| Commit message (Collapse) | Author | Age |
|
|
|
|
| |
- the key has been capitalized so compare with "Body" and not "body"
- comparison between strings is better done with == than with "is"
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
The email_as_string function, and the related RFC3156_canonicalize
function, are now used by the ForwardCommand and are not specific
anymore to the crypto routine. So we move them to the global helper
module.
fix an import removal mistake while moving email_as_string function: StringIO was not only used by email_as_string
|
| |
|
| |
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| | |
regarding utf8 chars in the config being read as str, not unicode
in python v2.7. This pathc introduces an additional
`alot.helper.string_decode` around the translated tagname read from the
config, and fixes issue #693. In the long run, we expect configobj
to be fully backweards compatible.
|
| |
| |
| |
| |
| | |
that will make the message widget display the given text instead
of the message's body text
|
|/
|
|
| |
cf issue #701
|
|
|
|
|
|
|
|
|
|
| |
This ensures the default initial command as given by the
`initial_command` config option will be interpreted as
command*line* as expected, and not as single command.
As a result, one can now use e.g.
initial_command = search A ; search B
issue: #698
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
parse_mailto parses mailto strings into headers,body and
mailto_to_envelope constructs an alot.db.envelope.Envelope
for a give mailto string
|
| |
|
| |
|
| |
|
|
|
|
| |
Add a .eml extension to let $EDITOR known the file as a mail.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This ensures that calls to commands.commandfactory
are done just before these are applied:
In command sequences, some previous cmd might have changed
the mode, which in turn influences the commandfactory
as some commandstrings are interpreted differently in different modes.
This patch ensures that one command has completed before a next
one gets instantiated.
Also, this adds a global error handler for command sequence deferreds.
cf issue #629
|
| |
|
|
|
|
|
|
| |
... in search.RefinePromptCommand,
bufferlist.BufferCloseCommand
and globals.BufferCloseCommand
|
| |
|
|
|
|
|
|
|
| |
When a pre or post command hook is called it gets
the Command instance it belongs to as keyword parameter "cmd".
This allows e.g. the post hook of TagCommands to access the
query string used in this particular instance.
|
|
|
|
| |
cf #629
|
| |
|
| |
|
|
|
|
| |
and obsolete commands.globals.CommandSequenceCommand
|
|
|
|
|
|
|
|
| |
being run
Conflicts:
alot/commands/globals.py
alot/ui.py
|
|
|
|
|
| |
that sets a whitelist of tag strings to be displayed
cf issue #662
|
|
|
|
| |
Expand docstring and explain handling of ambiguous keyid
|
|
|
|
|
|
|
|
|
| |
Previously when Context.get_key raised gpgme.ERR_AMBIGUOUS_NAME crypto.get_key
would not verify the ambiguous keys are actually valid for
signing/encryption.
Now when gpgme.get_key raises ERR_AMBIGUOUS_NAME we elimitate those invalid
keys. If there is only one valid key left we return it, otherwise re-raise
|
| |
|
|
|
|
| |
(and so avoid having a leading comma)
|
|
|
|
| |
So the tags are in a predictable order.
|
|
|
|
|
| |
So we don't have to remember to type the extra comma before typing a new
tag.
|
|\
| |
| | |
Check for "Re:" at start of subject in case insensitive manner
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
To avoid a subject ending up like:
"Re: RE: RE: RE: original subject"
That I've seen a few times.
The fix is to lower case the subject and the strings to compare against
before calling startwith()
|
|\ \ |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
If one replaces some widget without letting urwid know,
one has to manually trigger it to redraw the screen.
We enforce this to happen after focus_first and focus_last
commands, which might replace widgets in search buffers.
For sake of performance this is *not* done after all other
focus movements.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
In case a search buffer displays more than 200 threads
and one calls `move last`, this will now result in a
new search with reversed search order displayed using a
reversed PipeWalker.
This makes it unnecessary to read all thread id's
when focussing the last element and thus speeds up the UI
considerably.
|