lib/n_d_index_file: re-use thread-id of existing message

This prevents the message document getting multiple thread-id terms
when there are multiple files with the same message-id.

This change shifts some thread ids, requiring adjustments to other tests.

(cherry picked from commit 3f4de98e7c8c70f9a86a4f899147126f79907bd9)

3 files changed, 31 insertions, 17 deletions

diff --git a/lib/add-message.cc b/lib/add-message.cc
index 485debad..b47aa501 100644
--- a/lib/add-message.cc
+++ b/lib/add-message.cc
@@ -407,14 +407,17 @@ static notmuch_status_t
 _notmuch_database_link_message (notmuch_database_t *notmuch,
 				notmuch_message_t *message,
 				notmuch_message_file_t *message_file,
-				bool is_ghost)
+				bool is_ghost,
+				bool is_new)
 {
     void *local = talloc_new (NULL);
     notmuch_status_t status;
     const char *thread_id = NULL;
 
     /* Check if the message already had a thread ID */
-    if (notmuch->features & NOTMUCH_FEATURE_GHOSTS) {
+    if (! is_new) {
+	thread_id = notmuch_message_get_thread_id (message);
+    } else if (notmuch->features & NOTMUCH_FEATURE_GHOSTS) {
 	if (is_ghost)
 	    thread_id = notmuch_message_get_thread_id (message);
     } else {
@@ -539,7 +542,7 @@ notmuch_database_index_file (notmuch_database_t *notmuch,
 	}
 
 	ret = _notmuch_database_link_message (notmuch, message,
-					      message_file, is_ghost);
+					      message_file, is_ghost, is_new);
 	if (ret)
 	    goto DONE;
 
diff --git a/test/T357-index-decryption.sh b/test/T357-index-decryption.sh
index 1ed5f28c..1951ef78 100755
--- a/test/T357-index-decryption.sh
+++ b/test/T357-index-decryption.sh
@@ -112,12 +112,10 @@ test_expect_equal \
     "$expected"
 
 # try inserting it with decryption, should appear as a single copy
-# (note: i think thread id skips 4 because of duplicate message-id
-# insertion, above)
 test_begin_subtest "message cleartext is present with insert --decrypt=true"
 notmuch insert --folder=sent --decrypt=true <<<"$contents"
-output=$(notmuch search wumpus)
-expected='thread:0000000000000005   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 002 (encrypted inbox unread)'
+output=$(notmuch search wumpus | notmuch_search_sanitize)
+expected='thread:XXX   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 002 (encrypted inbox unread)'
 test_expect_equal \
     "$output" \
     "$expected"
@@ -127,9 +125,9 @@ test_expect_equal \
 test_begin_subtest 'tagging all messages'
 test_expect_success 'notmuch tag +blarney "encrypted message"'
 test_begin_subtest "verify that tags have not changed"
-output=$(notmuch search tag:blarney)
-expected='thread:0000000000000001   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 001 (blarney encrypted inbox)
-thread:0000000000000005   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 002 (blarney encrypted inbox unread)'
+output=$(notmuch search tag:blarney | notmuch_search_sanitize)
+expected='thread:XXX   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 001 (blarney encrypted inbox)
+thread:XXX   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 002 (blarney encrypted inbox unread)'
 test_expect_equal \
     "$output" \
     "$expected"
@@ -138,14 +136,14 @@ test_expect_equal \
 test_begin_subtest 'reindex old messages'
 test_expect_success 'notmuch reindex --decrypt=true tag:encrypted and not property:index.decryption=success'
 test_begin_subtest "reindexed encrypted message, including cleartext"
-output=$(notmuch search wumpus)
+output=$(notmuch search wumpus | notmuch_search_sanitize)
 test_expect_equal \
     "$output" \
     "$expected"
 
 # and the same search, but by property ($expected is untouched):
 test_begin_subtest "emacs search by property for both messages"
-output=$(notmuch search property:index.decryption=success)
+output=$(notmuch search property:index.decryption=success | notmuch_search_sanitize)
 test_expect_equal \
     "$output" \
     "$expected"
@@ -154,7 +152,7 @@ test_expect_equal \
 test_begin_subtest 'reindex in auto mode'
 test_expect_success 'notmuch reindex tag:encrypted and property:index.decryption=success'
 test_begin_subtest "reindexed encrypted messages, should not have changed"
-output=$(notmuch search wumpus)
+output=$(notmuch search wumpus | notmuch_search_sanitize)
 test_expect_equal \
     "$output" \
     "$expected"
@@ -188,9 +186,9 @@ test_expect_equal \
 
 # ensure that the tags remain even when we are dropping the cleartext.
 test_begin_subtest "verify that tags remain without cleartext"
-output=$(notmuch search tag:blarney)
-expected='thread:0000000000000001   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 001 (blarney encrypted inbox)
-thread:0000000000000005   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 002 (blarney encrypted inbox unread)'
+output=$(notmuch search tag:blarney | notmuch_search_sanitize)
+expected='thread:XXX   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 001 (blarney encrypted inbox)
+thread:XXX   2000-01-01 [1/1] Notmuch Test Suite; test encrypted message for cleartext index 002 (blarney encrypted inbox unread)'
 test_expect_equal \
     "$output" \
     "$expected"
@@ -199,7 +197,7 @@ test_begin_subtest "index cleartext without keeping session keys"
 test_expect_success "notmuch reindex --decrypt=nostash tag:blarney"
 
 test_begin_subtest "Ensure that the indexed terms are present"
-output=$(notmuch search wumpus)
+output=$(notmuch search wumpus | notmuch_search_sanitize)
 test_expect_equal \
     "$output" \
     "$expected"
diff --git a/test/T670-duplicate-mid.sh b/test/T670-duplicate-mid.sh
index 4e5672ab..7ae60595 100755
--- a/test/T670-duplicate-mid.sh
+++ b/test/T670-duplicate-mid.sh
@@ -6,6 +6,19 @@ add_message '[id]="duplicate"' '[subject]="message 1" [filename]=copy1'
 add_message '[id]="duplicate"' '[subject]="message 2" [filename]=copy2'
 
 add_message '[id]="duplicate"' '[subject]="message 0" [filename]=copy0'
+
+test_begin_subtest 'at most 1 thread-id per xapian document'
+db=${MAIL_DIR}/.notmuch/xapian
+for doc in $(xapian-delve -1 -t '' "$db" | grep '^[1-9]'); do
+    xapian-delve -1 -r "$doc" "$db" | grep -c '^G'
+done > OUTPUT.raw
+sort -u < OUTPUT.raw > OUTPUT
+cat <<EOF > EXPECTED
+0
+1
+EOF
+test_expect_equal_file EXPECTED OUTPUT
+
 test_begin_subtest 'search: first indexed subject preserved'
 cat <<EOF > EXPECTED
 thread:XXX   2001-01-05 [1/1(3)] Notmuch Test Suite; message 1 (inbox unread)