Huh, fascinating stuff. I'm not an sqlite developer but I can shed light on
some of your questions.
On Tue, 12 Feb 2019 at 09:54, Edwin Török wrote:
> A very conservative interpretation of various fsync bugs in various OS
> kernels [2][5] would suggest that:
>
> #1. the list of known OS issues
This is mainly for my curiosity. Is there any particular reason that one
can't add a unique column on an alter table? With a default value of null
they would all have unique values by default.
Any insight into this would be great. Perhaps there' something obvious I'm
missing.
-- Mark
Hey guys!
According to "Persistent Loadable Extensions” topic on
https://www.sqlite.org/loadext.html if the initialization procedure returns
SQLITE_OK_LOAD_PERMANENTLY (256) the extension should persist on the database
file instead of just belong to the current connection.
So I’ve downloaded
Hi,
I was wondering what changes SQLite3 would need in light of the fsync
problems discovered by the PostgreSQL community (see "How is it
possible that PostgreSQL used fsync incorrectly for 20 years, and what
we'll do about it" talk [1]).
[2] lists that MySQL and MongoDB did some changes, has
Ah yes. I forgot about that. I don't use WITHOUT ROWID typically, so it
didn't enter the thought process.
On Mon, Feb 11, 2019 at 2:29 PM J. King wrote:
> On February 11, 2019 2:19:27 PM EST, Stephen Chrzanowski <
> pontia...@gmail.com> wrote:
> >This is a dump from a MySQL table I created a
On February 11, 2019 2:19:27 PM EST, Stephen Chrzanowski
wrote:
>This is a dump from a MySQL table I created a few years ago. I'm not
>moving this particular database into SQLite, but, from what I learned
>today
>about MySQL dumps and the commenting system, I was kind of interested
>on
>how
This is a dump from a MySQL table I created a few years ago. I'm not
moving this particular database into SQLite, but, from what I learned today
about MySQL dumps and the commenting system, I was kind of interested on
how SQLite would handle the rest of the following statement:
CREATE TABLE
Simon Slavin, on Monday, February 11, 2019 11:53 AM, wrote...
>On 11 Feb 2019, at 4:51pm, Jose Isaias Cabrera wrote:
>
>> Thanks. Yes, sometimes I should revise 3 or 4 times before asking. :-) As
>> soon as I sent it, I figured it out.
>
>It wouldn't help. You can read it ten time, rewrite it
On 11 Feb 2019, at 4:51pm, Jose Isaias Cabrera wrote:
> Thanks. Yes, sometimes I should revise 3 or 4 times before asking. :-) As
> soon as I sent it, I figured it out.
It wouldn't help. You can read it ten time, rewrite it three times, and
explain it to your pet duck. You still won't
J. King, on Monday, February 11, 2019 11:25 AM, wrote...
>On February 11, 2019 11:16:32 AM EST, Jose Isaias Cabrera
>wrote:
>>
>>This SQL,
>>
>>SELECT a.*,b.*,c.Area,d.Bus_Area FROM Master_Project_List AS a
>>LEFT JOIN Master_Project_List_Extra AS b ON a.ProjID = b.ProjID
>>LEFT
On February 11, 2019 11:16:32 AM EST, Jose Isaias Cabrera
wrote:
>
>This SQL,
>
>SELECT a.*,b.*,c.Area,d.Bus_Area FROM Master_Project_List AS a
>LEFT JOIN Master_Project_List_Extra AS b ON a.ProjID = b.ProjID
>LEFT JOIN Bus_IT_Areas_ORGs AS c ON a.IT_OBS = c.IT_OBS
>LEFT JOIN
Never mind. Sorry guys for the wasted bandwidth.
From: sqlite-users on behalf of
Jose Isaias Cabrera
Sent: Monday, February 11, 2019 11:16 AM
To: sqlite-users@mailinglists.sqlite.org
Subject: [sqlite] Why Error: ambiguous column name: ProjID
This SQL,
This SQL,
SELECT a.*,b.*,c.Area,d.Bus_Area FROM Master_Project_List AS a
LEFT JOIN Master_Project_List_Extra AS b ON a.ProjID = b.ProjID
LEFT JOIN Bus_IT_Areas_ORGs AS c ON a.IT_OBS = c.IT_OBS
LEFT JOIN Business_OBS_List AS d ON a.Business_OBS = d.Bus_OBS
WHERE ProjID IN
Dominique, what I said was that it is undefined behaviour in C++ to return
a *value* in a void function. That is still true.
On Mon, Feb 11, 2019 at 2:49 PM Dominique Devienne
wrote:
> On Mon, Feb 11, 2019 at 1:11 PM Clemens Ladisch
> wrote:
>
> > Peter da Silva wrote:
> > > I am pretty sure
On 2/11/19, J. King wrote:
>
> Rowids are signed integers and can be negative. The documentation does
> advise (somewhere...) against using negative rowids because they are larger
> (when stored) than the typical used range of positive ones, but that's it.
Exactly. Negative rowids work fine.
On Mon, Feb 11, 2019 at 1:11 PM Clemens Ladisch wrote:
> Peter da Silva wrote:
> > I am pretty sure that the code is not legal C
>
> Indeed; C99 and C11 say in 6.3.2.2:
> | The (nonexistent) value of a void expression (an expression that has
> | type void) shall not be used in any way [...]
>
On February 11, 2019 8:35:57 AM EST, John Smith wrote:
>Hi,
>
>
>I read in SQLite documentation that if I define column of type INTEGER
>PRIMARY KEY then this column will become an alias to SQLite internal
>64-bit integer index that uniquely identifies the row (hence ‘rowid’).
>
>I also read that
Hi,
I read in SQLite documentation that if I define column of type INTEGER PRIMARY
KEY then this column will become an alias to SQLite internal 64-bit integer
index that uniquely identifies the row (hence ‘rowid’).
I also read that the initial default value that will be used for such column
On 9/2/62 03:31, Dominique Pellé wrote:
David Raymond wrote:
SQLite version 3.27.1 is now available on the SQLite website:
https://sqlite.org/
https://sqlite.org/download.html
https://sqlite.org/releaselog/3_27_1.html
Release notes https://sqlite.org/releaselog/3_27_1.html say:
Peter da Silva wrote:
> I am pretty sure that the code is not legal C
Indeed; C99 and C11 say in 6.3.2.2:
| The (nonexistent) value of a void expression (an expression that has
| type void) shall not be used in any way [...]
and in 6.8.6.4:
| A return statement with an expression shall not appear
> On Feb 11, 2019, at 6:33 AM, Peter da Silva wrote:
>
> I am pretty sure that the code is not legal C because it's using the return
> value of a void function, as well as returning a value from a void
> function. Compilers that "do what I mean" and accept it are in error. It's
> certainly
I am pretty sure that the code is not legal C because it's using the return
value of a void function, as well as returning a value from a void
function. Compilers that "do what I mean" and accept it are in error. It's
certainly possible that some obscure clause in some C standard blesses it
but I
I'm using cl.exe v19.12.
To summarize:
SQLite 3.24 compiles fine with warning as error enabled with cl.exe v19.12
SQLite 3.27.1 does not compile with warning as error enabled with cl.exe
v19.12.
To me, it looks like a simple fix to avoid writing "return " in void
functions even if is a void
On Mon, Feb 11, 2019 at 11:31 AM Jonas Bülow wrote:
> Sorry, I missed some information. It is the MSVC v15.5 compiler that
> complains:
>
> sqlite3.c(58167): error C2220: warning treated as error - no 'object' file
> generated [c:\work\sqlite-amalgamation-3270100\sqlite3.vcxproj]
>
Sorry, I missed some information. It is the MSVC v15.5 compiler that
complains:
sqlite3.c(58167): error C2220: warning treated as error - no 'object' file
generated [c:\work\sqlite-amalgamation-3270100\sqlite3.vcxproj]
sqlite3.c(58167): warning C4098: 'sqlite3PagerSnapshotUnlock': 'void'
function
On Mon, Feb 11, 2019 at 11:16 AM Jonas Bülow wrote:
> > Just tried to update my sqlite version from 3.24 to 3.27.1 and the
> > compiler complained about a void function returning a value. I don't know
> > about C, but in C++ this is undefined behaviour and the clang compiler
> > sometimes
The subject should say "3.27.1". Sorry!
On Mon, Feb 11, 2019 at 11:14 AM Jonas Bülow wrote:
> Just tried to update my sqlite version from 3.24 to 3.27.1 and the
> compiler complained about a void function returning a value. I don't know
> about C, but in C++ this is undefined behaviour and the
Just tried to update my sqlite version from 3.24 to 3.27.1 and the compiler
complained about a void function returning a value. I don't know about C,
but in C++ this is undefined behaviour and the clang compiler sometimes
generate an ud2 instruction for such code.
It's on line 58165 in
Ignore this.
Was mailed over 1w ago and only came through and I have figured this out
after
studying the CTE documentation on the SQLite site.
RBS
On Wed, Feb 6, 2019 at 7:24 PM Bart Smissaert
wrote:
>
> I can select the rank as in the previous e-mail with this recursive query:
>
> with
29 matches
Mail list logo