class Oj::Doc

The Doc class is used to parse and navigate a JSON document. The model it employs is that of a document that while open can be navigated and values extracted. Once the document is closed the document can not longer be accessed. This allows the parsing and data extraction to be extremely fast compared to other JSON parses.

An Oj::Doc class is not created directly but the _open()_ class method is used to open a document and the yield parameter to the block of the open() call is the Doc instance. The Doc instance can be moved across, up, and down the JSON document. At each element the data associated with the element can be extracted. It is also possible to just provide a path to the data to be extracted and retrieve the data in that manner.

For many of the methods a path is used to describe the location of an element. Paths follow a subset of the XPath syntax. The slash ('/') character is the separator. Each step in the path identifies the next branch to take through the document. A JSON object will expect a key string while an array will expect a positive index. A .. step indicates a move up the JSON document.

@example

json = %Q{[
  {
    "one"   : 1,
    "two"   : 2
  },
  {
    "three" : 3,
    "four"  : 4
  }
]}
# move and get value
Oj::Doc.open(json) do |doc|
  doc.move('/1/two')  
  # doc location is now at the 'two' element of the hash that is the first element of the array.
  doc.fetch()
end
#=> 2

# Now try again using a path to Oj::Doc.fetch() directly and not using a block.
doc = Oj::Doc.open(json)
doc.fetch('/2/three')  #=> 3
doc.close()

Public Class Methods

open(json) { |doc| ... } → Object click to toggle source

Parses a JSON document String and then yields to the provided block if one is given with an instance of the Oj::Doc as the single yield parameter. If a block is not given then an Oj::Doc instance is returned and must be closed with a call to the close() method when no longer needed.

@param [String] json JSON document string @yieldparam [Oj::Doc] doc parsed JSON document @yieldreturn [Object] returns the result of the yield as the result of the method call @example

Oj::Doc.open('[1,2,3]') { |doc| doc.size() }  #=> 4
# or as an alternative
doc = Oj::Doc.open('[1,2,3]')
doc.size()  #=> 4
doc.close()
static VALUE
doc_open(VALUE clas, VALUE str) {
    char        *json;
    size_t      len;
    VALUE       obj;
    int         given = rb_block_given_p();
    int         allocate;

    Check_Type(str, T_STRING);
    len = RSTRING_LEN(str) + 1;
    allocate = (SMALL_XML < len || !given);
    if (allocate) {
        json = ALLOC_N(char, len);
    } else {
        json = ALLOCA_N(char, len);
    }
    memcpy(json, StringValuePtr(str), len);
    obj = parse_json(clas, json, given, allocate);
    if (given && allocate) {
        xfree(json);
    }
    return obj;
}
open_file(filename) { |doc| ... } → Object click to toggle source

Parses a JSON document from a file and then yields to the provided block if one is given with an instance of the Oj::Doc as the single yield parameter. If a block is not given then an Oj::Doc instance is returned and must be closed with a call to the close() method when no longer needed.

@param [String] filename name of file that contains a JSON document @yieldparam [Oj::Doc] doc parsed JSON document @yieldreturn [Object] returns the result of the yield as the result of the method call @example

File.open('array.json', 'w') { |f| f.write('[1,2,3]') }
Oj::Doc.open_file(filename) { |doc| doc.size() }  #=> 4
# or as an alternative
doc = Oj::Doc.open_file(filename)
doc.size()  #=> 4
doc.close()
static VALUE
doc_open_file(VALUE clas, VALUE filename) {
    char        *path;
    char        *json;
    FILE        *f;
    size_t      len;
    VALUE       obj;
    int         given = rb_block_given_p();
    int         allocate;

    Check_Type(filename, T_STRING);
    path = StringValuePtr(filename);
    if (0 == (f = fopen(path, "r"))) {
        rb_raise(rb_eIOError, "%s", strerror(errno));
    }
    fseek(f, 0, SEEK_END);
    len = ftell(f);
    allocate = (SMALL_XML < len || !given);
    if (allocate) {
        json = ALLOC_N(char, len + 1);
    } else {
        json = ALLOCA_N(char, len + 1);
    }
    fseek(f, 0, SEEK_SET);
    if (len != fread(json, 1, len, f)) {
        fclose(f);
        rb_raise(rb_const_get_at(Oj, rb_intern("LoadError")), 
                 "Failed to read %lu bytes from %s.", (unsigned long)len, path);
    }
    fclose(f);
    json[len] = '\0';
    obj = parse_json(clas, json, given, allocate);
    if (given && allocate) {
        xfree(json);
    }
    return obj;
}
parse(p1) click to toggle source

@see ::open

static VALUE
doc_open(VALUE clas, VALUE str) {
    char        *json;
    size_t      len;
    VALUE       obj;
    int         given = rb_block_given_p();
    int         allocate;

    Check_Type(str, T_STRING);
    len = RSTRING_LEN(str) + 1;
    allocate = (SMALL_XML < len || !given);
    if (allocate) {
        json = ALLOC_N(char, len);
    } else {
        json = ALLOCA_N(char, len);
    }
    memcpy(json, StringValuePtr(str), len);
    obj = parse_json(clas, json, given, allocate);
    if (given && allocate) {
        xfree(json);
    }
    return obj;
}

Public Instance Methods

clone() click to toggle source
static VALUE
doc_not_implemented(VALUE self) {
    rb_raise(rb_eNotImpError, "Not implemented.");
    return Qnil;
}
close() → nil click to toggle source

Closes an open document. No further calls to the document will be valid after closing. @example

doc = Oj::Doc.open('[1,2,3]')
doc.size()  #=> 4
doc.close()
static VALUE
doc_close(VALUE self) {
    Doc         doc = self_doc(self);

    rb_gc_unregister_address(&doc->self);
    DATA_PTR(doc->self) = 0;
    if (0 != doc) {
        xfree(doc->json);
        doc_free(doc);
        xfree(doc);
    }
    return Qnil;
}
dump(path=nil) → String click to toggle source

Dumps the document or nodes to a new JSON document. It uses the default options for generating the JSON. @param [String] path if provided it identified the top of the branch to dump to JSON @param [String] filename if provided it is the filename to write the output to @example

Oj::Doc.open('[3,[2,1]]') { |doc|
    doc.dump('/2')
}
#=> "[2,1]"
static VALUE
doc_dump(int argc, VALUE *argv, VALUE self) {
    Doc         doc = self_doc(self);
    Leaf        leaf;
    const char  *path = 0;
    const char  *filename = 0;

    if (1 <= argc) {
        if (Qnil != *argv) {
            Check_Type(*argv, T_STRING);
            path = StringValuePtr(*argv);
        }
        if (2 <= argc) {
            Check_Type(argv[1], T_STRING);
            filename = StringValuePtr(argv[1]);
        }
    }
    if (0 != (leaf = get_doc_leaf(doc, path))) {
        VALUE  rjson;

        if (0 == filename) {
            char       buf[4096];
            struct _Out out;

            out.buf = buf;
            out.end = buf + sizeof(buf) - 10;
            out.allocated = 0;
            out.omit_nil = oj_default_options.dump_opts.omit_nil;
            oj_dump_leaf_to_json(leaf, &oj_default_options, &out);
            rjson = rb_str_new2(out.buf);
            if (out.allocated) {
                xfree(out.buf);
            }
        } else {
            oj_write_leaf_to_file(leaf, filename, &oj_default_options);
            rjson = Qnil;
        }
        return rjson;
    }
    return Qnil;
}
dup() click to toggle source
static VALUE
doc_not_implemented(VALUE self) {
    rb_raise(rb_eNotImpError, "Not implemented.");
    return Qnil;
}
each_child(path=nil) { |doc| ... } → nil click to toggle source

Yields to the provided block for each immediate child node with the identified location of the JSON document as the root. The parameter passed to the block on yield is the Doc instance after moving to the child location. @param [String] path if provided it identified the top of the branch to process the chilren of @yieldparam [Doc] Doc at the child location @example

Oj::Doc.open('[3,[2,1]]') { |doc|
    result = []
    doc.each_value('/2') { |doc| result << doc.where? }
    result
}
#=> ["/2/1", "/2/2"]
static VALUE
doc_each_child(int argc, VALUE *argv, VALUE self) {
    if (rb_block_given_p()) {
        Leaf           save_path[MAX_STACK];
        Doc            doc = self_doc(self);
        const char     *path = 0;
        size_t         wlen;

        wlen = doc->where - doc->where_path;
        if (0 < wlen) {
            memcpy(save_path, doc->where_path, sizeof(Leaf) * (wlen + 1));
        }
        if (1 <= argc) {
            Check_Type(*argv, T_STRING);
            path = StringValuePtr(*argv);
            if ('/' == *path) {
                doc->where = doc->where_path;
                path++;
            }
            if (0 != move_step(doc, path, 1)) {
                if (0 < wlen) {
                    memcpy(doc->where_path, save_path, sizeof(Leaf) * (wlen + 1));
                }
                return Qnil;
            }
        }
        if (COL_VAL == (*doc->where)->value_type && 0 != (*doc->where)->elements) {
            Leaf       first = (*doc->where)->elements->next;
            Leaf       e = first;

            doc->where++;
            do {
                *doc->where = e;
                rb_yield(self);
                e = e->next;
            } while (e != first);
        }
        if (0 < wlen) {
            memcpy(doc->where_path, save_path, sizeof(Leaf) * (wlen + 1));
        }
    }
    return Qnil;
}
each_leaf(path=nil) → nil click to toggle source

Yields to the provided block for each leaf node with the identified location of the JSON document as the root. The parameter passed to the block on yield is the Doc instance after moving to the child location. @param [String] path if provided it identified the top of the branch to process the leaves of @yieldparam [Doc] Doc at the child location @example

Oj::Doc.open('[3,[2,1]]') { |doc|
    result = {}
    doc.each_leaf() { |d| result[d.where?] = d.fetch() }
    result
}
#=> ["/1" => 3, "/2/1" => 2, "/2/2" => 1]
static VALUE
doc_each_leaf(int argc, VALUE *argv, VALUE self) {
    if (rb_block_given_p()) {
        Leaf           save_path[MAX_STACK];
        Doc            doc = self_doc(self);
        const char     *path = 0;
        size_t         wlen;

        wlen = doc->where - doc->where_path;
        if (0 < wlen) {
            memcpy(save_path, doc->where_path, sizeof(Leaf) * (wlen + 1));
        }
        if (1 <= argc) {
            Check_Type(*argv, T_STRING);
            path = StringValuePtr(*argv);
            if ('/' == *path) {
                doc->where = doc->where_path;
                path++;
            }
            if (0 != move_step(doc, path, 1)) {
                if (0 < wlen) {
                    memcpy(doc->where_path, save_path, sizeof(Leaf) * (wlen + 1));
                }
                return Qnil;
            }
        }
        each_leaf(doc, self);
        if (0 < wlen) {
            memcpy(doc->where_path, save_path, sizeof(Leaf) * (wlen + 1));
        }
    }
    return Qnil;
}
each_value(path=nil) { |val| ... } → nil click to toggle source

Yields to the provided block for each leaf value in the identified location of the JSON document. The parameter passed to the block on yield is the value of the leaf. Only those leaves below the element specified by the path parameter are processed. @param [String] path if provided it identified the top of the branch to process the leaf values of @yieldparam [Object] val each leaf value @example

Oj::Doc.open('[3,[2,1]]') { |doc|
    result = []
    doc.each_value() { |v| result << v }
    result
}
#=> [3, 2, 1]

Oj::Doc.open('[3,[2,1]]') { |doc|
    result = []
    doc.each_value('/2') { |v| result << v }
    result
}
#=> [2, 1]
static VALUE
doc_each_value(int argc, VALUE *argv, VALUE self) {
    if (rb_block_given_p()) {
        Doc            doc = self_doc(self);
        const char     *path = 0;
        Leaf           leaf;

        if (1 <= argc) {
            Check_Type(*argv, T_STRING);
            path = StringValuePtr(*argv);
        }
        if (0 != (leaf = get_doc_leaf(doc, path))) {
            each_value(doc, leaf);
        }
    }
    return Qnil;
}
fetch(path=nil) → nil, true, false, Fixnum, Float, String, Array, Hash click to toggle source

Returns the value at the location identified by the path or the current location if the path is nil or not provided. This method will create and return an Array or Hash if that is the type of Object at the location specified. This is more expensive than navigating to the leaves of the JSON document. @param [String] path path to the location to get the type of if provided @example

Oj::Doc.open('[1,2]') { |doc| doc.fetch() }      #=> [1, 2]
Oj::Doc.open('[1,2]') { |doc| doc.fetch('/1') }  #=> 1
static VALUE
doc_fetch(int argc, VALUE *argv, VALUE self) {
    Doc         doc;
    Leaf        leaf;
    VALUE       val = Qnil;
    const char  *path = 0;

    doc = self_doc(self);
    if (1 <= argc) {
        Check_Type(*argv, T_STRING);
        path = StringValuePtr(*argv);
        if (2 == argc) {
            val = argv[1];
        }
    }
    if (0 != (leaf = get_doc_leaf(doc, path))) {
        val = leaf_value(doc, leaf);
    }
    return val;
}
home() → nil click to toggle source

Moves the document marker or location to the hoot or home position. The same operation can be performed with a #move('/'). @example

Oj::Doc.open('[1,2,3]') { |doc| doc.move('/2'); doc.home(); doc.where? }  #=> '/'
static VALUE
doc_home(VALUE self) {
    Doc doc = self_doc(self);

    *doc->where_path = doc->data;
    doc->where = doc->where_path;

    return oj_slash_string;
}
local_key() → String, Fixnum, nil click to toggle source

Returns the final key to the current location. @example

Oj::Doc.open('[1,2,3]') { |doc| doc.move('/2'); doc.local_key() }      #=> 2
Oj::Doc.open('{"one":3}') { |doc| doc.move('/one'); doc.local_key() }  #=> "one"
Oj::Doc.open('[1,2,3]') { |doc| doc.local_key() }                      #=> nil
static VALUE
doc_local_key(VALUE self) {
    Doc         doc = self_doc(self);
    Leaf        leaf = *doc->where;
    VALUE       key = Qnil;

    if (T_HASH == leaf->parent_type) {
        key = rb_str_new2(leaf->key);
        key = oj_encode(key);
    } else if (T_ARRAY == leaf->parent_type) {
        key = LONG2NUM(leaf->index);
    }
    return key;
}
move(path) → nil click to toggle source

Moves the document marker to the path specified. The path can an absolute path or a relative path. @param [String] path path to the location to move to @example

Oj::Doc.open('{"one":[1,2]') { |doc| doc.move('/one/2'); doc.where? }  #=> "/one/2"
static VALUE
doc_move(VALUE self, VALUE str) {
    Doc         doc = self_doc(self);
    const char  *path;
    int         loc;

    Check_Type(str, T_STRING);
    path = StringValuePtr(str);
    if ('/' == *path) {
        doc->where = doc->where_path;
        path++;
    }
    if (0 != (loc = move_step(doc, path, 1))) {
        rb_raise(rb_eArgError, "Failed to locate element %d of the path %s.", loc, path);
    }
    return Qnil;
}
size() → Fixnum click to toggle source

Returns the number of nodes in the JSON document where a node is any one of the basic JSON components. @return Returns the size of the JSON document. @example

Oj::Doc.open('[1,2,3]') { |doc| doc.size() }  #=> 4
static VALUE
doc_size(VALUE self) {
    return ULONG2NUM(((Doc)DATA_PTR(self))->size);
}
type(path=nil) → Class click to toggle source

Returns the Class of the data value at the location identified by the path or the current location if the path is nil or not provided. This method does not create the Ruby Object at the location specified so the overhead is low. @param [String] path path to the location to get the type of if provided @example

Oj::Doc.open('[1,2]') { |doc| doc.type() }      #=> Array
Oj::Doc.open('[1,2]') { |doc| doc.type('/1') }  #=> Fixnum
static VALUE
doc_type(int argc, VALUE *argv, VALUE self) {
    Doc         doc = self_doc(self);
    Leaf        leaf;
    const char  *path = 0;
    VALUE       type = Qnil;

    if (1 <= argc) {
        Check_Type(*argv, T_STRING);
        path = StringValuePtr(*argv);
    }
    if (0 != (leaf = get_doc_leaf(doc, path))) {
        switch (leaf->rtype) {
        case T_NIL:    type = rb_cNilClass;       break;
        case T_TRUE:   type = rb_cTrueClass;     break;
        case T_FALSE:  type = rb_cFalseClass;   break;
        case T_STRING: type = rb_cString;      break;
#ifdef RUBY_INTEGER_UNIFICATION
        case T_FIXNUM: type = rb_cInteger;     break;
#else
        case T_FIXNUM: type = rb_cFixnum;      break;
#endif
        case T_FLOAT:  type = rb_cFloat;        break;
        case T_ARRAY:  type = rb_cArray;        break;
        case T_HASH:   type = rb_cHash;  break;
        default:                               break;
        }
    }
    return type;
}
where?() → String click to toggle source

Returns a String that describes the absolute path to the current location in the JSON document.

static VALUE
doc_where(VALUE self) {
    Doc doc = self_doc(self);

    if (0 == *doc->where_path || doc->where == doc->where_path) {
        return oj_slash_string;
    } else {
        Leaf   *lp;
        Leaf   leaf;
        size_t size = 3; // leading / and terminating \0
        char   *path;
        char   *p;

        for (lp = doc->where_path; lp <= doc->where; lp++) {
            leaf = *lp;
            if (T_HASH == leaf->parent_type) {
                size += esc_strlen((*lp)->key) + 1;
            } else if (T_ARRAY == leaf->parent_type) {
                size += ((*lp)->index < 100) ? 3 : 11;
            }
        }
        path = ALLOCA_N(char, size);
        p = path;
        for (lp = doc->where_path; lp <= doc->where; lp++) {
            leaf = *lp;
            if (T_HASH == leaf->parent_type) {
                p = append_key(p, (*lp)->key);
            } else if (T_ARRAY == leaf->parent_type) {
                p = ulong_fill(p, (*lp)->index);
            }
            *p++ = '/';
        }
        *--p = '\0';
        return rb_str_new(path, p - path);
    }
}