Domingo Alvarez Duarte wrote:
> it seems that the json* functions ignore the collation completely,
> is that by design or is it a forgotten implementation ?
The JSON functions use JSON rules.
RFC 7159 section 8.3 does not mention collations.
Regards,
Clemens
ith
zero or more entries for each record to allow quick searches.
Cheers !
?
> Fri Sep 11 2015 5:58:36 pm CEST CEST from "Richard Hipp"
>Subject: [sqlite] Feedback request: JSON support in SQLite
>
> Draft documentation for the current design of JSON support in SQL
.b due to collation nocase here
___
Cheers !
> Fri Sep 11 2015 5:58:36 pm CEST CEST from "Richard Hipp"
>Subject: [sqlite] Feedback request: JSON support in SQLite
>
> Draft documentation for the current design of JSON support in SQLite
> can be seen on
Hello !
That's what json_extract does !
json_extract('[2,3,4,5,6]', '$[3]'); --> returns 5;
Cheers !
> Fri Sep 11 2015 8:40:36 pm CEST CEST from "Nelson, Erik - 2"
> Subject: Re: [sqlite] Feedback request:
>JSON support in SQLite
>
> Richard Hipp wrote o
On 2015-09-11 07:50 PM, Keith Christian wrote:
> Pardon me, but: At what point does the code required for the
> inclusion of a multitude of supported data formats exceed the core
> purpose of the executable?
>
> At some point, obtaining a desired output format (from the potentially
> dozens
parameter from the sqlite3_context
?
The data to be returned by these two new functions is alread on the
sqlite3_context structure.
Cheers !
?
>
>
>
> Fri Sep 11 2015 5:58:36 pm CEST CEST from "Richard Hipp"
>Subject: [sqlite] Feedback request: JSON sup
On Fri, Sep 11, 2015 at 5:58 PM, Richard Hipp wrote:
> Your feedback is encouraged.
>
https://www.sqlite.org/draft/c3ref/value_subtype.html does not say what
happens or which value we get,
should one call sqlite3_value_subtype on a sqlite_value* which which
no sqlite3_result_subtype()
was made.
On Fri, Sep 11, 2015 at 5:58 PM, Richard Hipp wrote:
> Draft documentation for the current design of JSON support in SQLite [...]
Your feedback is encouraged.
>
"Experiments have so far been unable to find a binary encoding that is
significantly smaller or faster than a plain text encoding"
> On Sep 11, 2015, at 6:31 PM, Richard Hipp wrote:
>
> a passing fad
On the other hand, we can now embrace NormalFormZero without undue
embarrassment. Swell.
Richard Hipp wrote on Friday, September 11, 2015 11:59 AM
>
> Draft documentation for the current design of JSON support in SQLite
> can be seen on-line at
>
> https://www.sqlite.org/draft/json1.html
This looks really good!
With the understanding that json_extract() already has the
> On Sep 11, 2015, at 5:58 PM, Richard Hipp wrote:
>
> JSON support in SQLite
JSON, eh? No MERGE. No analytics. But serialization of the week is covered.
Sweet :D
On Fri, Sep 11, 2015 at 8:58 AM, Richard Hipp wrote:
> Draft documentation for the current design of JSON support in SQLite
> can be seen on-line at
>
> https://www.sqlite.org/draft/json1.html
>
> Your feedback is encouraged.
>
> All features described in the document above are implemented
On 9/11/15, Dominique Devienne wrote:
> On Fri, Sep 11, 2015 at 5:58 PM, Richard Hipp wrote:
>
>> Your feedback is encouraged.
>>
>
> https://www.sqlite.org/draft/c3ref/value_subtype.html does not say what
> happens or which value we get,
> should one call sqlite3_value_subtype on a
On 9/11/15, Petite Abeille wrote:
>
> we can now embrace NormalFormZero without undue
> embarrassment.
Well put. :-)
--
D. Richard Hipp
drh at sqlite.org
On 9/11/15, Petite Abeille wrote:
> serialization of the week
The json.org website has been up since 2002. JSON itself predates
that. It is roughly the same age as SQLite itself and is older than
SQLite3. I'm thinking that maybe JSON is not just a passing fad.
Could be wrong though.
--
Draft documentation for the current design of JSON support in SQLite
can be seen on-line at
https://www.sqlite.org/draft/json1.html
Your feedback is encouraged.
All features described in the document above are implemented and
working in the latest trunk version of SQLite, which you can
Pardon me, but: At what point does the code required for the
inclusion of a multitude of supported data formats exceed the core
purpose of the executable?
At some point, obtaining a desired output format (from the potentially
dozens available) might be offloaded to a different executable that
On Fri, Sep 11, 2015 at 11:31 AM, Richard Hipp wrote:
> On 9/11/15, Petite Abeille wrote:
> > serialization of the week
>
> The json.org website has been up since 2002. JSON itself predates
> that. It is roughly the same age as SQLite itself and is older than
> SQLite3. I'm thinking that
On 2015-09-11 9:31 AM, Richard Hipp wrote:
> On 9/11/15, Petite Abeille wrote:
>> serialization of the week
>
> The json.org website has been up since 2002. JSON itself predates
> that. It is roughly the same age as SQLite itself and is older than
> SQLite3. I'm thinking that maybe JSON is
19 matches
Mail list logo
