IRC channel logs

<elflng>(because it sounds like you have a 64 bit executable thats downcompiled, rather than a 32 bit executable thats upcompiled)

<bonz060>Here's what I have: `/gnu/store/l21xwala4flah4j1cv7vin9gh986js60-python2-2.4.6/bin/python: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked, interpreter /gnu/store/z4li262il798hbl0l1h1k3a5g7r6bffa-glibc-2.31/lib/ld-linux.so.2, for GNU/Linux 2.6.32, not stripped`

<elflng>is the ld-linux 64 or 32?

<elflng>all of the linked libs for mine are 32

<elflng>... look at the 2.6 . there are totally different methods for the mult and expr

<elflng>the main difference is everything is cast to longs to start with, which avoids the problem.

<elflng>so upcasting and then downcasting if necessary should fix the problem.

<elflng>i cant really spend time on this today - its erev Shabbos - but ill look bli neder tomorrow night.

<civodul>rekado: re https://medium.com/@QuantStack/open-software-packaging-for-science-61cecee7fc23 the QuantStack folks have a nice narrative around "scientific software" in general

<civodul>conda/mamba, Jupyter, Binder, etc.

<civodul>also xtensor & co.

<civodul>numpy and all

<civodul>and all these tools are extremely popular

<civodul>which is nice, it's nice that "Python Science" is incrementally succeeding at putting MATLAB and all that stuff to the trash

<bonz060>elflng: That's ok. I've had a power outage where I live so I've been AFK my whole day :(

<bonz060>Here's what I have: https://paste.debian.net/1157704/ <-- tthe ld-linux is 32 bit

<bonz060>I'll hack around this over the weekend.

<bonz060>Tbh, it's really cool working on something that existed when I was but a young kid :)

<bonz060>That's really cool heheh

<civodul>a youngster, lucky you! ;-)

IRC channel logs

2020-07-24.log