Horton

Thread

  1. psql Week 1

    Peter Eisentraut <peter_e@gmx.net> — 1999-10-05T18:12:29Z

    Here's the summary of my wheelings and dealings. No code yet, but trust
me, you don't want it.

* Changed one file into many
* Changed few functions into countless
* Added GNU long option support (added test to configure.in,
  definition to config.h.in)
  [ This feature is under protest by Tom L. ]
* Extra arguments are now taken as dbname and username
  (it used to fail if you have more than one extra)
* Added switch -V (--version) to display client and server version and
  warn about possible incompatibilities
* Added \copyright command. Changed welcome message accordingly.
* Added a few long slash commands equivalent to short ones (e.g.,
  \list = \l)
* Rewrote backslash command parser from scratch. Can now quote
  arguments. (Only doublequotes. Single quotes to come.)
* Added message output channel as alternative to query output and
  stderr. Might be useful to funnel output in scripts.
* SQL command help is now generated directly from the SGML sources at
  build time.
  [ Must have perl. Might be a problem on Windows. Perhaps package
  preparsed version as well. ]
* \connect now asks for password when appropriate
* Added switch -U allowing you to specify username on cmd line
* -? prints out default username, host, port, etc. in help screen
* PSQLRC env variable can override the name of your startup script
* PSQL_EDITOR can set your prefered editor for \e and \E (overriding
  EDITOR and VISUAL)
* Fixed flat tire on bike ...
* when \connect fails, it now keeps the previous connection (formerly
  aborted program)
* Custom prompts in tcsh style
  (No, I am not partial to tcsh. In fact, I don't even use it. But
  using % as escape character rather than \ saves a lot of headaches.)
* Increased abstraction of input routines.
  [ Cheers to Matthew D.! ]
* Started to clean up \copy. Can now specify delimiters and "with
  oids". Still needs some work though, especially regarding binary and
  quoting/escaping. I'll probably end up writing a better strtok()
  before this is all over.

More in a week . . .

-- 
Peter Eisentraut - peter_e@gmx.net
http://yi.org/peter-e

  2. Re: [HACKERS] psql Week 1

    Oleg Bartunov <oleg@sai.msu.su> — 1999-10-05T21:10:21Z

    Peter,

what  I miss on pgsql is ability to keep commands I issued in history file
as bash does for example. Say, ~/.pgsqlhistory
Of course, not to save passwords.

	Regards,
		Oleg

On Tue, 5 Oct 1999, Peter Eisentraut wrote:

> Date: Tue, 5 Oct 1999 20:12:29 +0200 (CEST)
> From: Peter Eisentraut <peter_e@gmx.net>
> To: pgsql-hackers@postgreSQL.org
> Subject: [HACKERS] psql Week 1
> 
> Here's the summary of my wheelings and dealings. No code yet, but trust
> me, you don't want it.
> 
> * Changed one file into many
> * Changed few functions into countless
> * Added GNU long option support (added test to configure.in,
>   definition to config.h.in)
>   [ This feature is under protest by Tom L. ]
> * Extra arguments are now taken as dbname and username
>   (it used to fail if you have more than one extra)
> * Added switch -V (--version) to display client and server version and
>   warn about possible incompatibilities
> * Added \copyright command. Changed welcome message accordingly.
> * Added a few long slash commands equivalent to short ones (e.g.,
>   \list = \l)
> * Rewrote backslash command parser from scratch. Can now quote
>   arguments. (Only doublequotes. Single quotes to come.)
> * Added message output channel as alternative to query output and
>   stderr. Might be useful to funnel output in scripts.
> * SQL command help is now generated directly from the SGML sources at
>   build time.
>   [ Must have perl. Might be a problem on Windows. Perhaps package
>   preparsed version as well. ]
> * \connect now asks for password when appropriate
> * Added switch -U allowing you to specify username on cmd line
> * -? prints out default username, host, port, etc. in help screen
> * PSQLRC env variable can override the name of your startup script
> * PSQL_EDITOR can set your prefered editor for \e and \E (overriding
>   EDITOR and VISUAL)
> * Fixed flat tire on bike ...
> * when \connect fails, it now keeps the previous connection (formerly
>   aborted program)
> * Custom prompts in tcsh style
>   (No, I am not partial to tcsh. In fact, I don't even use it. But
>   using % as escape character rather than \ saves a lot of headaches.)
> * Increased abstraction of input routines.
>   [ Cheers to Matthew D.! ]
> * Started to clean up \copy. Can now specify delimiters and "with
>   oids". Still needs some work though, especially regarding binary and
>   quoting/escaping. I'll probably end up writing a better strtok()
>   before this is all over.
> 
> More in a week . . .
> 
> -- 
> Peter Eisentraut - peter_e@gmx.net
> http://yi.org/peter-e
> 
> 
> ************
> 

_____________________________________________________________
Oleg Bartunov, sci.researcher, hostmaster of AstroNet,
Sternberg Astronomical Institute, Moscow University (Russia)
Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/
phone: +007(095)939-16-83, +007(095)939-23-83

  3. Re: [HACKERS] psql Week 1

    Bruce Momjian <maillist@candle.pha.pa.us> — 1999-10-05T21:30:37Z

    > Peter,
> 
> what  I miss on pgsql is ability to keep commands I issued in history file
> as bash does for example. Say, ~/.pgsqlhistory
> Of course, not to save passwords.

readline has the capability.  We would just need to enable it.  Good
idea.

-- 
  Bruce Momjian                        |  http://www.op.net/~candle
  maillist@candle.pha.pa.us            |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

  4. Re: [HACKERS] psql Week 1

    Roland Roberts <roberts@panix.com> — 1999-10-06T03:16:30Z

    -----BEGIN PGP SIGNED MESSAGE-----


    Peter> * \connect now asks for password when appropriate

Does this include the initial connect?  I has password authentication
enabled and think it would be nice if psql just prompted me rather
than failed....

roland
- -- 
		       PGP Key ID: 66 BC 3B CD
Roland B. Roberts, PhD                  Custom Software Solutions
roberts@panix.com                      76-15 113th Street, Apt 3B
rbroberts@acm.org                          Forest Hills, NY 11375

-----BEGIN PGP SIGNATURE-----
Version: 2.6.3a
Charset: noconv
Comment: Processed by Mailcrypt 3.5.4, an Emacs/PGP interface

iQCVAwUBN/q/B+oW38lmvDvNAQHUwwP/dLhX5AP05+v1lcpVEzx3gSK+9vWxySfx
lK521D3fsMrWmUQOYn0mqEtLPv/bVUcYgAT+rL3L6dPdvXAHNg1rz64dmZwcsHhy
Erm7GSH4OfCh6msNAhlF0vwgJQams+uRTbYf9AZ3UA6OBgTUCrQ3zR7Q4PSVM9O+
+v2y2Y+FJlo=
=hgOk
-----END PGP SIGNATURE-----

  5. Re: [HACKERS] psql Week 1

    Matthew N. Dodd <winter@jurai.net> — 1999-10-06T04:59:21Z

    On Tue, 5 Oct 1999, Peter Eisentraut wrote:
> * when \connect fails, it now keeps the previous connection (formerly
>   aborted program)

This is bad.

Consider some process that connects to various DBs and does stuff like
drop foo.

If connect fails assume that the user no longer wanted to be connected to
the old connection.

-- 
| Matthew N. Dodd  | '78 Datsun 280Z | '75 Volvo 164E | FreeBSD/NetBSD  |
| winter@jurai.net |       2 x '84 Volvo 245DL        | ix86,sparc,pmax |
| http://www.jurai.net/~winter | This Space For Rent  | ISO8802.5 4ever |

  6. Re: [HACKERS] psql Week 1

    Peter Eisentraut <peter_e@gmx.net> — 1999-10-06T16:26:49Z

    On Oct 6, Oleg Bartunov mentioned:

> what  I miss on pgsql is ability to keep commands I issued in history file
> as bash does for example. Say, ~/.pgsqlhistory

Consider it done. (Because it is done.)

> Of course, not to save passwords.

Usernames and passwords are entered through a different channel. This will
not be a problem.

-- 
Peter Eisentraut - peter_e@gmx.net
http://yi.org/peter-e/

  7. Re: [HACKERS] psql Week 1

    Peter Eisentraut <peter_e@gmx.net> — 1999-10-06T16:29:09Z

    On Oct 6, Matthew N. Dodd mentioned:

> On Tue, 5 Oct 1999, Peter Eisentraut wrote:
> > * when \connect fails, it now keeps the previous connection (formerly
> >   aborted program)
> 
> This is bad.
> 
> Consider some process that connects to various DBs and does stuff like
> drop foo.
> 
> If connect fails assume that the user no longer wanted to be connected to
> the old connection.

I forgot to mention that this only happens in interactive mode for the
very reason you cited. I do not see a problem there.

-- 
Peter Eisentraut - peter_e@gmx.net
http://yi.org/peter-e/