The link below should take you to the source code for Reado2_mfs.

The link is interesting for three Reasons:

1. reado2_mfs implements the READO call by caching records on a client machine. I have tested it with moderate (500k records) files and it seems stable. I have found that it can speed up reports. If you add it to SYSLISTS, it will not write T* and W* records to disk, so large selects should run faster. This is a work in progress, feel free to comment.

2. The cache is implemented with .Net collections objects and their default COM wrappers. I found that Hashtables, Arrays, sorted arrays and Queues all work as OLE objects. This seems useful.

3. The source is published with Google Documents. I did not use their API, but with the correct Google license it is possible to create and store documents, spreadsheets, schedules, etc from within an application. They are using a REST model, where every document (resource) has a unique identifier. This model is analogous to our model of using the key to store the record. I have a hunch that integrating Google apps as an extension of OpenInsight could be as useful as integrating .Net

See

Reado2_mfs

- Bob

Hi Bob,

Good stuff!

pre.code {

  background-color: #E5E5E5;

  border: 1px solid #000000;

  width: 450px;

  padding: 5px;

  font-family: courier, verdana, arial, serif;

  margin: 0px 10px auto;

}

Line 94: status$=true

Should that $ be there??

Cheers, M@

[img]http://www.vernonsystems.com/images/logo_main_ani.gif[/img]

fixed.

Thanks!

cool and interesting idea .. thanks

Hi Bob,

Sorry, but only just noticed - that $ probably belongs at the end of TRUE to make TRUE$ - yes?

M@

[img]http://www.vernonsystems.com/images/logo_main_ani.gif[/img]

Hi Matt.

I fixed several typos in the Reado2_mfs code . It should compile and run now.

View this thread on the Works forum...