« first day (1585 days earlier)      last day (3374 days later) » 

3:19 AM
@IonicăBizău I have a machine with Mavericks on it, I can take a look tomorrow
 
3:32 AM
@IonicăBizău its not an OSX issue, its a zsh issue
your git() function works on OSX with bash
your users error is reproduceable with zsh on either linux or os X
tested with bash 4.2.53 and zsh 5.0.5 on linux and bash 3.2.53 and zsh 5.0.5 on OS X
casey@convect ingest_cm1 % git status
>>> command git status
git:15: command not found: command git status
127
git:19: = not found
on linux w/ zsh
 
 
2 hours later…
5:49 AM
@casey Thanks for feedback. Have you any idea why this appears?
 
 
2 hours later…
7:59 AM
@IonicăBizău I don't see why you can't.
@casey is git installed?
 
8:40 AM
@FaheemMitha I think I will just switch on git hooks and it will be better I guess...
 
 
1 hour later…
9:42 AM
@FaheemMitha I don't have access to a machine that doesn't have git installed. Its one of my absolutely essential packages.
 
10:35 AM
@casey if git is installed, why is it saying "command not found: command git status" I wonder.
 
 
1 hour later…
12:03 PM
So much time since I came here
how everybody is doing ?
 
12:18 PM
Hi @Kiwy. Less people here than there used to be. How are you? How's the job?
 
12:44 PM
@FaheemMitha that's the bug he needs to figure out.
 
12:56 PM
@FaheemMitha still fucking awesome
I try to be here from time to time but it's hard, I have so much work :D I can't spend more than 15 minutes each day :(
 
1:55 PM
@FaheemMitha both git status and command git status work normally, just not inside of that git() function as defined
for me anyway, in zsh
 
2:43 PM
@Kiwy well, the internet is overrated. glad to know you are doing well.
 
3:00 PM
@terdon hello
 
@AndyK Hi andy
So, you want the fcvs for some and the fmvs or whatever for others?
 
There we go
Thanks for your help
I've never done text processing before
 
You're welcome, it's a fun problem.
You a bioinformatician? It looks like biological data of some sort.
 
data from Cassandra
or I guess
so.
I'ml looking for the fck=35 and the fck=78, the closest fmd's
 
OK
And for the others the closest fcv?
 
3:04 PM
and for the fck 84,85 , 86 and 83, I'm looking for the related fcv
 
OK. So can you have an input file that looks like this?
fck=35 fmd
fck=78 fcv
bnv=12 fcv
 
if they do not exist, I will need an empty column
it will be more like this
fck=35;fmd=1421687191;fck=78;fmd=1421687966;fck=84;fcv=10;fck=85;fcv=true;fck=86‌​;fcv=true;fck=83;fcv=demelog
 
@AndyK I mean can you have a list of search strings and associated target (fmd or fcv)
What you're showing me is the actual input file you want to parse.
 
what I'm showing you is the result
the whole file is like thos
 
@AndyK Yes, OK. But if you can have your search patterns saved in another file, it will make everything easier.
 
3:07 PM
this
yes of course
I tried with SED and AWK but they are not strong enough for that , in case of multiple search
one last thing
For fck=83, I can have either one of these string : naismc, decoide, decoccm, travide, travccm or mariccm
Would you need a sample of my file?
 
@casey Yes, I see.
@casey ok
 
@AndyK What do you mean? Will that be naismc=foo for example or fmd=naismc?
And yes, ideally, we'd like to see a sample of your input file (showing all possible permutations, things like multiple matches per line, none per line etc) and the output you would like to see from that file.
 
Hi @terdon. How's things?
 
@terdon: It will be like fmd=naismc
 
How do I diff (or cmp) the contents of two .tar.gz archives?
And how do I diff (or cmp) the contents of two .sql.gz archives?
 
3:12 PM
And awk is certainly strong enough, it just has clunkier syntax for this type of thing. I'm writing a perl solution.
 
@Terdon : How can I send it to you?
 
@AndyK add it to your question. But what do you mean with the naismc, decoide, decoccm, etc? Are those key names (decoccm=foo) or values (foo=decoccm)?
Hey @FaheemMitha fine, watching the snow out my window :)
 
values
like fmd=decoccm or fmd=travide
keynames are fck and fmd
 
@AndyK Ah, that should be fine then.
 
do you need that sample?
and where do you need me to send it , please?
 
3:19 PM
@AndyK It would help if you just added it to your question.
Have a look at my answer though and let me know if that's OK
 
I think we are almost there ,@terdon
if you look at my desired result , you can see that :

fck=35;fmd=1421687191;fck=78;fmd=1421687966;fck=84;fcv=10;fck=85;fcv=true;fck=86;fcv=true;fck=83;fcv=demelog
fck 35 and 78 go with fmd
 
@terdon Still in France?
 
the closest fmd
actually
and for fck 84, 85, 86 and 83, I need the fcv
the next fcv
and if you are in Paris, I buy you a beer.
 
@FaheemMitha Not any more. I moved back to Greece.
@AndyK Heh, thanks. I'll let you know next time I'm in town :)
 
@terdon Oh, got a job at some uni there?
How are things in Greece right now?
 
3:24 PM
@FaheemMitha No, still looking.
@FaheemMitha Interesting...
 
@terdon Ah, Ok.
@terdon Hmm. That's ambiguous...
 
For your solution
bnv=12;;fck=35;fmd=1421687191 fmd=1111111111;fck=78;fcv=de724a544277d79c14d19809fe51ab71;
 
@AndyK Oh, yeah, that just depends on the 1st input file. Forgot to update my answer, hang on.
 
you see fck=35 is ok but fck=78 , I've got a fcv
instead of a fmd
 
@AndyK See updated answer. Basically, for each search string (e.g. fck-35), you need a line in list.txt that has the string and what you want to find. For example:
fck=84 fcv
The script reads the targets from the first file and looks for them in the second.
 
3:26 PM
okay
Let me have a look. I keep you updated
what fields of bio informatics , are you looking at?
 
@FaheemMitha Very. We have a new government who are making very nice noises but we have no idea how they'll be able to implement what they're promising. If they manage it, I will be very pleased but so far we have no idea how.
@AndyK Genomics, Evolution and perhaps systems biology (networks).
 
@terdon All politicians are crooks and liars. But you know that already.
@terdon considered moving into straight programming/system stuff?
 
do you know this association ? hackyourphd.org/en
maybe they can help you out
say to them you are coming from Andy from la Paillasse
bottle in the sea but if you don't try, nothing happens
 
@AndyK Hmm, interesting. Thanks.
La Paillasse?
 
3:31 PM
@FaheemMitha Nah, not really. And yes, I don't trust politicians. The good news is that most of the guys in our new goverment have never held office before.
 
@terdon I wonder if that is a recommendation. At least they are presumably not seasoned scoundrels.
@terdon want to stay in research then?
 
@FaheemMitha Exactly. At least they haven't dirtied their hands yet
@AndyK Hmm, that also looks interesting, thanks.
@FaheemMitha Dunno. Probably. Maybe. Perhaps :)
 
@terdon well, nice to know you have strong opinions on the subject. :-)
 
Yeah...
@AndyK try that script out and let me know if you need anything else.
 
how does that work? Do I need to redirect the results to another file?
yeah you need to
 
3:44 PM
@AndyK Yes, for example parser.pl list.txt file.txt > output.txt
 
done
 
4:14 PM
@terdon, many thanks mate. That works. few stuff here and there to correct but it will be fine. Please let me know if you are in paris, I buy you a beer.
 
@AndyK Heh :) Glad it worked. I was living in Marseille for the past 4 years but have recently moved away so that beer might have to wait. I'll settle for the accept and gladly.
 
you need to stick in research
in dark times like we are living now, seekers of truth are always needed
 
@AndyK I probably will. Or academia anyway. What I don't want is to find myself spending all my time writing grants instead of doing research. Sigh
 
ah ...
you need to read one book
friendly advice: mastery by robert greene
it will not help you with the grant but for the rest :)
 
OK, I'll check it out.
 
4:33 PM
@AndyK you have strange notions of what happens in research. :-)
And it's always dark times. It's planet Earth.
 
4:49 PM
Lol. I got misunderstood. My bad.
 
5:02 PM
@AndyK ?
 
yes @farheem
yes @faheem
 

« first day (1585 days earlier)      last day (3374 days later) »