rtt: Add test suite coverage for RFC 4103

[AlexisHadj]

Introduces comprehensive testing for RFC 4103 Real-Time Text (RTT) within the asterisk-testsuite framework.

1. Implements automated RTT content verification.
2. Adds multi-language text stream handling validations.
3. Integrates tests with combinations of disabled/ enabled audio.

[AlexisHadj]

cherry-pick-to: 20
cherry-pick-to: 22
cherry-pick-to: 23
asterisk-test-pr: 1128

[github-actions]

Workflow Check completed successfully

[github-actions]

Workflow Check completed successfully

[github-actions]

Workflow Check completed successfully

[github-actions]

Workflow Check completed successfully

[henningw]

Feedback about the directory structure would be appreciated, we probably should adapt it better to the existing structure.
The tests were also briefly discussed in this comment asterisk/asterisk#1128 (comment)

[mbradeen]

Feedback about the directory structure would be appreciated, we probably should adapt it better to the existing structure. The tests were also briefly discussed in this comment asterisk/asterisk#1128 (comment)

I would probably put these in tests/channels/pjsip/rtp/rtt as they are pjsip specific.

Running the tests locally, I found a few of them failed against the 1128 branch:

=== TEST RESULTS ===

PATH: /usr/src/asterisk/testsuite/.venv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin

--> tests/rtt/rtt_red --- PASSED
--> tests/rtt/rtt_red_audio --- PASSED
--> tests/rtt/rtt_red_t140 --- FAILED
--> tests/rtt/rtt_red_t140_audio --- FAILED
--> tests/rtt/rtt_t140 --- FAILED
--> tests/rtt/rtt_t140_audio --- PASSED
--> tests/rtt/rtt_t140_red --- FAILED
--> tests/rtt/rtt_t140_red_audio --- PASSED

Is this expected? The failing tests have a bunch of output from pjsua but I'm not sure what the particular failure reason is.

[mbradeen]

Sorry, just saw asterisk/asterisk#1968

Interestingly text only works in my own pjsua based UA - I will compare traces with 1968

[henningw]

Feedback about the directory structure would be appreciated, we probably should adapt it better to the existing structure. The tests were also briefly discussed in this comment asterisk/asterisk#1128 (comment)

I would probably put these in tests/channels/pjsip/rtp/rtt as they are pjsip specific.

Running the tests locally, I found a few of them failed against the 1128 branch:

=== TEST RESULTS ===

PATH: /usr/src/asterisk/testsuite/.venv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin

--> tests/rtt/rtt_red --- PASSED
--> tests/rtt/rtt_red_audio --- PASSED
--> tests/rtt/rtt_red_t140 --- FAILED
--> tests/rtt/rtt_red_t140_audio --- FAILED
--> tests/rtt/rtt_t140 --- FAILED
--> tests/rtt/rtt_t140_audio --- PASSED
--> tests/rtt/rtt_t140_red --- FAILED
--> tests/rtt/rtt_t140_red_audio --- PASSED

Is this expected? The failing tests have a bunch of output from pjsua but I'm not sure what the particular failure reason is.

Hello @mbradeen, thanks for the feedback. Yes, we will adapt the directory structure, make sense.
About the failed test - yes this is related to the linked issue. Its something specific how the pjsua does the deactivation of the audio, you will see in the traces for sure.

[mbradeen]

My test softphone, which uses pjsip underneath, only sends a text m line in rtt mode. This bypasses the problem in 1968. So rtt can work in text-only mode with Asterisk as long as you don't have a disabled audio stream. I will take a look at 1968 as part of this overall review.

[henningw]

My test softphone, which uses pjsip underneath, only sends a text m line in rtt mode. This bypasses the problem in 1968. So rtt can work in text-only mode with Asterisk as long as you don't have a disabled audio stream. I will take a look at 1968 as part of this overall review.

This is the point, exactly. For (a bit outdated) more test details refer e.g. to this comment from April link .

[mbradeen]

asterisk/asterisk#2025 is required for this test to run on our CI