[Bug 277664] japanese/mh: inc command fails to connect with APOP on amd64

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=277664

--- Comment #1 from WATANABE Kazuhiro <CQG00620@nifty.ne.jp> ---
Hello.

Thanks for your report!  I've reproduced the same problem on my amd64
environment.

 * Without the patch

 $ uname -srm
 FreeBSD 14.0-RELEASE-p5 amd64
 $ inc -apop -user bena -host localhost
 Password (localhost:bena): 
 inc: -ERR [AUTH] authentication failure
 $ 

qpopper(8) says:

| Mar 25 16:34:16 aquarius-vm qpopper[2688]: Debugging turned on (-d)
[pop_init.c:824]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: (v4.1.0) Servicing request from
"localhost" at 127.0.0.1 [pop_init.c:1422]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: before TLS; tls_support==0
[popper.c:199]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: Skipped TLS Init [popper.c:223]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: (v4.1.0) Intro [popper.c:273]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: +OK Qpopper (version 4.1.0) at
aquarius-vm.sign.local starting.  <2688.1711352056@aquarius-vm.sign.local>
[popper.c:286]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: Qpopper ready for input from
(null) at localhost [127.0.0.1] [popper.c:320]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: Received (42): "APOP bena
b22a9ce93c7af8d4bd0ff102de926ca2" [pop_get_command.c:105]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: apop "bena" [pop_apop.c:214]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: ...checknonauthfile didn't match
user bena in file /usr/local/etc/qpopper/popusers [pop_pass.c:232]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: home (10): '/home/bena'
[pop_apop.c:233]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: bena at localhost (127.0.0.1):
-ERR [AUTH] authentication failure [pop_apop.c:369]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: apop returned 0; CurrentState now
halt [popper.c:364]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: +OK Pop server at
aquarius-vm.sign.local signing off. [popper.c:386]
| Mar 25 16:34:16 aquarius-vm qpopper[2688]: (v4.1.0) Ending request from
"bena" at (localhost) 127.0.0.1 [popper.c:412]

 * With the patch

inc(1) works correctly with APOP.

 $ inc -apop -user bena -host localhost
 Password (localhost:bena): 
 inc: no mail to incorporate
 $ 

qpopper(8) says:

| Mar 25 16:35:12 aquarius-vm qpopper[2789]: Debugging turned on (-d)
[pop_init.c:824]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: (v4.1.0) Servicing request from
"localhost" at 127.0.0.1 [pop_init.c:1422]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: before TLS; tls_support==0
[popper.c:199]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: Skipped TLS Init [popper.c:223]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: (v4.1.0) Intro [popper.c:273]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: +OK Qpopper (version 4.1.0) at
aquarius-vm.sign.local starting.  <2789.1711352112@aquarius-vm.sign.local>
[popper.c:286]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: Qpopper ready for input from
(null) at localhost [127.0.0.1] [popper.c:320]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: Received (42): "APOP bena
1a02a9120ba982e1e148029e3b51cfe7" [pop_get_command.c:105]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: apop "bena" [pop_apop.c:214]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: ...checknonauthfile didn't match
user bena in file /usr/local/etc/qpopper/popusers [pop_pass.c:232]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: home (10): '/home/bena'
[pop_apop.c:233]
| Mar 25 16:35:12 aquarius-vm qpopper[2789]: APOP authentication ok for "bena"
[pop_apop.c:373]
(snip)

 * Rewrite the patch

For more portability, I rewrite the patch like this:

 --- h/md5.h.orig       1999-02-09 11:30:00 UTC
 +++ h/md5.h
 @@ -20,11 +20,15 @@ typedef unsigned short int UINT2;
  typedef unsigned short int UINT2;

  /* UINT4 defines a four byte word */
 +#if 0
  #ifndef __alpha
  typedef unsigned long int UINT4;
  #else
  typedef unsigned int UINT4;
  #endif
 +#endif
 +#include <sys/types.h>
 +typedef uint32_t UINT4;

  /* PROTO_LIST is defined depending on how PROTOTYPES is defined above.
  If using PROTOTYPES, then PROTO_LIST returns the list, otherwise it

I've tested the new patch on 14.0-RELEASE/amd64, 14.0-RELEASE/i386,
and 15-CURRENT/amd64 environment.  It works fine.

-- 
You are receiving this mail because:
You are the assignee for the bug.