pidgin/pidgin

Fix several leaks in tests

14 months ago, Elliott Sales de Andrade
4e1bf25f5575
Fix several leaks in tests

All of these are specific to tests, not the library code.

For the moment, `protocol_xfer` still leaks connections (and anything they hold on to) because it is very difficult to disentangle them from the connection manager in the partially implemented state they are in.

This fixes leaks of options in the account option test (these two leaks occur for every test since they all leak the option):
```
61 (48 direct, 13 indirect) bytes in 1 blocks are definitely lost in loss record 133 of 276
at 0x4848464: calloc (vg_replace_malloc.c:1340)
by 0x49F75F0: g_malloc0 (gmem.c:163)
by 0x48C3B2E: purple_account_option_new (purpleaccountoption.c:78)
by 0x4014AF: test_purple_account_option_copy_int (test_account_option.c:67)
by 0x4A1CC7D: UnknownInlinedFun (gtestutils.c:2933)
by 0x4A1CC7D: g_test_run_suite_internal (gtestutils.c:3021)
by 0x4A1C9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A1C9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A1D181: g_test_run_suite (gtestutils.c:3115)
by 0x4A156EC: UnknownInlinedFun (gtestutils.c:2234)
by 0x4A156EC: g_test_run (gtestutils.c:2221)
by 0x401721: main (test_account_option.c:157)

61 (48 direct, 13 indirect) bytes in 1 blocks are definitely lost in loss record 134 of 276
at 0x4848464: calloc (vg_replace_malloc.c:1340)
by 0x49F75F0: g_malloc0 (gmem.c:163)
by 0x48C3BC7: purple_account_option_copy (purpleaccountoption.c:93)
by 0x4014BF: test_purple_account_option_copy_int (test_account_option.c:68)
by 0x4A1CC7D: UnknownInlinedFun (gtestutils.c:2933)
by 0x4A1CC7D: g_test_run_suite_internal (gtestutils.c:3021)
by 0x4A1C9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A1C9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A1D181: g_test_run_suite (gtestutils.c:3115)
by 0x4A156EC: UnknownInlinedFun (gtestutils.c:2234)
by 0x4A156EC: g_test_run (gtestutils.c:2221)
by 0x401721: main (test_account_option.c:157)
```
leaks in the credential manager test (times 3 for read/write/cancel tests):
```
69 (16 direct, 53 indirect) bytes in 1 blocks are definitely lost in loss record 2,427 of 3,503
at 0x484386F: malloc (vg_replace_malloc.c:393)
by 0x4A58168: g_malloc (gmem.c:130)
by 0x4A6FAB5: g_slice_alloc (gslice.c:1074)
by 0x4A700EC: g_slice_alloc0 (gslice.c:1100)
by 0x4A3BECB: g_error_allocate (gerror.c:710)
by 0x4A3C93F: UnknownInlinedFun (gerror.c:724)
by 0x4A3C93F: g_error_new_valist (gerror.c:766)
by 0x4BEE558: g_task_return_new_error (gtask.c:1941)
by 0x48D82C0: purple_credential_manager_read_password_async (purplecredentialmanager.c:492)
by 0x403634: test_purple_credential_manager_no_provider_read_password_idle (test_credential_manager.c:329)
by 0x4A4ECB1: g_idle_dispatch (gmain.c:6124)
by 0x4A4FCBE: UnknownInlinedFun (gmain.c:3444)
by 0x4A4FCBE: g_main_context_dispatch (gmain.c:4162)
by 0x4AA5597: g_main_context_iterate.constprop.0 (gmain.c:4238)
by 0x4A4F28E: g_main_loop_run (gmain.c:4438)
by 0x40369F: test_purple_credential_manager_no_provider_read_password_async (test_credential_manager.c:345)
by 0x4A7DC7D: UnknownInlinedFun (gtestutils.c:2933)
by 0x4A7DC7D: g_test_run_suite_internal (gtestutils.c:3021)
by 0x4A7D9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A7D9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A7E181: g_test_run_suite (gtestutils.c:3115)
by 0x4A766EC: UnknownInlinedFun (gtestutils.c:2234)
by 0x4A766EC: g_test_run (gtestutils.c:2221)
by 0x4048F6: main (test_credential_manager.c:695)
```
a leak in the image test:
```
161 bytes in 1 blocks are definitely lost in loss record 260 of 274
at 0x484386F: malloc (vg_replace_malloc.c:393)
by 0x4A55363: g_try_malloc (gmem.c:286)
by 0x4A3D630: UnknownInlinedFun (gfileutils.c:819)
by 0x4A3D630: UnknownInlinedFun (gfileutils.c:924)
by 0x4A3D630: g_file_get_contents (gfileutils.c:1027)
by 0x401890: test_image_new_from_file (test_image.c:144)
by 0x4A7DC7D: UnknownInlinedFun (gtestutils.c:2933)
by 0x4A7DC7D: g_test_run_suite_internal (gtestutils.c:3021)
by 0x4A7D9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A7E181: g_test_run_suite (gtestutils.c:3115)
by 0x4A766EC: UnknownInlinedFun (gtestutils.c:2234)
by 0x4A766EC: g_test_run (gtestutils.c:2221)
by 0x40195D: main (test_image.c:172)
```
a leak in queued output stream test:
```
72 (40 direct, 32 indirect) bytes in 1 blocks are definitely lost in loss record 219 of 396
at 0x49D51EF: g_type_create_instance (gtype.c:1909)
by 0x49BAC1F: g_object_new_internal (gobject.c:2228)
by 0x49BC247: g_object_new_with_properties (gobject.c:2391)
by 0x49BCFF0: g_object_new (gobject.c:2037)
by 0x402003: test_queued_output_stream_push_bytes_async_error (test_queued_output_stream.c:219)
by 0x4A7DC7D: UnknownInlinedFun (gtestutils.c:2933)
by 0x4A7DC7D: g_test_run_suite_internal (gtestutils.c:3021)
by 0x4A7D9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A7E181: g_test_run_suite (gtestutils.c:3115)
by 0x4A766EC: UnknownInlinedFun (gtestutils.c:2234)
by 0x4A766EC: g_test_run (gtestutils.c:2221)
by 0x402429: main (test_queued_output_stream.c:280)
```
and protocol xfer tests (times 3 for each test that creates a test protocol object):
```
112 (48 direct, 64 indirect) bytes in 1 blocks are definitely lost in loss record 3,430 of 3,698
at 0x49D51EF: g_type_create_instance (gtype.c:1909)
by 0x49BAC1F: g_object_new_internal (gobject.c:2228)
by 0x49BC247: g_object_new_with_properties (gobject.c:2391)
by 0x49BCFF0: g_object_new (gobject.c:2037)
by 0x40291C: test_purple_protocol_xfer_send_file_func (test_protocol_xfer.c:146)
by 0x4A7DC7D: UnknownInlinedFun (gtestutils.c:2933)
by 0x4A7DC7D: g_test_run_suite_internal (gtestutils.c:3021)
by 0x4A7D9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A7E181: g_test_run_suite (gtestutils.c:3115)
by 0x4A766EC: UnknownInlinedFun (gtestutils.c:2234)
by 0x4A766EC: g_test_run (gtestutils.c:2221)
by 0x402B64: main (test_protocol_xfer.c:195)
```
and util tests (times 3 for each call to `purple_text_strip_mnemonic` in the test):
```
5 bytes in 1 blocks are definitely lost in loss record 5 of 247
at 0x484386F: malloc (vg_replace_malloc.c:393)
by 0x49F7168: g_malloc (gmem.c:130)
by 0x491975B: purple_text_strip_mnemonic (util.c:895)
by 0x4015B0: test_util_text_strip_mnemonic (test_util.c:49)
by 0x4A1CC7D: UnknownInlinedFun (gtestutils.c:2933)
by 0x4A1CC7D: g_test_run_suite_internal (gtestutils.c:3021)
by 0x4A1C9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A1C9E4: g_test_run_suite_internal (gtestutils.c:3038)
by 0x4A1D181: g_test_run_suite (gtestutils.c:3115)
by 0x4A156EC: UnknownInlinedFun (gtestutils.c:2234)
by 0x4A156EC: g_test_run (gtestutils.c:2221)
by 0x401901: main (test_util.c:224)
```
and these leaks in any test that initializes the test UI:
```
4,104 bytes in 1 blocks are possibly lost in loss record 3,451 of 3,457
at 0x484386F: malloc (vg_replace_malloc.c:393)
by 0x5235B67: sqlite3MemMalloc.lto_priv.0 (sqlite3.c:25493)
by 0x5232797: UnknownInlinedFun (sqlite3.c:29181)
by 0x5232797: UnknownInlinedFun (sqlite3.c:29227)
by 0x5232797: sqlite3Malloc.lto_priv.0 (sqlite3.c:29221)
by 0x523BD8B: pcache1Alloc.lto_priv.0 (sqlite3.c:53546)
by 0x5249A8B: UnknownInlinedFun (sqlite3.c:53675)
by 0x5249A8B: allocateTempSpace (sqlite3.c:70848)
by 0x52625A6: sqlite3VdbeExec.lto_priv.0 (sqlite3.c:93857)
by 0x525CBEE: UnknownInlinedFun (sqlite3.c:87995)
by 0x525CBEE: UnknownInlinedFun (sqlite3.c:88056)
by 0x525CBEE: sqlite3_step (sqlite3.c:88045)
by 0x529B324: sqlite3_exec (sqlite3.c:131002)
by 0x48FD558: purple_sqlite3_run_migration (purplesqlite3.c:37)
by 0x48FDBB4: purple_sqlite3_run_migrations_from_resources (purplesqlite3.c:195)
by 0x48FDED9: purple_sqlite_history_adapter_run_migrations (purplesqlitehistoryadapter.c:69)
by 0x48FE7F0: purple_sqlite_history_adapter_activate (purplesqlitehistoryadapter.c:287)
by 0x48DB656: purple_history_adapter_activate (purplehistoryadapter.c:181)
by 0x48DC9BC: purple_history_manager_set_active (purplehistorymanager.c:308)
by 0x402BA8: test_ui_init_history (test_ui.c:132)
by 0x402C80: test_ui_purple_init (test_ui.c:167)
by 0x4027BB: main (test_contact.c:88)

4,368 bytes in 1 blocks are possibly lost in loss record 3,453 of 3,457
at 0x484386F: malloc (vg_replace_malloc.c:393)
by 0x5235B67: sqlite3MemMalloc.lto_priv.0 (sqlite3.c:25493)
by 0x5232797: UnknownInlinedFun (sqlite3.c:29181)
by 0x5232797: UnknownInlinedFun (sqlite3.c:29227)
by 0x5232797: sqlite3Malloc.lto_priv.0 (sqlite3.c:29221)
by 0x523BD8B: pcache1Alloc.lto_priv.0 (sqlite3.c:53546)
by 0x5240077: UnknownInlinedFun (sqlite3.c:53634)
by 0x5240077: pcache1FetchStage2 (sqlite3.c:54104)
by 0x5243E9C: UnknownInlinedFun (sqlite3.c:52671)
by 0x5243E9C: getPageNormal.lto_priv.0 (sqlite3.c:60628)
by 0x524A510: UnknownInlinedFun (sqlite3.c:60805)
by 0x524A510: btreeGetPage.lto_priv.0 (sqlite3.c:70289)
by 0x524C2F6: UnknownInlinedFun (sqlite3.c:71257)
by 0x524C2F6: sqlite3BtreeBeginTrans.lto_priv.0 (sqlite3.c:71647)
by 0x5266B3A: sqlite3VdbeExec.lto_priv.0 (sqlite3.c:93532)
by 0x525CBEE: UnknownInlinedFun (sqlite3.c:87995)
by 0x525CBEE: UnknownInlinedFun (sqlite3.c:88056)
by 0x525CBEE: sqlite3_step (sqlite3.c:88045)
by 0x48FD715: purple_sqlite3_get_schema_version (purplesqlite3.c:79)
by 0x48FD9DD: purple_sqlite3_run_migrations_from_resources (purplesqlite3.c:146)
by 0x48FDED9: purple_sqlite_history_adapter_run_migrations (purplesqlitehistoryadapter.c:69)
by 0x48FE7F0: purple_sqlite_history_adapter_activate (purplesqlitehistoryadapter.c:287)
by 0x48DB656: purple_history_adapter_activate (purplehistoryadapter.c:181)
by 0x48DC9BC: purple_history_manager_set_active (purplehistorymanager.c:308)
by 0x402BA8: test_ui_init_history (test_ui.c:132)
by 0x402C80: test_ui_purple_init (test_ui.c:167)
by 0x4027BB: main (test_contact.c:88)

4,368 bytes in 1 blocks are possibly lost in loss record 3,454 of 3,457
at 0x484386F: malloc (vg_replace_malloc.c:393)
by 0x5235B67: sqlite3MemMalloc.lto_priv.0 (sqlite3.c:25493)
by 0x5232797: UnknownInlinedFun (sqlite3.c:29181)
by 0x5232797: UnknownInlinedFun (sqlite3.c:29227)
by 0x5232797: sqlite3Malloc.lto_priv.0 (sqlite3.c:29221)
by 0x523BD8B: pcache1Alloc.lto_priv.0 (sqlite3.c:53546)
by 0x5240077: UnknownInlinedFun (sqlite3.c:53634)
by 0x5240077: pcache1FetchStage2 (sqlite3.c:54104)
by 0x5243E9C: UnknownInlinedFun (sqlite3.c:52671)
by 0x5243E9C: getPageNormal.lto_priv.0 (sqlite3.c:60628)
by 0x52499BC: UnknownInlinedFun (sqlite3.c:60805)
by 0x52499BC: UnknownInlinedFun (sqlite3.c:70289)
by 0x52499BC: btreeGetUnusedPage (sqlite3.c:70432)
by 0x524F504: allocateBtreePage.lto_priv.0 (sqlite3.c:74604)
by 0x5256209: btreeCreateTable.lto_priv.0 (sqlite3.c:77830)
by 0x5265EB2: UnknownInlinedFun (sqlite3.c:77849)
by 0x5265EB2: sqlite3VdbeExec.lto_priv.0 (sqlite3.c:96382)
by 0x525CBEE: UnknownInlinedFun (sqlite3.c:87995)
by 0x525CBEE: UnknownInlinedFun (sqlite3.c:88056)
by 0x525CBEE: sqlite3_step (sqlite3.c:88045)
by 0x529B324: sqlite3_exec (sqlite3.c:131002)
by 0x48FD558: purple_sqlite3_run_migration (purplesqlite3.c:37)
by 0x48FDBB4: purple_sqlite3_run_migrations_from_resources (purplesqlite3.c:195)
by 0x48FDED9: purple_sqlite_history_adapter_run_migrations (purplesqlitehistoryadapter.c:69)
by 0x48FE7F0: purple_sqlite_history_adapter_activate (purplesqlitehistoryadapter.c:287)
by 0x48DB656: purple_history_adapter_activate (purplehistoryadapter.c:181)
by 0x48DC9BC: purple_history_manager_set_active (purplehistorymanager.c:308)
by 0x402BA8: test_ui_init_history (test_ui.c:132)
by 0x402C80: test_ui_purple_init (test_ui.c:167)
by 0x4027BB: main (test_contact.c:88)
```

Testing Done:
Ran tests in valgrind, and all above leaks were gone except for noted `protocol_xfer` issues, and a bunch of leaks of the `PurpleBuddy`-`PurpleContact` compatibility bindings, which will go away in their entirety eventually.

Reviewed at https://reviews.imfreedom.org/r/2385/
/*
* Purple - Internet Messaging Library
* Copyright (C) Pidgin Developers <devel@pidgin.im>
*
* Purple is the legal property of its developers, whose names are too numerous
* to list here. Please refer to the COPYRIGHT file distributed with this
* source distribution.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, see <https://www.gnu.org/licenses/>.
*/
#if !defined(PURPLE_GLOBAL_HEADER_INSIDE) && !defined(PURPLE_COMPILATION)
# error "only <purple.h> may be included directly"
#endif
#ifndef PURPLE_PROTOCOL_IM_H
#define PURPLE_PROTOCOL_IM_H
#include <glib.h>
#include <glib-object.h>
#include <libpurple/connection.h>
#include <libpurple/purpleprotocol.h>
#define PURPLE_TYPE_PROTOCOL_IM (purple_protocol_im_get_type())
/**
* PurpleProtocolIM:
*
* #PurpleProtocolIM describes the API that protocols need to implement for
* handling one on one conversations.
*
* Since: 3.0.0
*/
G_DECLARE_INTERFACE(PurpleProtocolIM, purple_protocol_im, PURPLE, PROTOCOL_IM,
PurpleProtocol)
/**
* PurpleProtocolIMInterface:
* @send: This protocol function should return a positive value on
* success. If the message is too big to be sent, return
* <literal>-E2BIG</literal>. If the account is not connected,
* return <literal>-ENOTCONN</literal>. If the protocol is unable
* to send the message for another reason, return some other
* negative value. You can use one of the valid #errno values, or
* just big something. If the message should not be echoed to the
* conversation window, return 0.
* @send_typing: If this protocol requires the #PURPLE_IM_TYPING message to be
* sent repeatedly to signify that the user is still typing, then
* the protocol should return the number of seconds to wait before
* sending a subsequent notification. Otherwise the protocol
* should return 0.
*
* The protocol IM interface that needs to be implemented to send one to one
* messages.
*/
struct _PurpleProtocolIMInterface {
/*< private >*/
GTypeInterface parent;
/*< public >*/
gint (*send)(PurpleProtocolIM *im, PurpleConnection *connection,
PurpleMessage *msg);
guint (*send_typing)(PurpleProtocolIM *im, PurpleConnection *connection,
const gchar *name, PurpleIMTypingState state);
/*< private >*/
gpointer reserved[4];
};
G_BEGIN_DECLS
/**
* purple_protocol_im_send:
* @im: The #PurpleProtocolIM instance.
* @connection: The #PurpleConnection to use.
* @msg: The #PurpleMessage to send.
*
* Sends @msg out over @connection.
*
* Returns: >= 0 on success, or < 0 on error. If 0 is returned the message
* should not be output locally.
*/
gint purple_protocol_im_send(PurpleProtocolIM *im,
PurpleConnection *connection,
PurpleMessage *msg);
/**
* purple_protocol_im_send_typing:
* @im: The #PurpleProtocolIM instance.
* @connection: The #PurpleConversation to use.
* @name: The name of the user to send a typing notification to.
* @state: The #PurpleIMTypingState Value.
*
* If this protocol requires the #PURPLE_IM_TYPING message to be sent
* repeatedly to signify that the user is still typing, then the protocol
* should return the number of seconds to wait before sending a subsequent
* notification. Otherwise the protocol should return 0.
*
* Returns: A quiet-period, specified in seconds, where Purple will not send
* any additional typing notification messages. Most protocols should
* return 0, which means that no additional #PURPLE_IM_TYPING messages
* need to be sent. If this is 5, for example, then Purple will wait
* five seconds, and if the Purple user is still typing then Purple
* will send another #PURPLE_IM_TYPING message.
*/
guint purple_protocol_im_send_typing(PurpleProtocolIM *im,
PurpleConnection *connection,
const gchar *name,
PurpleIMTypingState state);
G_END_DECLS
#endif /* PURPLE_PROTOCOL_IM_H */