gistfile1.diff

diff --git a/Bio/GenBank/Scanner.py b/Bio/GenBank/Scanner.py
index cc72e7f..dd9d7f0 100644
--- a/Bio/GenBank/Scanner.py
+++ b/Bio/GenBank/Scanner.py
@@ -1183,7 +1183,69 @@ class GenBankScanner(InsdcScanner):
             return
         except StopIteration:
             raise ValueError("Problem in misc lines before sequence")
+
+class ImgtScanner(EmblScanner):
+    """For extracting chunks of information in IMGT/LIGM-DB flatfiles"""
+
+    # RECORD_START = "ID   "
+    # HEADER_WIDTH = 5
+    FEATURE_START_MARKERS = ["FH   Key             Location/Qualifiers",
+                             "FH   Key             Location/Qualifiers (from EMBL)",
+                             "FH   Key                 Location/Qualifiers",
+                             "FH"]
+    FEATURE_END_MARKERS = ["XX"] #XX can also mark the end of many things!
+    FEATURE_QUALIFIER_INDENTS = {"FH   Key             Location/Qualifiers":21,
+                                 "FH   Key             Location/Qualifiers (from EMBL)":21,
+                                 "FH   Key                 Location/Qualifiers":25}
+    FEATURE_QUALIFIER_SPACERS = {"FH   Key             Location/Qualifiers":
+                                                        "FT                   ",
+                                 "FH   Key             Location/Qualifiers (from EMBL)":
+                                                        "FT                   ",
+                                 "FH   Key                 Location/Qualifiers":
+                                                        "FT                       "}
+    # FEATURE_QUALIFIER_INDENTS = [21,25]
+    # FEATURE_QUALIFIER_SPACERS = [("FT" + " " * (f-2),f) for f in FEATURE_QUALIFIER_INDENTS]
+    FEATURE_QUALIFIER_INDENT = 0    # set dynamically
+    FEATURE_QUALIFIER_SPACER = ""   # set dynamically
+    # SEQUENCE_HEADERS=["SQ", "CO"] #Remove trailing spaces
+
+    def parse_header(self):
+        """Return list of strings making up the header
+
+        New line characters are removed.
+
+        Assumes you have just read in the ID/LOCUS line.
+        """
+        assert self.line[:self.HEADER_WIDTH]==self.RECORD_START, \
+               "Not at start of record"
         
+        header_lines = []
+        while True:
+            line = self.handle.readline()
+            if not line:
+                raise ValueError("Premature end of line during sequence data")
+            line = line.rstrip()
+            if line in self.FEATURE_START_MARKERS:
+                if self.debug : print "Found header table"
+                # NOTE: if first FH found has no info on indentation, we can't
+                #       define the qualifier indent/spacer
+                if line != 'FH':
+                    # determine if the IMGT record indentation is 21 or 25 chars
+                    self.FEATURE_QUALIFIER_INDENT = self.FEATURE_QUALIFIER_INDENTS[line]
+                    self.FEATURE_QUALIFIER_SPACER = self.FEATURE_QUALIFIER_SPACERS[line]
+                break
+            #if line[:self.HEADER_WIDTH]==self.FEATURE_START_MARKER[:self.HEADER_WIDTH]:
+            #    if self.debug : print "Found header table (?)"
+            #    break
+            if line[:self.HEADER_WIDTH].rstrip() in self.SEQUENCE_HEADERS:
+                if self.debug : print "Found start of sequence"
+                break
+            if line == "//":
+                raise ValueError("Premature end of sequence data marker '//' found")
+            header_lines.append(line)
+        self.line = line
+        return header_lines
+
 if __name__ == "__main__":
     from StringIO import StringIO
 
diff --git a/Bio/SeqIO/InsdcIO.py b/Bio/SeqIO/InsdcIO.py
index 752ed57..c10d74d 100644
--- a/Bio/SeqIO/InsdcIO.py
+++ b/Bio/SeqIO/InsdcIO.py
@@ -26,7 +26,7 @@ http://www.ddbj.nig.ac.jp/
 """
 
 from Bio.Seq import UnknownSeq
-from Bio.GenBank.Scanner import GenBankScanner, EmblScanner
+from Bio.GenBank.Scanner import GenBankScanner, EmblScanner, ImgtScanner
 from Bio import Alphabet
 from Interfaces import SequentialSequenceWriter
 from Bio import SeqFeature
@@ -61,6 +61,17 @@ def EmblIterator(handle):
     #This calls a generator function:
     return EmblScanner(debug=0).parse_records(handle)
 
+def ImgtIterator(handle):
+    """Breaks up an IMGT/LIGM-DB flatfile into SeqRecord objects.
+
+    Every section from the LOCUS line to the terminating // becomes
+    a single SeqRecord with associated annotation and features.
+    
+    Note that for genomes or chromosomes, there is typically only
+    one record."""
+    #This calls a generator function:
+    return ImgtScanner(debug=0).parse_records(handle)
+
 def GenBankCdsFeatureIterator(handle, alphabet=Alphabet.generic_protein):
     """Breaks up a Genbank file into SeqRecord objects for each CDS feature.
 
diff --git a/Bio/SeqIO/__init__.py b/Bio/SeqIO/__init__.py
index 03d281a..cb4e91f 100644
--- a/Bio/SeqIO/__init__.py
+++ b/Bio/SeqIO/__init__.py
@@ -325,6 +325,7 @@ _FormatToIterator = {"fasta" : FastaIO.FastaIterator,
                      "embl" : InsdcIO.EmblIterator,
                      "embl-cds" : InsdcIO.EmblCdsFeatureIterator,
                      "ig" : IgIO.IgIterator,
+                     "imgt": InsdcIO.ImgtIterator,
                      "swiss" : SwissIO.SwissIterator,
                      "phd" : PhdIO.PhdIterator,
                      "ace" : AceIO.AceIterator,
diff --git a/Bio/SeqIO/_index.py b/Bio/SeqIO/_index.py
index 0901e61..e15493c 100644
--- a/Bio/SeqIO/_index.py
+++ b/Bio/SeqIO/_index.py
@@ -401,6 +401,23 @@ class EmblDict(_SequentialSeqFileDict):
                         break
                 self._record_key(key, offset)
 
+class ImgtDict(_SequentialSeqFileDict):
+    """Indexed dictionary like access to an IMGT/LIGM-DB flatfile."""
+    def __init__(self, filename, alphabet, key_function):
+        _IndexedSeqFileDict.__init__(self, filename, alphabet, key_function)
+        self._format = "imgt"
+        handle = self._handle
+        marker_re = re.compile("^ID ")
+        self._marker_re = marker_re #saved for the get_raw method
+        while True:
+            offset = handle.tell()
+            line = handle.readline()
+            if not line : break #End of file
+            if marker_re.match(line):
+                # We assume the first word after ID is the accession
+                key = line.split()[1]
+                self._record_key(key, offset)
+
 class SwissDict(_SequentialSeqFileDict):
     """Indexed dictionary like access to a SwissProt file."""
     def __init__(self, filename, alphabet, key_function):
@@ -594,6 +611,7 @@ _FormatToIndexedDict = {"ace" : AceDict,
                         "genbank" : GenBankDict,
                         "gb" : GenBankDict, #alias of the above
                         "ig" : IntelliGeneticsDict,
+                        "imgt" : ImgtDict,
                         "phd" : PhdDict,
                         "pir" : PirDict,
                         "sff" : SffDict,