rsyntaxtree 2.0.0 → 2.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9ccd433053f296c9189c9eceb87f61c91ed815ff6b1be77f1cd2115a73eaf1cd
-  data.tar.gz: f769d68efb73c1aafc9dc9af9f0e4e7cbd7a0c741caf253905ebe11dc3c454be
+  metadata.gz: 231de78b2e34e37f4a5bbccd641839e8cf5ae1babc2d8d4362d86dcf782d28b4
+  data.tar.gz: a4d04e71e6587592540400aa9a7e23d9dadae0d19912078d5d6cc5ebd65c117d
 SHA512:
-  metadata.gz: 0e6218bc0ddf1f4c49e8fac2c8972cbd3e0eb251fe7087150ea73f63b054fe546483588e582a0f01e52ab26c20fd883ab2c67da07adf1f3120755447a7956945
-  data.tar.gz: b0810c0df9ede2cd1c36100ae9a3c055ece2d59261f8278ef8de8b79e07921702c7b8db83fa43f81810184c363b860bae3d9a9c1f45b4a02307f5f8220354fa8
+  metadata.gz: f97797690b6d8e0ad8404c00a10c061c89d2213e582915a33fc9d739305d5fb10cdd0afde6bffe0b16f504b5d5854d7dcb0249f50a22b10599638fb22180aed3
+  data.tar.gz: 7f5814cc9bf0875b21982cf6c143397d467fb900a7973c8f6213934387ba4f80e0605ddccec63f02bc7bacc6d182a9da1bda3237f909414ba4c64f63a1da1d12

data/CHANGELOG.md

@@ -1,5 +1,22 @@
 # Changelog
 
+## [2.1.0] - 2026-08-29
+
+### Changed
+- The JSON output's format name is `json`: `-f json`, writing `syntree.json`.
+  The old name `lsif` collided with an established code-intelligence format
+  of the same initials, and called itself an interchange format when only
+  one tool had ever written or read it. It remains an accepted alias — the
+  CLI says so on stderr — until 3.0. The document now identifies itself in a
+  top-level `format` object (`name: "rsyntaxtree-json"`, schema 0.4.0) in
+  place of the `lsif` key.
+
+### Added
+- A feature matrix in the JSON output carries its rows: cells split on the
+  tab stops, rules as rows of their own, and a matrix inside a cell
+  recursing. It used to arrive as an empty segment marked "matrix", its
+  attribute-value pairs recoverable only by re-parsing the raw label.
+
 ## [2.0.0] - 2026-08-28
 
 The removals announced in 1.10.0 and 1.12.0, and the fixes that came out of

data/README.md

@@ -69,7 +69,7 @@ Connect nodes with lines or arrows:
 
 ### Multiple Output Formats
 
-Generate trees in PNG, SVG, PDF, TikZ, or LSIF (JSON) format.
+Generate trees in PNG, SVG, PDF, TikZ, or JSON format — the last carrying the laid-out tree as structured data.
 
 ## Web Interface
 
@@ -170,7 +170,7 @@ Usage:
 | Option | Description | Default |
 |--------|-------------|---------|
 | `-o, --outdir` | Output directory | `./` |
-| `-f, --format` | Output format: png, svg, pdf, lsif, tikz | `png` |
+| `-f, --format` | Output format: png, svg, pdf, json, tikz | `png` |
 | `-l, --leafstyle` | Leaf style: auto, triangle, bar, nothing | `auto` |
 | `-n, --fontstyle` | Font style: sans, serif, cjk, mono | `sans` |
 | `-s, --fontsize` | Font size: 6-26 | `16` |

data/bin/rsyntaxtree

@@ -9,7 +9,7 @@ require_relative '../lib/rsyntaxtree'
 require_relative '../lib/rsyntaxtree/utils'
 require_relative '../lib/rsyntaxtree/format_converter'
 
-# "png, pdf, svg, lsif, or tikz" — built once from the library's
+# "png, pdf, svg, json, or tikz" — built once from the library's
 # own list so the three places that name the formats cannot drift apart.
 FORMAT_LIST_MESSAGE = FORMATS.join(", ").sub(/, ([^,]+)\z/, ', or \1')
 
@@ -17,7 +17,7 @@ FORMAT_LIST_MESSAGE = FORMATS.join(", ").sub(/, ([^,]+)\z/, ', or \1')
 CONFIG_VALIDATORS = {
   outdir: ->(v) { FileTest.directory?(v) ? nil : "must be an existing directory path" },
   outfilename: ->(v) { v.is_a?(String) ? nil : "must be a string" },
-  format: ->(v) { FORMATS.include?(v.to_s) ? nil : "must be #{FORMAT_LIST_MESSAGE}" },
+  format: ->(v) { FORMATS.include?(v.to_s) || v.to_s == "lsif" ? nil : "must be #{FORMAT_LIST_MESSAGE}" },
   leafstyle: ->(v) { OPTION_VALUES[:leafstyle].include?(v.to_s) ? nil : "must be auto, triangle, bar, or nothing" },
   fontstyle: ->(v) { OPTION_VALUES[:fontstyle].include?(v.to_s) ? nil : "must be sans, serif, cjk, or mono" },
   fontsize: ->(v) { v.is_a?(Integer) && v >= 6 && v <= 26 ? nil : "must be in the range of 6-26" },
@@ -133,6 +133,13 @@ opts = Optimist.options do
   opt :help, "This is a custom help message", short: :h
 end
 
+# The name the JSON format wore from 1.11.0 to 2.0.0. Accepted until 3.0,
+# and said out loud, the way JPG and GIF were seen out.
+if opts[:format] == "lsif"
+  warn "Warning: the format name 'lsif' is now 'json'. The old name will be removed in RSyntaxTree 3.0."
+  opts[:format] = "json"
+end
+
 Optimist.die :outdir, "must be an exsting directory path" unless FileTest.directory?(opts[:outdir])
 Optimist.die :direction, "must be one of: #{OPTION_VALUES[:direction].join(', ')}" unless OPTION_VALUES[:direction].include?(opts[:direction])
 Optimist.die :derivation, "must be one of: #{OPTION_VALUES[:derivation].join(', ')}" unless OPTION_VALUES[:derivation].include?(opts[:derivation])

data/lib/rsyntaxtree/{lsif_graph.rb → json_graph.rb}

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 #==========================
-# lsif_graph.rb
+# json_graph.rb
 #==========================
 #
 # Generates LSIF (Linguistic Structure Interchange Format) JSON output.
@@ -12,7 +12,7 @@ require_relative 'base_graph'
 require_relative 'utils'
 
 module RSyntaxTree
-  class LsifGraph < BaseGraph
+  class JSONGraph < BaseGraph
     attr_accessor :width, :height
 
     def initialize(element_list, params, global)
@@ -30,7 +30,7 @@ module RSyntaxTree
       @global = global
     end
 
-    def lsif_data
+    def json_data
       metrics = parse_list
       @width = metrics[:width] + @global[:h_gap_between_nodes] * 2
       @height = metrics[:height] + @global[:height_connector_to_text] / 2
@@ -241,14 +241,7 @@ module RSyntaxTree
       element.content.each do |l|
         next unless l[:type] == :text
 
-        segments = l[:elements].map do |e|
-          next if e[:decoration].include?(:whitespace) && e[:text].strip.empty?
-
-          {
-            text: e[:text].gsub(WHITESPACE_BLOCK, " ").gsub('&#62;', '>').gsub('&#60;', '<'),
-            decorations: e[:decoration].map(&:to_s).reject { |d| d == "whitespace" }
-          }
-        end.compact
+        segments = l[:elements].map { |e| build_segment(e) }.compact
         lines << { segments: segments } unless segments.empty?
       end
 
@@ -258,6 +251,43 @@ module RSyntaxTree
       }
     end
 
+    def build_segment(run)
+      return nil if run[:decoration].include?(:whitespace) && run[:text].strip.empty?
+
+      segment = {
+        text: run[:text].gsub(WHITESPACE_BLOCK, " ").gsub('&#62;', '>').gsub('&#60;', '<'),
+        decorations: run[:decoration].map(&:to_s).reject { |d| d == "whitespace" }
+      }
+      # A feature matrix used to reach this document as an empty segment
+      # marked "matrix", its attribute-value pairs recoverable only by
+      # re-parsing the raw label — which is no way to score whether a model
+      # wrote CASE nom where CASE nom was wanted. The rows come through now:
+      # cells split on the tab stops, rules kept as rows of their own, a
+      # matrix inside a cell recursing.
+      segment[:matrix] = { rows: build_matrix_rows(run[:matrix]) } if run[:decoration].include?(:matrix) && run[:matrix]
+      segment
+    end
+
+    def build_matrix_rows(matrix_lines)
+      matrix_lines.filter_map do |line|
+        case line[:type]
+        when :border then { rule: "single" }
+        when :bborder then { rule: "double" }
+        when :text
+          cells = [[]]
+          line[:elements].each do |run|
+            if run[:decoration].include?(:tabstop)
+              cells << []
+            else
+              segment = build_segment(run)
+              cells.last << segment if segment
+            end
+          end
+          { cells: cells }
+        end
+      end
+    end
+
     def map_enclosure(enclosure)
       case enclosure
       when :brackets then "brackets"
@@ -273,8 +303,14 @@ module RSyntaxTree
 
     def build_json
       data = {
-        lsif: {
-          version: "0.3.0",
+        # What this document is, said by a field rather than by the key, so a
+        # reader checks ["format"]["name"] and the key never has to change
+        # again. Documents up to schema 0.3.0 said it with a top-level "lsif"
+        # key instead — a name that collided with the code-intelligence
+        # format of the same initials.
+        format: {
+          name: "rsyntaxtree-json",
+          version: "0.4.0",
           generator: "rsyntaxtree #{RSyntaxTree::VERSION}",
           level: "rendered"
         },

data/lib/rsyntaxtree/notation_core.md

@@ -87,7 +87,7 @@ Options (the command line spells them --like-this):
 
 | option     | what it decides |
 |------------|-----------------|
-| format     | png, svg, pdf, tikz, lsif |
+| format     | png, svg, pdf, tikz, json (the tree as structured data) |
 | fontstyle  | sans, serif, mono, cjk |
 | fontsize   | 6–26 |
 | color      | modern, traditional, off, gray |

data/lib/rsyntaxtree/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module RSyntaxTree
-  VERSION = "2.0.0"
+  VERSION = "2.1.0"
 end

data/lib/rsyntaxtree.rb

@@ -34,12 +34,12 @@ WHITESPACE_BLOCK = "￭"
 # Every format the library can produce, and the file extension each one
 # writes. The CLI builds its --format validation, help text, and output
 # filename from these, so a new format is added here and nowhere else.
-FORMATS = %w[png pdf svg lsif tikz].freeze
+FORMATS = %w[png pdf svg json tikz].freeze
 FORMAT_EXTENSIONS = {
   "png" => "png",
   "pdf" => "pdf",
   "svg" => "svg",
-  "lsif" => "lsif.json",
+  "json" => "json",
   "tikz" => "tex" # forest code, to be included in a LaTeX document
 }.freeze
 
@@ -51,7 +51,10 @@ FORMAT_EXTENSIONS = {
 # purpose (see switched_on?); unknown keys are not rejected, because callers
 # pass their own extra parameters through.
 OPTION_VALUES = {
-  format: FORMATS,
+  # lsif is the name the JSON format wore from 1.11.0 to 2.0.0, accepted as
+  # an alias until 3.0. It collided with a code-intelligence format of the
+  # same initials, and claimed an interchange nobody else had implemented.
+  format: FORMATS + ["lsif"],
   leafstyle: %w[auto triangle bar nothing],
   fontstyle: ["sans", "serif", "cjk", "mono", "noto-sans", "noto-serif", "noto-sans-mono", "cjk zenhei"],
   color: %w[modern traditional gray grey off none on true false],
@@ -152,7 +155,7 @@ require_relative 'rsyntaxtree/utils'
 require_relative 'rsyntaxtree/element'
 require_relative 'rsyntaxtree/elementlist'
 require_relative 'rsyntaxtree/svg_graph'
-require_relative 'rsyntaxtree/lsif_graph'
+require_relative 'rsyntaxtree/json_graph'
 require_relative 'rsyntaxtree/tikz_generator'
 require_relative 'rsyntaxtree/version'
 require_relative 'rsyntaxtree/string_parser'
@@ -263,6 +266,8 @@ module RSyntaxTree
                             else
                               "off"
                             end
+        when :format
+          new_params[key] = value.to_s == "lsif" ? "json" : value.to_s
         when :hyphen
           new_params[key] = value.to_s == "literal" ? "literal" : "markup"
         when :fontsize
@@ -461,7 +466,7 @@ module RSyntaxTree
       case @params[:format]
       when "png" then raster_surface_for(draw_svg, &:finish)
       when "pdf" then pdf_surface_for(draw_svg, StringIO.new, &:finish)
-      when "lsif" then draw_lsif
+      when "json" then draw_json
       when "tikz" then draw_tikz
       else draw_svg
       end
@@ -535,11 +540,16 @@ module RSyntaxTree
       graph.svg_data
     end
 
-    def draw_lsif
+    def draw_json
       sp = StringParser.new(@params[:data].gsub('&', '&'), @params[:fontset], @params[:fontsize], @global)
       sp.parse
-      graph = LsifGraph.new(sp.get_elementlist, @params, @global)
-      graph.lsif_data
+      graph = JSONGraph.new(sp.get_elementlist, @params, @global)
+      graph.json_data
+    end
+
+    # The name the JSON output wore from 1.11.0 to 2.0.0; removed in 3.0.
+    def draw_lsif
+      draw_json
     end
 
     def draw_tikz(standalone: false, font: nil)

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: rsyntaxtree
 version: !ruby/object:Gem::Version
-  version: 2.0.0
+  version: 2.1.0
 platform: ruby
 authors:
 - Yoichiro Hasebe
@@ -126,7 +126,7 @@ files:
 - lib/rsyntaxtree/element.rb
 - lib/rsyntaxtree/elementlist.rb
 - lib/rsyntaxtree/format_converter.rb
-- lib/rsyntaxtree/lsif_graph.rb
+- lib/rsyntaxtree/json_graph.rb
 - lib/rsyntaxtree/markup_parser.rb
 - lib/rsyntaxtree/notation_core.md
 - lib/rsyntaxtree/notation_examples.md