Skip to content
Commits on Dec 26, 2015
Commits on Sep 12, 2011
Commits on Sep 10, 2011
  1. (api new) Reworded everything in terms of “metadata” and “blobs.”

    committed
    This commit rewrites large swathes of the codebase to semantically apply the new terminologies of
    the “blobs with metadata” story.
    
    In the process, some other relatively minor cleaning was preformed:
    - Replaced some calls that were superfluously indirected through structs with their direct
      equivalents. Yes, indirection through structs looks nice, but there’s a limit to how much we want
      to throw away free performance.
    - Changed some minor stylistic details throughout the codebase, while in the process of revamping
      everything.
      - Comma-first style (which I’ve been using for a long time elsewhere, so it escapes me why this
        codebase wasn’t already conformant)
      - Caught a few straggling instances of broken threesp-rule
    - Making more widespread use of the technique of directly returning a designated-initialized struct
    - Removed several superfluous getters. Putting more onus on the consumer to error-check when
      directly accessing struct members. Will involve more documentation, but also allows for less
      code and fewer tests. Net win.
    
    Squashes: 376ee11..b59de90
Commits on Sep 9, 2011
  1. (- meta fix) Referring people to #elliottcable instead of #Paws.Nucleus

    committed
    This isn’t the *official* channel, but it’s the preferred one for the moment. Until Freenode gives us ownership
    of #Paws, and there’s enough of a language to worry about having its own separate community from my own, it’s
    easier to concentrate everything in a single channel.
Commits on Aug 28, 2011
  1. (fix API) Renaming `AST` to `Script`.

    committed
    Only had to modify the implementation very slightly to make sense with the new `Script` semantics,
    studded-`TERMINAL`s and all.
Commits on Jul 15, 2011
  1. (- doc fix) Updating the IRC channel.

    committed
    (##Paws is now a private, invite-only, social channel; #Paws.Nucleus is the appropriate on-topic channel for Paws discussion.)
  2. (doc temp) Temporarily removing all documentation.

    committed
    Quite a lot of the documentation (such as it was) was getting very out-of-sync with the Nuclear design itself, to
    say nothing of the actual codebase. It all needs to be rewritten, but I can’t stop to do that right now; so I’m
    removing it outright pending a rewrite.
    
    A lot of this will go back in verbatim, or with minor changes; it just needs ratification as “still applicable.”
Commits on Jun 9, 2011
  1. Merge branch '`label`' into `execution`

    committed
    Conflicts:
    	README.markdown
    	Source/Paws.c
Commits on Jun 8, 2011
  1. (incomplete new) Lots of preliminary work on `AST.c` and the `executi…

    committed
    …on` nuketype.
    
    Most of this work was recovered from an old, corrupted git stash.
Commits on Jun 7, 2011
  1. @judofyr
  2. @judofyr
  3. (new) Implementing `Magazine` and `Unique`.

    committed
    Primarily of interest:
    - `Magazine->of()`, depending on
    - `_magazine__hash()` and
    - `Unique->of()`
Commits on May 23, 2011
  1. (- new api) Getting started with basic boilerplate and infrastructure…

    committed
    … for `label` and `Magazine.c`.
Commits on May 22, 2011
Commits on May 21, 2011
  1. Revert "(re style -) Properly re-indenting preprocessor directives in…

    committed
    … `execution/`." and "(new dup incomplete) Copying over the old `routine.c` and `ast.c`."
    
    This reverts commit 8bf9f5f.
    This reverts commit 569e20b.
Commits on May 19, 2011
  1. (style re) A few major style refactorings:

    committed
    - Far-left-edge structure-indentation *always* supercedes alignment-indentation (additively, of course)
    - Taking `…(void);` into account in min-3 calculations.
    - No longer taking `extern`/`static` keywords into account for min-3 calculations.
    - A few other, minor, consistency fixes.
Commits on May 17, 2011
  1. (- re style) Moving the `initialize` functions into the “family funct…

    committed
    …ions” section.
    
    Semantically speaking, `initialize()` operates on `struct a_type *`, not `a_type` itself. Until it’s been initialized, it’s not really an object of that type, now is it? Since it doesn’t operate on fully-operational instances of `a_type`, I’m considering them a family function instead of an instance function.
Commits on May 16, 2011
Commits on Apr 21, 2011
  1. (- tests) Updating our reference to the lastest version of `CEST()`, …

    committed
    …making `SOMEDAY;` available.
Commits on Apr 20, 2011
Commits on Apr 19, 2011
Commits on Mar 7, 2011
Something went wrong with that request. Please try again.