def main(cls, args):
# db = SimpleDB("tblmgrtest", 400, 8)
fm = FileMgr(File("tblmgrtest"), 400)
lm = LogMgr(fm, "simpledb.log")
bm = BufferMgr(fm, lm, 8)
tx = Transaction(fm, lm, bm)
tm = TableMgr(True, tx)
sch = Schema()
sch.addIntField("A")
sch.addStringField("B", 9)
tm.createTable("MyTable", sch, tx)
layout = tm.getLayout("MyTable", tx)
size = layout.slotSize()
sch2 = layout.schema()
print("MyTable has slot size " + str(size))
print("Its fields are:")
for fldname in sch2.fields():
if sch2.type(fldname) == INTEGER:
_type = "int"
else:
strlen = sch2.length(fldname)
_type = "varchar(" + str(strlen) + ")"
print(fldname + ": " + _type)
tx.commit()
def main(cls, args):
sch = Schema()
sch.addIntField("A")
sch.addStringField("B", 9)
layout = Layout(sch)
for fldname in layout.schema().fields():
offset = layout.offset(fldname)
print(fldname + " has offset " + str(offset))
def fieldType(self, fldname):
schema = Schema()
if self.lex.matchKeyword("int"):
self.lex.eatKeyword("int")
schema.addIntField(fldname)
else:
self.lex.eatKeyword("varchar")
self.lex.eatDelim('(')
strLen = self.lex.eatIntConstant()
self.lex.eatDelim(')')
schema.addStringField(fldname, strLen)
return schema
def main(cls, args):
# db = SimpleDB("recordtest", 400, 8)
fm = FileMgr(File("recordtest"), 400)
lm = LogMgr(fm, "simpledb.log")
bm = BufferMgr(fm, lm, 8)
tx = Transaction(fm, lm, bm)
sch = Schema()
sch.addIntField("A")
sch.addStringField("B", 9)
layout = Layout(sch)
for fldname in layout.schema().fields():
offset = layout.offset(fldname)
print(fldname + " has offset " + str(offset))
blk = tx.append("testfile")
tx.pin(blk)
rp = RecordPage(tx, blk, layout)
rp.format()
print("Filling the page with random records.")
slot = rp.insertAfter(-1)
while slot >= 0:
n = random.randint(0, 50)
rp.setInt(slot, "A", n)
rp.setString(slot, "B", "rec" + str(n))
print("inserting into slot " + str(slot) + ": {" + str(n) + ", " + "rec" + str(n) + "}")
slot = rp.insertAfter(slot)
print("Deleting these records, whose A-values are less than 25.")
count = 0
slot = rp.nextAfter(-1)
while slot >= 0:
a = rp.getInt(slot, "A")
b = rp.getString(slot, "B")
if a < 25:
count += 1
print("slot " + str(slot) + ": {" + str(a) + ", " + str(b) + "}")
rp.delete(slot)
slot = rp.nextAfter(slot)
print(str(count) + " values under 25 were deleted.\n")
print("Here are the remaining records.")
slot = rp.nextAfter(-1)
while slot >= 0:
a = rp.getInt(slot, "A")
b = rp.getString(slot, "B")
print("slot " + str(slot) + ": {" + str(a) + ", " + str(b) + "}")
slot = rp.nextAfter(slot)
tx.unpin(blk)
tx.commit()
def main(cls, args):
# db = SimpleDB("scantest2")
fm = FileMgr(File("scantest2"), 400)
lm = LogMgr(fm, "simpledb.log")
bm = BufferMgr(fm, lm, 8)
tx = Transaction(fm, lm, bm)
sch1 = Schema()
sch1.addIntField("A")
sch1.addStringField("B", 9)
layout1 = Layout(sch1)
us1 = TableScan(tx, "T1", layout1)
us1.beforeFirst()
n = 200
print("Inserting " + str(n) + " records into T1.")
for i in range(n):
us1.insert()
us1.setInt("A", i)
us1.setString("B", "bbb" + str(i))
us1.close()
sch2 = Schema()
sch2.addIntField("C")
sch2.addStringField("D", 9)
layout2 = Layout(sch2)
us2 = TableScan(tx, "T2", layout2)
us2.beforeFirst()
print("Inserting " + str(n) + " records into T2.")
for i in range(n):
us2.insert()
us2.setInt("C", n - i - 1)
us2.setString("D", "ddd" + str((n - i - 1)))
us2.close()
s1 = TableScan(tx, "T1", layout1)
s2 = TableScan(tx, "T2", layout2)
s3 = ProductScan(s1, s2)
# selecting all records where A=C
t = Term(Expression("A"), Expression("C"))
pred = Predicate(t)
print("The predicate is " + pred.__str__())
s4 = SelectScan(s3, pred)
# projecting on [B,D]
c = ["B", "D"]
s5 = ProjectScan(s4, c)
while s5.next():
print(s5.getString("B") + " " + s5.getString("D"))
s5.close()
tx.commit()
def main(cls, args):
# db = SimpleDB("tabletest", 400, 8)
fm = FileMgr(File("tabletest"), 400)
lm = LogMgr(fm, "simpledb.log")
bm = BufferMgr(fm, lm, 8)
tx = Transaction(fm, lm, bm)
sch = Schema()
sch.addIntField("A")
sch.addStringField("B", 9)
layout = Layout(sch)
for fldname in layout.schema().fields():
offset = layout.offset(fldname)
print(fldname + " has offset " + str(offset))
print("Filling the table with 50 random records.")
ts = TableScan(tx, "T", layout)
for i in range(50):
ts.insert()
n = random.randint(0, 50)
ts.setInt("A", n)
ts.setString("B", "rec" + str(n))
print("inserting into slot " + ts.getRid().__str__() + ": {" +
str(n) + ", " + "rec" + str(n) + "}")
print("Deleting these records, whose A-values are less than 25.")
count = 0
ts.beforeFirst()
while ts.next():
a = ts.getInt("A")
b = ts.getString("B")
if a < 25:
count += 1
print("slot " + ts.getRid().__str__() + ": {" + str(a) + ", " +
b + "}")
ts.delete()
print(str(count) + " values under 10 were deleted.\n")
print("Here are the remaining records.")
ts.beforeFirst()
while ts.next():
a = ts.getInt("A")
b = ts.getString("B")
print("slot " + ts.getRid().__str__() + ": {" + str(a) + ", " + b +
"}")
ts.close()
tx.commit()
def main(cls, args):
# db = SimpleDB("producttest")
fm = FileMgr(File("producttest"), 400)
lm = LogMgr(fm, "simpledb.log")
bm = BufferMgr(fm, lm, 8)
tx = Transaction(fm, lm, bm)
sch1 = Schema()
sch1.addIntField("A")
sch1.addStringField("B", 9)
layout1 = Layout(sch1)
ts1 = TableScan(tx, "T1", layout1)
sch2 = Schema()
sch2.addIntField("C")
sch2.addStringField("D", 9)
layout2 = Layout(sch2)
ts2 = TableScan(tx, "T2", layout2)
ts1.beforeFirst()
n = 200
print("Inserting " + str(n) + " records into T1.")
for i in range(n):
ts1.insert()
ts1.setInt("A", i)
ts1.setString("B", "aaa" + str(i))
ts1.close()
ts2.beforeFirst()
print("Inserting " + str(n) + " records into T2.")
for i in range(n):
ts2.insert()
ts2.setInt("C", n - i - 1)
ts2.setString("D", "bbb" + str((n - i - 1)))
ts2.close()
s1 = TableScan(tx, "T1", layout1)
s2 = TableScan(tx, "T2", layout2)
s3 = ProductScan(s1, s2)
while s3.next():
print(s3.getString("B"))
s3.close()
tx.commit()
def main(cls, args):
# db = SimpleDB("scantest1")
fm = FileMgr(File("scantest1"), 400)
lm = LogMgr(fm, "simpledb.log")
bm = BufferMgr(fm, lm, 8)
tx = Transaction(fm, lm, bm)
sch1 = Schema()
sch1.addIntField("A")
sch1.addStringField("B", 9)
layout = Layout(sch1)
s1 = TableScan(tx, "T", layout)
s1.beforeFirst()
n = 200
print("Inserting " + str(n) + " random records.")
for i in range(n):
s1.insert()
k = random.randint(0, 50)
s1.setInt("A", k)
s1.setString("B", "rec" + str(k))
s1.close()
s2 = TableScan(tx, "T", layout)
# selecting all records where A=10
c = Constant(10)
t = Term(Expression("A"), Expression(c))
pred = Predicate(t)
print("The predicate is " + pred.__str__())
s3 = SelectScan(s2, pred)
fields = ["B"]
s4 = ProjectScan(s3, fields)
while s4.next():
print(s4.getString("B"))
s4.close()
tx.commit()
def createIdxLayout(self):
sch = Schema()
sch.addIntField("block")
sch.addIntField("id")
if self.tblSchema.type(self.fldname) == INTEGER:
sch.addIntField("dataval")
else:
fldlen = self.tblSchema.length(self.fldname)
sch.addStringField("dataval", fldlen)
return Layout(sch)
def __init__(self, isNew, tx):
tcatSchema = Schema()
tcatSchema.addStringField("tblname", TableMgr.MAX_NAME)
tcatSchema.addIntField("slotsize")
self.tcatLayout = Layout(tcatSchema)
fcatSchema = Schema()
fcatSchema.addStringField("tblname", TableMgr.MAX_NAME)
fcatSchema.addStringField("fldname", TableMgr.MAX_NAME)
fcatSchema.addIntField("type")
fcatSchema.addIntField("length")
fcatSchema.addIntField("offset")
self.fcatLayout = Layout(fcatSchema)
if isNew:
self.createTable("tblcat", tcatSchema, tx)
self.createTable("fldcat", fcatSchema, tx)
def main(cls, args):
# db = SimpleDB("metadatamgrtest", 400, 8)
fm = FileMgr(File("metadatamgrtest"), 400)
lm = LogMgr(fm, "simpledb.log")
bm = BufferMgr(fm, lm, 8)
tx = Transaction(fm, lm, bm)
mdm = MetadataMgr(True, tx)
sch = Schema()
sch.addIntField("A")
sch.addStringField("B", 9)
# Part 1: Table Metadata
mdm.createTable("MyTable", sch, tx)
layout = mdm.getLayout("MyTable", tx)
size = layout.slotSize()
sch2 = layout.schema()
print("MyTable has slot size " + str(size))
print("Its fields are:")
for fldname in sch2.fields():
if sch2.type(fldname) == INTEGER:
_type = "int"
else:
strlen = sch2.length(fldname)
_type = "varchar(" + str(strlen) + ")"
print(fldname + ": " + _type)
# Part 2: Statistics Metadata
ts = TableScan(tx, "MyTable", layout)
for i in range(50):
ts.insert()
n = random.randint(0, 50)
ts.setInt("A", n)
ts.setString("B", "rec" + str(n))
si = mdm.getStatInfo("MyTable", layout, tx)
print("B(MyTable) = " + str(si.blocksAccessed()))
print("R(MyTable) = " + str(si.recordsOutput()))
print("V(MyTable,A) = " + str(si.distinctValues("A")))
print("V(MyTable,B) = " + str(si.distinctValues("B")))
# Part 3: View Metadata
viewdef = "select B from MyTable where A = 1"
mdm.createView("viewA", viewdef, tx)
v = mdm.getViewDef("viewA", tx)
print("View def = " + v)
# Part 4: Index Metadata
mdm.createIndex("indexA", "MyTable", "A", tx)
mdm.createIndex("indexB", "MyTable", "B", tx)
idxmap = mdm.getIndexInfo("MyTable", tx)
ii = idxmap.get('A')
print(ii)
print("B(indexA) = " + str(ii.blocksAccessed()))
print("R(indexA) = " + str(ii.recordsOutput()))
print("V(indexA,A) = " + str(ii.distinctValues("A")))
print("V(indexA,B) = " + str(ii.distinctValues("B")))
ii = idxmap.get("B")
print("B(indexB) = " + str(ii.blocksAccessed()))
print("R(indexB) = " + str(ii.recordsOutput()))
print("V(indexB,A) = " + str(ii.distinctValues("A")))
print("V(indexB,B) = " + str(ii.distinctValues("B")))
tx.commit()
class GroupByPlan(Plan):
#
# * Create a groupby plan for the underlying query.
# * The grouping is determined by the specified
# * collection of group fields,
# * and the aggregation is computed by the
# * specified collection of aggregation functions.
# * @param p a plan for the underlying query
# * @param groupfields the group fields
# * @param aggfns the aggregation functions
# * @param tx the calling transaction
#
def __init__(self, tx, p, groupfields, aggfns):
super(GroupByPlan, self).__init__()
self.p = SortPlan(tx, p, groupfields)
self.groupfields = groupfields
self.aggfns = aggfns
self.sch = Schema()
for fldname in groupfields:
self.sch.add(fldname, p.schema())
for fn in aggfns:
self.sch.addIntField(fn.fieldName())
#
# * This method opens a sort plan for the specified plan.
# * The sort plan ensures that the underlying records
# * will be appropriately grouped.
# * @see Plan#open()
#
def open(self):
s = self.p.open()
return GroupByScan(s, self.groupfields, self.aggfns)
#
# * Return the number of blocks required to
# * compute the aggregation,
# * which is one pass through the sorted table.
# * It does <i>not</i> include the one-time cost
# * of materializing and sorting the records.
# * @see Plan#blocksAccessed()
#
def blocksAccessed(self):
return self.p.blocksAccessed()
#
# * Return the number of groups. Assuming equal distribution,
# * this is the product of the distinct values
# * for each grouping field.
# * @see Plan#recordsOutput()
#
def recordsOutput(self):
numgroups = 1
for fldname in self.groupfields:
numgroups *= self.p.distinctValues(fldname)
return numgroups
#
# * Return the number of distinct values for the
# * specified field. If the field is a grouping field,
# * then the number of distinct values is the same
# * as in the underlying query.
# * If the field is an aggregate field, then we
# * assume that all values are distinct.
# * @see Plan#distinctValues(String)
#
def distinctValues(self, fldname):
if self.p.schema().hasField(fldname):
return self.p.distinctValues(fldname)
else:
return self.recordsOutput()
#
# * Returns the schema of the output table.
# * The schema consists of the group fields,
# * plus one field for each aggregation function.
# * @see Plan#schema()
#
def schema(self):
return self.sch