mirror of
https://github.com/taglib/taglib.git
synced 2025-07-24 07:54:32 -04:00
comment writing support and more tests for mod and xm
This commit is contained in:
Mathias Panzenböck
@ -24,10 +24,327 @@
|
||||
#include "xmfile.h"
|
||||
#include "modfileprivate.h"
|
||||
|
||||
#include <string.h>
|
||||
#include <algorithm>
|
||||
|
||||
using namespace TagLib;
|
||||
using namespace XM;
|
||||
using TagLib::uint;
|
||||
using TagLib::ushort;
|
||||
using TagLib::ulong;
|
||||
|
||||
/*!
|
||||
* The *Reader classes are helpers to make handling of the stripped XM
|
||||
* format more easy. In the stripped XM format certain headersizes might
|
||||
* be smaller than one would expect. The fields that are not included
|
||||
* are then just some predefined valued (e.g. 0).
|
||||
*
|
||||
* Using these classes this code:
|
||||
*
|
||||
* if(headerSize >= 4)
|
||||
* {
|
||||
* if(!readU16L(value1)) ERROR();
|
||||
* if(headerSize >= 8)
|
||||
* {
|
||||
* if(!readU16L(value2)) ERROR();
|
||||
* if(headerSize >= 12)
|
||||
* {
|
||||
* if(!readString(value3, 22)) ERROR();
|
||||
* ...
|
||||
* }
|
||||
* }
|
||||
* }
|
||||
*
|
||||
* Becomes:
|
||||
*
|
||||
* StructReader header;
|
||||
* header.u16L(value1).u16L(value2).string(value3, 22). ...;
|
||||
* if(header.read(*this, headerSize) < std::min(header.size(), headerSize))
|
||||
* ERROR();
|
||||
*
|
||||
* Maybe if this is useful to other formats this cleasses can be moved to
|
||||
* their onw public files.
|
||||
*/
|
||||
class Reader
|
||||
{
|
||||
public:
|
||||
virtual ~Reader()
|
||||
{
|
||||
}
|
||||
|
||||
/*!
|
||||
* Reads associated values from \a file, but never reads more
|
||||
* then \a limit bytes.
|
||||
*/
|
||||
virtual uint read(TagLib::File &file, uint limit) = 0;
|
||||
|
||||
/*!
|
||||
* Returns the number of bytes this reader would like to read.
|
||||
*/
|
||||
virtual uint size() const = 0;
|
||||
};
|
||||
|
||||
class SkipReader : public Reader
|
||||
{
|
||||
public:
|
||||
SkipReader(uint size) : m_size(size)
|
||||
{
|
||||
}
|
||||
|
||||
uint read(TagLib::File &file, uint limit)
|
||||
{
|
||||
uint count = std::min(m_size, limit);
|
||||
file.seek(count, TagLib::File::Current);
|
||||
return count;
|
||||
}
|
||||
|
||||
uint size() const
|
||||
{
|
||||
return m_size;
|
||||
}
|
||||
|
||||
private:
|
||||
uint m_size;
|
||||
};
|
||||
|
||||
template<typename T>
|
||||
class ValueReader : public Reader
|
||||
{
|
||||
public:
|
||||
ValueReader(T &value) : value(value)
|
||||
{
|
||||
}
|
||||
|
||||
protected:
|
||||
T &value;
|
||||
};
|
||||
|
||||
class StringReader : public ValueReader<String>
|
||||
{
|
||||
public:
|
||||
StringReader(String &string, uint size) :
|
||||
ValueReader<String>(string), m_size(size)
|
||||
{
|
||||
}
|
||||
|
||||
uint read(TagLib::File &file, uint limit)
|
||||
{
|
||||
ByteVector data = file.readBlock(std::min(m_size,limit));
|
||||
uint count = data.size();
|
||||
int index = data.find((char) 0);
|
||||
if(index > -1)
|
||||
{
|
||||
data.resize(index);
|
||||
}
|
||||
data.replace((char) 0xff, ' ');
|
||||
value = data;
|
||||
return count;
|
||||
}
|
||||
|
||||
uint size() const
|
||||
{
|
||||
return m_size;
|
||||
}
|
||||
|
||||
private:
|
||||
uint m_size;
|
||||
};
|
||||
|
||||
class ByteReader : public ValueReader<uchar>
|
||||
{
|
||||
public:
|
||||
ByteReader(uchar &byte) : ValueReader<uchar>(byte) {}
|
||||
|
||||
uint read(TagLib::File &file, uint limit)
|
||||
{
|
||||
ByteVector data = file.readBlock(std::min(1U,limit));
|
||||
if(data.size() > 0)
|
||||
{
|
||||
value = data[0];
|
||||
}
|
||||
return data.size();
|
||||
}
|
||||
|
||||
uint size() const
|
||||
{
|
||||
return 1;
|
||||
}
|
||||
};
|
||||
|
||||
template<typename T>
|
||||
class NumberReader : public ValueReader<T>
|
||||
{
|
||||
public:
|
||||
NumberReader(T &value, bool bigEndian) :
|
||||
ValueReader<T>(value), bigEndian(bigEndian)
|
||||
{
|
||||
}
|
||||
|
||||
protected:
|
||||
bool bigEndian;
|
||||
};
|
||||
|
||||
class U16Reader : public NumberReader<ushort>
|
||||
{
|
||||
public:
|
||||
U16Reader(ushort &value, bool bigEndian)
|
||||
: NumberReader<ushort>(value, bigEndian) {}
|
||||
|
||||
uint read(TagLib::File &file, uint limit)
|
||||
{
|
||||
ByteVector data = file.readBlock(std::min(2U,limit));
|
||||
value = data.toUShort(bigEndian);
|
||||
return data.size();
|
||||
}
|
||||
|
||||
uint size() const
|
||||
{
|
||||
return 2;
|
||||
}
|
||||
};
|
||||
|
||||
class U32Reader : public NumberReader<ulong>
|
||||
{
|
||||
public:
|
||||
U32Reader(ulong &value, bool bigEndian = true) :
|
||||
NumberReader<ulong>(value, bigEndian)
|
||||
{
|
||||
}
|
||||
|
||||
uint read(TagLib::File &file, uint limit)
|
||||
{
|
||||
ByteVector data = file.readBlock(std::min(4U,limit));
|
||||
value = data.toUInt(bigEndian);
|
||||
return data.size();
|
||||
}
|
||||
|
||||
uint size() const
|
||||
{
|
||||
return 4;
|
||||
}
|
||||
};
|
||||
|
||||
class StructReader : public Reader
|
||||
{
|
||||
public:
|
||||
StructReader()
|
||||
{
|
||||
m_readers.setAutoDelete(true);
|
||||
}
|
||||
|
||||
/*!
|
||||
* Add a nested reader. This reader takes ownership.
|
||||
*/
|
||||
StructReader &reader(Reader *reader)
|
||||
{
|
||||
m_readers.append(reader);
|
||||
return *this;
|
||||
}
|
||||
|
||||
/*!
|
||||
* Don't read anything but skip \a size bytes.
|
||||
*/
|
||||
StructReader &skip(uint size)
|
||||
{
|
||||
m_readers.append(new SkipReader(size));
|
||||
return *this;
|
||||
}
|
||||
|
||||
/*!
|
||||
* Read a string of \a size characters (bytes) into \a string.
|
||||
*/
|
||||
StructReader &string(String &string, uint size)
|
||||
{
|
||||
m_readers.append(new StringReader(string, size));
|
||||
return *this;
|
||||
}
|
||||
|
||||
/*!
|
||||
* Read a byte into \a byte.
|
||||
*/
|
||||
StructReader &byte(uchar &byte)
|
||||
{
|
||||
m_readers.append(new ByteReader(byte));
|
||||
return *this;
|
||||
}
|
||||
|
||||
/*!
|
||||
* Read a unsigned 16 Bit integer into \a number. The byte order
|
||||
* is controlled by \a bigEndian.
|
||||
*/
|
||||
StructReader &u16(ushort &number, bool bigEndian)
|
||||
{
|
||||
m_readers.append(new U16Reader(number, bigEndian));
|
||||
return *this;
|
||||
}
|
||||
|
||||
/*!
|
||||
* Read a unsigned 16 Bit little endian integer into \a number.
|
||||
*/
|
||||
StructReader &u16L(ushort &number)
|
||||
{
|
||||
return u16(number, false);
|
||||
}
|
||||
|
||||
/*!
|
||||
* Read a unsigned 16 Bit big endian integer into \a number.
|
||||
*/
|
||||
StructReader &u16B(ushort &number)
|
||||
{
|
||||
return u16(number, true);
|
||||
}
|
||||
|
||||
/*!
|
||||
* Read a unsigned 32 Bit integer into \a number. The byte order
|
||||
* is controlled by \a bigEndian.
|
||||
*/
|
||||
StructReader &u32(ulong &number, bool bigEndian)
|
||||
{
|
||||
m_readers.append(new U32Reader(number, bigEndian));
|
||||
return *this;
|
||||
}
|
||||
|
||||
/*!
|
||||
* Read a unsigned 32 Bit little endian integer into \a number.
|
||||
*/
|
||||
StructReader &u32L(ulong &number)
|
||||
{
|
||||
return u32(number, false);
|
||||
}
|
||||
|
||||
/*!
|
||||
* Read a unsigned 32 Bit big endian integer into \a number.
|
||||
*/
|
||||
StructReader &u32B(ulong &number)
|
||||
{
|
||||
return u32(number, true);
|
||||
}
|
||||
|
||||
uint size() const
|
||||
{
|
||||
uint size = 0;
|
||||
for(List<Reader*>::ConstIterator i = m_readers.begin(); i != m_readers.end(); ++ i)
|
||||
{
|
||||
size += (*i)->size();
|
||||
}
|
||||
return size;
|
||||
}
|
||||
|
||||
uint read(TagLib::File &file, uint limit)
|
||||
{
|
||||
uint sumcount = 0;
|
||||
for(List<Reader*>::Iterator i = m_readers.begin(); limit > 0 && i != m_readers.end(); ++ i)
|
||||
{
|
||||
uint count = (*i)->read(file, limit);
|
||||
limit -= count;
|
||||
sumcount += count;
|
||||
}
|
||||
return sumcount;
|
||||
}
|
||||
|
||||
private:
|
||||
List<Reader*> m_readers;
|
||||
};
|
||||
|
||||
class XM::File::FilePrivate
|
||||
{
|
||||
@ -83,7 +400,105 @@ bool XM::File::save()
|
||||
writeString(d->tag.title(), 20);
|
||||
seek(1, Current);
|
||||
writeString(d->tag.trackerName(), 20);
|
||||
// TODO: write comment as instrument and sample names
|
||||
seek(2, Current);
|
||||
ulong headerSize = 0;
|
||||
if(!readU32L(headerSize))
|
||||
return false;
|
||||
seek(2+2+2, Current);
|
||||
|
||||
ushort patternCount = 0;
|
||||
ushort instrumentCount = 0;
|
||||
if(!readU16L(patternCount) || !readU16L(instrumentCount))
|
||||
return false;
|
||||
|
||||
seek(60 + headerSize);
|
||||
|
||||
// need to read patterns again in order to seek to the instruments:
|
||||
for(ushort i = 0; i < patternCount; ++ i)
|
||||
{
|
||||
ulong patternHeaderLength = 0;
|
||||
if(!readU32L(patternHeaderLength) || patternHeaderLength < 4)
|
||||
return false;
|
||||
|
||||
ushort dataSize = 0;
|
||||
StructReader pattern;
|
||||
pattern.skip(3).u16L(dataSize);
|
||||
|
||||
uint count = pattern.read(*this, patternHeaderLength - 4U);
|
||||
if(count != std::min(patternHeaderLength - 4U, (ulong)pattern.size()))
|
||||
return false;
|
||||
|
||||
seek(patternHeaderLength - (4 + count) + dataSize, Current);
|
||||
}
|
||||
|
||||
StringList lines = d->tag.comment().split("\n");
|
||||
uint sampleNameIndex = instrumentCount;
|
||||
for(ushort i = 0; i < instrumentCount; ++ i)
|
||||
{
|
||||
ulong instrumentHeaderSize = 0;
|
||||
if(!readU32L(instrumentHeaderSize) || instrumentHeaderSize < 4)
|
||||
return false;
|
||||
|
||||
uint len = std::min(22UL, instrumentHeaderSize - 4U);
|
||||
if(i > lines.size())
|
||||
writeString(String::null, len);
|
||||
else
|
||||
writeString(lines[i], len);
|
||||
|
||||
ushort sampleCount = 0;
|
||||
long offset = 0;
|
||||
if(instrumentHeaderSize >= 29U)
|
||||
{
|
||||
seek(1, Current);
|
||||
if(!readU16L(sampleCount))
|
||||
return false;
|
||||
|
||||
if(sampleCount > 0)
|
||||
{
|
||||
ulong sampleHeaderSize = 0;
|
||||
if(instrumentHeaderSize < 33U || !readU32L(sampleHeaderSize))
|
||||
return false;
|
||||
// skip unhandeled header proportion:
|
||||
seek(instrumentHeaderSize - 33, Current);
|
||||
|
||||
for(ushort j = 0; j < sampleCount; ++ j)
|
||||
{
|
||||
if(sampleHeaderSize > 4U)
|
||||
{
|
||||
ulong length = 0;
|
||||
if(!readU32L(length))
|
||||
return false;
|
||||
offset += length;
|
||||
|
||||
seek(std::min(sampleHeaderSize, 14UL), Current);
|
||||
if(sampleHeaderSize > 18U)
|
||||
{
|
||||
uint len = std::min(sampleHeaderSize - 18U, 22UL);
|
||||
if(sampleNameIndex >= lines.size())
|
||||
writeString(String::null, len);
|
||||
else
|
||||
writeString(lines[sampleNameIndex ++], len);
|
||||
seek(sampleHeaderSize - (18U + len), Current);
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
seek(sampleHeaderSize, Current);
|
||||
}
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
offset = instrumentHeaderSize - 29;
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
offset = instrumentHeaderSize - (4 + len);
|
||||
}
|
||||
seek(offset, Current);
|
||||
}
|
||||
|
||||
return true;
|
||||
}
|
||||
|
||||
@ -92,97 +507,143 @@ void XM::File::read(bool)
|
||||
if(!isOpen())
|
||||
return;
|
||||
|
||||
READ_ASSERT(readBlock(17) == "Extended Module: ");
|
||||
seek(0);
|
||||
ByteVector magic = readBlock(17);
|
||||
// it's all 0x00 for stripped XM files:
|
||||
READ_ASSERT(magic == "Extended Module: " || magic == ByteVector(17, 0));
|
||||
|
||||
READ_STRING(d->tag.setTitle, 20);
|
||||
READ_BYTE_AS(mark);
|
||||
READ_ASSERT(mark == 0x1A);
|
||||
READ_BYTE_AS(escape);
|
||||
// in stripped XM files this is 0x00:
|
||||
READ_ASSERT(escape == 0x1A || escape == 0x00);
|
||||
|
||||
READ_STRING(d->tag.setTrackerName, 20);
|
||||
READ_U16L(d->properties.setVersion);
|
||||
|
||||
READ_U32L_AS(headerSize);
|
||||
READ_U16L(d->properties.setTableLength);
|
||||
READ_U16L(d->properties.setRestartPosition);
|
||||
READ_U16L(d->properties.setChannels);
|
||||
READ_U16L_AS(patternCount);
|
||||
READ_ASSERT(headerSize >= 4);
|
||||
|
||||
ushort tableLength = 0;
|
||||
ushort restartPosition = 0;
|
||||
ushort channels = 0;
|
||||
ushort patternCount = 0;
|
||||
ushort instrumentCount = 0;
|
||||
ushort flags = 0;
|
||||
ushort tempo = 0;
|
||||
ushort bpmSpeed = 0;
|
||||
|
||||
StructReader header;
|
||||
header.u16L(tableLength)
|
||||
.u16L(restartPosition)
|
||||
.u16L(channels)
|
||||
.u16L(patternCount)
|
||||
.u16L(instrumentCount)
|
||||
.u16L(flags)
|
||||
.u16L(tempo)
|
||||
.u16L(bpmSpeed);
|
||||
|
||||
uint count = header.read(*this, headerSize - 4U);
|
||||
uint size = std::min(headerSize - 4U, (ulong)header.size());
|
||||
|
||||
READ_ASSERT(count == size);
|
||||
|
||||
d->properties.setTableLength(tableLength);
|
||||
d->properties.setRestartPosition(restartPosition);
|
||||
d->properties.setChannels(channels);
|
||||
d->properties.setPatternCount(patternCount);
|
||||
READ_U16L_AS(instrumentCount);
|
||||
d->properties.setInstrumentCount(instrumentCount);
|
||||
READ_U16L(d->properties.setFlags);
|
||||
READ_U16L(d->properties.setTempo);
|
||||
READ_U16L(d->properties.setBpmSpeed);
|
||||
d->properties.setFlags(flags);
|
||||
d->properties.setTempo(tempo);
|
||||
d->properties.setBpmSpeed(bpmSpeed);
|
||||
|
||||
seek(60 + headerSize);
|
||||
|
||||
|
||||
// read patterns:
|
||||
for(ushort i = 0; i < patternCount; ++ i)
|
||||
{
|
||||
READ_U32L_AS(patternHeaderLength);
|
||||
READ_BYTE_AS(patternType);
|
||||
READ_U16L_AS(rowCount);
|
||||
READ_U16L_AS(patternDataSize);
|
||||
READ_ASSERT(patternHeaderLength >= 4);
|
||||
|
||||
uchar packingType = 0;
|
||||
ushort rowCount = 0;
|
||||
ushort dataSize = 0;
|
||||
StructReader pattern;
|
||||
pattern.byte(packingType).u16L(rowCount).u16L(dataSize);
|
||||
|
||||
seek(patternHeaderLength - (4+1+2+2) + patternDataSize, Current);
|
||||
uint count = pattern.read(*this, patternHeaderLength - 4U);
|
||||
READ_ASSERT(count == std::min(patternHeaderLength - 4U, (ulong)pattern.size()));
|
||||
|
||||
seek(patternHeaderLength - (4 + count) + dataSize, Current);
|
||||
}
|
||||
|
||||
StringList intrumentNames;
|
||||
StringList sampleNames;
|
||||
|
||||
// read instruments:
|
||||
for(ushort i = 0; i < instrumentCount; ++ i)
|
||||
{
|
||||
long pos = tell();
|
||||
READ_U32L_AS(instrumentSize);
|
||||
READ_U32L_AS(instrumentHeaderSize);
|
||||
READ_ASSERT(instrumentHeaderSize >= 4);
|
||||
|
||||
String instrumentName;
|
||||
uchar instrumentType = 0;
|
||||
uchar instrumentType = 0;
|
||||
ushort sampleCount = 0;
|
||||
|
||||
if(instrumentSize > 4)
|
||||
{
|
||||
READ_ASSERT(readString(instrumentName, std::min(22UL, instrumentSize-4)));
|
||||
|
||||
if(instrumentSize >= (4+22+1))
|
||||
{
|
||||
READ_ASSERT(readByte(instrumentType));
|
||||
StructReader instrument;
|
||||
instrument.string(instrumentName, 22).byte(instrumentType).u16L(sampleCount);
|
||||
|
||||
if(instrumentSize >= (4+22+1+2))
|
||||
{
|
||||
READ_ASSERT(readU16L(sampleCount));
|
||||
}
|
||||
}
|
||||
}
|
||||
// 4 for instrumentHeaderSize
|
||||
uint count = 4 + instrument.read(*this, instrumentHeaderSize - 4U);
|
||||
READ_ASSERT(count == std::min(instrumentHeaderSize, (ulong)instrument.size() + 4));
|
||||
|
||||
ulong sumSampleLength = 0;
|
||||
ulong sampleHeaderSize = 0;
|
||||
long offset = 0;
|
||||
if(sampleCount > 0)
|
||||
{
|
||||
if(!readU32L(sampleHeaderSize))
|
||||
{
|
||||
setValid(false);
|
||||
return;
|
||||
}
|
||||
// wouldn't know which header size to assume otherwise:
|
||||
READ_ASSERT(instrumentHeaderSize >= count + 4 && readU32L(sampleHeaderSize));
|
||||
// skip unhandeled header proportion:
|
||||
seek(instrumentHeaderSize - count - 4, Current);
|
||||
|
||||
seek(pos + instrumentSize);
|
||||
|
||||
long sampleheaderPos = tell();
|
||||
for(ushort j = 0; j < sampleCount; ++ j)
|
||||
{
|
||||
seek(sampleheaderPos + sampleHeaderSize * j);
|
||||
READ_U32L_AS(length);
|
||||
READ_U32L_AS(loopStart);
|
||||
READ_U32L_AS(loopLength);
|
||||
READ_BYTE_AS(volume);
|
||||
READ_BYTE_AS(finetune);
|
||||
READ_BYTE_AS(sampleType);
|
||||
READ_BYTE_AS(panning);
|
||||
READ_BYTE_AS(noteNumber);
|
||||
READ_BYTE_AS(compression);
|
||||
READ_STRING_AS(sampleName, 22);
|
||||
|
||||
sumSampleLength += length;
|
||||
ulong length = 0;
|
||||
ulong loopStart = 0;
|
||||
ulong loopLength = 0;
|
||||
uchar volume = 0;
|
||||
uchar finetune = 0;
|
||||
uchar sampleType = 0;
|
||||
uchar panning = 0;
|
||||
uchar noteNumber = 0;
|
||||
uchar compression = 0;
|
||||
String sampleName;
|
||||
StructReader sample;
|
||||
sample.u32L(length)
|
||||
.u32L(loopStart)
|
||||
.u32L(loopLength)
|
||||
.byte(volume)
|
||||
.byte(finetune)
|
||||
.byte(sampleType)
|
||||
.byte(panning)
|
||||
.byte(noteNumber)
|
||||
.byte(compression)
|
||||
.string(sampleName, 22);
|
||||
|
||||
uint count = sample.read(*this, sampleHeaderSize);
|
||||
READ_ASSERT(count == std::min(sampleHeaderSize, (ulong)sample.size()));
|
||||
// skip unhandeled header proportion:
|
||||
seek(sampleHeaderSize - count, Current);
|
||||
|
||||
offset += length;
|
||||
sampleNames.append(sampleName);
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
offset = instrumentHeaderSize - count;
|
||||
}
|
||||
intrumentNames.append(instrumentName);
|
||||
seek(pos + instrumentSize + sampleHeaderSize * sampleCount + sumSampleLength);
|
||||
seek(offset, Current);
|
||||
}
|
||||
|
||||
String comment(intrumentNames.toString("\n"));
|
||||
|
||||
Reference in New Issue
Block a user