This is BBPdbFilter.m in view mode; [Download] [Up]
/* BBPdbFilter.m
*
* This format filter does handle read/write of molecules in the
* Protein Data Bank (.pdb or .PDB) fileformat.
*
* For interface-info see the header file. The comments in this file mostly
* cover only the real implementation details.
*
* Written by: Thomas Engel
* Created: 01.11.1994 (Copyleft)
* Last modified: 01.11.1994
*/
#import "BBPdbFilter.h"
#import "BBFileFilterManager.h"
#import "../BBAppManager.h"
#import "../AtomLibrary.subproj/BBAtomLibraryManager.h"
#import "../BBBeaker.h"
#import "../BBMolecule.h"
#import "../BBAtom.h"
#import "../BBBond.h"
#import <misckit/MiscString.h>
#define MYBUFFERSIZE 500
#define SINGLEBONDTOKEN "E"
#define SINGLEFIXEDBONDTOKEN "Ef"
#define DOUBLEBONDTOKEN "D"
#define TRIPLEBONDTOKEN "T"
#define DOUBLEDELOCATEDBONDTOKEN "Dd"
// I hate statics but here they might be useful.
// The other variables have been added to the intercase to allow debugging.
static char lineBuffer[MYBUFFERSIZE];
static FILE *inStream;
@implementation BBPdbFilter
- (float)canReadFile:(const char*)filename
{
// Let's see if we can handle the file. Well this is a stupid check.
// Right now we only check the file extention. If its "pdb" or "PDB" we
// will do it.
// Normally we should return (0, 0.5, 1) according to the percentage of
// data we understand.
id pathString;
id aString;
float answer;
pathString = [MiscString new];
[pathString setPath:filename];
aString = [pathString fileExtension];
if( [aString casecmp:"pdb"] == 0 )
answer = (float)YES;
else answer = (float)NO;
[pathString free];
[aString free];
return answer;
}
- readBeaker:aBeaker fromFile:(const char*)filename
{
// A simple parser for Protein Data Bank Molecules.
// The tokenList is only defined as long as its data is not allready
// parsed or it contains the parts that have to be parsed. Freeing
// the list is not enough! We have to reset the id to nil to make the
// check work properly.
BOOL goAhead = YES;
inStream = fopen( filename, "r" );
if( inStream == NULL ) return self;
// So lets parse the data get one line and do it. We will
// not take care of all the settings! We'll say that the current line
// is already parsed because we wan another one to get read.
mol = [BBMolecule new];
theLine = [MiscString new];
tokenList = nil;
do
{
// Read a new line if no already done. Let the MiscString create the
// tokenList. We will only free it.
if( !tokenList )
{
if( !fgets( lineBuffer, MYBUFFERSIZE, inStream )) break;
[theLine setStringValue:lineBuffer];
tokenList = [theLine tokenize:" \t\n\r" into:nil];
}
// Now lets read till we have to stop.
goAhead = [self readNewLine];
}
while( goAhead );
[aBeaker addMolecule:mol];
[theLine free];
fclose( inStream );
return self;
}
- (BOOL)readNewLine
{
// Lets check the tokens..skip empty lines and comments
// theLine and tokenList are already set acording to our lineBuffer
// If there is nothing to parse we want to see the next line..
BOOL goAhead = YES;
id token;
id aString;
token = [tokenList objectAt:0];
if( [tokenList count] == 0 )
{
[[tokenList freeObjects] free];
tokenList = nil;
}
else if( [token compareStr:"COMPND"] == 0 )
{
aString = [theLine substringFrom:11 to:70];
[aString trimTailWhiteSpaces];
[(BBMolecule *)mol setName:[aString stringValue]];
[aString free];
[[tokenList freeObjects] free];
tokenList = nil;
}
else if( [token compareStr:"ATOM"] == 0 )
{
[[tokenList freeObjects] free];
tokenList = nil;
goAhead = [self readAtomData];
}
else if( [token compareStr:"HETATM"] == 0 )
{
[[tokenList freeObjects] free];
tokenList = nil;
goAhead = [self readAtomData];
}
else if( [token compareStr:"CONECT"] == 0 )
{
[[tokenList freeObjects] free];
tokenList = nil;
// goAhead = [self readBondData];
}
// Currently we do not evaluate every possible token.
// We will simple ingore everything else and just free the tokenList
// if this has not already happened.
else
{
[[tokenList freeObjects] free];
tokenList = nil;
}
return goAhead;
}
- (BOOL)readAtomData
{
// Reading the atom data
// Right here we will continue to read lines until a new token comes.
// As a token we will take everything that does not match a atom definition
id anAtom;
id basicAtom;
id atomLibrary;
id string;
id symbol;
atomLibrary = [[NXApp delegate] atomLibrary];
// Well now let's check the single details of the current theLine...
// the ATOM entries..
symbol = [theLine substringFrom:13 to:16];
[symbol trimTailWhiteSpaces];
string = [MiscString new];
[self checkSymbol:symbol andUpdate:string];
// We have to find the basicAtom too.
anAtom = [BBAtom new];
basicAtom = [atomLibrary findBasicAtomWithSymbol:[symbol stringValue]];
[anAtom setMotherAtom:basicAtom];
[anAtom setSymbolExtention:[string stringValue]];
[string free];
[symbol free];
string = [theLine substringFrom:31 to:38];
[string trimTailWhiteSpaces];
[anAtom setXPos:[string doubleValue]];
[string free];
string = [theLine substringFrom:39 to:46];
[string trimTailWhiteSpaces];
[anAtom setYPos:[string doubleValue]];
[string free];
string = [theLine substringFrom:47 to:54];
[string trimTailWhiteSpaces];
[anAtom setZPos:[string doubleValue]];
[string free];
[mol addAtom:anAtom];
return YES;
}
- (BOOL)readBondData
{
// Reading the bond data.
// Right here we will continue to read lines until a new token comes.
// As a token we will take everything that does not match a bond definition
BOOL goAhead = YES;
BBBond * aBond;
id atomList;
id fromAtom;
id toAtom;
id token;
char type;
atomList = [mol atomList];
// Check for comments, blank lines and new tokens.
// If there are not enough elements it has to be a token.
// Well kind of stupid but it should work for every LOOK mol
// otherwise lets init a new bond from this line:
// We will set the name to some usefull value.
aBond = [BBBond new];
if( [token compareStr:SINGLEBONDTOKEN] == 0 )
type = BOND_SINGLE;
else if( [token compareStr:SINGLEFIXEDBONDTOKEN] == 0 )
type = BOND_SINGLE;
else if( [token compareStr:DOUBLEBONDTOKEN] == 0 )
type = BOND_DOUBLE;
else if( [token compareStr:TRIPLEBONDTOKEN] == 0 )
type = BOND_TRIPLE;
else if( [token compareStr:DOUBLEDELOCATEDBONDTOKEN] == 0 )
type = BOND_DOUBLE_DELOCATED;
else
type = BOND_SINGLE;
fromAtom = [atomList objectAt:[[tokenList objectAt:1] intValue]];
toAtom = [atomList objectAt:[[tokenList objectAt:2] intValue]];
[aBond connect:fromAtom to:toAtom with:type];
// We will use the token string object to compose our Bond name.
[token setStringValue:[fromAtom symbol]];
[token catStringValue:"±"];
[token catStringValue:[toAtom symbol]];
[aBond setName:[token stringValue]];
[mol addBond:aBond];
// Lets free the list and make the line as read. We don't need it
// anymore
[[tokenList freeObjects] free];
tokenList = nil;
return YES;
}
- checkSymbol:symbol andUpdate:symbolExtention
{
// PDB symbol contain counts and special symbols at their end.
// We will move them to the symbolExtentions.
int i;
id aString;
for( i=0; i<[symbol length]; i++ )
if( [symbol charAt:i] < 'A' ||
[symbol charAt:i] > 'z' ||
([symbol charAt:i] < 'a' && i > 0 ))
break;
aString = [symbol substringFrom:i to:[symbol length]-1];
[symbolExtention setStringValue:[aString stringValue]];
[aString free];
aString = [symbol substringFrom:0 to:i-1];
[symbol setStringValue:[aString stringValue]];
[aString free];
return self;
}
- writeBeaker:aBeaker toFile:(const char *)filename
{
return self;
}
@end
/*
* History: 01.11.94 Created this object to parse ATOm and BOND info only.
*
*
* Bugs: - many PDB marks not supported
*/These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Netfuture.ch.