b00a5c1aab
Add a test to check if ByteVector is detached correctly when being replaced.
2016-12-05 10:15:26 +09:00
f6a604f54b
#include guards in CMake generated headers.
2016-12-02 17:26:43 +09:00
489e2e6cbb
Update NEWS.
2016-12-01 15:25:30 +09:00
9336c82da3
Fix possible Ogg packet losses.
2016-12-01 11:32:01 +09:00
cfbaf34597
Prevent the segment table of Ogg pages from exceeding the size limit.
2016-12-01 10:51:59 +09:00
046c98230f
Remove Utils::floatByteOrder() and use systemByteOrder() instead.
...
We can safely assume that the integer and float byte orders are the same on IEEE754 compliant systems.
2016-11-29 14:58:39 +09:00
4381bd75f3
Add a test for #743/#779.
2016-11-29 10:53:33 +09:00
6df61cf2af
Small fix in style.
2016-11-29 10:38:11 +09:00
e9ef40fe7f
Merge pull request #779 from supermihi/master
...
Fixes #743 by not overwriting existing Xiph comment in FLAC::File::save
2016-11-28 21:03:39 +09:00
2786aa7463
Fixes #743 by not overwriting existing Xiph comment in FLAC::File::save
2016-11-27 19:17:13 +01:00
d3062f3af4
A bit more tolerant check to return itself in String::substr().
2016-11-26 13:05:14 +09:00
7871afec37
Merge pull request #778 from martin-flaska/optimization
...
String::substr optimization
2016-11-25 16:38:32 -05:00
c9a0754e3b
tstring: String::substr optimization when returning itself as a substring
...
Use copy ctor to return in a case whole string is being returned.
The intention was to optimize String::stripWhiteSpace for no-strip case
(without any leading or trailing white space removal).
copyFromUTF16 was used in any case previously and allocated duplicate
buffer for the same string - no implicit sharing.
Signed-off-by: Martin Flaska <martin.flaska@legrand.us >
2016-11-25 15:32:26 +01:00
6cfb11bb12
test_string: Make 'stripWhiteSpace' test more complex
...
No string without leading/trailing spaces was used in the test.
Signed-off-by: Martin Flaska <martin.flaska@legrand.us >
2016-11-25 13:56:39 +01:00
ad075a56f9
Suppress MSVC warnings in test.
2016-11-24 14:45:22 +09:00
f80a7c0d83
Merge pull request #776 from mathbunnyru/small_improvements
...
Small improvements
2016-11-23 23:31:49 -05:00
5e1d9fad31
Small fixes
2016-11-24 02:05:19 +03:00
eff28c55bf
Increment fixes
2016-11-22 01:10:28 +03:00
d5b9d7b8a7
Update NEWS.
2016-11-18 13:55:43 +09:00
ce77fbb0e7
Merge pull request #772 from TsudaKageyu/vorbis-fields
...
Fix handling of lowercase 'metadata_block_picture' field
2016-11-18 13:52:38 +09:00
b98a984b66
Fix handling of lowercase 'metadata_block_picture' fields in Vorbis comments.
...
Also refactored some redundant code for parsing pictures.
2016-11-11 00:07:32 +09:00
f9a747dceb
Avoid adding fields with invalid keys to Vorbis Comments.
...
According to the spec, '\x7F' is not allowed.
2016-11-10 23:35:14 +09:00
7b8d576bde
Don't decode redundant UTF-8 sequences in Win32.
...
Linux and OS X are working well and won't be affected.
2016-11-10 17:12:58 +09:00
2651372291
Separate some tests to make them more specific.
2016-11-09 15:51:33 +09:00
499f6db977
Check invalid Unicode APE keys properly.
2016-11-09 00:29:03 +09:00
9d58e9f8e8
Removed a utility function which is used only at one place.
2016-11-08 23:27:55 +09:00
56cd3695f7
Add README.md.
2016-11-08 22:56:42 +09:00
d81d894d41
tolower() depends on the current locale.
...
It's much easier to write our own function than to use locales properly.
2016-11-08 21:39:53 +09:00
e390cbac52
Update NEWS.
2016-11-08 21:17:00 +09:00
253c61e37d
Merge pull request #765 from TsudaKageyu/zero-length-atom
...
Proper handling of MP4 atoms with zero length.
2016-11-08 21:12:47 +09:00
1848b3bc6f
Merge pull request #759 from ufleisch/mp4_classical
...
Support new classical music frames introduced with iTunes 12.5, #758 .
2016-11-08 21:12:31 +09:00
aae23f3c07
Initialize all the data members of ID3v2::ChapterFrame.
2016-11-07 14:12:38 +09:00
da9df1b2a8
Values of FILE_* macros are guaranteed in Win32.
2016-11-07 00:42:12 +09:00
70334edd19
Add List::swap() and Map::swap().
2016-11-04 16:43:14 +09:00
f5ca097379
Proper handling of MP4 atoms with zero length.
...
If the size of an atom is 0, it designates the last atom which extends to the end of the file.
2016-11-02 15:44:50 +09:00
eb6d058ab9
Remove a useless branch.
...
longLength <= LONG_MAX is always true if sizeof(long) == sizeof(long long).
2016-11-01 16:03:15 +09:00
e6a69e24bc
Add a common function to generate a long string to test.
2016-10-31 20:01:52 +09:00
dcab8ed90e
Allow ScopedFileCopy to be const.
2016-10-31 10:29:13 +09:00
f5ce498182
Suppress MSVC warnings about narrowing conversions.
2016-10-30 23:51:35 +09:00
ee7fa78011
Update NEWS.
2016-10-30 22:51:15 +09:00
873c917081
Assume that SetFilePointerEx() and GetFileSizeEx() are always available.
...
This drops support for Windows 9x and NT 4.0 or older.
2016-10-30 22:36:18 +09:00
d3bd8fb7ff
Assume that CreateFileW() is always available.
...
This drops support for Windows 9x.
2016-10-30 22:25:34 +09:00
005441faaa
Prevent overflows, just in case.
2016-10-28 15:25:50 +09:00
935534aa53
Backport a test from taglib2 branch.
2016-10-28 15:19:35 +09:00
65a24bbc51
Remove some useless seek()/tell() from RIFF::File.
2016-10-28 14:22:50 +09:00
711b35cc6e
Encourage compilers to optimize out debug() and debugData().
...
It's sort of like a throwback, but I found that debug(const String &s) {} doesn't prevent a String from being constructed and the error messages from being embedded.
2016-10-28 11:19:51 +09:00
d53ca6f736
Update NEWS.
2016-10-27 15:40:14 +09:00
aa5f9bb221
Suppress some warnings in test.
2016-10-27 15:23:24 +09:00
d2b3547254
Add a test for File::truncate().
2016-10-27 15:15:22 +09:00
06ca9a099d
Check if file size <= LONG_MAX.
2016-10-27 15:10:34 +09:00
