|
4.0 Extract Configuration
|
The package supports multiple extracts. The configuration file for each
Extract is located in a plain text format and can be found at
{INSTALLATION_DIRECTORY}
/
{EXTRACTS_DIRECTORY}
/{EXTRACT_NAME}. The EXTRACT_NAME is selected
when an Extract is
defined
. It is a good idea to keep a backup of
Extract configurations in this directory.
One of the important functions of setting up an Extract is mapping
the fields from the RETS server (Source) into the fields of the
Target. Depending on the Target type, the mapping approach
is different. Please see the documentation related to setting up a
Target
for a discussion on mapping.
The following items are used to configure an Extract:
- SOURCE - The Source to use.
- TARGET - The Target to use.
- USER - The user to use during downloading. This is only used in
situations where the target supports the concept. In this version, only
Open-Realty supports different users.
- REFRESH - A boolean value that controls whether listings will be
replaced if they already exist. Defaults to "true".
- MLS_ONLY - A boolean value that controls whether user supplied
information will be retained. If set to true, downloads always result in
information in the target containing only MLS listing data. Any user
supplied information will be lost. This setting does not apply to all
Target types. Defaults to "false".
- CLASS_NAME_STYLE - Only used for Open-Realty targets. Selects the
name style to use for RETS class names. Defaults to "" signifying SystemName.
- TRACE - A boolean value that controls whether detailed information
is shown to the user during downloading. Defaults to "true".
- USE_PROXY - A boolean value that controls whether a proxy is used
to retrieve images from the RETS server. If the RETS server does not support
the Location option, this should be set to "false" to avoid the RETS GetObject
overhead experienced with proxies. If the RETS server supportes Location,
it probably has a decicated media server, making the use of proxies as faster
method for retrieving images. The RETS Location option is auto-detected and
stored as
MEDIA_PROXY
setting of the SOURCE. Defaults to "true".
- BATCH_SIZE - If the RETS server supports "pagination" (RETS OFFSET
option), the size of each batch. If the RETS server does not support
"pagination", this value cannot be changed by the user because large values
can corrupt the query string to the RETS server. Defaults to "10".
- MAX_IMAGE_COUNT - The total numer of images to expect from any one
listing. Defaults to "6".
- COLUMN_LIST - Only valid for XML and CSV Target types. The list of
columns to download. It is not recommended that users modify this variable.
- WORKING_FILE_PATH - Only valid if the server does not support
pagination. This is the name of the directoy used to temporarily store results
while performing pagination on the client side. If the path for this setting
does not exist, all manipulation will be performed in memory which can result
in running out of memory if your query is large.
- CACHE_SIZE - Only valid if the server does not support pagination
and the WORKING_FILE_PATH esists. This is the size of the memory buffer to
use when downloading listings. Defaults to 2048K.
- LIMIT - Allows you to define the maximum number of records pulled
from the server. Defaults to 0 (no limit).
A Batch Control File is created for for each
extract and is located in a plain text format and can be found at
{INSTALLATION_DIRECTORY}
/
{BCF_DIRECTORY}
/{EXTRACT_NAME}. More information about Batch Control Files can be
found in the section dedicated to
Command Line Usage
.
If you specify a query with only one field and that field is also the
UNIQUE_KEY
, then all pagination logic will be bypassed. There are no settings that can
force this situation outside of the column choices. This feature is known as
"Index Mode" and is handy for creating lists of IDs for different situations
like "all pending", "all active", etc.
|
|
|
