- Nuke empty_string. It is a reminanent from the time where RETURN_FALSE()
used to return "" and not bool(false). It's not worth keeping it because
STR_FREE() and zval_dtor() always have

- Nuke empty_string. It is a reminanent from the time where RETURN_FALSE()
used to return "" and not bool(false). It's not worth keeping it because
STR_FREE() and zval_dtor() always have to check for it and it slows down
the general case. In addition, it seems that empty_string has been abused
quite a lot, and was used not only for setting zval's but generally in
PHP code instead of "", which wasn't the intention. Last but not least,
nuking empty_string should improve stability as I doubt every place
correctly checked if they are not mistakenly erealloc()'ing it or
calling efree() on it.
NOTE: Some code is probably broken. Each extension maintainer should
check and see that my changes are OK. Also, I haven't had time to touch
PECL yet. Will try and do it tomorrow.

show more ...

fix #29012 (Potential race during first connection) thanks to cjbj at hotmail dot com

fix #28978 (multiple OCIBindByName on the same placeholder eats memory)

hmm. this comma is supposed to be a semicolon.

Wordsize fixes

this should really fix compile failure with gcc 2.96

Fix compile failure

Only use Z_TYPE* with zvals..

* email address update

* update my email address

fix bug #27149
(offset should start with 1)

-remove obsolete types
-changes in config.m4 to make oci8 to compile with Oracle 10g
(thanks to cjbj at hotmail dot com for reporting both problems)

add missing alias for ocifreecursor()

fix yet another possible segfault

two fixes:
-oci_lob_load shouldn't reset LOB's length to zero
-Oracle feels bad, but reports no error, when trying to erase LOB with offset bigger than LOB

change to use different allocator (Sascha, thanks =) )

fix this annoying notice about limited range

improve debug messages

oops =(
forgot these brackets..

fix oci_field_type_raw's proto
oci_collection_element_get should return FALSE only if there is no such element and NULL, if it's null

add missing protos
and some minor changes

MFB
oci_lob_save() should be removed indeed, but we're already in feature freeze =(

*change constant name, coz it doesn't comply with others
*add constant's alias, coz others do have aliases
*change protos (oops =))
*add notices to collection functions
*change coll_e

*change constant name, coz it doesn't comply with others
*add constant's alias, coz others do have aliases
*change protos (oops =))
*add notices to collection functions
*change coll_element_get & coll_assign to inform user, that this type of element is not supported yet.

show more ...

fix potential segfault in oci_lob_write()
change oci_lob_trim() - truncate length should be 0 by default.

Fix bug #26892, ORA-21301 in ocinewcollection() call.
pEnv should be initialized in object mode too.