b4 and lore and '/' in a message id

I've run into problems with a specific user that happens to send patches out with a '/' in the message id field.

Quoted 2 lines · Greg KH

I've run into problems with a specific user that happens to send patches out with a '/' in the message id field.

There's some interesting interaction between nginx, varnish, and public-inbox that results in this behaviour. There is a way to solve it, but I don't like the approach, so I'm going to work with upstream to see what my options are.

I've run into problems with a specific user that happens to send patches out with a '/' in the message id field.

I think this is a valid thing, but b4 and lore don't seem to handle this very well.

Here's their latest patch:

$ b4 am YTRuXTu/kePBDwAF@user
Looking up https://lore.kernel.org/r/YTRuXTu%2FkePBDwAF%40user
That message-id is not known.

And yes, if you go to the above link, it is not found.

But, if you go to 'all' it is found:
https://lore.kernel.org/all/YTRuXTu%2FkePBDwAF@user

Any hints on what to do here?

Note, this is the second time this has happened. First time I wrongly blamed the developer that they submitted the patch incorrectly, this time they graciously pointed out that they had sent it correctly, something was wrong with my side :)

Saurav, how are you sending these patches out, using 'git send-email' or something else? What is generating that message id?

thanks,

greg k-h

Quoted 24 lines · Greg KH

I've run into problems with a specific user that happens to send patches out with a '/' in the message id field.

I think this is a valid thing, but b4 and lore don't seem to handle this very well.

Here's their latest patch:

$ b4 am YTRuXTu/kePBDwAF@user
Looking up https://lore.kernel.org/r/YTRuXTu%2FkePBDwAF%40user
That message-id is not known.

And yes, if you go to the above link, it is not found.

But, if you go to 'all' it is found:
https://lore.kernel.org/all/YTRuXTu%2FkePBDwAF@user

Any hints on what to do here?

Note, this is the second time this has happened. First time I wrongly blamed the developer that they submitted the patch incorrectly, this time they graciously pointed out that they had sent it correctly, something was wrong with my side :)

Quoted 7 lines

Saurav, how are you sending these patches out, using 'git send-email' or something else? What is generating that message id?

thanks,

greg k-h

Hi Greg,

I am using mutt to send the patch. Below is muttrc settings.

unset use_domain # because joe@localhost is just embarrassing

set realname = "SAURAV GIREPUNJE"

set from = "[email protected]"

set use_from = yes

I am using "mutt -H" command to send the patch and that is creating issue on message id.
See below link:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=833192

I've run into problems with a specific user that happens to send patches out with a '/' in the message id field.

I think this is a valid thing, but b4 and lore don't seem to handle this very well.

Quoted 5 lines · Greg KH

I've run into problems with a specific user that happens to send patches out with a '/' in the message id field.

I think this is a valid thing, but b4 and lore don't seem to handle this very well.

The latest public-inbox version allows us to do the right thing here, so things work as expected now:

	$ b4 am YTRuXTu/kePBDwAF@user
	Looking up https://lore.kernel.org/r/YTRuXTu%2FkePBDwAF%40user
	Analyzing 5 messages in the thread
	Checking attestation on all messages, may take a moment...
	---
	  ✓ [PATCH] staging: r8188eu: include: remove duplicate declaration.
	  ---
	  ✓ Signed: DKIM/gmail.com
	---
	Total patches: 1
	---
	 Link: https://lore.kernel.org/r/YTRuXTu/kePBDwAF@user
	 Base: applies clean to current tree
		   git am ./20210905_saurav_girepunje_staging_r8188eu_include_remove_duplicate_declaration.mbx

Best regards,
-K