tabry 0.1.1 → 0.1.2

data/treesitter/src/tree_sitter/parser.h

@@ -0,0 +1,223 @@
+#ifndef TREE_SITTER_PARSER_H_
+#define TREE_SITTER_PARSER_H_
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <stdbool.h>
+#include <stdint.h>
+#include <stdlib.h>
+
+#define ts_builtin_sym_error ((TSSymbol)-1)
+#define ts_builtin_sym_end 0
+#define TREE_SITTER_SERIALIZATION_BUFFER_SIZE 1024
+
+typedef uint16_t TSStateId;
+
+#ifndef TREE_SITTER_API_H_
+typedef uint16_t TSSymbol;
+typedef uint16_t TSFieldId;
+typedef struct TSLanguage TSLanguage;
+#endif
+
+typedef struct {
+  TSFieldId field_id;
+  uint8_t child_index;
+  bool inherited;
+} TSFieldMapEntry;
+
+typedef struct {
+  uint16_t index;
+  uint16_t length;
+} TSFieldMapSlice;
+
+typedef struct {
+  bool visible;
+  bool named;
+  bool supertype;
+} TSSymbolMetadata;
+
+typedef struct TSLexer TSLexer;
+
+struct TSLexer {
+  int32_t lookahead;
+  TSSymbol result_symbol;
+  void (*advance)(TSLexer *, bool);
+  void (*mark_end)(TSLexer *);
+  uint32_t (*get_column)(TSLexer *);
+  bool (*is_at_included_range_start)(const TSLexer *);
+  bool (*eof)(const TSLexer *);
+};
+
+typedef enum {
+  TSParseActionTypeShift,
+  TSParseActionTypeReduce,
+  TSParseActionTypeAccept,
+  TSParseActionTypeRecover,
+} TSParseActionType;
+
+typedef union {
+  struct {
+    uint8_t type;
+    TSStateId state;
+    bool extra;
+    bool repetition;
+  } shift;
+  struct {
+    uint8_t type;
+    uint8_t child_count;
+    TSSymbol symbol;
+    int16_t dynamic_precedence;
+    uint16_t production_id;
+  } reduce;
+  uint8_t type;
+} TSParseAction;
+
+typedef struct {
+  uint16_t lex_state;
+  uint16_t external_lex_state;
+} TSLexMode;
+
+typedef union {
+  TSParseAction action;
+  struct {
+    uint8_t count;
+    bool reusable;
+  } entry;
+} TSParseActionEntry;
+
+struct TSLanguage {
+  uint32_t version;
+  uint32_t symbol_count;
+  uint32_t alias_count;
+  uint32_t token_count;
+  uint32_t external_token_count;
+  uint32_t state_count;
+  uint32_t large_state_count;
+  uint32_t production_id_count;
+  uint32_t field_count;
+  uint16_t max_alias_sequence_length;
+  const uint16_t *parse_table;
+  const uint16_t *small_parse_table;
+  const uint32_t *small_parse_table_map;
+  const TSParseActionEntry *parse_actions;
+  const char * const *symbol_names;
+  const char * const *field_names;
+  const TSFieldMapSlice *field_map_slices;
+  const TSFieldMapEntry *field_map_entries;
+  const TSSymbolMetadata *symbol_metadata;
+  const TSSymbol *public_symbol_map;
+  const uint16_t *alias_map;
+  const TSSymbol *alias_sequences;
+  const TSLexMode *lex_modes;
+  bool (*lex_fn)(TSLexer *, TSStateId);
+  bool (*keyword_lex_fn)(TSLexer *, TSStateId);
+  TSSymbol keyword_capture_token;
+  struct {
+    const bool *states;
+    const TSSymbol *symbol_map;
+    void *(*create)(void);
+    void (*destroy)(void *);
+    bool (*scan)(void *, TSLexer *, const bool *symbol_whitelist);
+    unsigned (*serialize)(void *, char *);
+    void (*deserialize)(void *, const char *, unsigned);
+  } external_scanner;
+};
+
+/*
+ *  Lexer Macros
+ */
+
+#define START_LEXER()           \
+  bool result = false;          \
+  bool skip = false;            \
+  bool eof = false;             \
+  int32_t lookahead;            \
+  goto start;                   \
+  next_state:                   \
+  lexer->advance(lexer, skip);  \
+  start:                        \
+  skip = false;                 \
+  lookahead = lexer->lookahead;
+
+#define ADVANCE(state_value) \
+  {                          \
+    state = state_value;     \
+    goto next_state;         \
+  }
+
+#define SKIP(state_value) \
+  {                       \
+    skip = true;          \
+    state = state_value;  \
+    goto next_state;      \
+  }
+
+#define ACCEPT_TOKEN(symbol_value)     \
+  result = true;                       \
+  lexer->result_symbol = symbol_value; \
+  lexer->mark_end(lexer);
+
+#define END_STATE() return result;
+
+/*
+ *  Parse Table Macros
+ */
+
+#define SMALL_STATE(id) id - LARGE_STATE_COUNT
+
+#define STATE(id) id
+
+#define ACTIONS(id) id
+
+#define SHIFT(state_value)            \
+  {{                                  \
+    .shift = {                        \
+      .type = TSParseActionTypeShift, \
+      .state = state_value            \
+    }                                 \
+  }}
+
+#define SHIFT_REPEAT(state_value)     \
+  {{                                  \
+    .shift = {                        \
+      .type = TSParseActionTypeShift, \
+      .state = state_value,           \
+      .repetition = true              \
+    }                                 \
+  }}
+
+#define SHIFT_EXTRA()                 \
+  {{                                  \
+    .shift = {                        \
+      .type = TSParseActionTypeShift, \
+      .extra = true                   \
+    }                                 \
+  }}
+
+#define REDUCE(symbol_val, child_count_val, ...) \
+  {{                                             \
+    .reduce = {                                  \
+      .type = TSParseActionTypeReduce,           \
+      .symbol = symbol_val,                      \
+      .child_count = child_count_val,            \
+      __VA_ARGS__                                \
+    },                                           \
+  }}
+
+#define RECOVER()                    \
+  {{                                 \
+    .type = TSParseActionTypeRecover \
+  }}
+
+#define ACCEPT_INPUT()              \
+  {{                                \
+    .type = TSParseActionTypeAccept \
+  }}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif  // TREE_SITTER_PARSER_H_

data/treesitter/tabry-compile.js

@@ -0,0 +1,394 @@
+#!/usr/bin/env node
+// TODO: run through grammar and take stock of what is NOT implemented here. look at TODOs.
+// TODO: tests, maybe more syntax checks
+const fs = require('fs');
+const YAML = require('yaml');
+const Parser = require('tree-sitter');
+const TabryGrammar = require('./bindings/node/index.js');
+
+///////////////////////////////////////////////////////////////////
+
+// HELPERS
+
+function die(err) {
+  console.error(`ERROR: ${err}`);
+  process.exit(1);
+}
+
+function handleChildren(state, node) {
+  for (const child of node.namedChildren) {
+    const handler = 'handle' + child.type.replace(/((_|^)\w)/g, k => (k[1] || k[0]).toUpperCase());
+    if (handlers[handler]) {
+      handlers[handler](state, child);
+    } else {
+      console.error(`WARNING: no handler for type ${child.type} (make fn ${handler})`)
+    }
+  }
+}
+
+const childrenOfType = (node, type) => node.namedChildren.filter(x => x.type == type);
+const firstChildOfType = (node, type) => childrenOfType(node, type)[0];
+
+function pickMultiple(node, mappings) {
+  const result = {};
+  for (const name in mappings) {
+    result[name] = childrenOfType(node, mappings[name]);
+  }
+  return result;
+}
+function pickFirst(node, mappings) {
+  const result = {};
+  for (const name in mappings) {
+    result[name] = firstChildOfType(node, mappings[name]);
+  }
+  return result;
+}
+
+function pick(node, firsts, multiples) {
+  return {...pickFirst(node, firsts), ...pickMultiple(node, multiples)};
+}
+
+function textFromString(node) {
+  if (node.type !== 'string') {
+    throw new Error("node is not a string");
+  }
+  let text = node.text;
+  if (text[0] == '"') {
+    try {
+      // TODO: this may notbe all technically correct, may want to just replace
+      // \\ and \" with a regex (carefully constructed, probably equal to the
+      // one in grammar.js)
+      const replacedNewlines = text.replace(/\n/g, "\\n");
+      text = JSON.parse(replacedNewlines);
+    } catch (e) {
+    }
+  }
+  return unindent(text);
+}
+
+// Remove indentation up to the minimum indentation level of the string.
+// e.g.:
+//   desc "
+//        ---- Hello! ----
+//      My thing:
+//        * xyz
+//        * abc
+//   "
+// Will have indentation removed such that My thing has no indentation "---- Hello! ----"
+// has two spaes of indentation.
+function unindent(text) {
+  if (text.includes("\n")) {
+    const lines = text.split("\n").filter(l => l.trim().length !== 0);
+    const minIndent = Math.min.apply(Math, lines.map(l => l.match(/^( *)/)[0].length));
+    const removeIndent = new RegExp(`^ {${minIndent}}`);
+    text = text.split("\n").map(l => l.replace(removeIndent, '')).join("\n");
+    text = text.replace(/^\s*\n/, '').replace(/\s*$/, '');
+  }
+  return text;
+}
+
+function nameFromAt(node) {
+  if (node.type !== 'at_identifier') {
+    throw new Error("node is not a at_identifier");
+  }
+  return node.text.slice(1);
+}
+
+function safePush(object, field, value) {
+  if (!object[field]) {
+    object[field] = [];
+  }
+  object[field].push(value);
+}
+
+function safeSet(object, field, key, value) {
+  if (!object[field]) {
+    object[field] = {}
+  }
+  if (object[field][key]) {
+    die(`ERROR! duplicate ${field} -- ${key}`);
+  }
+  object[field][key] = value;
+}
+
+function checkContext(state, node, allowable) {
+  if (!allowable.includes(state.context)) {
+    die(`ERROR: ${node.type} found in context ${state.context}. Allowable: ${allowable}`);
+  }
+}
+
+/// CREATE* (part II of handlers)
+
+function createFlag(state, {nameAndAliases, desc, mods, ats, block, arg}) {
+  const [name, ...aliases] = textFromString(nameAndAliases).split(",");
+  const flag = {name};
+  if (aliases.length) {
+    flag.aliases = aliases;
+  }
+
+  for (const at of ats) {
+    safePush(flag, 'options', {type: 'include', value: nameFromAt(at)});
+  }
+
+  setMods(flag, mods, {reqd: 'required'});
+
+  if (desc) {
+    flag.description = textFromString(desc);
+  }
+  if (arg) {
+    flag.arg = true;
+  }
+
+  safePush(state.currentNode, 'flags', flag);
+
+  if (block) {
+    handleChildren({...state, currentNode: flag, context: 'flag'}, block);
+  }
+}
+
+function createIncludeObjectOnSub(state, {at}) {
+  const name = nameFromAt(at);
+  for (const type of ['arg', 'flag', 'sub']) {
+    safePush(state.currentNode, type + 's', {include: name});
+  }
+}
+
+function createSub(state, {nameAndAliases, ats, block, desc}) {
+  const [name, ...aliases] = textFromString(nameAndAliases).split(",");
+  const sub = {name};
+  safePush(state.currentNode, 'subs', sub);
+
+  if (aliases.length) {
+    sub.aliases = aliases;
+  }
+  if (desc) {
+    sub.description = textFromString(desc);
+  }
+
+  const newState = {...state, currentNode: sub, context: 'sub'};
+
+  for (const at of ats) {
+    createIncludeObjectOnSub(newState, {at});
+  }
+
+  if (block) {
+    handleChildren(newState, block);
+  }
+}
+
+function createOpts(state, {type, value}) {
+  safePush(state.currentNode, 'options', { type: type, value: value});
+}
+
+function setMods(object, mods, tabryToYaml) {
+  for (const mod of mods) {
+    const yamlName = tabryToYaml[mod.text];
+    if (yamlName) {
+      object[yamlName] = true
+    } else {
+      die(`Unknown mod ${mod.text}`);
+    }
+  }
+}
+
+function createArg(state, {name, desc, mods, ats, block, varargs}) {
+  const arg = {};
+  if (name) {
+    arg.name = textFromString(name);
+  }
+  setMods(arg, mods, {opt: 'optional'})
+
+  for (const at of ats) {
+    safePush(arg, 'options', {type: 'include', value: nameFromAt(at)});
+  }
+
+  if (desc) {
+    arg.description = textFromString(desc);
+  }
+  if (varargs) {
+    arg.varargs = true;
+  }
+
+  safePush(state.currentNode, 'args', arg);
+
+  if (block) {
+    handleChildren({...state, currentNode: arg, context: 'arg'}, block);
+  }
+}
+
+function createOptionInclude(state, {type, at, block}) {
+  const list = [];
+  const include = { options: list };
+  safeSet(state.output, 'option_includes', nameFromAt(at), list);
+  handleChildren({...state, currentNode: include, context: 'option_include'}, block);
+}
+
+function createArgInclude(state, {type, at, block}) {
+  const include = {};
+  safeSet(state.output, 'arg_includes', nameFromAt(at), include);
+  handleChildren({...state, currentNode: include, context: 'arg_include'}, block);
+}
+
+/// STATEMENT HANDLERS
+
+const handlers = {
+  handleDescStatement(state, node) {
+    checkContext(state, node, ['main', 'sub', 'arg', 'flag']);
+    state.currentNode.description = textFromString(firstChildOfType(node, 'string'));
+  },
+
+  handleTitleStatement(state, node) {
+    checkContext(state, node, ['arg']);
+    state.currentNode.title = textFromString(firstChildOfType(node, 'string'));
+  },
+
+  handleFlagStatement(state, node, arg=false) {
+    checkContext(state, node, ['sub', 'main', 'arg_include']);
+    const {mods, desc, names, ats, block} = pick(node, {
+      block: 'block',
+      names: 'flag_name_list',
+      desc: 'string',
+    }, {
+      mods: 'flag_modifier',
+      ats: 'at_identifier',
+    });
+
+    for (const nameAndAliases of names.namedChildren) {
+      createFlag(state, {nameAndAliases, desc, mods, ats, block, arg});
+    }
+  },
+
+  handleFlagargStatement(state, node) {
+    handlers.handleFlagStatement(state, node, true);
+  },
+
+  handleCmdStatement(state, node) {
+    checkContext(state, node, ['main']);
+    state.output.cmd = textFromString(firstChildOfType(node, 'string'));
+  },
+
+  handleSubStatement(state, node) {
+    checkContext(state, node, ['main', 'sub', 'arg_include']);
+    const {names, ats, block, desc} = pick(node, {
+      block: 'block',
+      names: 'sub_name_list',
+      desc: 'string',
+    }, {
+      ats: 'at_identifier',
+    });
+
+    for (const nameAndAliases of names.namedChildren) {
+      createSub(state, {nameAndAliases, ats, block, desc});
+    }
+  },
+
+  handleArgStatement(state, node) {
+    checkContext(state, node, ['main', 'sub', 'arg_include']);
+    const {mods, type, names, ats, block, desc} = pick(node, {
+      type: 'arg_type',
+      block: 'block',
+      desc: 'string',
+      names: 'arg_name_list',
+    }, {
+      mods: 'arg_modifier',
+      ats: 'at_identifier',
+    });
+    const varargs = type.text === 'varargs';
+    if (varargs && names && names.length > 1) {
+      die(`Args statement may have a maximum of one name`);
+    }
+    const effectiveNames = names ? names.namedChildren : [undefined];
+    for (const name of effectiveNames) {
+      createArg(state, {name, mods, ats, block, desc, varargs});
+    }
+  },
+
+  handleOptsConstStatement(state, node) {
+    checkContext(state, node, ['arg', 'flag', 'option_include']);
+    const {names} = pick(node, {}, {names: 'string'});
+    for (const name of names) {
+      createOpts(state, {type: 'const', value: textFromString(name)});
+    }
+  },
+
+  handleOptsShellStatement(state, node) {
+    checkContext(state, node, ['arg', 'flag', 'option_include']);
+    const {value} = pick(node, {value: 'string'});
+    createOpts(state, {type: 'shell', value: textFromString(value)});
+  },
+
+  handleOptsFileStatement(state, node) {
+    checkContext(state, node, ['arg', 'flag', 'option_include']);
+    createOpts(state, {type: 'file'});
+  },
+
+  handleOptsDirStatement(state, node) {
+    checkContext(state, node, ['arg', 'flag', 'option_include']);
+    createOpts(state, {type: 'dir'});
+  },
+
+  handleIncludeStatement(state, node) {
+    checkContext(state, node, ['arg', 'flag', 'option_include',
+      'arg_include', 'sub', 'main']);
+    const {at} = pick(node, {at: 'at_identifier'});
+    if (['arg_include', 'sub', 'main'].includes(state.context)) {
+      createIncludeObjectOnSub(state, {at});
+    } else {
+      createOpts(state, {type: 'include', value: nameFromAt(at)});
+    }
+  },
+
+  handleDefargsStatement(state, node) {
+    checkContext(state, node, ['main']);
+    const {at, block} = pick(node, {at: 'at_identifier', block: 'block'});
+    createArgInclude(state, {at, block});
+  },
+  handleDefoptsStatement(state, node) {
+    checkContext(state, node, ['main']);
+    const {at, block} = pick(node, {at: 'at_identifier', block: 'block'});
+    createOptionInclude(state, {at, block});
+  },
+
+  handleERROR(state, node) {
+    die(`Tree-sitter parse error. Try running npx tree-sitter parse. State:\n${JSON.stringify(state)}`);
+  },
+}
+
+
+function parseText(text) {
+  const parser = new Parser();
+  parser.setLanguage(TabryGrammar);
+  const tree = parser.parse(text);
+  return tree;
+}
+
+function transformText(text) {
+  const tree = parseText(text);
+  const output = {cmd: null, main: {}};
+  const state = {
+    output,
+    context: 'main', currentNode: output.main,
+  };
+  handleChildren(state, tree.rootNode);
+  return output;
+}
+
+///// MAIN
+
+if (require.main === module) {
+  const filename = process.argv[2];
+  const outputFn = process.argv[3];
+  if (!filename) { die("usage: compile.js file.tabry [output.json] # or output.yml"); }
+  const text = fs.readFileSync(filename).toString();
+  const output = transformText(text);
+
+  if (!outputFn) {
+    console.log(YAML.stringify(output, null, {aliasDuplicateObjects: false, version: '1.1'}));
+  } else if (outputFn.match(/\.yml/)) {
+    fs.writeFileSync(outputFn, YAML.stringify(output, null, {aliasDuplicateObjects: false, version: '1.1'}));
+  } else {
+    fs.writeFileSync(outputFn, JSON.stringify(output));
+  }
+}
+
+module.exports = {transformText};

data/treesitter/tabry-compile.test.js

@@ -0,0 +1,51 @@
+const tabryCompile = require('./tabry-compile');
+const fs = require('fs');
+const yaml = require('yaml');
+
+function loadTreesitterExamplesFromTxtFile(filename) {
+  const examples = [];
+  const lines = fs.readFileSync(filename).toString().split("\n");
+  let nameMode = false;
+  // ============
+  // Example name
+  // ============
+  // example stuff
+  // ---
+  // example stuff2
+  for (const l of lines) {
+    if (l.match(/^===+/)) {
+      nameMode = !nameMode;
+    } else if (nameMode) {
+      examples.push({name: l, text: ''});
+    } else if (examples.length) {
+      const lastExample = examples[examples.length - 1];
+      lastExample.text += l + "\n";
+    }
+  };
+
+  // result: {name: "example_name", text: "example stuff"}
+  return examples.map(ex => ({
+    name: ex.name,
+    filename: ex.name.toLowerCase().replace(/[^a-zA-Z0-9]/g, '_'),
+    text: ex.text.split("\n---\n")[0]
+  }));
+}
+
+const corpusFile = __dirname + "/corpus/examples_from_language_reference.txt";
+const examples = loadTreesitterExamplesFromTxtFile(corpusFile);
+const fixtureDir = __dirname + "/jest_fixtures/examples_from_language_reference/";
+
+describe('examples from language reference', () => {
+  for (const example of examples) {
+    describe(example.name, () => {
+      it("parses correctly", () => {
+        const fixture = `${fixtureDir}${example.filename}.yml`;
+        const expected = yaml.parse(fs.readFileSync(fixture).toString());
+
+        const actual = tabryCompile.transformText(example.text);
+
+        expect(actual).toEqual(expected);
+      });
+    });
+  }
+});

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: tabry
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.1.2
 platform: ruby
 authors:
 - Evan Battaglia
@@ -165,6 +165,41 @@ files:
 - spec/tabry/runner_spec.rb
 - spec/tabry/usage_generator_spec.rb
 - tabry.gemspec
+- treesitter/Cargo.toml
+- treesitter/README.md
+- treesitter/binding.gyp
+- treesitter/bindings/node/binding.cc
+- treesitter/bindings/node/index.js
+- treesitter/bindings/rust/build.rs
+- treesitter/bindings/rust/lib.rs
+- treesitter/corpus/arg.txt
+- treesitter/corpus/at.txt
+- treesitter/corpus/comment.txt
+- treesitter/corpus/desc.txt
+- treesitter/corpus/examples_from_language_reference.txt
+- treesitter/corpus/flag.txt
+- treesitter/corpus/flag_desc_inline.txt
+- treesitter/corpus/opts.txt
+- treesitter/corpus/rapture.txt
+- treesitter/grammar.js
+- treesitter/jest_fixtures/examples_from_language_reference/argument_titles.yml
+- treesitter/jest_fixtures/examples_from_language_reference/arguments_and_possible_options__arg_.yml
+- treesitter/jest_fixtures/examples_from_language_reference/flags__flag__flagarg__reqd_flagarg_.yml
+- treesitter/jest_fixtures/examples_from_language_reference/getting_started.yml
+- treesitter/jest_fixtures/examples_from_language_reference/includes.yml
+- treesitter/jest_fixtures/examples_from_language_reference/multi_line_descriptions.yml
+- treesitter/jest_fixtures/examples_from_language_reference/optional_args_and_varargs__opt_arg__varargs__opt_varargs_.yml
+- treesitter/jest_fixtures/examples_from_language_reference/options.yml
+- treesitter/jest_fixtures/examples_from_language_reference/subcommands__sub__1.yml
+- treesitter/jest_fixtures/examples_from_language_reference/subcommands__sub__2.yml
+- treesitter/package.json
+- treesitter/parser_compile.sh
+- treesitter/src/grammar.json
+- treesitter/src/node-types.json
+- treesitter/src/parser.c
+- treesitter/src/tree_sitter/parser.h
+- treesitter/tabry-compile.js
+- treesitter/tabry-compile.test.js
 homepage: https://github.com/evanbattaglia/tabry
 licenses:
 - MIT