gnumed-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnumed-devel] LDAP vs. SQL


From: Karsten Hilbert
Subject: Re: [Gnumed-devel] LDAP vs. SQL
Date: Fri, 1 Aug 2003 10:17:30 +0200
User-agent: Mutt/1.3.22.1i

> I always thought that it
> should represent something like a named gateway to functional defined part
> of the backend.
It still is. Make that "functionally defined aggregation of
data sources and a common API to that aggregation". 

> In the bootstrap-scripts
>  it as a part or whole of a particular database. 
To be precise boostrap*.conf should say
"schema-for-service-such-and-such" but that's just too long.

> Now, with XML-RPC servers and different data sources for just one group of
> data (contacts, drug-DB etc.)
No. One server == one "service" == one business object. But
this business object may draw from several sources.

> client -> service -> one or more databases
> client -> service -> XML-RPC and/or one or more databases 
This is what I am thinking of.

> One reason for this question is the need to choose between different drug
> databases at runtime. This is currently only possible if one 
> defines one service for every database or if the databases are combined in
> one database. Good enough for now but not for the future.

drug_server = xml_rpc.connect(url=drugref.org, ...)
drug_server.select(cfg.get(drugs, 'data source'))
drug_server.searchByBrand(...)
...
drug_sources = drug_server.get_know_sources()
choice = user_select(drug_sources)
drug_server.select(choice)

The frontend does not assume anything about how drug_server
handles connecting to the various possible sources it knows
about. Another possibility is:

drug_server = xml_rpc.connect(url=drugref.org,...)
drug_server.searchByBrand(...)
some_url = user_input() # url=www.amis.de
try:
    drug_server = xml_rpc.connect(url=some_url, ...)
except:
    print "cannot change drug source to some_url"

Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346




reply via email to

[Prev in Thread] Current Thread [Next in Thread]