEK1xHG_dWgUBT6EfBDz6wA0F7xSPTA,11358
-logdetective-0.5.9.dist-info/METADATA,sha256=YZbrICuAKXVD4LEEH6orwX-fuX3i3hpSsKuNa1nosoI,14737
-logdetective-0.5.9.dist-info/WHEEL,sha256=fGIA9gx4Qxk2KDKeNJCbOEwSrmLtjWCwzBz351GyrPQ,88
-logdetective-0.5.9.dist-info/entry_points.txt,sha256=3K_vXja6PmcA8sNdUi63WdImeiNhVZcEGPTaoJmltfA,63
-logdetective-0.5.9.dist-info/RECORD,,