Browse files

Clean up OpenEdge README and add it to main list of drivers

  • Loading branch information...
abevoelker committed Aug 9, 2012
1 parent b91d77d commit 6a8083ea4a5e57e21c4608ba58e0347434442cd3
Showing with 10 additions and 16 deletions.
  1. +4 −0 README.markdown
  2. +6 −16 do_openedge/README.markdown
@@ -44,6 +44,10 @@ At present the following drivers are available:
<td>SQL Server</td>
+ <tr>
+ <td>OpenEdge</td>
+ <td>-</td>
+ <td>x</td></tr>
There is experimental support for [Rubinius][rubinius].
@@ -17,19 +17,18 @@ This driver is currently provided only for JRuby.
operators. In other words, you cannot query rows based on their value. You
will have to query the rows using a different column value (there is no
issue doing reads or writes with `BLOB`/`CLOB` fields). See [ProKB P91964][0]
- for more info. This causes one ByteArray spec to fail.
+ for more info.
* The 10.2B JDBC driver causes `DECIMAL`/`NUMERIC` SQL types to round up to the
nearest integer and then truncate all digits after the decimal point. According
to [ProKB P187898][1], it appears to be a regression bug in the JDBC driver.
- This causes one of the BigDecimal specs to fail.
## Synopsis
An example of usage:
@connection ="openedge://localhost:4000/sports2000")
@reader = @connection.create_command('SELECT * FROM State').execute_reader
-! && @reader.values
The `Connection` constructor should be passed either a DataObjects-style URI or
JDBC-style URI:
@@ -41,7 +40,7 @@ Note that the DataDirect proprietary-style JDBC URI tokenized with `;`s:
-are **NOT** supported (pull requests accepted).
+is **not** supported.
## Requirements
@@ -86,32 +85,23 @@ To run individual specs:
jruby -S rake spec SPEC=spec/connection_spec.rb
-Note that the *typecast tests must be run individually*. There is apparently
-some type of concurrency issue when trying to run many at the same time.
-The error that the JDBC driver returns when this happens (on 10.2B) is:
- [DataDirect][OpenEdge JDBC Driver][OpenEdge] Failure to acquire exclusive schema lock for DDL operation. (7872)
### Spec data setup
The specs require an empty database to use for running tests against
(the database will be written/read from by the tests). Here are
-some commands to be ran from `proenv` to create and bootstrap a
+some commands to be ran from `proenv` to create an empty
database that can be used for testing (note the use of UTF-8,
which is required for proper multibyte string support):
prodb test empty
proutil test -C convchar convert utf-8
- proserve test -S 4000 -cpinternal utf-8 -cpstream utf-8
-Note that I think the SQL engine is unaffected by the codepage
-parameters, but it doesn't hurt to be careful.
+ proserve test -S 4000
## License
This code is licensed under an **MIT License**. Please see the
accompanying `LICENSE` file.

0 comments on commit 6a8083e

Please sign in to comment.