Kevin McCarthy [Tue, 20 Aug 2019 22:17:31 +0000 (15:17 -0700)]
Update autocrypt keyring documentation.
Recommend setting $autocrypt_dir to your normal keyring directory if
you want to use your existing key.
Trying to copy it over leads to signature verification issues. Even
if I reversed the order (which is much less clean), that would just
lead to missing key signature errors for Autocrypt messages instead.
Kevin McCarthy [Fri, 28 Jun 2019 22:09:51 +0000 (15:09 -0700)]
Add size display configuration variables.
Although it would be nice to "open this up" more, performance and
security/stability issues would arise.
Based on the thread in mutt-dev, I hope these vars will satisfy most
customization needs:
$size_show_bytes
$size_show_mb
$size_show_fractions
$size_units_on_left
Kevin McCarthy [Thu, 8 Aug 2019 01:34:58 +0000 (18:34 -0700)]
Turn off macro processing during autocrypt initialization.
The muttrc can push events into the macro buffer, with the assumption
that mailbox will be opened next. This will interfere with the
prompts uses during first run. The only issue is that macros won't
work inside the folder browser, if invoked.
Kevin McCarthy [Thu, 8 Aug 2019 01:24:15 +0000 (18:24 -0700)]
Allow nested setting of OPTIGNOREMACROEVENTS.
This will be needed for the first-run of autocrypt initialization.
The initialization involves a number of prompts, and can even use the
folder browser. The browser could in turn connect to IMAP which might
invoke a certificate prompt, login prompt, etc.
Kevin McCarthy [Fri, 19 Jul 2019 19:54:32 +0000 (12:54 -0700)]
Add autocrypt line to the compose menu.
Remove the hardcoded HDR_ATTACH offset calcuation, and add an explicit
enum for the "-- Attachments" line to make loops and padding array
sizes easier.
Add security and recommendataion fields on the line.
Add mutt_autocrypt_ui_recommendation, following the autocrypt spec
guidelines.
Kevin McCarthy [Fri, 12 Jul 2019 01:06:56 +0000 (18:06 -0700)]
Add autocrypt header parsing to mutt_parse_rfc822_line().
Convert parse_parameters() for autocrypt header usage:
* change to use a BUFFER to accomodate large autocrypt keydata
attribute values.
* Autocrypt header parameters are not rfc2231 compliant. Rather
than rolling another very similar function, just change the
existing one to allow space separation.
Kevin McCarthy [Sun, 28 Jul 2019 17:13:22 +0000 (10:13 -0700)]
Add a comment to the OPTIGNOREMACROEVENTS km_dokey() change.
The option was added in commit 53900afa, and its actual purpose was
to separate out an "unget" event buffer from the "macro" buffer, to
solve a problem with certificate prompts.
The safest approach in a low-level function like km_dokey() was to
return an error if new macros were generated when the option is set.
However, this results in an unbuffered username/password prompt being
aborted.
Currently the only users of unbuffered input are the SSL certificate
prompts, which use menu->dialog mode (and thus mutt_getch() directly)
and username/password prompts. So the only affected cases are
editor-menu prompts, and returning the pressed keys is likely less
surprising than aborting the prompt.
If other unbuffered menus are created in the future, we may want to
add a check for which menu mode is being used.
Kevin McCarthy [Fri, 26 Jul 2019 01:18:08 +0000 (18:18 -0700)]
Fix accidental fall-through for <quote-char> if aborted.
If <quote-char> was aborted, it was falling through to the
<transpose-chars> function.
This problem was introduced 19 years when the <transpose-chars> case
was moved below <quote-char>. Previously it fell through to the
default case which beeped.
Andrew Marks [Fri, 5 Jul 2019 15:44:21 +0000 (08:44 -0700)]
Added * option to unattachments command
The * option clears all previous attachments settings. A
list_free_generic method is added to muttlib to enable generic freeing
of specific LIST structures. free_attachments_data is used with
list_free_generic to clear four LISTs which manage allowed and excluded
inline and attached "attachments"
The User-Agent: header can be fun and interesting and useful for
debugging, but it also leaks quite a bit of information about the user
and their software stack.
This represents a potential security risk (attackers can target the
particular stack) and also an anonymity risk (a user trying to
preserve their anonymity by sending mail from a non-associated account
might reveal quite a lot of information if their choice of mail user
agent is exposed).
Users who want to configure `user_agent` to `yes` can still do so, but
it makes sense to have safer defaults.
Kevin McCarthy [Thu, 27 Jun 2019 22:35:12 +0000 (15:35 -0700)]
Remove unnecessary "" checks for DT_STR and DT_PATH MuttVars.
MuttVars of those types are set via safe_strdup(), which returns NULL
if the original is "". Thus Var implies *Var.
A good portion of the code relies on that axiom, but over the years
some (Var && *Var) checks have crept in, including from me.
This was partially because of the INITVAL("") that were in the code,
which implied (incorrectly) the initial value could be "". Commit 2f91d43e removed those to make it more clear.
This commit removes the *Var checks to make it even clearer, and help
avoid them creeping back in again.
Kevin McCarthy [Thu, 27 Jun 2019 18:06:19 +0000 (11:06 -0700)]
Convert $header_cache_pagesize to type DT_LNUM.
Prior to commit 4bc76c2f there was no LNUM type, and so the workaround
was to store it as a string, converting in the hcache_open_gdbm()
call.
This will not affect the user interface or config file, because DT_NUM
and DT_LNUM read in a string from the config file and convert to a
number. Quotes are used for escaping style, not passed through to the
variable setter.
So essentially this simply moves the conversion to parse_set(), and
provides feedback for a non-numeric type immediately.
Kevin McCarthy [Tue, 25 Jun 2019 21:25:33 +0000 (14:25 -0700)]
Convert Commands to use the union pointer_long_t too.
As with MuttVars, Commands was using an unsigned long to hold pointers
as well as enums. Convert to use the same union type of MuttVars.
Adjust command functions data parameter type to the union. Since
these are used outside init.c, relocate the union definition to
mutt.h.
Although the number of functions affected was long, most of them did
not need much adjustment. Many of them made no use of the parameter.
Those that did were easily cast into an added "data" variable at the
top.
Kevin McCarthy [Tue, 25 Jun 2019 01:07:54 +0000 (18:07 -0700)]
Convert MuttVars.data and .init to use a union type.
They were using an "unsigned long" and casting to a pointer when
needed. Obviously this has "worked" for a long time, but it's not
correct to assume a pointer can fit in unsigned long.
Replace with a union contain "void *p" and "long l". Fortunately, the
only parts making direct use of MuttVars are in init.h and init.c, so
we just need to update those manipulation functions.
In general I don't like single letter variables, but brevity is worth
it in this case.
Kevin McCarthy [Sat, 22 Jun 2019 20:33:11 +0000 (13:33 -0700)]
Remove mutt_rfc1524_expand_filename() return value and checks.
The return value was of dubious value. It returned 0 only for the
case that a nametemplate was specified and it already matched the
supplied oldfile.
However, just because the nametemplate matched does not mean
attachment handling in send-mode should skip the mutt_adv_mktemp()
conversion, which includes a call to mutt_sanitize_filename(). We
didn't do so if *no* nametemplate was supplied.
Remove the return value from the function, and remove the checks and
"special handing" in attach.c calls.
Kevin McCarthy [Sat, 22 Jun 2019 19:54:40 +0000 (12:54 -0700)]
Remove unnecessary strcmp for mutt_view_attachment().
mutt_rfc1524_expand_filename() runs the result through
mutt_adv_mktemp(), which will sanitize and relocate the filename under
$tmpdir. The strcmp() is unneeded and distracting to the program
logic; none of the other routines perform or need this check.
Kevin McCarthy [Sat, 22 Jun 2019 17:09:02 +0000 (10:09 -0700)]
Fix send-mode printing when expand_filename() returns 1.
It will return this when a nametemplate entry already matched the
passed in filename. So this bug required both a print entry and an
already matching nametemplate entry to trigger.
Kevin McCarthy [Fri, 21 Jun 2019 20:28:11 +0000 (13:28 -0700)]
Remove UL "" initialization from init.h.
mutt_init() calls mutt_set_default() followed by
mutt_restore_default(). The mutt_restore_default() calls
mutt_str_replace(), which translates "" into 0.
Therefore assigning "" to the option->init field simply wastes space
and (incorrectly) implies the option->data will be non-NULL by default.
Kevin McCarthy [Tue, 11 Jun 2019 20:15:22 +0000 (13:15 -0700)]
Add $fcc_before_send, defaulting unset.
When set, the message will be Fcc'ed the same as sent. $fcc_clear and
$fcc_attach will be ignored. This is because of the difficulty of
unwinding changes, notably Protected Headers, without potentially
breaking signatures.
Kevin McCarthy [Mon, 10 Jun 2019 21:51:50 +0000 (14:51 -0700)]
Allow imap_cmd_finish() to both expunge and fetch new mail.
Since commit dd327606 changed check_status setting to use bit
operators, and imap_check_mailbox() can call imap_cmd_finish() twice,
there is no reason to delay the processing of new mail until a second
call.
imap_read_headers() deals with msn_end < msg_begin, so remove
the (count > idata->max_msn) check. This will allow the reopen flag
to be reset if somehow it's not the case.
Kevin McCarthy [Fri, 7 Jun 2019 20:43:36 +0000 (13:43 -0700)]
Improve imap_append_message() error message handling.
If the rc is IMAP_CMD_BAD, then either idata->buf is stale or an error
message has already been printed (in cmd_handle_untagged()).
Use imap_next_word() to skip over the next two words instead of
directly skipping over SEQLEN, in case the buffer is in a different
format. We don't want to jump over the end of string.
Skip the mutt_error() if there is nothing to print.
Kevin McCarthy [Thu, 6 Jun 2019 23:03:42 +0000 (16:03 -0700)]
Enable the idata->check_status using bit operations.
Commit e3f66d7e fixed dropped new mail notications, removing the
unsetting of idata->reopen IMAP_NEWMAIL_PENDING in imap_cmd_finish()
when an EXPUNGE was processed.
However, imap_cmd_finish() can be called twice by
imap_check_mailbox(). First as part of the imap_exec(), and manually
again just below.
Now that the IMAP_NEWMAIL_PENDING still exists, a second call could
overwrite idata->check_status if both reopen flags were set.
This unfortunately affects update_index(), which behaves differently
for MUTT_REOPENED.
I need to change the return value of mx_check_mailbox() in master to
preserve all the bits, so the index can both notify of new mail and
update_index() properly.
For stable, the best fix is to use bit operators to enable the
check_status flags in imap_cmd_finish() (and cmd_parse_fetch for
flags), and keep the imap_check_mailbox() priority of setting its
return value (it prioritizes IMAP_EXCHANGE_PENDING).
Kevin McCarthy [Thu, 6 Jun 2019 20:38:03 +0000 (13:38 -0700)]
Improve robustness of imap_append_message().
First, check the imap_cmd_step() return value instead of looking at
idata->buf for "OK". If the connection bombed and imap_cmd_step()
returned IMAP_CMD_BAD, the value of idata->buf is stale.
If the server returned "+ OK" for the command continuation request
response, the call to imap_code(idata->buf) would even end up
returning true, despite that the append failed! (See #110, although
at the time of commit I can only hypothesize this is what is
happening.)
Second, check the status of the writes. flush_buffer() was not
passing the rc from mutt_socket_write_n(), which was further making
the above disaster scenerio possible.