unable to run this in aws lambda

[Cryptogevity]

I try to run this in my aws lambda, and i want to change the output path to /tmp because AWS lambda only allow write in /tmp folder

and i build a docker image for this with this command

RUN mkdir -p /tmp/audio/tmp /tmp/data/transcripts /tmp/data/audio
&& sed -i 's|temp_audio_dir: "data/audio/tmp/"|temp_audio_dir: "/tmp/audio/tmp"|' $(python3 -c "import podcastfy; print(podcastfy.file.rsplit('/', 1)[0])")/conversation_config.yaml
&& sed -i 's|transcripts: "./data/transcripts"|transcripts: "/tmp/data/transcripts"|' $(python3 -c "import podcastfy; print(podcastfy.file.rsplit('/', 1)[0])")/conversation_config.yaml
&& sed -i 's|audio: "./data/audio"|audio: "/tmp/data/audio"|' $(python3 -c "import podcastfy; print(podcastfy.file.rsplit('/', 1)[0])")/conversation_config.yaml

to change the configuration yaml

however, i still got this issue

podcastfy.client - ERROR - An error occurred in the process_content function: [Errno 30] Read-only file system: '/var/lang/lib/python3.11/site-packages/podcastfy/tmp'

seems like the /tmp is still under the current folder, is there anyway to configure the output path to absolute path ?

[souzatharsis]

Hi, could you please share your configuration yml file? <http://linkedin.com/in/tharsissouza>

…

On Fri, Nov 22, 2024 at 2:26 PM Cryptogevity Team ***@***.***> wrote: I try to run this in my aws lambda, and i want to change the output path to /tmp because AWS lambda only allow write in /tmp folder and i build a docker image for this with this command RUN mkdir -p /tmp/audio/tmp /tmp/data/transcripts /tmp/data/audio && sed -i 's|temp_audio_dir: "data/audio/tmp/"|temp_audio_dir: "/tmp/audio/tmp"|' $(python3 -c "import podcastfy; print(podcastfy.*file*.rsplit('/', 1)[0])")/conversation_config.yaml && sed -i 's|transcripts: "./data/transcripts"|transcripts: "/tmp/data/transcripts"|' $(python3 -c "import podcastfy; print(podcastfy. *file*.rsplit('/', 1)[0])")/conversation_config.yaml && sed -i 's|audio: "./data/audio"|audio: "/tmp/data/audio"|' $(python3 -c "import podcastfy; print(podcastfy.*file*.rsplit('/', 1)[0])")/conversation_config.yaml to change the configuration yaml however, i still got this issue podcastfy.client - ERROR - An error occurred in the process_content function: [Errno 30] Read-only file system: '/var/lang/lib/python3.11/site-packages/podcastfy/tmp' seems like the /tmp is still under the current folder, is there anyway to configure the output path to absolute path ? — Reply to this email directly, view it on GitHub <#189>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADTMY3KVVRVDGDK7H6QSPKL2B5SNRAVCNFSM6AAAAABSJ2AEG2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGY4DGOJXHEYTCNQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[Cryptogevity]

i think i found the issue, i created duplicate /tmp folder caused it points to the wrong folder, thanks for the reply

[Cryptogevity]

I just double checked my code and i believe my code is correct, i used the exact same code

`
instructor_config = {
"word_count": 1000, # Longer to allow for detailed arguments
"conversation_style": ["enthusiastic"], # Appropriate for academic discourse
"roles_person1": "main summarizer", # Presents the main argument
"roles_person2": "audience participator", # Challenges the thesis
"dialogue_structure": [
"summarize the student data, mention a few termologies and concepts that students are struggling with",
], # Mimics a structured debate format
"podcast_name": "podcast ",
"podcast_tagline": "podcast",
"engagement_techniques": [
"normal questions"
], # Techniques to stimulate critical thinking
"creativity": 0.2 # Low creativity to maintain focus on facts and logic
}

formatted_text = "\n".join(text_output)

# Print the formatted text
generate_podcast(text=formatted_text, conversation_config=instructor_config)

`

to generate the podcast. The old version I used is 0.2.17, and it works totally fine in AWS lambda, and the latest version 0.4.0, i noticed that you have a new folder under podcast data/audio/tmp/tmpibdcmvin, maybe some code change as well, this new version will try to write something into

/podcastfy/tmp during the podcast generation.

Could you please look into this?

[souzatharsis]

Are you still getting the same error message? podcastfy.client - ERROR - An error occurred in the process_content function: [Errno 30] Read-only file system: '/var/lang/lib/python3.11/site-packages/podcastfy/tmp'

…

On Fri, Nov 22, 2024, 6:17 PM Cryptogevity Team ***@***.***> wrote: Reopened #189 <#189>. — Reply to this email directly, view it on GitHub <#189 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADTMY3PB4WTQUD2F3RYIJVT2B6NNRAVCNFSM6AAAAABSJ2AEG2VHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJVGQYDGOJTHE3TMMI> . You are receiving this because you commented.Message ID: ***@***.***>

[Cryptogevity]

if I use 0.2.17 i won't see this error.

but i use 0.4.0 i see that error

[souzatharsis]

Hi, I see that you are not passing your custom tmp folder in the config file. That should be the issue.

…

On Fri, Nov 22, 2024, 7:36 PM Cryptogevity Team ***@***.***> wrote: if I use 0.2.17 i won't see this error. but i use 0.4.0 i see that error — Reply to this email directly, view it on GitHub <#189 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADTMY3LJJD42KYC7UHOSBHT2B6WWZAVCNFSM6AAAAABSJ2AEG2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIOJUHE4DKMJTGE> . You are receiving this because you commented.Message ID: ***@***.***>

[Cryptogevity]

Thanks for your reply, may i know how to pass the tmp folder in the config

from my code

RUN mkdir -p /tmp/audio/tmp /tmp/data/transcripts /tmp/data/audio && sed -i 's|temp_audio_dir: "data/audio/tmp/"|temp_audio_dir: "/tmp/audio/tmp"|' $(python3 -c "import podcastfy; print(podcastfy.file.rsplit('/', 1)[0])")/conversation_config.yaml && sed -i 's|transcripts: "./data/transcripts"|transcripts: "/tmp/data/transcripts"|' $(python3 -c "import podcastfy; print(podcastfy.file.rsplit('/', 1)[0])")/conversation_config.yaml && sed -i 's|audio: "./data/audio"|audio: "/tmp/data/audio"|' $(python3 -c "import podcastfy; print(podcastfy.file.rsplit('/', 1)[0])")/conversation_config.yaml

i already change the config inside conversion_config.yaml, are there any other config that i need to change ?

any code samples will be very helpful.