The Perl Toolchain Summit needs more sponsors. If your company depends on Perl, please support this very important event.
 / 
App-YAML-Filter-0.012
1 ++

Changes for version 0.012

  • Change: 90495a14bc41cb68e324598f14456d940b95cb59 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-10-19 17:25:41 +0000
    • release v0.012
  • Change: 4f38e20b004d974de63d88c746d37efeceb990d4 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-10-19 00:00:30 +0000
    • add yfrom script to output YAML from JSON or CSV
    • yfrom and yto are the sentinels to get data into the internal format used by yq and the other, future tools.
  • Change: 3976d7917a0d619f419796c9436404d5ac7d3648 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-10-18 22:47:19 +0000
    • remove references to Test::Most
  • Change: 198acbb5f380baa5ece2d0e5b62f5e9bff9bba28 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-10-18 22:37:17 +0000
    • add Marpa::R2 prereq
    • We need to only use one of these prereqs, or split up this dist into multiple things...
  • Change: 7fc32e215a8f21ee0e01c0f1e67accf1090476ec Author: Doug Bell <madcityzen@gmail.com> Date : 2014-10-18 22:31:17 +0000
    • add travis config
  • Change: f58890bc17ac6e9c4d8515cb05e5e76175fd86a4 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-10-18 22:29:03 +0000
    • add yto script to convert from YAML to other formats
    • Currently only JSON is supported, and only if the user has JSON::XS (from CPAN) or JSON::PP (core since 5.14)
  • Change: bd9747ee0204fc41d6ddfb61792b3ba8984963b7 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-10-18 22:28:36 +0000
    • always use 5.10 features
    • We require 5.10 for recursive regex, so always enable "say" and "state"
  • Change: 6f3001a9d64f5aff8dc5eafbb4ecfc020bbbbb7f Author: Doug Bell <madcityzen@gmail.com> Date : 2014-10-18 22:27:58 +0000
    • move away from Test::Most
    • Our own Import::Base is also being used, so we only need one
  • Change: 57be0bd53f8f8722652d0ce961e22e5284f54607 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-23 23:39:41 +0000
    • add an initial Marpa::R2 parser
    • This simply does the parse tree and nothing else. The actual interpreting is just a Simple Matter of Programming.
  • Change: 0236f1459c694758d352d7dfa3fa97dccab8352f Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-22 23:58:55 +0000
    • start on a parse-tree-based P::RD implementation
    • This implementation lets us prepare the filter first, then run through the resulting parse tree (much like an AST).
  • Change: 54ca82c841e505bc90995d64c698a7ade9eedd62 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-22 23:57:45 +0000
    • int rule is unnecessary since float covers ints
    • Though this leaves us open to array indexing with floats, which probably won't work... Maybe we need intlike and numberlike...
  • Change: bf97a7fa40e3dc1c3cb4427e560c17e9d7baf92a Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-18 23:19:23 +0000
    • add some preprocessing to fix pipes
    • This gets around the order of Parse::RecDescent's interpretation, but it's not a good plan. The tree parse is a much better plan.
  • Change: 0cdca4825e889993598214e2035793e95d74852e Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-13 01:44:38 +0000
    • fix comma combinator
    • Fun story: This interpreted method will not work for pipe combinators, since by the time the pipe is being interpreted, both sides of the pipe have already been evaluated.
    • P::RD does make it very easy to build a parse tree though, so perhaps we'll start from that and build an interpreter of that parse tree.
  • Change: fdd6bb511c5941f986acdd510171653b10f44651 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-13 01:37:54 +0000
    • undef is a valid value, so return an empty list
  • Change: db8f2e3ffdada6c45e5d5595dcd170dc03e1968a Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-13 00:49:29 +0000
    • put use lines at the top of the parse
    • The initial actions is for global stuff, much like a module's initial lines
  • Change: 3cef00015fe3994994b5cf92a02383528e296469 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-13 00:20:08 +0000
    • fix the order of <|> and <=|>=
    • We must put the <= first, so that > doesn't get matched and leave a dangling "=" that doesn't match anything
  • Change: 07f5af20483ce34d09e678f3a2ab7cdc6a90e9ab Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-13 00:19:24 +0000
    • use eval to get rid of the literal quote characters
    • Internally, quoted strings don't have the opening and closing quotes, they're just for the parser.
  • Change: b858df8e7f1d13cd4649be8304c83ef6271cd0f5 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-13 00:19:01 +0000
    • add the missing sort function
  • Change: 36e22f1d46e9dc83ccee349b9c657be34bedc54b Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-13 00:18:06 +0000
    • fix the missing empty function
  • Change: d6e1cb1221a2be98cce3830f3a8d4ce7a50e3727 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-13 00:16:51 +0000
    • fix number parsing (int, float, bin, hex, oct)
    • We still have to use eval to get Perl to turn them into its internal format.
  • Change: 7f425ef5ae7c98402034a4557f2a3acd2d446473 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-13 00:16:38 +0000
    • add list object to differentiate arrays
    • Most things can return lists, but some things want to return an array. Since Parse::RecDescent can only return scalars, we need some way of differentiating between a filter or expression returning multiple things, and a filter or expression returning an array.
    • A list is just a blessed array and can be flattened. When we only want one thing, a list will take one item, while an array is one thing.
    • This is a lot like Perl's way of doing it, except we have to use arrayrefs for both circumstances.
  • Change: d0cd11a7cc4e08ab35d77a1e89b664091119385e Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-05 22:50:57 +0000
    • SurgicalPodWeaver -> PodWeaver
    • SurgicalPodWeaver is broken, and all our files should be using PodWeaver anyway.
  • Change: a10a27d56cc2d5605546c61c61ae6e931d7ce40a Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-05 22:50:57 +0000
    • initial test Parse::RecDescent implementation
    • I'm going to throw this away, because I absolutely do not understand it anymore. Going to try a different tactic...
  • Change: 5188d986f0e379b6c67efdd8c67e5de79cb8edbb Author: Doug Bell <madcityzen@gmail.com> Date : 2014-07-05 22:50:57 +0000
    • load the underlying interpreter at runtime
  • Change: a26cf9e93cf0e66c957b7668abfecc667ab826c9 Author: Doug Bell <doug.bell@baml.com> Date : 2014-05-23 15:50:34 +0000
    • add sort(EXPR) function to sort by string
    • This does not support numeric sorting. Like group_by(EXPR), this buffers the output. Like all functions that use scope, this probably will not work in assignment.
    • I'm not sure how I want to do numeric sorting. Probably as a first argument of some kind. Maybe if sort took two EXPRs and we added the <=> and cmp binary operators, and the magic $a and $b variables...
    • Fixes #12
  • Change: 36e761fdb4c87f0e4fc2a3288f2c3adbf7f152b8 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-05-22 02:10:42 +0000
    • move the interpreter to a module
    • We're going to try Parse::RecDescent and maybe Marpa::R2 as well...
  • Change: 65b6d78d7fbdcbaed14401c5005ae708b83fa135 Author: Doug Bell <doug.bell@baml.com> Date : 2014-05-21 18:18:08 +0000
    • add <,<=,>,>= numeric comparison operators
    • Also clarify the behavior of undef in comparisons. These allow warnings if either side is undef. We should not suppress these warnings, though we may have a way to ensure more robust coding in the future...
    • Fixes #30
  • Change: 42c5051ea50d0547fd3606de3c48afb66d8bd451 Author: Doug Bell <doug.bell@baml.com> Date : 2014-05-21 17:23:08 +0000
    • test and document the == and != operators
  • Change: 5572be058f3306cef40a615cde767e948c36a3d5 Author: Doug Bell <doug.bell@baml.com> Date : 2014-05-21 17:11:44 +0000
    • add length( EXPR ) function
    • Tries to DWIM as much as possible, though I think numbers are going to be a bit weird (will get the number of base-10 digits no matter what, I suspect, but that's not a use-case I need right now).
    • I won't be doing a scalar() alias, but I might consider a count() alias.
    • Fixes #28
  • Change: c4d9194ab9275455a151d3ddd531c7921a79aa0d Author: Doug Bell <doug.bell@baml.com> Date : 2014-05-21 16:35:24 +0000
    • add keys(EXPR) function
    • The keys function gets the keys of the HASH or ARRAY pointed to by EXPR.
    • This function returns an arrayref, but I'm not sure if that's the right idea. We can flatten an arrayref using `keys() | .[]` I guess, so maybe that's the right solution...
    • Fixes #27
  • Change: d98b61c25f86620579595b29e203f28d356f6290 Author: Doug Bell <madcityzen@gmail.com> Date : 2014-05-17 03:03:28 +0000
    • clarify and add some comments during the parsing

Documentation

yfrom
Build YAML from another format (like JSON or CSV)
yq
Filter YAML through a command-line program
yto
Change YAML to another format (like JSON)

Modules

App::YAML::Filter
Filter YAML with a series of rules
App::YAML::Filter::Base
Base module for App::YAML::Filter
App::YAML::Filter::Marpa
App::YAML::Filter::RecDescent
A Parse::RecDescent-based parser for programs
App::YAML::Filter::RecDescentTree
A Parse::RecDescent-based parser using a parse tree
App::YAML::Filter::Regex
A regex-based parser for programs
App::YAML::Filter::Test
Base module for App::YAML::Filter tests

Other files