| Data Flow of
Sabre Data Source |
| Data Flow Using the Intelligent Work Station (IWS)
Implementation |
| The Intelligent Work Station (IWS) initiates data
transactions between your office and Sabre. Usually the IWS is still working as a display
terminal. This causes a pre-defined message size limitation with no inherent message
security and transparent line protocol (ALC or X.25). The host TPF application consists of
a core business application and Sabre Data Source utilities that provide the necessary
security and line protocols. The TPF application coordinates with the IWS application to
protect your data from the time you access it until its delivery and acknowledgment by
your receiving application. |
| The Process Flow of Sabre Data Source and Data Flow
describe the flow and the message format in Sabre Data Source, which allows developers to
write the IWS application code with their own parsing methods. |
| Process Flow
of Sabre Data Source |
| The following steps indicate the process flow of
Sabre Data
Source. Gateway and Local Area Network (LAN) access, acquiring of pool terminal address,
and sign-in procedures are not included in this process narrative, yet are required before
any Sabre Data Source queries. |
| The Sabre Data Source protocol requires the symbol
'&&' to precede the Sabre input message. This shows the Sabre host that the entry
is a standard transaction format that requires a Sabre Data Source structured output
message. A input sample message is: |
| '&&JX PNR01' |
| During transmission to the Sabre host, the keyboard locks
as the IWS application waits for the response activation. |
| The Sabre host application processes the message, creates
an output response, and sends it through the Sabre Data Source utility. |
| The Sabre Data Source utility reblocks the message into a
1024-byte message block. If the message is multi-block, Sabre sends the first block
immediately to the device. Sabre saves any remaining block or blocks within the AAA area. |
| The IWS receives the message and determines if it is the
last block. The End of Message characters (``) always terminate the last block. If it is
not the last block, the IWS sends the Move Down message ('&&MD') to get the next
block. If a communication line error occurs, use '&&MT' to re-transmit the whole
message starting from the first block. |
| Host Sabre Data Source intercepts '&&MD' and sends
the next block saved in the 'AAA' area. The '&&MD' and '&&MT' are internal
functions within Sabre. |
| The IWS receives the remaining blocks and repeats the
sequences until all parts of the message are received. If the IWS attempts to send
'&&MD' or '&&MT' and there are no more blocks to transmit from the host, a
Sabre Data Source error message occurs. |
| After the IWS receives and processes the message, it
proceeds with the next data request and the process repeats from step 1 again. |
| The IWS can advise the Sabre Data Source utility how many
message blocks to return, from 1 to 16, or 'A' for ALL, without issuing subsequent
'&&MD' commands. This provides the local application flexibility based upon the
amount of data its buffer can manage and will positively impact end-to-end response time
for data inquiries. If no specification is made, one message block will be assumed. The
block size option '&#&', where #=1 to 16 or A, immediately follows the initial
&& within a request. Some sample input entries using this feature are: |
| '&&&8&JX PNR01&02' (to receive 8K of
data prior to '&&MD' command, if required) |
| '&&&16&HOTDFW/12JUN-2NT1' (to receive 16K
of data prior to '&&MD' command, if required) |
|
| NOTES
The
symbol & is the equivalent of ASCII - x'86', ALC -x '2F'. |
| The end of message character is a double segment delimiter,
ASCII character - x'89', or ALC character - x'2E2F'. |
|
