From bac7ce7184b9db310835f504e54b63242a6c22d3 Mon Sep 17 00:00:00 2001
From: Eric Coissac <eric.coissac@metabarcoding.org>
Date: Thu, 2 Jun 2016 19:10:33 +0200
Subject: [PATCH] Start of the implementation of the export methods

---
 python/obitools3/format/__init__.py |  0
 python/obitools3/format/fasta.pxd   |  9 +++++
 python/obitools3/format/fasta.pyx   | 52 +++++++++++++++++++++++++++++
 python/obitools3/format/header.pxd  |  5 +++
 python/obitools3/format/header.pyx  | 27 +++++++++++++++
 5 files changed, 93 insertions(+)
 create mode 100644 python/obitools3/format/__init__.py
 create mode 100644 python/obitools3/format/fasta.pxd
 create mode 100644 python/obitools3/format/fasta.pyx
 create mode 100644 python/obitools3/format/header.pxd
 create mode 100644 python/obitools3/format/header.pyx

diff --git a/python/obitools3/format/__init__.py b/python/obitools3/format/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/python/obitools3/format/fasta.pxd b/python/obitools3/format/fasta.pxd
new file mode 100644
index 0000000..b2d2c2d
--- /dev/null
+++ b/python/obitools3/format/fasta.pxd
@@ -0,0 +1,9 @@
+from .header cimport HeaderFormat
+from cython.view cimport array as cvarray
+
+cdef class FastaFormat:
+
+    cdef HeaderFormat headerFormater
+    
+    cdef size_t  sequenceBufferLength
+    cdef cvarray sequenceBuffer
diff --git a/python/obitools3/format/fasta.pyx b/python/obitools3/format/fasta.pyx
new file mode 100644
index 0000000..2460e11
--- /dev/null
+++ b/python/obitools3/format/fasta.pyx
@@ -0,0 +1,52 @@
+cimport cython
+
+cdef class FastaFormat:
+	
+	def __init__(self, list tags=[], bint printNAKeys=False):
+		self.headerFormater = HeaderFormat(True,
+										   tags,
+										   printNAKeys)
+
+		self.sequenceBufferLength=1000
+		self.sequenceBuffer = cvarray(shape=(1000,),
+									  itemsize=sizeof(char), 
+									  format="c",
+									  mode="c",
+									  allocate_buffer=True)
+	@cython.boundscheck(False)	
+	def __call__(self, dict data):
+		cdef bytes brawseq      = <bytes> data['sequence']
+		cdef size_t lseq        = len(brawseq)
+		cdef size_t needed_size = lseq + 1
+		cdef char[:] seq
+		cdef char[:] fasta
+		cdef size_t k=0
+		needed_size += needed_size/ 50
+		
+		if needed_size > self.sequenceBufferLength:
+			self.sequenceBufferLength=needed_size
+			self.sequenceBuffer = cvarray(shape=(needed_size,),
+										  itemsize=sizeof(char), 
+										  format="c",
+										  mode="c",
+										  allocate_buffer=True)
+			
+		seq   = brawseq
+		fasta = self.sequenceBuffer
+		
+		for i in range(0,lseq,60):
+			if i+60 <= lseq:
+				fasta[k:(k+60)]=seq[i:(i+60)]
+				fasta[k+60]='\n'
+				k+=61
+			else:
+				fasta[k:(k+lseq-i)]=seq[i:lseq]
+				k+=lseq-i
+				
+		
+				
+			
+		
+		
+		
+		
\ No newline at end of file
diff --git a/python/obitools3/format/header.pxd b/python/obitools3/format/header.pxd
new file mode 100644
index 0000000..92539f2
--- /dev/null
+++ b/python/obitools3/format/header.pxd
@@ -0,0 +1,5 @@
+cdef class HeaderFormat:
+
+    cdef str    start
+    cdef size_t headerBufferLength
+    
\ No newline at end of file
diff --git a/python/obitools3/format/header.pyx b/python/obitools3/format/header.pyx
new file mode 100644
index 0000000..9431c3a
--- /dev/null
+++ b/python/obitools3/format/header.pyx
@@ -0,0 +1,27 @@
+
+cdef class HeaderFormat:
+    
+    def __init__(self, bint fastaHeader=True, list tags=[], bint printNAKeys=False):
+        
+        self.tags           = tags 
+        self.printNaKeys    = printNAKeys
+        
+        if fastaHeader:
+            self.start=">"
+        else:
+            self.start="@"
+        
+        self.headerBufferLength = 1000
+        self.headerBuffer = []
+        
+    def __call__(self, dict data):
+        cdef str header
+        
+        if data['definition'] is not None:
+            header = "%s%s %s" % (self.start,data['id'],
+                                  data['definition'])
+        else:
+            header = "%s%s" % (self.start,data['id'])
+            
+        return header
+