[zeromq-dev] cppzmq revival and RFC on design goals and supported platforms
Simon.Giesecke at btc-ag.com
Simon.Giesecke at btc-ag.com
Fri May 25 09:42:07 CEST 2018
Hi Harald,
* test framework:
Personally I have no particular preference for googletest, and at first sight I think we could go with doctest as well. I agree that not needing a build step is a plus, in particular for a project like cppzmq, which is a header-only library itself. However, someone would need to migrate the existing tests. Would you be willing to do that? That would be great. Maybe we should first seek if someone else has a particular preference for googletest or against doctest, which should be taken into account.
* const:
Thanks for making this clear, now I understand. I am aware of the discussion, but was not aware of the catchphrase "east const". I agree this should be done consistently. I had a lot of discussions about this in the past as well, and I think there is no reason to diverge from the recommendation in the C++ Core Guidelines (which means “west const”, right? ;) ). We should start some coding conventions, which need not be comprehensive, but list anything that was subject to debate, like this.
Best regards
Simon
Von: zeromq-dev [mailto:zeromq-dev-bounces at lists.zeromq.org] Im Auftrag von Harald Achitz
Gesendet: Donnerstag, 24. Mai 2018 10:33
An: ZeroMQ development list <zeromq-dev at lists.zeromq.org>
Betreff: Re: [zeromq-dev] cppzmq revival and RFC on design goals and supported platforms
Hi Simon
I am on mobile so a bit restricted. But in short.
ad testing, if gtest is used it should be an installed libary, but google has forgotten what ABI compatility is so they recommand downloading and adding and rebuilding gtest into the project you use.
scaling this to my work this would mean >100 times download and rebuild for nothing, google does not care about build times, this is a solved problem (original zitat Titus Winters)
but I have to care.
alternative:
https://github.com/onqtam/doctest<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fonqtam%2Fdoctest&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7C2916f028766c46380f1f08d5c1510f2d%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C1%7C636627476260537915&sdata=XiZoDgchvi7RXRP9HTnNtXuA9gsR4RLFgUGp4nNWXTQ%3D&reserved=0>
just add the header, now download from the interenet required
if you want to know how to accelerate testing with doctest, I have written about that recently
https://a4z.bitbucket.io/blog/2018/05/17/Speed-up-your-test-cycles-with-CMake.html<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fa4z.bitbucket.io%2Fblog%2F2018%2F05%2F17%2FSpeed-up-your-test-cycles-with-CMake.html&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7C2916f028766c46380f1f08d5c1510f2d%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C1%7C636627476260537915&sdata=A476WGPWGDs%2Bx5727sTYwsc54E%2B4ynaXCiVlOZGIH%2Fo%3D&reserved=0>
east const, for exampel row 165 and the code below
zmq_pollitem_t const *items
instead of
const zmq_pollitem_t * items
this is a nerdy, and also funny, discussion in the C++ cummunity, but I also find it unproductive.
Stroustrup / Sutter and co have stated their opinion in the CppCore guildelines
https://github.com/isocpp/CppCoreGuidelines/blob/master/CppCoreGuidelines.md#Rl-const<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fisocpp%2FCppCoreGuidelines%2Fblob%2Fmaster%2FCppCoreGuidelines.md%23Rl-const&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7C2916f028766c46380f1f08d5c1510f2d%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C1%7C636627476260537915&sdata=oIUByoPyYSa6vwzRpG%2Bcyth3TPb6gzI%2Fze3eP5LVjBk%3D&reserved=0>
pre C++11, somewhen needs to be the cut.
make a branch, put the old version there, forget it
I hope this gives you a bit more info and helps.
Regards,
Harald (now with pain in the fingers:-)
On Thu, 24 May 2018 at 10:10, <Simon.Giesecke at btc-ag.com<mailto:Simon.Giesecke at btc-ag.com>> wrote:
Hi Harald,
thanks for your mail. Unfortunately, I fear you need to elaborate a bit more on your points so that I understand them.
* It seems google is infecting anyone with it’s monolithic build philosophy.
Too sad that you decided to fetch gtest.
What do you mean by “monolithic build philosophy”? What problems do you see here? Which test framework would you have used? Do you think it is worthwhile to migrate?
Note that for “building” cppzmq without tests, nothing should have changed (actually, there is nothing to build since it is header-only).
* Going east const might feel some more enlightened than others. But now the file has 2 styles.
Sorry, but I totally don’t get what you mean with this remark. Can you explain this?
* I also think that pre C++ 11 Support pollutes the code. It doesn’t provide the same interface and functionality anyway. So why care?
The addition of parts that require C++11 support started before I got into contact with cppzmq. I also think that it would be cleaner to use the same standard version everywhere, but I don’t want to break existing users. That’s why I asked if there are users that still require pre-C++11 support.
Best regards
Simon
Von: zeromq-dev [mailto:zeromq-dev-bounces at lists.zeromq.org<mailto:zeromq-dev-bounces at lists.zeromq.org>] Im Auftrag von Harald Achitz
Gesendet: Donnerstag, 24. Mai 2018 09:40
An: ZeroMQ development list <zeromq-dev at lists.zeromq.org<mailto:zeromq-dev at lists.zeromq.org>>
Betreff: Re: [zeromq-dev] cppzmq revival and RFC on design goals and supported platforms
Some critics, hope this is ok.
It seems google is infecting anyone with it’s monolithic build philosophy.
Too sad that you decided to fetch gtest.
Going east const might feel some more enlightened than others. But now the file has 2 styles.
I also think that pre C++ 11 Support pollutes the code. It doesn’t provide the same interface and functionality anyway. So why care?
I appreciate your efforts and I am happy that someone takes care about this.
But it also seems to be the right time to focus on a own adopted version of this header.
Regards
Harald
On Thu, 24 May 2018 at 09:13, <Simon.Giesecke at btc-ag.com<mailto:Simon.Giesecke at btc-ag.com>> wrote:
Hi Ernest,
thanks for your mail!
I am not sure if I get exactly what you want to achieve by detaching. Maybe open an issue on github with some code sketch of how you would use that, and/or a PR that implements it? I am not sure what kind of “hack” you refer to. If it is something that could be easily misused, it might be better not to include it.
Regarding your second question, well, the community is in charge ;) I personally am not working on the zguide. If you would like to improve that, it would be very welcome.
Best wishes
Simon
Von: zeromq-dev [mailto:zeromq-dev-bounces at lists.zeromq.org<mailto:zeromq-dev-bounces at lists.zeromq.org>] Im Auftrag von Ernest Zed
Gesendet: Donnerstag, 24. Mai 2018 06:36
An: ZeroMQ development list <zeromq-dev at lists.zeromq.org<mailto:zeromq-dev at lists.zeromq.org>>
Betreff: Re: [zeromq-dev] cppzmq revival and RFC on design goals and supported platforms
Hi,
It is a blessed move... so far, I'm missing few things, one of it, ability to detach pointer from received message and pass ownership to another object. I know it is not achievable by zmq, but there is a hack to implement it.
Second, who is in charge of C++ examples? As I've reported before, the Paranoid Pirate example doesnt work.
Sincerely,
Ernest
On Wed, May 23, 2018 at 10:09 PM, Gyorgy Szekely <hoditohod at gmail.com<mailto:hoditohod at gmail.com>> wrote:
Hi Simon,
This is great news! We're using cppzmq in a message broker and an accompanying communication library for 2 years now.
I fully agree with the declared goals. libzmq has a simple and concise API with object oriented mindset. It works well on its own, but cppzmq makes it a whole lot easier. What's particularly good about it:
- type safety and RAII: it's very straigtforward to think in classes that properly clean-up resources at destruction
- higher level functions: multipart messages are really nice, though the API is/was a bit inconsistent (socket.send(msg) vs, msg.send(socket))
- header only, it's very easy to use. Header only libraries usually mean template heavy monsters, but fortunately not in this case
What I personally really like is it's a thin wrapper and doesn't want to be more than libzmq. Methods usually map 1-to-1 to libzmq calls, there's no hidden trickery and the documentation at api.zeromq.org<https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapi.zeromq.org&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7C90d876b632924bc4519b08d5c12feb4f%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C0%7C636627333917558531&sdata=JrTOCmytDVgVHiCOffOAv3pQ6I%2BWOJ7hVeiefD0CjQw%3D&reserved=0> is fully relevant.
I haven't checked the recent updates (yet), but I found a few strange bits while working with cppzmq. Like the above mentioned sending inconsistency, or having to cast the socket to void* to use it in a pollset. Apart from that I completely agree with the direction. This is how a thin C++ wrapper should look like for a good base C API.
BTW, we're using the lib on Ubuntu16.04 64bit / G++ 5.3, no issues so far.
Regards,
Gyorgy
On Wed, May 23, 2018 at 6:07 PM, <Simon.Giesecke at btc-ag.com<mailto:Simon.Giesecke at btc-ag.com>> wrote:
Hi,
Pawel Kurdybacha (kurdybacha) and me (sigiesec) have recently started to "revive" cppzmq (https://github.com/zeromq/cppzmq<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fzeromq%2Fcppzmq&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7C90d876b632924bc4519b08d5c12feb4f%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C0%7C636627333917558531&sdata=31MtvPSLH6oBgRyHbkUQOeRCF1hAWBbqjK%2F31hFzutk%3D&reserved=0>), the light-weight C++ wrapper around libzmq. We added CI for Windows/MSVC, Linux and MacOS, implemented tests, cleaned up the CMake infrastructure, formatted the source code consistently and added some overview documentation.
If you are using cppzmq or are interested in using it, we encourage you to have a look at the recent changes.
One particular point we would like to seek feedback on are the design goals, which have recently been documented for the first time. I tried to extrapolate them from the actual design, and from the reasons we chose to use cppzmq in comparison to other alternatives. These are part of the https://github.com/zeromq/cppzmq/blob/master/README.md<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fzeromq%2Fcppzmq%2Fblob%2Fmaster%2FREADME.md&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7C90d876b632924bc4519b08d5c12feb4f%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C0%7C636627333917558531&sdata=JuSiHQECJNykVjawgnuhXl6%2FYl6BYiuUDp6lksqEqyc%3D&reserved=0> file:
* cppzmq maps the libzmq C API to C++ concepts. In particular:
* it is type-safe (the libzmq C API exposes various class-like concepts as void*)
* it provides exception-based error handling (the libzmq C API provides errno-based error handling)
* it provides RAII-style classes that automate resource management (the libzmq C API requires the user to take care to free resources explicitly)
* cppzmq is a light-weight, header-only binding. You only need to include the header file zmq.hpp (and maybe zmq_addon.hpp) to use it.
* zmq.hpp is meant to contain direct mappings of the abstractions provided by the libzmq C API, while zmq_addon.hpp provides additional higher-level abstractions.
We would like to here from you if you agree with these design goals. If you have any opposing views, proposals for improvement or extension of the design goals, please share them on the mailing list or by sending a PR.
Another part of the README is a section on the supported platforms. Please review this section, in particular if you do not use MacOS, Linux or Windows/MSVC with a recent compiler. If you successfully use a different platform, please send a PR to include this in the list of "Additional platforms that are known to work". Support for non-C++11 compilers is already partial only, and might be removed completely, unless there are users that still require such support.
Of course, you are also invited to contribute extensions, new features, cleanup, further tests, etc. to cppzmq.
Best regards
Simon
--
i.A. Simon Giesecke
BTC Business Technology Consulting AG
Kurfürstendamm 33
10719 Berlin
E-Mail: Simon.Giesecke at btc-ag.com<mailto:Simon.Giesecke at btc-ag.com>
Rechtliche Hinweise:
www.btc-ag.com/impressum.htm<http://www.btc-ag.com/impressum.htm>
Handelsregister: Amtsgericht Oldenburg HRB 4717
Aufsichtsratsvorsitzender: Michael Heidkamp
Vorstand: Dr. Jörg Ritter, Dirk Thole
_______________________________________________
zeromq-dev mailing list
zeromq-dev at lists.zeromq.org<mailto:zeromq-dev at lists.zeromq.org>
https://lists.zeromq.org/mailman/listinfo/zeromq-dev<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.zeromq.org%2Fmailman%2Flistinfo%2Fzeromq-dev&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7C90d876b632924bc4519b08d5c12feb4f%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C0%7C636627333917558531&sdata=krDx3Oxa5X3cOdIjY7SapwP2jkDRjtOjmbybsYnhkCA%3D&reserved=0>
_______________________________________________
zeromq-dev mailing list
zeromq-dev at lists.zeromq.org<mailto:zeromq-dev at lists.zeromq.org>
https://lists.zeromq.org/mailman/listinfo/zeromq-dev<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.zeromq.org%2Fmailman%2Flistinfo%2Fzeromq-dev&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7C90d876b632924bc4519b08d5c12feb4f%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C0%7C636627333917714776&sdata=5aDm6Y6OCvXfIywP8jozDgBReWdTwxG%2BPJWIYU84ri4%3D&reserved=0>
_______________________________________________
zeromq-dev mailing list
zeromq-dev at lists.zeromq.org<mailto:zeromq-dev at lists.zeromq.org>
https://lists.zeromq.org/mailman/listinfo/zeromq-dev<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.zeromq.org%2Fmailman%2Flistinfo%2Fzeromq-dev&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7Ccaee78efa4e34195850f08d5c149a74d%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C0%7C636627444456244154&sdata=lsglpxnyjw5Zw3WWsCl7WQzQBQi3WhKtvkdhG15adk0%3D&reserved=0>
_______________________________________________
zeromq-dev mailing list
zeromq-dev at lists.zeromq.org<mailto:zeromq-dev at lists.zeromq.org>
https://lists.zeromq.org/mailman/listinfo/zeromq-dev<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.zeromq.org%2Fmailman%2Flistinfo%2Fzeromq-dev&data=02%7C01%7Csimon.giesecke%40btc-ag.com%7C2916f028766c46380f1f08d5c1510f2d%7Cc064efb078954eebb406a40bc377bc7d%7C0%7C1%7C636627476260537915&sdata=YSHs1VvABp3b5NErZ8M1ov1JS89QuL%2BUm7JN8f9WLwI%3D&reserved=0>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.zeromq.org/pipermail/zeromq-dev/attachments/20180525/5f313d1d/attachment.htm>
More information about the zeromq-dev
mailing list