Class: Rugged::Blob

Inherits:

RuggedObject

• Object
• RuggedObject
• Rugged::Blob

Defined in:

lib/rugged/blob.rb,
ext/rugged/rugged_blob.c

Defined Under Namespace

Classes: HashSignature

Class Method Summary

• .from_buffer(repository, buffer) ⇒ Object

  Write a blob to repository with the contents specified in buffer, where buffer is a String.

• .from_disk(repository, file_path) ⇒ Object

  Write the file specified in file_path to a blob in repository.

• .from_io(repository, io[, hint_path]) ⇒ Object

  Write a loose blob to the repository from an IO provider of data.

• .from_workdir(repository, file_path) ⇒ Object

  Write the file specified in file_path to a blob in repository.

• .merge_files(*args) ⇒ Object
• .to_buffer(*args) ⇒ Object

Instance Method Summary

• #binary? ⇒ Boolean

  Determine if the blob content is most certainly binary or not.

• #content(max_bytes = -1) ⇒ String

  Return up to max_bytes from the contents of a blob as bytes String.

• #diff(other, options = {}) ⇒ Object

  Directly generate a Rugged::Patch from the difference between blob and other.

• #hashsig(options = 0) ⇒ Object
• #loc ⇒ Integer

  Return the number of lines for this blob, assuming the blob is plaintext (i.e. not binary).

• #similarity(other) ⇒ Object
• #rawsize ⇒ Integer

  Return the size in bytes of the blob.

• #sloc ⇒ Integer

  Return the number of non-empty code lines for the blob, assuming the blob is plaintext (i.e. not binary).

• #text(max_lines = -1, encoding = Encoding.default_external) ⇒ String

  Return up to max_lines of text from a blob as a String.

Class Method Details

.from_buffer(repository, buffer) ⇒ Object

Write a blob to repository with the contents specified in buffer, where buffer is a String. The encoding of buffer is ignored and bytes are copied as-is.

# File 'ext/rugged/rugged_blob.c', line 137

static VALUE rb_git_blob_from_buffer(VALUE self, VALUE rb_repo, VALUE rb_buffer)
{
	int error;
	git_oid oid;
	git_repository *repo;

	Check_Type(rb_buffer, T_STRING);
	rugged_check_repo(rb_repo);

	Data_Get_Struct(rb_repo, git_repository, repo);

	error = git_blob_create_frombuffer(&oid, repo, RSTRING_PTR(rb_buffer), RSTRING_LEN(rb_buffer));
	rugged_exception_check(error);

	return rugged_create_oid(&oid);
}

.from_disk(repository, file_path) ⇒ Object

Write the file specified in file_path to a blob in repository. The repository can be bare or not.

Example:

Blob.from_disk(repo, '/var/repos/blob.h') #=> '5b5b025afb0b4c913b4c338a42934a3863bf3643'

# File 'ext/rugged/rugged_blob.c', line 192

static VALUE rb_git_blob_from_disk(VALUE self, VALUE rb_repo, VALUE rb_path)
{
	int error;
	git_oid oid;
	git_repository *repo;

	FilePathValue(rb_path);
	rugged_check_repo(rb_repo);

	Data_Get_Struct(rb_repo, git_repository, repo);

	error = git_blob_create_fromdisk(&oid, repo, StringValueCStr(rb_path));
	rugged_exception_check(error);

	return rugged_create_oid(&oid);
}

.from_io(repository, io[, hint_path]) ⇒ Object

Write a loose blob to the repository from an IO provider of data.

The repository can be bare or not.

The data provider io should respond to a read(size) method. Generally any instance of a class based on Ruby’s IO class should work(ex. File). On each read call it should return a String with maximum size of size.

NOTE: If an exception is raised in the io object’s read method, no blob will be created.

Provided the hint_path parameter is given, its value will help to determine what git filters should be applied to the object before it can be placed to the object database.

File.open('/path/to/file') do |file|
  Blob.from_io(repo, file, 'hint/blob.h') #=> '42cab3c0cde61e2b5a2392e1eadbeffa20ffa171'
end

# File 'ext/rugged/rugged_blob.c', line 244

static VALUE rb_git_blob_from_io(int argc, VALUE *argv, VALUE klass)
{
	VALUE rb_repo, rb_io, rb_hint_path, rb_buffer, rb_read_args[2];
	const char * hint_path = NULL;
	git_writestream *stream;
	int error = 0, exception = 0, max_length = 4096;
	git_oid oid;
	git_repository *repo;

	rb_scan_args(argc, argv, "21", &rb_repo, &rb_io, &rb_hint_path);

	rugged_check_repo(rb_repo);
	Data_Get_Struct(rb_repo, git_repository, repo);

	if (!NIL_P(rb_hint_path)) {
		FilePathValue(rb_hint_path);
		hint_path = StringValueCStr(rb_hint_path);
	}

	error = git_blob_create_fromstream(&stream, repo, hint_path);
	if (error)
		goto cleanup;

	rb_read_args[0] = rb_io;
	rb_read_args[1] = INT2FIX(max_length);

	do {
		rb_buffer = rb_protect(rb_read_check, (VALUE)rb_read_args, &exception);

		if (exception)
			goto cleanup;

		if (NIL_P(rb_buffer))
			break;

		error = stream->write(stream, RSTRING_PTR(rb_buffer), RSTRING_LEN(rb_buffer));
		if (error)
			goto cleanup;
	} while (RSTRING_LEN(rb_buffer) == max_length);

	error = git_blob_create_fromstream_commit(&oid, stream);

cleanup:

	if (exception)
		rb_jump_tag(exception);

	rugged_exception_check(error);

	return rugged_create_oid(&oid);
}

.from_workdir(repository, file_path) ⇒ Object

Write the file specified in file_path to a blob in repository. file_path must be relative to the repository’s working folder. The repository cannot be bare.

Blob.from_workdir(repo, 'src/blob.h') #=> '9d09060c850defbc7711d08b57def0d14e742f4e'

# File 'ext/rugged/rugged_blob.c', line 164

static VALUE rb_git_blob_from_workdir(VALUE self, VALUE rb_repo, VALUE rb_path)
{
	int error;
	git_oid oid;
	git_repository *repo;

	FilePathValue(rb_path);
	rugged_check_repo(rb_repo);

	Data_Get_Struct(rb_repo, git_repository, repo);

	error = git_blob_create_fromworkdir(&oid, repo, StringValueCStr(rb_path));
	rugged_exception_check(error);

	return rugged_create_oid(&oid);
}

.merge_files(*args) ⇒ Object

# File 'ext/rugged/rugged_blob.c', line 593

static VALUE rb_git_blob_merge_files(int argc, VALUE *argv, VALUE klass)
{
	VALUE rb_repo, rb_ancestor, rb_ours, rb_theirs, rb_options, rb_result = Qnil;

	git_repository *repo = NULL;
	rugged_merge_file_input ancestor = RUGGED_MERGE_FILE_INPUT_INIT,
		ours = RUGGED_MERGE_FILE_INPUT_INIT,
		theirs = RUGGED_MERGE_FILE_INPUT_INIT;
	git_blob *ancestor_blob = NULL, *our_blob = NULL, *their_blob = NULL;
	git_merge_file_options opts = GIT_MERGE_FILE_OPTIONS_INIT;
	git_merge_file_result result = {0};
	int error;

	rb_scan_args(argc, argv, "41", &rb_repo, &rb_ancestor, &rb_ours, &rb_theirs, &rb_options);

	if (!NIL_P(rb_repo)) {
		rugged_check_repo(rb_repo);
		Data_Get_Struct(rb_repo, git_repository, repo);
	}

	if (!NIL_P(rb_options))
		rugged_parse_merge_file_options(&opts, rb_options);

	if (!NIL_P(rb_ancestor))
		rugged_parse_merge_file_input(&ancestor, repo, rb_ancestor);
	if (!NIL_P(rb_ours))
		rugged_parse_merge_file_input(&ours, repo, rb_ours);
	if (!NIL_P(rb_theirs))
		rugged_parse_merge_file_input(&theirs, repo, rb_theirs);

	if ((error = rugged_load_merge_file_input(&ancestor_blob, repo, &ancestor)) < 0 ||
		(error = rugged_load_merge_file_input(&our_blob, repo, &ours)) < 0 ||
		(error = rugged_load_merge_file_input(&their_blob, repo, &theirs)) < 0 ||
		(error = git_merge_file(&result, &ancestor.parent, &ours.parent, &theirs.parent, &opts)) < 0)
		goto done;

	rb_result = rb_merge_file_result_fromC(&result);

done:
	git_blob_free(ancestor_blob);
	git_blob_free(our_blob);
	git_blob_free(their_blob);
	git_merge_file_result_free(&result);

	rugged_exception_check(error);
	return rb_result;
}

.to_buffer(*args) ⇒ Object

# File 'ext/rugged/rugged_blob.c', line 496

static VALUE rb_git_blob_to_buffer(int argc, VALUE *argv, VALUE self)
{
	VALUE rb_repo, rb_sha1, rb_max_bytes;
	VALUE rb_ret;

	git_repository *repo = NULL;
	git_blob *blob = NULL;

	size_t size;
	const char *content;

	rb_scan_args(argc, argv, "21", &rb_repo, &rb_sha1, &rb_max_bytes);

	rugged_check_repo(rb_repo);
	Data_Get_Struct(rb_repo, git_repository, repo);

	blob = (git_blob *)rugged_object_get(repo, rb_sha1, GIT_OBJ_BLOB);

	content = git_blob_rawcontent(blob);
	size = git_blob_rawsize(blob);

	if (!NIL_P(rb_max_bytes)) {
		int maxbytes;

		Check_Type(rb_max_bytes, T_FIXNUM);
		maxbytes = FIX2INT(rb_max_bytes);

		if (maxbytes >= 0 && (size_t)maxbytes < size)
			size = (size_t)maxbytes;
	}

	rb_ret = rb_ary_new();

	rb_ary_push(rb_ret, rb_str_new(content, size));
	rb_ary_push(rb_ret, INT2FIX(git_blob_rawsize(blob)));

	git_object_free((git_object*)blob);

	/* TODO: LOC */

	return rb_ret;
}

Instance Method Details

#binary? ⇒ Boolean

Determine if the blob content is most certainly binary or not.

The heuristic used to guess if a file is binary is taken from core git: Searching for NUL bytes and looking for a reasonable ratio of printable to non-printable characters among the first 4000 bytes.

Returns:

• (Boolean)

# File 'ext/rugged/rugged_blob.c', line 385

static VALUE rb_git_blob_is_binary(VALUE self)
{
	git_blob *blob;
	TypedData_Get_Struct(self, git_blob, &rugged_object_type, blob);
	return git_blob_is_binary(blob) ? Qtrue : Qfalse;
}

#content(max_bytes = -1) ⇒ String

Return up to max_bytes from the contents of a blob as bytes String. If max_bytes is less than 0, the full string is returned.

This string is tagged with the ASCII-8BIT encoding: the bytes are returned as-is, since Git is encoding agnostic.

Returns:

• (String)

# File 'ext/rugged/rugged_blob.c', line 83

static VALUE rb_git_blob_content_GET(int argc, VALUE *argv, VALUE self)
{
	git_blob *blob;
	size_t size;
	const char *content;
	VALUE rb_max_bytes;

	TypedData_Get_Struct(self, git_blob, &rugged_object_type, blob);
	rb_scan_args(argc, argv, "01", &rb_max_bytes);

	content = git_blob_rawcontent(blob);
	size = git_blob_rawsize(blob);

	if (!NIL_P(rb_max_bytes)) {
		int maxbytes;

		Check_Type(rb_max_bytes, T_FIXNUM);
		maxbytes = FIX2INT(rb_max_bytes);

		if (maxbytes >= 0 && (size_t)maxbytes < size)
			size = (size_t)maxbytes;
	}

	/*
	 * since we don't really ever know the encoding of a blob
	 * lets default to the binary encoding (ascii-8bit)
	 */
	return rb_str_new(content, size);
}

#diff(other, options = {}) ⇒ Object

Directly generate a Rugged::Patch from the difference between blob and other.

other can either be another Rugged::Blob instance, a string, or nil (treated as an empty blob).

The following options can be passed in the options Hash:

:max_size

An integer specifying the maximum byte size of a blob before a it will be treated as binary. The default value is 512MB.

:context_lines

The number of unchanged lines that define the boundary of a hunk (and to display before and after the actual changes). The default is 3.

:interhunk_lines

The maximum number of unchanged lines between hunk boundaries before the hunks will be merged into a one. The default is 0.

:reverse

If true, the sides of the diff will be reversed.

:force_text

If true, all files will be treated as text, disabling binary attributes & detection.

:ignore_whitespace

If true, all whitespace will be ignored.

:ignore_whitespace_change

If true, changes in amount of whitespace will be ignored.

:ignore_whitespace_eol

If true, whitespace at end of line will be ignored.

:patience

If true, the “patience diff” algorithm will be used (currently unimplemented).

:skip_binary_check

If true, diff deltas will be generated without spending time on binary detection. This is useful to improve performance in cases where the actual file content difference is not needed.

:old_path

An optional string to treat blob as if it had this filename.

:new_path

An optional string to treat other as if it had this filename.

# File 'ext/rugged/rugged_blob.c', line 444

static VALUE rb_git_blob_diff(int argc, VALUE *argv, VALUE self)
{
	git_blob *blob;
	git_diff_options opts = GIT_DIFF_OPTIONS_INIT;
	git_patch *patch;
	const char *old_path = NULL, *new_path = NULL;
	VALUE rb_other, rb_options;
	int error;

	rb_scan_args(argc, argv, "10:", &rb_other, &rb_options);
	if (!NIL_P(rb_options)) {
		VALUE rb_value;

		rb_value = rb_hash_aref(rb_options, CSTR2SYM("old_path"));
		if (!NIL_P(rb_value)) {
			Check_Type(rb_value, T_STRING);
			old_path = StringValueCStr(rb_value);
		}

		rb_value = rb_hash_aref(rb_options, CSTR2SYM("new_path"));
		if (!NIL_P(rb_value)) {
			Check_Type(rb_value, T_STRING);
			new_path = StringValueCStr(rb_value);
		}

		rugged_parse_diff_options(&opts, rb_options);
	}

	TypedData_Get_Struct(self, git_blob, &rugged_object_type, blob);

	if (NIL_P(rb_other)) {
		error = git_patch_from_blobs(&patch, blob, old_path, NULL, new_path, &opts);
	} else if (rb_obj_is_kind_of(rb_other, rb_cRuggedBlob)) {
		git_blob *other_blob;

		TypedData_Get_Struct(rb_other, git_blob, &rugged_object_type, other_blob);

		error = git_patch_from_blobs(&patch, blob, old_path, other_blob, new_path, &opts);
	} else if (TYPE(rb_other) == T_STRING) {
		const char * buffer = StringValueCStr(rb_other);

		error = git_patch_from_blob_and_buffer(&patch, blob, old_path, buffer, RSTRING_LEN(rb_other), new_path, &opts);
	} else {
		rb_raise(rb_eTypeError, "wrong argument type %s (expected Rugged::Blob, String, or nil)",
			rb_obj_classname(rb_other));
	}

	rugged_exception_check(error);

	return rugged_patch_new(self, patch);
}

#hashsig(options = 0) ⇒ Object

# File 'lib/rugged/blob.rb', line 14

def hashsig(options = 0)
  @hashsig ||= HashSignature.new(self, options)
end

#loc ⇒ Integer

Return the number of lines for this blob, assuming the blob is plaintext (i.e. not binary)

Returns:

• (Integer)

# File 'ext/rugged/rugged_blob.c', line 303

static VALUE rb_git_blob_loc(VALUE self)
{
	git_blob *blob;
	const char *data, *data_end;
	size_t loc = 0;

	TypedData_Get_Struct(self, git_blob, &rugged_object_type, blob);

	data = git_blob_rawcontent(blob);
	data_end = data + git_blob_rawsize(blob);

	if (data == data_end)
		return INT2FIX(0);

	for (; data < data_end; ++data) {
		if (data[0] == '\n') {
			loc++;
		}
		else if (data[0] == '\r') {
			if (data + 1 < data_end && data[1] == '\n')
				data++;
			loc++;
		}
	}

	if (data[-1] != '\n' && data[-1] != '\r')
		loc++;

	return INT2FIX(loc);
}

#similarity(other) ⇒ Object

# File 'lib/rugged/blob.rb', line 18

def similarity(other)
  other_sig = case other
  when HashSignature
    other
  when String
    HashSignature.new(other)
  when Blob
    other.hashsig
  else
    raise TypeError, "Expected a Rugged::Blob, String or Rugged::Blob::HashSignature"
  end

  HashSignature.compare(self.hashsig, other_sig)
end

#rawsize ⇒ Integer

Return the size in bytes of the blob. This is the real, uncompressed size and the length of blob.content, not the compressed size.

Returns:

• (Integer)

# File 'ext/rugged/rugged_blob.c', line 121

static VALUE rb_git_blob_rawsize(VALUE self)
{
	git_blob *blob;
	TypedData_Get_Struct(self, git_blob, &rugged_object_type, blob);

	return INT2FIX(git_blob_rawsize(blob));
}

#sloc ⇒ Integer

Return the number of non-empty code lines for the blob, assuming the blob is plaintext (i.e. not binary)

Returns:

• (Integer)

# File 'ext/rugged/rugged_blob.c', line 342

static VALUE rb_git_blob_sloc(VALUE self)
{
	git_blob *blob;
	const char *data, *data_end;
	size_t sloc = 0;

	TypedData_Get_Struct(self, git_blob, &rugged_object_type, blob);

	data = git_blob_rawcontent(blob);
	data_end = data + git_blob_rawsize(blob);

	if (data == data_end)
		return INT2FIX(0);

	/* go through the whole blob, counting lines
	 * that are not empty */
	while (data < data_end) {
		if (*data++ == '\n') {
			while (data < data_end && isspace(*data))
				data++;

			sloc++;
		}
	}

	/* last line without trailing '\n'? */
	if (data[-1] != '\n')
		sloc++;

	return INT2FIX(sloc);
}

#text(max_lines = -1, encoding = Encoding.default_external) ⇒ String

Return up to max_lines of text from a blob as a String. If max_lines is less than 0, the full string is returned.

The string is created with the given encoding, defaulting to Encoding.default_external.

When limiting the size of the text with max_lines, the string is expected to have an ASCII-compatible encoding, and is checked for the newline \n character.

Returns:

• (String)

# File 'ext/rugged/rugged_blob.c', line 36

static VALUE rb_git_blob_text_GET(int argc, VALUE *argv, VALUE self)
{
	git_blob *blob;
	size_t size;
	const char *content;
	VALUE rb_max_lines, rb_encoding;

	TypedData_Get_Struct(self, git_blob, &rugged_object_type, blob);
	rb_scan_args(argc, argv, "02", &rb_max_lines, &rb_encoding);

	content = git_blob_rawcontent(blob);
	size = git_blob_rawsize(blob);

	if (!NIL_P(rb_max_lines)) {
		size_t i = 0;
		int lines = 0, maxlines;

		Check_Type(rb_max_lines, T_FIXNUM);
		maxlines = FIX2INT(rb_max_lines);

		if (maxlines >= 0) {
			while (i < size && lines < maxlines) {
				if (content[i++] == '\n')
					lines++;
			}
			size = (size_t)i;
		}

	}

	if (!NIL_P(rb_encoding)) {
		return rb_enc_str_new(content, size, rb_to_encoding(rb_encoding));
	}

	return rb_external_str_new(content, size);
}