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>
*
* 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 <pidgin.h> may be included directly"
#endif
#ifndef PURPLE_PERSON_H
#define PURPLE_PERSON_H
#include <glib.h>
#include <glib-object.h>
#include <gdk-pixbuf/gdk-pixbuf.h>
#include <libpurple/purpletags.h>
G_BEGIN_DECLS
#define PURPLE_TYPE_PERSON (purple_person_get_type())
G_DECLARE_FINAL_TYPE(PurplePerson, purple_person, PURPLE, PERSON, GObject)
#include <libpurple/purplecontactinfo.h>
/**
* PurplePerson:
*
* A collection of [class@Purple.ContactInfo] that contains a user selectable
* custom avatar and alias.
*
* Since: 3.0.0
*/
/**
* purple_person_new:
*
* Creates a new [class@Purple.Person].
*
* Returns: (transfer full): The new instance.
*
* Since: 3.0.0
*/
PurplePerson *purple_person_new(void);
/**
* purple_person_get_id:
* @person: The instance.
*
* Gets the id of @person. This is created internally and is read-only.
*
* Returns: The id of @person.
*
* Since: 3.0.0
*/
const gchar *purple_person_get_id(PurplePerson *person);
/**
* purple_person_get_alias:
* @person: The instance.
*
* Gets the alias of @person that was set by the libpurple user.
*
* This will be %NULL if no alias is set.
*
* Returns: (nullable): The alias of @person or %NULL.
*
* Since: 3.0.0
*/
const gchar *purple_person_get_alias(PurplePerson *person);
/**
* purple_person_set_alias:
* @person: The instance.
* @alias: (nullable): The new alias.
*
* Sets the alias of @person to @alias.
*
* This should only be called in direct response to a user interaction to set a
* custom alias. Protocol plugins should only be setting the alias of
* [class@Purple.Contact].
*
* If @alias is %NULL, then the previous alias is cleared.
*
* Since: 3.0.0
*/
void purple_person_set_alias(PurplePerson *person, const gchar *alias);
/**
* purple_person_get_avatar_for_display:
* @person: The instance.
*
* Gets the avatar that should be displayed for @person. If @person has an
* avatar set, it will be returned, otherwise this will fallback to the value
* of [property@Purple.ContactInfo:avatar] for
* [property@Purple.Person:priority-contact-info].
*
* Returns: (transfer none) (nullable): The avatar to display.
*
* Since: 3.0.0
*/
GdkPixbuf *purple_person_get_avatar_for_display(PurplePerson *person);
/**
* purple_person_get_avatar:
* @person: The instance.
*
* Gets the avatar of @person or %NULL if no avatar is set.
*
* Returns: (transfer none) (nullable): The avatar for @person.
*
* Since: 3.0.0
*/
GdkPixbuf *purple_person_get_avatar(PurplePerson *person);
/**
* purple_person_set_avatar:
* @person: The instance.
* @avatar: (nullable): The new avatar.
*
* Sets the avatar of @person to @avatar. If @avatar is %NULL then the previous
* avatar is cleared.
*
* This should only be called in direct response to a user interaction to set a
* custom avatar. Protocol plugins should only be setting the avatars of
* [class@Purple.Contact].
*
* Since: 3.0.0
*/
void purple_person_set_avatar(PurplePerson *person, GdkPixbuf *avatar);
/**
* purple_person_get_tags:
* @person: The instance.
*
* Gets the [class@Purple.Tags] instance for @person.
*
* Returns: (transfer none): The tags for @person.
*
* Since: 3.0.0
*/
PurpleTags *purple_person_get_tags(PurplePerson *person);
/**
* purple_person_get_name_for_display:
* @person: The instance.
*
* Gets the name that should be displayed for @person. See
* [property@Purple.Person:name-for-display] for more information.
*
* Returns: (transfer none): The name to display for @person.
*
* Since: 3.0.0
*/
const char *purple_person_get_name_for_display(PurplePerson *person);
/**
* purple_person_add_contact:
* @person: The instance.
* @info: The [class@Purple.ContactInfo] to add.
*
* Adds @info to @person.
*
* Duplicate contacts are currently allowed, but that may change at a later
* time.
*
* Since: 3.0.0
*/
void purple_person_add_contact_info(PurplePerson *person, PurpleContactInfo *info);
/**
* purple_person_remove_contact_info:
* @person: The instance.
* @info: The [class@Purple.ContactInfo] to remove.
*
* Removes @info from @person.
*
* Returns: %TRUE if @info was found and removed otherwise %FALSE.
*
* Since: 3.0.0
*/
gboolean purple_person_remove_contact_info(PurplePerson *person, PurpleContactInfo *info);
/**
* purple_person_get_priority_contact_info:
* @person: The instance.
*
* Gets the contact info for the most available contact info in @person.
*
* Returns: (transfer none) (nullable): The priority contact or %NULL if
* @person does not have any contacts.
*
* Since: 3.0.0
*/
PurpleContactInfo *purple_person_get_priority_contact_info(PurplePerson *person);
/*
* purple_person_has_contacts:
* @person: The instance.
*
* Gets whether or not @person has any contacts.
*
* Returns: %TRUE if @person has at least one contact, otherwise %FALSE.
*
* Since: 3.0.0
*/
gboolean purple_person_has_contacts(PurplePerson *person);
/**
* purple_person_matches:
* @person: The instance.
* @needle: (nullable): The string to match on.
*
* Checks if the alias matches @needle. This also checks @needle against
* [method@Purple.ContactInfo.matches] for each [class@Purple.ContactInfo] that
* @person is tracking.
*
* If @needle is %NULL or empty string, %TRUE will be returned.
*
* Returns: %TRUE if @person matches @needle in any way.
*
* Since: 3.0.0
*/
gboolean purple_person_matches(PurplePerson *person, const char *needle);
G_END_DECLS
#endif /* PURPLE_PERSON_H */