IRC channel logs

2015-08-01.log

back to list of logs

<amz3>sneek: later tell nalaginrut I'm trying to improve the definition of dynamic ffi bindings for my lib (the article on my blog in french). What do you think of this way of declaring bindings, can it work with other kind of c code? https://friendpaste.com/45Hol9olJmAdGnIK4mZIjQ
<sneek>Okay.
<amz3>any comment is welcome, I tried to make the paste self-explaning. The implementation is missing, but it' 2:00 AM at GMT+2 so kind of time to sleep :)
<amz3>so far so good I happy with the result, I'll add a layer of cosmectic syntax regarding the definition of the record for a given data structure
<amz3>nite!
<amz3>héllo guile :)
<amz3>I pushed my small ffi wrappers on my blog
<amz3>you can find it at http://www.hyperdev.fr/notes/2015/06/27/guile-dynamic-foreign-function-interface/
<amz3>scroll to the bottom to find the file I shared yesterday (decription of the API+implementation)
<amz3>the C file for which the bindings are done can be copy/pasted from the same page
<amz3>I added the implementation in this version, that's why it's published and replace the old article which was messy ramblings
<amz3>This experience, I think, stress the importance of writing small procedures to go forward
<adhoc>hi amz3
<amz3>hi adhoc
<amz3>(it's much shorter that the initial code :)
<adhoc>amz3: did you link me to your blog on FFI between C and Guile last week ?
<adhoc>unfortunately I didn't bookmark it.
<amz3>adhoc: I did not
<amz3>well it's in french and now a mix of english and french
<amz3>this is/was more like a "learning in progress" kind of post
<amz3>now with the procedure I've written things are more clear hopefully it's the case for others too
<amz3>the sad thing is that I don't know how to contribute otherwise to this subject directly to guile project
<amz3>nalaginrut advocates for that we have support similar to LUA (which I understand as support parsing C headers)
<amz3>the problem with this approach is that, it's only simple on the paper. From my experience using python-cffi it's not as easy as dropping an upstream header in a directory and call (automatic-dynamic-bindings "my-header.h")
<amz3>I'm rewriting the bindings of wiredtiger using the thing is describe in the blog, I'll see if it's good or not
<amz3>so far so good but I only bound two functions :))
<amz3>/luckily/ it's a quite repetitive task
<adhoc>amz3: i'm still finding my feet
<adhoc>have to start somewhere =)
<amz3>np
<amz3>ACTION bound the third function :D
<amz3>(i did a nap tough :p)