Tcl

HTTP GET -- Read Response from Stream

Demonstrates how to send an HTTP GET for a web page, and then read the response from a Stream. The purpose is to fulfill a situation such as the following: "I have a URL and just want the <head> </head> portion of the HTML. This would avoid having to download a potentially enormous web page just to get at header information, such as the <styles>."

Chilkat Tcl Downloads

Download Chilkat for Tcl

Tcl


load ./chilkat.dll

set success 0

# This example requires the Chilkat API to have been previously unlocked.
# See Global Unlock Sample for sample code.

set rest [new_CkRest]

# In this example, we'll get the web page at http://www.nytimes.com/pages/business/index.html
# The domain is "www.nytimes.com", and the path is "/pages/business/index.html"

# If we have only the full URL to begin with, it can be loaded into the Chilkat URL object to
# access the parts:
set url [new_CkUrl]

CkUrl_ParseUrl $url "http://www.nytimes.com/pages/business/index.html"

set bAutoReconnect 1
set success [CkRest_Connect $rest [CkUrl_host $url] [CkUrl_get_Port $url] [CkUrl_get_Ssl $url] $bAutoReconnect]

# Send the GET request (This sends the GET request, but does not read the response.)
set success [CkRest_SendReqNoBody $rest "GET" [CkUrl_path $url]]
if {$success != 1} then {
    puts [CkRest_lastErrorText $rest]
    delete_CkRest $rest
    delete_CkUrl $url
    exit
}

# Read the response header.
set responseStatusCode [CkRest_ReadResponseHeader $rest]
if {$responseStatusCode < 0} then {
    puts [CkRest_lastErrorText $rest]
    delete_CkRest $rest
    delete_CkUrl $url
    exit
}

puts "Response status code = $responseStatusCode"

# We expect a 200 response status.
if {$responseStatusCode != 200} then {

    # If the response status code is not 200, we could check for a redirect status code and 
    # then follow it, read the entire response (as shown here), or just call rest.Disconnect
    set errResponse [CkRest_readRespBodyString $rest]
    if {[CkRest_get_LastMethodSuccess $rest] != 1} then {
        puts [CkRest_lastErrorText $rest]
    }     else {
        puts "$errResponse"
    }

    delete_CkRest $rest
    delete_CkUrl $url
    exit
}

set bodyStream [new_CkStream]

# Set a 10 second read timeout for the stream. 
# (Give up if no data arrives within 10 seconds after calling a read method.)
CkStream_put_ReadTimeoutMs $bodyStream 10000

# Create a background thread task to read the response body (which feeds
# it to the bodyStream object.)
# readResponseBodyTask is a CkTask
set readResponseBodyTask [CkRest_ReadRespBodyStreamAsync $rest $bodyStream 1]

# Start the task. 
set success [CkTask_Run $readResponseBodyTask]

# Read the HTTP response body until the "</head>" is seen, or until
# the end-of-stream is reached.

set sbBody [new_CkStringBuilder]

set exitLoop 0
while {expr !$exitLoop  &&  [[CkStream_get_EndOfStream $bodyStream] != 1]} {

    set bodyText [CkStream_readString $bodyStream]
    if {[CkStream_get_LastMethodSuccess $bodyStream] == 1} then {
        CkStringBuilder_Append $sbBody $bodyText
        if {D]CkStringBuilder_Contains $sbBody "</head>" 0[/L} then {
            set exitLoop 1
        }

    }     else {
        set exitLoop 1
    }

}

# Cancel the remainder of the task...
CkTask_Cancel $readResponseBodyTask

delete_CkTask $readResponseBodyTask

# Ensure we're disconnected from the server.
set maxWaitMs 50
CkRest_Disconnect $rest $maxWaitMs

puts "----"
puts [CkStringBuilder_getAsString $sbBody]
puts "----"
puts "Successfully received the body up to the desired point."

delete_CkRest $rest
delete_CkUrl $url
delete_CkStream $bodyStream
delete_CkStringBuilder $sbBody