Rewrite of findall.py to clean code. Configuration tested using

[erange.git] / ReadDataset.py

diff --git a/ReadDataset.py b/ReadDataset.py
index 850a5ec602a7b6d7a9a22ac98c4bf3eef4b873e5..c9d2a0bf2b4eb56e0fa906be2319fcac866c30b3 100644 (file)
--- a/ReadDataset.py
+++ b/ReadDataset.py
@@ -207,15 +207,27 @@ class ReadDataset():
         return sql
 
 
+    def getMemCursor(self):
+        """ returns a cursor to memory database for low-level (SQL)
+        access to the data.
+        """
+        return self.memcon.cursor()
+
+
+    def getFileCursor(self):
+        """ returns a cursor to file database for low-level (SQL)
+        access to the data.
+        """
+        return self.dbcon.cursor()
+
+
     def hasIndex(self):
-        """ check whether the RDS file has at least one index.
+        """ return True if the RDS file has at least one index.
         """
         stmt = "select count(*) from sqlite_master where type='index'"
         count = int(self.execute(stmt, returnResults=True)[0][0])
-        if count > 0:
-            return True
 
-        return False
+        return count > 0
 
 
     def initializeTables(self, dbConnection, cache=100000):
@@ -237,20 +249,6 @@ class ReadDataset():
         dbConnection.commit()
 
 
-    def getFileCursor(self):
-        """ returns a cursor to file database for low-level (SQL)
-        access to the data.
-        """
-        return self.dbcon.cursor()
-
-
-    def getMemCursor(self):
-        """ returns a cursor to memory database for low-level (SQL)
-        access to the data.
-        """
-        return self.memcon.cursor()
-
-
     def getMetadata(self, valueName=""):
         """ returns a dictionary of metadata.
         """
@@ -309,7 +307,7 @@ class ReadDataset():
 
 
     def getChromosomes(self, table="uniqs", fullChrom=True):
-        """ returns a list of distinct chromosomes in table.
+        """ returns a sorted list of distinct chromosomes in table.
         """
         statement = "select distinct chrom from %s" % table
         sql = self.getSqlCursor()
@@ -330,7 +328,7 @@ class ReadDataset():
         return results
 
 
-    def getMaxCoordinate(self, chrom, verbose=False, doUniqs=True,
+    def getMaxCoordinate(self, chrom, doUniqs=True,
                          doMulti=False, doSplices=False):
         """ returns the maximum coordinate for reads on a given chromosome.
         """
@@ -347,9 +345,6 @@ class ReadDataset():
             multiMax = self.getMaxStartCoordinateInTable(chrom, "multi")
             maxCoord = max(multiMax, maxCoord)
 
-        if verbose:
-            print "%s maxCoord: %d" % (chrom, maxCoord)
-
         return maxCoord
 
 
@@ -375,9 +370,9 @@ class ReadDataset():
         and which can be restricted by chromosome or custom-flag.
         Returns unique reads by default, but can return multireads
         with doMulti set to True.
-        
-        Need to rethink original design 1: Cannot have pairID without exporting as a readIDDict
+
         """
+        #TODO: Need to rethink original design 1: Cannot have pairID without exporting as a readIDDict
 
         whereQuery = self.getReadWhereQuery(chrom, flag, flagLike, start, stop, hasMismatch, strand, readLike)
         if findallOptimize:
@@ -421,27 +416,16 @@ class ReadDataset():
         if findallOptimize:
             if self.memBacked:
                 self.memcon.row_factory = None
-                sql = self.memcon.cursor()
             else:
                 self.dbcon.row_factory = None
-                sql = self.dbcon.cursor()
 
             stmt.append("order by start")
         elif readIDDict:
-            if self.memBacked:
-                sql = self.memcon.cursor()
-            else:
-                sql = self.dbcon.cursor()
-
             stmt.append("order by readID, start")
         else:
-            if self.memBacked:
-                sql = self.memcon.cursor()
-            else:
-                sql = self.dbcon.cursor()
-
             stmt.append("order by chrom, start")
 
+        sql = self.getSqlCursor()
         sqlQuery = string.join(stmt)
         sql.execute(sqlQuery)
 
@@ -602,10 +586,7 @@ class ReadDataset():
         whereQuery = self.getReadWhereQuery(chrom, flag, flagLike, start, stop, hasMismatch, strand, splice=True)
         selectClause = "select ID, chrom, startL, stopL, startR, stopR, readID"
         selectQuery = self.getReadSelectQuery(selectClause, noSense, withWeight, withFlag, withMismatch)
-        if self.memBacked:
-            sql = self.memcon.cursor()
-        else:
-            sql = self.dbcon.cursor()
+        sql = self.getSqlCursor()
 
         stmt = "%s from splices %s order by chrom, startL" % (selectQuery, whereQuery)
         sql.execute(stmt)
@@ -718,7 +699,7 @@ class ReadDataset():
 
 
     def getTableEntryCount(self, table, chrom="", rmin="", rmax="", restrict="", distinct=False, startField="start"):
-        """ returns the number of row in the uniqs table.
+        """ returns the number of row in the specified table.
         """
         whereClause = []
         count = 0
@@ -741,10 +722,7 @@ class ReadDataset():
         else:
             whereQuery = ""
 
-        if self.memBacked:
-            sql = self.memcon.cursor()
-        else:
-            sql = self.dbcon.cursor()
+        sql = self.getSqlCursor()
 
         if distinct:
             sql.execute("select count(distinct chrom+%s+sense) from %s %s" % (startField, table, whereQuery))
@@ -803,11 +781,7 @@ class ReadDataset():
             limitPart = "LIMIT %d" % limit
 
         sqlQuery = "%s group by readID %s" % (selectPart, limitPart)
-        if self.memBacked:
-            sql = self.memcon.cursor()
-        else:
-            sql = self.dbcon.cursor()
-
+        sql = self.getSqlCursor()
         sql.execute(sqlQuery)
         result = sql.fetchall()