FileMaker - Asterisk Communication - database

Does anyone had accomplished this?.
The big picture would be to develop the entire asterisk GUI from filemaker, but right now I'm asking you help to connect both.
Asterisk controls our entire Call Center. I would like the info from incoming calls and queues to be written in a FileMaker database.

Disclaimer: I don't know the first thing about FileMaker. But, if it's like any other programming language (which from what I know, I'm not sure that's true) then let's look at the options on how we'd accomplish this generically with other programming languages...
If you just want the results of your call, the CDRs (call detail records), you can configure Asterisk to output custom CDRs in cdr_custom.conf (check it out if you'd generated the sample configurations)
Here's an example cdr_custom.conf:
[mappings]
Simple.csv => ${CSV_QUOTE(${EPOCH})},${CSV_QUOTE(${CDR(src)})},${CSV_QUOTE(${CDR(dst)})}
It will drop a file typically in /var/log/asterisk/ if you haven't changed it otherwise in your configuration.
Then, either restart asterisk, or more gracefully just reload the cdr module:
asterisk*CLI> cdr show status
asterisk*CLI> module reload cdr_custom.so
Using the resulting file, parse the CSV and format it in a friendly fashion for Filemaker / "your language of choice".
If you're looking for real-time information about calls, it does get more complicated. Probably for just reporting purposes, you can use the Asterisk AMI (Asterisk Manager Interface). (Canonical wiki page linked)
This is a TCP IP application, open a socket to it, and you're good to go. There's also the AJAM interface (Asynchronous Javascript Asterisk Manger). Which you can make HTTP calls to.
Lastly, if you want to do further processing during the routing of the call via the dialplan, you'd want to use AGI (Asterisk Gateway Interface) which is called from the dialplan, and is all over STDIO.

Actually you can create a ODBC connection to the asterisk databases and use filemaker to access the tables direct. It will give you a 'live' connection and save you all the import <-> export fuss. If you google on filemaker odbc you'll get results on setting this up, it works quite easy (not always fast depending on your query but certainly a lot quicker than the manual method)

Related

How to automatically save received pdf files from gmail into a database?

I would like to know if this scenario would be possible in any programming language combined with any database technology.
I would like to automatically save received pdf files that are attached in emails into a database. Is this possible? Is there any library or framework available to do so?
Yes, I would recommend using Google Apps Script for this. The approach you should follow is to use the GmailApp class (Documentation here) to get the messages you need, you can use methods like getInboxThreads() (Documentation), to retrieve the messages.
After you've found the message and retrieved the attachment (which you can do withgetAttachments() (Documentation)), you can use the JDBC Service to connect with external databases. The specifics here depend a lot on what database you want to connect with, but the documentation will lead you in the right direction.

Win32 development. Standards

A call to all Win32 developers... I'm developing an application in C using plain Win32. I wanted to ask about Windows development standards regarding these things:
Is there a standard Windows error log api? For example if my client uses my app and it crashes, I would like them to send me the error log and I would prefer this being a standard location so they can maybe access it with a standard Windows log utility.
My app needs to store settings information. I think the registry is the standard utility for this task. Is that right?
My app needs to store and retrieve files that it downloaded from the internet - images, executables etc. Is Application Data/myapp the standard location to store this type of information?
My app needs a very straight-forward database - I'm using CSV for this. I basically need to store and retrieve this type of data so I'm just serializing a .csv file from Application Data/myapp. Is there a better Windows standard way of doing this?
That's all for now :). Thanks!
Is there a standard Windows error log api?
There is the Windows Event Log, but I don't think you want a typical user having to go into it to extract your logged information.
You probably don't want to log by default, unless you're shipping questionable pre-release code. When a user is experiencing problems, then you have them turn logging on. In this case, I recommend placing the file somewhere that typical users have experience with, like My Documents.
By the way, if you're writing a standalone application and want the best possible information in the event of a crash, look into minidumps. Here is a Codeproject sample.
My app needs to store settings information
Yep, registry.
My app needs to store and retrieve files
Yes, App Data. Just be sure to use SHGetFolderPath and CSIDL_APPDATA.
My app needs a very straight-forward database
There's nothing wrong with CSV for simple data. You could store the data in XML and use MSXML to process it, if you prefer. I've used SQlite in the past when I needed fast, lightweight storage of more complicated data.

BizTalk 2006 - Copy a received file to a new directory

I want to be able to copy the file I have which comes in as XML into a new folder location on the server. Essentially I want to hold a back up of the input files in a new folder.
What I have done so far is try to follow what has been said on this forum post - link text
At first I tried the last method which didn't do anything (file renaming while reading). So I tried one of the other options and altered the orchestration and put a Send shape just after the Receive shape. So the same message that comes in is sent out to the logical port. I export the MSI, and I have created a Send Port in the Admin console which has been set to point to my copy location. It copies the file but it continues to create one every second. The Event Viewer also reports warnings saying "The file exists". I have set the Copy Mode of the port to 'overwrite' and 'Create New', both are not working.
I have looked on Google but nothing helps - BTW I support BizTalk but I have no idea how pipelines, ports work. So any help would be appreciated.
thanks for the quick responses.
As David has suggested I want to be able to track the message off the wire before BizTalk does any processing with it.
I have tried to the CodePlex link that Ben supplied and its points to 'Atomic-Scope's BizTalk Message Archiving Pipeline Component' which looks like my client will have to pay for. I have downloaded the trial and will see if I have any luck.
David - I agree that the orchestration should represent the business flow and making a copy of a file isn't part of the business process. I just assumed when I started tinkering around I could do it myself in the orchestration as suggested on the link I posted.
I'd also rather not rely on the BizTalk tracking within the message box database as I suppose the tracked messages will need to be pruned on a regular basis. Is that correct or am I talking nonsense?
However is there a way I can do what Atomic-Scope have done which may be cheaper?
**Hi again, I have figured it out from David's original post as indicated I also created a Send port which just has a "Filter" expression like - BTS.ReceivePortName == ReceivePortName
Thanks all**
As the post you linked to suggests there are several ways of achieving this sort of result.
The first question is: What do you need to track?
It sounds like there are two possible answers to that question in your case, which I'll address seperately.
You need to track the message as received off the wire before BizTalk touches it
This scenario often arises where you need to be able to prove that your BizTalk solution is not the source of any message corruption or degradation being seen in messages.
There are two common approaches to this:
Use a pipeline component such as the one as Ben Runchey suggests
There is another example of a pipeline component for archiving here on codebetter.com. It looks good - just be careful if you use other components, and where you place this component, that you are still following BizTalk streaming model proper practices. BizTalk pipelines are all forwardonly streaming, meaning that your stream is readonly once, and all the work on them the happens in an eventing manner.
This is a good approach, but with the following caveats:
You need to be careful about the streaming employed within the pipeline component
You are not actually tracking the on the wire message - what your pipeline actually sees is the message after it has gone through the BizTalk adapter (e.g. HTTP adapter, File etc...)
Rely upon BizTalk's out of the box tracking
BizTalk automatically persists all messages to the message box database and if you turn on BizTalk tracking you can make BizTalk keep these messages around.
The main downside here is that enabling this tracking will result in some performance degradation on your server - depending on the exact scenario, this may not be a huge hit, but it can be signifigant.
You can track the message after it has gone through the initial receive pipeline
With this approach there are two main options, to use a pure messaging send port subscribing to the receive port, to use an orchestration send port.
I personally do not like the idea of using an orchestration send port. Orchestrations are generally best used to model the business flow needed. Unless this archiving is part of the business flow as understood by standard users, it could simply confuse what does what in your solution.
The approach I tend to use is to create a messaging send port in the BizTalk admin console that subscribes to your receive port. The send port will then just use a standard BizTalk file adapter, with a pass through pipeline.
I think you should look at the Biztalk Message Archiving pipeline component. You can find it on Codeplex (http://www.codeplex.com/btsmsgarchcomp).
You will have to create a new pipeline and deploy it to your biztalk group. Then update your receive pipeline to archive the file to a location that the host this receive location is running under has access to.

How do I create a database on the computer in flash?

I want to create a a application in actionscipt 3.0 that allows the user to listen to music and read descriptions of the music. For this to happen i suppose there should be a database where the textbits and music is located and then flash fetch the info when the correct buttons are pushed. The database will contain up to 100 tracks and textbits.
The application will function on a stand that won't have a connection to the internet.
What is the easiest way to do this in actionscript 3.0?
If any of you are familiar with UML and thinks this might help in understanding the problem, then here is use-case and flow-chart:
alt text http://img135.imageshack.us/img135/1498/flowchart2.jpg
alt text http://img27.imageshack.us/img27/1000/usercase.jpg
Thanks in advance.
The easiest way to do what you're asking is probably to store the files in a directory on the machine the application is going to be running on, and then design an XML structure for storing your data. The XML is easily loaded in to Flash at runtime and is easily edittable.
Your other option would be running a database server on the machine, creating web services that run locally and push/pull the data from the database, and then call those services from your Flash application.
The first option is most definitely the easiest and should be able to provide exactly what you need. The second would be more geared towards a distributed Flash application where you needed a central data repository for the clients.
If you're building an AIR application, you can use the integrated SQLITE database. But, i agree with Justin, the easiest way is to use a XML file.
You can probably consider using "Local Shared Objects" which is a kind of cookie, with bigger capacity (100Kb by default, but you can change it). Compared to other solutions already proposed, it has then advantage of not requiring any web server.

Writing data into a database using a fully REST web service

How would one create a REST web service to write a row into a databse table. Use the follwoing scenario:
The table is called Customer - the
data to be inserted into the row would
be the name, addresss, telephone
number, email.
I think its impossible to describe the whole thing end to end in Java or C#, and I would never expect that, but here are the questions I have popping into my head as I prepare for coding:
How the URI would look (eg. if you use this URL - http://www.example.com/)?
What info would go into the HTTP envelope?
Would I use POST when writing to the database in this way?
Do I use a resource to store the posted data from the client? Is this even necessary if the data is being written to a database anyway?
When the data to be writeen into the db is recieved by the server - how do I physically insert it into the database - do I call some method on the server to actually write the data (in Java)? - this doesn't seem to fit with truely REST architecture - shunning RPC calls.
Should I even be bothering writing to a DB - should I be storing my data as a resource?
As you can see I need a few issues clearing in my head. Any help much appreciated.
First of all, I'm not either java nor c# expert and I don't exactly know what means do these languages have to support REST design, but in general:
http://www.example.com/customers - customers is a collection of resources and you want to add a new resource to this collection
It depends on various things - you should probably set the content-type header (according to the data format in which you are sending the representation) and set some authentication headers if you need it.
Yes, you always use POST to create a new entry in a collection of resources.
I don't fully understand this question, to be honest. What do you mean by "inmediately writing data into the database"?
REST is primarily just a style of communication between server and a client. It doesn't say anything about how you should handle the data received by using it. The usual way how modern web approaches (MVC style frameworks) solve it, is by routing every REST action to a method of some class (usually a controller instance) where you handle the received parameters (eg. save them to the database) and generate a response to be sent back.
For a very brief and very clear introduction to REST have a look at this short video.
RESTful Web Services, published by O'Reilly and Associates, seems to fit the bill you're looking for.
As far as doing it in Java, Sun has a page on it.

Resources