UN-3987 [FIX] Send LLMWhisperer V2 params under the correct names

docker/scripts/uv-lock-gen/uv-lock.sh

@@ -1,28 +1,23 @@
 #!/bin/bash
 set -o pipefail
 
-# Extract local path dependencies from [tool.uv.sources] in a pyproject.toml.
-# Returns resolved pyproject.toml paths for each local dependency.
+# Resolved pyproject.toml paths for a directory's local path dependencies.
+# The lockfile is read as well as the pyproject: it records the whole resolved
+# graph, so a change further down it (sdk1 -> connectors) is still seen, while
+# the pyproject covers a dependency added but not yet locked.
 get_local_dep_pyprojects() {
     local dir="$1"
-    local file_path="$dir/pyproject.toml"
 
-    grep -A1 'path\s*=' "$file_path" 2>/dev/null \
-        | grep -oP 'path\s*=\s*"\K[^"]+' \
-        | while read -r rel_path; do
-            # Resolve relative to the service directory
-            local dep_pyproject
-            if [[ "$dir" == "." ]]; then
-                dep_pyproject="$rel_path/pyproject.toml"
-            else
-                dep_pyproject="$dir/$rel_path/pyproject.toml"
-            fi
-            # Normalize the path
-            dep_pyproject=$(realpath --relative-to=. "$dep_pyproject" 2>/dev/null || echo "$dep_pyproject")
-            if [[ -f "$dep_pyproject" ]]; then
-                echo "$dep_pyproject"
-            fi
-        done
+    {
+        grep -oP 'path\s*=\s*"\K[^"]+' "$dir/pyproject.toml" 2>/dev/null
+        grep -oP 'source = \{ (editable|directory) = "\K[^"]+' "$dir/uv.lock" 2>/dev/null
+    } | while read -r rel_path; do
+        local dep_pyproject
+        dep_pyproject=$(realpath --relative-to=. "$dir/$rel_path/pyproject.toml" 2>/dev/null) || continue
+        if [[ -f "$dep_pyproject" ]]; then
+            echo "$dep_pyproject"
+        fi
+    done | sort -u
 }
 
 # Check if a directory's own pyproject.toml or any of its local
@@ -107,6 +102,11 @@ directories=(
     "unstract/core"
     "unstract/flags"
     "unstract/connectors"
+    "unstract/sdk1"
+    "unstract/tool-registry"
+    "unstract/tool-sandbox"
+    "unstract/workflow-execution"
+    "tool-sidecar"
     "workers"
 )
 

unstract/sdk1/pyproject.toml

@@ -47,7 +47,7 @@ dependencies = [
     "pdfplumber>=0.11.2",
     "redis>=5.2.1",
     # # LLMWhisperer client
-    "llmwhisperer-client>=2.6.2",
+    "llmwhisperer-client>=2.8.1",
     # # Core utilities (Redis Sentinel-aware client factory, etc.)
     "unstract-core",
 ]

unstract/sdk1/src/unstract/sdk1/adapters/x2text/llm_whisperer_v2/src/constants.py

@@ -14,6 +14,12 @@ class OutputModes(Enum):
     TEXT = "text"
 
 
+class LineSplitterStrategies(Enum):
+    LEFT_PRIORITY = "left-priority"
+    MID_PRIORITY = "mid-priority"
+    RIGHT_PRIORITY = "right-priority"
+
+
 class HTTPMethod(Enum):
     GET = "GET"
     POST = "POST"
@@ -60,12 +66,14 @@ class WhispererConfig:
     MEDIAN_FILTER_SIZE = "median_filter_size"
     GAUSSIAN_BLUR_RADIUS = "gaussian_blur_radius"
     LINE_SPLITTER_TOLERANCE = "line_splitter_tolerance"
-    LINE_SPLITTER_STRATEGY = "line_spitter_strategy"
+    LINE_SPLITTER_STRATEGY = "line_splitter_strategy"
     HORIZONTAL_STRETCH_FACTOR = "horizontal_stretch_factor"
     PAGES_TO_EXTRACT = "pages_to_extract"
     MARK_VERTICAL_LINES = "mark_vertical_lines"
     MARK_HORIZONTAL_LINES = "mark_horizontal_lines"
+    # Misspelling retained: correcting the JSON schema key means migrating saved configs
     PAGE_SEPARATOR = "page_seperator"
+    PAGE_SEPARATOR_PARAM = "page_separator"
     URL_IN_POST = "url_in_post"
     TAG = "tag"
     USE_WEBHOOK = "use_webhook"
@@ -78,6 +86,7 @@ class WhispererConfig:
     INCLUDE_LINE_CONFIDENCE = "include_line_confidence"
     EXTRACT_ALL_LINES = "extract_all_lines"
     LINES = "lines"
+    FILE_NAME = "file_name"
 
 
 class WhisperStatus:

unstract/sdk1/src/unstract/sdk1/adapters/x2text/llm_whisperer_v2/src/helper.py

@@ -15,6 +15,7 @@
 from unstract.sdk1.adapters.utils import AdapterUtils
 from unstract.sdk1.adapters.x2text.constants import X2TextConstants
 from unstract.sdk1.adapters.x2text.llm_whisperer_v2.src.constants import (
+    LineSplitterStrategies,
     Modes,
     OutputModes,
     WhispererConfig,
@@ -171,6 +172,20 @@ def get_whisperer_params(
         Returns:
             dict[str, Any]: Query params
         """
+        line_splitter_strategy = config.get(
+            WhispererConfig.LINE_SPLITTER_STRATEGY,
+            WhispererDefaults.LINE_SPLITTER_STRATEGY,
+        )
+        # This setting was never sent to the service, so stored values were never
+        # validated. The service rejects an unknown strategy with a 400.
+        if line_splitter_strategy not in {s.value for s in LineSplitterStrategies}:
+            logger.warning(
+                "Unsupported line splitter strategy '%s', falling back to '%s'",
+                line_splitter_strategy,
+                WhispererDefaults.LINE_SPLITTER_STRATEGY,
+            )
+            line_splitter_strategy = WhispererDefaults.LINE_SPLITTER_STRATEGY
+
         params = {
             WhispererConfig.MODE: config.get(WhispererConfig.MODE, Modes.FORM.value),
             WhispererConfig.OUTPUT_MODE: config.get(
@@ -180,10 +195,7 @@ def get_whisperer_params(
                 WhispererConfig.LINE_SPLITTER_TOLERANCE,
                 WhispererDefaults.LINE_SPLITTER_TOLERANCE,
             ),
-            WhispererConfig.LINE_SPLITTER_STRATEGY: config.get(
-                WhispererConfig.LINE_SPLITTER_STRATEGY,
-                WhispererDefaults.LINE_SPLITTER_STRATEGY,
-            ),
+            WhispererConfig.LINE_SPLITTER_STRATEGY: line_splitter_strategy,
             WhispererConfig.HORIZONTAL_STRETCH_FACTOR: config.get(
                 WhispererConfig.HORIZONTAL_STRETCH_FACTOR,
                 WhispererDefaults.HORIZONTAL_STRETCH_FACTOR,
@@ -200,7 +212,7 @@ def get_whisperer_params(
                 WhispererConfig.MARK_HORIZONTAL_LINES,
                 WhispererDefaults.MARK_HORIZONTAL_LINES,
             ),
-            WhispererConfig.PAGE_SEPARATOR: config.get(
+            WhispererConfig.PAGE_SEPARATOR_PARAM: config.get(
                 WhispererConfig.PAGE_SEPARATOR,
                 WhispererDefaults.PAGE_SEPARATOR,
             ),
@@ -260,6 +272,8 @@ def send_whisper_request(
         params = LLMWhispererHelper.get_whisperer_params(
             config=config, extra_params=extra_params
         )
+        # Recorded against the extraction for cross referencing in usage reports
+        params[WhispererConfig.FILE_NAME] = Path(input_file_path).name
         response: requests.Response
         try:
             input_file_data = BytesIO(fs.read(path=input_file_path, mode="rb"))

unstract/sdk1/src/unstract/sdk1/adapters/x2text/llm_whisperer_v2/src/static/json_schema.json

@@ -58,6 +58,7 @@
     "line_splitter_strategy": {
       "type": "string",
       "title": "Line Splitter Strategy",
+      "enum": ["left-priority", "mid-priority", "right-priority"],
       "default":"left-priority",
       "description": "An advanced option for customizing the line splitting process."
     },

unstract/sdk1/tests/test_llm_whisperer_v2_params.py

@@ -0,0 +1,49 @@
+"""Tests for the query params the LLMWhisperer V2 adapter sends."""
+
+import pytest
+from unstract.sdk1.adapters.x2text.llm_whisperer_v2.src.dto import (
+    WhispererRequestParams,
+)
+from unstract.sdk1.adapters.x2text.llm_whisperer_v2.src.helper import LLMWhispererHelper
+
+
+def _params(config: dict) -> dict:
+    return LLMWhispererHelper.get_whisperer_params(
+        config=config, extra_params=WhispererRequestParams()
+    )
+
+
+def test_line_splitter_strategy_from_config() -> None:
+    """The key stored by the adapter's JSON schema is the one that is read."""
+    params = _params({"line_splitter_strategy": "right-priority"})
+
+    assert params["line_splitter_strategy"] == "right-priority"
+
+
+@pytest.mark.parametrize("strategy", ["left-priority", "mid-priority", "right-priority"])
+def test_supported_line_splitter_strategies_pass_through(strategy: str) -> None:
+    """Every value the service accepts reaches it unchanged."""
+    params = _params({"line_splitter_strategy": strategy})
+
+    assert params["line_splitter_strategy"] == strategy
+
+
+@pytest.mark.parametrize(
+    "stored", ["", "  ", "left_priority", "LEFT-PRIORITY", "nonsense"]
+)
+def test_unsupported_line_splitter_strategy_falls_back(
+    stored: str, caplog: pytest.LogCaptureFixture
+) -> None:
+    """A stored value the service would reject keeps the previous behaviour."""
+    params = _params({"line_splitter_strategy": stored})
+
+    assert params["line_splitter_strategy"] == "left-priority"
+    assert "Unsupported line splitter strategy" in caplog.text
+
+
+def test_page_separator_read_under_legacy_config_key() -> None:
+    """Existing configs store the misspelled key but the client kwarg is correct."""
+    params = _params({"page_seperator": "<<< {{page_no}} >>>"})
+
+    assert params["page_separator"] == "<<< {{page_no}} >>>"
+    assert "page_seperator" not in params