Discussion:
Bytea on Windows Update??
Jason Moehlman
2007-05-08 13:20:43 UTC
Permalink
Hello Marc,

I was wondering if there was any update on the Windows bytea
pgescapebytea issue?

Thanks....Jason
Aliacta Support
2007-05-21 11:49:54 UTC
Permalink
Hi Jason,

Any luck with the code I sent you?

Cheers,

Marc
Jason Moehlman
2007-05-21 19:21:18 UTC
Permalink
Hello Marc,

I have created the 15 executables, and plan to test each with 2 files.

Would it help for me to submit the files from good or bad tests along
with the originals for results/examination?

Assuming I can keep it all separated, I anticipate finishing this
testing tonight.

Thanks again.

Jason
Post by Aliacta Support
Hi Jason,
Any luck with the code I sent you?
Cheers,
Marc
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
Jason Moehlman
2007-05-22 02:27:56 UTC
Permalink
Hello,

I hope this makes sense and matches your testing experience.


Version Mac Linux Windows
V1 JPG OK BAD BAD
V1 PDF OK BAD BAD
V2 JPG BAD BAD BAD
V2 PDF BAD BAD BAD
V3 JPG OK OK OK
V3 PDF OK OK OK
V4 JPG OK OK OK
V4 PDF OK OK OK
V5 JPG OK OK OK
V5 PDF OK OK OK


Thanks.

Jason
Post by Jason Moehlman
Hello Marc,
I have created the 15 executables, and plan to test each with 2 files.
Would it help for me to submit the files from good or bad tests along
with the originals for results/examination?
Assuming I can keep it all separated, I anticipate finishing this
testing tonight.
Thanks again.
Jason
Post by Aliacta Support
Hi Jason,
Any luck with the code I sent you?
Cheers,
Marc
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
Aliacta Support
2007-05-23 09:29:09 UTC
Permalink
Post by Jason Moehlman
Hello,
I hope this makes sense and matches your testing experience.
Version Mac Linux Windows
V1 JPG OK BAD BAD
V1 PDF OK BAD BAD
V2 JPG BAD BAD BAD
V2 PDF BAD BAD BAD
V3 JPG OK OK OK
V3 PDF OK OK OK
V4 JPG OK OK OK
V4 PDF OK OK OK
V5 JPG OK OK OK
V5 PDF OK OK OK
Thanks Jason!

V1 was supposed to work on Linux as well according to your previous
message since it is the same as the method of the current release.

I'm a bit surprised V3 to V5 all work the same since there are
endianness differences in them. I'll try to devise a test with a
huge number of files before using one of them for the official
release. If you have time to test these 3 more that would be cool as
well.

Cheers,

Marc
Jason Moehlman
2007-05-23 12:20:13 UTC
Permalink
Hello Marc,

I was a little surprised also based on what I thought would happen.
I will do some more testing also with a bigger range of files, I just
wanted to get something back to you. I will also double check
everything I did to make sure I didn't miss something.

If it makes a difference I have not run any of this on a PPC Mac,
just Intel MacBook Pro at this point. PostgreSQL 8.1.3 is also
running on an Intel Mac for these tests. I can check them on a G5 later.

I tried to isolate the tests as much as possible, separate DB's,
separate project files with the change in escaping routine, separate
executables numbered for the version, etc. I will follow up once I
recheck and add some additional files.

Jason
Post by Aliacta Support
Post by Jason Moehlman
Hello,
I hope this makes sense and matches your testing experience.
Version Mac Linux Windows
V1 JPG OK BAD BAD
V1 PDF OK BAD BAD
V2 JPG BAD BAD BAD
V2 PDF BAD BAD BAD
V3 JPG OK OK OK
V3 PDF OK OK OK
V4 JPG OK OK OK
V4 PDF OK OK OK
V5 JPG OK OK OK
V5 PDF OK OK OK
Thanks Jason!
V1 was supposed to work on Linux as well according to your previous
message since it is the same as the method of the current release.
I'm a bit surprised V3 to V5 all work the same since there are
endianness differences in them. I'll try to devise a test with a
huge number of files before using one of them for the official
release. If you have time to test these 3 more that would be cool
as well.
Cheers,
Marc
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
Jason Moehlman
2007-05-23 14:40:52 UTC
Permalink
Hello Marc,

I was able to verify everything is in order as far as versions,
executables, and results. I re-ran each of the files/tests and
checked each individually.

I also quickly dumped in a windows exe in a zip file as another
test. The size of the files is around 11MB total, I thought that
would have a reasonable chance of hitting a large number of byte
sequences.

These tests were all on Intel based machines. G5 tests to follow at
a later time.

Version Mac Linux Windows
V1 JPG OK BAD BAD
V1 PDF OK BAD BAD
V1 ZIP OK BAD BAD
V2 JPG BAD BAD BAD
V2 PDF BAD BAD BAD
V2 ZIP BAD BAD BAD
V3 JPG OK OK OK
V3 PDF OK OK OK
V3 ZIP OK OK OK
V4 JPG OK OK OK
V4 PDF OK OK OK
V4 ZIP OK OK OK
V5 JPG OK OK OK
V5 PDF OK OK OK
V5 ZIP OK OK OK

My only explanation for the original difference for V1 on Linux is I
may have made a mistake about it working, if so I appologize. I am
much more confident about these tests and the way I conducted them.

If there is anything else I can provide please let me know.

Have a great day.

Jason
Post by Jason Moehlman
Hello Marc,
I was a little surprised also based on what I thought would
happen. I will do some more testing also with a bigger range of
files, I just wanted to get something back to you. I will also
double check everything I did to make sure I didn't miss something.
If it makes a difference I have not run any of this on a PPC Mac,
just Intel MacBook Pro at this point. PostgreSQL 8.1.3 is also
running on an Intel Mac for these tests. I can check them on a G5 later.
I tried to isolate the tests as much as possible, separate DB's,
separate project files with the change in escaping routine,
separate executables numbered for the version, etc. I will follow
up once I recheck and add some additional files.
Jason
Post by Aliacta Support
Post by Jason Moehlman
Hello,
I hope this makes sense and matches your testing experience.
Version Mac Linux Windows
V1 JPG OK BAD BAD
V1 PDF OK BAD BAD
V2 JPG BAD BAD BAD
V2 PDF BAD BAD BAD
V3 JPG OK OK OK
V3 PDF OK OK OK
V4 JPG OK OK OK
V4 PDF OK OK OK
V5 JPG OK OK OK
V5 PDF OK OK OK
Thanks Jason!
V1 was supposed to work on Linux as well according to your
previous message since it is the same as the method of the current
release.
I'm a bit surprised V3 to V5 all work the same since there are
endianness differences in them. I'll try to devise a test with a
huge number of files before using one of them for the official
release. If you have time to test these 3 more that would be cool
as well.
Cheers,
Marc
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
Aliacta Support
2007-05-23 15:50:15 UTC
Permalink
Thanks a lot for your time and effort Jason. I guess you can use
either V3, 4, or 5 then while you wait for the next official release
of pgSQL4RB. ;-)

Marc
Post by Jason Moehlman
Hello Marc,
I was able to verify everything is in order as far as versions,
executables, and results. I re-ran each of the files/tests and
checked each individually.
I also quickly dumped in a windows exe in a zip file as another
test. The size of the files is around 11MB total, I thought that
would have a reasonable chance of hitting a large number of byte
sequences.
These tests were all on Intel based machines. G5 tests to follow
at a later time.
Version Mac Linux Windows
V1 JPG OK BAD BAD
V1 PDF OK BAD BAD
V1 ZIP OK BAD BAD
V2 JPG BAD BAD BAD
V2 PDF BAD BAD BAD
V2 ZIP BAD BAD BAD
V3 JPG OK OK OK
V3 PDF OK OK OK
V3 ZIP OK OK OK
V4 JPG OK OK OK
V4 PDF OK OK OK
V4 ZIP OK OK OK
V5 JPG OK OK OK
V5 PDF OK OK OK
V5 ZIP OK OK OK
My only explanation for the original difference for V1 on Linux is
I may have made a mistake about it working, if so I appologize. I
am much more confident about these tests and the way I conducted them.
If there is anything else I can provide please let me know.
Have a great day.
Jason
Post by Jason Moehlman
Hello Marc,
I was a little surprised also based on what I thought would
happen. I will do some more testing also with a bigger range of
files, I just wanted to get something back to you. I will also
double check everything I did to make sure I didn't miss something.
If it makes a difference I have not run any of this on a PPC Mac,
just Intel MacBook Pro at this point. PostgreSQL 8.1.3 is also
running on an Intel Mac for these tests. I can check them on a G5 later.
I tried to isolate the tests as much as possible, separate DB's,
separate project files with the change in escaping routine,
separate executables numbered for the version, etc. I will follow
up once I recheck and add some additional files.
Jason
Post by Aliacta Support
Post by Jason Moehlman
Hello,
I hope this makes sense and matches your testing experience.
Version Mac Linux Windows
V1 JPG OK BAD BAD
V1 PDF OK BAD BAD
V2 JPG BAD BAD BAD
V2 PDF BAD BAD BAD
V3 JPG OK OK OK
V3 PDF OK OK OK
V4 JPG OK OK OK
V4 PDF OK OK OK
V5 JPG OK OK OK
V5 PDF OK OK OK
Thanks Jason!
V1 was supposed to work on Linux as well according to your
previous message since it is the same as the method of the
current release.
I'm a bit surprised V3 to V5 all work the same since there are
endianness differences in them. I'll try to devise a test with a
huge number of files before using one of them for the official
release. If you have time to test these 3 more that would be
cool as well.
Cheers,
Marc
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
Jason Moehlman
2007-05-21 18:40:14 UTC
Permalink
Sorry I have not emailed back sooner, I have been working on the
release of my v2 framework. I tried setting this up today when I
realized how long it had been, I think the new framework is ready on
my end.

I ran into a few small false starts due to my incredible lack of
ingenuity. I think I am on the right track now. I will let you know
in a day or so.

Can you confirm that there are only escaping routines in here to test
also. I just want to make sure I am not missing anything as part of
the test and the tab completion is not very helpful if you don't know
at least part of the function name your looking for.

Thanks again.

Jason
Post by Aliacta Support
Hi Jason,
Any luck with the code I sent you?
Cheers,
Marc
_______________________________________________
Postgresql mailing list
http://aliacta.com/mailman/listinfo/postgresql_aliacta.com
Loading...