[R-sig-DB] [PATCH] segfault in RSQLite 0.5-4

Seth Falcon @|@|con @end|ng |rom |hcrc@org
Wed Jul 18 00:28:06 CEST 2007


Seth Falcon <sfalcon using fhcrc.org> writes:

> "Ashish Kulkarni" <ashish.kulkarni using kalyptorisk.com> writes:
>
>> Now that you've upgraded to 3.4.0, I have some patches which 
>> use the sqlite3_prepare_v2 API, which is recommended now: see 
>>
>> http://www.sqlite.org/capi3ref.html#sqlite3_prepare_v2
>>
>> SQLITE_SCHEMA handling is done internally in that case, and
>> the legacy behavior that made corrected_sqlite3_step() necessary
>> is no longer needed.
>>
>> I've attached patches against both the trunk and the 2.5 branch.
>
> Thanks, I will take a look later this week.

Had a quick look and I have few comments:

1. Your patch has weird line endings.  I get:

    file trunk_sqlite_prepare_v2.patch 
    trunk_sqlite_prepare_v2.patch: ASCII C program text, with CRLF, LF line terminators

   Please use unix line endings only.

2. There is a unit test suite.  To run it, first install the RUnit
   package, install patched RSQLite and then:

   cd RSQLite/inst/UnitTest
   R --slave < runalltests.R

   ## on unix-alike you can just do:
   make test

   Your patch causes one of the tests to fail.  I haven't looked into
   the failure yet, but perhaps you can?

+ seth

-- 
Seth Falcon | Computational Biology | Fred Hutchinson Cancer Research Center
http://bioconductor.org




More information about the R-sig-DB mailing list