Thread: tsearch core path, v0.58
http://www.sigaev.ru/misc/tsearch_core-0.58.gz Changes since 0.52 version: 1) Introduce dictionary's template which contains only methods of dictionary and can be managed only by superuser. CREATE TEXT SEARCH DICTIONARY dictname TEMPLATE dicttmplname [OPTION opt_text ] ; CREATE TEXT SEARCH DICTIONARY TEMPLATE dicttmplname LEXIZE lexize_function [INIT init_function ] ; DROP TEXT SEARCH DICTIONARY TEMPLATE [IF EXISTS] dicttmplname [CASCADE] ALTER TEXT SEARCH DICTIONARY TEMPLATE dicttmplname RENAME TO newname; psql has \dFt command operated templates 2) parser and dictionary template could be managed only by superuser (due to security reasons pointed by Tom). So, they don't have owner columns and removed ALTER .. PARSER .. OWNER TO command 4) As Bruce suggests, GUC variable tsearch_conf_name is renamed to default_text_search_config and trigger tsearch is renamed to tsvector_update_trigger 5) remove cfglocale and cfgdefault columns in configuration. So, CREATE/ALTER .. CONFIGURATION hasn't AS DEFAULT and LOCALE options. Instead of that initdb tries to find suitable configuration name for selected locale. Or it uses -T, --text-search-config=CFG switch. 6) pg_dump, psql are changed accordingly. -- Teodor Sigaev E-mail: teodor@sigaev.ru WWW: http://www.sigaev.ru/
Your patch has been added to the PostgreSQL unapplied patches list at:
http://momjian.postgresql.org/cgi-bin/pgpatches
It will be applied as soon as one of the PostgreSQL committers reviews
and approves it.
---------------------------------------------------------------------------
Teodor Sigaev wrote:
> http://www.sigaev.ru/misc/tsearch_core-0.58.gz
>
> Changes since 0.52 version:
>
> 1) Introduce dictionary's template which contains only methods of dictionary and
> can be managed only by superuser.
> CREATE TEXT SEARCH DICTIONARY dictname
> TEMPLATE dicttmplname
> [OPTION opt_text ]
> ;
>
> CREATE TEXT SEARCH DICTIONARY TEMPLATE dicttmplname
> LEXIZE lexize_function
> [INIT init_function ]
> ;
>
> DROP TEXT SEARCH DICTIONARY TEMPLATE [IF EXISTS] dicttmplname [CASCADE]
> ALTER TEXT SEARCH DICTIONARY TEMPLATE dicttmplname RENAME TO newname;
>
> psql has \dFt command operated templates
>
> 2) parser and dictionary template could be managed only by superuser (due to
> security reasons pointed by Tom). So, they don't have owner columns and removed
> ALTER .. PARSER .. OWNER TO command
>
> 4) As Bruce suggests, GUC variable tsearch_conf_name is renamed to
> default_text_search_config and trigger tsearch is renamed to tsvector_update_trigger
>
> 5) remove cfglocale and cfgdefault columns in configuration. So, CREATE/ALTER ..
> CONFIGURATION hasn't AS DEFAULT and LOCALE options. Instead of that initdb tries
> to find suitable configuration name for selected locale. Or it uses -T,
> --text-search-config=CFG switch.
>
> 6) pg_dump, psql are changed accordingly.
>
>
> --
> Teodor Sigaev E-mail: teodor@sigaev.ru
> WWW: http://www.sigaev.ru/
>
> ---------------------------(end of broadcast)---------------------------
> TIP 1: if posting/reading through Usenet, please send an appropriate
> subscribe-nomail command to majordomo@postgresql.org so that your
> message can get through to the mailing list cleanly
--
Bruce Momjian <bruce@momjian.us> http://momjian.us
EnterpriseDB http://www.enterprisedb.com
+ If your life is a hard drive, Christ can be your backup. +
I did my first minimal review of this patch. First, it is massive ---- 26k line diff for the new commands and functionality, and 31k for adding snowball stemmer. I am glad Oleg and Teodor wrote this because they have been around this code for a while and are available to fix any problems we find. The patch looks well structured. The majority is just standard glue to add new commands, like CREATE/DROP, grammar, system catalogs, pg_dump, cache entries, regression tests, etc. --------------------------------------------------------------------------- Teodor Sigaev wrote: > http://www.sigaev.ru/misc/tsearch_core-0.58.gz > > Changes since 0.52 version: > > 1) Introduce dictionary's template which contains only methods of dictionary and > can be managed only by superuser. > CREATE TEXT SEARCH DICTIONARY dictname > TEMPLATE dicttmplname > [OPTION opt_text ] > ; > > CREATE TEXT SEARCH DICTIONARY TEMPLATE dicttmplname > LEXIZE lexize_function > [INIT init_function ] > ; > > DROP TEXT SEARCH DICTIONARY TEMPLATE [IF EXISTS] dicttmplname [CASCADE] > ALTER TEXT SEARCH DICTIONARY TEMPLATE dicttmplname RENAME TO newname; > > psql has \dFt command operated templates > > 2) parser and dictionary template could be managed only by superuser (due to > security reasons pointed by Tom). So, they don't have owner columns and removed > ALTER .. PARSER .. OWNER TO command > > 4) As Bruce suggests, GUC variable tsearch_conf_name is renamed to > default_text_search_config and trigger tsearch is renamed to tsvector_update_trigger > > 5) remove cfglocale and cfgdefault columns in configuration. So, CREATE/ALTER .. > CONFIGURATION hasn't AS DEFAULT and LOCALE options. Instead of that initdb tries > to find suitable configuration name for selected locale. Or it uses -T, > --text-search-config=CFG switch. > > 6) pg_dump, psql are changed accordingly. > > > -- > Teodor Sigaev E-mail: teodor@sigaev.ru > WWW: http://www.sigaev.ru/ > > ---------------------------(end of broadcast)--------------------------- > TIP 1: if posting/reading through Usenet, please send an appropriate > subscribe-nomail command to majordomo@postgresql.org so that your > message can get through to the mailing list cleanly -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://www.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Teodor Sigaev wrote: > http://www.sigaev.ru/misc/tsearch_core-0.58.gz > > Changes since 0.52 version: > > 1) Introduce dictionary's template which contains only methods of dictionary and > can be managed only by superuser. > CREATE TEXT SEARCH DICTIONARY dictname > TEMPLATE dicttmplname > [OPTION opt_text ] > ; > > CREATE TEXT SEARCH DICTIONARY TEMPLATE dicttmplname > LEXIZE lexize_function > [INIT init_function ] > ; I am finding the above syntax confusing. If TEMPLATE appears before the dictionary name, it is a template, but after, it is using a template. Can we use a different word instead of TEMPLATE, and have a USING clause to reference the template? -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://www.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Teodor Sigaev <teodor@sigaev.ru> writes: > http://www.sigaev.ru/misc/tsearch_core-0.58.gz What is src/backend/utils/tsearch/dict_ispell/parse.h ? Well, I know what it *is*: it's bison output. The question is what is it doing here? It doesn't seem to be used, and if it is used then I do not see the bison grammar file it's made from. BTW, I would like to shorten some of the path names in this fileset. Is there a reason not to combine src/backend/utils/adt/tsearch, src/backend/utils/tsearch, and src/backend/utils/tsearch/dict_ispell into one place, perhaps src/backend/tsearch? regards, tom lane
"Tom Lane" <tgl@sss.pgh.pa.us> writes: > BTW, I would like to shorten some of the path names in this fileset. > Is there a reason not to combine src/backend/utils/adt/tsearch, > src/backend/utils/tsearch, and src/backend/utils/tsearch/dict_ispell > into one place, perhaps src/backend/tsearch? I for one would like to see all the data types in a file in adt. Even if it's just a single file containing mostly just glue functions into code in sc/backend/tsearch. -- Gregory Stark EnterpriseDB http://www.enterprisedb.com
Gregory Stark <stark@enterprisedb.com> writes:
> "Tom Lane" <tgl@sss.pgh.pa.us> writes:
>> BTW, I would like to shorten some of the path names in this fileset.
>> Is there a reason not to combine src/backend/utils/adt/tsearch,
>> src/backend/utils/tsearch, and src/backend/utils/tsearch/dict_ispell
>> into one place, perhaps src/backend/tsearch?
> I for one would like to see all the data types in a file in adt. Even if it's
> just a single file containing mostly just glue functions into code in
> sc/backend/tsearch.
The existing file layout doesn't meet that expectation either, since all
the functions were for some reason dumped into a *subdirectory* of adt.
That was where my too-weird, too-much-typing reflex kicked in ...
regards, tom lane
Tom Lane wrote: > Teodor Sigaev <teodor@sigaev.ru> writes: >> http://www.sigaev.ru/misc/tsearch_core-0.58.gz > > What is src/backend/utils/tsearch/dict_ispell/parse.h ? oops - that is unused file after some experiments. Will be removed from patch. > > Well, I know what it *is*: it's bison output. The question is what is > it doing here? It doesn't seem to be used, and if it is used then I > do not see the bison grammar file it's made from. > > BTW, I would like to shorten some of the path names in this fileset. > Is there a reason not to combine src/backend/utils/adt/tsearch, > src/backend/utils/tsearch, and src/backend/utils/tsearch/dict_ispell > into one place, perhaps src/backend/tsearch? There is no any strong reason - src/backend/utils/tsearch contains fucntions about processing text, they are rather complex. -- Teodor Sigaev E-mail: teodor@sigaev.ru WWW: http://www.sigaev.ru/