[phpMyAdmin Developers] about import chunk size

[Hugues Peccatte]

Le mar. 18 août 2015 à 21:33, Dan Ungureanu <udan1107 at gmail.com> a écrit :

On Tue, Aug 18, 2015 at 9:11 PM, Marc Delisle <marc at infomarc.info> wrote:

Le 2015-08-18 12:31, Hugues Peccatte a écrit :
> Le lun. 17 août 2015 à 22:54, Marc Delisle <marc at infomarc.info
> <mailto:marc at infomarc.info>> a écrit :
>
>     Hi,
>     does anyone remember why the import chunk size is 200?
>
>     See https <https
<https://github.com/phpmyadmin/phpmyadmin/issues/11386>://
<https://github.com/phpmyadmin/phpmyadmin/issues/11386>github.com
<https://github.com/phpmyadmin/phpmyadmin/issues/11386>/
<https://github.com/phpmyadmin/phpmyadmin/issues/11386>phpmyadmin
<https://github.com/phpmyadmin/phpmyadmin/issues/11386>/
<https://github.com/phpmyadmin/phpmyadmin/issues/11386>phpmyadmin
<https://github.com/phpmyadmin/phpmyadmin/issues/11386>/issues/11386
<https://github.com/phpmyadmin/phpmyadmin/issues/11386>>://
>
>
> Hi,
>
> I saw this when I rewrote this part, but I've no idea about the reason
> to read only 200 characters...
> I think that we can increase this value.
>
> H.

Hi Hugues,
looking at your comment in
https <https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4caf>://
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4caf>github.com
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4caf>/
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4caf>phpmyadmin
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4caf>/
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4caf>phpmyadmin
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4caf>/commit/2da2c4caf
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4caf>

you wanted to reduce the size of data to read, maybe to improve parsing
speed?

--
Marc Delisle | phpMyAdmin

_______________________________________________
Developers mailing list
Developers at phpmyadmin.net
https <https://lists.phpmyadmin.net/mailman/listinfo/developers>://
<https://lists.phpmyadmin.net/mailman/listinfo/developers>
lists.phpmyadmin.net
<https://lists.phpmyadmin.net/mailman/listinfo/developers>/mailman/
<https://lists.phpmyadmin.net/mailman/listinfo/developers>listinfo
<https://lists.phpmyadmin.net/mailman/listinfo/developers>/developers
<https://lists.phpmyadmin.net/mailman/listinfo/developers>

I asked myself this too, but then I saw this commit [1] and I decided to
leave it this way. I believe that there is no performance loss or gain if
the size will be increased, because the parser continues from where it
left. There is a special case involving DELIMITER statements which may
cause a small rewind, but it's negligible. At least, this is what the new
parser does, I believe the old one did the same.

[1] https <https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4c>://
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4c>github.com
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4c>/
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4c>phpmyadmin
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4c>/
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4c>phpmyadmin
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4c>/commit/2da2c4c
<https://github.com/phpmyadmin/phpmyadmin/commit/2da2c4c>

Best regards,

Dan Ungureanu



Oops... So I would be the owner of this change...
I believe it was because of the multibytes conversion functions that were
too long on "big" string. So I preferred to work on small string, but there
are side effects...

We should check the log history, but if the string functions had been
rewritten after this, we can think about removing this length limitation.

H.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.phpmyadmin.net/pipermail/developers/attachments/20150821/f7b38b5e/attachment.html>