pidgin/pidgin

Fix several leaks in tests
default tip
2 hours 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_MARKUP_H
#define PURPLE_MARKUP_H
#include <glib.h>
G_BEGIN_DECLS
/**
* purple_markup_html_to_xhtml:
* @html: The HTML markup.
* @dest_xhtml: The destination XHTML output.
* @dest_plain: The destination plain-text output.
*
* Converts HTML markup to XHTML.
*/
void purple_markup_html_to_xhtml(const char *html, char **dest_xhtml,
char **dest_plain);
/**
* purple_markup_strip_html:
* @str: The string to strip HTML from.
*
* Strips HTML tags from a string.
*
* Returns: The new string without HTML. You must g_free this string
* when finished with it.
*/
char *purple_markup_strip_html(const char *str);
/**
* purple_markup_linkify:
* @str: The string to linkify.
*
* Adds the necessary HTML code to turn URIs into HTML links in a string.
*
* Returns: The new string with all URIs surrounded in standard
* HTML &lt;a href="whatever"&gt;&lt;/a&gt; tags. You must g_free()
* this string when finished with it.
*/
char *purple_markup_linkify(const char *str);
/**
* purple_unescape_text:
* @text: The string in which to unescape any HTML entities
*
* Unescapes HTML entities to their literal characters in the text.
* For example "&amp;amp;" is replaced by '&amp;' and so on. Also converts
* numerical entities (e.g. "&amp;\#38;" is also '&amp;').
*
* This function currently supports the following named entities:
* "&amp;amp;", "&amp;lt;", "&amp;gt;", "&amp;copy;", "&amp;quot;",
* "&amp;reg;", "&amp;apos;"
*
* purple_unescape_html() is similar, but also converts "&lt;br&gt;" into "\n".
*
* See purple_unescape_html()
*
* Returns: The text with HTML entities literalized. You must g_free
* this string when finished with it.
*/
char *purple_unescape_text(const char *text);
/**
* purple_unescape_html:
* @html: The string in which to unescape any HTML entities
*
* Unescapes HTML entities to their literal characters and converts
* "&lt;br&gt;" to "\n". See purple_unescape_text() for more details.
*
* See purple_unescape_text()
*
* Returns: The text with HTML entities literalized. You must g_free
* this string when finished with it.
*/
char *purple_unescape_html(const char *html);
/**
* purple_markup_slice:
* @str: The input NUL terminated, HTML, UTF-8 (or ASCII) string.
* @x: The character offset into an unformatted version of str to begin at.
* @y: The character offset (into an unformatted version of str) of one past
* the last character to include in the slice.
*
* Returns a newly allocated substring of the HTML UTF-8 string "str".
* The markup is preserved such that the substring will have the same
* formatting as original string, even though some tags may have been
* opened before "x", or may close after "y". All open tags are closed
* at the end of the returned string, in the proper order.
*
* Note that x and y are in character offsets, not byte offsets, and
* are offsets into an unformatted version of str. Because of this,
* this function may be sensitive to changes in GtkIMHtml and may break
* when used with other UI's. libpurple users are encouraged to report and
* work out any problems encountered.
*
* Returns: The HTML slice of string, with all formatting retained.
*/
char *purple_markup_slice(const char *str, guint x, guint y);
/**
* purple_markup_get_tag_name:
* @tag: The string starting a HTML tag.
*
* Returns a newly allocated string containing the name of the tag
* located at "tag". Tag is expected to point to a '<', and contain
* a '>' sometime after that. If there is no '>' and the string is
* not NUL terminated, this function can be expected to segfault.
*
* Returns: A string containing the name of the tag.
*/
char *purple_markup_get_tag_name(const char *tag);
/**
* purple_markup_unescape_entity:
* @text: A string containing an HTML entity.
* @length: If not %NULL, the string length of the entity is stored in this location.
*
* Returns a constant string of the character representation of the HTML
* entity pointed to by @text. For example, purple_markup_unescape_entity("&amp;amp;")
* will return "&amp;". The @text variable is expected to point to an '&amp;',
* the first character of the entity. If given an unrecognized entity, the function
* returns %NULL.
*
* Note that this function, unlike purple_unescape_html(), does not search
* the string for the entity, does not replace the entity, and does not
* return a newly allocated string.
*
* Returns: A constant string containing the character representation of the given entity.
*/
const char * purple_markup_unescape_entity(const char *text, int *length);
G_END_DECLS
#endif /* PURPLE_MARKUP_H */