You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@doris.apache.org by 924060929 <92...@qq.com.INVALID> on 2022/02/19 04:20:26 UTC
The new scalable regression testing framework(#8125)
In the past, developer of Apache Doris send pull request and wait for regression testing, which is execute in Baidu internally.
There are some problem:
1. test case is not shared to developer. so we can not quickly accumulate test case to ensure the stable of Apache Doris
2. waiting too long. if test failed, developer must change code and wait again
So, I implement a scalable regression testing framework(in pull request #8125).
Developer can run regression testing in local enviroment, and contribute more test case.
In order to readable, writeable, normalization and scalable, I use groovy to implement a DSL which named Action, developer can use Action to write various test case, for example
1. sql action can execute a sql
2. qt action can execute sql and compare result to tsv file
3. stream load action can load data to doris
4. explain action can check query plan
welcome to contribute test case, improvement test framework, and update document.
Re: The new scalable regression testing framework(#8125)
Posted by Shuo Wang <wa...@gmail.com>.
Great job, man!
ling miao <li...@apache.org> 于2022年2月21日周一 11:25写道：
> First, we need to open source the P0 test set (From Baidu)~
>
> Ling Miao
>
> 924060929 <92...@qq.com.invalid> 于2022年2月19日周六 18:43写道：
>
> > Thinks for greate suggestions.
> >
> >
> > The print result function is missing, and can run parallely
> > and randomly.
> > But it is easy to implement, I can support in next pull request.
> >
> >
> >
> >
> > ------------------ 原始邮件 ------------------
> > 发件人: "1 0"<dataroaring@gmail.com>;
> > 发送时间: 2022年2月19日(星期六) 下午4:30
> > 收件人: "dev"<dev@doris.apache.org>;
> > 主题: Re: The new scalable regression testing framework(#8125)
> >
> >
> >
> >
> >
> It's a great job. I have a suggestion on writing a case more easily. We
> >
> >
> should accept cases in sql and check the result by comparing the content of
> >
> >
> stdout and stderr. This way, anyone can contribute a case easily, for
> >
> >
> example I just execute sql and put output. The cases can also be used
> > easily by other tests like fuzzy.
> >
> >
> >
> Can the framework run parallely and randomly? We need to run functional and
> >
> >
> regression tests parallely and randomly to discover more problems.
> >
> > Thanks,
> > Dataroaring.
>
Re:Re: Re: The new scalable regression testing framework(#8125)
Posted by 陈明雨 <mo...@163.com>.
>We need the contributor to sign the iCLA when submitting no trivial patch.
>Otherwise, we cannot merge the patch into the main branch.
Most of the Apache project contribution guidelines that we have consulted before do not require PR contributors to sign iCLA,
but only committers to sign iCLA before being granted commit privileges.
However, the guidance on the Apache website does recommend that contributors also sign the iCLA to avoid potential legal risks.
We will evaluate this issue more rigorously in our future code merge process and try to push contributors to sign iCLA for non-trivial code contributions.
> if the contributor do the contribution in his
> daily work, we need to do the IP clearance with the SGA or CCLA.
OK, I see.
And Thanks to Willem for the reminder.!
--
此致！Best Regards
陈明雨 Mingyu Chen
Email:
chenmingyu@apache.org
在 2022-02-22 15:13:40，"Willem Jiang" <wi...@gmail.com> 写道：
>We need the contributor to sign the iCLA when submitting no trivial patch.
>Otherwise, we cannot merge the patch into the main branch.
>
>For the question2, if the contributor do the contribution in his
>daily work, we need to do the IP clearance with the SGA or CCLA.
>
>Please let me know if you have any question about it.
>
>Willem Jiang
>
>Twitter: willemjiang
>Weibo: 姜宁willem
>
>On Tue, Feb 22, 2022 at 11:29 AM 陈明雨 <mo...@163.com> wrote:
>>
>> Hi Willem:
>>
>>
>> I have a few questions.
>> 1. Do all contributors need to sign iCLA?
>> Currently most Contributors in Doris have not signed iCLA, only Committer or PPMC members have signed it.
>> 2. this feature is from a contributor in Meituan, Inc, does it also need IP clearance?
>> I refer to the instructions here[1]. But didn't find how to start the process...
>> And it also says: Incubating projects should follow the procedure described in the incubator mentor guide and report status via STATUS tracking.
>> but still not find the way to start...
>>
>>
>>
>> [1] https://incubator.apache.org/ip-clearance/
>>
>>
>>
>>
>> --
>>
>> 此致！Best Regards
>> 陈明雨 Mingyu Chen
>>
>> Email:
>> chenmingyu@apache.org
>>
>>
>>
>>
>>
>> 在 2022-02-22 08:47:40，"Willem Jiang" <wi...@gmail.com> 写道：
>> >You can find the IP Clearance guide here[1]
>> >
>> >[1]https://incubator.apache.org/guides/ip_clearance.html
>> >
>> >Willem Jiang
>> >
>> >Twitter: willemjiang
>> >Weibo: 姜宁willem
>> >
>> >On Tue, Feb 22, 2022 at 8:40 AM Willem Jiang <wi...@gmail.com> wrote:
>> >>
>> >> Please make sure the contributor signed the CLA.
>> >> If these test framework is from Baidu, we need to go through the IP clearance[1]
>> >>
>> >> [1]https://incubator.apache.org/ip-clearance/
>> >>
>> >> Willem Jiang
>> >>
>> >> Twitter: willemjiang
>> >> Weibo: 姜宁willem
>> >>
>> >> On Mon, Feb 21, 2022 at 11:25 AM ling miao <li...@apache.org> wrote:
>> >> >
>> >> > First, we need to open source the P0 test set (From Baidu)~
>> >> >
>> >> > Ling Miao
>> >> >
>> >> > 924060929 <92...@qq.com.invalid> 于2022年2月19日周六 18:43写道：
>> >> >
>> >> > > Thinks for greate suggestions.
>> >> > >
>> >> > >
>> >> > > The print result function is missing, and can run parallely
>> >> > > and randomly.
>> >> > > But it is easy to implement, I can support in next pull request.
>> >> > >
>> >> > >
>> >> > >
>> >> > >
>> >> > > ------------------ 原始邮件 ------------------
>> >> > > 发件人: "1 0"<dataroaring@gmail.com>;
>> >> > > 发送时间: 2022年2月19日(星期六) 下午4:30
>> >> > > 收件人: "dev"<dev@doris.apache.org>;
>> >> > > 主题: Re: The new scalable regression testing framework(#8125)
>> >> > >
>> >> > >
>> >> > >
>> >> > >
>> >> > > It's a great job. I have a suggestion on writing a case more easily. We
>> >> > >
>> >> > > should accept cases in sql and check the result by comparing the content of
>> >> > >
>> >> > > stdout and stderr. This way, anyone can contribute a case easily, for
>> >> > >
>> >> > > example I just execute sql and put output. The cases can also be used
>> >> > > easily by other tests like fuzzy.
>> >> > >
>> >> > >
>> >> > > Can the framework run parallely and randomly? We need to run functional and
>> >> > >
>> >> > > regression tests parallely and randomly to discover more problems.
>> >> > >
>> >> > > Thanks,
>> >> > > Dataroaring.
>> >
>> >---------------------------------------------------------------------
>> >To unsubscribe, e-mail: dev-unsubscribe@doris.apache.org
>> >For additional commands, e-mail: dev-help@doris.apache.org
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: dev-unsubscribe@doris.apache.org
>For additional commands, e-mail: dev-help@doris.apache.org
Re: Re: The new scalable regression testing framework(#8125)
Posted by Willem Jiang <wi...@gmail.com>.
We need the contributor to sign the iCLA when submitting no trivial patch.
Otherwise, we cannot merge the patch into the main branch.
For the question2, if the contributor do the contribution in his
daily work, we need to do the IP clearance with the SGA or CCLA.
Please let me know if you have any question about it.
Willem Jiang
Twitter: willemjiang
Weibo: 姜宁willem
On Tue, Feb 22, 2022 at 11:29 AM 陈明雨 <mo...@163.com> wrote:
>
> Hi Willem:
>
>
> I have a few questions.
> 1. Do all contributors need to sign iCLA?
> Currently most Contributors in Doris have not signed iCLA, only Committer or PPMC members have signed it.
> 2. this feature is from a contributor in Meituan, Inc, does it also need IP clearance?
> I refer to the instructions here[1]. But didn't find how to start the process...
> And it also says: Incubating projects should follow the procedure described in the incubator mentor guide and report status via STATUS tracking.
> but still not find the way to start...
>
>
>
> [1] https://incubator.apache.org/ip-clearance/
>
>
>
>
> --
>
> 此致！Best Regards
> 陈明雨 Mingyu Chen
>
> Email:
> chenmingyu@apache.org
>
>
>
>
>
> 在 2022-02-22 08:47:40，"Willem Jiang" <wi...@gmail.com> 写道：
> >You can find the IP Clearance guide here[1]
> >
> >[1]https://incubator.apache.org/guides/ip_clearance.html
> >
> >Willem Jiang
> >
> >Twitter: willemjiang
> >Weibo: 姜宁willem
> >
> >On Tue, Feb 22, 2022 at 8:40 AM Willem Jiang <wi...@gmail.com> wrote:
> >>
> >> Please make sure the contributor signed the CLA.
> >> If these test framework is from Baidu, we need to go through the IP clearance[1]
> >>
> >> [1]https://incubator.apache.org/ip-clearance/
> >>
> >> Willem Jiang
> >>
> >> Twitter: willemjiang
> >> Weibo: 姜宁willem
> >>
> >> On Mon, Feb 21, 2022 at 11:25 AM ling miao <li...@apache.org> wrote:
> >> >
> >> > First, we need to open source the P0 test set (From Baidu)~
> >> >
> >> > Ling Miao
> >> >
> >> > 924060929 <92...@qq.com.invalid> 于2022年2月19日周六 18:43写道：
> >> >
> >> > > Thinks for greate suggestions.
> >> > >
> >> > >
> >> > > The print result function is missing, and can run parallely
> >> > > and randomly.
> >> > > But it is easy to implement, I can support in next pull request.
> >> > >
> >> > >
> >> > >
> >> > >
> >> > > ------------------ 原始邮件 ------------------
> >> > > 发件人: "1 0"<dataroaring@gmail.com>;
> >> > > 发送时间: 2022年2月19日(星期六) 下午4:30
> >> > > 收件人: "dev"<dev@doris.apache.org>;
> >> > > 主题: Re: The new scalable regression testing framework(#8125)
> >> > >
> >> > >
> >> > >
> >> > >
> >> > > It's a great job. I have a suggestion on writing a case more easily. We
> >> > >
> >> > > should accept cases in sql and check the result by comparing the content of
> >> > >
> >> > > stdout and stderr. This way, anyone can contribute a case easily, for
> >> > >
> >> > > example I just execute sql and put output. The cases can also be used
> >> > > easily by other tests like fuzzy.
> >> > >
> >> > >
> >> > > Can the framework run parallely and randomly? We need to run functional and
> >> > >
> >> > > regression tests parallely and randomly to discover more problems.
> >> > >
> >> > > Thanks,
> >> > > Dataroaring.
> >
> >---------------------------------------------------------------------
> >To unsubscribe, e-mail: dev-unsubscribe@doris.apache.org
> >For additional commands, e-mail: dev-help@doris.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@doris.apache.org
For additional commands, e-mail: dev-help@doris.apache.org
Re:Re: The new scalable regression testing framework(#8125)
Posted by 陈明雨 <mo...@163.com>.
Hi Willem:
I have a few questions.
1. Do all contributors need to sign iCLA?
Currently most Contributors in Doris have not signed iCLA, only Committer or PPMC members have signed it.
2. this feature is from a contributor in Meituan, Inc, does it also need IP clearance?
I refer to the instructions here[1]. But didn't find how to start the process...
And it also says: Incubating projects should follow the procedure described in the incubator mentor guide and report status via STATUS tracking.
but still not find the way to start...
[1] https://incubator.apache.org/ip-clearance/
--
此致！Best Regards
陈明雨 Mingyu Chen
Email:
chenmingyu@apache.org
在 2022-02-22 08:47:40，"Willem Jiang" <wi...@gmail.com> 写道：
>You can find the IP Clearance guide here[1]
>
>[1]https://incubator.apache.org/guides/ip_clearance.html
>
>Willem Jiang
>
>Twitter: willemjiang
>Weibo: 姜宁willem
>
>On Tue, Feb 22, 2022 at 8:40 AM Willem Jiang <wi...@gmail.com> wrote:
>>
>> Please make sure the contributor signed the CLA.
>> If these test framework is from Baidu, we need to go through the IP clearance[1]
>>
>> [1]https://incubator.apache.org/ip-clearance/
>>
>> Willem Jiang
>>
>> Twitter: willemjiang
>> Weibo: 姜宁willem
>>
>> On Mon, Feb 21, 2022 at 11:25 AM ling miao <li...@apache.org> wrote:
>> >
>> > First, we need to open source the P0 test set (From Baidu)~
>> >
>> > Ling Miao
>> >
>> > 924060929 <92...@qq.com.invalid> 于2022年2月19日周六 18:43写道：
>> >
>> > > Thinks for greate suggestions.
>> > >
>> > >
>> > > The print result function is missing, and can run parallely
>> > > and randomly.
>> > > But it is easy to implement, I can support in next pull request.
>> > >
>> > >
>> > >
>> > >
>> > > ------------------ 原始邮件 ------------------
>> > > 发件人: "1 0"<dataroaring@gmail.com>;
>> > > 发送时间: 2022年2月19日(星期六) 下午4:30
>> > > 收件人: "dev"<dev@doris.apache.org>;
>> > > 主题: Re: The new scalable regression testing framework(#8125)
>> > >
>> > >
>> > >
>> > >
>> > > It's a great job. I have a suggestion on writing a case more easily. We
>> > >
>> > > should accept cases in sql and check the result by comparing the content of
>> > >
>> > > stdout and stderr. This way, anyone can contribute a case easily, for
>> > >
>> > > example I just execute sql and put output. The cases can also be used
>> > > easily by other tests like fuzzy.
>> > >
>> > >
>> > > Can the framework run parallely and randomly? We need to run functional and
>> > >
>> > > regression tests parallely and randomly to discover more problems.
>> > >
>> > > Thanks,
>> > > Dataroaring.
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: dev-unsubscribe@doris.apache.org
>For additional commands, e-mail: dev-help@doris.apache.org
Re: The new scalable regression testing framework(#8125)
Posted by Willem Jiang <wi...@gmail.com>.
You can find the IP Clearance guide here[1]
[1]https://incubator.apache.org/guides/ip_clearance.html
Willem Jiang
Twitter: willemjiang
Weibo: 姜宁willem
On Tue, Feb 22, 2022 at 8:40 AM Willem Jiang <wi...@gmail.com> wrote:
>
> Please make sure the contributor signed the CLA.
> If these test framework is from Baidu, we need to go through the IP clearance[1]
>
> [1]https://incubator.apache.org/ip-clearance/
>
> Willem Jiang
>
> Twitter: willemjiang
> Weibo: 姜宁willem
>
> On Mon, Feb 21, 2022 at 11:25 AM ling miao <li...@apache.org> wrote:
> >
> > First, we need to open source the P0 test set (From Baidu)~
> >
> > Ling Miao
> >
> > 924060929 <92...@qq.com.invalid> 于2022年2月19日周六 18:43写道：
> >
> > > Thinks for greate suggestions.
> > >
> > >
> > > The print result function is missing, and can run parallely
> > > and randomly.
> > > But it is easy to implement, I can support in next pull request.
> > >
> > >
> > >
> > >
> > > ------------------ 原始邮件 ------------------
> > > 发件人: "1 0"<dataroaring@gmail.com>;
> > > 发送时间: 2022年2月19日(星期六) 下午4:30
> > > 收件人: "dev"<dev@doris.apache.org>;
> > > 主题: Re: The new scalable regression testing framework(#8125)
> > >
> > >
> > >
> > >
> > > It's a great job. I have a suggestion on writing a case more easily. We
> > >
> > > should accept cases in sql and check the result by comparing the content of
> > >
> > > stdout and stderr. This way, anyone can contribute a case easily, for
> > >
> > > example I just execute sql and put output. The cases can also be used
> > > easily by other tests like fuzzy.
> > >
> > >
> > > Can the framework run parallely and randomly? We need to run functional and
> > >
> > > regression tests parallely and randomly to discover more problems.
> > >
> > > Thanks,
> > > Dataroaring.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@doris.apache.org
For additional commands, e-mail: dev-help@doris.apache.org
Re: The new scalable regression testing framework(#8125)
Posted by Willem Jiang <wi...@gmail.com>.
Please make sure the contributor signed the CLA.
If these test framework is from Baidu, we need to go through the IP clearance[1]
[1]https://incubator.apache.org/ip-clearance/
Willem Jiang
Twitter: willemjiang
Weibo: 姜宁willem
On Mon, Feb 21, 2022 at 11:25 AM ling miao <li...@apache.org> wrote:
>
> First, we need to open source the P0 test set (From Baidu)~
>
> Ling Miao
>
> 924060929 <92...@qq.com.invalid> 于2022年2月19日周六 18:43写道：
>
> > Thinks for greate suggestions.
> >
> >
> > The print result function is missing, and can run parallely
> > and randomly.
> > But it is easy to implement, I can support in next pull request.
> >
> >
> >
> >
> > ------------------ 原始邮件 ------------------
> > 发件人: "1 0"<dataroaring@gmail.com>;
> > 发送时间: 2022年2月19日(星期六) 下午4:30
> > 收件人: "dev"<dev@doris.apache.org>;
> > 主题: Re: The new scalable regression testing framework(#8125)
> >
> >
> >
> >
> > It's a great job. I have a suggestion on writing a case more easily. We
> >
> > should accept cases in sql and check the result by comparing the content of
> >
> > stdout and stderr. This way, anyone can contribute a case easily, for
> >
> > example I just execute sql and put output. The cases can also be used
> > easily by other tests like fuzzy.
> >
> >
> > Can the framework run parallely and randomly? We need to run functional and
> >
> > regression tests parallely and randomly to discover more problems.
> >
> > Thanks,
> > Dataroaring.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@doris.apache.org
For additional commands, e-mail: dev-help@doris.apache.org
Re: The new scalable regression testing framework(#8125)
Posted by ling miao <li...@apache.org>.
> Do you have any plan on opening source the P0 test set? It should be a
very
good start for community to build ci/cd pipeline.
Sure, I will prepare to open source P0 test after our test framework finish
~
Ling Miao
1 0 <da...@gmail.com> 于2022年3月2日周三 09:58写道：
> Hi ling,
>
> Do you have any plan on opening source the P0 test set? It should be a very
> good start for community to build ci/cd pipeline.
>
> ling miao <li...@apache.org> 於 2022年2月21日週一 上午11:25寫道：
>
> > First, we need to open source the P0 test set (From Baidu)~
> >
> > Ling Miao
> >
> > 924060929 <92...@qq.com.invalid> 于2022年2月19日周六 18:43写道：
> >
> > > Thinks for greate suggestions.
> > >
> > >
> > > The print result function is missing, and can run parallely
> > > and randomly.
> > > But it is easy to implement, I can support in next pull request.
> > >
> > >
> > >
> > >
> > > ------------------ 原始邮件 ------------------
> > > 发件人: "1 0"<dataroaring@gmail.com>;
> > > 发送时间: 2022年2月19日(星期六) 下午4:30
> > > 收件人: "dev"<dev@doris.apache.org>;
> > > 主题: Re: The new scalable regression testing framework(#8125)
> > >
> > >
> > >
> > >
> > >
> >
> It's a great job. I have a suggestion on writing a case more easily. We
> > >
> > >
> >
> should accept cases in sql and check the result by comparing the content of
> > >
> > >
> >
> stdout and stderr. This way, anyone can contribute a case easily, for
> > >
> > >
> >
> example I just execute sql and put output. The cases can also be used
> > > easily by other tests like fuzzy.
> > >
> > >
> > >
> >
> Can the framework run parallely and randomly? We need to run functional and
> > >
> > >
> >
> regression tests parallely and randomly to discover more problems.
> > >
> > > Thanks,
> > > Dataroaring.
> >
>
Re: The new scalable regression testing framework(#8125)
Posted by 1 0 <da...@gmail.com>.
Hi ling,
Do you have any plan on opening source the P0 test set? It should be a very
good start for community to build ci/cd pipeline.
ling miao <li...@apache.org> 於 2022年2月21日週一 上午11:25寫道：
> First, we need to open source the P0 test set (From Baidu)~
>
> Ling Miao
>
> 924060929 <92...@qq.com.invalid> 于2022年2月19日周六 18:43写道：
>
> > Thinks for greate suggestions.
> >
> >
> > The print result function is missing, and can run parallely
> > and randomly.
> > But it is easy to implement, I can support in next pull request.
> >
> >
> >
> >
> > ------------------ 原始邮件 ------------------
> > 发件人: "1 0"<dataroaring@gmail.com>;
> > 发送时间: 2022年2月19日(星期六) 下午4:30
> > 收件人: "dev"<dev@doris.apache.org>;
> > 主题: Re: The new scalable regression testing framework(#8125)
> >
> >
> >
> >
> >
> It's a great job. I have a suggestion on writing a case more easily. We
> >
> >
> should accept cases in sql and check the result by comparing the content of
> >
> >
> stdout and stderr. This way, anyone can contribute a case easily, for
> >
> >
> example I just execute sql and put output. The cases can also be used
> > easily by other tests like fuzzy.
> >
> >
> >
> Can the framework run parallely and randomly? We need to run functional and
> >
> >
> regression tests parallely and randomly to discover more problems.
> >
> > Thanks,
> > Dataroaring.
>
Re: The new scalable regression testing framework(#8125)
Posted by ling miao <li...@apache.org>.
First, we need to open source the P0 test set (From Baidu)~
Ling Miao
924060929 <92...@qq.com.invalid> 于2022年2月19日周六 18:43写道：
> Thinks for greate suggestions.
>
>
> The print result function is missing, and can run parallely
> and randomly.
> But it is easy to implement, I can support in next pull request.
>
>
>
>
> ------------------ 原始邮件 ------------------
> 发件人: "1 0"<dataroaring@gmail.com>;
> 发送时间: 2022年2月19日(星期六) 下午4:30
> 收件人: "dev"<dev@doris.apache.org>;
> 主题: Re: The new scalable regression testing framework(#8125)
>
>
>
>
> It's a great job. I have a suggestion on writing a case more easily. We
>
> should accept cases in sql and check the result by comparing the content of
>
> stdout and stderr. This way, anyone can contribute a case easily, for
>
> example I just execute sql and put output. The cases can also be used
> easily by other tests like fuzzy.
>
>
> Can the framework run parallely and randomly? We need to run functional and
>
> regression tests parallely and randomly to discover more problems.
>
> Thanks,
> Dataroaring.
回复： The new scalable regression testing framework(#8125)
Posted by 924060929 <92...@qq.com.INVALID>.
Thinks for greate suggestions.
The print result function is missing, and can run parallely and randomly.
But it is easy to implement, I can support in next pull request.
------------------ 原始邮件 ------------------
发件人: "1 0"<dataroaring@gmail.com>;
发送时间: 2022年2月19日(星期六) 下午4:30
收件人: "dev"<dev@doris.apache.org>;
主题: Re: The new scalable regression testing framework(#8125)
It's a great job. I have a suggestion on writing a case more easily. We
should accept cases in sql and check the result by comparing the content of
stdout and stderr. This way, anyone can contribute a case easily, for
example I just execute sql and put output. The cases can also be used
easily by other tests like fuzzy.
Can the framework run parallely and randomly? We need to run functional and
regression tests parallely and randomly to discover more problems.
Thanks,
Dataroaring.
Re: The new scalable regression testing framework(#8125)
Posted by 1 0 <da...@gmail.com>.
It's a great job. I have a suggestion on writing a case more easily. We
should accept cases in sql and check the result by comparing the content of
stdout and stderr. This way, anyone can contribute a case easily, for
example I just execute sql and put output. The cases can also be used
easily by other tests like fuzzy.
Can the framework run parallely and randomly? We need to run functional and
regression tests parallely and randomly to discover more problems.
Thanks,
Dataroaring.
Re:The new scalable regression testing framework(#8125)
Posted by 陈明雨 <mo...@163.com>.
That is a GREATE job!
We have discussed privately some details about this testing framework, including some POC demos. I think it is a very good framework.
This framework is currently a standalone part, so I'll go ahead and merge this PR in so that people can do quick iterations on it.
TODO:
1. English document
2. Contribution Guide: How to add test
--
此致！Best Regards
陈明雨 Mingyu Chen
Email:
chenmingyu@apache.org
At 2022-02-19 12:20:26, "924060929" <92...@qq.com.INVALID> wrote:
>In the past, developer of Apache Doris send pull request and wait for regression testing, which is execute in Baidu internally.
>There are some problem:
>1. test case is not shared to developer. so we can not quickly accumulate test case to ensure the stable of Apache Doris
>2. waiting too long. if test failed, developer must change code and wait again
>
>
>So, I implement a scalable regression testing framework(in pull request #8125).
>Developer can run regression testing in local enviroment, and contribute more test case.
>
>
>In order to readable, writeable, normalization and scalable, I use groovy to implement a DSL which named Action, developer can use Action to write various test case, for example
>1. sql action can execute a sql
>2. qt action can execute sql and compare result to tsv file
>3. stream load action can load data to doris
>4. explain action can check query plan
>
>
>welcome to contribute test case, improvement test framework, and update document.