lib/parse-sexp: expand queries

The code here is just gluing together _notmuch_query_expand with the
existing sexp parser infrastructure.
This commit is contained in:
David Bremner 2021-08-24 08:17:33 -07:00
parent 4083fd8bec
commit afe85e6578
3 changed files with 118 additions and 10 deletions

View file

@ -144,6 +144,11 @@ MODIFIERS
*Modifiers* refer to any prefixes (first elements of compound queries) *Modifiers* refer to any prefixes (first elements of compound queries)
that are neither operators nor fields. that are neither operators nor fields.
``(matching`` |q1| |q2| ... |qn| ``)`` ``(of`` |q1| |q2| ... |qn| ``)``
Match all messages have the same values of the current field as
those matching all of |q1| ... |qn|. Supported in most term [#not-path]_ or
phrase fields. Most commonly used in the ``thread`` field.
``(regex`` *atom* ``)`` ``(rx`` *atom* ``)`` ``(regex`` *atom* ``)`` ``(rx`` *atom* ``)``
Interpret *atom* as a POSIX.2 regular expression (see Interpret *atom* as a POSIX.2 regular expression (see
:manpage:`regex(7)`). This applies in term fields and a subset [#not-phrase]_ of :manpage:`regex(7)`). This applies in term fields and a subset [#not-phrase]_ of
@ -176,6 +181,9 @@ EXAMPLES
Match the *phrase* "quick" followed by "fox" in phrase fields (or Match the *phrase* "quick" followed by "fox" in phrase fields (or
outside a field). Match the literal string in a term field. outside a field). Match the literal string in a term field.
``(folder (of (id 1234@invalid)))``
Match any message in the same folder as the one with Message-Id "1234@invalid"
``(id 1234@invalid blah@test)`` ``(id 1234@invalid blah@test)``
Matches Message-Id "1234@invalid" *or* Message-Id "blah@test" Matches Message-Id "1234@invalid" *or* Message-Id "blah@test"
@ -193,6 +201,14 @@ EXAMPLES
Match messages whose subject contains "quick brown fox", but also Match messages whose subject contains "quick brown fox", but also
"brown fox quicksand". "brown fox quicksand".
``(thread (of (id 1234@invalid)))``
Match any message in the same thread as the one with Message-Id "1234@invalid"
``(thread (matching (from bob@example.com) (to bob@example.com)))``
Match any (messages in) a thread containing a message from
"bob@example.com" and a (possibly distinct) message to "bob at
example.com")
``(to (or bob@example.com mallory@example.org))`` ``(or (to bob@example.com) (to mallory@example.org))`` ``(to (or bob@example.com mallory@example.org))`` ``(or (to bob@example.com) (to mallory@example.org))``
Match in the "To" or "Cc" headers, "bob@example.com", Match in the "To" or "Cc" headers, "bob@example.com",
"mallory@example.org", and also "bob@example.com.au" since it "mallory@example.org", and also "bob@example.com.au" since it
@ -216,6 +232,10 @@ NOTES
.. [#not-body] Due the the way ``body`` is implemented in notmuch, .. [#not-body] Due the the way ``body`` is implemented in notmuch,
this modifier is not supported in the ``body`` field. this modifier is not supported in the ``body`` field.
.. [#not-path] Due to the way recursive ``path`` queries are implemented
in notmuch, this modifier is not supported in the
``path`` field.
.. |q1| replace:: :math:`q_1` .. |q1| replace:: :math:`q_1`
.. |q2| replace:: :math:`q_2` .. |q2| replace:: :math:`q_2`
.. |qn| replace:: :math:`q_n` .. |qn| replace:: :math:`q_n`

View file

@ -15,6 +15,8 @@ typedef enum {
SEXP_FLAG_WILDCARD = 1 << 3, SEXP_FLAG_WILDCARD = 1 << 3,
SEXP_FLAG_REGEX = 1 << 4, SEXP_FLAG_REGEX = 1 << 4,
SEXP_FLAG_DO_REGEX = 1 << 5, SEXP_FLAG_DO_REGEX = 1 << 5,
SEXP_FLAG_EXPAND = 1 << 6,
SEXP_FLAG_DO_EXPAND = 1 << 7,
} _sexp_flag_t; } _sexp_flag_t;
/* /*
@ -46,29 +48,33 @@ static _sexp_prefix_t prefixes[] =
{ "and", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "and", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_NONE }, SEXP_FLAG_NONE },
{ "attachment", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "attachment", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD }, SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD | SEXP_FLAG_EXPAND },
{ "body", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "body", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_FIELD }, SEXP_FLAG_FIELD },
{ "from", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "from", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX | SEXP_FLAG_EXPAND },
{ "folder", Xapian::Query::OP_OR, Xapian::Query::MatchNothing, { "folder", Xapian::Query::OP_OR, Xapian::Query::MatchNothing,
SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX | SEXP_FLAG_EXPAND },
{ "id", Xapian::Query::OP_OR, Xapian::Query::MatchNothing, { "id", Xapian::Query::OP_OR, Xapian::Query::MatchNothing,
SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX },
{ "is", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "is", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX | SEXP_FLAG_EXPAND },
{ "matching", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_DO_EXPAND },
{ "mid", Xapian::Query::OP_OR, Xapian::Query::MatchNothing, { "mid", Xapian::Query::OP_OR, Xapian::Query::MatchNothing,
SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX },
{ "mimetype", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "mimetype", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD }, SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD | SEXP_FLAG_EXPAND },
{ "not", Xapian::Query::OP_AND_NOT, Xapian::Query::MatchAll, { "not", Xapian::Query::OP_AND_NOT, Xapian::Query::MatchAll,
SEXP_FLAG_NONE }, SEXP_FLAG_NONE },
{ "of", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_DO_EXPAND },
{ "or", Xapian::Query::OP_OR, Xapian::Query::MatchNothing, { "or", Xapian::Query::OP_OR, Xapian::Query::MatchNothing,
SEXP_FLAG_NONE }, SEXP_FLAG_NONE },
{ "path", Xapian::Query::OP_OR, Xapian::Query::MatchNothing, { "path", Xapian::Query::OP_OR, Xapian::Query::MatchNothing,
SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX },
{ "property", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "property", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX | SEXP_FLAG_EXPAND },
{ "regex", Xapian::Query::OP_INVALID, Xapian::Query::MatchAll, { "regex", Xapian::Query::OP_INVALID, Xapian::Query::MatchAll,
SEXP_FLAG_SINGLE | SEXP_FLAG_DO_REGEX }, SEXP_FLAG_SINGLE | SEXP_FLAG_DO_REGEX },
{ "rx", Xapian::Query::OP_INVALID, Xapian::Query::MatchAll, { "rx", Xapian::Query::OP_INVALID, Xapian::Query::MatchAll,
@ -76,13 +82,13 @@ static _sexp_prefix_t prefixes[] =
{ "starts-with", Xapian::Query::OP_WILDCARD, Xapian::Query::MatchAll, { "starts-with", Xapian::Query::OP_WILDCARD, Xapian::Query::MatchAll,
SEXP_FLAG_SINGLE }, SEXP_FLAG_SINGLE },
{ "subject", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "subject", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX | SEXP_FLAG_EXPAND },
{ "tag", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "tag", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX | SEXP_FLAG_EXPAND },
{ "thread", Xapian::Query::OP_OR, Xapian::Query::MatchNothing, { "thread", Xapian::Query::OP_OR, Xapian::Query::MatchNothing,
SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX }, SEXP_FLAG_FIELD | SEXP_FLAG_BOOLEAN | SEXP_FLAG_WILDCARD | SEXP_FLAG_REGEX | SEXP_FLAG_EXPAND },
{ "to", Xapian::Query::OP_AND, Xapian::Query::MatchAll, { "to", Xapian::Query::OP_AND, Xapian::Query::MatchAll,
SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD }, SEXP_FLAG_FIELD | SEXP_FLAG_WILDCARD | SEXP_FLAG_EXPAND },
{ } { }
}; };
@ -210,6 +216,32 @@ _sexp_parse_regex (notmuch_database_t *notmuch,
val, output, msg); val, output, msg);
} }
static notmuch_status_t
_sexp_expand_query (notmuch_database_t *notmuch,
const _sexp_prefix_t *prefix, const _sexp_prefix_t *parent,
const sexp_t *sx, Xapian::Query &output)
{
Xapian::Query subquery;
notmuch_status_t status;
std::string msg;
if (! (parent->flags & SEXP_FLAG_EXPAND)) {
_notmuch_database_log (notmuch, "'%s' unsupported inside '%s'\n", prefix->name, parent->name);
return NOTMUCH_STATUS_BAD_QUERY_SYNTAX;
}
status = _sexp_combine_query (notmuch, NULL, prefix->xapian_op, prefix->initial, sx, subquery);
if (status)
return status;
status = _notmuch_query_expand (notmuch, parent->name, subquery, output, msg);
if (status) {
_notmuch_database_log (notmuch, "error expanding query %s\n", msg.c_str ());
}
return status;
}
/* Here we expect the s-expression to be a proper list, with first /* Here we expect the s-expression to be a proper list, with first
* element defining and operation, or as a special case the empty * element defining and operation, or as a special case the empty
* list */ * list */
@ -286,6 +318,10 @@ _sexp_to_xapian_query (notmuch_database_t *notmuch, const _sexp_prefix_t *parent
return _sexp_parse_regex (notmuch, prefix, parent, sx->list->next->val, output); return _sexp_parse_regex (notmuch, prefix, parent, sx->list->next->val, output);
} }
if (prefix->flags & SEXP_FLAG_DO_EXPAND) {
return _sexp_expand_query (notmuch, prefix, parent, sx->list->next, output);
}
return _sexp_combine_query (notmuch, parent, prefix->xapian_op, prefix->initial, return _sexp_combine_query (notmuch, parent, prefix->xapian_op, prefix->initial,
sx->list->next, output); sx->list->next, output);
} }

View file

@ -637,4 +637,56 @@ notmuch search --output=threads '*' | grep '7$' > EXPECTED
notmuch search --output=threads --query=sexp '(thread (rx 7$))' > OUTPUT notmuch search --output=threads --query=sexp '(thread (rx 7$))' > OUTPUT
test_expect_equal_file EXPECTED OUTPUT test_expect_equal_file EXPECTED OUTPUT
test_begin_subtest "Basic query that matches no messages"
count=$(notmuch count from:keithp and to:keithp)
test_expect_equal 0 "$count"
test_begin_subtest "Same query against threads"
notmuch search --query=sexp '(and (thread (of (from keithp))) (thread (matching (to keithp))))' \
| notmuch_search_sanitize > OUTPUT
cat<<EOF > EXPECTED
thread:XXX 2009-11-18 [7/7] Lars Kellogg-Stedman, Mikhail Gusarov, Keith Packard, Carl Worth; [notmuch] Working with Maildir storage? (inbox signed unread)
EOF
test_expect_equal_file EXPECTED OUTPUT
test_begin_subtest "Mix thread and non-threads query"
notmuch search --query=sexp '(and (thread (matching keithp)) (to keithp))' | notmuch_search_sanitize > OUTPUT
cat<<EOF > EXPECTED
thread:XXX 2009-11-18 [1/7] Lars Kellogg-Stedman| Mikhail Gusarov, Keith Packard, Carl Worth; [notmuch] Working with Maildir storage? (inbox signed unread)
EOF
test_expect_equal_file EXPECTED OUTPUT
test_begin_subtest "Compound subquery"
notmuch search --query=sexp '(thread (of (from keithp) (subject Maildir)))' | notmuch_search_sanitize > OUTPUT
cat<<EOF > EXPECTED
thread:XXX 2009-11-18 [7/7] Lars Kellogg-Stedman, Mikhail Gusarov, Keith Packard, Carl Worth; [notmuch] Working with Maildir storage? (inbox signed unread)
EOF
test_expect_equal_file EXPECTED OUTPUT
test_begin_subtest "empty subquery"
notmuch search --query=sexp '(thread (of))' 1>OUTPUT 2>&1
notmuch search '*' > EXPECTED
test_expect_equal_file EXPECTED OUTPUT
test_begin_subtest "illegal expansion"
notmuch search --query=sexp '(id (of ego))' 1>OUTPUT 2>&1
cat<<EOF > EXPECTED
notmuch search: Syntax error in query
'of' unsupported inside 'id'
EOF
test_expect_equal_file EXPECTED OUTPUT
test_begin_subtest "(folder (of subquery))"
notmuch search --query=sexp --output=messages '(folder (of (id yun3a4cegoa.fsf@aiko.keithp.com)))' > OUTPUT
cat <<EOF > EXPECTED
id:yun1vjwegii.fsf@aiko.keithp.com
id:yun3a4cegoa.fsf@aiko.keithp.com
id:1258509400-32511-1-git-send-email-stewart@flamingspork.com
id:1258506353-20352-1-git-send-email-stewart@flamingspork.com
id:20091118010116.GC25380@dottiness.seas.harvard.edu
id:20091118005829.GB25380@dottiness.seas.harvard.edu
id:cf0c4d610911171136h1713aa59w9cf9aa31f052ad0a@mail.gmail.com
EOF
test_expect_equal_file EXPECTED OUTPUT
test_done test_done