Re: RFI: Database URIs

On Nov 25, 2013, at 11:08 AM, Jens Rehsack <[email protected]> wrote:

> Let’s go - shoot:
> 
>    # specify most possible flags via driver flags
>    $dbh = DBI->connect ("dbi:CSV:", undef, undef, {
>        f_schema         => undef,
>        f_dir            => "data",
>        f_dir_search     => [],
>        f_ext            => ".csv/r",
>        f_lock           => 2,
>        f_encoding       => "utf8",
> 
>        csv_eol          => "\r\n",
>        csv_sep_char     => ",",
>        csv_quote_char   => '"',
>        csv_escape_char  => '"',
>        csv_class        => "Text::CSV_XS",
>        csv_null         => 1,
>        csv_tables       => {
>            info => { f_file => "info.csv" }
>            },
> 
>        RaiseError       => 1,
>        PrintError       => 1,
>        FetchHashKeyName => "NAME_lc",
>        }) or die $DBI::errstr;
> 
> And keep in mind, csv_tables can be more complex and there’re other
> attributes like it eg. in DBD::AnyData.

Well, how you would want to handle params would be up to you. No, they are not great for hashes, but do-able.

    db:csv?f_dir=data&f_dir_search=foo&f_dir_search=bar&f_ext=.csv/r&f_lock=2&f_encoding=utf8&csv_eol=%0D%0A&csv_sep_char=,&csv_quote_char=%22&csv_escape_char=%22&csv+class=Text::CSV_XS&csv_null=1&RaiseError=1&PrintError=1&FetchHashKeyName=NAME_lc&csv_tables=%7B%22info%22%3A%20%7B%22f_file%22%3A%22info_csv%22%7D%7D

So yeah, one would need to do some sort of parsting of nested data (JSON in the csv_tables example here), though arrays work okay (e.g., f_dir_search).

OTOH, can you specify all this stuff in a DSN parseable by parse_dsn(), either?

But my point is not to replace the connect() API, but to create a standard of sorts for representing connection info in a URL, to make it easier to specify how to connect to things on the command-line. Yeah, if you want to do everything, it will require more work, but that would be up to the code that handles each driver, which is to say the URI subclass for a particular DBMS.

But this discussion orthogonal to my original questions, which were:

* Should I use a hierarchical scheme like JDBC? I’m leaning heavily toward this now, just need a decent prefix, I'm thinking "dbms", e.g., "dbms:csv:foo".

* Do I have the metadata wrong for any of the DBMSes I have so far added support for? Right now that’s just the ports they listen on by default:

 https://github.com/theory/uri-db/blob/master/t/db.t#L9

Best,

David