summaryrefslogtreecommitdiffstats
path: root/doc
diff options
context:
space:
mode:
Diffstat (limited to 'doc')
-rw-r--r--doc/README.csv7
-rw-r--r--doc/README.http24
-rw-r--r--doc/README.standard15
-rw-r--r--doc/README.xml21
4 files changed, 67 insertions, 0 deletions
diff --git a/doc/README.csv b/doc/README.csv
new file mode 100644
index 0000000..5f77c09
--- /dev/null
+++ b/doc/README.csv
@@ -0,0 +1,7 @@
+This I/O Handler was put together primarily as a demonstration of
+how easy it is to write handler routines for astmanproxy.
+
+However, it could be handy if you need CSV output, too. :) There
+is no _read routine at this time; I would welcome someone to write
+one and submit it to the project!
+
diff --git a/doc/README.http b/doc/README.http
new file mode 100644
index 0000000..63268a9
--- /dev/null
+++ b/doc/README.http
@@ -0,0 +1,24 @@
+This provides a very basic interface to http; it should be considered
+poorly built, lame, and dangerous until further notice.
+
+That said, if you GET or POST form data name/value pairs to the proxy port
+using HTTP, the proxy will treat it as Name: Value input and will act on
+it, and will respond back in whatever format you want to use.
+
+A useful scenario might be to use http input and xml output.
+
+The following form can be used as a test:
+
+<HTML>
+<FORM ACTION="http://myhost:1234" METHOD=POST>
+Action: <input name="Action" type="text" value="Ping"><br>
+ActionID: <input name="ActionID" type="text" value="Foo"><br>
+<input type=submit><br>
+</FORM>
+</HTML>
+
+See samples/httpast.html for a sample file that implements this.
+
+It would be best to use this with autofilter=on since right now you are
+only going to get one response block back, and it might as well be
+relevant. :)
diff --git a/doc/README.standard b/doc/README.standard
new file mode 100644
index 0000000..40699d7
--- /dev/null
+++ b/doc/README.standard
@@ -0,0 +1,15 @@
+This is the standard input/output handler. This handler implements the
+traditional Asterisk Manager interface I/O format, which can be
+described as follows:
+
+Each block of input and output should be described in terms of headers,
+each terminated with \r\n (cr+lf) and consisting of a name/value pair,
+and each block of headers should be terminated with two \r\n sequences:
+
+Name1: Value1\r\nName2: Value2\r\n\r\n\r\n
+
+The "standard" IO handler is used by astmanproxy to communicate with
+Asterisk server sessions. Clients connecting to astmanproxy can use
+the "standard" handler if they wish, for either input, output or both,
+or may use any other available handler.
+
diff --git a/doc/README.xml b/doc/README.xml
new file mode 100644
index 0000000..0e34c97
--- /dev/null
+++ b/doc/README.xml
@@ -0,0 +1,21 @@
+The XML input handler is homegrown and very basic at this point. The
+following represents valid XML input:
+
+<AsteriskManagerInput>
+ <Action Value="Ping"/>
+ <ActionID Value="foo"/>
+</AsteriskManagerInput>
+
+<AsteriskManagerInput>
+ <ProxyAction Value="SetOutputFormat"/>
+ <Outputformat Value="CSV"/>
+</AsteriskManagerInput>
+
+It's entirely possible that other formats will break the parser entirely.
+I am toying with going with a lightweight XML parser like MiniXML that may
+give some more flexibility at the price of having a dependency. It is
+very likely, though, that the current parser will work fine for most
+applications.
+
+DCT 6/23/2005
+