compiling errors

[ScenicBiotech]

Hi Maarten,

it's me again :) I went through the config step of screen-analyzer, and everything looks good.

Now, when I run make I get the following:
make_screen-analyzer.log

should I set some environment variables?

Best,

Francesco

[mhekkel]

You're almost there. Hang on. I'm afraid you installed the wrong yarn. You should install the one from https://classic.yarnpkg.com/en/docs/install/#debian-stable And then run 'yarn' in the source directory before running make again.

…

-maarten Op 30-09-2020 om 16:31 schreef ScenicBiotech:

Hi Maarten, it's me again :) I went through the config step of screen-analyzer, and everything looks good. Now, when I run make I get the following: make_screen-analyzer.log <https://github.com/mhekkel/screen-analyzer/files/5306205/make_screen-analyzer.log> should I set some environment variables? Best, Francesco — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#4>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADNA47E2TEHILKIURSUAP4LSIM6MHANCNFSM4R7FNXLA>.

-- Maarten L. Hekkelman Cataloniëstraat 3 6663NJ Lent http://www.hekkelman.com/ +31 24 348 0192

[ScenicBiotech]

It should be the last few issues by now. I followed your indications, and installed yarn v1.22.5, and run yarn prior to running make. Everything looks fine. after running make, quite a number of targets get compiled successfully, but I get still one error:

make_screen-analyzer_2.log

let me know what you think of this one.

Best,

Francesco

[mhekkel]

ah yes... you are using boost 1.73, right? Something changed in Boost ASIO and apparently I do not deal with that properly. I'll have to write a fix for this. Might well take until Monday, since my weekend just started.

…

-maarten Op 30-09-2020 om 17:04 schreef ScenicBiotech:

It should be the last few issues by now. I followed your indications, and installed yarn v1.22.5, and run yarn prior to running make. Everything looks fine. after running make, quite a number of targets get compiled successfully, but I get still one error: make_screen-analyzer_2.log <https://github.com/mhekkel/screen-analyzer/files/5306444/make_screen-analyzer_2.log> let me know what you think of this one. Best, Francesco — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#4 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADNA47EJMSQ6DZWNPVBF2YTSINCGBANCNFSM4R7FNXLA>.

-- Maarten L. Hekkelman Cataloniëstraat 3 6663NJ Lent http://www.hekkelman.com/ +31 24 348 0192

[mhekkel]

Goedemorgen Francesco, OK, you have a problem. boost::asio::io_context is a class that was added to boost in version 1.67. You said earlier you manually installed boost 1.73. However, the error message indicates the code is compiled using an older version of boost, since the compiler claims boost::asio::io_context is not known. That means you have both the new and the old version of boost installed and now the old headers are being picked up. That's not going to work. So either use a clean build environment (a virtual machine e.g.) or throw away the development version of the system boost you have installed.

…

-maarten Op 30-09-2020 om 17:04 schreef ScenicBiotech:

It should be the last few issues by now. I followed your indications, and installed yarn v1.22.5, and run yarn prior to running make. Everything looks fine. after running make, quite a number of targets get compiled successfully, but I get still one error: make_screen-analyzer_2.log <https://github.com/mhekkel/screen-analyzer/files/5306444/make_screen-analyzer_2.log> let me know what you think of this one. Best, Francesco — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#4 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADNA47EJMSQ6DZWNPVBF2YTSINCGBANCNFSM4R7FNXLA>.

-- Maarten L. Hekkelman Cataloniëstraat 3 6663NJ Lent http://www.hekkelman.com/ +31 24 348 0192

[ScenicBiotech]

Good morning Maarten,

I solved the problem with boost. I got rid of the installation through package manager and I installed boost 1.73.
So I am past the last error I showed you. now, I get this:

make_screen-analyzer_3.log

let me know.
Best,

Francesco

[mhekkel]

Include the line #include <iomanip> at the top of utils.cpp

…

-maarten Op 05-10-2020 om 09:40 schreef ScenicBiotech:

Good morning Maarten, I solved the problem with boost. I got rid of the installation through package manager and I installed boost 1.73. So I am past the last error I showed you. now, I get this: make_screen-analyzer_3.log <https://github.com/mhekkel/screen-analyzer/files/5325897/make_screen-analyzer_3.log> let me know. Best, Francesco — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#4 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADNA47DMVK4BRBA74BFGRJ3SJFZ6JANCNFSM4R7FNXLA>.

-- Maarten L. Hekkelman Cataloniëstraat 3 6663NJ Lent http://www.hekkelman.com/ +31 24 348 0192

[ScenicBiotech]

Hi Maarten, not there yet. Now, everything compiles correctly. Then, at then end of the process it complains that :
undefined reference to `boost::system::generic_category()'.

see detailed log hereunder.
make_screen-analyzer_4.log

I don't have an easy way to create a VM, so I would prefer not to start from a clean env but instead disentangle the dependency hell in the current system. If this becomes really impossible, I will talk to our IT and ask for support.

Best,

Francesco.

[mhekkel]

First of all, did you 'make clean' before getting rid of the system boost? If not, you compiled some sources against the old header files and the symbols cannot be found in the newer libs. If that's not the case, the linker might pick up the boost libraries from the wrong location. Anyway, boost::system is used by older versions of boost, so perhaps you could start by making clean and making again. Just to be sure.

…

-maarten Op 05-10-2020 om 12:26 schreef ScenicBiotech:

Hi Maarten, not there yet. Now, everything compiles correctly. Then, at then end of the process it complains that : undefined reference to `boost::system::generic_category()'. see detailed log hereunder. make_screen-analyzer_4.log <https://github.com/mhekkel/screen-analyzer/files/5326708/make_screen-analyzer_4.log> I don't have an easy way to create a VM, so I would prefer not to start from a clean env but instead disentangle the dependency hell in the current system. If this becomes really impossible, I will talk to our IT and ask for support. Best, Francesco. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#4 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADNA47ECZFH3B3ACEPVSPTLSJGNOFANCNFSM4R7FNXLA>.

-- Maarten L. Hekkelman Cataloniëstraat 3 6663NJ Lent http://www.hekkelman.com/ +31 24 348 0192