Tests: USBSerial: Handle minimal printf limitations

[fkjagodzinski]

Summary of changes

In a line coding test, host changes the line coding based on data received from DUT via the USBSerial::printf(). A fixed-size payload, equal to LINE_CODING_STRLEN, is required by the host side. When the minimal printf is used, fixed-width messages can not be easily generated on the fly. Zero-fill or width specifiers are not supported by the minimal printf.

Update the host side to stop reading on a defined payload delimiter and allow variable-width messages.

For more details, please see a related issue #12718.

Impact of changes

Migration actions required

Documentation

None

---

Pull request type

[x] Patch update (Bug fix / Target update / Docs update / Test update / Refactor)
[] Feature update (New feature / Functionality change / New API)
[] Major update (Breaking change E.g. Return code change / API behaviour change)

---

Test results

[] No Tests required for this change (E.g docs only update)
[x] Covered by existing mbed-os tests (Greentea or Unittest)
[] Tests / results supplied as part of this PR

---

Reviewers

@jamesbeyond, @evedon, @maciejbocianski

---

[ciarmcom]

@fkjagodzinski, thank you for your changes.
@maciejbocianski @jamesbeyond @evedon @ARMmbed/mbed-os-maintainers please review.

[evedon]

I think we should have the same test for minimal-printf and standard C printf and change the test assertion TEST_ASSERT_EQUAL_INT(LINE_CODING_STRLEN, rc);
You could test that rc is between a minimal value and LINE_CODING_STRLEN.

Pull request has been modified.

[fkjagodzinski]

I can see I forgot to mention the reasoning behind the fixed-size USBSerial payloads. Sorry about that. Commit message and code comment already updated.

Actually, it is the host side that requires fixed-size payloads. Please see

mbed-os/TESTS/host_tests/usb_device_serial.py

Lines 272 to 281 in f0d04c8

	payload = six.ensure_str(mbed_serial.read(LINE_CODING_STRLEN))
	while len(payload) == LINE_CODING_STRLEN:
	baud, bits, parity, stop = (int(i) for i in payload.split(','))
	new_line_coding = {
	'baudrate': baud,
	'bytesize': self._BYTESIZES[bits],
	'parity': self._PARITIES[parity],
	'stopbits': self._STOPBITS[stop]}
	mbed_serial.apply_settings(new_line_coding)
	payload = six.ensure_str(mbed_serial.read(LINE_CODING_STRLEN))

Thanks for the review @evedon.

[fkjagodzinski]

Actually, I can see we could use https://pyserial.readthedocs.io/en/latest/pyserial_api.html#serial.Serial.read_until on the host side.

[evedon]

Actually, I can see we could use https://pyserial.readthedocs.io/en/latest/pyserial_api.html#serial.Serial.read_until on the host side.

It would be better to change the host side not to expect fixed-size USBSerial payloads.

[fkjagodzinski]

@evedon I updated the patch as discussed.

[fkjagodzinski]

@rajkan01 could you also review, please?

[evedon]

Minor comment but code change looks good.
Thanks!

[evedon]

This is not needed in this file

[mbed-ci]

Test run: SUCCESS

Summary: 4 of 4 test jobs passed
Build number : 1
Build artifacts

[rajkan01]

LGTM

[0xc0170]

Will be merged as soon as we have 6.0.0 beta1 released