LAE

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

Hi guys,

I need your help, to understand how can I use my Phython node to find match using more than one field.

This is my current code. At this moment, I'm using only one field to match: PREFIX vs. net_b_number_normalized.

######################################
import braininfo

def setup(brainNodeControlObj, BrainNodeClass):
''' Must return a class that inherits from BrainNodeClass. '''
class BrainNode(BrainNodeClass):
def initialize(self):
''' Called at node initialization, before first pump.'''
super(BrainNode, self).initialize()

self.lookupList = list()
self.destHash = list()
inRec = self.inputs[1].read()
while inRec:
self.lookupList.append(inRec)
self.destHash.append(inRec["PREFIX"])
inRec = self.inputs[1].read()

metadata = self.inputs[0].metadata
metadata.append("UOOM_EVT", "String")
metadata.append("REPEAT_EVT","String")
metadata.append("PRECOCHAMADA_EVT", "String")
metadata.append("UOOM1_MIN_SEC", "String")
metadata.append("REPEAT1_MIN_SEC", "String")
metadata.append("PRECOCHAMADA1_MIN_SEC", "String")
metadata.append("CHARGE_TYPE1_MIN_SEC", "String")
metadata.append("UOOM1_SEC", "String")
metadata.append("REPEAT1_SEC", "String")
metadata.append("PRECOCHAMADA1_SEC", "String")
metadata.append("CHARGE_TYPE1_SEC", "String")
metadata.append("UOOM2_MIN_SEC", "String")
metadata.append("REPEAT2_MIN_SEC", "String")
metadata.append("PRECOCHAMADA2_MIN_SEC", "String")
metadata.append("UOOM2_SEC", "String")
metadata.append("REPEAT2_SEC", "String")
metadata.append("PRECOCHAMADA2_SEC", "String")
metadata.append("UOOM3_MIN_SEC", "String")
metadata.append("REPEAT3_MIN_SEC", "String")
metadata.append("PRECOCHAMADA3_MIN_SEC", "String")
metadata.append("UOOM3_SEC", "String")
metadata.append("REPEAT3_SEC", "String")
metadata.append("PRECOCHAMADA3_SEC", "String")
metadata.append("OUTCTP", "String")
metadata.append("OPERADORA", "String")
metadata.append("DIRECCAOCHAMADA", "String")
metadata.append("PREFIX", "String")
metadata.append("TIER_ID", "String")
metadata.append("POI_ID", "String")
metadata.append("TRUNK_ID", "String")
self.outputs[0].metadata = metadata

def finalize(self, val):
''' Called at node end, after last pump call. '''
super(BrainNode, self).finalize(val)

def pump(self, quant):
''' Will continue to be called until False is returned.

Only a small amount of work should be done in pump without checking
quant.permitsRunning. When that returns False, return from pump
with a True value to be called again, a False value when finished.'''
while quant.permitsRunning(self):
# your code here
inRec = self.inputs[0].read()

if not inRec:
return False

dest = inRec['net_b_number_normalized']

found = False
i=0
while not found and i<len(self.destHash):
# rec = self.lookupList[i]
# prefix = rec["COSTCODE"]
prefix = self.destHash[i]
if dest.startswith(prefix):
found = True
rec = self.lookupList[i]
i = i+1

outRec = self.outputs[0].newRecord()
outRec._copyFrom(inRec)
if found:
outRec["UOOM_EVT"] = rec["UOOM_EVT"]
outRec["REPEAT_EVT"] = rec["REPEAT_EVT"]
outRec["PRECOCHAMADA_EVT"] = rec["PRECOCHAMADA_EVT"]
outRec["UOOM1_MIN_SEC"] = rec["UOOM1_MIN_SEC"]
outRec["REPEAT1_MIN_SEC"] = rec["REPEAT1_MIN_SEC"]
outRec["PRECOCHAMADA1_MIN_SEC"] = rec["PRECOCHAMADA1_MIN_SEC"]
outRec["CHARGE_TYPE1_MIN_SEC"] = rec["CHARGE_TYPE1_MIN_SEC"]
outRec["UOOM1_SEC"] = rec["UOOM1_SEC"]
outRec["REPEAT1_SEC"] = rec["REPEAT1_SEC"]
outRec["PRECOCHAMADA1_SEC"] = rec["PRECOCHAMADA1_SEC"]
outRec["CHARGE_TYPE1_SEC"] = rec["CHARGE_TYPE1_SEC"]
outRec["UOOM2_MIN_SEC"] = rec["UOOM2_MIN_SEC"]
outRec["REPEAT2_MIN_SEC"] = rec["REPEAT2_MIN_SEC"]
outRec["PRECOCHAMADA2_MIN_SEC"] = rec["PRECOCHAMADA2_MIN_SEC"]
outRec["UOOM2_SEC"] = rec["UOOM2_SEC"]
outRec["REPEAT2_SEC"] = rec["REPEAT2_SEC"]
outRec["PRECOCHAMADA2_SEC"] = rec["PRECOCHAMADA2_SEC"]
outRec["UOOM3_MIN_SEC"] = rec["UOOM3_MIN_SEC"]
outRec["REPEAT3_MIN_SEC"] = rec["REPEAT3_MIN_SEC"]
outRec["PRECOCHAMADA3_MIN_SEC"] = rec["PRECOCHAMADA3_MIN_SEC"]
outRec["UOOM3_SEC"] = rec["UOOM3_SEC"]
outRec["REPEAT3_SEC"] = rec["REPEAT3_SEC"]
outRec["PRECOCHAMADA3_SEC"] = rec["PRECOCHAMADA3_SEC"]
outRec["OUTCTP"] = rec["OUTCTP"]
outRec["OPERADORA"] = rec["OPERADORA"]
outRec["DIRECCAOCHAMADA"] = rec["DIRECCAOCHAMADA"]
outRec["PREFIX"] = rec["PREFIX"]
outRec["TIER_ID"] = rec["TIER_ID"]
outRec["POI_ID"] = rec["POI_ID"]
outRec["TRUNK_ID"] = rec["TRUNK_ID"]
self.outputs[0].write(outRec)

# finished your quantum, you want pump called again
return True

# If you implement your logic here, you don't need code below. If you want
# to inherit, and add more logic at next level, uncomment below code

# py3File = brainNodeControlObj.properties.getString("ls.brain .node.BrainPython.python3ImplementationFile")
# braininfo.loadModule(py3File, "py3", brainNodeControlObj)
# import py3
# BrainNodeImpl = py3.setup(brainNodeControlObj, BrainNode)
# return BrainNodeImpl

return BrainNode
##################################################

How can I introduce another field to this match, for instance: TRUNK_ID.
How can I do it? With this field, is a one-on-one match; not like the field that I'm using right now, the prefix, which is a match by the destination number that fits.

Many thanks for your help,

Regards
Natanael

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: xathras

Natanael,

Could you create a dummy graph and upload to here with a BRD with some data?
As an FYI, when you paste the code online it loses all the indentation, so its not as straight forward as looking at your code. Python is pretty anal about indentation (if you were not aware).

Also, is there a reason why you are doing in python and not a join node?

Regards
Wayne
Attachments:
example.brg

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

Hi Wayne,

Thanks for your fast reply.
I'm using python, because the field I'm using for match (destination numbers vs. prefix) is not a field where the match is direct.
For example: imagine, in a call, the b-number 707202303.
My ref data, with whom I want to cross ref, has the prefix 7072.
So I'm using python to see the hightest exact match, by seeing first the entire b-number, and then, it takes out the last number, and tries to find the match...and so on.

Attached is the sample brg

Regards,
Natanael
ZON
Attachments:
MDA Cost sample.zip

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: mmarinelli

There is an prototype node in the Lavastorm Labs forum which may be useful here. The Substring Lookup node will allow you to match keys where one is a substring of the other. The node is described here: http://community.lavastorm.com/showt...bstring-Lookup, and can be downloaded as part of the Lavastorm Labs prototype library here: http://community.lavastorm.com/showt...brary-Released

These nodes are experimental so they may not have the stability or performance profile of an official Lavastorm release, and they are not officially supported, but I think at the very least the code for this node should point you in the right direction.

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: xathras

Hmm interesting. I did a puzzle like this in my uni days, but not in python. Let me have a think on this one. Regex would of been a good fit for this scenario i think.
I used to work as a billing data analyst for a teleco company and had to complete an exercise to look up dialing codes to regions to get the best fit. Maybe i still have my notes

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

I'm not allow to access the link you sent.

Yeh, it would be great if you can find your notes.
As soon as you have any news, let me know.

Natanael
ZON

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: mmarinelli

Natanael, in order to obtain the prototype library you have to join the Lab Experimenter group (apologies, I had assumed that you were already a member). This link provides instructions on how to request membership for this group: http://community.lavastorm.com/showt...ining-Overview

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

Ok, will do.

But, if you can help me alter the python, it would be great, because I use that node in almost every project that I have.

Natanael

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

Hi,

I've tried to use the node Substring Lookup, but is not running when I click F5.
I've put in the InputKey field net_b_number_normalized and in lookupkey field PREFIX, and in match type "start with".

Is not working...can you help?

Regards,
Natanael

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: mmarinelli

Sure, I can try to help, but it would be helpful to know how exactly it is not working. Could you provide an error message / log or a screen shot of what you're seeing?

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

See the image below...can I do that?
Multiple fields?

And, what if I want that the final table, brings fields from the lookup key...how do I do it?

Regards,
Natanael
Attachments:
Doc1.zip

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: mmarinelli

The node you are using isn't coded to handle multiple input fields on each side of the join. It currently only works with a single field from each data set, but you could alter the code to use arrays for each side when performing a join. If you don't want to change the code, you could probably accomplish what you're trying to do using a sequence of nodes (one for each field you want to compare).

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

Ok.

Sorry to ask, but did you managed to look at my phyton node?
I think for me it would be easier to change my phyton node for now, and after, try to include this new node.

Cheers,
Natanael

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

Hi,

I tried to use the new node again, only with one field, but it turned out error.
>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<
ERROR processing data:
heap allocation failed: m_pchData

Installing log (11/09/2012 12:05:45) ...
2012-11-09 12:50:34; Level: 4; Type: 0; Desc: "heap allocation failed: m_pchData"; File: "../../../include/jigsaw\Buffer.h"; Line: 27; Context: ""
Uninstalling log (11/09/2012 12:50:34) ...
>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<

Do you know why?

Regards,
Natanael
ZON

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: rboccuzzi

Natanael, I tried taking a look at your sample graph attached above, but you didn't include the necessary base libraries, so I wasn't able to see what you are doing.

From your Python error above, it sounds like you just ran out of memory. Looking at your Python code above, it seems you are loading the entire second input into memory. How many records are on your second input? I looks like you loaded them into a hash until you ran out of space. You could try loading the MINIMUM you need into memory, so maybe just the thing you are matching against and an ID, and then you can do a straight up join after? This would limit what you are storing into memory in your hash, and you should be able to get a lot more rows in memory.

Also, it sounds like you might be trying to do something very similar to what is discussed in this thread here, matching country prefix codes. I think the libraries that Mick posted might do what you are looking for.

Cheers
Rich