During his unconference session at OpenWorld 2007, Lutz Hartmann used rlwrap to give SQL*Plus and RMAN command line history and basic editing functionality. Like the Windows Process Explorer post I wrote recently, this is another example of a gizmo I’ve used in the past then completely forgotten about, so I’m grateful to Lutz for reminding me. To see how I install and configure it click here.



Update: Someone and just told me my rlwrap post is now ranked higher than Howard’s on Google. This is really a “duck and cover” event. 🙁

Author: Tim...

DBA, Developer, Author, Trainer.

14 thoughts on “rlwrap…”

  1. Hi Tim,
    As I told Lutz a long time ago, aliasing sqlplus is a dubious practice, especially if you are a dba. For example, I often notice that CTRL-C does not work to interrupt evil scripts with rlwrap. Also rlwrap support only interactive mode, so if you do a kind of sqlplus <<EOF … select … EOF, it will not do what you expect. Try a different alias, like

    alias SQLPLUS=’rlwrap sqlplus’
    alias RMAN=’rlwrap rman’



  2. Not sure about other shells, but bash and ksh don’t use aliases if executed in non interactive mode ( and that is the case with shell scripts ), so , just curious , what can be issue with here documents…

    Best regards


  3. I agree with Lutz that it is sensible to use an alias with a different name than the rlwrapped command. However, rlwrap (now at 0.30) *does* work as expected when its in- or output are redirected . And if CTRL-C doesn’t work correctly in a rlwrapped command, it’s a bug: please report it!.


    Hans (rlwrap author)

  4. Hi.

    Maxim: That’s interesting.

    Fidel: As I said in my post, “another example of a gizmo I’ve used in the past then completely forgotten about”. I’m not claiming it’s new. I wrote the post to remind myself because this is something I’ve used before then forgotten about.

    Hans: I’ll take a look at it and get back to you.



  5. No-one’s claiming you did say it was new. Quite obviously it isn’t, as your own prose makes quite clear. Fidel’s point was, I think, rather that it was documented a long time ago on an Oracle website near you and you might have had the grace to acknowledge as much. Or, having had your memory refreshed, the grace to go back and edit the original piece so that it contains an acknowledgment. Particularly since the original site says rather more about what you might do with the product than you manage to do.

    There’s an acknowledgment, for example, in the Doris articles currently on the new Dizwell site that yours was the first documentation I’d ever seen explaining how two fixes I already knew about -and indeed, had previously documented separately- could be combined to produce a fix for installing Oracle 11g on Fedora 8. No-one’s claiming you invented those fixes. No-one’s claiming I did by documenting them in the different, separate and earlier contexts. Fact remains, the combination of the two was new, and you were the first to mention it, so it warranted at least a passing mention, which it got.

    Acknowledging the source of one’s novelties is at least an honourable thing to do, dude.

  6. There is a big difference here. You read my article and used the information in it to create something new. Depending on your opinion that may or may not be worthy of a credit.

    In the case of this article, I didn’t know you had written about it until the comment mentioned it. I also didn’t know Lutz had blogged about it until Laurent pointed it out in his comment. Since neither of you were top of my Google search using the term “rlwrap”, and it was so easy to set up, I had no reason to search for additional information, which is why I didn’t find your site. In this situation is is totally not necessary to credit you as you provided no help in the writing of this article.



  7. It’s OK Tim. You can’t do basic research. We know that now. You don’t have to keep explaining. But you will keep repeating what has already been better said elsewhere. Your choice.

  8. Howard: By your own admission, having a PhD proves I can do research. I don’t need to prove anything to you or anyone else.

    On the other hand you consistently prove your pathetic nature by dragging on a subject when you are completely wrong. Maybe the reason I have to repeat myself is that my original argument is unavailable now because you’ve thrown your toys out of the pram and turned off your blog. Looks like the reaction of someone who knows they are beaten to me. Accusing me of repeating myself is a great statement from someone who’s spent the last two days repeating their rubbish arguments constantly.

    I think a quick look through the comments on the blog posts around the blogsphere say everything that needs to be said on these subjects. At some point, even the most stubborn and stupid of individuals must realize that they are fighting a losing battle. But then I know you are arrogant enough to think you are right and the whole world is wrong.



  9. Maxim: Looks like you’re correct. The aliases don’t appear to work for non-interactive sessions.

    Hans: I’ve tested your comments and it looks like you are on the money.

    I’ve altered the article. It still includes the warnings, but also my comments contradicting some of the them. Thanks for your help.



  10. tim: I reckon it is good that you are spreading the work about rlwrap and other utilities keep up the good work. As for Howards comments on this and other web pages it just seems like sour grapes. I have read howards article before the web page was taken away and he didn’t acknowledge the inventor of rlwrap as being hans (http://utopia.knoware.nl/~hlub/rlwrap/) so it seems that howard also can’t research!?

  11. Sammy: my articles **linked** to Hans’ website! I also made a specific point of congratulating the website owner on his witty choice of site name (in case you don’t get the joke, “utopia” means “nowhere” and the link was to utopia.knoware.nl). I think linking to the site, expressing appreciation of the site’s owner’s wit and gratitude in abundance for the software in the first place counts as acknowledgement enough. I’m not sure, however, why you are berating me for my lack of “research skills” in any case, since you’re actually commenting on my *attribution* skills or supposed lack thereof. But such nuances are probably lost on someone who demonstrably hasn’t quite mastered the tricky arts of punctuation, spelling and grammar, I suppose.

    And Tim: by my own admission, yes: having a PhD says you USED to be able to do basic research. The current evidence, however, is that you’ve lost that skill and now prefer the less productive approach of merely and unoriginally re-inventing wheels.

    Oh, and PS: it’s not difficult to rank higher than me in Google when I’ve deliberately made sure all links to my work at Google have been removed!

  12. Howard: The argument is old, boring and incorrect. You’re just miffed because you are wrong, and now you’re trying to turn the argument in another direction to try and fool everyone into thinking you aren’t. You are no longer barking on about how you deserve credit, which you don’t, but trying to focus your argument on how much or little research I do. Seems like it’s a good job I wrote m own article, since some publishers choose to yank there work for no reason.

    As for the Google rankings, at the time my comment was made, you stuff was still in the Google index, so you are late to the party with that comment.



Comments are closed.