Re: [SLUG] who wrote the most code?
On Tue, 2004-02-24 at 14:59, Andrew Bennetts wrote: On Tue, Feb 24, 2004 at 02:06:46PM +1100, Jamie Wilkinson wrote: A few weeks ago, Benno asked me if I had a way to use cvs annotate to work out percentages of code written, and after a short bit of hacking, there was this: http://statcvs.sf.net/ Cute. And looking at the sample reports, there is some chance it does the right thing. Rob -- GPG key available at: http://www.robertcollins.net/keys.txt. signature.asc Description: This is a digitally signed message part -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
quote who=Jamie Wilkinson http://statcvs.sf.net/ unless this project contains a php jukebox and photo gallery, it's just wasting space; i have already proved that it is trivially implemented on the command line. It's on sourceforget, so obviously it's a *framework* for building CVS statistics. - Jeff -- GVADEC 2004: Kristiansand, Norwayhttp://2004.guadec.org/ What's up with that word though... it's like something you did to frogs in grammar school. - Ani DiFranco on bisexuality -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
This one time, at band camp, Robert Collins wrote:
On Tue, 2004-02-24 at 14:06, Jamie Wilkinson wrote:
A few weeks ago, Benno asked me if I had a way to use cvs annotate to
work out percentages of code written, and after a short bit of hacking,
there was this:
a=; t=0; cvs ann 2/dev/null | cut -c15-21 | sort | uniq -c | while read i; do
u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f1 -d' '`; t=$((${t-0} + $s)); echo
$t $u $s ; done | sort -rn | while read i ; do if [ x$a = x ]; then a=`echo $i
| cut -f1 -d' '`; fi ; u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f3 -d' '`;
s=$(($s * 100)); echo $u $(($s / $a)); done
but I was looking at the subversion repo at work today, and wondered
the same thing... so with a bit of mangling here it is:
a=; t=0; find . -type f ! -regex '.*\.svn.*' -exec svn ann {} \; 2/dev/null | awk
'{print $2}' | sort | uniq -c | while read i; do u=`echo $i | cut -f2 -d' '`;
s=`echo $i | cut -f1 -d' '`; t=$((${t-0} + $s)); echo $t $u $s ; done | sort -rn |
while read i ; do if [ x$a = x ]; then a=`echo $i | cut -f1 -d' '`; fi ;
u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f3 -d' '`; s=$(($s * 100)); echo $u
$(($s / $a)); done
Fun, eh? :-)
Pity it's completely wrong.
How? Thats an exercise for the reader. Hint: your script does what you
intended it to do, but the oresult is /not/ LOC per person.
check the manpage for the above command, you'll find that it never
claims to output LOC per person. therefore it's completely right. QED.
P.S. your (and Micksa's) email clients are broken.
--
[EMAIL PROTECTED] http://spacepants.org/jaq.gpg
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
On Tue, 2004-02-24 at 19:10, Jamie Wilkinson wrote: P.S. your (and Micksa's) email clients are broken. Oh? In what way? Rob -- GPG key available at: http://www.robertcollins.net/keys.txt. signature.asc Description: This is a digitally signed message part -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
This one time, at band camp, Robert Collins wrote: On Tue, 2004-02-24 at 19:10, Jamie Wilkinson wrote: P.S. your (and Micksa's) email clients are broken. Oh? In what way? (You're doing it again! :-) Your replies are going to the list and to me, which suggests that your mail client doesn't acknowledge the Mail-Followup-To header. -- [EMAIL PROTECTED] http://spacepants.org/jaq.gpg -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
On Tue Feb 24, 2004 at 19:21:53 +1100, Jamie Wilkinson wrote: This one time, at band camp, Robert Collins wrote: On Tue, 2004-02-24 at 19:10, Jamie Wilkinson wrote: P.S. your (and Micksa's) email clients are broken. Oh? In what way? (You're doing it again! :-) Your replies are going to the list and to me, which suggests that your mail client doesn't acknowledge the Mail-Followup-To header. Which RFC is that in? Or is it just a DJBism? (http://cr.yp.to/proto/replyto.html) Benno -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
On Tue, 2004-02-24 at 19:21, Jamie Wilkinson wrote: This one time, at band camp, Robert Collins wrote: On Tue, 2004-02-24 at 19:10, Jamie Wilkinson wrote: P.S. your (and Micksa's) email clients are broken. Oh? In what way? (You're doing it again! :-) Your replies are going to the list and to me, which suggests that your mail client doesn't acknowledge the Mail-Followup-To header. It's simpler than that I think... I hit reply-to-all as reflex - there's no reply to list icon in evolution. ;) Is that header added by the listserver, or your MUA? (And if the listserve, how does it discrimination between subscribers and non subscribers ?) Rob -- GPG key available at: http://www.robertcollins.net/keys.txt. signature.asc Description: This is a digitally signed message part -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
Jamie Wilkinson wrote: This one time, at band camp, Robert Collins wrote: On Tue, 2004-02-24 at 19:10, Jamie Wilkinson wrote: P.S. your (and Micksa's) email clients are broken. Oh? In what way? (You're doing it again! :-) Your replies are going to the list and to me, which suggests that your mail client doesn't acknowledge the Mail-Followup-To header. That wasn't my mail client, it was me clicking reply to all. Sorry. Mick. -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
On Tue, Feb 24, 2004, Robert Collins wrote: It's simpler than that I think... I hit reply-to-all as reflex - there's no reply to list icon in evolution. ;) Reply-to-all would reply to the Mail-Followup-To header only ... assuming that the client respected Mail-Followup-To. What mail clients *do* respect Mail-Followup-To, aside from mutt? It's always seemed to me to be a way to coerce the replies of my fellow mutt users, rather than anyone else. Is that header added by the listserver, or your MUA? (And if the listserve, how does it discrimination between subscribers and non subscribers ?) It's added by the MUA. -Mary -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
This one time, at band camp, Robert Collins wrote: On Tue, 2004-02-24 at 19:21, Jamie Wilkinson wrote: This one time, at band camp, Robert Collins wrote: On Tue, 2004-02-24 at 19:10, Jamie Wilkinson wrote: P.S. your (and Micksa's) email clients are broken. Oh? In what way? (You're doing it again! :-) Your replies are going to the list and to me, which suggests that your mail client doesn't acknowledge the Mail-Followup-To header. It's simpler than that I think... I hit reply-to-all as reflex - there's no reply to list icon in evolution. ;) Oh :( that sucks. Is that header added by the listserver, or your MUA? (And if the listserve, how does it discrimination between subscribers and non subscribers ?) It's set in my MUA. -- [EMAIL PROTECTED] http://spacepants.org/jaq.gpg -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
This one time, at band camp, Benno wrote: On Tue Feb 24, 2004 at 19:21:53 +1100, Jamie Wilkinson wrote: This one time, at band camp, Robert Collins wrote: On Tue, 2004-02-24 at 19:10, Jamie Wilkinson wrote: P.S. your (and Micksa's) email clients are broken. Oh? In what way? (You're doing it again! :-) Your replies are going to the list and to me, which suggests that your mail client doesn't acknowledge the Mail-Followup-To header. Which RFC is that in? Or is it just a DJBism? It's not in an RFC, afaik. I didn't know it was a djbism, that sucks :-) (but then you use Maildir). Googling for rfc mail-followup-to seems to have a few useful links about it, and it seems there's a group strongly opposing its use too (that is, until something like usenet's Followup-To from RFC 1036). -- [EMAIL PROTECTED] http://spacepants.org/jaq.gpg -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
This one time, at band camp, Robert Collins wrote: It's simpler than that I think... I hit reply-to-all as reflex - there's no reply to list icon in evolution. ;) though there is a reply to list menu action... i'd file a bug. -- [EMAIL PROTECTED] http://spacepants.org/jaq.gpg -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
On Tue, Feb 24, 2004, Jamie Wilkinson wrote: This one time, at band camp, Robert Collins wrote: It's simpler than that I think... I hit reply-to-all as reflex - there's no reply to list icon in evolution. ;) though there is a reply to list menu action... i'd file a bug. Fixed in 1.5, according to Jan. It's the dammit-they-fixed-it-in-unstable bind. -Mary -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
This one time, at band camp, Jamie Wilkinson wrote: Googling for rfc mail-followup-to seems to have a few useful links about it, and it seems there's a group strongly opposing its use too (that is, until something like usenet's Followup-To from RFC 1036). There's this nifty page at http://pm-doc.sourceforge.net/pm-tips-body.html that has a few links to papers about the more offensive mail headers and what they really mean. there's also a pile of cool procmail stuff in there... -- [EMAIL PROTECTED] http://spacepants.org/jaq.gpg -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
quote who=Mary Gardiner On Tue, Feb 24, 2004, Jamie Wilkinson wrote: This one time, at band camp, Robert Collins wrote: It's simpler than that I think... I hit reply-to-all as reflex - there's no reply to list icon in evolution. ;) though there is a reply to list menu action... i'd file a bug. Fixed in 1.5, according to Jan. It's the dammit-they-fixed-it-in-unstable bind. Here's my patch for /usr/share/evolution/1.4/ui/evolution-mail-message.xml to add reply-to-list as a keybinding to evo 1.4. ctrl-l is the new keybinding (already in 1.5) J. -- Jan Schmidt [EMAIL PROTECTED] If Darl McBride had his way, he would have banned marriage too, because it obviously is against the remunerative interests of prostitutes - Bruce Perens --- evolution-mail-message.xml.orig 2004-02-17 14:35:25.0 +1100 +++ evolution-mail-message.xml 2004-02-17 14:35:42.0 +1100 @@ -105,7 +105,7 @@ cmd name=MessageReplyList _tip=Compose a reply to the mailing list of the selected message - accel=/ + accel=*Control*l/ cmd name=MessageReplySender _tip=Compose a reply to the sender of the selected message -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
Jan Schmidt wrote: quote who=Mary Gardiner On Tue, Feb 24, 2004, Jamie Wilkinson wrote: This one time, at band camp, Robert Collins wrote: It's simpler than that I think... I hit reply-to-all as reflex - there's no reply to list icon in evolution. ;) though there is a reply to list menu action... i'd file a bug. Fixed in 1.5, according to Jan. It's the dammit-they-fixed-it-in-unstable bind. Here's my patch for /usr/share/evolution/1.4/ui/evolution-mail-message.xml to add reply-to-list as a keybinding to evo 1.4. ctrl-l is the new keybinding (already in 1.5) Last time I saw this, I forgot to ask if they also made it a button in the toolbar. So, Jan, is it also a button in the toolbar? -- Pete -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] who wrote the most code?
A few weeks ago, Benno asked me if I had a way to use cvs annotate to
work out percentages of code written, and after a short bit of hacking,
there was this:
a=; t=0; cvs ann 2/dev/null | cut -c15-21 | sort | uniq -c | while read i; do u=`echo
$i | cut -f2 -d' '`; s=`echo $i | cut -f1 -d' '`; t=$((${t-0} + $s)); echo $t $u $s ;
done | sort -rn | while read i ; do if [ x$a = x ]; then a=`echo $i | cut -f1 -d'
'`; fi ; u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f3 -d' '`; s=$(($s * 100));
echo $u $(($s / $a)); done
but I was looking at the subversion repo at work today, and wondered
the same thing... so with a bit of mangling here it is:
a=; t=0; find . -type f ! -regex '.*\.svn.*' -exec svn ann {} \; 2/dev/null | awk
'{print $2}' | sort | uniq -c | while read i; do u=`echo $i | cut -f2 -d' '`; s=`echo
$i | cut -f1 -d' '`; t=$((${t-0} + $s)); echo $t $u $s ; done | sort -rn | while read
i ; do if [ x$a = x ]; then a=`echo $i | cut -f1 -d' '`; fi ; u=`echo $i | cut -f2
-d' '`; s=`echo $i | cut -f3 -d' '`; s=$(($s * 100)); echo $u $(($s / $a)); done
Fun, eh? :-)
--
[EMAIL PROTECTED] http://spacepants.org/jaq.gpg
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
On Tue, Feb 24, 2004 at 02:06:46PM +1100, Jamie Wilkinson wrote: A few weeks ago, Benno asked me if I had a way to use cvs annotate to work out percentages of code written, and after a short bit of hacking, there was this: http://statcvs.sf.net/ ;) -Andrew. -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
RE: [SLUG] who wrote the most code?
Where's the structure, indentation, and documentation, aka comments in
the code ? Or do we open a a new OSSP for this project to get it cleaned
up. But then how would we measure who wrote the most code of the code
measuring code? :-) Ho hum.
-Original Message-
From: Jamie Wilkinson [mailto:[EMAIL PROTECTED]
Sent: Tuesday, 24 February 2004 2:07 PM
To: [EMAIL PROTECTED]
Subject: [SLUG] who wrote the most code?
A few weeks ago, Benno asked me if I had a way to use cvs annotate to
work out percentages of code written, and after a short bit of hacking,
there was this:
a=; t=0; cvs ann 2/dev/null | cut -c15-21 | sort | uniq -c | while read
i; do u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f1 -d' '`;
t=$((${t-0} + $s)); echo $t $u $s ; done | sort -rn | while read i ; do
if [ x$a = x ]; then a=`echo $i | cut -f1 -d' '`; fi ; u=`echo $i |
cut -f2 -d' '`; s=`echo $i | cut -f3 -d' '`; s=$(($s * 100)); echo $u
$(($s / $a)); done
but I was looking at the subversion repo at work today, and wondered
the same thing... so with a bit of mangling here it is:
a=; t=0; find . -type f ! -regex '.*\.svn.*' -exec svn ann {} \;
2/dev/null | awk '{print $2}' | sort | uniq -c | while read i; do
u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f1 -d' '`; t=$((${t-0} +
$s)); echo $t $u $s ; done | sort -rn | while read i ; do if [ x$a =
x ]; then a=`echo $i | cut -f1 -d' '`; fi ; u=`echo $i | cut -f2 -d'
'`; s=`echo $i | cut -f3 -d' '`; s=$(($s * 100)); echo $u $(($s / $a));
done
Fun, eh? :-)
--
[EMAIL PROTECTED]
http://spacepants.org/jaq.gpg
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
RE: [SLUG] who wrote the most code?
LOC - Surely this metric really doesn't count for much - boom tish.
A better metric might be who went through the LOC and reduced them to
fewer, more robust LOC. A far harder metric - but obvious if you are on
the dev team.
Remember that 94.251% of all statistics are lies.
Stu
On Tue, 2004-02-24 at 16:38, Ron Daniel wrote:
Where's the structure, indentation, and documentation, aka comments in
the code ? Or do we open a a new OSSP for this project to get it cleaned
up. But then how would we measure who wrote the most code of the code
measuring code? :-) Ho hum.
-Original Message-
From: Jamie Wilkinson [mailto:[EMAIL PROTECTED]
Sent: Tuesday, 24 February 2004 2:07 PM
To: [EMAIL PROTECTED]
Subject: [SLUG] who wrote the most code?
A few weeks ago, Benno asked me if I had a way to use cvs annotate to
work out percentages of code written, and after a short bit of hacking,
there was this:
a=; t=0; cvs ann 2/dev/null | cut -c15-21 | sort | uniq -c | while read
i; do u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f1 -d' '`;
t=$((${t-0} + $s)); echo $t $u $s ; done | sort -rn | while read i ; do
if [ x$a = x ]; then a=`echo $i | cut -f1 -d' '`; fi ; u=`echo $i |
cut -f2 -d' '`; s=`echo $i | cut -f3 -d' '`; s=$(($s * 100)); echo $u
$(($s / $a)); done
but I was looking at the subversion repo at work today, and wondered
the same thing... so with a bit of mangling here it is:
a=; t=0; find . -type f ! -regex '.*\.svn.*' -exec svn ann {} \;
2/dev/null | awk '{print $2}' | sort | uniq -c | while read i; do
u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f1 -d' '`; t=$((${t-0} +
$s)); echo $t $u $s ; done | sort -rn | while read i ; do if [ x$a =
x ]; then a=`echo $i | cut -f1 -d' '`; fi ; u=`echo $i | cut -f2 -d'
'`; s=`echo $i | cut -f3 -d' '`; s=$(($s * 100)); echo $u $(($s / $a));
done
Fun, eh? :-)
--
[EMAIL PROTECTED]
http://spacepants.org/jaq.gpg
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
This one time, at band camp, Andrew Bennetts wrote: On Tue, Feb 24, 2004 at 02:06:46PM +1100, Jamie Wilkinson wrote: A few weeks ago, Benno asked me if I had a way to use cvs annotate to work out percentages of code written, and after a short bit of hacking, there was this: http://statcvs.sf.net/ unless this project contains a php jukebox and photo gallery, it's just wasting space; i have already proved that it is trivially implemented on the command line. -- [EMAIL PROTECTED] http://spacepants.org/jaq.gpg -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
Jamie Wilkinson wrote: This one time, at band camp, Andrew Bennetts wrote: On Tue, Feb 24, 2004 at 02:06:46PM +1100, Jamie Wilkinson wrote: A few weeks ago, Benno asked me if I had a way to use cvs annotate to work out percentages of code written, and after a short bit of hacking, there was this: http://statcvs.sf.net/ unless this project contains a php jukebox and photo gallery, it's just wasting space; i have already proved that it is trivially implemented on the command line. trivially implemented? Is that a troll? That was the longest single line of shell code I've ever seen ;) In any case that program appears to gather different statistics from your shell script. Yours counts lines of code per developer as last modified by them on the head. statcvs counts lines of checkins (to oversimplify it). statcvs also makes pretty graphs :) Not that I'm defending statcvs though. It's written in Java. Mick. -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] who wrote the most code?
On Tue, 2004-02-24 at 14:06, Jamie Wilkinson wrote:
A few weeks ago, Benno asked me if I had a way to use cvs annotate to
work out percentages of code written, and after a short bit of hacking,
there was this:
a=; t=0; cvs ann 2/dev/null | cut -c15-21 | sort | uniq -c | while read i; do
u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f1 -d' '`; t=$((${t-0} + $s)); echo
$t $u $s ; done | sort -rn | while read i ; do if [ x$a = x ]; then a=`echo $i |
cut -f1 -d' '`; fi ; u=`echo $i | cut -f2 -d' '`; s=`echo $i | cut -f3 -d' '`;
s=$(($s * 100)); echo $u $(($s / $a)); done
but I was looking at the subversion repo at work today, and wondered
the same thing... so with a bit of mangling here it is:
a=; t=0; find . -type f ! -regex '.*\.svn.*' -exec svn ann {} \; 2/dev/null | awk
'{print $2}' | sort | uniq -c | while read i; do u=`echo $i | cut -f2 -d' '`;
s=`echo $i | cut -f1 -d' '`; t=$((${t-0} + $s)); echo $t $u $s ; done | sort -rn |
while read i ; do if [ x$a = x ]; then a=`echo $i | cut -f1 -d' '`; fi ; u=`echo
$i | cut -f2 -d' '`; s=`echo $i | cut -f3 -d' '`; s=$(($s * 100)); echo $u $(($s /
$a)); done
Fun, eh? :-)
Pity it's completely wrong.
How? Thats an exercise for the reader. Hint: your script does what you
intended it to do, but the result is /not/ LOC per person.
Rob
--
GPG key available at: http://www.robertcollins.net/keys.txt.
signature.asc
Description: This is a digitally signed message part
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
