From: Mark Walters <markwalters1009@gmail.com>
To: Tomi Ollila <tomi.ollila@iki.fi>, Michal Sojka <sojkam1@fel.cvut.cz>,
	notmuch@notmuchmail.org
Subject: Re: [PATCH v5 0/7] notmuch search --output=sender/recipients
In-Reply-To: <m24mukyyoz.fsf@guru.guru-group.fi>
References: <1414713573-21461-1-git-send-email-sojkam1@fel.cvut.cz>
	<878ujwvctt.fsf@qmul.ac.uk> <8761f0v8v7.fsf@qmul.ac.uk>
	<87vbn06yry.fsf@steelpick.2x.cz>
	<m24mukyyoz.fsf@guru.guru-group.fi>
User-Agent: Notmuch/0.18.1+86~gef5e66a (http://notmuchmail.org) Emacs/23.4.1
	(x86_64-pc-linux-gnu)
Date: Fri, 31 Oct 2014 16:54:14 +0000
Message-ID: <87wq7gtc21.fsf@qmul.ac.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Precedence: list


On Fri, 31 Oct 2014, Tomi Ollila <tomi.ollila@iki.fi> wrote:
> On Fri, Oct 31 2014, Michal Sojka <sojkam1@fel.cvut.cz> wrote:
>
>> On Fri, Oct 31 2014, Mark Walters wrote:
>>> My only query is in the text output: should the name part be printed as
>>> a quoted string. For example currently I get a line of the form
>>>
>>> Bloggs, Fred <fred@example.com>
>>
>> Good point.
>
> There has been some discussion on this issue on IRC channel, and the
> opinion of most (seen so far) is that output the parts without quoting
> (i.e. just like done in this patch series)...

Just to confirm I am happy either way with the quoting. I think it would
be good not to change the text format after it goes into mainline
though.=20

Best wishes

Mark





>
> Taken the other example from Mark's earlyer email:
>
> Bloggs <the king>, Fred <fred@example.com>
>
> echo '^^^' | sed 's/.*</</' would leave only the address part (with <>:s)=
 (*)
>
> echo '^^^' | sed 's/ <[^<]*$//' would leave only the name part
>
> and regexp '\(.*\) <\(.*\)>' or pcre-compatible /(.*)\s<(.*)>/
> would capture name & address parts...
>
> (all of the above untested, though;)
>
> In case instead of 'name <addr>' there is only 'addr', then above
> sed lines return the same (full) string and the regexps just don't
> match.
>
> There were some suggestions how the text output could be changed on IRC;
> if anyone wishes to bring those forward, please do so :D
>
> So, IMO, this issue is not showstopper in this series (if anything is);
> patches 1-6 looks good to me on paper, but I have not tested those yet.
>
> Tomi
>
> (*) echo '^^^' | sed 's/.*<//; s/>.*//' would drop <>'s from first example
>
>
>>
>> On Fri, Oct 31 2014, Mark Walters wrote:
>>> Hi
>>>
>>> I attach a patch which does the quoting for real names but I don't know
>>> if we want it: it changes (example taken from the test suite)
>>>
>>> Fran=C3=A7ois Boulogne to
>>>
>>> =3D?iso-8859-1?q?Fran=3DE7ois?=3D Boulogne
>>>
>>> (which is what was in the original email)
>>>
>>> Plausibly the best thing is just to leave the series as is, so the
>>> text output is readable and parseable in the common cases.
>>>
>>> Anyway the patch is attached if anyone wants to experiment.
>>>
>>> Best wishes
>>>
>>> Mark
>>>
>>> From 53b1ced2d6a9fbbba93448325f795e6b99faa240 Mon Sep 17 00:00:00 2001
>>> From: Mark Walters <markwalters1009@gmail.com>
>>> Date: Fri, 31 Oct 2014 10:11:40 +0000
>>> Subject: [PATCH] search: quote real names for output=3Dsender/recipient=
 in text
>>>  format
>>>
>>> This quotes the real name (when gmime thinks appropriate) for the text
>>> output. For the other outputs the real name is separate from the
>>> address so the consumer can do any quoting necessary.
>>> ---
>>>  notmuch-search.c           |    8 ++++----
>>>  test/T090-search-output.sh |    8 ++++----
>>>  2 files changed, 8 insertions(+), 8 deletions(-)
>>>
>>> diff --git a/notmuch-search.c b/notmuch-search.c
>>> index eae749a..8eac161 100644
>>> --- a/notmuch-search.c
>>> +++ b/notmuch-search.c
>>> @@ -47,6 +47,7 @@ typedef struct {
>>>  typedef struct {
>>>      const char *name;
>>>      const char *addr;
>>> +    const char *string;
>>>  } mailbox_t;
>>>=20=20
>>>  /* Return two stable query strings that identify exactly the matched
>>> @@ -255,15 +256,13 @@ print_mailbox (const search_options_t *opt, const=
 mailbox_t *mailbox)
>>>  {
>>>      const char *name =3D mailbox->name;
>>>      const char *addr =3D mailbox->addr;
>>> +    const char *string =3D mailbox->string;
>>>      sprinter_t *format =3D opt->format;
>>>=20=20
>>>      if (format->is_text_printer) {
>>>  	char *mailbox_str;
>>>=20=20
>>> -	if (name && *name)
>>> -	    mailbox_str =3D talloc_asprintf (format, "%s <%s>", name, addr);
>>> -	else
>>> -	    mailbox_str =3D talloc_strdup (format, addr);
>>> +	mailbox_str =3D talloc_strdup (format, string);
>>>=20=20
>>>  	if (! mailbox_str) {
>>>  	    fprintf (stderr, "Error: out of memory\n");
>>> @@ -309,6 +308,7 @@ process_address_list (const search_options_t *opt, =
GHashTable *addrs,
>>>  	    mailbox_t mbx =3D {
>>>  		.name =3D internet_address_get_name (address),
>>>  		.addr =3D internet_address_mailbox_get_addr (mailbox),
>>> +		.string =3D internet_address_to_string (address, TRUE),
>>
>> I'd prefer having the second parameter (encode) FALSE. This will still
>> add quotes when necessary, but does not encode non-ascii characters so
>> the result would be human readable.
>>
>> Another question is whether to add .string to mailbox_t. In this patch
>> it doesn't matter, but if --output=3Dcount patch will be merged, this
>> would mean that memory consumption doubles, because with --output=3Dcount
>> the addresses are kept in memory and printed only after the search is
>> completed. It would be therefore better to construct a new
>> InternetAddressMailbox from name and addr in print_mailbox() and perform
>> the conversion to string there. Thoughts?
>>
>> Thanks,
>> -Michal
>>
>>>  	    };
>>>=20=20
>>>  	    if (is_duplicate (opt, addrs, mbx.name, mbx.addr))
>>> diff --git a/test/T090-search-output.sh b/test/T090-search-output.sh
>>> index 841a721..776e3f4 100755
>>> --- a/test/T090-search-output.sh
>>> +++ b/test/T090-search-output.sh
>>> @@ -390,7 +390,7 @@ test_expect_equal_file OUTPUT EXPECTED
>>>  test_begin_subtest "--output=3Dsender"
>>>  notmuch search --output=3Dsender '*' >OUTPUT
>>>  cat <<EOF >EXPECTED
>>> -Fran=C3=A7ois Boulogne <boulogne.f@gmail.com>
>>> +=3D?iso-8859-1?q?Fran=3DE7ois?=3D Boulogne <boulogne.f@gmail.com>
>>>  Olivier Berger <olivier.berger@it-sudparis.eu>
>>>  Chris Wilson <chris@chris-wilson.co.uk>
>>>  Carl Worth <cworth@cworth.org>
>>> @@ -437,7 +437,7 @@ test_begin_subtest "--output=3Drecipients"
>>>  notmuch search --output=3Drecipients '*' >OUTPUT
>>>  cat <<EOF >EXPECTED
>>>  Allan McRae <allan@archlinux.org>
>>> -Discussion about the Arch User Repository (AUR) <aur-general@archlinux=
.org>
>>> +"Discussion about the Arch User Repository (AUR)" <aur-general@archlin=
ux.org>
>>>  olivier.berger@it-sudparis.eu
>>>  notmuch@notmuchmail.org
>>>  notmuch <notmuch@notmuchmail.org>
>>> @@ -449,9 +449,9 @@ test_expect_equal_file OUTPUT EXPECTED
>>>  test_begin_subtest "--output=3Dsender --output=3Drecipients"
>>>  notmuch search --output=3Dsender --output=3Drecipients '*' >OUTPUT
>>>  cat <<EOF >EXPECTED
>>> -Fran=C3=A7ois Boulogne <boulogne.f@gmail.com>
>>> +=3D?iso-8859-1?q?Fran=3DE7ois?=3D Boulogne <boulogne.f@gmail.com>
>>>  Allan McRae <allan@archlinux.org>
>>> -Discussion about the Arch User Repository (AUR) <aur-general@archlinux=
.org>
>>> +"Discussion about the Arch User Repository (AUR)" <aur-general@archlin=
ux.org>
>>>  Olivier Berger <olivier.berger@it-sudparis.eu>
>>>  olivier.berger@it-sudparis.eu
>>>  Chris Wilson <chris@chris-wilson.co.uk>
>>> --=20
>>> 1.7.10.4
>> _______________________________________________
>> notmuch mailing list
>> notmuch@notmuchmail.org
>> http://notmuchmail.org/mailman/listinfo/notmuch