v3 of pre-gmime-3.0 cleanup

This obsoletes [1]. Since that series, a bit of the actual port to gmime-3.0 has crept in, in order to be able to mark tests as broken with respect a particular gmime version. The series in progress [2] has grown to 21 patches, and this is roughly the first half. Currently there about 7 failing

[Patch v3 09/11] test/multipart: reorganize creation of multipart message

We want to have the bodies of the multipart available in a file on their own for planned modifications to tests. --- test/T190-multipart.sh | 40 +--- 1 file changed, 25 insertions(+), 15 deletions(-) diff --git a/test/T190-multipart.sh

[Patch v3 02/11] cli/reply: direct all output for text format to gmime stream

Interleaving printfs with writes to the gmime stream worked when the gmime stream was backed by the FILE *stdout, but that is no longer the case. Create one stream and pass it into the two functions where needed, as well well as replacing printfs with g_mime_stream_printf. --- notmuch-client.h |

[Patch v3 03/11] cli/show: use single stream for printf / gmime object output

This is again motivated by the need to transition away from GMimeStreamFile for output to stdout. format_part_mbox is left alone for now, as this cannot be mixed in with output using gmime object output. --- notmuch-client.h | 1 + notmuch-show.c | 70

[Patch v3 01/11] util: convenience function to create gmime stream for stdout

It turns out that our use of GMimeStreamPipe has only succeeded because gmime has been ignoring some seek failures; this will no longer be the case in gmime 3.0, so we use a GMimeStreamPipe, which does not assume seekability, wrapped in a buffering stream. --- util/Makefile.local | 2 +-

[Patch v3 08/11] test: define GMime version dependant breakage

We have some tests where the gmime 3 behaviour seems like a bug fix, others where it's less clear, so we allow both possibilities. --- configure| 5 + test/test-lib.sh | 16 2 files changed, 21 insertions(+) diff --git a/configure b/configure index

[Patch v3 11/11] test: mark test as broken in gmime 3.0

Currently I'm not sure what the intent of this test is, so it's not clear if the new answer is better or worse than the old one. --- test/T310-emacs.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/test/T310-emacs.sh b/test/T310-emacs.sh index ef599849..2ef43925 100755 ---

[Patch v3 10/11] test: mark inclusion of headers as broken in gmime-2.x

The output of headers here reflects an underlying bug / quirk of gmime-2.6. --- test/T190-multipart.sh | 28 1 file changed, 4 insertions(+), 24 deletions(-) diff --git a/test/T190-multipart.sh b/test/T190-multipart.sh index 91a632c6..94bb0570 100755 ---

[Patch v3 05/11] cli/reply: fix two memory leaks, document a third

internet_address_list_to_string returns an allocated string, which needs to be freed with g_free. g_free can handle a NULL argument, so we follow the usage elsewhere of calling it unconditionally. The third leak we leave as it would require restructuring of add_recipients_from_message, and is

[Patch v3 06/11] test/thread-naming: remove excess escaping from sender address.

This is another case where the behaviour of gmime-2.6 and gmime-3.0 seems to differ. It may be that we prefer the more lax parsing of the previous version, but that should be tested seperately. --- test/T200-thread-naming.sh | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff

[Patch v3 07/11] configure: add optional support for gmime-3.0

This is only the changes to make configure work; it won't compile with gmime-3.0 yet. --- configure | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/configure b/configure index ed7c2280..91aeba51 100755 --- a/configure +++ b/configure @@ -484,8 +484,13 @@ fi