Re: improvements in Unicode tables generation code

From: Peter Eisentraut <peter(dot)eisentraut(at)enterprisedb(dot)com>
To: Heikki Linnakangas <hlinnaka(at)iki(dot)fi>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: improvements in Unicode tables generation code
Date: 2021-07-20 11:57:16
Message-ID: 18e34084-aab1-1b4c-edd1-c4f9fb04f714@enterprisedb.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 23.06.21 10:55, Peter Eisentraut wrote:
>>> v1-0001-Make-Unicode-makefile-more-parallel-safe.patch
>>>
>>> The makefile rule that calls UCS_to_most.pl was written incorrectly for
>>> parallel make.  The script writes all output files in one go, but the
>>> rule as written would call the command once for each output file in
>>> parallel.
>>
>> This could use a comment. At a quick glance, I don't understand what
>> all the $(wordlist ...) magic does.
>>
>> Perhaps we should change the script or Makefile so that it doesn't
>> create all the maps in one go?
>
> I agree, either comment it better or just write one file at a time. I'll
> take another look at that.

Here is a patch that does it one file (pair) at a time. The other rules
besides UCS_to_most.pl actually had the same problem, since they produce
two output files, so running in parallel called each script twice. In
this patch, all of that is heavily refactored and works correctly now.
Note that UCS_to_most.pl already accepted a command-line argument to
specify which encoding to work with.

Attachment Content-Type Size
0001-Make-Unicode-makefile-parallel-safe.patch text/plain 6.8 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Amit Kapila 2021-07-20 12:00:29 Re: Next Steps with Hash Indexes
Previous Message Greg Nancarrow 2021-07-20 11:42:50 Re: row filtering for logical replication