view QueryADB.py @ 395:bc7a821004bb api-inversion

Invert audioDB::status / audiodb_status().

To do that without breaking abstractions, we actually need a new field
in the status structure, storing the size of the data region.
Previously, this was computed in the audioDB::status request from the
database header, but I'm assuming that "user" code doesn't have access
to such internals.  While we're at it, name some intermediate values in
audioDB::status() so that I don't get confused.

Here's the thing, though: we need to make sure that the adb_t * that we
have from audiodb_open() or audiodb_create() is propagated all the way
through into the C++ routines that implement library functions -- in
particular those which actually write to the database; otherwise we
won't have a consistent view in memory of the header on-disk (as the adb
header that will have been written to disk won't be the same as the one
in memory).

We can do that, by altering the "API" audioDB constructors to take the
adb_t * argument, and setting the adb field in the audioDB object that
we've already introduced to that.  But now we need to be careful a
couple of times: if we have one, then audioDB::initTables() mustn't
stomp on it; also, if we're only constructing an audioDB instance to
fulfil an API request, we mustn't audiodb_close() the one we have when
we destroy the audioDB object, because the adb_t * is the one we have
passed in and are going to reuse in later calls to the API.

The good news is that we can be careful in just these ways with minimal
code.  The really good news is that once the inversion is complete, all
of this horribleness will automatically go away (as there will be no
code which constructs audioDB objects to fulfil API functions).  Hooray!
It's almost like it was all planned this way.

#!/usr/bin/python

# Python code/library to query the audioDB via the SOAP web interface.
# by Malcolm Slaney, August/September 2008
# [e-mail]

import sys, socket
from xml.dom.minidom import parseString

global debug
debug = False
global dbName
dbName = 'tutorial.adb'

# From: http://www.informit.com/articles/article.aspx?p=686162&seqNum=2
#serverHost = 'research-hm3.corp.sk1.yahoo.com'
serverHost = 'localhost'
serverPort = 14475

# Start the server on serverHost with
#	./audioDB -s 14475

# Here are the templates used for the different kinds of queries.  We'll fill in the
# desired parameters are we go.
LIST_TEMPLATE = """
<?xml version="1.0" encoding="UTF-8"?>
<SOAP-ENV:Envelope
 xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/"
 xmlns:SOAP-ENC="http://schemas.xmlsoap.org/soap/encoding/"
 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
 xmlns:xsd="http://www.w3.org/2001/XMLSchema"
 xmlns:adb="http://tempuri.org/adb.xsd">
 <SOAP-ENV:Body>
  <adb:liszt>
   <dbName>%s</dbName>
   <lisztOffset>0</lisztOffset>
   <lisztLength>100000</lisztLength>
  </adb:liszt>
 </SOAP-ENV:Body>
</SOAP-ENV:Envelope>
"""

SEQUENCE_TEMPLATE = """
<?xml version="1.0" encoding="UTF-8"?>
<SOAP-ENV:Envelope
xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/"
xmlns:SOAP-ENC="http://schemas.xmlsoap.org/soap/encoding/"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns:xsd="http://www.w3.org/2001/XMLSchema"
xmlns:adb="http://tempuri.org/adb.xsd">
<SOAP-ENV:Body>
 <adb:sequenceQueryByKey>
  <dbName>%s</dbName>
  <featureFileName>%s</featureFileName>
  <queryType>%s</queryType>
  <trackFileName></trackFileName>
  <timesFileName></timesFileName>
  <queryPoint>%s</queryPoint>
  <pointNN>%s</pointNN>			<!-- Number of nearest neighbors to retrieve -->
  <trackNN>%s</trackNN>			<!-- Number of tracks to retrieve -->
  <sequenceLength>%s</sequenceLength>	<!-- Number of frames in a shingle -->
  <radius>%s</radius>			<!-- Distance radius to search -->
  <absolute-threshold>-4.0</absolute-threshold>
  <usingQueryPoint>1</usingQueryPoint>
  <lsh-exact>0</lsh-exact>
 </adb:sequenceQueryByKey>
</SOAP-ENV:Body>
</SOAP-ENV:Envelope>
"""

STATUS_TEMPLATE = """
<?xml version="1.0" encoding="UTF-8"?>
<SOAP-ENV:Envelope
 xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/"
 xmlns:SOAP-ENC="http://schemas.xmlsoap.org/soap/encoding/"
 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
 xmlns:xsd="http://www.w3.org/2001/XMLSchema"
 xmlns:adb="http://tempuri.org/adb.xsd">
 <SOAP-ENV:Body>
  <adb:status>
   <dbName>%s</dbName>
  </adb:status>
 </SOAP-ENV:Body>
</SOAP-ENV:Envelope>
"""

SHINGLE_QUERY_TEMPLATE = """
<?xml version="1.0" encoding="UTF-8"?>
<SOAP-ENV:Envelope
 xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/"
 xmlns:SOAP-ENC="http://schemas.xmlsoap.org/soap/encoding/"
 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
 xmlns:xsd="http://www.w3.org/2001/XMLSchema"
 xmlns:adb="http://tempuri.org/adb.xsd">
 <SOAP-ENV:Body>
  <adb:shingleQuery>
   <dbName>%s</dbName>
   <qVector>
   <dim>%s</dim>
      %s
      %s
   </qVector>
   <keyList></keyList>
   <timesFileName></timesFileName>
   <queryType>32</queryType>
   <queryPos>%s</queryPos>
   <pointNN>%s</pointNN>
   <trackNN>%s</trackNN>
   <sequenceLength>1</sequenceLength>
   <radius>%s</radius>
   <absolute-threshold>%s</absolute-threshold>
   <relative-threshold>%s</relative-threshold>
   <exhaustive>%s</exhaustive>
   <lsh-exact>%s</lsh-exact>
  </adb:shingleQuery>
 </SOAP-ENV:Body>
</SOAP-ENV:Envelope>
"""

###############  List Query - Show the files in the database ###########
# Return a list of (key identifier, frame length) pairs.
def RunListQuery():
	global debug, dbName
	message = LIST_TEMPLATE%(dbName)
	
	response = SendXMLCommand(message)
	return response

# Construct a list from the two DOM entries passed.  Used with the map routine to 
# assemble the output.
def ParseListConstruct(f,l):
	el = [f.firstChild.data.encode('latin-1'),l.firstChild.data.encode('latin-1')]
	return el
	
def ParseListXML(response):
	dom = parseString(response)
	fileElements = dom.getElementsByTagName('Rkey')
	# print fileElements
	lenElements = dom.getElementsByTagName('Rlen')
	# print lenElements
	return map(ParseListConstruct, fileElements, lenElements)

###############  Status Query - Show the status of the database ###########
# Return a dictionary with the status fields
def GetDomElement(dom, field):
	els = dom.getElementsByTagName(field)
	if len(els) > 0:
		return els[0].firstChild.data.encode('latin-1')
	else:
		return ""

def RunStatusQuery():
	global debug, dbName
	message = STATUS_TEMPLATE%(dbName)

	response = SendXMLCommand(message)
	# print response
	dom = parseString(response)
	status = {}
	status['numFiles'] = GetDomElement(dom, 'numFiles')
	status['dim'] = GetDomElement(dom, 'dim')
	status['length'] = GetDomElement(dom, 'length')
	status['dudCount'] = GetDomElement(dom, 'dudCount')
	status['nullCount'] = GetDomElement(dom, 'nullCount')
	return status



###############  Shingle/Matrix Query - Show the data closest to shingle range ###########
#
# Encode features as a matrix with dim columns, and (optionally) powers as a matrix with one column
# dim   # number of columns
# f1,1 f1,2...f1,dim # first row
# f2,1 f2,2...f2,dim
# ...
# fN,1 fN,2...fN,dim # last row
# p1,1         # first row's power
# p2,1
# ...
# pN,1         # last row's power
#
def RunShingleQuery():
	global debug, dbName
	featureDim = '3'
	queryVector='<v>1.0</v><v>0.5</v><v>0.25</v><v>0.5</v><v>1.5</v><v>1.0</v>'  # two rows of features
	powerVector='<p>-1.0</p><p>-1.0</p>'   # one power feature per row
	queryPos = '0'  # where in the feature sequence to start the shingle query
	pointNN = '10'  # how many near points to return per track
	trackNN = '10'  # how many near tracks to return
	radius = '1.0'  # search radius
	absoluteThreshold = '-4.5'  # absolute silence threshold in Bels (query and database shingles)
	relativeThreshold = '0'     # relative silence threshold in Bels between features, 0 = ignore
	exhaustive = '0'            # 1 = perform query using all subsequences of features of length sequenceLength
	lshExact = '0'              # if using an index then compute exact distances after LSH retrieval
	message = SHINGLE_QUERY_TEMPLATE
	message = SHINGLE_QUERY_TEMPLATE%(dbName, featureDim, queryVector, powerVector, queryPos, pointNN, trackNN, radius, absoluteThreshold, relativeThreshold, exhaustive, lshExact);
	# print message
	print message
	response = SendXMLCommand(message)
	ParseShingleXML(response)
		

###############  Sequence Query - Show the data closest to one query ###########
def RunSequenceQuery(argv):
	global debug, dbName	
	if len(argv) > 2:
		dbKey = argv[2]
		qType = '32'			# nSequence
		qPos = argv[3]
		pointNN = '10'
		trackNN = '5'
		seqLen = argv[4]
		queryRadius = '0.5'
	else:
		dbKey = 'tmp/3.chr'
		qType = '32'			# nSequence
		qPos = '110'
		pointNN = '10'
		trackNN = '5'
		seqLen = '20'
		queryRadius = '0.4'

	message = SEQUENCE_TEMPLATE
	message = SEQUENCE_TEMPLATE%(dbName, dbKey, qType, qPos, pointNN, trackNN, seqLen, queryRadius)
	# print message
	response = SendXMLCommand(message)
	ParseShingleXML(response)

def ParseShingleXML(response):
	# Grab all the responses
	#	See http://diveintopython.org/xml_processing/parsing_xml.html
	dom = parseString(response)
	resultList = []
	for node in dom.getElementsByTagName('Rlist'):
		# print node.toxml()
		resultList.append(node.firstChild.data.encode('latin-1'))

	distanceList = []
	for node in dom.getElementsByTagName('Dist'):
		# print node.toxml()
		distanceList.append(node.firstChild.data.encode('latin-1'))

	positionList = []
	for node in dom.getElementsByTagName('Spos'):
		# print node.toxml()
		positionList.append(node.firstChild.data.encode('latin-1'))

	# print resultList
	# print distanceList
	# print positionList

	# Print out a summary of the most similar results
	for i in range(0,len(resultList)):
		if i > 0 and resultList[i] != resultList[i-1]:
			print
		print positionList[i], distanceList[i], resultList[i]

	dom.unlink()

###############  XML and Network Utilities ###########
# Send one XML SOAP command to the server.  Get back the response.

def SendXMLCommand(message):
	global debug
	if debug:
		print message
		print

	#Create a socket
	sSock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)

	#Connect to server
	sSock.connect((serverHost, serverPort))

	#Send messages
	sSock.send(message)
	data = ""
	# Now loop, while getting all the data we can get from the socket.
	while True:
		c = sSock.recv(4096)
		if c == "":
			break
		data += c
	if data == "":
		print "No response from the audioDB server"
		sys.exit(0)
	# Split off the HTTP header and the data
	header,response = data.split("\r\n\r\n", 1)
	if debug:
		print 'Client received: ',response

	sSock.close()
	return response

	
###############  Main Program - Figure out which query we want ###########

# Argument processing scheme described at: http://docs.python.org/lib/module-getopt.html

import sys
if __name__=="__main__":
	cmdname = sys.argv[0]
	if len(sys.argv) == 1:
		print "Syntax: " + sys.argv[0] + " -q feature_file pos len"
		sys.exit(1)

	queryType = sys.argv[1]
	if queryType == '-s' or queryType == 'status':
		response = RunStatusQuery()
		for k, v in response.iteritems():
			print k, v
	elif queryType == '-q' or queryType == 'query':
		RunSequenceQuery(sys.argv)
	elif queryType == '-l' or queryType == 'list':
		response = RunListQuery()
		# print response
		results = ParseListXML(response)
		for (f,l) in results:
			print "%s\t%s" % (f,l)
	elif queryType == '-v' or queryType == 'vector':
		response = RunShingleQuery()