From c77c3607828dd2987f107e603f8618f1a51eb81f Mon Sep 17 00:00:00 2001
From: Eric Coissac <eric.coissac@metabarcoding.org>
Date: Wed, 20 Jun 2007 07:39:40 +0000
Subject: [PATCH] git-svn-id:
 https://www.grenoble.prabi.fr/svn/LECASofts/ecoPCR/branches/refactoring@76
 60f365c0-8329-0410-b2a4-ec073aeeaa1d

---
 tools/ecoPCRFormat.py | 63 +++++++++++++++++++++++++++++++++++--------
 1 file changed, 52 insertions(+), 11 deletions(-)

diff --git a/tools/ecoPCRFormat.py b/tools/ecoPCRFormat.py
index 9a3205d..0f237ae 100755
--- a/tools/ecoPCRFormat.py
+++ b/tools/ecoPCRFormat.py
@@ -171,15 +171,14 @@ def readNodeTable(file):
         
     return taxonomy,ranks,index
 
-def scientificNameIterator(file):
+def nameIterator(file):
     file = universalOpen(file)
     names = ColumnFile(file, 
                        sep='|', 
                        types=(int,str,
                               str,str))
     for taxid,name,unique,classname,white in names:
-        if classname == 'scientific name':
-            yield taxid,name
+        yield taxid,name,classname
         
 def mergedNodeIterator(file):
     file = universalOpen(file)
@@ -201,8 +200,12 @@ def readTaxonomyDump(taxdir):
     taxonomy,ranks,index = readNodeTable('%s/nodes.dmp' % taxdir)
     
     print >>sys.stderr,"Adding scientific name..."
-    for taxid,name in scientificNameIterator('%s/names.dmp' % taxdir):
-        taxonomy[index[taxid]].append(name)
+
+    alternativeName=[]
+    for taxid,name,classname in nameIterator('%s/names.dmp' % taxdir):
+        alternativeName.append((name,classname,index[taxid]))
+        if classname == 'scientific name':
+            taxonomy[index[taxid]].append(name)
         
     print >>sys.stderr,"Adding taxid alias..."
     for taxid,current in mergedNodeIterator('%s/merged.dmp' % taxdir):
@@ -212,7 +215,7 @@ def readTaxonomyDump(taxdir):
     for taxid in deletedNodeIterator('%s/delnodes.dmp' % taxdir):
         index[taxid]=None
     
-    return taxonomy,ranks,index
+    return taxonomy,ranks,alternativeName,index
 
 
 #####
@@ -405,6 +408,22 @@ def ecoRankPacker(rank):
     
     return packed
                 
+def ecoNamePacker(name):
+    
+    namelength = len(name[0])
+    classlength= len(name[1])
+    totalSize =  namelength + classlength + 4 + 4 + 4 + 4
+    
+    packed = struct.pack('> I I I I I %ds %ds' % (namelength,classlength),
+                         totalSize,
+                         int(name[1]=='scientific name'),
+                         namelength,
+                         classlength,
+                         name[2],
+                         name[0],
+                         name[1])
+    
+    return packed
     
 def ecoSeqWriter(file,input,taxindex,parser):
     output = open(file,'wb')
@@ -462,18 +481,40 @@ def ecoRankWriter(file,ranks):
         output.write(ecoRankPacker(rank))
 
     output.close()
+
+def nameCmp(n1,n2):
+    name1=n1[0].upper()
+    name2=n2[0].upper()
+    if name1 < name2:
+        return -1
+    elif name1 > name2:
+        return 1
+    return 0
+
+
+def ecoNameWriter(file,names):
+    output = open(file,'wb')
+    output.write(struct.pack('> I',len(names)))
+
+    names.sort(nameCmp)
+    
+    for name in names:
+        output.write(ecoNamePacker(name))
+
+    output.close()
     
 def ecoDBWriter(prefix,taxonomy,seqFileNames,parser):
     
     ecoRankWriter('%s.rdx' % prefix, taxonomy[1])
     ecoTaxWriter('%s.tdx' % prefix, taxonomy[0])
-    
+    ecoNameWriter('%s.ndx' % prefix, taxonomy[2])
+  
     filecount = 0
     for filename in seqFileNames:
         filecount+=1
         sk=ecoSeqWriter('%s_%03d.sdx' % (prefix,filecount), 
                      filename, 
-                     taxonomy[2], 
+                     taxonomy[3], 
                      parser)
         if sk:
             print >>sys.stderr,"Skipped entry :"
@@ -526,9 +567,9 @@ def printHelp():
     print "-----------------------------------"
     print "ecoPCRFormat.py [option] <argument>"
     print "-----------------------------------"
-    print "-e    --embl        :[E]mbl format file name"
-    print "-f    --fasta       :[F]asta format file name"
-    print "-g    --genbank     :[G]enbank format file name"
+    print "-e    --embl        :[E]mbl format"
+    print "-f    --fasta       :[F]asta format"
+    print "-g    --genbank     :[G]enbank format"
     print "-h    --help        :[H]elp - print this help"
     print "-n    --name        :[N]ame of the new database created"
     print "-t    --taxonomy    :[T]axonomy - path to the taxonomy database"