Silly Q, did you blow away the pip cache before committing the layer? That always trips me up.
Cheers Andrew On Tue, Aug 17, 2021 at 10:56 Mich Talebzadeh <mich.talebza...@gmail.com> wrote: > With no additional python packages etc we get 1.4GB compared to 2.19GB > before > > REPOSITORY TAG IMAGE ID > CREATED SIZE > spark/spark-py 3.1.1_sparkpy_3.7-scala_2.12-java8only faee4dbb95dd > Less than a second ago 1.41GB > spark/spark-py 3.1.1_sparkpy_3.7-scala_2.12-java8 ba3c17bc9337 4 > hours ago 2.19GB > > root@233a81199b43:/opt/spark/work-dir# pip list > Package Version > ------------- ------- > asn1crypto 0.24.0 > cryptography 2.6.1 > entrypoints 0.3 > keyring 17.1.1 > keyrings.alt 3.1.1 > pip 21.2.4 > pycrypto 2.6.1 > PyGObject 3.30.4 > pyxdg 0.25 > SecretStorage 2.3.1 > setuptools 57.4.0 > six 1.12.0 > wheel 0.32.3 > > > HTH > > > view my Linkedin profile > <https://www.linkedin.com/in/mich-talebzadeh-ph-d-5205b2/> > > > > *Disclaimer:* Use it at your own risk. Any and all responsibility for any > loss, damage or destruction of data or any other property which may arise > from relying on this email's technical content is explicitly disclaimed. > The author will in no case be liable for any monetary damages arising from > such loss, damage or destruction. > > > > > On Tue, 17 Aug 2021 at 16:24, Mich Talebzadeh <mich.talebza...@gmail.com> > wrote: > >> Yes, I will double check. it includes java 8 in addition to base java 11. >> >> in addition it has these Python packages for now (added for my own needs >> for now) >> >> root@ce6773017a14:/opt/spark/work-dir# pip list >> Package Version >> ------------- ------- >> asn1crypto 0.24.0 >> cryptography 2.6.1 >> cx-Oracle 8.2.1 >> entrypoints 0.3 >> keyring 17.1.1 >> keyrings.alt 3.1.1 >> numpy 1.21.2 >> pip 21.2.4 >> py4j 0.10.9 >> pycrypto 2.6.1 >> PyGObject 3.30.4 >> pyspark 3.1.2 >> pyxdg 0.25 >> PyYAML 5.4.1 >> SecretStorage 2.3.1 >> setuptools 57.4.0 >> six 1.12.0 >> wheel 0.32.3 >> >> >> HTH >> >> >> view my Linkedin profile >> <https://www.linkedin.com/in/mich-talebzadeh-ph-d-5205b2/> >> >> >> >> *Disclaimer:* Use it at your own risk. Any and all responsibility for >> any loss, damage or destruction of data or any other property which may >> arise from relying on this email's technical content is explicitly >> disclaimed. The author will in no case be liable for any monetary damages >> arising from such loss, damage or destruction. >> >> >> >> >> On Tue, 17 Aug 2021 at 16:17, Maciej <mszymkiew...@gmail.com> wrote: >> >>> Quick question ‒ is this actual output? If so, do we know what accounts >>> 1.5GB overhead for PySpark image. Even without --no-install-recommends >>> this seems like a lot (if I recall correctly it was around 400MB for >>> existing images). >>> >>> >>> On 8/17/21 2:24 PM, Mich Talebzadeh wrote: >>> >>> Examples: >>> >>> *docker images* >>> >>> REPOSITORY TAG IMAGE ID >>> CREATED SIZE >>> >>> spark/spark-py 3.1.1_sparkpy_3.7-scala_2.12-java8 ba3c17bc9337 2 >>> minutes ago 2.19GB >>> >>> spark 3.1.1-scala_2.12-java11 4595c4e78879 18 >>> minutes ago 635MB >>> >>> >>> view my Linkedin profile >>> <https://www.linkedin.com/in/mich-talebzadeh-ph-d-5205b2/> >>> >>> >>> >>> *Disclaimer:* Use it at your own risk. Any and all responsibility for >>> any loss, damage or destruction of data or any other property which may >>> arise from relying on this email's technical content is explicitly >>> disclaimed. The author will in no case be liable for any monetary damages >>> arising from such loss, damage or destruction. >>> >>> >>> >>> >>> On Tue, 17 Aug 2021 at 10:31, Mich Talebzadeh <mich.talebza...@gmail.com> >>> wrote: >>> >>>> 3.1.2_sparkpy_3.7-scala_2.12-java11 >>>> >>>> 3.1.2_sparkR_3.6-scala_2.12-java11 >>>> Yes let us go with that and remember that we can change the tags >>>> anytime. The accompanying release note should detail what is inside the >>>> image downloaded. >>>> >>>> +1 for me >>>> >>>> >>>> view my Linkedin profile >>>> <https://www.linkedin.com/in/mich-talebzadeh-ph-d-5205b2/> >>>> >>>> >>>> >>>> *Disclaimer:* Use it at your own risk. Any and all responsibility for >>>> any loss, damage or destruction of data or any other property which may >>>> arise from relying on this email's technical content is explicitly >>>> disclaimed. The author will in no case be liable for any monetary damages >>>> arising from such loss, damage or destruction. >>>> >>>> >>>> >>>> >>>> On Tue, 17 Aug 2021 at 09:51, Maciej <mszymkiew...@gmail.com> wrote: >>>> >>>>> On 8/17/21 4:04 AM, Holden Karau wrote: >>>>> >>>>> These are some really good points all around. >>>>> >>>>> I think, in the interest of simplicity, well start with just the 3 >>>>> current Dockerfiles in the Spark repo but for the next release (3.3) we >>>>> should explore adding some more Dockerfiles/build options. >>>>> >>>>> Sounds good. >>>>> >>>>> However, I'd consider adding guest lang version to the tag names, i.e. >>>>> >>>>> 3.1.2_sparkpy_3.7-scala_2.12-java11 >>>>> >>>>> 3.1.2_sparkR_3.6-scala_2.12-java11 >>>>> >>>>> and some basics safeguards in the layers, to make sure that these are >>>>> really the versions we use. >>>>> >>>>> On Mon, Aug 16, 2021 at 10:46 AM Maciej <mszymkiew...@gmail.com> >>>>> wrote: >>>>> >>>>>> I have a few concerns regarding PySpark and SparkR images. >>>>>> >>>>>> First of all, how do we plan to handle interpreter versions? Ideally, >>>>>> we should provide images for all supported variants, but based on the >>>>>> preceding discussion and the proposed naming convention, I assume it is >>>>>> not >>>>>> going to happen. If that's the case, it would be great if we could fix >>>>>> interpreter versions based on some support criteria (lowest supported, >>>>>> lowest non-deprecated, highest supported at the time of release, etc.) >>>>>> >>>>>> Currently, we use the following: >>>>>> >>>>>> - for R use buster-cran35 Debian repositories which install R 3.6 >>>>>> (provided version already changed in the past and broke image build ‒ >>>>>> SPARK-28606). >>>>>> - for Python we depend on the system provided python3 packages, >>>>>> which currently provides Python 3.7. >>>>>> >>>>>> which don't guarantee stability over time and might be hard to >>>>>> synchronize with our support matrix. >>>>>> >>>>>> Secondly, omitting libraries which are required for the full >>>>>> functionality and performance, specifically >>>>>> >>>>>> - Numpy, Pandas and Arrow for PySpark >>>>>> - Arrow for SparkR >>>>>> >>>>>> is likely to severely limit usability of the images (out of these, >>>>>> Arrow is probably the hardest to manage, especially when you already >>>>>> depend >>>>>> on system packages to provide R or Python interpreter). >>>>>> >>>>>> On 8/14/21 12:43 AM, Mich Talebzadeh wrote: >>>>>> >>>>>> Hi, >>>>>> >>>>>> We can cater for multiple types (spark, spark-py and spark-r) and >>>>>> spark versions (assuming they are downloaded and available). >>>>>> The challenge is that these docker images built are snapshots. They >>>>>> cannot be amended later and if you change anything by going inside >>>>>> docker, >>>>>> as soon as you are logged out whatever you did is reversed. >>>>>> >>>>>> For example, I want to add tensorflow to my docker image. These are >>>>>> my images >>>>>> >>>>>> REPOSITORY TAG IMAGE ID >>>>>> CREATED SIZE >>>>>> eu.gcr.io/axial-glow-224522/spark-py java8_3.1.1 >>>>>> cfbb0e69f204 5 days ago 2.37GB >>>>>> eu.gcr.io/axial-glow-224522/spark 3.1.1 >>>>>> 8d1bf8e7e47d 5 days ago 805MB >>>>>> >>>>>> using image ID I try to log in as root to the image >>>>>> >>>>>> *docker run -u0 -it cfbb0e69f204 bash* >>>>>> >>>>>> root@b542b0f1483d:/opt/spark/work-dir# pip install keras >>>>>> Collecting keras >>>>>> Downloading keras-2.6.0-py2.py3-none-any.whl (1.3 MB) >>>>>> |████████████████████████████████| 1.3 MB 1.1 MB/s >>>>>> Installing collected packages: keras >>>>>> Successfully installed keras-2.6.0 >>>>>> WARNING: Running pip as the 'root' user can result in broken >>>>>> permissions and conflicting behaviour with the system package manager. It >>>>>> is recommended to use a virtual environment instead: >>>>>> https://pip.pypa.io/warnings/venv >>>>>> root@b542b0f1483d:/opt/spark/work-dir# pip list >>>>>> Package Version >>>>>> ------------- ------- >>>>>> asn1crypto 0.24.0 >>>>>> cryptography 2.6.1 >>>>>> cx-Oracle 8.2.1 >>>>>> entrypoints 0.3 >>>>>> *keras 2.6.0 <--- it is here* >>>>>> keyring 17.1.1 >>>>>> keyrings.alt 3.1.1 >>>>>> numpy 1.21.1 >>>>>> pip 21.2.3 >>>>>> py4j 0.10.9 >>>>>> pycrypto 2.6.1 >>>>>> PyGObject 3.30.4 >>>>>> pyspark 3.1.2 >>>>>> pyxdg 0.25 >>>>>> PyYAML 5.4.1 >>>>>> SecretStorage 2.3.1 >>>>>> setuptools 57.4.0 >>>>>> six 1.12.0 >>>>>> wheel 0.32.3 >>>>>> root@b542b0f1483d:/opt/spark/work-dir# exit >>>>>> >>>>>> Now I exited from the image and try to log in again >>>>>> (pyspark_venv) hduser@rhes76: /home/hduser/dba/bin/build> docker run >>>>>> -u0 -it cfbb0e69f204 bash >>>>>> >>>>>> root@5231ee95aa83:/opt/spark/work-dir# pip list >>>>>> Package Version >>>>>> ------------- ------- >>>>>> asn1crypto 0.24.0 >>>>>> cryptography 2.6.1 >>>>>> cx-Oracle 8.2.1 >>>>>> entrypoints 0.3 >>>>>> keyring 17.1.1 >>>>>> keyrings.alt 3.1.1 >>>>>> numpy 1.21.1 >>>>>> pip 21.2.3 >>>>>> py4j 0.10.9 >>>>>> pycrypto 2.6.1 >>>>>> PyGObject 3.30.4 >>>>>> pyspark 3.1.2 >>>>>> pyxdg 0.25 >>>>>> PyYAML 5.4.1 >>>>>> SecretStorage 2.3.1 >>>>>> setuptools 57.4.0 >>>>>> six 1.12.0 >>>>>> wheel 0.32.3 >>>>>> >>>>>> *Hm that keras is not there*. The docker Image cannot be altered >>>>>> after build! So once the docker image is created that is just a snapshot. >>>>>> However, it will still have tons of useful stuff for most >>>>>> users/organisations. My suggestions is to create for a given type (spark, >>>>>> spark-py etc): >>>>>> >>>>>> >>>>>> 1. One vanilla flavour for everyday use with few useful packages >>>>>> 2. One for medium use with most common packages for ETL/ELT stuff >>>>>> 3. One specialist for ML etc with keras, tensorflow and anything >>>>>> else needed >>>>>> >>>>>> >>>>>> These images should be maintained as we currently maintain spark >>>>>> releases with accompanying documentation. Any reason why we cannot >>>>>> maintain >>>>>> ourselves? >>>>>> >>>>>> HTH >>>>>> >>>>>> view my Linkedin profile >>>>>> <https://www.linkedin.com/in/mich-talebzadeh-ph-d-5205b2/> >>>>>> >>>>>> >>>>>> >>>>>> *Disclaimer:* Use it at your own risk. Any and all responsibility >>>>>> for any loss, damage or destruction of data or any other property which >>>>>> may >>>>>> arise from relying on this email's technical content is explicitly >>>>>> disclaimed. The author will in no case be liable for any monetary damages >>>>>> arising from such loss, damage or destruction. >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> On Fri, 13 Aug 2021 at 17:26, Holden Karau <hol...@pigscanfly.ca> >>>>>> wrote: >>>>>> >>>>>>> So we actually do have a script that does the build already it's >>>>>>> more a matter of publishing the results for easier use. Currently the >>>>>>> script produces three images spark, spark-py, and spark-r. I can >>>>>>> certainly >>>>>>> see a solid reason to publish like with a jdk11 & jdk8 suffix as well if >>>>>>> there is interest in the community. If we want to have a say >>>>>>> spark-py-pandas for a Spark container image with everything necessary >>>>>>> for >>>>>>> the Koalas stuff to work then I think that could be a great PR from >>>>>>> someone >>>>>>> to add :) >>>>>>> >>>>>>> On Fri, Aug 13, 2021 at 1:00 AM Mich Talebzadeh < >>>>>>> mich.talebza...@gmail.com> wrote: >>>>>>> >>>>>>>> should read PySpark >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> view my Linkedin profile >>>>>>>> <https://www.linkedin.com/in/mich-talebzadeh-ph-d-5205b2/> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> *Disclaimer:* Use it at your own risk. Any and all responsibility >>>>>>>> for any loss, damage or destruction of data or any other property >>>>>>>> which may >>>>>>>> arise from relying on this email's technical content is explicitly >>>>>>>> disclaimed. The author will in no case be liable for any monetary >>>>>>>> damages >>>>>>>> arising from such loss, damage or destruction. >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> On Fri, 13 Aug 2021 at 08:51, Mich Talebzadeh < >>>>>>>> mich.talebza...@gmail.com> wrote: >>>>>>>> >>>>>>>>> Agreed. >>>>>>>>> >>>>>>>>> I have already built a few latest for Spark and PYSpark on 3.1.1 >>>>>>>>> with Java 8 as I found out Java 11 does not work with Google BigQuery >>>>>>>>> data >>>>>>>>> warehouse. However, to hack the Dockerfile one finds out the hard way. >>>>>>>>> >>>>>>>>> For example how to add additional Python libraries like tensorflow >>>>>>>>> etc. Loading these libraries through Kubernetes is not practical as >>>>>>>>> unzipping and installing it through --py-files etc will take >>>>>>>>> considerable >>>>>>>>> time so they need to be added to the dockerfile at the built time in >>>>>>>>> directory for Python under Kubernetes >>>>>>>>> >>>>>>>>> /opt/spark/kubernetes/dockerfiles/spark/bindings/python >>>>>>>>> >>>>>>>>> RUN pip install pyyaml numpy cx_Oracle tensorflow .... >>>>>>>>> >>>>>>>>> Also you will need curl to test the ports from inside the docker >>>>>>>>> >>>>>>>>> RUN apt-get update && apt-get install -y curl >>>>>>>>> RUN ["apt-get","install","-y","vim"] >>>>>>>>> >>>>>>>>> As I said I am happy to build these specific dockerfiles plus the >>>>>>>>> complete documentation for it. I have already built one for Google >>>>>>>>> (GCP). >>>>>>>>> The difference between Spark and PySpark version is that in >>>>>>>>> Spark/scala a >>>>>>>>> fat jar file will contain all needed. That is not the case with >>>>>>>>> Python I am >>>>>>>>> afraid. >>>>>>>>> >>>>>>>>> HTH >>>>>>>>> >>>>>>>>> >>>>>>>>> view my Linkedin profile >>>>>>>>> <https://www.linkedin.com/in/mich-talebzadeh-ph-d-5205b2/> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *Disclaimer:* Use it at your own risk. Any and all responsibility >>>>>>>>> for any loss, damage or destruction of data or any other property >>>>>>>>> which may >>>>>>>>> arise from relying on this email's technical content is explicitly >>>>>>>>> disclaimed. The author will in no case be liable for any monetary >>>>>>>>> damages >>>>>>>>> arising from such loss, damage or destruction. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> On Fri, 13 Aug 2021 at 08:13, Bode, Meikel, NMA-CFD < >>>>>>>>> meikel.b...@bertelsmann.de> wrote: >>>>>>>>> >>>>>>>>>> Hi all, >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> I am Meikel Bode and only an interested reader of dev and user >>>>>>>>>> list. Anyway, I would appreciate to have official docker images >>>>>>>>>> available. >>>>>>>>>> >>>>>>>>>> Maybe one could get inspiration from the Jupyter docker stacks >>>>>>>>>> and provide an hierarchy of different images like this: >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> https://jupyter-docker-stacks.readthedocs.io/en/latest/using/selecting.html#image-relationships >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> Having a core image only supporting Java, an extended supporting >>>>>>>>>> Python and/or R etc. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> Looking forward to the discussion. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> Best, >>>>>>>>>> >>>>>>>>>> Meikel >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> *From:* Mich Talebzadeh <mich.talebza...@gmail.com> >>>>>>>>>> *Sent:* Freitag, 13. August 2021 08:45 >>>>>>>>>> *Cc:* dev <dev@spark.apache.org> >>>>>>>>>> *Subject:* Re: Time to start publishing Spark Docker Images? >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> I concur this is a good idea and certainly worth exploring. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> In practice, preparing docker images as deployable will throw >>>>>>>>>> some challenges because creating docker for Spark is not really a >>>>>>>>>> singular >>>>>>>>>> modular unit, say creating docker for Jenkins. It involves different >>>>>>>>>> versions and different images for Spark and PySpark and most likely >>>>>>>>>> will >>>>>>>>>> end up as part of Kubernetes deployment. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> Individuals and organisations will deploy it as the first cut. >>>>>>>>>> Great but I equally feel that good documentation on how to build a >>>>>>>>>> consumable deployable image will be more valuable. FRom my own >>>>>>>>>> experience >>>>>>>>>> the current documentation should be enhanced, for example how to >>>>>>>>>> deploy >>>>>>>>>> working directories, additional Python packages, build with >>>>>>>>>> different Java >>>>>>>>>> versions (version 8 or version 11) etc. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> HTH >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> view my Linkedin profile >>>>>>>>>> <https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.linkedin.com%2Fin%2Fmich-talebzadeh-ph-d-5205b2%2F&data=04%7C01%7CMeikel.Bode%40bertelsmann.de%7Cd97d97be540246aa975308d95e260c99%7C1ca8bd943c974fc68955bad266b43f0b%7C0%7C0%7C637644339790679755%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=0CkL3HZo9FNVUOnLQ4CYs29Z9HfrwE4xDqLgVmMbr10%3D&reserved=0> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> *Disclaimer:* Use it at your own risk. Any and all >>>>>>>>>> responsibility for any loss, damage or destruction of data or any >>>>>>>>>> other >>>>>>>>>> property which may arise from relying on this email's technical >>>>>>>>>> content is >>>>>>>>>> explicitly disclaimed. The author will in no case be liable for any >>>>>>>>>> monetary damages arising from such loss, damage or destruction. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> On Fri, 13 Aug 2021 at 01:54, Holden Karau <hol...@pigscanfly.ca> >>>>>>>>>> wrote: >>>>>>>>>> >>>>>>>>>> Awesome, I've filed an INFRA ticket to get the ball rolling. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> On Thu, Aug 12, 2021 at 5:48 PM John Zhuge <jzh...@apache.org> >>>>>>>>>> wrote: >>>>>>>>>> >>>>>>>>>> +1 >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> On Thu, Aug 12, 2021 at 5:44 PM Hyukjin Kwon <gurwls...@gmail.com> >>>>>>>>>> wrote: >>>>>>>>>> >>>>>>>>>> +1, I think we generally agreed upon having it. Thanks Holden for >>>>>>>>>> headsup and driving this. >>>>>>>>>> >>>>>>>>>> +@Dongjoon Hyun <dongj...@apache.org> FYI >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> 2021년 7월 22일 (목) 오후 12:22, Kent Yao <yaooq...@gmail.com>님이 작성: >>>>>>>>>> >>>>>>>>>> +1 >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> Bests, >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> *Kent Yao* >>>>>>>>>> >>>>>>>>>> @ Data Science Center, Hangzhou Research Institute, NetEase Corp. >>>>>>>>>> >>>>>>>>>> *a spark* *enthusiast* >>>>>>>>>> >>>>>>>>>> *kyuubi >>>>>>>>>> <https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fyaooqinn%2Fkyuubi&data=04%7C01%7CMeikel.Bode%40bertelsmann.de%7Cd97d97be540246aa975308d95e260c99%7C1ca8bd943c974fc68955bad266b43f0b%7C0%7C0%7C637644339790679755%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=ZkE%2BAK4%2BUO9JsDzZlAfY5gsATCVm5hidLCp7EGxAWiY%3D&reserved=0>**is >>>>>>>>>> a unified* *multi-tenant* *JDBC interface for large-scale data >>>>>>>>>> processing and analytics,* *built on top of* *Apache Spark >>>>>>>>>> <https://eur02.safelinks.protection.outlook.com/?url=http%3A%2F%2Fspark.apache.org%2F&data=04%7C01%7CMeikel.Bode%40bertelsmann.de%7Cd97d97be540246aa975308d95e260c99%7C1ca8bd943c974fc68955bad266b43f0b%7C0%7C0%7C637644339790689711%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=4YYZ61B6datdx2GsxqnEUOpYuJUn35egYRQSVnUxtF0%3D&reserved=0>* >>>>>>>>>> *.* >>>>>>>>>> *spark-authorizer >>>>>>>>>> <https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fyaooqinn%2Fspark-authorizer&data=04%7C01%7CMeikel.Bode%40bertelsmann.de%7Cd97d97be540246aa975308d95e260c99%7C1ca8bd943c974fc68955bad266b43f0b%7C0%7C0%7C637644339790689711%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=P6TMaSh7UeXVyv79RiRqdBpipaIjh2o3DhRs0GGhWF4%3D&reserved=0>**A >>>>>>>>>> Spark SQL extension which provides SQL Standard Authorization for* >>>>>>>>>> >>>>>>>>>> -- It's dark in this basement.