bio2.bam.header source code

1 /*
2     New style BAM reader. This file is part of Sambamba.
3     Copyright (C) 2017 Pjotr Prins <pjotr.prins@thebird.nl>
4 
5     Sambamba is free software; you can redistribute it and/or modify
6     it under the terms of the GNU General Public License as published
7     by the Free Software Foundation; either version 2 of the License,
8     or (at your option) any later version.
9 
10     Sambamba is distributed in the hope that it will be useful, but
11     WITHOUT ANY WARRANTY; without even the implied warranty of
12     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
13     General Public License for more details.
14 
15     You should have received a copy of the GNU General Public License
16     along with this program; if not, write to the Free Software
17     Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
18     02111-1307 USA
19 
20 */
21 
22 // This is a complete rewrite of Artem Tarasov's original reader.
23 
24 module bio2.bam.header;
25 
26 /*
27 import std.conv;
28 import core.stdc.stdio: fopen, fread, fclose;
29 import std.typecons;
30 import std.bitmanip;
31 
32 import bio.bam.cigar;
33 */
34 
35 import std.exception;
36 import std.file;
37 import std.stdio;
38 import std.string;
39 
40 import bio.bam.constants;
41 
42 import bio2.bgzf;
43 import bio2.bgzf_writer;
44 import bio2.constants;
45 
46 struct RefSequence {
47   size_d length;
48   string name;
49 }
50 
51 struct BamHeader {
52   string id;
53   string text;
54   RefSequence[] refs;
55 
56   @disable this(this); // disable copy semantics;
57 }
58 
59 void fetch_bam_header(ref BamHeader header, ref BgzfStream stream) {
60   // stderr.writeln("Fetching BAM header");
61   ubyte[4] ubyte4;
62   stream.read(ubyte4);
63   enforce(ubyte4 == BAM_MAGIC,"Invalid file format: expected BAM magic number");
64   immutable text_size = stream.read!int();
65   // stderr.writeln("Text size ",text_size.sizeof," ",text_size);
66   immutable text = stream.read!string(text_size);
67   header = BamHeader(BAM_MAGIC,text);
68   immutable n_refs = stream.read!int();
69   // stderr.writeln("Fetching ",n_refs," references");
70   foreach(int n_ref; 0..n_refs) {
71     immutable l_name = stream.read!int();
72     // stderr.writeln("!!",l_name);
73     auto ref_name = stream.read!string(l_name);
74     immutable l_ref = stream.read!int(); // length of reference sequence (bps)
75     // stderr.writeln(l_name," ",ref_name," ",l_ref);
76     header.refs ~= RefSequence(l_ref,ref_name[0..l_name-1]); // drop zero terminator
77   }
78 }
79 
80 void write_bam_header(ref BgzfWriter bw, ref BamHeader header) {
81   // stderr.writeln("Writing BAM header");
82   ubyte[4] magic = cast(ubyte[])BAM_MAGIC;
83   bw.write(magic);
84   // stderr.writeln("Text size ",int.sizeof," ",header.text.length);
85   bw.write!int(cast(int)header.text.length);
86   bw.write(header.text);
87   auto n_refs = cast(int)header.refs.length;
88   bw.write!int(cast(int)header.refs.length);
89   // stderr.writeln("Writing ",n_refs," references");
90   foreach(int n_ref; 0..n_refs) {
91     immutable refseq = header.refs[n_ref];
92     bw.write!int(cast(int)(refseq.name.length+1));  // incl. zero terminator
93     // stderr.writeln("!!",refseq.name.length+1);
94     bw.write(refseq.name);
95     bw.write!ubyte(cast(ubyte)'\0');
96     bw.write!int(cast(int)refseq.length);
97     // stderr.writeln(refseq.name.length+1," ",refseq.name," ",refseq.length);
98   }
99   // stderr.writeln("!!");
100   bw.flush_block();
101 }