void CGapAnalysis::x_AddGapsFromBases(
const CSeqMap_CI & seqmap_ci,
TSeqIdConstRef bioseq_seq_id,
TSeqPos iBioseqLength,
TFlag fFlags)
{
const TSeqPos begin_pos = seqmap_ci.GetPosition();
// get location representing this segment's bases
CRef<CSeq_loc> loc_of_bases(
new CSeq_loc(
*SerialClone(*bioseq_seq_id),
begin_pos,
(begin_pos + seqmap_ci.GetLength() - 1)));
CSeqVector seq_vec(
*loc_of_bases, *seqmap_ci.GetScope(), CBioseq_Handle::eCoding_Iupac);
const char kGapChar = seq_vec.GetGapChar(
CSeqVectorTypes::eCaseConversion_upper);
// a simple "runs of unknown bases" algo
size_t size_of_curr_gap = 0;
size_t start_pos_of_curr_gap = kInvalidSeqPos;
CSeqVector_CI seq_vec_ci = seq_vec.begin();
for( ; seq_vec_ci; ++seq_vec_ci ) {
if( *seq_vec_ci == kGapChar ) {
++size_of_curr_gap;
if( start_pos_of_curr_gap == kInvalidSeqPos ) {
start_pos_of_curr_gap = (begin_pos + seq_vec_ci.GetPos());
}
} else if( size_of_curr_gap > 0 ) {
_ASSERT(start_pos_of_curr_gap != kInvalidSeqPos);
AddGap(
eGapType_UnknownBases, bioseq_seq_id, size_of_curr_gap,
iBioseqLength,
start_pos_of_curr_gap, (begin_pos + seq_vec_ci.GetPos()),
fFlags);
size_of_curr_gap = 0;
start_pos_of_curr_gap = kInvalidSeqPos;
}
}
if( size_of_curr_gap > 0 ) {
_ASSERT(start_pos_of_curr_gap != kInvalidSeqPos);
AddGap(
eGapType_UnknownBases, bioseq_seq_id, size_of_curr_gap,
iBioseqLength,
start_pos_of_curr_gap, (begin_pos + seq_vec_ci.GetPos()),
fFlags);
}
}
void CGapList::AddGap(uint16 part)
{
if (part >= m_iPartCount) {
wxFAIL;
return;
}
uint64 gapstart = part * PARTSIZE;
uint64 gapend = gapstart + GetPartSize(part) - 1;
AddGap(gapstart, gapend);
m_partsComplete[part] = incomplete;
}
void CMetfile::FillGap(unsigned int start, unsigned int end)
{
int todelete = 0;
unsigned long newStart, newEnd;
GAP * nextGap;
newStart = 0;
newEnd = 0;
GAP * currentGap = gaplist;
while (currentGap)
{
nextGap = currentGap -> next;
if (start <= currentGap -> start)
{
if (end > currentGap -> start)
{
if (end < currentGap -> end)
{
currentGap -> start = end;
}
else
{
todelete++;
}
}
}
else if(start < currentGap -> end)
{
if (end < currentGap -> end)
{
newStart = end;
newEnd = currentGap -> end;
}
currentGap -> end = start;
}
if (todelete)
{
if (gaplist == currentGap)
{
gaplist = currentGap -> next;
}
free(currentGap);
todelete = 0;
}
else if(newEnd)
{
AddGap(newStart, newEnd);
newEnd = 0;
}
currentGap = nextGap;
}
}
void CGapList::Init(uint64 fileSize, bool isEmpty)
{
m_filesize = fileSize;
m_iPartCount = fileSize / PARTSIZE + 1;
m_sizeLastPart = fileSize % PARTSIZE;
// file with size of n * PARTSIZE
if (m_sizeLastPart == 0
&& fileSize) { // that's only for pre-init in ctor
m_sizeLastPart = PARTSIZE;
m_iPartCount--;
}
m_gaplist.clear();
m_partsComplete.clear();
if (isEmpty) {
m_partsComplete.resize(m_iPartCount, incomplete);
AddGap(0, fileSize - 1);
m_totalGapSize = fileSize;
} else {
m_partsComplete.resize(m_iPartCount, complete);
m_totalGapSize = 0;
}
m_totalGapSizeValid = true;
}
void CGapAnalysis::AddBioseqGaps(
const CBioseq_Handle & bioseq_h,
TAddFlag add_flags,
TFlag fFlags,
size_t max_resolve_count)
{
// get CSeq_id of CBioseq
TSeqIdConstRef pSeqId = bioseq_h.GetSeqId();
const TSeqPos bioseq_len = bioseq_h.GetBioseqLength();
// fFlags control what we look at
CSeqMap::TFlags seq_map_flags = 0;
if( add_flags & fAddFlag_IncludeSeqGaps ) {
seq_map_flags |= CSeqMap::fFindGap;
}
if( add_flags & fAddFlag_IncludeUnknownBases ) {
seq_map_flags |= CSeqMap::fFindData;
}
TSeqPos end_of_last_segment = 0; // exclusive
bool all_segments_and_in_order = true;
SSeqMapSelector selector;
selector.SetFlags(seq_map_flags).SetResolveCount(max_resolve_count);
CSeqMap_CI seqmap_ci(bioseq_h, selector);
for( ; seqmap_ci; ++seqmap_ci ) {
if( seqmap_ci.GetPosition() != end_of_last_segment ) {
all_segments_and_in_order = false;
}
end_of_last_segment = seqmap_ci.GetEndPosition();
CSeqMap::ESegmentType seg_type = seqmap_ci.GetType();
switch(seg_type) {
case CSeqMap::eSeqGap:
_ASSERT(add_flags & fAddFlag_IncludeSeqGaps);
AddGap(
eGapType_SeqGap, pSeqId,
seqmap_ci.GetLength(),
bioseq_len,
seqmap_ci.GetPosition(), seqmap_ci.GetEndPosition(),
fFlags);
break;
case CSeqMap::eSeqData:
_ASSERT(add_flags & fAddFlag_IncludeUnknownBases);
x_AddGapsFromBases(
seqmap_ci, pSeqId,
bioseq_len, fFlags);
break;
default:
NCBI_USER_THROW_FMT(
"This segment type is not supported at this time: " <<
static_cast<int>(seg_type) );
}
}
if( end_of_last_segment != bioseq_len ) {
all_segments_and_in_order = false;
}
if( ! all_segments_and_in_order ) {
ERR_POST(
Warning << "Not all segments on bioseq '"
<< pSeqId->AsFastaString() << "' were in order "
"or some positions appear to have been skipped. "
"One possible reason is that there were far references for "
"which no attempt was made to resolve due to max resolve count "
"being reached.");
}
}
// задание списка изломов
void key_SetGaps(void)
{
if (bKey == bKEY_ENTER)
{
if (enKeyboard == KBD_ENTER)
{
if (boGapsFlag == false)
{ BlockProgram2(wSET_GAPS_FLAG, 1); return; }
else
{
enKeyboard = KBD_POSTINPUT1; // начинаем просмотр списка изломов
ShowHi(szGaps);
}
}
else if (enKeyboard == KBD_POSTINPUT2)
{
tiT.bDay = GetCharLo(0,1);
if ((tiT.bDay > 0) && (tiT.bDay <= 31))
{
enKeyboard = KBD_INPUT3;
szLo[2] = '.';
}
else Beep();
}
else if (enKeyboard == KBD_POSTINPUT3)
{
tiT.bMonth = GetCharLo(3,4);
tiT.bYear = 0;
if ((tiT.bMonth == 0) || (tiT.bMonth > 12))
Beep();
else if (tiT.bDay > GetDaysInMonthM(tiT.bMonth))
{
enKeyboard = KBD_INPUT2;
LongBeep();
ShowLo(szMaskGap);
}
else
{
enKeyboard = KBD_INPUT4;
tiT.bSecond = 0; // тип излома (режим работы)
ShowGapName(tiT.bSecond);
szLo[6] = '.';
}
}
else if (enKeyboard == KBD_INPUT4)
{
enKeyboard = KBD_POSTENTER; // переходим в режим ввода
if (AddGap() == 1) // добавляем новый излома в список
{
if (++ibX == bGAPS) // если список заполнен
{
ibX = 0; // переходим на начало
enKeyboard = KBD_SHOW; // переходим в режим просмотра
}
}
else Beep(); // такой излом есть в списке
}
else if (enKeyboard == KBD_SHOW) // режим проcмотра
{
ibX++; // переходим на следующий излом
if (ibX == GetGapSize()) // дошли до конца списка
{
if (GetGapSize() == bGAPS)
ibX = 0; // если список заполнен - продолжаем просмотр с начала
else
enKeyboard = KBD_POSTENTER; // если в списке есть свободное место - переходим в режим ввода
}
}
if ((enKeyboard == KBD_POSTINPUT1) || (enKeyboard == KBD_INPUT2))
{ // начинаем или продолжаем просмотр списка изломов
ibX = 0; // переходим на начало
if (GetGapSize() == 0)
enKeyboard = KBD_POSTENTER; // если список пустой - переходим в режим ввода
else
enKeyboard = KBD_SHOW; // если список не пустой - переходим в режим просмотра
}
if (enKeyboard == KBD_POSTENTER) // режим ввода
{
enKeyboard = KBD_INPUT2; // подготовка к вводу излома
ShowLo(szMaskGap);
}
else if (enKeyboard == KBD_SHOW) // режим просмотра
ShowGap(ibX); // просмотр текущего излома
}
else if (bKey < 10)
{
if ((enKeyboard == KBD_INPUT2) || (enKeyboard == KBD_POSTINPUT2))
{
enKeyboard = KBD_POSTINPUT2;
ShiftLo(0,1);
}
else
if ((enKeyboard == KBD_INPUT3) || (enKeyboard == KBD_POSTINPUT3))
{
enKeyboard = KBD_POSTINPUT3;
ShiftLo(3,4);
}
}
else if (bKey == bKEY_MINUS)
{
if (enKeyboard == KBD_SHOW) // в режиме просмотра
{
DelGap(); // удаляем излом из списка
LongBeep();
if (GetGapSize() == 0)
{
enKeyboard = KBD_INPUT2;
ShowHi(szGaps);
ShowLo(szMaskGap);
}
else ShowGap(ibX);
}
else Beep();
}
else if (bKey == bKEY_POINT)
{
if (enKeyboard == KBD_POSTINPUT2)
{
tiT.bDay = GetCharLo(0,1);
if ((tiT.bDay > 0) && (tiT.bDay <= 31))
{
enKeyboard = KBD_INPUT3;
szLo[2] = '.';
}
else Beep();
}
else if (enKeyboard == KBD_INPUT4)
{
if (++tiT.bSecond >= 12) tiT.bSecond = 0;
ShowGapName(tiT.bSecond);
szLo[6] = '.';
}
else Beep();
}
else Beep();
}
void CMetfile::Read()
{
ClearGaplist();
unsigned char * buffer = new unsigned char[512];
long length = 0;
length = mf_metfile -> Length();
tagcount = 0;
if (1 == mf_metfile -> Read( & version, 1))
{
if (debug)
{
printf("Version=0x%02x: Length=%lu\n", version, length);
}
isnewstyle = false;
switch (version)
{
case 0xe0:
case 0xe1:
isnewstyle = (version == 0xe1);
parts = 0;
if (!isnewstyle)
{
if (debug)
{
printf("...is not newstyle\n");
}
mf_metfile -> Seek(24);
if (4 == mf_metfile -> Read(buffer, 4))
{
if ((buffer[0] == 0) && (buffer[1] == 0) && (buffer[2] == 2) && (buffer[3] == 1))
{
isnewstyle = true;
}
mf_metfile -> Seek(1);
if (4 == mf_metfile -> Read( & date, 4))
{
if (16 == mf_metfile -> Read(checkid, 16))
{
for (int i = 0 ; i < 16 ; i++)
{
sprintf(mf_filehash + i * 2, "%02x", checkid[i]);
}
if (2 == mf_metfile -> Read( & parts, 2))
{
if (parts_hashtable)
{
free(parts_hashtable);
parts_hashtable = NULL;
}
if (parts)
{
int blocklength, position = 0, bytes = 0;
blocklength = 16 * parts;
parts_hashtable = (unsigned char *) malloc(blocklength);
while (position < blocklength)
{
bytes = mf_metfile -> Read(parts_hashtable + position, 16);
if (!bytes)
{
blocklength = - blocklength;
}
else
{
position += bytes;
}
}
if (blocklength <= 0)
{
blocklength = - blocklength;
if (debug)
{
printf("\nerror: %d(%d)%d ", blocklength, position, bytes);
}
}
}
}
else
{
tagcount = - 4;
}
}
else
{
tagcount = - 3;
}
}
else
{
tagcount = - 2;
}
}
else
{
tagcount = - 1;
}
}
else
{
if (debug)
{
printf("...is newstyle...");
}
if (4 == mf_metfile -> Read(buffer, 4))
{
if (debug)
{
printf("temp-ok %02x,%02x,%02x,%02x\n", buffer[0], buffer[1], buffer[2], buffer[3]);
}
if (! (buffer[0] | buffer[1] | buffer[2] | buffer[3]))
{
if (16 == mf_metfile -> Read(checkid, 16))
{
if (2 == mf_metfile -> Read( & parts, 2))
{
if (parts_hashtable)
{
free(parts_hashtable);
parts_hashtable = NULL;
}
if (parts)
{
parts_hashtable = (unsigned char *) malloc(16 * parts);
int i = 0;
while (i < parts)
{
if (16 == mf_metfile -> Read( & hash_current, 16))
{
memcpy(parts_hashtable + i * 16, & hash_current, 16);
i++;
}
else
{
i = parts;
tagcount = - 13;
}
}
}
}
else
{
parts = 0;
tagcount = - 12;
}
}
else
{
tagcount = - 11;
}
}
else
{
mf_metfile -> Seek(2);
if (4 == mf_metfile -> Read( & date, 4))
{
if (16 == mf_metfile -> Read(checkid, 16))
{
for (int i = 0 ; i < 16 ; i++)
{
sprintf(mf_filehash + i * 2, "%02x", checkid[i]);
}
}
else
{
tagcount = - 23;
}
}
else
{
tagcount = - 22;
}
}
}
else
{
if (debug)
{
printf("temp-error(-21)\n");
}
tagcount = - 21;
}
}
if (!tagcount)
{
if (4 == mf_metfile -> Read( & tagcount, 4))
{
int gapindex = 0, gapstatus = 0, gapstart = 0, gapend = 0, i = 0;
while (i < tagcount)
{
if (debug)
{
printf("ctag(%dv%d): ", 1 + i, tagcount);
}
CMetfileTag * CTag = new CMetfileTag(mf_metfile);
if (!CTag)
{
printf("error\n");
}
if (debug)
{
printf(" T=%d L=%d SpT=0x%02x StL=%d ",
CTag -> type, CTag -> length, CTag -> specialtagtype, CTag -> stringlength);
}
switch (CTag -> type)
{
case 2:
case 3:
switch (CTag -> specialtagtype)
{
case 0:
switch (CTag -> name[0])
{
case 0x09:
gapstart = CTag -> value;
gapstatus |= 1;
gapindex = atoi((char *) CTag -> name + 1);
break;
case 0x0a:
gapstatus |= 2;
gapend = CTag -> value;
if (gapindex != atoi((char *) CTag -> name + 1))
{
printf("gaperror: startindex(%d) != endindex(%d)\n", gapindex, atoi((char *) CTag -> name + 1));
}
else if(gapstatus == 3)
{
AddGap(gapstart, gapend);
}
else
{
printf("gaperror: gapstatus(%d)!=3\n", gapstatus);
}
gapstatus = 0;
break;
case 0xfe:
//ed2k-last-seen-complete
time_t timet0;
timet0 = (time_t) CTag -> value;
if (debug)
{
printf(" ct=%s ", ctime( & timet0));
}
break;
default:
gapstatus = 0;
break;
}
break;
case 1:
gapstatus = 0;
//filename
if (mf_filename == NULL)
{
mf_filename = (char *) malloc(strlen((char *) CTag -> string) + 1);
strcpy(mf_filename, (char *) CTag -> string);
}
break;
case 2:
gapstatus = 0;
//filesize
mf_filesize = CTag -> value;
break;
case 0x12:
gapstatus = 0;
//partfile
if (mf_partfile == NULL)
{
mf_partfile = (char *) malloc(strlen((char *) CTag -> string) + 1);
strcpy(mf_partfile, (char *) CTag -> string);
}
break;
default:
gapstatus = 0;
break;
}
if (debug)
{
printf("\n");
}
break;
default:
gapstatus = 0;
if (debug)
{
printf("\n");
}
break;
}
delete CTag;
i++;
}
}
else
{
tagcount = - 31;
}
}
if (debug)
{
printf("Date=%s", asctime(localtime((time_t *) & date)));
printf(" Parts=%d tagcount=%d\n", parts, tagcount);
}
break;
default:
break;
}
}
delete[] buffer;
SortGaplist();
}