feat: sdk scripts#1848
Merged
Merged
Conversation
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
🚧 Files skipped from review as they are similar to previous changes (1)
WalkthroughThis PR adds a standalone Python SDK smoke-test runner for testing OpenRAG SaaS deployments. The implementation includes project configuration, async test harness, CLI entry point with reporting, a suite registry, shared test helpers, and ordered check suites covering settings, models, documents, search, chat, filters, and error scenarios. ChangesPython SDK Smoke-Test Runner
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Possibly related PRs
Suggested reviewers
🚥 Pre-merge checks | ✅ 3 | ❌ 2❌ Failed checks (1 warning, 1 inconclusive)
✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
github-actions
Bot
added
enhancement
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 7
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@scripts/test_scripts/sdk/python/checks/check_chat.py`:
- Around line 48-51: Move the call to _register_chat_cleanup(ctx,
stream.chat_id) so it runs as soon as stream.chat_id is available (i.e.,
immediately after the stream yields/sets chat_id) and before the other
assertions; ensure you only call _register_chat_cleanup after confirming
stream.chat_id is not None (so check/guard for chat_id first), then perform the
remaining asserts on collected and stream.text in the same scope (function/check
in check_chat.py).
In `@scripts/test_scripts/sdk/python/checks/check_documents.py`:
- Around line 68-78: Register the created file for cleanup right after make_doc
by calling register_doc_cleanup(ctx, path) so the document is removed even if
later assertions fail; in delete_by_filename, add that registration before
calling ctx.client.documents.ingest, and then simplify the post-ingest
assertions by removing the unreachable else branch (assume ingest succeeded or
raised), keeping the delete call (ctx.client.documents.delete) and asserting
result.success is True and result.deleted_chunks > 0.
In `@scripts/test_scripts/sdk/python/checks/check_errors.py`:
- Around line 25-30: Narrow the exception handling in invalid_settings_value and
the bogus filter_id checks to catch ValidationError instead of any OpenRAGError:
replace the broad except OpenRAGError with except ValidationError to mark the
test as passing, and if a different OpenRAGError is raised, re-raise or fail the
test (i.e., let non-ValidationError OpenRAGError propagate or explicitly raise
an AssertionError), referencing the invalid_settings_value function and the
bogus filter_id test blocks so the checks only accept HTTP 400/validation
failures.
In `@scripts/test_scripts/sdk/python/checks/check_filters.py`:
- Around line 61-63: scope_setup currently calls ctx.client.documents.ingest
twice without waiting for index readiness, causing filters.filter_id_in_search
and filters.filter_id_in_chat to run against non-searchable docs; after calling
ctx.client.documents.ingest (either after each ingest or once after both), add a
readiness sync step — e.g., call the client’s document-ready or
wait-for-indexing API or poll search until the ingested documents are
discoverable — before returning from scope_setup so the subsequent scoped checks
are deterministic.
- Around line 94-100: The current filter checks (e.g., filter_id_in_search) only
assert that the out-of-scope file ("beta") is not present, which can pass when
the search returns no results; update these functions to also assert that the
expected in-scope file ("alpha") is actually returned. In filter_id_in_search
(and the analogous function around lines 102-117), after asserting
results.results is not None, add a positive assertion that at least one result
has filename == scope["alpha"] (with a clear failure message), and keep the
existing negative assertion for scope["beta"]; this ensures the filter both
includes expected in-scope evidence and excludes out-of-scope items.
In `@scripts/test_scripts/sdk/python/checks/check_models.py`:
- Around line 14-16: The except block catching OpenRAGError in check_models.py
turns all provider errors into Skip; change it to only convert to Skip when the
error indicates credentials not configured by checking e.status_code == 400 and
include the provider in the message, otherwise re-raise the original
OpenRAGError so 401/403/5xx failures surface; update the except OpenRAGError as
e handler to conditionally raise Skip(...) from e only for e.status_code == 400
and use plain raise for other status codes.
In `@scripts/test_scripts/sdk/python/checks/check_settings.py`:
- Around line 13-18: The current "no-op" roundtrip in update_roundtrip can
mutate shared state because chunk_size is set to 1000 when
current.knowledge.chunk_size is falsy; instead, read current via
ctx.client.settings.get() and only include "chunk_size" in the payload passed to
ctx.client.settings.update(...) when current.knowledge.chunk_size is explicitly
set (not None/falsy). Update the logic around variables current and chunk_size
so that if current.knowledge.chunk_size is None/absent you call update with an
empty payload (or omit the chunk_size key) to avoid changing the deployment
config.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: a9b0275a-4def-480b-9970-6f4ae014aa25
⛔ Files ignored due to path filters (1)
scripts/test_scripts/sdk/python/uv.lockis excluded by!**/*.lock
📒 Files selected for processing (15)
scripts/test_scripts/sdk/python/.env.examplescripts/test_scripts/sdk/python/.gitignorescripts/test_scripts/sdk/python/README.mdscripts/test_scripts/sdk/python/checks/__init__.pyscripts/test_scripts/sdk/python/checks/check_chat.pyscripts/test_scripts/sdk/python/checks/check_documents.pyscripts/test_scripts/sdk/python/checks/check_errors.pyscripts/test_scripts/sdk/python/checks/check_filters.pyscripts/test_scripts/sdk/python/checks/check_models.pyscripts/test_scripts/sdk/python/checks/check_search.pyscripts/test_scripts/sdk/python/checks/check_settings.pyscripts/test_scripts/sdk/python/checks/common.pyscripts/test_scripts/sdk/python/harness.pyscripts/test_scripts/sdk/python/main.pyscripts/test_scripts/sdk/python/pyproject.toml
Comment on lines
+48
to
+51
| assert collected, "text_stream yielded no text" | ||
| assert stream.text, "stream.text not accumulated" | ||
| assert stream.chat_id is not None, "stream.chat_id not populated after iteration" | ||
| _register_chat_cleanup(ctx, stream.chat_id) |
Contributor
There was a problem hiding this comment.
Register chat cleanup before assertions in the stream context-manager check.
Cleanup is registered only after assertions. If one of those assertions fails, the conversation can be left behind. Register cleanup immediately after stream.chat_id is available, then run assertions.
Suggested patch
async def stream_context_manager(ctx: Context) -> None:
async with ctx.client.chat.stream(
message="Say 'hello' and nothing else."
) as stream:
collected = ""
async for text in stream.text_stream:
collected += text
- assert collected, "text_stream yielded no text"
- assert stream.text, "stream.text not accumulated"
- assert stream.chat_id is not None, "stream.chat_id not populated after iteration"
- _register_chat_cleanup(ctx, stream.chat_id)
+ assert stream.chat_id is not None, "stream.chat_id not populated after iteration"
+ _register_chat_cleanup(ctx, stream.chat_id)
+ assert collected, "text_stream yielded no text"
+ assert stream.text, "stream.text not accumulated"📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| assert collected, "text_stream yielded no text" | |
| assert stream.text, "stream.text not accumulated" | |
| assert stream.chat_id is not None, "stream.chat_id not populated after iteration" | |
| _register_chat_cleanup(ctx, stream.chat_id) | |
| async def stream_context_manager(ctx: Context) -> None: | |
| async with ctx.client.chat.stream( | |
| message="Say 'hello' and nothing else." | |
| ) as stream: | |
| collected = "" | |
| async for text in stream.text_stream: | |
| collected += text | |
| assert stream.chat_id is not None, "stream.chat_id not populated after iteration" | |
| _register_chat_cleanup(ctx, stream.chat_id) | |
| assert collected, "text_stream yielded no text" | |
| assert stream.text, "stream.text not accumulated" |
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@scripts/test_scripts/sdk/python/checks/check_chat.py` around lines 48 - 51,
Move the call to _register_chat_cleanup(ctx, stream.chat_id) so it runs as soon
as stream.chat_id is available (i.e., immediately after the stream yields/sets
chat_id) and before the other assertions; ensure you only call
_register_chat_cleanup after confirming stream.chat_id is not None (so
check/guard for chat_id first), then perform the remaining asserts on collected
and stream.text in the same scope (function/check in check_chat.py).
Comment on lines
+25
to
+30
| async def invalid_settings_value(ctx: Context) -> None: | ||
| try: | ||
| await ctx.client.settings.update({"chunk_size": -999999}) | ||
| except OpenRAGError: | ||
| return | ||
| raise AssertionError("invalid settings value did not raise OpenRAGError") |
Contributor
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# 1) Show the relevant portion of the file under review
sed -n '1,140p' scripts/test_scripts/sdk/python/checks/check_errors.py | nl -ba | sed -n '1,140p'
# 2) Locate where OpenRAGError / ValidationError are defined or mapped in the repo
echo "---- Searching for exception definitions ----"
rg -n "class ValidationError|ValidationError\(" -S . || true
rg -n "class OpenRAGError|OpenRAGError\(" -S . || true
rg -n "status[_-]?code.*400|400.*ValidationError|ValidationError.*400|404.*NotFoundError|NotFoundError.*404" -S . || true
# 3) Show where the checks are registered / invoked (so we understand harness expectations)
rg -n "CHECKS|invalid_settings_value|bogus_filter_id_in_search|bogus_filter_id_in_chat" -S scripts/test_scripts/sdk/python/checks/check_errors.py || true
# 4) If SDK package exists in repo, search for openrag_sdk.exceptions mapping
rg -n "openrag_sdk\.exceptions" -S . | head -n 50 || trueRepository: langflow-ai/openrag
Length of output: 106
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# 1) Show the relevant portion of the file under review
sed -n '1,140p' scripts/test_scripts/sdk/python/checks/check_errors.py | nl -ba | sed -n '1,140p'
# 2) Locate where OpenRAGError / ValidationError are defined or mapped in the repo
echo "---- Searching for exception definitions ----"
rg -n "class ValidationError|ValidationError\(" -S . || true
rg -n "class OpenRAGError|OpenRAGError\(" -S . || true
rg -n "status[_-]?code.*400|400.*ValidationError|ValidationError.*400|404.*NotFoundError|NotFoundError.*404" -S . || true
# 3) Show where the checks are registered / invoked (so we understand harness expectations)
rg -n "CHECKS|invalid_settings_value|bogus_filter_id_in_search|bogus_filter_id_in_chat" -S scripts/test_scripts/sdk/python/checks/check_errors.py || true
# 4) If SDK package exists in repo, search for openrag_sdk.exceptions mapping
rg -n "openrag_sdk\.exceptions" -S . | head -n 50 || trueRepository: langflow-ai/openrag
Length of output: 106
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# 1) Show relevant portion with line numbers from sed (no `nl`)
echo "---- check_errors.py lines 1-140 ----"
sed -n '1,140p' scripts/test_scripts/sdk/python/checks/check_errors.py | cat -n | sed -n '1,140p'
# 2) Locate where OpenRAGError / ValidationError are defined or mapped in the repo
echo "---- Searching for exception definitions/mappings ----"
rg -n "class ValidationError|ValidationError\(" -S . || true
rg -n "class OpenRAGError|OpenRAGError\(" -S . || true
rg -n "status[_-]?code.*400|400.*ValidationError|ValidationError.*400|404.*NotFoundError|NotFoundError.*404" -S . || true
# 3) Show where the checks are registered / invoked
echo "---- Searching for check functions/registry in check_errors.py ----"
rg -n "CHECKS|invalid_settings_value|bogus_filter_id_in_search|bogus_filter_id_in_chat" -S scripts/test_scripts/sdk/python/checks/check_errors.py || true
# 4) If SDK package exists in repo, search for openrag_sdk.exceptions mapping usage
echo "---- Searching for openrag_sdk.exceptions usage ----"
rg -n "openrag_sdk\.exceptions" -S . | head -n 50 || trueRepository: langflow-ai/openrag
Length of output: 50376
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Small, targeted excerpts (line-numbered where possible without `nl`)
echo "---- sdks/python/openrag_sdk/exceptions.py (1-120) ----"
sed -n '1,160p' sdks/python/openrag_sdk/exceptions.py | cat -n
echo "---- sdks/python/openrag_sdk/client.py (180-280) ----"
sed -n '180,280p' sdks/python/openrag_sdk/client.py | cat -n
echo "---- check_errors.py (20-90) ----"
sed -n '20,95p' scripts/test_scripts/sdk/python/checks/check_errors.py | cat -nRepository: langflow-ai/openrag
Length of output: 7077
Tighten expected exception type in SDK error checks to avoid false positives
invalid_settings_value (25-30) and the bogus filter_id checks (59-66, 69-76) currently treat any OpenRAGError as success, which can mask auth/rate-limit/server failures. The SDK maps HTTP 400 to ValidationError; narrow these checks to ValidationError (and fail if a different OpenRAGError is raised).
Proposed fix
-from openrag_sdk.exceptions import NotFoundError, OpenRAGError
+from openrag_sdk.exceptions import NotFoundError, OpenRAGError, ValidationError
@@
async def invalid_settings_value(ctx: Context) -> None:
try:
await ctx.client.settings.update({"chunk_size": -999999})
- except OpenRAGError:
+ except ValidationError:
return
+ except OpenRAGError as e:
+ raise AssertionError(
+ f"expected ValidationError, got {type(e).__name__}: {e}"
+ ) from e
raise AssertionError("invalid settings value did not raise OpenRAGError")
@@
async def bogus_filter_id_in_search(ctx: Context) -> None:
try:
await ctx.client.search.query(
"anything", filter_id=f"does-not-exist-{uuid.uuid4().hex}"
)
- except OpenRAGError:
+ except ValidationError:
return
+ except OpenRAGError as e:
+ raise AssertionError(
+ f"expected ValidationError, got {type(e).__name__}: {e}"
+ ) from e
raise AssertionError("search with a bogus filter_id did not raise OpenRAGError")
@@
async def bogus_filter_id_in_chat(ctx: Context) -> None:
try:
await ctx.client.chat.create(
message="hi", filter_id=f"does-not-exist-{uuid.uuid4().hex}"
)
- except OpenRAGError:
+ except ValidationError:
return
+ except OpenRAGError as e:
+ raise AssertionError(
+ f"expected ValidationError, got {type(e).__name__}: {e}"
+ ) from e
raise AssertionError("chat with a bogus filter_id did not raise OpenRAGError")🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@scripts/test_scripts/sdk/python/checks/check_errors.py` around lines 25 - 30,
Narrow the exception handling in invalid_settings_value and the bogus filter_id
checks to catch ValidationError instead of any OpenRAGError: replace the broad
except OpenRAGError with except ValidationError to mark the test as passing, and
if a different OpenRAGError is raised, re-raise or fail the test (i.e., let
non-ValidationError OpenRAGError propagate or explicitly raise an
AssertionError), referencing the invalid_settings_value function and the bogus
filter_id test blocks so the checks only accept HTTP 400/validation failures.
Comment on lines
+61
to
+63
| await ctx.client.documents.ingest(file_path=str(alpha)) | ||
| await ctx.client.documents.ingest(file_path=str(beta)) | ||
|
|
Contributor
There was a problem hiding this comment.
scope_setup ingests documents without readiness sync, making scoped checks flaky.
The setup immediately proceeds after ingest. If indexing is async, filters.filter_id_in_search and filters.filter_id_in_chat can run before docs are searchable, causing nondeterministic outcomes.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@scripts/test_scripts/sdk/python/checks/check_filters.py` around lines 61 -
63, scope_setup currently calls ctx.client.documents.ingest twice without
waiting for index readiness, causing filters.filter_id_in_search and
filters.filter_id_in_chat to run against non-searchable docs; after calling
ctx.client.documents.ingest (either after each ingest or once after both), add a
readiness sync step — e.g., call the client’s document-ready or
wait-for-indexing API or poll search until the ingested documents are
discoverable — before returning from scope_setup so the subsequent scoped checks
are deterministic.
Comment on lines
+94
to
+100
| async def filter_id_in_search(ctx: Context) -> None: | ||
| scope = ctx.shared["scope"] | ||
| results = await ctx.client.search.query("animals", filter_id=scope["filter_id"]) | ||
| assert results.results is not None | ||
| leaked = [r.filename for r in results.results if r.filename == scope["beta"]] | ||
| assert not leaked, f"filter leaked: search returned out-of-scope file {leaked}" | ||
|
|
Contributor
There was a problem hiding this comment.
Scoped filter checks can pass vacuously; add positive in-scope assertions.
Both checks only verify that out-of-scope (beta) is absent. They still pass when retrieval is empty or broken. Add assertions that in-scope (alpha) evidence is actually returned.
Suggested patch
async def filter_id_in_search(ctx: Context) -> None:
scope = ctx.shared["scope"]
results = await ctx.client.search.query("animals", filter_id=scope["filter_id"])
assert results.results is not None
+ assert len(results.results) > 0, "scoped search returned no results"
+ filenames = [r.filename for r in results.results]
+ assert scope["alpha"] in filenames, (
+ f"scoped search did not return expected in-scope file {scope['alpha']}: {filenames}"
+ )
- leaked = [r.filename for r in results.results if r.filename == scope["beta"]]
+ leaked = [name for name in filenames if name == scope["beta"]]
assert not leaked, f"filter leaked: search returned out-of-scope file {leaked}"
@@
async def filter_id_in_chat(ctx: Context) -> None:
scope = ctx.shared["scope"]
response = await ctx.client.chat.create(
message="What animals appear in these documents?",
filter_id=scope["filter_id"],
@@
- assert response.sources is not None
+ assert response.sources is not None, "chat returned no sources collection"
+ assert len(response.sources) > 0, "scoped chat returned no sources"
source_names = [s.filename for s in response.sources]
+ assert scope["alpha"] in source_names, (
+ f"scoped chat did not cite expected in-scope file {scope['alpha']}: {source_names}"
+ )
assert scope["beta"] not in source_names, (
f"filter leaked: chat cited out-of-scope file in {source_names}"
)Also applies to: 102-117
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@scripts/test_scripts/sdk/python/checks/check_filters.py` around lines 94 -
100, The current filter checks (e.g., filter_id_in_search) only assert that the
out-of-scope file ("beta") is not present, which can pass when the search
returns no results; update these functions to also assert that the expected
in-scope file ("alpha") is actually returned. In filter_id_in_search (and the
analogous function around lines 102-117), after asserting results.results is not
None, add a positive assertion that at least one result has filename ==
scope["alpha"] (with a clear failure message), and keep the existing negative
assertion for scope["beta"]; this ensures the filter both includes expected
in-scope evidence and excludes out-of-scope items.
Comment on lines
+14
to
+16
| except OpenRAGError as e: | ||
| # SaaS deployments typically enable only a subset of providers. | ||
| raise Skip(f"provider '{provider}' not available: {e}") from e |
Contributor
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Locate the target file and show the relevant section
FILE="scripts/test_scripts/sdk/python/checks/check_models.py"
echo "=== Target snippet ($FILE) ==="
nl -ba "$FILE" | sed -n '1,120p'
# Search for OpenRAGError definition and any mention of status_code/provider-unavailable semantics
echo
echo "=== Search for OpenRAGError definition ==="
rg -n "class OpenRAGError|OpenRAGError\(" -S . | head -n 50
# If OpenRAGError is in a Python package, locate it
echo
echo "=== Search for 'status_code' on OpenRAGError ==="
rg -n "OpenRAGError.*status_code|status_code.*OpenRAGError|def .*status_code" -S . | head -n 100
# Find all occurrences of 'provider.*not available' or Skip usage patterns
echo
echo "=== Search for Skip(...) related to provider availability ==="
rg -n "Skip\\(.*provider|provider.*not available|not available: \\{e\\}" -S . | head -n 100
# Show any other Skip conversions from OpenRAGError near similar code
echo
echo "=== Search for 'except OpenRAGError' blocks ==="
rg -n "except OpenRAGError" -S . | head -n 100Repository: langflow-ai/openrag
Length of output: 204
🏁 Script executed:
#!/bin/bash
set -euo pipefail
FILE="scripts/test_scripts/sdk/python/checks/check_models.py"
echo "=== Target snippet ($FILE) ==="
# Prefer cat -n (available in most environments); fallback to awk for line numbers
if command -v cat >/dev/null 2>&1; then
cat -n "$FILE" | sed -n '1,160p'
else
awk 'NR<=160{printf "%5d\t%s\n", NR, $0}' "$FILE"
fi
echo
echo "=== Search for OpenRAGError definition and constructor usage ==="
rg -n "class OpenRAGError|def __init__\(.*OpenRAGError|OpenRAGError\(" -S . | head -n 80
echo
echo "=== Search for status_code references near OpenRAGError ==="
rg -n "status_code" -S . | head -n 120
echo
echo "=== Search for provider-not-available / Skip patterns ==="
rg -n "Skip\\(.*provider|provider.*not available|not available: \\{e\\}|except OpenRAGError" -S scripts/test_scripts/sdk/python/checks . | head -n 120Repository: langflow-ai/openrag
Length of output: 50376
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== OpenRAGError exception definition ==="
FILE="sdks/python/openrag_sdk/exceptions.py"
rg -n "class OpenRAGError" "$FILE"
start=$(rg -n "class OpenRAGError" "$FILE" | head -n1 | cut -d: -f1)
end=$((start+120))
sed -n "${start},${end}p" "$FILE" | cat -n
echo
echo "=== Where OpenRAGError is raised with status_code ==="
FILE="sdks/python/openrag_sdk/client.py"
rg -n "raise OpenRAGError\\(.*status_code|raise OpenRAGError" "$FILE"
# show a small window around the first match
line=$(rg -n "raise OpenRAGError\\(.*status_code|raise OpenRAGError" "$FILE" | head -n1 | cut -d: -f1)
start=$((line-40))
end=$((line+60))
sed -n "${start},${end}p" "$FILE" | cat -n
echo
echo "=== check_errors.py existing handling of OpenRAGError ==="
FILE="scripts/test_scripts/sdk/python/checks/check_errors.py"
if [ -f "$FILE" ]; then
rg -n "except OpenRAGError" "$FILE"
# show around each match
rg -n "except OpenRAGError" "$FILE" | cut -d: -f1 | while read -r ln; do
start=$((ln-10)); end=$((ln+25));
echo "--- context around line $ln ---"
sed -n "${start},${end}p" "$FILE" | cat -n
echo
done
else
echo "File not found: $FILE"
fiRepository: langflow-ai/openrag
Length of output: 7839
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Search for endpoint/handler for listing models per provider
rg -n "models\\.list|list\\(provider\\)|/models|models\\b.*provider|provider.*models" src sdks plugins scripts tests -S --hidden 2>/dev/null | head -n 200
# Specifically find where "provider" is used in a models route
rg -n "provider.*models|models.*provider" src -S --hidden 2>/dev/null | head -n 200
# Find routes mentioning "models" and "provider"
rg -n "APIRouter|router\\.|@.*models|/models" src -S --hidden 2>/dev/null | head -n 200Repository: langflow-ai/openrag
Length of output: 13291
🏁 Script executed:
#!/bin/bash
set -euo pipefail
FILE="src/api/v1/models.py"
echo "=== File header (first 140 lines) ==="
sed -n '1,140p' "$FILE" | cat -n
echo
echo "=== Search for list_models_endpoint and provider-not-available handling ==="
rg -n "list_models_endpoint|_fetch_models|enabled|not available|unavailable|disabled|Skip|HTTPException|status_code|404|400|403" "$FILE" -S
# Print the main endpoint function region
# (Try to cover from near list_models_endpoint declaration down ~120 lines)
line=$(rg -n "def list_models_endpoint" "$FILE" | head -n1 | cut -d: -f1)
start=$((line-20)); end=$((line+140))
echo
echo "=== Around list_models_endpoint (lines $start-$end) ==="
sed -n "${start},${end}p" "$FILE" | cat -n
# Print the _fetch_models helper region
line=$(rg -n "async def _fetch_models" "$FILE" | head -n1 | cut -d: -f1)
start=$((line-20)); end=$((line+160))
echo
echo "=== Around _fetch_models (lines $start-$end) ==="
sed -n "${start},${end}p" "$FILE" | cat -nRepository: langflow-ai/openrag
Length of output: 11627
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Inspect sdks/python/openrag_sdk/models.py list() implementation ==="
FILE="sdks/python/openrag_sdk/models.py"
rg -n "class Models|async def list|models\\.list" "$FILE"
line=$(rg -n "async def list" "$FILE" | head -n1 | cut -d: -f1)
start=$((line-40)); end=$((line+140))
sed -n "${start},${end}p" "$FILE" | cat -n
echo
echo "=== Inspect scripts/test_scripts/sdk/python/README.md (models section) ==="
FILE="scripts/test_scripts/sdk/python/README.md"
# print around the models row/description
rg -n "models\\b|unavailable providers" "$FILE" | head -n 50
line=$(rg -n "\\|\\s*`models`|models\\s+\\|" "$FILE" | head -n1 | cut -d: -f1 || true)
if [ -n "${line:-}" ]; then
start=$((line-10)); end=$((line+20))
sed -n "${start},${end}p" "$FILE" | cat -n
fiRepository: langflow-ai/openrag
Length of output: 172
🏁 Script executed:
#!/bin/bash
set -euo pipefail
FILE="sdks/python/openrag_sdk/models.py"
echo "=== File exists? ==="
ls -l "$FILE" || true
echo
echo "=== Search within models.py for v1/models or list method ==="
rg -n "\"/v1/models/\\{provider\\}\"|/api/v1/models/\\{provider\\}|models\\.list\\(|def list\\b|async def list\\b|list_models|list_\\w*models" "$FILE" -S
echo
echo "=== Print relevant region around any list_...models function ==="
# Try common candidates
for pat in "list_models" "list_" "models" ; do
line=$(rg -n "${pat}" "$FILE" | head -n1 | cut -d: -f1 || true)
if [ -n "${line:-}" ]; then
start=$((line-40)); end=$((line+160));
echo "--- candidate pattern '${pat}' at line ${line}: showing ${start}-${end} ---"
sed -n "${start},${end}p" "$FILE" | cat -n
break
fi
done
echo
echo "=== Show top of models.py (first 260 lines) ==="
sed -n '1,260p' "$FILE" | cat -nRepository: langflow-ai/openrag
Length of output: 229
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Search for /api/v1/models/{provider} usage in Python SDK ==="
rg -n "\"/api/v1/models/\\{provider\\}\"|/api/v1/models/\\{provider\\}|models/\\{provider\\}\" -S sdks/python/openrag_sdk | head -n 50
echo
echo "=== Search for list_models method implementation ==="
rg -n "list_models|models\\.list|def list\\(|async def list\\(|class Models" -S sdks/python/openrag_sdk | head -n 100
echo
echo "=== Inspect sdks/python/openrag_sdk/client.py around models listing ==="
FILE="sdks/python/openrag_sdk/client.py"
# find the line that contains the endpoint string (best effort)
line=$(rg -n "/api/v1/models/\\{provider\\}" "$FILE" | head -n1 | cut -d: -f1 || true)
if [ -n "${line:-}" ]; then
start=$((line-60)); end=$((line+80))
sed -n "${start},${end}p" "$FILE" | cat -n
else
# fallback: show the section containing "models" methods
rg -n "class Models|def models|async def list|models\\b" "$FILE" | head -n 30
fi
echo
echo "=== Inspect tests/integration/sdk/test_models.py ==="
FILE="tests/integration/sdk/test_models.py"
sed -n '1,200p' "$FILE" | cat -n
echo
echo "=== Inspect scripts/test_scripts/sdk/python/README.md models row ==="
FILE="scripts/test_scripts/sdk/python/README.md"
rg -n "models`|unavailable providers|models\\b.*unavailable" "$FILE" -n -S || trueRepository: langflow-ai/openrag
Length of output: 208
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Search for models endpoint string in Python SDK ==="
rg -n "/api/v1/models/" -S sdks/python/openrag_sdk | head -n 50
echo
echo "=== Search for Models client class/methods in Python SDK ==="
rg -n "class Models|def list|def.*models|async def" -S sdks/python/openrag_sdk | head -n 120
echo
echo "=== Inspect sdks/python/openrag_sdk/client.py around first models endpoint usage ==="
FILE="sdks/python/openrag_sdk/client.py"
line=$(rg -n "/api/v1/models/" "$FILE" -S | head -n1 | cut -d: -f1 || true)
if [ -n "${line:-}" ]; then
start=$((line-80)); end=$((line+80))
sed -n "${start},${end}p" "$FILE" | cat -n
else
echo "No /api/v1/models/ usage found in $FILE"
fi
echo
echo "=== Inspect sdks/python/openrag_sdk/client.py Models method block (fallback) ==="
rg -n "class Models|models\\b|list\\(" "$FILE" -S | head -n 40
echo
echo "=== Inspect tests/integration/sdk/test_models.py ==="
FILE="tests/integration/sdk/test_models.py"
sed -n '1,200p' "$FILE" | cat -n
echo
echo "=== Inspect scripts/test_scripts/sdk/python/README.md around models row/description ==="
FILE="scripts/test_scripts/sdk/python/README.md"
rg -n "models\\b|unavailable providers" -S "$FILE" | head -n 80
line=$(rg -n "models\\b" "$FILE" -S | head -n1 | cut -d: -f1 || true)
if [ -n "${line:-}" ]; then
start=$((line-10)); end=$((line+30))
sed -n "${start},${end}p" "$FILE" | cat -n
fiRepository: langflow-ai/openrag
Length of output: 4722
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# 1) Inspect sdks/python/openrag_sdk/client.py around ModelsClient.list and request error handling
FILE="sdks/python/openrag_sdk/client.py"
echo "=== client.py: first 120 lines ==="
sed -n '1,120p' "$FILE" | cat -n
echo
echo "=== client.py: around ModelsClient.list (search hit line) ==="
line=$(rg -n "class ModelsClient|async def list\\(self, provider" "$FILE" | head -n1 | cut -d: -f1)
start=1
end=$((line+120))
sed -n "${start},${end}p" "$FILE" | cat -n
echo
echo "=== client.py: around _request/_handle_error mapping ==="
# show section containing _handle_error (already saw earlier, but re-show bounded)
line=$(rg -n "def _handle_error" "$FILE" | head -n1 | cut -d: -f1)
start=$((line-40)); if [ $start -lt 1 ]; then start=1; fi
end=$((line+120))
sed -n "${start},${end}p" "$FILE" | cat -n
# 2) Integration test expectations
echo
echo "=== tests/integration/sdk/test_models.py ==="
sed -n '1,120p' tests/integration/sdk/test_models.py | cat -n
# 3) Test-script README models row
echo
echo "=== scripts/test_scripts/sdk/python/README.md (models row/description) ==="
FILE="scripts/test_scripts/sdk/python/README.md"
# print lines around the "models" row
rg -n "\|\s*`models`\s*\|" "$FILE" -n || true
line=$(rg -n "\|\s*`models`\s*\|" "$FILE" | head -n1 | cut -d: -f1 || true)
if [ -n "${line:-}" ]; then
start=$((line-5)); if [ $start -lt 1 ]; then start=1; fi
end=$((line+15))
sed -n "${start},${end}p" "$FILE" | cat -n
else
# fallback: show around first "models" mention
line=$(rg -n "models.*unavailable|models.*skipped|models.*provider" "$FILE" | head -n1 | cut -d: -f1 || true)
if [ -n "${line:-}" ]; then
start=$((line-5)); if [ $start -lt 1 ]; then start=1; fi
end=$((line+15))
sed -n "${start},${end}p" "$FILE" | cat -n
fi
fiRepository: langflow-ai/openrag
Length of output: 15715
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# 1) Inspect src/services/models_service.py for error handling / status mapping patterns
FILE="src/services/models_service.py"
echo "=== models_service.py: search for 'status_code' and HTTPException / JSONResponse ==="
rg -n "status_code|HTTPException|JSONResponse|raise .*Exception\\(|not configured|not available|unavailable|disabled" "$FILE" -S || true
echo
echo "=== models_service.py: search for 400/404/403/500 literals near provider fetches ==="
rg -n "status_code\\s*=\\s*(400|404|403|500)|\\b400\\b|\\b404\\b|\\b403\\b|\\b500\\b" "$FILE" -S | head -n 80
echo
echo "=== src/api/v1/models.py: confirm missing-credential cases only return 400 ==="
rg -n "status_code=400|not configured|Set it in Settings" src/api/v1/models.py -S
# 2) Inspect any router/mcp mapping that could return 404 for missing provider models endpoint
echo
echo "=== Search for '/v1/models/{provider}' in routes to see if 404 is possible elsewhere ==="
rg -n "/v1/models/\\{provider\\}|/api/v1/models/\\{provider\\}|VALID_PROVIDERS|Not Found" src -S | head -n 80Repository: langflow-ai/openrag
Length of output: 1476
Restrict “provider not available” SKIPs to HTTP 400 only in check_models.py.
check_models.py currently converts any OpenRAGError into Skip, which can mask real failures (e.g., 401/403 auth errors and >=500 server/upstream errors) as “provider not available.” The backend /v1/models/{provider} uses 400 for “provider credentials not configured” cases and 500 for other failures—so skip should allowlist only e.status_code == 400 and re-raise otherwise.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@scripts/test_scripts/sdk/python/checks/check_models.py` around lines 14 - 16,
The except block catching OpenRAGError in check_models.py turns all provider
errors into Skip; change it to only convert to Skip when the error indicates
credentials not configured by checking e.status_code == 400 and include the
provider in the message, otherwise re-raise the original OpenRAGError so
401/403/5xx failures surface; update the except OpenRAGError as e handler to
conditionally raise Skip(...) from e only for e.status_code == 400 and use plain
raise for other status codes.
Comment on lines
+13
to
+18
| # Safe no-op: re-set the current chunk_size and read it back, so the | ||
| # deployment's configuration is never actually changed. | ||
| current = await ctx.client.settings.get() | ||
| chunk_size = current.knowledge.chunk_size or 1000 | ||
|
|
||
| result = await ctx.client.settings.update({"chunk_size": chunk_size}) |
Contributor
There was a problem hiding this comment.
update_roundtrip is not actually a no-op when chunk_size is unset.
Line 16 (current.knowledge.chunk_size or 1000) can overwrite instance config (e.g., None/0 becomes 1000), which contradicts the non-destructive intent in Lines 13-15 and can mutate shared SaaS state.
Proposed fix
-from harness import Check, Context
+from harness import Check, Context, Skip
@@
async def update_roundtrip(ctx: Context) -> None:
@@
- chunk_size = current.knowledge.chunk_size or 1000
+ chunk_size = current.knowledge.chunk_size
+ if chunk_size is None:
+ raise Skip("knowledge.chunk_size is unset; skipping destructive roundtrip")
result = await ctx.client.settings.update({"chunk_size": chunk_size})🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@scripts/test_scripts/sdk/python/checks/check_settings.py` around lines 13 -
18, The current "no-op" roundtrip in update_roundtrip can mutate shared state
because chunk_size is set to 1000 when current.knowledge.chunk_size is falsy;
instead, read current via ctx.client.settings.get() and only include
"chunk_size" in the payload passed to ctx.client.settings.update(...) when
current.knowledge.chunk_size is explicitly set (not None/falsy). Update the
logic around variables current and chunk_size so that if
current.knowledge.chunk_size is None/absent you call update with an empty
payload (or omit the chunk_size key) to avoid changing the deployment config.
Contributor
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
scripts/test_scripts/sdk/python/main.py (1)
69-74:⚠️ Potential issue | 🟡 Minor | ⚡ Quick winValidate timeout is strictly positive at parse time.
--timeoutcurrently accepts0or negative values (Line 69), which are propagated intoConfig(Line 125) and then into the SDK client timeout. Fail fast in CLI parsing with a> 0check to avoid runtime-side failures and clearer operator feedback.Suggested patch
+def _positive_float(value: str) -> float: + v = float(value) + if v <= 0: + raise argparse.ArgumentTypeError("--timeout must be > 0") + return v + def parse_config(argv: list[str]) -> Config: @@ parser.add_argument( "--timeout", - type=float, + type=_positive_float, default=120.0, help="Per-request timeout in seconds (default: 120; streaming chat " "routinely exceeds the SDK's 30s default)", )Also applies to: 125-125
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@scripts/test_scripts/sdk/python/main.py` around lines 69 - 74, The CLI currently allows non-positive timeouts; update the "--timeout" parsing to enforce a strictly positive value by adding a validator (e.g., positive_float) and using it as the add_argument type or by checking args.timeout immediately after parse_args and calling parser.error if <= 0; also add the same guard when constructing Config (the Config constructor / assignment at the Config creation site) to raise a ValueError if timeout <= 0 so invalid values cannot propagate into the SDK client.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@scripts/test_scripts/sdk/python/main.py`:
- Around line 69-74: The CLI currently allows non-positive timeouts; update the
"--timeout" parsing to enforce a strictly positive value by adding a validator
(e.g., positive_float) and using it as the add_argument type or by checking
args.timeout immediately after parse_args and calling parser.error if <= 0; also
add the same guard when constructing Config (the Config constructor / assignment
at the Config creation site) to raise a ValueError if timeout <= 0 so invalid
values cannot propagate into the SDK client.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 6d322c9b-0bb0-4265-9643-df671d0c05ef
📒 Files selected for processing (1)
scripts/test_scripts/sdk/python/main.py
* FIxed no langflow ingestion * style: ruff autofix (auto) * add fallback for mime types * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Standalone uv project under scripts/test_scripts/sdk/python that exercises every openrag-sdk functionality (settings, models, documents, search, chat, knowledge filters, error handling) against a remote instance using IBM SaaS header auth (X-Username / X-Api-Key), with pass/fail/skip reporting to console, markdown, JSON, and a run log.
github-actions
Bot
added
backend
github-actions
Bot
added
enhancement
github-actions
Bot
added
enhancement
github-actions
Bot
added
enhancement
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (2)
scripts/test_scripts/sdk/python/checks/check_filters.py (2)
60-62:⚠️ Potential issue | 🟠 Major | ⚡ Quick winAdd readiness synchronization after ingestion in
scope_setup.Line 60-Line 61 ingest files and immediately proceed; if indexing is async, downstream scoped checks become flaky/non-deterministic.
#!/bin/bash set -euo pipefail echo "== Inspect ingest API for wait/readiness options ==" fd -i 'documents.py' | xargs -I{} sh -c ' echo "--- {} ---" rg -n -C3 "async def ingest|def ingest|wait|ready|poll|timeout" "{}" ' echo echo "== Inspect smoke checks for readiness patterns after ingest ==" rg -n --type=py -C2 "documents\\.ingest\\(|wait|poll|ready|search\\.query\\(" scripts/test_scripts/sdk/python/checksExpected verification result: either
documents.ingestis non-blocking (confirming this issue), or it blocks by default (then this can be downgraded).🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@scripts/test_scripts/sdk/python/checks/check_filters.py` around lines 60 - 62, The scope_setup currently calls ctx.client.documents.ingest for alpha and beta and proceeds immediately; add a readiness synchronization step after each ingest so downstream scoped checks aren't flaky: after calling ctx.client.documents.ingest(...) in scope_setup, poll for document availability (e.g., repeatedly call the search/query API or any provided readiness API until the newly ingested document(s) return in results or a timeout elapses) and only then return from scope_setup; reference ctx.client.documents.ingest and the project’s search/query or indexing readiness method (or implement a small poll loop using ctx.client.documents.search/query) to confirm indexing completion before continuing.
93-99:⚠️ Potential issue | 🟠 Major | ⚡ Quick winScoped filter assertions are vacuous without positive in-scope checks.
Line 97-Line 98 and Line 114-Line 116 only prove
betais absent. These checks still pass when retrieval is empty/broken. Assert thatalphais present in both search and chat sources.Suggested patch
async def filter_id_in_search(ctx: Context) -> None: scope = ctx.shared["scope"] results = await ctx.client.search.query("animals", filter_id=scope["filter_id"]) assert results.results is not None - leaked = [r.filename for r in results.results if r.filename == scope["beta"]] + filenames = [r.filename for r in results.results] + assert scope["alpha"] in filenames, ( + f"scoped search did not return expected in-scope file {scope['alpha']}: {filenames}" + ) + leaked = [name for name in filenames if name == scope["beta"]] assert not leaked, f"filter leaked: search returned out-of-scope file {leaked}" @@ async def filter_id_in_chat(ctx: Context) -> None: @@ assert response.sources is not None source_names = [s.filename for s in response.sources] + assert scope["alpha"] in source_names, ( + f"scoped chat did not cite expected in-scope file {scope['alpha']}: {source_names}" + ) assert scope["beta"] not in source_names, ( f"filter leaked: chat cited out-of-scope file in {source_names}" )Also applies to: 112-116
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@scripts/test_scripts/sdk/python/checks/check_filters.py` around lines 93 - 99, The tests only assert that the out-of-scope file (scope["beta"]) is absent, which allows a broken/empty retrieval to pass; update the scoped-filter tests to also assert that the in-scope file (scope["alpha"]) is actually returned. In filter_id_in_search, after fetching results via ctx.client.search.query("animals", filter_id=scope["filter_id"]), add a positive assertion that at least one result.filename equals scope["alpha"] (e.g., check any(r.filename == scope["alpha"]) and assert True with a helpful message). Do the analogous change in the chat-related test (the function handling chat/query with filter_id) to assert the chat response includes or references scope["alpha"] (check response content/messages for scope["alpha"] or an identifier indicating the alpha file) so both search and chat tests verify presence of in-scope entries.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Duplicate comments:
In `@scripts/test_scripts/sdk/python/checks/check_filters.py`:
- Around line 60-62: The scope_setup currently calls ctx.client.documents.ingest
for alpha and beta and proceeds immediately; add a readiness synchronization
step after each ingest so downstream scoped checks aren't flaky: after calling
ctx.client.documents.ingest(...) in scope_setup, poll for document availability
(e.g., repeatedly call the search/query API or any provided readiness API until
the newly ingested document(s) return in results or a timeout elapses) and only
then return from scope_setup; reference ctx.client.documents.ingest and the
project’s search/query or indexing readiness method (or implement a small poll
loop using ctx.client.documents.search/query) to confirm indexing completion
before continuing.
- Around line 93-99: The tests only assert that the out-of-scope file
(scope["beta"]) is absent, which allows a broken/empty retrieval to pass; update
the scoped-filter tests to also assert that the in-scope file (scope["alpha"])
is actually returned. In filter_id_in_search, after fetching results via
ctx.client.search.query("animals", filter_id=scope["filter_id"]), add a positive
assertion that at least one result.filename equals scope["alpha"] (e.g., check
any(r.filename == scope["alpha"]) and assert True with a helpful message). Do
the analogous change in the chat-related test (the function handling chat/query
with filter_id) to assert the chat response includes or references
scope["alpha"] (check response content/messages for scope["alpha"] or an
identifier indicating the alpha file) so both search and chat tests verify
presence of in-scope entries.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: e57f9d82-d7d9-4437-a311-12bd664b83c0
📒 Files selected for processing (10)
scripts/test_scripts/sdk/python/README.mdscripts/test_scripts/sdk/python/checks/check_chat.pyscripts/test_scripts/sdk/python/checks/check_documents.pyscripts/test_scripts/sdk/python/checks/check_errors.pyscripts/test_scripts/sdk/python/checks/check_filters.pyscripts/test_scripts/sdk/python/checks/check_models.pyscripts/test_scripts/sdk/python/checks/check_search.pyscripts/test_scripts/sdk/python/checks/common.pyscripts/test_scripts/sdk/python/harness.pyscripts/test_scripts/sdk/python/main.py
✅ Files skipped from review due to trivial changes (1)
- scripts/test_scripts/sdk/python/README.md
🚧 Files skipped from review as they are similar to previous changes (7)
- scripts/test_scripts/sdk/python/checks/check_models.py
- scripts/test_scripts/sdk/python/checks/check_errors.py
- scripts/test_scripts/sdk/python/checks/check_search.py
- scripts/test_scripts/sdk/python/checks/check_documents.py
- scripts/test_scripts/sdk/python/main.py
- scripts/test_scripts/sdk/python/checks/check_chat.py
- scripts/test_scripts/sdk/python/harness.py
mfortman11
approved these changes
Jun 12, 2026
github-actions
Bot
added
enhancement
…f sensitive information' Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
github-actions
Bot
added
enhancement
ricofurtado
pushed a commit
that referenced
this pull request
Jun 12, 2026
* sdk scripts * Update main.py * fix: put correct file type in langflow-less ingestion (#1842) * FIxed no langflow ingestion * style: ruff autofix (auto) * add fallback for mime types * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> * feat: add Python SDK smoke-test suite for remote/IBM SaaS deployments Standalone uv project under scripts/test_scripts/sdk/python that exercises every openrag-sdk functionality (settings, models, documents, search, chat, knowledge filters, error handling) against a remote instance using IBM SaaS header auth (X-Username / X-Api-Key), with pass/fail/skip reporting to console, markdown, JSON, and a run log. * style: ruff autofix (auto) * Potential fix for pull request finding 'CodeQL / Clear-text logging of sensitive information' Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com> --------- Co-authored-by: Lucas Oliveira <[email protected]> Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
ricofurtado
added a commit
that referenced
this pull request
Jun 12, 2026
…dingly (#1851) * fix: handle document processing errors and update task statuses accordingly * fix: add OCR handling for raster images during file ingestion * style: ruff autofix (auto) * fix: google to support WEBHOOK_BASE_URL (#1849) * Update connector.py * update env in settings * fix webhook url of sharepoint * feat: crd report scaleddown phase status (#1846) * crd report scaleddown phase status * improve status message * chore: release 0.4.0.dev0 of sdks (#1847) * release 0.4.1 of sdks * dev 0 * dev 0 * 0.4.0-dev0 * feat: sdk scripts (#1848) * sdk scripts * Update main.py * fix: put correct file type in langflow-less ingestion (#1842) * FIxed no langflow ingestion * style: ruff autofix (auto) * add fallback for mime types * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> * feat: add Python SDK smoke-test suite for remote/IBM SaaS deployments Standalone uv project under scripts/test_scripts/sdk/python that exercises every openrag-sdk functionality (settings, models, documents, search, chat, knowledge filters, error handling) against a remote instance using IBM SaaS header auth (X-Username / X-Api-Key), with pass/fail/skip reporting to console, markdown, JSON, and a run log. * style: ruff autofix (auto) * Potential fix for pull request finding 'CodeQL / Clear-text logging of sensitive information' Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com> --------- Co-authored-by: Lucas Oliveira <[email protected]> Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com> * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Edwin Jose <[email protected]> Co-authored-by: ming <[email protected]> Co-authored-by: Lucas Oliveira <[email protected]> Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
Vchen7629
pushed a commit
to Vchen7629/openrag
that referenced
this pull request
Jun 15, 2026
* sdk scripts * Update main.py * fix: put correct file type in langflow-less ingestion (langflow-ai#1842) * FIxed no langflow ingestion * style: ruff autofix (auto) * add fallback for mime types * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> * feat: add Python SDK smoke-test suite for remote/IBM SaaS deployments Standalone uv project under scripts/test_scripts/sdk/python that exercises every openrag-sdk functionality (settings, models, documents, search, chat, knowledge filters, error handling) against a remote instance using IBM SaaS header auth (X-Username / X-Api-Key), with pass/fail/skip reporting to console, markdown, JSON, and a run log. * style: ruff autofix (auto) * Potential fix for pull request finding 'CodeQL / Clear-text logging of sensitive information' Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com> --------- Co-authored-by: Lucas Oliveira <[email protected]> Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
Vchen7629
pushed a commit
to Vchen7629/openrag
that referenced
this pull request
Jun 15, 2026
…dingly (langflow-ai#1851) * fix: handle document processing errors and update task statuses accordingly * fix: add OCR handling for raster images during file ingestion * style: ruff autofix (auto) * fix: google to support WEBHOOK_BASE_URL (langflow-ai#1849) * Update connector.py * update env in settings * fix webhook url of sharepoint * feat: crd report scaleddown phase status (langflow-ai#1846) * crd report scaleddown phase status * improve status message * chore: release 0.4.0.dev0 of sdks (langflow-ai#1847) * release 0.4.1 of sdks * dev 0 * dev 0 * 0.4.0-dev0 * feat: sdk scripts (langflow-ai#1848) * sdk scripts * Update main.py * fix: put correct file type in langflow-less ingestion (langflow-ai#1842) * FIxed no langflow ingestion * style: ruff autofix (auto) * add fallback for mime types * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> * feat: add Python SDK smoke-test suite for remote/IBM SaaS deployments Standalone uv project under scripts/test_scripts/sdk/python that exercises every openrag-sdk functionality (settings, models, documents, search, chat, knowledge filters, error handling) against a remote instance using IBM SaaS header auth (X-Username / X-Api-Key), with pass/fail/skip reporting to console, markdown, JSON, and a run log. * style: ruff autofix (auto) * Potential fix for pull request finding 'CodeQL / Clear-text logging of sensitive information' Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com> --------- Co-authored-by: Lucas Oliveira <[email protected]> Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com> * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Edwin Jose <[email protected]> Co-authored-by: ming <[email protected]> Co-authored-by: Lucas Oliveira <[email protected]> Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
ricofurtado
pushed a commit
that referenced
this pull request
Jun 15, 2026
* sdk scripts * Update main.py * fix: put correct file type in langflow-less ingestion (#1842) * FIxed no langflow ingestion * style: ruff autofix (auto) * add fallback for mime types * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> * feat: add Python SDK smoke-test suite for remote/IBM SaaS deployments Standalone uv project under scripts/test_scripts/sdk/python that exercises every openrag-sdk functionality (settings, models, documents, search, chat, knowledge filters, error handling) against a remote instance using IBM SaaS header auth (X-Username / X-Api-Key), with pass/fail/skip reporting to console, markdown, JSON, and a run log. * style: ruff autofix (auto) * Potential fix for pull request finding 'CodeQL / Clear-text logging of sensitive information' Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com> --------- Co-authored-by: Lucas Oliveira <[email protected]> Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
ricofurtado
added a commit
that referenced
this pull request
Jun 15, 2026
…dingly (#1851) * fix: handle document processing errors and update task statuses accordingly * fix: add OCR handling for raster images during file ingestion * style: ruff autofix (auto) * fix: google to support WEBHOOK_BASE_URL (#1849) * Update connector.py * update env in settings * fix webhook url of sharepoint * feat: crd report scaleddown phase status (#1846) * crd report scaleddown phase status * improve status message * chore: release 0.4.0.dev0 of sdks (#1847) * release 0.4.1 of sdks * dev 0 * dev 0 * 0.4.0-dev0 * feat: sdk scripts (#1848) * sdk scripts * Update main.py * fix: put correct file type in langflow-less ingestion (#1842) * FIxed no langflow ingestion * style: ruff autofix (auto) * add fallback for mime types * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> * feat: add Python SDK smoke-test suite for remote/IBM SaaS deployments Standalone uv project under scripts/test_scripts/sdk/python that exercises every openrag-sdk functionality (settings, models, documents, search, chat, knowledge filters, error handling) against a remote instance using IBM SaaS header auth (X-Username / X-Api-Key), with pass/fail/skip reporting to console, markdown, JSON, and a run log. * style: ruff autofix (auto) * Potential fix for pull request finding 'CodeQL / Clear-text logging of sensitive information' Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com> --------- Co-authored-by: Lucas Oliveira <[email protected]> Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com> * style: ruff autofix (auto) --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Edwin Jose <[email protected]> Co-authored-by: ming <[email protected]> Co-authored-by: Lucas Oliveira <[email protected]> Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
feat: sdk scripts
Summary by CodeRabbit
New Features
Tests
Documentation