Comments on “Install memcached on OSX via MacPorts” 2010-06-20T21:41:21-04:00 Chyrp Git Tip: Ignore Changes to Tracked Files,2010-06-20:/id/181//comment_21573 2010-06-20T21:41:21-04:00 2010-06-20T21:41:21-04:00 Rob Wilkerson <p>Nope. In fact, I find myself finding all kinds of reasons to use it. Great functional technique, <span class="caps">IMO</span>.</p> Git Tip: Ignore Changes to Tracked Files,2010-06-18:/id/181//comment_21551 2010-06-18T16:42:05-04:00 2010-06-18T16:42:05-04:00 Rick Guyer <p>This would be useful. I’m going to try it. Have you noticed any negative side effects since you wrote this post?</p> Add to Google Reader Extension for Safari 5,2010-06-17:/id/185//comment_21540 2010-06-17T19:48:45-04:00 2010-06-17T19:48:45-04:00 Rob Wilkerson <p>Yep, this was brought to my attention a couple of hours ago by someone else. The compiled extension included the git repository, a <span class="caps">README</span> that didn’t impact functionality as well as a copy of the compiled extension itself. What can I say, I’m learning. :-)</p> <p>A new copy is available that should be much smaller. 16K, to be precise. You can download it using the link above (the target file hasn’t changed, just its size).</p> <p>Thanks for pointing that out.</p> Add to Google Reader Extension for Safari 5,2010-06-17:/id/185//comment_21538 2010-06-17T18:19:51-04:00 2010-06-17T18:19:51-04:00 ajbpearce <p>hey rob, I love the functionality that this extension purports to grant, but i am somewhat reluctant to install it as the .safariextz file i have downloaded was 21mb as opposed to Chupa’s original which was just 31kb. This leads me to suspect something must have snuck into your complied file by accident?</p> CakePHP: Adding Dynamic Content to Layouts,2010-06-04:/id/159//comment_21290 2010-06-04T05:35:39-04:00 2010-06-04T05:35:39-04:00 bla <p>Hey,</p> <p>can you make the <span class="caps">SQL</span>-Code public?</p> Chyrp Theme Updates,2010-06-03:/id/15//comment_21282 2010-06-03T15:27:13-04:00 2010-06-03T15:27:13-04:00 Mike891 <p>Working on more changes above the comments, but thanks. The guy that wrote the software made it pretty easy. :-)</p> Ignore Git's Suspicious Patch Lines,2010-06-01:/id/106//comment_21234 2010-06-01T00:22:52-04:00 2010-06-01T00:22:52-04:00 Victor <p>I had the same problem. I need the pre-commit hook but I just dont need the trailing space verification so I edit the pre-commit file commenting with “#” character the three lines that do the trailing space verification</p> <p>if (s/^\+//) { $lineno++; chomp; # if (/\s$/) { # bad_line(“trailing whitespace”, $_); # } if (/^\s* \t/) { bad_line(“indent SP followed by a <span class="caps">TAB</span>”, $_); } if (/^([<>])\1{6} |^={7}$/) { bad_line(“unresolved merge conflict”, $_); } }</p> Ignore Git's Suspicious Patch Lines,2010-05-26:/id/106//comment_21034 2010-05-26T18:45:33-04:00 2010-05-26T18:45:33-04:00 JOsh Beauregard http://http// <p>I just had this problem it was due to the pre-hooks file being executable. so what you have to do is change the permissions on <REPODIR>/.git/hooks/pre-commit to be non executable</p> Install "Non-Stable" PEAR Packages,2010-05-06:/id/125//comment_19895 2010-05-06T21:03:00-04:00 2010-05-06T21:03:00-04:00 Mike <p>Thanks for the tip – very useful.</p> Remove Conflicted Files from Dropbox,2010-05-06:/id/152//comment_19887 2010-05-06T11:00:31-04:00 2010-05-06T11:00:31-04:00 Rob Wilkerson <p>@alvin – Me too. I ended up just scripting these commands for easy (and regular) execution. :-)</p> Remove Conflicted Files from Dropbox,2010-05-05:/id/152//comment_19876 2010-05-05T17:34:18-04:00 2010-05-05T17:34:18-04:00 Alvin <p>very slick, will try that in Ubuntu; hope I get the syntax right; perhaps I will try it first on a dummy directory to make sure I got it right. I have 3-4 computers on all the time, so always getting this problem. </p> Linux: Make Your Scroll Wheel Double Click,2010-05-05:/id/94//comment_19872 2010-05-05T12:56:39-04:00 2010-05-05T12:56:39-04:00 Steve <p>I used this tip on my wife’s computer; it worked great on 8.04 but not on subsequent versions (although Rob said it worked for him on 8.10). The problem might be that starting with 8.10, Nautilus seems to have its own (undocumented?) scroll wheel click behavior: a single click of the scroll wheel behaves like a double-click when Nautilus is in Icon view (although scroll wheel clicking a folder opens it as a new tab rather than replacing the contents of the current window). A scroll wheel click doesn’t appear to do anything in List view.</p> <p>If anyone has any insight into this I’d love to disable the default behavior and use Rob’s tip (or something similar) instead.</p> <p>Thanks!</p> Ignore Git's Suspicious Patch Lines,2010-04-23:/id/106//comment_19543 2010-04-23T15:56:31-04:00 2010-04-23T15:56:31-04:00 mars <p>As a <span class="caps">PHP</span> programmer, I agree with your frustration, because <span class="caps">PHP</span> ignores whitespace. I would guess the pre-commit hook is in there for programmers who use other languages where whitespace matters such as Python. The <span class="caps">PEP</span> 8 that defines the Python coding style guide ( suggests an 80 character width for all lines.</p> <p>Whether or not a restriction of 80 char/line makes any sense in this day and age, well, no comment. I’m not a Python programmer (yet) but I know enthusiasts, and they don’t seem to mind it too much.</p> <p>I’m not sure if the Python interpreter enforces the 80 char limit, but if it does, you can see where trailing whitespace might become a problem.</p> <p>This theory can be backed up by the comments found in .git/hooks/pre-commit that also check for a mixture of tabs and spaces at the beginning of lines, which I know the Python interpreter will whine about if invoked with certain options.</p> <p>Furthermore, I’m sure there are plenty of other examples where <span class="caps">GIT</span> is being used to version documents that require a specific number of characters per line, whitespace or otherwise (although none come immediately to mind :-)</p> <p>This is the <span class="caps">ONLY</span> reason I can currently think of having that particular check in the pre-commit hook. However I can think of NO reason why it is there by default.</p> <p>As for your solution, you might want to be careful by simply making pre-commit non-executable.</p> <p>For one, if you look in the actual script, there is another error there that spits the message “unresolved merge conflict”, which is decidedly more important, and I wouldn’t recommend trying to bypass it.</p> <p>Secondly, you can add your own pre-commit hooks which can be very handy sometimes.</p> <p>Personally, my workaround was just to comment out the conditional that throws the “trailing whitespace” error (in my version, it’s lines 58-60 — otherwise identifiable by the regex:</p> <p>/\s$/</p> <p>also, use # to comment in Perl which has no block commenting to my knowledge)</p> Change Key Bindings in Pidgin,2010-04-22:/id/114//comment_19511 2010-04-22T02:04:54-04:00 2010-04-22T02:04:54-04:00 Jon Chamberlain <p>o. m. g. this has caused me so much hair-pulling and frustration over the past couple of weeks. Thanks so much for documenting this – this is definitely more of a bug than a feature. My few remaining hairs thank you also! ;-)</p> Synchronizing Firefox Through Dropbox,2010-04-01:/id/118//comment_19167 2010-04-01T00:36:11-04:00 2010-04-01T00:36:11-04:00 QA <p>Did you know that Dropbox has a repository of feature requests where you can vote for user-submitted suggestions? Why not go there and vote for a Firefox profile sync feature? Ideally we shouldn’t all have to spend time rigging together a home-made sync solution for FF.</p> <p></p> <p>In the meantime we have this clever workaround that Rob W. has been kind enough to post, so thanks Rob!</p> Synchronizing Firefox Through Dropbox,2010-03-26:/id/118//comment_18873 2010-03-26T02:12:32-04:00 2010-03-26T02:12:32-04:00 silversrt <p>Do you have copy writer for so good articles? If so please give me contacts, because this really rocks! :)</p> Labels. They're Not Just for Forms Anymore.,2010-03-21:/id/150//comment_18630 2010-03-21T00:05:27-04:00 2010-03-21T00:05:27-04:00 Nick <p>I’d use a definition list for that.</p> Remove Console Output Limits in Eclipse,2010-03-17:/id/99//comment_18444 2010-03-17T12:17:33-04:00 2010-03-17T12:17:33-04:00 Alex <p>How do i remove console box… if i’m using eclipse?</p> <p>come on please …</p> Learning CakePHP: Foreign Key Constraint Violation,2010-02-24:/id/167//comment_17245 2010-02-24T09:04:36-04:00 2010-02-24T09:04:36-04:00 Rob Wilkerson <p>Readability. <code>locations</code> is really just a lookup table. Whenever possible, I like to be able to open a table that uses lookup data and have it be meaningful without requiring the extra step of writing <span class="caps">SQL</span> to join the lookup. In this case, opening the <code>events</code> table would show a location like <code>MAINTENT</code>. Just by looking at the event record, I can tell where it’s happening.</p> <p>You’re right that Cake may be smart enough if it was an <code>int</code>, but I use <span class="caps">UUID</span>s. That’s just a <code>char</code> datatype so it may not behave any differently. It’s a good point to test, though. I may try it.</p> Learning CakePHP: Foreign Key Constraint Violation,2010-02-24:/id/167//comment_17235 2010-02-24T06:11:56-04:00 2010-02-24T06:11:56-04:00 Dieter_be <p>Why do you use the varchar datatype?<br /> I don’t allow users to enter “no relation” so I’m not sure, but if you use a proper datatype for id’s and foreign keys (like unsigned int), Cake might be smart and use null instead of an empty string for unset FK’s.<br /> If it doesn’t do that, I suggest to file a bug report.</p>