8a24435db2
GUI: Update translation files.
2010-07-11 16:33:49 +03:00
d5946a8359
Merge branch 'gui-load-xml'
2010-07-11 16:30:22 +03:00
c9d63fa454
GUI: Remember user-selected base path.
...
Remember the base path user selects when opening files from loaded
XML report.
2010-07-11 16:20:19 +03:00
b8793b1529
GUI: Ask file location from user if not found.
...
When loading report from XML there is no full paths so the file's real
path is not known and cppcheck cannot open it. So if the file has no
absolute path then we ask where the file is located from the user.
2010-07-11 14:22:55 +03:00
f894b19209
GUI: Add errors read from XML report to GUI.
2010-07-11 02:02:08 +03:00
9a5166075f
GUI: Fix line numbers and filenames in errors.
...
When converting to use new ErrorItem and ErrorLine I made few mistakes
in how I handled the data. And for some reason there was not even
warnings about converting integers to QStrings.
2010-07-11 01:07:40 +03:00
352941f5df
GUI: Fix transporting errors.
...
Need to register integer list as new metatype so that Qt's type
system knows how to use it. Adding also additional constructors
for the ErrorItem.
2010-07-11 01:04:53 +03:00
faa483b8d0
GUI: Use integer list instead of variant list for line numbers.
2010-07-10 20:54:33 +03:00
2f0202d105
GUI: More refactoring to use ErrorItem and ErrorLine.
2010-07-10 20:30:31 +03:00
0e9d0e9bde
GUI: Refactor error data passing to use own class.
2010-07-10 18:20:45 +03:00
5e14abf735
GUI: Read errors from report XML file.
...
This commits adds new "Open XML" item to File-menu. Selecting this
menuitem allows user to select report file to open. When the file is
read the error data is printed to debug output. Later patches will
implement adding error data back to the GUI.
2010-07-10 16:37:36 +03:00
81449a823a
GUI: Refactoring xmlreport to use pointer to stream writing class.
2010-07-10 13:53:44 +03:00
78b8acfc88
Makefile: debug mode
2010-07-10 11:35:37 +02:00
3b391398c1
1.44: updated Makefile for release
2010-07-10 11:31:07 +02:00
72263d64b0
Changelog: updated for release
2010-07-10 11:30:03 +02:00
94a8eba989
1.44: updated version information
2010-07-10 11:29:07 +02:00
2e03e60218
Manual: Better description of -D. As suggested by kwin. Ticket #1212
2010-07-10 08:12:23 +02:00
4b0e3edfa4
Unit Testing: Test that suppressions work
2010-07-09 13:27:15 +02:00
abbd557761
Fixed #1836 (--suppresions file.txt fails due to wrong line edings)
2010-07-09 12:42:08 +02:00
2531bb73d7
Fixed #1834 (False positive: invalid iterator when erase() is used in if-else)
2010-07-09 10:50:24 +02:00
a60047227d
Settings: added '#include <cctype>'
2010-07-09 10:18:35 +02:00
b5d81d0a19
Update man page
2010-07-08 17:51:28 +03:00
c34c3ee742
Unused functions: Refactoring
2010-07-08 13:01:53 +02:00
1876cd482b
Suppressions: Validate given id
2010-07-08 12:06:27 +02:00
11a72461f3
Fixed #1835 (false positive: uninitialized variable when using ?)
2010-07-08 11:16:49 +02:00
ee2f785529
GUI: Remove custom lrelease target.
...
Remove the custom lrelease target since it fails to work in
Fedora 13. In Fedora 13 there is no 'lrelease' binary but
'lrelease-qt4'.
2010-07-08 10:50:46 +03:00
566b4b4beb
Variable usage: fixed false positive reported in #1729
2010-07-08 08:42:34 +02:00
2d6dfa57e1
Variable usage: Better aliasing support (Ticket #1729 )
2010-07-08 08:00:50 +02:00
49626e427e
Realloc: changed the error message. I think it's better to mention realloc in the message.
2010-07-07 20:28:15 +02:00
b5da0b8ed2
Fixed #1650 (Cppcheck deadlock)
...
http://sourceforge.net/apps/trac/cppcheck/ticket/1650
2010-07-07 15:42:39 +03:00
f5d3042635
Add more unit tests
2010-07-07 14:59:42 +03:00
f60a469932
Fixed #1831 (GUI: no progress bar when rechecking files)
...
Need to initialize the checking also when rechecking.
2010-07-07 12:25:47 +03:00
a1793edabe
Fixed #1825 (*log(0) error)
2010-07-07 09:03:40 +02:00
4ebf4caf66
Variable usage: only warn about variables that have variable id > 0
2010-07-07 08:50:34 +02:00
2bb07a55bd
Fixed #1803 (False positive: unused variable: this)
2010-07-07 08:39:18 +02:00
a107cd1b59
Fixed #1729 (False positive: variable is assigned a value that is never used (pointer aliasing))
2010-07-07 08:26:24 +02:00
b0f1d5669b
Fixed #1829 (### Internal error in Cppcheck. Please report it.)
2010-07-07 07:40:11 +02:00
dc2a0a6468
Fixed #1789 (false positive: memory leak (reallocation in subfunction through parameter))
2010-07-06 21:36:50 +02:00
1b20f8d27d
Fixed #1824 (false positive: unitialised variable)
2010-07-06 13:18:28 +02:00
1799e4a575
Fixed #1827 (### Internal error in Cppcheck. Please report it.)
2010-07-06 08:26:30 +02:00
c949a29d1e
Fixed #1821 (simplifyTypedef: better typedef support)
2010-07-05 23:03:54 +02:00
18a75edd20
Fixed #1823 (tokenize && wrong)
2010-07-05 22:41:02 +02:00
b02fc037ed
Buffer Overrun: Fixed false positive when variable is reassigned in called function
2010-07-05 22:19:27 +02:00
ae3557fa92
Fixed #1820 (False positive: memory leak (auto deallocated class))
2010-07-05 14:01:25 +02:00
bf7ce98178
GUI: updated swedish translation
2010-07-05 13:36:44 +02:00
b63c932320
GUI: updated swedish translation
2010-07-05 13:29:58 +02:00
b173d5f117
Fixed #1817 (False positive: Resource leak (casting))
2010-07-05 13:16:33 +02:00
c811acaa50
Fixed #1818 (False positive: Dangerous usage of strncpy (copying a constant string))
2010-07-05 12:45:39 +02:00
643d400716
GUI: Refactoring progress/finished handling.
...
Progress signal had also item count with it and then the handler determined
that check is ready when max count of progress was done. Also progressbar
was practically reset in every progress signal. This was simply fragile
code.
After this patch progress signal has only the current progress count. Total
count of items is given when initializing the checking. And there is
separate function for handling check finishing.
This also fixes the bug that progressbar was not hidden after checking
or when interrupting the checking.
2010-07-04 18:54:41 +03:00
acbf5af586
GUI: Fixup TS translation files after they got messed up earlier.
...
The lupdate got confused of removing "Show security errors" and there being
existing string "Show style errors". lupdate thought the "Show style errors"
was a new version ("similar enough") of "Show security errors" string
instead of new string to translate. Or actually existing translated string.
So many translations got "Show style errors" translated by earlier
translation of "Show security errors".
This commit fixes the mess by manually removing "Show security errors"
references and restoring correct translations of "Show style errors" from
earlier file versions.
Good lesson to learn about lupdate behavior.
2010-07-04 01:15:09 +03:00
Kimmo Varis
Daniel Marjamäki
Reijo Tomperi
Kimmo Varis
Robert Reif