On Fri, Nov 30 2012, david@tethera.net wrote: > From: Jani Nikula <jani@nikula.org> > > Basic test of functionality, along with all combinations of options. > > Modified extensively by David Bremner <david@tethera.net> > > The choice of @ as a tag is intended to be non-alphanumeric, but still > not too much trouble in the shell and in the old sup dump/restore format. > --- > > Mark: good catch. > > I decided to save restore the tags rather than have > multiple input and output files here. > > test/tagging | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 50 insertions(+) > > diff --git a/test/tagging b/test/tagging > index 980ff92..75552e8 100755 > --- a/test/tagging > +++ b/test/tagging > @@ -46,6 +46,56 @@ test_expect_equal "$output" "\ > thread:XXX 2001-01-05 [1/1] Notmuch Test Suite; One (:\" inbox tag1 unread) > thread:XXX 2001-01-05 [1/1] Notmuch Test Suite; Two (inbox tag1 tag4 unread)" > > +test_begin_subtest "--batch" > +notmuch tag --batch <<EOF > +# %20 is a space in tag > +-:"%20 -tag1 +tag5 +tag6 -- One > ++tag1 -tag1 -tag4 +tag4 -- Two > +-tag6 One > ++tag5 Two > +EOF > +output=$(notmuch search \* | notmuch_search_sanitize) > +test_expect_equal "$output" "\ > +thread:XXX 2001-01-05 [1/1] Notmuch Test Suite; One (inbox tag5 unread) > +thread:XXX 2001-01-05 [1/1] Notmuch Test Suite; Two (inbox tag4 tag5 unread)" > + > +cat > batch.in <<EOF > +# %20 is a space in tag > ++%40 -tag5 +tag6 -- One > ++tag1 -tag1 -tag4 +tag4 -- Two > +-tag5 +tag6 Two > +EOF > + > +cat > batch.expected <<EOF > +thread:XXX 2001-01-05 [1/1] Notmuch Test Suite; One (@ inbox tag6 unread) > +thread:XXX 2001-01-05 [1/1] Notmuch Test Suite; Two (inbox tag4 tag6 unread) > +EOF > + > +test_begin_subtest "--input" > +notmuch dump > backup.tags.$test_count > +notmuch tag --input=batch.in > +notmuch search \* | notmuch_search_sanitize > OUTPUT.$test_count > +notmuch restore < backup.tags.$test_count > +test_expect_equal_file batch.expected OUTPUT.$test_count 2 things for all of these test_expect_equal_file ...:s Generally, argument order is OUTPUT EXPECTED (the diff is harder to grasp -- but changing the diff order in test_expect_equal_file is pretty easy... (& I have some patches in preparation for that...) In case of failure. test_expect_equal_file does if diff -q "$file1" "$file2" >/dev/null ; then test_ok_ "$test_subtest_name" else testname=$this_test.$test_count cp "$file1" "$testname.$basename1" cp "$file2" "$testname.$basename2" i.e. the $test_count is already in the files -- in these cases that is duplicated... Tomi > +test_begin_subtest "--batch --input" > +notmuch dump > backup.tags.$test_count > +notmuch tag --batch --input=batch.in > +notmuch search \* | notmuch_search_sanitize > OUTPUT.$test_count > +notmuch restore < backup.tags.$test_count > +test_expect_equal_file batch.expected OUTPUT.$test_count > + > +test_begin_subtest "--batch, blank lines and comments" > +notmuch dump | sort > EXPECTED.$test_count > +notmuch tag --batch <<EOF > +# this line is a comment; the next has only white space > + > + > +# the previous line is empty > +EOF > +notmuch dump | sort > OUTPUT.$test_count > +test_expect_equal_file EXPECTED.$test_count OUTPUT.$test_count > + > test_expect_code 1 "Empty tag names" 'notmuch tag + One' > > test_expect_code 1 "Tag name beginning with -" 'notmuch tag +- One' > -- > 1.7.10.4