Add guarded USB device controls and firmware flashing - #22
Conversation
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (2)
🚧 Files skipped from review as they are similar to previous changes (1)
📝 WalkthroughWalkthroughThe application adds connected-radio configuration and maintenance controls, verified ESP32 firmware discovery and flashing, threaded controller workflows, a device page, ChangesDevice Controls and Firmware Management
Estimated code review effort: 4 (Complex) | ~60 minutes Sequence Diagram(s)sequenceDiagram
participant Operator
participant DevicePage
participant MainWindow
participant FirmwareController
participant FirmwareService
participant ESP32Radio
Operator->>DevicePage: Select firmware operation
DevicePage->>MainWindow: Emit firmware request
MainWindow->>FirmwareController: Queue discovery or preparation
FirmwareController->>FirmwareService: Validate release and build bundle
FirmwareService-->>DevicePage: Report progress and verified bundle
DevicePage->>MainWindow: Confirm flash with USB identity
MainWindow->>ESP32Radio: Disconnect active serial connection
MainWindow->>FirmwareController: Queue flashing
FirmwareController->>FirmwareService: Flash verified bundle
FirmwareService->>ESP32Radio: Run esptool commands
FirmwareService-->>DevicePage: Return logs and completion status
MainWindow->>ESP32Radio: Reconnect after flashing
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches 💡 1📝 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 |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 13
🧹 Nitpick comments (5)
src/meshchat/services/device_config.py (1)
192-196: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick winValidate
section_nameagainst_WRITABLE_SECTIONSbefore the write.
hasattraccepts any attribute name on the config message. For a scalar attribute such asversion, or for a bound method name, line 197 raisesAttributeErrorinstead of the intendedValueError. The read path already restricts sections to_WRITABLE_SECTIONS(line 120). Apply the same allowlist on the write path so both directions share one contract.♻️ Proposed fix
def apply_section(node, section_name: str, changes: dict[str, Any]) -> None: + if section_name not in _WRITABLE_SECTIONS: + raise ValueError(f"Unknown configuration section: {section_name}") parent = node.localConfig if hasattr(node.localConfig, section_name) else node.moduleConfig if not hasattr(parent, section_name): raise ValueError(f"Unknown configuration section: {section_name}")🤖 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 `@src/meshchat/services/device_config.py` around lines 192 - 196, Update apply_section to validate section_name against _WRITABLE_SECTIONS before selecting or accessing the configuration parent, raising ValueError for any disallowed section. Keep the existing parent selection and valid-section write behavior unchanged, matching the read path’s allowlist contract.tests/test_firmware_service.py (1)
119-143: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winAdd one test that runs against the installed esptool parser.
Every flashing test replaces
sys.modules["esptool"]with a stub, so the asserted command names are never checked against a real esptool. This is why the v4/v5 command-name mismatch reached review. Add a guard test that asserts the installed esptool acceptschip-id,erase-flash, andwrite-flash.💚 Proposed test
def test_installed_esptool_accepts_the_command_names_used_by_flash_bundle(): esptool = pytest.importorskip("esptool") with pytest.raises(SystemExit) as exit_info: esptool.main(["--help"]) assert exit_info.value.code == 0 # Fails on esptool 4.x, where the commands are chip_id / erase_flash / write_flash. assert tuple(int(part) for part in esptool.__version__.split(".")[:1]) >= (5,)🤖 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 `@tests/test_firmware_service.py` around lines 119 - 143, Add a separate test near test_flash_update_uses_only_verified_update_offset that uses pytest.importorskip("esptool") without mocking sys.modules, invokes the installed parser through esptool.main(["--help"]) and verifies successful exit, then assert the installed esptool major version supports the hyphenated commands chip-id, erase-flash, and write-flash. Keep existing flashing tests and stubs unchanged.src/meshchat/controllers/firmware_controller.py (1)
22-23: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winLog worker exceptions in addition to emitting
completed.Each handler converts the exception to
str(exc)and discards the traceback. Firmware failures are hard to diagnose from a one-line message. The rest of the codebase useslog.exceptionfor the same pattern, for example inMeshtasticWorker.Add a module logger and call
log.exception(...)in eachexceptblock before emittingcompleted.Also applies to: 34-35, 45-46
🤖 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 `@src/meshchat/controllers/firmware_controller.py` around lines 22 - 23, Add a module-level logger and update every exception handler in the firmware controller, including the handlers around discover and the other referenced operations, to call log.exception(...) before emitting completed. Preserve each handler’s existing completion signal and error text.src/meshchat/ui/device/device_page.py (1)
230-241: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win
set_connectedclears_snapshotas a side effect, whichset_snapshotthen relies on ordering to survive.
set_snapshotassignsself._snapshot = snapshotat Line 240, then callsset_connected(...)at Line 241. Ifsnapshot.serial_portisNone,set_connected(False)resetsself._snapshotback toNone, but the following lines still populate the summary, identity fields, sections, and channels from that snapshot. The page then shows device data whileself._snapshotisNone.The current behavior is safe because the tabs are disabled and every action guards on
self._snapshot is None. The coupling is fragile. Move the_snapshotreset out ofset_connectedand clear it explicitly in the disconnect path.♻️ Proposed refactor
def set_connected(self, connected: bool) -> None: self._tabs.setEnabled(connected) self._refresh.setEnabled(connected) if not connected: - self._snapshot = None self._summary.setText("Connect a Meshtastic radio over USB / Serial to manage it.") self._flash_update.setEnabled(False) self._flash_full.setEnabled(False)Then clear the snapshot where disconnect is handled, for example in
MainWindow._on_disconnectedbeforeset_connected(False), or add a smallclear()method onDevicePage.🤖 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 `@src/meshchat/ui/device/device_page.py` around lines 230 - 241, Remove the _snapshot reset from DevicePage.set_connected so set_snapshot preserves the assigned snapshot while updating the connection state. Clear the device page snapshot explicitly in the disconnect flow, preferably in MainWindow._on_disconnected before calling set_connected(False), or through a dedicated DevicePage clear method.tests/test_device_page.py (1)
5-20: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winAdd coverage for the destructive and parsing paths of
DevicePage.The current tests cover tab labels and the summary text. The highest-risk logic is untested:
_confirm_flashtyped-confirmation gating, including the case where the typed phrase does not match and no signal is emitted._confirm_factory_resetfor bothfull=Falseandfull=True._read_widgetparsing of int, float, and repeated fields, including theValueErrorpath that_save_current_sectionconverts into a warning._load_channelclearing_channel_pskwhen the user switches channels, which is the guarantee that a PSK is never carried across channels.The snapshot factory also omits
sectionsandchannels, so_rebuild_sectionsand_rebuild_channelscurrently run against empty input only.🤖 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 `@tests/test_device_page.py` around lines 5 - 20, Expand tests around DevicePage to cover _confirm_flash’s matching and non-matching typed confirmation behavior, _confirm_factory_reset with both full=False and full=True, and _read_widget parsing for integer, float, repeated fields, and invalid input through _save_current_section’s warning path. Add representative sections and channels to _snapshot, and verify _load_channel clears _channel_psk when switching channels.
🤖 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 `@packaging/orcmesh.spec`:
- Around line 83-85: Update the esptool packaging setup near hiddenimports to
collect its non-Python data files with collect_data_files, ensuring the call is
made while datas is in scope and preserves the existing collect_submodules
behavior. Include the esptool target stub JSON files in the frozen build.
In `@pyproject.toml`:
- Line 18: Update the esptool dependency constraint in pyproject.toml to require
version 5.0 or newer while preserving the existing upper bound, so firmware.py’s
v5-only hyphenated commands remain supported.
In `@requirements.txt`:
- Line 12: Update the esptool version constraint in requirements.txt to use the
minimum version that provides the v5 command names, matching the consolidated
constraint defined for pyproject.toml. Preserve the existing upper bound unless
the consolidated requirement specifies otherwise.
In `@src/meshchat/controllers/firmware_controller.py`:
- Around line 84-86: Update the firmware worker around its flash/prepare
operation handlers and shutdown() to track whether an operation is in flight,
then ensure shutdown waits until that operation completes before returning. Do
not allow the thread to be destroyed while work remains active; if
MainWindow.closeEvent participates in shutdown, preserve the existing close flow
only after the worker is no longer busy.
In `@src/meshchat/controllers/meshtastic_controller.py`:
- Around line 952-962: The successful write path in set_owner must refresh the
device-control snapshot. After setOwner succeeds, call _emit_device_controls()
before or alongside emitting device_operation_completed, matching the behavior
of apply_device_section and update_channel; leave the existing error handling
unchanged.
- Around line 993-1002: Update _run_local_node_action to accept separate
display-label and verb-phrase parameters, using the label for the success
message and the verb phrase for the error message. Update every caller,
including reset_nodedb, set_fixed_position, and remove_fixed_position, to
provide grammatically correct values while preserving the existing callback and
error-handling behavior.
In `@src/meshchat/services/firmware.py`:
- Around line 296-301: Update the expected-chip validation around expected_chip
so every declared release platform is covered by an explicit mapping; reject any
platform absent from that mapping before flashing, rather than skipping
validation when expected_chip is None. Preserve the existing case-insensitive
chip-identity mismatch error for mapped platforms and ensure both write paths
remain gated by this preflight.
- Around line 56-63: Update both _json and _download to validate each URL before
calling urllib.request.urlopen: require HTTPS and restrict the hostname to the
known GitHub hosts used by the API and release downloads. Reject any other
scheme or host, including values from browser_download_url and the manifest
response, before fetching.
- Around line 271-285: Update the local esptool invocation in run so it no
longer uses contextlib.redirect_stdout or redirect_stderr, which mutate
process-global streams from the QThread. Pass an explicit output/error stream
supported by esptool, or use a thread-local capture mechanism, while preserving
writer.text collection and the existing FirmwareError handling.
- Around line 303-309: Update the esptool command names used by the firmware
flashing flow, including the preflight command and both branches of the flash
logic around full_install, to use the underscore-form names supported by the
pinned esptool v4 dependency (chip_id, erase_flash, and write_flash). Keep the
existing arguments, offsets, and execution order unchanged.
In `@src/meshchat/ui/device/device_page.py`:
- Around line 368-373: Update the enum and channel-role widget initialization
near the non-repeated enum branch and the corresponding logic around
_read_widget so an unknown field value is preserved rather than coerced to index
0. Only select a combo-box choice when findData returns a valid index; otherwise
retain the original value and ensure saving does not overwrite it with the first
choice.
- Around line 120-131: Update the button handler mapping in the device-page
button setup so the non-full “Factory Reset” action invokes
_confirm_factory_reset with full=False through a wrapper that ignores
QPushButton.clicked’s boolean argument; preserve the existing lambda for “Full
Factory Reset” and other handlers.
In `@src/meshchat/ui/main_window.py`:
- Around line 606-619: Clear _device_snapshot in the disconnect handling path,
specifically _on_disconnected, so stale serial-port data cannot authorize a
flash after the radio is gone. Update _on_firmware_flash_requested to detect an
already-disconnected controller and proceed with the pending flash handoff
directly instead of relying on disconnect to emit an event; preserve the
existing connected flow through self._controller.disconnect().
---
Nitpick comments:
In `@src/meshchat/controllers/firmware_controller.py`:
- Around line 22-23: Add a module-level logger and update every exception
handler in the firmware controller, including the handlers around discover and
the other referenced operations, to call log.exception(...) before emitting
completed. Preserve each handler’s existing completion signal and error text.
In `@src/meshchat/services/device_config.py`:
- Around line 192-196: Update apply_section to validate section_name against
_WRITABLE_SECTIONS before selecting or accessing the configuration parent,
raising ValueError for any disallowed section. Keep the existing parent
selection and valid-section write behavior unchanged, matching the read path’s
allowlist contract.
In `@src/meshchat/ui/device/device_page.py`:
- Around line 230-241: Remove the _snapshot reset from DevicePage.set_connected
so set_snapshot preserves the assigned snapshot while updating the connection
state. Clear the device page snapshot explicitly in the disconnect flow,
preferably in MainWindow._on_disconnected before calling set_connected(False),
or through a dedicated DevicePage clear method.
In `@tests/test_device_page.py`:
- Around line 5-20: Expand tests around DevicePage to cover _confirm_flash’s
matching and non-matching typed confirmation behavior, _confirm_factory_reset
with both full=False and full=True, and _read_widget parsing for integer, float,
repeated fields, and invalid input through _save_current_section’s warning path.
Add representative sections and channels to _snapshot, and verify _load_channel
clears _channel_psk when switching channels.
In `@tests/test_firmware_service.py`:
- Around line 119-143: Add a separate test near
test_flash_update_uses_only_verified_update_offset that uses
pytest.importorskip("esptool") without mocking sys.modules, invokes the
installed parser through esptool.main(["--help"]) and verifies successful exit,
then assert the installed esptool major version supports the hyphenated commands
chip-id, erase-flash, and write-flash. Keep existing flashing tests and stubs
unchanged.
🪄 Autofix
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: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: fafdc3cb-3931-4454-aee6-aa71d1920f74
📒 Files selected for processing (17)
README.mdROADMAP.mdTHIRD_PARTY_LICENSES.mdpackaging/orcmesh.specpyproject.tomlrequirements.txtsrc/meshchat/controllers/firmware_controller.pysrc/meshchat/controllers/meshtastic_controller.pysrc/meshchat/models/device_control.pysrc/meshchat/services/device_config.pysrc/meshchat/services/firmware.pysrc/meshchat/ui/device/__init__.pysrc/meshchat/ui/device/device_page.pysrc/meshchat/ui/main_window.pytests/test_device_config.pytests/test_device_page.pytests/test_firmware_service.py
| if full_install: | ||
| run(["erase-flash"]) | ||
| run(["write-flash", "0x0", str(bundle.factory_image)]) | ||
| run(["write-flash", bundle.ota_offset, str(bundle.ota_image)]) | ||
| run(["write-flash", bundle.filesystem_offset, str(bundle.filesystem_image)]) | ||
| else: | ||
| run(["write-flash", "0x10000", str(bundle.update_image)]) |
There was a problem hiding this comment.
🩺 Stability & Availability | 🔴 Critical | ⚡ Quick win
The hyphenated esptool commands require esptool v5.
chip-id, erase-flash, and write-flash are the v5 command names. esptool v5 renamed every command from _ to -, and the v4 parser accepts only the underscore names. The dependency pin allows 4.8. With esptool 4.x the preflight at line 288 fails in argparse, so every flash attempt aborts. The unit tests replace esptool with a stub, so they cannot catch this.
Details and the cross-file fix are in the consolidated comment.
🤖 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 `@src/meshchat/services/firmware.py` around lines 303 - 309, Update the esptool
command names used by the firmware flashing flow, including the preflight
command and both branches of the flash logic around full_install, to use the
underscore-form names supported by the pinned esptool v4 dependency (chip_id,
erase_flash, and write_flash). Keep the existing arguments, offsets, and
execution order unchanged.
| def _on_device_controls_updated(self, snapshot) -> None: | ||
| self._device_snapshot = snapshot | ||
| self._device_page.set_snapshot(snapshot) | ||
|
|
||
| def _on_firmware_flash_requested(self, bundle, full_install: bool, expected_usb) -> None: | ||
| snapshot = self._device_snapshot | ||
| if snapshot is None or not snapshot.serial_port: | ||
| QMessageBox.warning(self, "USB Radio Required", "Reconnect the radio over USB before flashing.") | ||
| return | ||
| self._flash_port = snapshot.serial_port | ||
| self._pending_flash = (bundle, full_install, expected_usb) | ||
| self._supervisor.cancel() | ||
| self._status_bar.showMessage("Releasing the USB port for firmware flashing…") | ||
| self._controller.disconnect() |
There was a problem hiding this comment.
🩺 Stability & Availability | 🟠 Major | ⚡ Quick win
A stale _device_snapshot can start a flash handoff that never resumes.
_device_snapshot is only assigned in _on_device_controls_updated. It is never cleared on disconnect, while DevicePage.set_connected(False) does clear the page state. Two consequences follow:
- If the radio is already disconnected,
_on_firmware_flash_requestedstill passes thesnapshot.serial_portguard at Line 612. It sets_pending_flashand_flash_port, then callsself._controller.disconnect().MeshtasticWorker.disconnectreturns immediately when the state is alreadyDISCONNECTED, sodisconnectedis never emitted._on_disconnectednever runs,_pending_flashstays set, and the flash never starts. DevicePage._confirm_flashhas already disabled both flash buttons.firmware_completednever fires, so the buttons are never re-enabled. The user has no feedback and no way to retry.
Clear _device_snapshot on disconnect, and start the flash directly when the controller is already disconnected.
🐛 Proposed fix
def _on_disconnected(self, reason: str) -> None:
...
self._device_page.set_connected(False)
+ self._device_snapshot = None
self._status_bar.showMessage(f"Disconnected — {reason}") def _on_firmware_flash_requested(self, bundle, full_install: bool, expected_usb) -> None:
snapshot = self._device_snapshot
if snapshot is None or not snapshot.serial_port:
QMessageBox.warning(self, "USB Radio Required", "Reconnect the radio over USB before flashing.")
return
self._flash_port = snapshot.serial_port
- self._pending_flash = (bundle, full_install, expected_usb)
self._supervisor.cancel()
+ if not self._is_connected:
+ # disconnect() is a no-op when already disconnected, so
+ # `disconnected` would never fire to resume the handoff.
+ self._firmware_controller.flash(bundle, self._flash_port, full_install, expected_usb)
+ return
+ self._pending_flash = (bundle, full_install, expected_usb)
self._status_bar.showMessage("Releasing the USB port for firmware flashing…")
self._controller.disconnect()🤖 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 `@src/meshchat/ui/main_window.py` around lines 606 - 619, Clear
_device_snapshot in the disconnect handling path, specifically _on_disconnected,
so stale serial-port data cannot authorize a flash after the radio is gone.
Update _on_firmware_flash_requested to detect an already-disconnected controller
and proceed with the pending flash handoff directly instead of relying on
disconnect to emit an event; preserve the existing connected flow through
self._controller.disconnect().
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 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 `@tests/test_firmware_service.py`:
- Around line 240-245: Update
test_installed_esptool_accepts_commands_used_by_flasher to wrap each
esptool.main([command, "--help"]) call in handling for SystemExit, asserting the
caught exception has code zero while allowing normal help termination. Preserve
the existing command iteration and version check.
🪄 Autofix
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: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: d11c06e7-7e28-41bf-b991-4ac98a10726f
📒 Files selected for processing (12)
packaging/orcmesh.specpyproject.tomlrequirements.txtsrc/meshchat/controllers/firmware_controller.pysrc/meshchat/controllers/meshtastic_controller.pysrc/meshchat/services/device_config.pysrc/meshchat/services/firmware.pysrc/meshchat/ui/device/device_page.pysrc/meshchat/ui/main_window.pytests/test_device_config.pytests/test_device_page.pytests/test_firmware_service.py
🚧 Files skipped from review as they are similar to previous changes (7)
- packaging/orcmesh.spec
- tests/test_device_config.py
- pyproject.toml
- src/meshchat/ui/device/device_page.py
- src/meshchat/ui/main_window.py
- src/meshchat/controllers/meshtastic_controller.py
- src/meshchat/services/device_config.py
| def test_installed_esptool_accepts_commands_used_by_flasher(): | ||
| esptool = pytest.importorskip("esptool") | ||
| assert int(version("esptool").split(".", 1)[0]) >= 5 | ||
| for command in ("chip-id", "erase-flash", "write-flash"): | ||
| esptool.main([command, "--help"]) | ||
|
|
There was a problem hiding this comment.
🎯 Functional Correctness | 🟠 Major | ⚡ Quick win
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
set -euo pipefail
python - <<'PY'
import esptool
for command in ("chip-id", "erase-flash", "write-flash"):
try:
esptool.main([command, "--help"])
except SystemExit as exc:
assert exc.code == 0, (command, exc.code)
else:
raise AssertionError(f"{command} did not exit from --help")
PYRepository: hardcoreerik/OrcMesh
Length of output: 278
🏁 Script executed:
set -euo pipefail
printf '%s\n' '--- candidate files ---'
git ls-files | grep -E '(^|/)(tests/test_firmware_service\.py|pyproject\.toml|requirements[^/]*|setup\.cfg|tox\.ini|.*\.ya?ml)$' | head -80
printf '%s\n' '--- test context ---'
sed -n '228,252p' tests/test_firmware_service.py
printf '%s\n' '--- esptool references ---'
rg -n -i 'esptool|pytest.importorskip' . --glob '!*.lock' --glob '!package-lock.json' | head -120Repository: hardcoreerik/OrcMesh
Length of output: 3176
🏁 Script executed:
set -euo pipefail
printf '%s\n' '--- dependency and test configuration ---'
sed -n '1,80p' pyproject.toml
printf '%s\n' '--- production esptool invocation ---'
sed -n '268,300p' src/meshchat/services/firmware.py
printf '%s\n' '--- esptool version-5 references in repository ---'
rg -n -C 3 'esptool\.main|SystemExit|--help' tests src pyproject.toml requirements.txtRepository: hardcoreerik/OrcMesh
Length of output: 5385
🌐 Web query:
espressif esptool v5.0.0 esptool.main standalone_mode SystemExit --help source
💡 Result:
In esptool v5.0.0 and subsequent versions, calling esptool.main to run commands programmatically often causes the Python process to exit because the underlying implementation is designed for CLI execution [1][2]. The behavior you are encountering is rooted in how Python command-line interfaces are typically built using the click library, where standalone_mode=True is the default [3]. In this mode, the application handles its own exceptions and automatically calls sys.exit upon completion or when displaying help messages [3]. Key points regarding this behavior in esptool v5.0.0: 1. Process Termination: When you call esptool.main with arguments (e.g., ['--chip', 'esp32', 'read_mac']), it executes as a standalone command, which forces the Python interpreter to terminate with a SystemExit exception (or simply exit the process) once the operation finishes [1][2]. 2. Programmatic Alternatives: To avoid this behavior when integrating esptool into your own scripts or applications, avoid calling esptool.main directly [4][2]. Instead, use the updated Python API provided in v5.0.0, which has been refactored for better programmatic access [5][6]. The official esptool documentation recommends using functions within the esptool.cmds module (e.g., esptool.cmds.connect_esp or esptool.cmds.detect_chip) for stable, non-terminating integration [4]. 3. Migration Context: Esptool v5.0.0 introduced significant breaking changes, including the refactoring of internal command dependencies and the removal of the.py suffix for console scripts [7][5]. If your workflow relies on legacy patterns, consult the official v5 Migration Guide for the recommended way to invoke tools and use the module interface [5][6]. If you must capture output or control execution flow without the process exiting, one workaround is to invoke the esptool script as a subprocess via subprocess.Popen rather than importing it as a module [2]. This isolates the exit behavior to the subprocess, allowing your main script to continue running [2].
Citations:
- 1: fix: stop exit 0 when being called programmatically espressif/esptool#1108
- 2: https://community.platformio.org/t/using-esptool-in-extrascript/45354
- 3: https://click.palletsprojects.com/en/7.x/api/?highlight=standalone_mode
- 4: https://docs.espressif.com/projects/esptool/en/latest/esp32/esptool/scripting.html
- 5: https://docs.espressif.com/projects/esptool/en/latest/esp32/migration-guide.html
- 6: https://docs.espressif.com/projects/esptool/en/latest/esp8266/migration-guide.html
- 7: https://github.com/espressif/esptool/releases/tag/v5.0.0
🌐 Web query:
site:github.com/espressif/esptool/blob/v5.0.0/esptool/__init__.py def main standalone_mode
💡 Result:
In esptool v5.0.0, main is defined at line 974:
def main(argv: list[str] | None = None, esp: ESPLoader | None = None):
args = expand_file_arguments(argv or sys.argv[1:])
cli(args=args, esp=esp)There is no standalone_mode parameter in this function; it invokes the Click command via cli(...). [1]
Handle the normal help exit from esptool.main().
Each --help invocation raises SystemExit(0) through Click's standalone CLI mode. Catch the expected exit and assert that its code is zero.
🤖 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 `@tests/test_firmware_service.py` around lines 240 - 245, Update
test_installed_esptool_accepts_commands_used_by_flasher to wrap each
esptool.main([command, "--help"]) call in handling for SystemExit, asserting the
caught exception has code zero while allowing normal help termination. Preserve
the existing command iteration and version check.
1 participant
Summary
Hardware verification
Tests
Summary by CodeRabbit