Graham Knop

Recent Actions

• Commented on Names and Numbers, Brand and Identity
  What currently exists as Perl 5 needs an identity. A real identity that people can both understand and promote. "A member of the Perl family" is not an identity. Changing the name is not just about changing our comparison to...
• Commented on Testing scripts in your distribution, portably
  Keep in mind that $FindBin::Bin can also contain spaces, so you should be quoting $MPATH as well. As I had mentioned in IRC, the full story with Windows argument quoting is a lot nastier, but at least for filenames, wrapping...
• Posted Using system or exec safely on Windows to Graham Knop

  Passing a list of arguments to another program on Windows in perl is much more complicated than it should be. There are several different issues that combine that lead to this.

  (mostly copied from a post I made on PerlMonks)

  First is that argument lists are always passed as a sing…

• Posted Converting Complex SVN Repositories to Git - Part 4 to Graham Knop
  • Part 2: Initial Import into Git
• Posted Converting Complex SVN Repositories to Git - Part 3 to Graham Knop

  Resolving Branches and Calculating Merges

  The most important part of the repository conversion I did was resolving all of the branches and calculating the merge points. The majority of the rest of the process is easily automated with other tools.

  The main part of this section wa…

• Posted Converting Complex SVN Repositories to Git - Part 2 to Graham Knop

  Initial Import into Git

  Creating a mirror

  SVN is slow, and git-svn is slower. The amount of network traffic needed by SVN makes everything slow, especially since git-svn needs to walk the history multiple times. Even if I made no mistakes and only had to run the import once, …

• Posted Converting Complex SVN Repositories to Git - Part 1 to Graham Knop

  In May and June, I worked on converting the DBIx::Class repository from SVN to Git. I’ve had a number of people ask me to describe the process and show…

• Commented on A history of Perl variables
  And CHILD_ERROR_NATURE should be CHILD_ERROR_NATIVE. I suppose I could have finished reading the post before replying....
• Commented on A history of Perl variables
  s/WARNINGS_BITS/WARNING_BITS/...
• Commented on Spot the error
  Or actually: rmap_all { $_ = $_ ? true : false if JSON::is_bool($_) } $arg; Probably should have checked the docs on Data::Rmap before replying....
• Commented on Spot the error
  You are also relying on the internals of both boolean and JSON's boolean values, and breaking them in the process. JSON (PP or XS) uses singleton objects for the booleans, which means you are trying to rebless the constants $JSON::true...
• Commented on So is wantarray() bad or not?
  I don't see a general problem with using wantarray to return different things as in your initial example, but using it for anything other than altering the return value seems very bad to me. For example, returning either an array...
• Commented on More bash completion help for perldoc
  This is exactly feature I was going to ask for after seeing your previous post. Thanks....

Comment Threads

• Steven Haryanto commented on Spot the error

  @Graham: I rebless instead of assigning to $_ (which should be clearer) because the code was converted from the one that uses Data::Walk. Now that I use Data::Rmap, I can use assignment. Thanks, my code is now improved.
  

• dagolden.com commented on Testing scripts in your distribution, portably

  I'm biased, but I would recommend Capture::Tiny instead of qx(). That also means you can use system() and stop worrying about spaces in paths. (Mostly stop worrying).

  I would use "bin/yourprogram" or "script/yourprogram" (depending on where you put it) and leave out FindBin.

  See https://gist.github.com/3795436 for what that looks like. I think that winds up much more portable.

  Probe::Perl instead of $^X if you're *really* paranoid, but I've largely stopped bothering with it.

• Robert commented on Names and Numbers, Brand and Identity

  Someone looking for a new language is going to try "Pumpkin Perl"? I doubt it.

• Matt S Trout (mst) commented on Names and Numbers, Brand and Identity

  Mike, I was the one who originally popularised 'perl5 is a velociraptor' - my YAPC::* closing keynotes summarising the state of the perl5 community are known as 'State of the Velociraptor'.

  However, I think the raptor meme belongs to the perl5+CPAN platform rather than to perl5 the language itself, so I chose not to try and force it onto the language.