Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-18 Thread Aleksander Alekseev
Hi Thomas, > 1. I didn't have --enable-cassert enabled before. Oops. > 2. It'll now dump a gdb backtrace for any core files found after a > check-world failure (if you can find your way to the build log...). > Thanks to Andres for the GDB scripting for this! > 3. It'll now push gcov results

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-17 Thread Thomas Munro
On Mon, Sep 18, 2017 at 2:39 PM, Andres Freund wrote: > E.g. very little of the new stuff in > https://codecov.io/gh/postgresql-cfbot/postgresql/commit/ceaa3dbece3c9b98abcaa28009320fde45a83f88 > is exercised. Hoist by my own petard. -- Thomas Munro

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-17 Thread Andres Freund
Hi, On 2017-09-18 14:26:53 +1200, Thomas Munro wrote: > A couple of new experimental features on commitfest.cputube.org: Yay. > 2. It'll now dump a gdb backtrace for any core files found after a > check-world failure (if you can find your way to the build log...). > Thanks to Andres for the

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-17 Thread Thomas Munro
Hi hackers, A couple of new experimental features on commitfest.cputube.org: 1. I didn't have --enable-cassert enabled before. Oops. 2. It'll now dump a gdb backtrace for any core files found after a check-world failure (if you can find your way to the build log...). Thanks to Andres for the

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Tom Lane
Thomas Munro writes: > On Wed, Sep 13, 2017 at 2:34 AM, Alvaro Herrera > wrote: >> Tom Lane wrote: >>> Can you clarify what went wrong for you on that one? I went to rebase it, >>> but I end up with the identical patch except for a few

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Thomas Munro
On Tue, Sep 12, 2017 at 12:45 AM, Tomas Vondra wrote: > That won't work until (2) is reliable enough. There are patches (for > example my "multivariate MCV lists and histograms") which fails to apply > only because the tool picks the wrong patch. Possibly because it

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Thomas Munro
On Wed, Sep 13, 2017 at 2:34 AM, Alvaro Herrera wrote: > Tom Lane wrote: >> Aleksander Alekseev writes: >> > I've ended up with this script [1]. It just generates a list of patches >> > that are in "Needs Review" status but don't apply or don't

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Andrew Dunstan
On 09/12/2017 11:30 AM, Tom Lane wrote: > Alvaro Herrera writes: >> Tom Lane wrote: >>> Aleksander Alekseev writes: === Apply Failed: 29 === https://commitfest.postgresql.org/14/1235/ (Support arrays over domain types) >>> Can

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Peter Geoghegan
On Tue, Sep 12, 2017 at 8:30 AM, Tom Lane wrote: > +1 ... seems like there's enough noise here that changing patch status > based on the results might be premature. Still, I applaud the effort. There are always going to be cases where the tool fails, unless there are more

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Andres Freund
Hi, On 2017-09-12 11:30:33 -0400, Tom Lane wrote: > One thing I'm a tad worried about is automatically running trojan-horsed > submissions. I hope the CI bot is tightly sandboxed. Well, that's part of the nice thing here. The "really dangerous stuff" is all running on a service that does so

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Tom Lane
Alvaro Herrera writes: > Tom Lane wrote: >> Aleksander Alekseev writes: >>> === Apply Failed: 29 === >>> https://commitfest.postgresql.org/14/1235/ (Support arrays over domain >>> types) >> Can you clarify what went wrong for you on that one?

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Alvaro Herrera
Tom Lane wrote: > Aleksander Alekseev writes: > > I've ended up with this script [1]. It just generates a list of patches > > that are in "Needs Review" status but don't apply or don't compile. Here > > is the current list: > > > === Apply Failed: 29 === > >

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Tom Lane
Aleksander Alekseev writes: > I've ended up with this script [1]. It just generates a list of patches > that are in "Needs Review" status but don't apply or don't compile. Here > is the current list: > === Apply Failed: 29 === >

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Thomas Munro
On Tue, Sep 12, 2017 at 10:03 PM, Aleksander Alekseev wrote: > Unless there are any objections I'm going to give these patches "Waiting > on Author" status today (before doing this I'll re-run the script to > make sure that the list is up-to-date). I'm also going to

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-12 Thread Aleksander Alekseev
Hi Tomas, > > I hope this observation will change your mind :) > > > > Not sure. But I'm mostly just a passenger here, not the driver. After working on this script for some time I got second thoughts regarding this idea as well. The reason for this is that the script is just a bunch of regular

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-11 Thread Erik Rijkers
On 2017-09-11 02:12, Thomas Munro wrote: On Mon, Sep 11, 2017 at 11:40 AM, Michael Paquier wrote: Thomas Munro has hacked up a prototype of application testing automatically if patches submitted apply and build: http://commitfest.cputube.org/ I should add: this is

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-11 Thread Tomas Vondra
On 09/11/2017 03:01 PM, Aleksander Alekseev wrote: > Hi Tomas, > >>> Unless there are any objections to give this idea a try I'm willing to >>> write and host a corresponding script. >>> >> That won't work until (2) is reliable enough. There are patches >> (for example my "multivariate MCV

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-11 Thread Aleksander Alekseev
Hi Tomas, > > Unless there are any objections to give this idea a try I'm willing to > > write and host a corresponding script. > > > That won't work until (2) is reliable enough. There are patches (for > example my "multivariate MCV lists and histograms") which fails to apply > only because the

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-11 Thread Tomas Vondra
On 09/11/2017 11:41 AM, Aleksander Alekseev wrote: > Hi Thomas, > > Great job! > +1 > Here is a crazy idea. What if we write a script that would automatically > return the patches that: > > 1) Are not in "Waiting on Author" status > 2) Don't apply OR don't pass `make installcheck-world` > >

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-11 Thread Ashutosh Bapat
On Mon, Sep 11, 2017 at 3:11 PM, Aleksander Alekseev wrote: > Hi Thomas, > > Great job! > > Here is a crazy idea. What if we write a script that would automatically > return the patches that: > > 1) Are not in "Waiting on Author" status > 2) Don't apply OR don't pass

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-11 Thread Aleksander Alekseev
Hi Thomas, Great job! Here is a crazy idea. What if we write a script that would automatically return the patches that: 1) Are not in "Waiting on Author" status 2) Don't apply OR don't pass `make installcheck-world` ... to the "Waiting on Author" status and describe the problem through the

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-10 Thread Thomas Munro
On Mon, Sep 11, 2017 at 12:10 PM, Jeff Janes wrote: > On Sun, Sep 10, 2017 at 4:40 PM, Michael Paquier >> http://commitfest.cputube.org/ > > This looks very interesting. But when I click on a "build: failing" icon, > it takes me to a generic

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-10 Thread Thomas Munro
On Mon, Sep 11, 2017 at 11:40 AM, Michael Paquier wrote: > Thomas Munro has hacked up a prototype of application testing > automatically if patches submitted apply and build: > http://commitfest.cputube.org/ > > I would recommend have a look at it from time to time if

Re: [HACKERS] Automatic testing of patches in commit fest

2017-09-10 Thread Jeff Janes
On Sun, Sep 10, 2017 at 4:40 PM, Michael Paquier wrote: > Hi all, > > Thomas Munro has hacked up a prototype of application testing > automatically if patches submitted apply and build: > http://commitfest.cputube.org/ > > I would recommend have a look at it from time