Gordon Messmer wrote:
Sam Varshavchik wrote:
I think that ctlfile should be opened for append mode.
...capital idea. Am I wrong to assume that all of the calls to open()
in the openctl() function should have the same flags?
I'm not sure what you mean. And I wonder what sense may have
to truncate and append at the same time. (And O_TRUNC is not
needed on systems where O_CREAT | O_EXCL make the open fail
in case the file already exists.)
[...]
- O_WRONLY | O_TRUNC | O_CREAT | O_EXCL,
+ O_WRONLY | O_TRUNC | O_CREAT | O_EXCL | O_APPEND,
PERMISSION);
An lseek(fd, 0, SEEK_END) upon return from the filter should suffice.
To make the question clear to all, I include an example of a cltfile
I used to get the first times my filter attempted to write to it.
Althought the ctlfile is flushed, its read/write pointer is not adjusted.
| sa...@tana.it
| fdns; north (north.tana.it [194.243.254.162])
| e
| t
| M0003024A.403B5AF6.0000359B
| rcou...@north.tana.it
| R
| N
| E1078236505 |
| p1077660505 | these 36 bytes overwrite the head of filter's output
| W1077646105 |
| ... tail of filter's output goes here
| ...
| A1077632052
| ...
where the first letter on each line is the data type
(as defined in comctlfile.h)
N = COMCTLFILE_DSN is written by AddRecipient() that issues a sync().
E = COMCTLFILE_EXPIRES is written by closectl() after filtering.
p = COMCTLFILE_FAXEXPIRES ditto
W = COMCTLFILE_WARNING ditto
A = COMCTLFILE_NEXTATTEMPT is not in submit2.C
34 messages in net.sourceforge.lists.courier-users[courier-users] ctlfile (was: Re: Cou...
.patch
