silent fail of revDataFromQuery() in standalone; works in IDE

[Dr. Hawkins]

On Sat, May 24, 2014 at 1:08 AM, Peter Haworth <pete at lcsql.com> wrote:

> What do you get in theRes  when it works? CREATE shouldn't return any data.
>

empty, unless there's  an error

I always use revExecuteSQL for any SQL statement other than SELECT,
>

I've found it easier to just use the one form.  I also have a
process/wrapper that I use for non-memory commands.

> although if revDataFromQuery works in the IDE  it should work in the
> standalone.
>

It has for two years . . . I


On Sat, May 24, 2014 at 1:23 AM, Pierre Sahores <sc at sahores-conseil.com>wrote:

> I don’t know how the below value are set in LC-Server preferences (PHP
> values shown below) :
>
> max_execution_time      30      30
> ...
> but i expect your script could die because it reaches one of the
> corresponding
>

Creating a simple table in an in memory database?  That can only be a few
milliseconds, if that.


On Sat, May 24, 2014 at 7:17 AM, Mike Bonner <bonnmike at gmail.com> wrote:

> I wonder if the compiler is for some reason not including the sqlite
> driver.
>

It's already opened the databases a this point, though.


>  MIght look at the standalone settings and manually include the
> sqlite drivers and rebuild.  Or put a try/catch around the revdatafromquery
> and see if you can get an error message out.
>

try/catch set me on the trail.  the database for the cacheDb wasn't a
number.  And by backtracing down about three levels (put in ck points,
compile run, repeat), I finally found the culprit.

I have a test for whether it's in the IDE or not, and the IDE  branch was
building an array to measure heights--so when I moved the normal place to
build it . . .

But  I still don't think that bombing out in a funciton with no error
message returned is acceptable . . .

-- 
Dr. Richard E. Hawkins, Esq.
(702) 508-8462