node/tools/doc/json.js

module.exports = doJSON;

// Take the lexed input, and return a JSON-encoded object
// A module looks like this: https://gist.github.com/1777387

var marked = require('marked');

function doJSON(input, filename, cb) {
  var root = {source: filename};
  var stack = [root];
  var depth = 0;
  var current = root;
  var state = null;
  var lexed = marked.lexer(input);
  lexed.forEach(function (tok) {
    var type = tok.type;
    var text = tok.text;

    // <!-- type = module -->
    // This is for cases where the markdown semantic structure is lacking.
    if (type === 'paragraph' || type === 'html') {
      var metaExpr = /<!--([^=]+)=([^\-]+)-->\n*/g;
      text = text.replace(metaExpr, function(_0, k, v) {
        current[k.trim()] = v.trim();
        return '';
      });
      text = text.trim();
      if (!text) return;
    }

    if (type === 'heading' &&
        !text.trim().match(/^example/i)) {
      if (tok.depth - depth > 1) {
        return cb(new Error('Inappropriate heading level\n'+
                            JSON.stringify(tok)));
      }

      // Sometimes we have two headings with a single
      // blob of description.  Treat as a clone.
      if (current &&
          state === 'AFTERHEADING' &&
          depth === tok.depth) {
        var clone = current;
        current = newSection(tok);
        current.clone = clone;
        // don't keep it around on the stack.
        stack.pop();
      } else {
        // if the level is greater than the current depth,
        // then it's a child, so we should just leave the stack
        // as it is.
        // However, if it's a sibling or higher, then it implies
        // the closure of the other sections that came before.
        // root is always considered the level=0 section,
        // and the lowest heading is 1, so this should always
        // result in having a valid parent node.
        var d = tok.depth;
        while (d <= depth) {
          finishSection(stack.pop(), stack[stack.length - 1]);
          d++;
        }
        current = newSection(tok);
      }

      depth = tok.depth;
      stack.push(current);
      state = 'AFTERHEADING';
      return;
    } // heading

    // Immediately after a heading, we can expect the following
    //
    // { type: 'code', text: 'Stability: ...' },
    //
    // a list: starting with list_start, ending with list_end,
    // maybe containing other nested lists in each item.
    //
    // If one of these isnt' found, then anything that comes between
    // here and the next heading should be parsed as the desc.
    var stability
    if (state === 'AFTERHEADING') {
      if (type === 'code' &&
          (stability = text.match(/^Stability: ([0-5])(?:\s*-\s*)?(.*)$/))) {
        current.stability = parseInt(stability[1], 10);
        current.stabilityText = stability[2].trim();
        return;
      } else if (type === 'list_start' && !tok.ordered) {
        state = 'AFTERHEADING_LIST';
        current.list = current.list || [];
        current.list.push(tok);
        current.list.level = 1;
      } else {
        current.desc = current.desc || [];
        if (!Array.isArray(current.desc)) {
          current.shortDesc = current.desc;
          current.desc = [];
        }
        current.desc.push(tok);
        state = 'DESC';
      }
      return;
    }

    if (state === 'AFTERHEADING_LIST') {
      current.list.push(tok);
      if (type === 'list_start') {
        current.list.level++;
      } else if (type === 'list_end') {
        current.list.level--;
      }
      if (current.list.level === 0) {
        state = 'AFTERHEADING';
        processList(current);
      }
      return;
    }

    current.desc = current.desc || [];
    current.desc.push(tok);

  });

  // finish any sections left open
  while (root !== (current = stack.pop())) {
    finishSection(current, stack[stack.length - 1]);
  }

  return cb(null, root)
}


// go from something like this:
// [ { type: 'list_item_start' },
//   { type: 'text',
//     text: '`settings` Object, Optional' },
//   { type: 'list_start', ordered: false },
//   { type: 'list_item_start' },
//   { type: 'text',
//     text: 'exec: String, file path to worker file.  Default: `__filename`' },
//   { type: 'list_item_end' },
//   { type: 'list_item_start' },
//   { type: 'text',
//     text: 'args: Array, string arguments passed to worker.' },
//   { type: 'text',
//     text: 'Default: `process.argv.slice(2)`' },
//   { type: 'list_item_end' },
//   { type: 'list_item_start' },
//   { type: 'text',
//     text: 'silent: Boolean, whether or not to send output to parent\'s stdio.' },
//   { type: 'text', text: 'Default: `false`' },
//   { type: 'space' },
//   { type: 'list_item_end' },
//   { type: 'list_end' },
//   { type: 'list_item_end' },
//   { type: 'list_end' } ]
// to something like:
// [ { name: 'settings',
//     type: 'object',
//     optional: true,
//     settings:
//      [ { name: 'exec',
//          type: 'string',
//          desc: 'file path to worker file',
//          default: '__filename' },
//        { name: 'args',
//          type: 'array',
//          default: 'process.argv.slice(2)',
//          desc: 'string arguments passed to worker.' },
//        { name: 'silent',
//          type: 'boolean',
//          desc: 'whether or not to send output to parent\'s stdio.',
//          default: 'false' } ] } ]

function processList(section) {
  var list = section.list;
  var values = [];
  var current;
  var stack = [];

  // for now, *just* build the heirarchical list
  list.forEach(function(tok) {
    var type = tok.type;
    if (type === 'space') return;
    if (type === 'list_item_start') {
      if (!current) {
        var n = {};
        values.push(n);
        current = n;
      } else {
        current.options = current.options || [];
        stack.push(current);
        var n = {};
        current.options.push(n);
        current = n;
      }
      return;
    } else if (type === 'list_item_end') {
      if (!current) {
        throw new Error('invalid list - end without current item\n' +
                        JSON.stringify(tok) + '\n' +
                        JSON.stringify(list));
      }
      current = stack.pop();
    } else if (type === 'text') {
      if (!current) {
        throw new Error('invalid list - text without current item\n' +
                        JSON.stringify(tok) + '\n' +
                        JSON.stringify(list));
      }
      current.textRaw = current.textRaw || '';
      current.textRaw += tok.text + ' ';
    }
  });

  // shove the name in there for properties, since they are always
  // just going to be the value etc.
  if (section.type === 'property' && values[0]) {
    values[0].textRaw = '`' + section.name + '` ' + values[0].textRaw;
  }

  // now pull the actual values out of the text bits.
  values.forEach(parseListItem);

  // Now figure out what this list actually means.
  // depending on the section type, the list could be different things.

  switch (section.type) {
    case 'ctor':
    case 'classMethod':
    case 'method':
      // each item is an argument, unless the name is 'return',
      // in which case it's the return value.
      section.signatures = section.signatures || [];
      var sig = {}
      section.signatures.push(sig);
      sig.params = values.filter(function(v) {
        if (v.name === 'return') {
          sig.return = v;
          return false;
        }
        return true;
      });
      parseSignature(section.textRaw, sig);
      break;

    case 'property':
      // there should be only one item, which is the value.
      // copy the data up to the section.
      var value = values[0] || {};
      delete value.name;
      section.typeof = value.type;
      delete value.type;
      Object.keys(value).forEach(function(k) {
        section[k] = value[k];
      });
      break;

    case 'event':
      // event: each item is an argument.
      section.params = values;
      break;
  }

  // section.listParsed = values;
  delete section.list;
}


// textRaw = "someobject.someMethod(a[, b=100][, c])"
function parseSignature(text, sig) {
  var params = text.match(paramExpr);
  if (!params) return;
  params = params[1];
  // the [ is irrelevant. ] indicates optionalness.
  params = params.replace(/\[/g, '');
  params = params.split(/,/)
  params.forEach(function(p, i, _) {
    p = p.trim();
    if (!p) return;
    var param = sig.params[i];
    var optional = false;
    var def;
    // [foo] -> optional
    if (p.charAt(p.length - 1) === ']') {
      optional = true;
      p = p.substr(0, p.length - 1);
      p = p.trim();
    }
    var eq = p.indexOf('=');
    if (eq !== -1) {
      def = p.substr(eq + 1);
      p = p.substr(0, eq);
    }
    if (!param) {
      param = sig.params[i] = { name: p };
    }
    // at this point, the name should match.
    if (p !== param.name) {
      console.error('Warning: invalid param "%s"', p);
      console.error(' > ' + JSON.stringify(param));
      console.error(' > ' + text);
    }
    if (optional) param.optional = true;
    if (def !== undefined) param.default = def;
  });
}


function parseListItem(item) {
  if (item.options) item.options.forEach(parseListItem);
  if (!item.textRaw) return;

  // the goal here is to find the name, type, default, and optional.
  // anything left over is 'desc'
  var text = item.textRaw.trim();
  // text = text.replace(/^(Argument|Param)s?\s*:?\s*/i, '');

  text = text.replace(/^, /, '').trim();
  var retExpr = /^returns?\s*:?\s*/i;
  var ret = text.match(retExpr);
  if (ret) {
    item.name = 'return';
    text = text.replace(retExpr, '');
  } else {
    var nameExpr = /^['`"]?([^'`": \{]+)['`"]?\s*:?\s*/;
    var name = text.match(nameExpr);
    if (name) {
      item.name = name[1];
      text = text.replace(nameExpr, '');
    }
  }

  text = text.trim();
  var defaultExpr = /\(default\s*[:=]?\s*['"`]?([^, '"`]*)['"`]?\)/i;
  var def = text.match(defaultExpr);
  if (def) {
    item.default = def[1];
    text = text.replace(defaultExpr, '');
  }

  text = text.trim();
  var typeExpr = /^\{([^\}]+)\}/;
  var type = text.match(typeExpr);
  if (type) {
    item.type = type[1];
    text = text.replace(typeExpr, '');
  }

  text = text.trim();
  var optExpr = /^Optional\.|(?:, )?Optional$/;
  var optional = text.match(optExpr);
  if (optional) {
    item.optional = true;
    text = text.replace(optExpr, '');
  }

  text = text.replace(/^\s*-\s*/, '');
  text = text.trim();
  if (text) item.desc = text;
}


function finishSection(section, parent) {
  if (!section || !parent) {
    throw new Error('Invalid finishSection call\n'+
                    JSON.stringify(section) + '\n' +
                    JSON.stringify(parent));
  }

  if (!section.type) {
    section.type = 'module';
    if (parent && (parent.type === 'misc')) {
      section.type = 'misc';
    }
    section.displayName = section.name;
    section.name = section.name.toLowerCase()
      .trim().replace(/\s+/g, '_');
  }

  if (section.desc && Array.isArray(section.desc)) {
    section.desc.links = section.desc.links || [];
    section.desc = marked.parser(section.desc);
  }

  if (!section.list) section.list = [];
  processList(section);

  // classes sometimes have various 'ctor' children
  // which are actually just descriptions of a constructor
  // class signature.
  // Merge them into the parent.
  if (section.type === 'class' && section.ctors) {
    section.signatures = section.signatures || [];
    var sigs = section.signatures;
    section.ctors.forEach(function(ctor) {
      ctor.signatures = ctor.signatures || [{}];
      ctor.signatures.forEach(function(sig) {
        sig.desc = ctor.desc;
      });
      sigs.push.apply(sigs, ctor.signatures);
    });
    delete section.ctors;
  }

  // properties are a bit special.
  // their "type" is the type of object, not "property"
  if (section.properties) {
    section.properties.forEach(function (p) {
      if (p.typeof) p.type = p.typeof;
      else delete p.type;
      delete p.typeof;
    });
  }

  // handle clones
  if (section.clone) {
    var clone = section.clone;
    delete section.clone;
    delete clone.clone;
    deepCopy(section, clone);
    finishSection(clone, parent);
  }

  var plur;
  if (section.type.slice(-1) === 's') {
    plur = section.type + 'es';
  } else if (section.type.slice(-1) === 'y') {
    plur = section.type.replace(/y$/, 'ies');
  } else {
    plur = section.type + 's';
  }

  // if the parent's type is 'misc', then it's just a random
  // collection of stuff, like the "globals" section.
  // Make the children top-level items.
  if (section.type === 'misc') {
    Object.keys(section).forEach(function(k) {
      switch (k) {
        case 'textRaw':
        case 'name':
        case 'type':
        case 'desc':
        case 'miscs':
          return;
        default:
          if (parent.type === 'misc') {
            return;
          }
          if (Array.isArray(k) && parent[k]) {
            parent[k] = parent[k].concat(section[k]);
          } else if (!parent[k]) {
            parent[k] = section[k];
          } else {
            // parent already has, and it's not an array.
            return;
          }
      }
    });
  }

  parent[plur] = parent[plur] || [];
  parent[plur].push(section);
}


// Not a general purpose deep copy.
// But sufficient for these basic things.
function deepCopy(src, dest) {
  Object.keys(src).filter(function(k) {
    return !dest.hasOwnProperty(k);
  }).forEach(function(k) {
    dest[k] = deepCopy_(src[k]);
  });
}

function deepCopy_(src) {
  if (!src) return src;
  if (Array.isArray(src)) {
    var c = new Array(src.length);
    src.forEach(function(v, i) {
      c[i] = deepCopy_(v);
    });
    return c;
  }
  if (typeof src === 'object') {
    var c = {};
    Object.keys(src).forEach(function(k) {
      c[k] = deepCopy_(src[k]);
    });
    return c;
  }
  return src;
}


// these parse out the contents of an H# tag
var eventExpr = /^Event(?::|\s)+['"]?([^"']+).*$/i;
var classExpr = /^Class:\s*([^ ]+).*?$/i;
var propExpr = /^(?:property:?\s*)?[^\.]+\.([^ \.\(\)]+)\s*?$/i;
var braceExpr = /^(?:property:?\s*)?[^\.\[]+(\[[^\]]+\])\s*?$/i;
var classMethExpr =
  /^class\s*method\s*:?[^\.]+\.([^ \.\(\)]+)\([^\)]*\)\s*?$/i;
var methExpr =
  /^(?:method:?\s*)?(?:[^\.]+\.)?([^ \.\(\)]+)\([^\)]*\)\s*?$/i;
var newExpr = /^new ([A-Z][a-z]+)\([^\)]*\)\s*?$/;
var paramExpr = /\((.*)\);?$/;

function newSection(tok) {
  var section = {};
  // infer the type from the text.
  var text = section.textRaw = tok.text;
  if (text.match(eventExpr)) {
    section.type = 'event';
    section.name = text.replace(eventExpr, '$1');
  } else if (text.match(classExpr)) {
    section.type = 'class';
    section.name = text.replace(classExpr, '$1');
  } else if (text.match(braceExpr)) {
    section.type = 'property';
    section.name = text.replace(braceExpr, '$1');
  } else if (text.match(propExpr)) {
    section.type = 'property';
    section.name = text.replace(propExpr, '$1');
  } else if (text.match(classMethExpr)) {
    section.type = 'classMethod';
    section.name = text.replace(classMethExpr, '$1');
  } else if (text.match(methExpr)) {
    section.type = 'method';
    section.name = text.replace(methExpr, '$1');
  } else if (text.match(newExpr)) {
    section.type = 'ctor';
    section.name = text.replace(newExpr, '$1');
  } else {
    section.name = text;
  }
  return section;
}