v0.46.0: security, Docker UID/GID, model discovery, i18n, cancel fix
* fix: decode HTML entities before markdown processing + zh/zh-Hant translations (#239)
Adds decode() helper in renderMd() to fix double-escaping of HTML entities
from LLM output (e.g. <code> becoming &lt;code&gt; instead
of rendering). XSS-safe: decode runs before esc(), only 5 entity patterns.
Also adds 40+ missing zh (Simplified Chinese) translation keys and a new
zh-Hant (Traditional Chinese) locale with 163 keys.
Fix applied: removed duplicate settings_label_notifications key in both
zh and zh-Hant locales.
Fixes #240
* fix: restore custom model list discovery with config api key (#238)
get_available_models() now reads api_key from config.yaml before env vars:
1. model.api_key
2. providers.<active>.api_key / providers.custom.api_key
3. env var fallbacks (HERMES_API_KEY, OPENAI_API_KEY, etc.)
Also adds OpenAI/Python User-Agent header and a regression test covering
authenticated /v1/models discovery.
Fixes users with LM Studio / Ollama custom endpoints configured in
config.yaml whose model picker silently collapsed to the default model.
* feat: Docker UID/GID matching to avoid root-owned .hermes files (#237)
Adds docker_init.bash with hermeswebuitoo/hermeswebui user pattern so
container files match the host user UID/GID. Prevents .hermes volume
mounts from being owned by root when using a non-root host user.
Configure via WANTED_UID and WANTED_GID env vars (default 1000/1000).
Readme updated with setup instructions.
Fix applied: removed duplicate WANTED_GID=1000 line in docker-compose.yml
that was overriding the ${GID:-1000} variable expansion.
* security: redact credentials from API responses and fix credential file permissions (#243)
Adds response-layer credential redaction to three endpoints:
- GET /api/session — messages[], tool_calls[], and title
- GET /api/session/export — download also redacted
- SSE done event — session payload in stream
- GET /api/memory — MEMORY.md and USER.md content
Adds api/startup.py with fix_credential_permissions() at server startup.
Adds 13 tests in tests/test_security_redaction.py.
Merged with #237 container detection changes in server.py.
* fix: cancel button now interrupts agent and cleans up UI state (#244)
Wires agent.interrupt() into cancel_stream() so the backend actually
stops tool execution when the user clicks Cancel, rather than only
stopping the SSE stream while the agent keeps running.
Changes:
- api/config.py: adds AGENT_INSTANCES dict (stream_id -> AIAgent)
- api/streaming.py: stores agent in AGENT_INSTANCES after creation,
checks CANCEL_FLAGS immediately after store (race condition fix),
calls agent.interrupt() in cancel_stream(), cleans up in finally block
- static/boot.js: removes stale setStatus(cancelling) call
- static/messages.js: setBusy(false)/setStatus('') unconditionally on cancel
Race condition fix: after storing agent in AGENT_INSTANCES, immediately
checks if CANCEL_FLAGS[stream_id] is already set (cancel arrived during
agent init) and interrupts before starting. Check is inside the same
STREAMS_LOCK acquisition, making it atomic.
New test file: tests/test_cancel_interrupt.py with 6 unit tests.
* docs: v0.46.0 release notes, bump version, update test counts
---------
Co-authored-by: Nathan Esquenazi <nesquena@gmail.com>
2026-04-11 17:17:52 +00:00
|
|
|
"""
|
|
|
|
|
Unit tests for cancel/interrupt functionality.
|
|
|
|
|
Tests the integration between cancel_stream() and agent.interrupt().
|
|
|
|
|
"""
|
|
|
|
|
import queue
|
|
|
|
|
import threading
|
|
|
|
|
from unittest.mock import Mock
|
|
|
|
|
|
|
|
|
|
from api.streaming import cancel_stream
|
2026-06-03 18:26:12 +00:00
|
|
|
from api.config import AGENT_INSTANCES, STREAMS, CANCEL_FLAGS, ACTIVE_RUNS, SESSION_AGENT_CACHE, SESSION_AGENT_CACHE_LOCK
|
v0.46.0: security, Docker UID/GID, model discovery, i18n, cancel fix
* fix: decode HTML entities before markdown processing + zh/zh-Hant translations (#239)
Adds decode() helper in renderMd() to fix double-escaping of HTML entities
from LLM output (e.g. <code> becoming &lt;code&gt; instead
of rendering). XSS-safe: decode runs before esc(), only 5 entity patterns.
Also adds 40+ missing zh (Simplified Chinese) translation keys and a new
zh-Hant (Traditional Chinese) locale with 163 keys.
Fix applied: removed duplicate settings_label_notifications key in both
zh and zh-Hant locales.
Fixes #240
* fix: restore custom model list discovery with config api key (#238)
get_available_models() now reads api_key from config.yaml before env vars:
1. model.api_key
2. providers.<active>.api_key / providers.custom.api_key
3. env var fallbacks (HERMES_API_KEY, OPENAI_API_KEY, etc.)
Also adds OpenAI/Python User-Agent header and a regression test covering
authenticated /v1/models discovery.
Fixes users with LM Studio / Ollama custom endpoints configured in
config.yaml whose model picker silently collapsed to the default model.
* feat: Docker UID/GID matching to avoid root-owned .hermes files (#237)
Adds docker_init.bash with hermeswebuitoo/hermeswebui user pattern so
container files match the host user UID/GID. Prevents .hermes volume
mounts from being owned by root when using a non-root host user.
Configure via WANTED_UID and WANTED_GID env vars (default 1000/1000).
Readme updated with setup instructions.
Fix applied: removed duplicate WANTED_GID=1000 line in docker-compose.yml
that was overriding the ${GID:-1000} variable expansion.
* security: redact credentials from API responses and fix credential file permissions (#243)
Adds response-layer credential redaction to three endpoints:
- GET /api/session — messages[], tool_calls[], and title
- GET /api/session/export — download also redacted
- SSE done event — session payload in stream
- GET /api/memory — MEMORY.md and USER.md content
Adds api/startup.py with fix_credential_permissions() at server startup.
Adds 13 tests in tests/test_security_redaction.py.
Merged with #237 container detection changes in server.py.
* fix: cancel button now interrupts agent and cleans up UI state (#244)
Wires agent.interrupt() into cancel_stream() so the backend actually
stops tool execution when the user clicks Cancel, rather than only
stopping the SSE stream while the agent keeps running.
Changes:
- api/config.py: adds AGENT_INSTANCES dict (stream_id -> AIAgent)
- api/streaming.py: stores agent in AGENT_INSTANCES after creation,
checks CANCEL_FLAGS immediately after store (race condition fix),
calls agent.interrupt() in cancel_stream(), cleans up in finally block
- static/boot.js: removes stale setStatus(cancelling) call
- static/messages.js: setBusy(false)/setStatus('') unconditionally on cancel
Race condition fix: after storing agent in AGENT_INSTANCES, immediately
checks if CANCEL_FLAGS[stream_id] is already set (cancel arrived during
agent init) and interrupts before starting. Check is inside the same
STREAMS_LOCK acquisition, making it atomic.
New test file: tests/test_cancel_interrupt.py with 6 unit tests.
* docs: v0.46.0 release notes, bump version, update test counts
---------
Co-authored-by: Nathan Esquenazi <nesquena@gmail.com>
2026-04-11 17:17:52 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
class TestCancelInterrupt:
|
|
|
|
|
"""Test suite for cancel/interrupt functionality"""
|
|
|
|
|
|
|
|
|
|
def setup_method(self):
|
|
|
|
|
"""Clean up before each test"""
|
|
|
|
|
AGENT_INSTANCES.clear()
|
|
|
|
|
STREAMS.clear()
|
|
|
|
|
CANCEL_FLAGS.clear()
|
2026-06-03 18:26:12 +00:00
|
|
|
ACTIVE_RUNS.clear()
|
|
|
|
|
with SESSION_AGENT_CACHE_LOCK:
|
|
|
|
|
SESSION_AGENT_CACHE.clear()
|
v0.46.0: security, Docker UID/GID, model discovery, i18n, cancel fix
* fix: decode HTML entities before markdown processing + zh/zh-Hant translations (#239)
Adds decode() helper in renderMd() to fix double-escaping of HTML entities
from LLM output (e.g. <code> becoming &lt;code&gt; instead
of rendering). XSS-safe: decode runs before esc(), only 5 entity patterns.
Also adds 40+ missing zh (Simplified Chinese) translation keys and a new
zh-Hant (Traditional Chinese) locale with 163 keys.
Fix applied: removed duplicate settings_label_notifications key in both
zh and zh-Hant locales.
Fixes #240
* fix: restore custom model list discovery with config api key (#238)
get_available_models() now reads api_key from config.yaml before env vars:
1. model.api_key
2. providers.<active>.api_key / providers.custom.api_key
3. env var fallbacks (HERMES_API_KEY, OPENAI_API_KEY, etc.)
Also adds OpenAI/Python User-Agent header and a regression test covering
authenticated /v1/models discovery.
Fixes users with LM Studio / Ollama custom endpoints configured in
config.yaml whose model picker silently collapsed to the default model.
* feat: Docker UID/GID matching to avoid root-owned .hermes files (#237)
Adds docker_init.bash with hermeswebuitoo/hermeswebui user pattern so
container files match the host user UID/GID. Prevents .hermes volume
mounts from being owned by root when using a non-root host user.
Configure via WANTED_UID and WANTED_GID env vars (default 1000/1000).
Readme updated with setup instructions.
Fix applied: removed duplicate WANTED_GID=1000 line in docker-compose.yml
that was overriding the ${GID:-1000} variable expansion.
* security: redact credentials from API responses and fix credential file permissions (#243)
Adds response-layer credential redaction to three endpoints:
- GET /api/session — messages[], tool_calls[], and title
- GET /api/session/export — download also redacted
- SSE done event — session payload in stream
- GET /api/memory — MEMORY.md and USER.md content
Adds api/startup.py with fix_credential_permissions() at server startup.
Adds 13 tests in tests/test_security_redaction.py.
Merged with #237 container detection changes in server.py.
* fix: cancel button now interrupts agent and cleans up UI state (#244)
Wires agent.interrupt() into cancel_stream() so the backend actually
stops tool execution when the user clicks Cancel, rather than only
stopping the SSE stream while the agent keeps running.
Changes:
- api/config.py: adds AGENT_INSTANCES dict (stream_id -> AIAgent)
- api/streaming.py: stores agent in AGENT_INSTANCES after creation,
checks CANCEL_FLAGS immediately after store (race condition fix),
calls agent.interrupt() in cancel_stream(), cleans up in finally block
- static/boot.js: removes stale setStatus(cancelling) call
- static/messages.js: setBusy(false)/setStatus('') unconditionally on cancel
Race condition fix: after storing agent in AGENT_INSTANCES, immediately
checks if CANCEL_FLAGS[stream_id] is already set (cancel arrived during
agent init) and interrupts before starting. Check is inside the same
STREAMS_LOCK acquisition, making it atomic.
New test file: tests/test_cancel_interrupt.py with 6 unit tests.
* docs: v0.46.0 release notes, bump version, update test counts
---------
Co-authored-by: Nathan Esquenazi <nesquena@gmail.com>
2026-04-11 17:17:52 +00:00
|
|
|
|
|
|
|
|
def teardown_method(self):
|
|
|
|
|
"""Clean up after each test"""
|
|
|
|
|
AGENT_INSTANCES.clear()
|
|
|
|
|
STREAMS.clear()
|
|
|
|
|
CANCEL_FLAGS.clear()
|
2026-06-03 18:26:12 +00:00
|
|
|
ACTIVE_RUNS.clear()
|
|
|
|
|
with SESSION_AGENT_CACHE_LOCK:
|
|
|
|
|
SESSION_AGENT_CACHE.clear()
|
v0.46.0: security, Docker UID/GID, model discovery, i18n, cancel fix
* fix: decode HTML entities before markdown processing + zh/zh-Hant translations (#239)
Adds decode() helper in renderMd() to fix double-escaping of HTML entities
from LLM output (e.g. <code> becoming &lt;code&gt; instead
of rendering). XSS-safe: decode runs before esc(), only 5 entity patterns.
Also adds 40+ missing zh (Simplified Chinese) translation keys and a new
zh-Hant (Traditional Chinese) locale with 163 keys.
Fix applied: removed duplicate settings_label_notifications key in both
zh and zh-Hant locales.
Fixes #240
* fix: restore custom model list discovery with config api key (#238)
get_available_models() now reads api_key from config.yaml before env vars:
1. model.api_key
2. providers.<active>.api_key / providers.custom.api_key
3. env var fallbacks (HERMES_API_KEY, OPENAI_API_KEY, etc.)
Also adds OpenAI/Python User-Agent header and a regression test covering
authenticated /v1/models discovery.
Fixes users with LM Studio / Ollama custom endpoints configured in
config.yaml whose model picker silently collapsed to the default model.
* feat: Docker UID/GID matching to avoid root-owned .hermes files (#237)
Adds docker_init.bash with hermeswebuitoo/hermeswebui user pattern so
container files match the host user UID/GID. Prevents .hermes volume
mounts from being owned by root when using a non-root host user.
Configure via WANTED_UID and WANTED_GID env vars (default 1000/1000).
Readme updated with setup instructions.
Fix applied: removed duplicate WANTED_GID=1000 line in docker-compose.yml
that was overriding the ${GID:-1000} variable expansion.
* security: redact credentials from API responses and fix credential file permissions (#243)
Adds response-layer credential redaction to three endpoints:
- GET /api/session — messages[], tool_calls[], and title
- GET /api/session/export — download also redacted
- SSE done event — session payload in stream
- GET /api/memory — MEMORY.md and USER.md content
Adds api/startup.py with fix_credential_permissions() at server startup.
Adds 13 tests in tests/test_security_redaction.py.
Merged with #237 container detection changes in server.py.
* fix: cancel button now interrupts agent and cleans up UI state (#244)
Wires agent.interrupt() into cancel_stream() so the backend actually
stops tool execution when the user clicks Cancel, rather than only
stopping the SSE stream while the agent keeps running.
Changes:
- api/config.py: adds AGENT_INSTANCES dict (stream_id -> AIAgent)
- api/streaming.py: stores agent in AGENT_INSTANCES after creation,
checks CANCEL_FLAGS immediately after store (race condition fix),
calls agent.interrupt() in cancel_stream(), cleans up in finally block
- static/boot.js: removes stale setStatus(cancelling) call
- static/messages.js: setBusy(false)/setStatus('') unconditionally on cancel
Race condition fix: after storing agent in AGENT_INSTANCES, immediately
checks if CANCEL_FLAGS[stream_id] is already set (cancel arrived during
agent init) and interrupts before starting. Check is inside the same
STREAMS_LOCK acquisition, making it atomic.
New test file: tests/test_cancel_interrupt.py with 6 unit tests.
* docs: v0.46.0 release notes, bump version, update test counts
---------
Co-authored-by: Nathan Esquenazi <nesquena@gmail.com>
2026-04-11 17:17:52 +00:00
|
|
|
|
|
|
|
|
def test_cancel_calls_agent_interrupt(self):
|
|
|
|
|
"""Verify that cancel_stream() calls agent.interrupt() when agent exists"""
|
|
|
|
|
# Setup
|
|
|
|
|
stream_id = "test_stream_123"
|
|
|
|
|
mock_agent = Mock()
|
|
|
|
|
mock_agent.interrupt = Mock()
|
|
|
|
|
|
|
|
|
|
STREAMS[stream_id] = queue.Queue()
|
|
|
|
|
CANCEL_FLAGS[stream_id] = threading.Event()
|
|
|
|
|
AGENT_INSTANCES[stream_id] = mock_agent
|
|
|
|
|
|
|
|
|
|
# Execute
|
|
|
|
|
result = cancel_stream(stream_id)
|
|
|
|
|
|
|
|
|
|
# Assert
|
|
|
|
|
assert result is True
|
|
|
|
|
mock_agent.interrupt.assert_called_once_with("Cancelled by user")
|
2026-04-20 23:54:40 +00:00
|
|
|
# CANCEL_FLAGS is eagerly popped after cancel (#776 fix) so the flag
|
|
|
|
|
# is no longer in the dict — verify the pop happened instead
|
|
|
|
|
assert stream_id not in CANCEL_FLAGS, \
|
|
|
|
|
"cancel_stream() should eagerly pop CANCEL_FLAGS after signalling"
|
v0.46.0: security, Docker UID/GID, model discovery, i18n, cancel fix
* fix: decode HTML entities before markdown processing + zh/zh-Hant translations (#239)
Adds decode() helper in renderMd() to fix double-escaping of HTML entities
from LLM output (e.g. <code> becoming &lt;code&gt; instead
of rendering). XSS-safe: decode runs before esc(), only 5 entity patterns.
Also adds 40+ missing zh (Simplified Chinese) translation keys and a new
zh-Hant (Traditional Chinese) locale with 163 keys.
Fix applied: removed duplicate settings_label_notifications key in both
zh and zh-Hant locales.
Fixes #240
* fix: restore custom model list discovery with config api key (#238)
get_available_models() now reads api_key from config.yaml before env vars:
1. model.api_key
2. providers.<active>.api_key / providers.custom.api_key
3. env var fallbacks (HERMES_API_KEY, OPENAI_API_KEY, etc.)
Also adds OpenAI/Python User-Agent header and a regression test covering
authenticated /v1/models discovery.
Fixes users with LM Studio / Ollama custom endpoints configured in
config.yaml whose model picker silently collapsed to the default model.
* feat: Docker UID/GID matching to avoid root-owned .hermes files (#237)
Adds docker_init.bash with hermeswebuitoo/hermeswebui user pattern so
container files match the host user UID/GID. Prevents .hermes volume
mounts from being owned by root when using a non-root host user.
Configure via WANTED_UID and WANTED_GID env vars (default 1000/1000).
Readme updated with setup instructions.
Fix applied: removed duplicate WANTED_GID=1000 line in docker-compose.yml
that was overriding the ${GID:-1000} variable expansion.
* security: redact credentials from API responses and fix credential file permissions (#243)
Adds response-layer credential redaction to three endpoints:
- GET /api/session — messages[], tool_calls[], and title
- GET /api/session/export — download also redacted
- SSE done event — session payload in stream
- GET /api/memory — MEMORY.md and USER.md content
Adds api/startup.py with fix_credential_permissions() at server startup.
Adds 13 tests in tests/test_security_redaction.py.
Merged with #237 container detection changes in server.py.
* fix: cancel button now interrupts agent and cleans up UI state (#244)
Wires agent.interrupt() into cancel_stream() so the backend actually
stops tool execution when the user clicks Cancel, rather than only
stopping the SSE stream while the agent keeps running.
Changes:
- api/config.py: adds AGENT_INSTANCES dict (stream_id -> AIAgent)
- api/streaming.py: stores agent in AGENT_INSTANCES after creation,
checks CANCEL_FLAGS immediately after store (race condition fix),
calls agent.interrupt() in cancel_stream(), cleans up in finally block
- static/boot.js: removes stale setStatus(cancelling) call
- static/messages.js: setBusy(false)/setStatus('') unconditionally on cancel
Race condition fix: after storing agent in AGENT_INSTANCES, immediately
checks if CANCEL_FLAGS[stream_id] is already set (cancel arrived during
agent init) and interrupts before starting. Check is inside the same
STREAMS_LOCK acquisition, making it atomic.
New test file: tests/test_cancel_interrupt.py with 6 unit tests.
* docs: v0.46.0 release notes, bump version, update test counts
---------
Co-authored-by: Nathan Esquenazi <nesquena@gmail.com>
2026-04-11 17:17:52 +00:00
|
|
|
|
|
|
|
|
def test_cancel_handles_interrupt_exception(self):
|
|
|
|
|
"""Verify that cancel_stream() handles interrupt() exceptions gracefully"""
|
|
|
|
|
stream_id = "test_stream_456"
|
|
|
|
|
mock_agent = Mock()
|
|
|
|
|
mock_agent.interrupt = Mock(side_effect=RuntimeError("Agent error"))
|
|
|
|
|
|
|
|
|
|
STREAMS[stream_id] = queue.Queue()
|
|
|
|
|
CANCEL_FLAGS[stream_id] = threading.Event()
|
|
|
|
|
AGENT_INSTANCES[stream_id] = mock_agent
|
|
|
|
|
|
|
|
|
|
# Should not raise exception
|
|
|
|
|
result = cancel_stream(stream_id)
|
|
|
|
|
|
|
|
|
|
# Assert
|
|
|
|
|
assert result is True
|
|
|
|
|
mock_agent.interrupt.assert_called_once()
|
2026-04-20 23:54:40 +00:00
|
|
|
assert stream_id not in CANCEL_FLAGS, \
|
|
|
|
|
"cancel_stream() should eagerly pop CANCEL_FLAGS even on interrupt exception"
|
v0.46.0: security, Docker UID/GID, model discovery, i18n, cancel fix
* fix: decode HTML entities before markdown processing + zh/zh-Hant translations (#239)
Adds decode() helper in renderMd() to fix double-escaping of HTML entities
from LLM output (e.g. <code> becoming &lt;code&gt; instead
of rendering). XSS-safe: decode runs before esc(), only 5 entity patterns.
Also adds 40+ missing zh (Simplified Chinese) translation keys and a new
zh-Hant (Traditional Chinese) locale with 163 keys.
Fix applied: removed duplicate settings_label_notifications key in both
zh and zh-Hant locales.
Fixes #240
* fix: restore custom model list discovery with config api key (#238)
get_available_models() now reads api_key from config.yaml before env vars:
1. model.api_key
2. providers.<active>.api_key / providers.custom.api_key
3. env var fallbacks (HERMES_API_KEY, OPENAI_API_KEY, etc.)
Also adds OpenAI/Python User-Agent header and a regression test covering
authenticated /v1/models discovery.
Fixes users with LM Studio / Ollama custom endpoints configured in
config.yaml whose model picker silently collapsed to the default model.
* feat: Docker UID/GID matching to avoid root-owned .hermes files (#237)
Adds docker_init.bash with hermeswebuitoo/hermeswebui user pattern so
container files match the host user UID/GID. Prevents .hermes volume
mounts from being owned by root when using a non-root host user.
Configure via WANTED_UID and WANTED_GID env vars (default 1000/1000).
Readme updated with setup instructions.
Fix applied: removed duplicate WANTED_GID=1000 line in docker-compose.yml
that was overriding the ${GID:-1000} variable expansion.
* security: redact credentials from API responses and fix credential file permissions (#243)
Adds response-layer credential redaction to three endpoints:
- GET /api/session — messages[], tool_calls[], and title
- GET /api/session/export — download also redacted
- SSE done event — session payload in stream
- GET /api/memory — MEMORY.md and USER.md content
Adds api/startup.py with fix_credential_permissions() at server startup.
Adds 13 tests in tests/test_security_redaction.py.
Merged with #237 container detection changes in server.py.
* fix: cancel button now interrupts agent and cleans up UI state (#244)
Wires agent.interrupt() into cancel_stream() so the backend actually
stops tool execution when the user clicks Cancel, rather than only
stopping the SSE stream while the agent keeps running.
Changes:
- api/config.py: adds AGENT_INSTANCES dict (stream_id -> AIAgent)
- api/streaming.py: stores agent in AGENT_INSTANCES after creation,
checks CANCEL_FLAGS immediately after store (race condition fix),
calls agent.interrupt() in cancel_stream(), cleans up in finally block
- static/boot.js: removes stale setStatus(cancelling) call
- static/messages.js: setBusy(false)/setStatus('') unconditionally on cancel
Race condition fix: after storing agent in AGENT_INSTANCES, immediately
checks if CANCEL_FLAGS[stream_id] is already set (cancel arrived during
agent init) and interrupts before starting. Check is inside the same
STREAMS_LOCK acquisition, making it atomic.
New test file: tests/test_cancel_interrupt.py with 6 unit tests.
* docs: v0.46.0 release notes, bump version, update test counts
---------
Co-authored-by: Nathan Esquenazi <nesquena@gmail.com>
2026-04-11 17:17:52 +00:00
|
|
|
|
|
|
|
|
def test_cancel_before_agent_ready(self):
|
|
|
|
|
"""Test cancel when agent not yet stored in AGENT_INSTANCES (race condition)"""
|
|
|
|
|
stream_id = "test_stream_789"
|
|
|
|
|
|
|
|
|
|
STREAMS[stream_id] = queue.Queue()
|
|
|
|
|
CANCEL_FLAGS[stream_id] = threading.Event()
|
|
|
|
|
# Note: AGENT_INSTANCES[stream_id] not set (simulating race condition)
|
|
|
|
|
|
|
|
|
|
# Should succeed even without agent
|
|
|
|
|
result = cancel_stream(stream_id)
|
|
|
|
|
|
|
|
|
|
# Assert
|
|
|
|
|
assert result is True
|
2026-04-20 23:54:40 +00:00
|
|
|
# CANCEL_FLAGS is eagerly popped; the agent thread checks the event
|
|
|
|
|
# object it already has a reference to — pop doesn't clear the event
|
|
|
|
|
assert stream_id not in CANCEL_FLAGS, \
|
|
|
|
|
"cancel_stream() should eagerly pop CANCEL_FLAGS even without an agent"
|
|
|
|
|
# Agent will check this flag (it holds a reference to the event object)
|
v0.46.0: security, Docker UID/GID, model discovery, i18n, cancel fix
* fix: decode HTML entities before markdown processing + zh/zh-Hant translations (#239)
Adds decode() helper in renderMd() to fix double-escaping of HTML entities
from LLM output (e.g. <code> becoming &lt;code&gt; instead
of rendering). XSS-safe: decode runs before esc(), only 5 entity patterns.
Also adds 40+ missing zh (Simplified Chinese) translation keys and a new
zh-Hant (Traditional Chinese) locale with 163 keys.
Fix applied: removed duplicate settings_label_notifications key in both
zh and zh-Hant locales.
Fixes #240
* fix: restore custom model list discovery with config api key (#238)
get_available_models() now reads api_key from config.yaml before env vars:
1. model.api_key
2. providers.<active>.api_key / providers.custom.api_key
3. env var fallbacks (HERMES_API_KEY, OPENAI_API_KEY, etc.)
Also adds OpenAI/Python User-Agent header and a regression test covering
authenticated /v1/models discovery.
Fixes users with LM Studio / Ollama custom endpoints configured in
config.yaml whose model picker silently collapsed to the default model.
* feat: Docker UID/GID matching to avoid root-owned .hermes files (#237)
Adds docker_init.bash with hermeswebuitoo/hermeswebui user pattern so
container files match the host user UID/GID. Prevents .hermes volume
mounts from being owned by root when using a non-root host user.
Configure via WANTED_UID and WANTED_GID env vars (default 1000/1000).
Readme updated with setup instructions.
Fix applied: removed duplicate WANTED_GID=1000 line in docker-compose.yml
that was overriding the ${GID:-1000} variable expansion.
* security: redact credentials from API responses and fix credential file permissions (#243)
Adds response-layer credential redaction to three endpoints:
- GET /api/session — messages[], tool_calls[], and title
- GET /api/session/export — download also redacted
- SSE done event — session payload in stream
- GET /api/memory — MEMORY.md and USER.md content
Adds api/startup.py with fix_credential_permissions() at server startup.
Adds 13 tests in tests/test_security_redaction.py.
Merged with #237 container detection changes in server.py.
* fix: cancel button now interrupts agent and cleans up UI state (#244)
Wires agent.interrupt() into cancel_stream() so the backend actually
stops tool execution when the user clicks Cancel, rather than only
stopping the SSE stream while the agent keeps running.
Changes:
- api/config.py: adds AGENT_INSTANCES dict (stream_id -> AIAgent)
- api/streaming.py: stores agent in AGENT_INSTANCES after creation,
checks CANCEL_FLAGS immediately after store (race condition fix),
calls agent.interrupt() in cancel_stream(), cleans up in finally block
- static/boot.js: removes stale setStatus(cancelling) call
- static/messages.js: setBusy(false)/setStatus('') unconditionally on cancel
Race condition fix: after storing agent in AGENT_INSTANCES, immediately
checks if CANCEL_FLAGS[stream_id] is already set (cancel arrived during
agent init) and interrupts before starting. Check is inside the same
STREAMS_LOCK acquisition, making it atomic.
New test file: tests/test_cancel_interrupt.py with 6 unit tests.
* docs: v0.46.0 release notes, bump version, update test counts
---------
Co-authored-by: Nathan Esquenazi <nesquena@gmail.com>
2026-04-11 17:17:52 +00:00
|
|
|
|
|
|
|
|
def test_cancel_nonexistent_stream(self):
|
|
|
|
|
"""Test cancel for a stream that doesn't exist"""
|
|
|
|
|
result = cancel_stream("nonexistent_stream")
|
|
|
|
|
assert result is False
|
|
|
|
|
|
2026-06-03 18:26:12 +00:00
|
|
|
def test_cancel_falls_back_to_active_run_registry(self):
|
|
|
|
|
"""Cancel should still work when STREAMS is gone but the worker is alive."""
|
|
|
|
|
from unittest.mock import patch
|
|
|
|
|
|
|
|
|
|
stream_id = "detached_stream_123"
|
|
|
|
|
session_id = "sess_detached_123"
|
|
|
|
|
mock_agent = Mock()
|
|
|
|
|
mock_agent.interrupt = Mock()
|
|
|
|
|
mock_agent.session_id = session_id
|
|
|
|
|
|
|
|
|
|
mock_session = Mock()
|
|
|
|
|
mock_session.session_id = session_id
|
|
|
|
|
mock_session.active_stream_id = stream_id
|
|
|
|
|
mock_session.pending_user_message = "hello"
|
|
|
|
|
mock_session.pending_attachments = ["file.txt"]
|
|
|
|
|
mock_session.pending_started_at = 1234567890.0
|
|
|
|
|
mock_session.messages = []
|
|
|
|
|
mock_session.save = Mock()
|
|
|
|
|
|
|
|
|
|
ACTIVE_RUNS[stream_id] = {
|
|
|
|
|
"session_id": session_id,
|
|
|
|
|
"started_at": 1234567890.0,
|
|
|
|
|
"phase": "running",
|
|
|
|
|
}
|
|
|
|
|
with SESSION_AGENT_CACHE_LOCK:
|
|
|
|
|
SESSION_AGENT_CACHE[session_id] = (mock_agent, "sig")
|
|
|
|
|
|
|
|
|
|
with patch("api.streaming.get_session", return_value=mock_session):
|
|
|
|
|
result = cancel_stream(stream_id)
|
|
|
|
|
|
|
|
|
|
assert result is True
|
2026-06-19 19:34:24 +00:00
|
|
|
assert ACTIVE_RUNS[stream_id]["phase"] == "cancelling"
|
2026-06-03 18:26:12 +00:00
|
|
|
mock_agent.interrupt.assert_called_once_with("Cancelled by user")
|
|
|
|
|
assert mock_session.active_stream_id is None
|
|
|
|
|
assert mock_session.pending_user_message is None
|
|
|
|
|
assert mock_session.pending_attachments == []
|
|
|
|
|
assert mock_session.pending_started_at is None
|
|
|
|
|
mock_session.save.assert_called_once()
|
|
|
|
|
|
v0.46.0: security, Docker UID/GID, model discovery, i18n, cancel fix
* fix: decode HTML entities before markdown processing + zh/zh-Hant translations (#239)
Adds decode() helper in renderMd() to fix double-escaping of HTML entities
from LLM output (e.g. <code> becoming &lt;code&gt; instead
of rendering). XSS-safe: decode runs before esc(), only 5 entity patterns.
Also adds 40+ missing zh (Simplified Chinese) translation keys and a new
zh-Hant (Traditional Chinese) locale with 163 keys.
Fix applied: removed duplicate settings_label_notifications key in both
zh and zh-Hant locales.
Fixes #240
* fix: restore custom model list discovery with config api key (#238)
get_available_models() now reads api_key from config.yaml before env vars:
1. model.api_key
2. providers.<active>.api_key / providers.custom.api_key
3. env var fallbacks (HERMES_API_KEY, OPENAI_API_KEY, etc.)
Also adds OpenAI/Python User-Agent header and a regression test covering
authenticated /v1/models discovery.
Fixes users with LM Studio / Ollama custom endpoints configured in
config.yaml whose model picker silently collapsed to the default model.
* feat: Docker UID/GID matching to avoid root-owned .hermes files (#237)
Adds docker_init.bash with hermeswebuitoo/hermeswebui user pattern so
container files match the host user UID/GID. Prevents .hermes volume
mounts from being owned by root when using a non-root host user.
Configure via WANTED_UID and WANTED_GID env vars (default 1000/1000).
Readme updated with setup instructions.
Fix applied: removed duplicate WANTED_GID=1000 line in docker-compose.yml
that was overriding the ${GID:-1000} variable expansion.
* security: redact credentials from API responses and fix credential file permissions (#243)
Adds response-layer credential redaction to three endpoints:
- GET /api/session — messages[], tool_calls[], and title
- GET /api/session/export — download also redacted
- SSE done event — session payload in stream
- GET /api/memory — MEMORY.md and USER.md content
Adds api/startup.py with fix_credential_permissions() at server startup.
Adds 13 tests in tests/test_security_redaction.py.
Merged with #237 container detection changes in server.py.
* fix: cancel button now interrupts agent and cleans up UI state (#244)
Wires agent.interrupt() into cancel_stream() so the backend actually
stops tool execution when the user clicks Cancel, rather than only
stopping the SSE stream while the agent keeps running.
Changes:
- api/config.py: adds AGENT_INSTANCES dict (stream_id -> AIAgent)
- api/streaming.py: stores agent in AGENT_INSTANCES after creation,
checks CANCEL_FLAGS immediately after store (race condition fix),
calls agent.interrupt() in cancel_stream(), cleans up in finally block
- static/boot.js: removes stale setStatus(cancelling) call
- static/messages.js: setBusy(false)/setStatus('') unconditionally on cancel
Race condition fix: after storing agent in AGENT_INSTANCES, immediately
checks if CANCEL_FLAGS[stream_id] is already set (cancel arrived during
agent init) and interrupts before starting. Check is inside the same
STREAMS_LOCK acquisition, making it atomic.
New test file: tests/test_cancel_interrupt.py with 6 unit tests.
* docs: v0.46.0 release notes, bump version, update test counts
---------
Co-authored-by: Nathan Esquenazi <nesquena@gmail.com>
2026-04-11 17:17:52 +00:00
|
|
|
def test_cancel_sets_cancel_event(self):
|
|
|
|
|
"""Verify that cancel_stream() sets the cancel_event flag"""
|
|
|
|
|
stream_id = "test_stream_event"
|
|
|
|
|
|
|
|
|
|
STREAMS[stream_id] = queue.Queue()
|
|
|
|
|
cancel_event = threading.Event()
|
|
|
|
|
CANCEL_FLAGS[stream_id] = cancel_event
|
|
|
|
|
|
|
|
|
|
result = cancel_stream(stream_id)
|
|
|
|
|
|
|
|
|
|
assert result is True
|
|
|
|
|
assert cancel_event.is_set()
|
|
|
|
|
|
|
|
|
|
def test_cancel_puts_sentinel_in_queue(self):
|
|
|
|
|
"""Verify that cancel_stream() puts cancel sentinel in queue"""
|
|
|
|
|
stream_id = "test_stream_queue"
|
|
|
|
|
q = queue.Queue()
|
|
|
|
|
|
|
|
|
|
STREAMS[stream_id] = q
|
|
|
|
|
CANCEL_FLAGS[stream_id] = threading.Event()
|
|
|
|
|
|
|
|
|
|
result = cancel_stream(stream_id)
|
|
|
|
|
|
|
|
|
|
assert result is True
|
|
|
|
|
# Check that cancel message was queued
|
|
|
|
|
assert not q.empty()
|
|
|
|
|
event_type, data = q.get_nowait()
|
|
|
|
|
assert event_type == 'cancel'
|
|
|
|
|
assert data['message'] == 'Cancelled by user'
|
2026-06-03 18:26:12 +00:00
|
|
|
|
|
|
|
|
def test_cancel_preserves_partial_text_when_interrupt_pops_buffers(self):
|
|
|
|
|
"""Regression (Codex pre-release finding on #3475): the partial-text /
|
|
|
|
|
reasoning / tool-call buffers must be snapshotted UNDER streams_lock
|
|
|
|
|
BEFORE agent.interrupt() runs. Otherwise the worker's finally block can
|
|
|
|
|
pop those live buffers (it does so under STREAMS_LOCK) the instant the
|
|
|
|
|
interrupt wakes it, and the cancelled turn silently loses its
|
|
|
|
|
already-streamed partial text.
|
|
|
|
|
|
|
|
|
|
We simulate that race deterministically: the mock agent's interrupt()
|
|
|
|
|
clears the live STREAM_* maps, mimicking the worker finally. The fix
|
|
|
|
|
must still persist the partial text captured before the interrupt.
|
|
|
|
|
"""
|
|
|
|
|
from unittest.mock import patch
|
|
|
|
|
from api.config import (
|
|
|
|
|
STREAM_PARTIAL_TEXT, STREAM_REASONING_TEXT, STREAM_LIVE_TOOL_CALLS,
|
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
stream_id = "race_stream_partial"
|
|
|
|
|
session_id = "sess_race_partial"
|
|
|
|
|
|
|
|
|
|
q = queue.Queue()
|
|
|
|
|
STREAMS[stream_id] = q
|
|
|
|
|
CANCEL_FLAGS[stream_id] = threading.Event()
|
|
|
|
|
STREAM_PARTIAL_TEXT[stream_id] = "partial answer so far"
|
|
|
|
|
STREAM_REASONING_TEXT[stream_id] = "thinking..."
|
|
|
|
|
STREAM_LIVE_TOOL_CALLS[stream_id] = [{"name": "search"}]
|
|
|
|
|
|
|
|
|
|
mock_agent = Mock()
|
|
|
|
|
mock_agent.session_id = session_id
|
|
|
|
|
|
|
|
|
|
def _interrupt(_msg):
|
|
|
|
|
# Mimic the worker's finally block popping the live buffers the
|
|
|
|
|
# moment the interrupt wakes it (it runs under STREAMS_LOCK in prod;
|
|
|
|
|
# here we just clear to model the worst-case post-interrupt state).
|
|
|
|
|
STREAM_PARTIAL_TEXT.pop(stream_id, None)
|
|
|
|
|
STREAM_REASONING_TEXT.pop(stream_id, None)
|
|
|
|
|
STREAM_LIVE_TOOL_CALLS.pop(stream_id, None)
|
|
|
|
|
|
|
|
|
|
mock_agent.interrupt = Mock(side_effect=_interrupt)
|
|
|
|
|
AGENT_INSTANCES[stream_id] = mock_agent
|
|
|
|
|
|
|
|
|
|
mock_session = Mock()
|
|
|
|
|
mock_session.session_id = session_id
|
|
|
|
|
mock_session.active_stream_id = stream_id
|
|
|
|
|
mock_session.pending_user_message = "q"
|
|
|
|
|
mock_session.pending_attachments = []
|
|
|
|
|
mock_session.pending_started_at = 1.0
|
|
|
|
|
mock_session.messages = []
|
|
|
|
|
mock_session.save = Mock()
|
|
|
|
|
|
|
|
|
|
with patch("api.streaming.get_session", return_value=mock_session):
|
|
|
|
|
result = cancel_stream(stream_id)
|
|
|
|
|
|
|
|
|
|
assert result is True
|
|
|
|
|
mock_agent.interrupt.assert_called_once_with("Cancelled by user")
|
|
|
|
|
# The cancelled turn must carry the partial text that was live BEFORE the
|
|
|
|
|
# interrupt popped the buffers. Find it in the appended messages.
|
|
|
|
|
appended = [m for m in mock_session.messages if isinstance(m, dict)]
|
|
|
|
|
joined = " ".join(str(m.get("content", "")) for m in appended)
|
|
|
|
|
assert "partial answer so far" in joined, (
|
|
|
|
|
"cancelled turn lost its already-streamed partial text — the snapshot "
|
|
|
|
|
"must be captured under streams_lock BEFORE agent.interrupt()"
|
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
def test_cancel_preserves_partial_text_on_detached_active_run_path(self):
|
|
|
|
|
"""Regression (Codex 2nd pre-release finding on #3475): the under-lock
|
|
|
|
|
snapshot must also cover the STREAMS-absent / ACTIVE_RUNS-present path.
|
|
|
|
|
When the browser SSE has detached (no STREAMS entry) but the worker is
|
|
|
|
|
still live in ACTIVE_RUNS, cancel resolves the agent from
|
|
|
|
|
SESSION_AGENT_CACHE; the partial-text snapshot must still be taken
|
|
|
|
|
before agent.interrupt() pops the live buffers.
|
|
|
|
|
"""
|
|
|
|
|
from unittest.mock import patch
|
|
|
|
|
from api.config import (
|
|
|
|
|
STREAM_PARTIAL_TEXT, STREAM_REASONING_TEXT, STREAM_LIVE_TOOL_CALLS,
|
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
stream_id = "detached_race_stream"
|
|
|
|
|
session_id = "sess_detached_race"
|
|
|
|
|
|
|
|
|
|
# NOTE: deliberately NO STREAMS entry — this is the detached path.
|
|
|
|
|
STREAM_PARTIAL_TEXT[stream_id] = "detached partial text"
|
|
|
|
|
STREAM_REASONING_TEXT[stream_id] = "detached reasoning"
|
|
|
|
|
STREAM_LIVE_TOOL_CALLS[stream_id] = [{"name": "tool"}]
|
|
|
|
|
|
|
|
|
|
mock_agent = Mock()
|
|
|
|
|
mock_agent.session_id = session_id
|
|
|
|
|
|
|
|
|
|
def _interrupt(_msg):
|
|
|
|
|
STREAM_PARTIAL_TEXT.pop(stream_id, None)
|
|
|
|
|
STREAM_REASONING_TEXT.pop(stream_id, None)
|
|
|
|
|
STREAM_LIVE_TOOL_CALLS.pop(stream_id, None)
|
|
|
|
|
|
|
|
|
|
mock_agent.interrupt = Mock(side_effect=_interrupt)
|
|
|
|
|
|
|
|
|
|
ACTIVE_RUNS[stream_id] = {
|
|
|
|
|
"session_id": session_id,
|
|
|
|
|
"started_at": 1.0,
|
|
|
|
|
"phase": "running",
|
|
|
|
|
}
|
|
|
|
|
with SESSION_AGENT_CACHE_LOCK:
|
|
|
|
|
SESSION_AGENT_CACHE[session_id] = (mock_agent, "sig")
|
|
|
|
|
|
|
|
|
|
mock_session = Mock()
|
|
|
|
|
mock_session.session_id = session_id
|
|
|
|
|
mock_session.active_stream_id = stream_id
|
|
|
|
|
mock_session.pending_user_message = "q"
|
|
|
|
|
mock_session.pending_attachments = []
|
|
|
|
|
mock_session.pending_started_at = 1.0
|
|
|
|
|
mock_session.messages = []
|
|
|
|
|
mock_session.save = Mock()
|
|
|
|
|
|
|
|
|
|
with patch("api.streaming.get_session", return_value=mock_session), \
|
|
|
|
|
patch("api.streaming._cached_agent_matches_session", return_value=True):
|
|
|
|
|
result = cancel_stream(stream_id)
|
|
|
|
|
|
|
|
|
|
assert result is True
|
|
|
|
|
mock_agent.interrupt.assert_called_once_with("Cancelled by user")
|
|
|
|
|
appended = [m for m in mock_session.messages if isinstance(m, dict)]
|
|
|
|
|
joined = " ".join(str(m.get("content", "")) for m in appended)
|
|
|
|
|
assert "detached partial text" in joined, (
|
|
|
|
|
"detached-path cancel lost its partial text — the under-lock snapshot "
|
|
|
|
|
"must cover the ACTIVE_RUNS-only path too, not just STREAMS-present"
|
|
|
|
|
)
|