Increase the maximum possible value of the page size

[petterreinholdtsen]

Firefox/Iceweasel uses a page size of 32k, but sqlite won't allow it because the maximum it allows by default is 8k (through the SQLITE_MAX_DEFAULT_PAGE_SIZE define). This patch increases this limit.

Patch by Laszlo Boszormenyi (GCS) gcs@debian.org to solve Debian bug http://bugs.debian.org/672573 .

Patch has been used by Debian package for a while, see https://udd.debian.org/patches.cgi?src=sqlcipher&version=3.4.1-2 .

[sjlombardo]

Hello @petterreinholdtsen I have reviewed the description of this change and I don't think that this is applicable at all. The SQLITE_MAX_DEFAULT_PAGE_SIZE is only used to determine the max size that SQLite would be used by default based on device characteristics:

/*
** Ordinarily, if no value is explicitly provided, SQLite creates databases
** with page size SQLITE_DEFAULT_PAGE_SIZE. However, based on certain
** device characteristics (sector-size and atomic write() support),
** SQLite may choose a larger value. This constant is the maximum value
** SQLite will choose on its own.
*/

Having this set to 8K would not impact the ability of Firefox/Iceweasel to use a larger page size. SQLITE_MAX_DEFAULT_PAGE_SIZE doesn't prevent an application from explicitly selecting a large page size. The applicable maximum page size (SQLITE_MAX_PAGE_SIZE) is already set to 65536.

Furthermore, unless I am mistaken, Firefox is already using both SQLite and SQLCipher without these patches on other platforms.

I understand this patch has been used for some time but I believe it is misguided and wouldn't do what is purported.

[petterreinholdtsen]

[Stephen Lombardo]

Furthermore, unless I am mistaken, Firefox is already using both SQLite and SQLCipher without these patches on other platforms. I understand this patch has been used for some time but I believe it is misguided and wouldn't do what is purported.

The rationale behind the patch is in <URL: https://bugs.debian.org/672573 > from 2012. I do not know why it was created or if it worked at all.

…

-- Happy hacking Peter Reinholdtsen

[sjlombardo]

Hello @petterreinholdtsen - I reviewed the history and rationale of that change. In my opinion it would be best to remove that patch. It will not work as intended.