Re:Psql-describe: \d ported to JavaScript
by gmac on Wednesday May 08, 2024 @02:21PM
(#40301349)
OP (and library author) here. This is a fair point, and one I freely admit to in the accompanying blog post. It’s probably not crucial to sync more than once per major Postgres version, but we’re still figuring out how we’ll deal with this.
Re:Psql-describe: \d ported to JavaScript
by williamjackson on Wednesday May 08, 2024 @01:47PM
(#40300964)
The fact that this is not a manual reimplementation, but a "transpilation" of the psql source from C to JavaScript, makes this super interesting to me.
From the Postgres master branch (17devel), we take exec_command_d, exec_command_list and exec_command_sf_sv from command.c, and all of describe.c and sql_help.c, from src/bin/psql.
We use plenty of RegExp search-and-replace to turn this C code into valid JS syntax.
We implement some C library functions, such as strlen and strchr, and some Postgres support functions, such as printTable and printQuery, in JavaScript.
Re:Psql-describe: \d ported to JavaScript
by ivanjermakov on Wednesday May 08, 2024 @03:07PM
(#40301781)
When I started my programming journey I thought I could just "translate" Pascal to C++ just with search/replace and minor adjustments. Turns out sometimes it's the best approach :)
Re:Psql-describe: \d ported to JavaScript
by arp242 on Wednesday May 08, 2024 @01:56PM
(#40301062)
I always thought that most \-commands in psql were just frontends for queries? For example:
% psql db_name
(1)=# \set ECHO_HIDDEN
(1)=# \d
********* QUERY **********
SELECT n.nspname as "Schema",
c.relname as "Name",
CASE c.relkind WHEN 'r' THEN 'table' WHEN 'v' THEN 'view' WHEN 'm' THEN 'materialized view' WHEN 'i' THEN 'index' WHEN 'S' TH
EN 'sequence' WHEN 't' THEN 'TOAST table' WHEN 'f' THEN 'foreign table' WHEN 'p' THEN 'partitioned table' WHEN 'I' THEN 'partit
ioned index' END as "Type",
pg_catalog.pg_get_userbyid(c.relowner) as "Owner"
FROM pg_catalog.pg_class c
LEFT JOIN pg_catalog.pg_namespace n ON n.oid = c.relnamespace
LEFT JOIN pg_catalog.pg_am am ON am.oid = c.relam
WHERE c.relkind IN ('r','p','v','m','S','f','')
AND n.nspname <> 'pg_catalog'
AND n.nspname !~ '^pg_toast'
AND n.nspname <> 'information_schema'
AND pg_catalog.pg_table_is_visible(c.oid)
ORDER BY 1,2;
**************************
List of relations
Schema │ Name │ Type │ Owner
────────┼────────────────────────────────────────┼──────────┼────────
...
I re-implemented quite a bit of that in a PostgreSQL management tool I built, based just on the ECHO_HIDDEN queries. Never even had to look at any C code.
I'm not sure if I entirely follow what translating C to JS gives you, other than a bit of (fairly simple) parsing of \cmd and flags? I didn't look too carefully at the code, so maybe I'm missing something.
Re:Psql-describe: \d ported to JavaScript
by arp242 on Wednesday May 08, 2024 @04:47PM
(#40302578)
Almost every single post you make mentions clickhouse. You've replied to a few of my posts now with this vaguely on-topic forceful injection of clickhouse.
Of course I'm not saying you can never bring up clickhouse, but at this point it's just becoming spam. High-effort spam, maybe. But still spam.
Re:Psql-describe: \d ported to JavaScript
by arp242 on Wednesday May 08, 2024 @06:43PM
(#40303466)
I am aware. But this thread is not about Clickhouse. And your comment contributed nothing to the general conversation either. The general pattern is really not appropriate.
Re:Psql-describe: \d ported to JavaScript
by arp242 on Wednesday May 08, 2024 @07:39PM
(#40303826)
When you keep "comparing" tons of stuff to the project you wrote then you're just spamming. Like I said, this is hardly the first time I've seen it, and that's just from random posts I've seen.
Re:Psql-describe: \d ported to JavaScript
by zX41ZdbW on Friday May 10, 2024 @12:17PM
(#40320703)
Sorry for distracting you. My apologies.
I will try to remember your login and not answer your comments in the future.
At the same time, it is possible that I will occasionally answer your comments by mistake.
Re:Psql-describe: \d ported to JavaScript
by gmac on Wednesday May 08, 2024 @02:25PM
(#40301378)
Some commands issue multiple queries and include logic for stitching the results into a nice table. There’s not a one-to-one mapping from commands to queries by any means.
Re:Psql-describe: \d ported to JavaScript
by arp242 on Wednesday May 08, 2024 @02:30PM
(#40301428)
Yeah, that's right; but the "glue code" I've had to write for this has been really minimal. Certainly less effort than translating C to JavaScript with regexps (that on its own is something of an accomplishment). I didn't port everything so maybe I missed some more complex bits.
I mean, if this works well for you then that's grand; just wanted to point out to others that \-commands are typically very "non-magic" and you don't need to go the route of translating C code.
Re:Psql-describe: \d ported to JavaScript
by stuaxo on Thursday May 09, 2024 @04:37AM
(#40306343)
Thw convertet is nice, I could imagine it being it's own library, with the generator, and a release including the generated code every postures release.
You might well never need to use it, but if you’re building an online query editor it could be very handy!
I'm not sure if I entirely follow what translating C to JS gives you, other than a bit of (fairly simple) parsing of \cmd and flags? I didn't look too carefully at the code, so maybe I'm missing something.
If you want to support all of them, including handling all the different Postgres versions, it becomes quite a lot to implement.
I personally don't like abbreviations and prefer full queries.
\-commands look like hacks and smell like bad practice to me. It is something from the past.
Similarly, I prefer always to write --a-long-command-line-parameter-name in my scripts instead of -wTf.
But we added shorthand commands to ClickHouse for convenience, to please Postgres users.
Of course I'm not saying you can never bring up clickhouse, but at this point it's just becoming spam. High-effort spam, maybe. But still spam.
Some have \-commands, and some have .-commands, which I found hacky - because it is a micro language notation in addition to the actual language.
I will try to remember your login and not answer your comments in the future. At the same time, it is possible that I will occasionally answer your comments by mistake.
Cheers, Alexey.
I mean, if this works well for you then that's grand; just wanted to point out to others that \-commands are typically very "non-magic" and you don't need to go the route of translating C code.