Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-02-27 Thread Heikki Linnakangas
On 27.02.2013 11:38, Etsuro Fujita wrote: Agreed. ISTM that the comment on parse_slash_copy() needs to be changed: * table name can be double-quoted and can have a schema part. * column names can be double-quoted. * filename can be single-quoted like SQL literals. * command can be s

Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-02-27 Thread Heikki Linnakangas
On 27.02.2013 08:32, Amit Kapila wrote: If there is semicolon at end, it takes it into account for creating filename. \copy foo to c:\data\foo.out; This problem was fixed in previous patch, but I think handling of quotes has changed this behavior. This is existing behavior, that creates a file

Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-02-27 Thread Etsuro Fujita
Hi, Thank you for the work! > From: Amit Kapila [mailto:amit.kap...@huawei.com] > On Wednesday, February 27, 2013 12:41 AM Heikki Linnakangas wrote: > > There's one oddity in the psql \copy syntax. This is actually present > > in earlier versions too, but I think we should fix it now that we >

Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-02-26 Thread Amit Kapila
On Wednesday, February 27, 2013 12:41 AM Heikki Linnakangas wrote: > On 22.02.2013 12:43, Etsuro Fujita wrote: > >> 1. "Broken pipe" is not handled in case of psql "\copy" command; > >> Issue are as follows: > >> Following are verified on SuSE-Linux 10.2. > >>

Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-02-26 Thread Heikki Linnakangas
On 22.02.2013 12:43, Etsuro Fujita wrote: 1. "Broken pipe" is not handled in case of psql "\copy" command; Issue are as follows: Following are verified on SuSE-Linux 10.2. 1) psql is exiting when "\COPY xxx TO" command is issued and command/script is not found

Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-02-22 Thread Amit Kapila
On Wednesday, February 20, 2013 5:25 PM Etsuro Fujita wrote: > Hi Amit, > > Thank you for the review. Etsuro-san, you are welcome. > > From: Amit Kapila [mailto:amit.kap...@huawei.com] > > > >> Test case issues: > > >> -- > > >> 1. "Broken pipe" is not handled in case of psql "

Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-02-22 Thread Etsuro Fujita
Hi Amit, Thank you for your careful review! > -Original Message- > From: Amit Kapila [mailto:amit.kap...@huawei.com] > Sent: Friday, February 22, 2013 7:18 PM > To: 'Etsuro Fujita'; 'pgsql-hackers' > Subject: RE: [HACKERS] Review : Add hooks for pre- a

Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-02-20 Thread Etsuro Fujita
Hi Amit, Thank you for the review. > From: Amit Kapila [mailto:amit.kap...@huawei.com] > >> Test case issues: > >> -- > >> 1. "Broken pipe" is not handled in case of psql "\copy" command; > >>     Issue are as follows: > >>         Following are verified on SuSE-Linux 10.2. > >>

Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-01-30 Thread Amit Kapila
On Wednesday, January 23, 2013 5:36 PM Etsuro Fujita wrote: > Hi Amit, > Thank you for your review.  I’ve rebased and updated the patch.  Please find attached the patch. >> Test case issues: >> -- >> 1. "Broken pipe" is not handled in case of psql "\copy" command; >>     Issue

Re: [HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2013-01-23 Thread Etsuro Fujita
Hi Amit, Thank you for your review. I've rebased and updated the patch. Please find attached the patch. > Code Review comments: > - > > 1. Modify the comment in function header of: parse_slash_copy (needs to modify for new syntax) Done. > 2. Comments for fu

[HACKERS] Review : Add hooks for pre- and post-processor executables for COPY and \copy

2012-12-18 Thread Amit Kapila
Basic stuff: - Rebase of Patch is required. - Compiles cleanly without any errors/warnings - Regression tests pass. What it does: - This patch is useful when COPY command input/output are stored in compression format or in any