data.py

from __future__ import absolute_import

import abc
import logging
import mimetypes
import os
import shutil
import string
import tempfile
import zipfile
from cgi import escape
from inspect import isclass

import paste
import six

from galaxy import util
from galaxy.datatypes.metadata import MetadataElement  # import directly to maintain ease of use in Datatype class definitions
from galaxy.util import compression_utils
from galaxy.util import FILENAME_VALID_CHARS
from galaxy.util import inflector
from galaxy.util import unicodify
from galaxy.util.bunch import Bunch
from galaxy.util.odict import odict
from galaxy.util.sanitize_html import sanitize_html

from . import dataproviders
from . import metadata

XSS_VULNERABLE_MIME_TYPES = [
    'image/svg+xml',  # Unfiltered by Galaxy and may contain JS that would be executed by some browsers.
    'application/xml',  # Some browsers will evalute SVG embedded JS in such XML documents.
]
DEFAULT_MIME_TYPE = 'text/plain'  # Vulnerable mime types will be replaced with this.

log = logging.getLogger(__name__)

# Valid first column and strand column values vor bed, other formats
col1_startswith = ['chr', 'chl', 'groupun', 'reftig_', 'scaffold', 'super_', 'vcho']
valid_strand = ['+', '-', '.']

DOWNLOAD_FILENAME_PATTERN_DATASET = "Galaxy${hid}-[${name}].${ext}"
DOWNLOAD_FILENAME_PATTERN_COLLECTION_ELEMENT = "Galaxy${hdca_hid}-[${hdca_name}__${element_identifier}].${ext}"


class DataMeta( abc.ABCMeta ):
    """
    Metaclass for Data class.  Sets up metadata spec.
    """
    def __init__( cls, name, bases, dict_ ):
        cls.metadata_spec = metadata.MetadataSpecCollection()
        for base in bases:  # loop through bases (class/types) of cls
            if hasattr( base, "metadata_spec" ):  # base of class Data (object) has no metadata
                cls.metadata_spec.update( base.metadata_spec )  # add contents of metadata spec of base class to cls
        metadata.Statement.process( cls )


@six.add_metaclass(DataMeta)
@dataproviders.decorators.has_dataproviders
class Data( object ):
    """
    Base class for all datatypes.  Implements basic interfaces as well
    as class methods for metadata.

    >>> class DataTest( Data ):
    ...     MetadataElement( name="test" )
    ...
    >>> DataTest.metadata_spec.test.name
    'test'
    >>> DataTest.metadata_spec.test.desc
    'test'
    >>> type( DataTest.metadata_spec.test.param )
    <class 'galaxy.model.metadata.MetadataParameter'>
    """
    edam_data = "data_0006"
    edam_format = "format_1915"
    # Data is not chunkable by default.
    CHUNKABLE = False

    #: Dictionary of metadata fields for this datatype
    metadata_spec = None

    # Add metadata elements
    MetadataElement( name="dbkey", desc="Database/Build", default="?", param=metadata.DBKeyParameter, multiple=False, no_value="?" )
    # Stores the set of display applications, and viewing methods, supported by this datatype
    supported_display_apps = {}
    # If False, the peek is regenerated whenever a dataset of this type is copied
    copy_safe_peek = True
    # The dataset contains binary data --> do not space_to_tab or convert newlines, etc.
    # Allow binary file uploads of this type when True.
    is_binary = True
    # Allow user to change between this datatype and others. If False, this datatype
    # cannot be changed from or into.
    allow_datatype_change = True
    # Composite datatypes
    composite_type = None
    composite_files = odict()
    primary_file_name = 'index'
    # A per datatype setting (inherited): max file size (in bytes) for setting optional metadata
    _max_optional_metadata_filesize = None

    # Trackster track type.
    track_type = None

    # Data sources.
    data_sources = {}

    def __init__(self, **kwd):
        """Initialize the datatype"""
        object.__init__(self, **kwd)
        self.supported_display_apps = self.supported_display_apps.copy()
        self.composite_files = self.composite_files.copy()
        self.display_applications = odict()

    def write_from_stream(self, dataset, stream):
        """Writes data from a stream"""
        fd = open(dataset.file_name, 'wb')
        while True:
            chunk = stream.read(1048576)
            if not chunk:
                break
            os.write(fd, chunk)
        os.close(fd)

    def set_raw_data(self, dataset, data):
        """Saves the data on the disc"""
        fd = open(dataset.file_name, 'wb')
        os.write(fd, data)
        os.close(fd)

    def get_raw_data( self, dataset ):
        """Returns the full data. To stream it open the file_name and read/write as needed"""
        try:
            return open(dataset.file_name, 'rb').read(-1)
        except OSError:
            log.exception('%s reading a file that does not exist %s', self.__class__.__name__, dataset.file_name)
            return ''

    def dataset_content_needs_grooming( self, file_name ):
        """This function is called on an output dataset file after the content is initially generated."""
        return False

    def groom_dataset_content( self, file_name ):
        """This function is called on an output dataset file if dataset_content_needs_grooming returns True."""
        pass

    def init_meta( self, dataset, copy_from=None ):
        # Metadata should be left mostly uninitialized.  Dataset will
        # handle returning default values when metadata is not set.
        # copy_from allows metadata to be passed in that will be
        # copied. (although this seems ambiguous, see
        # Dataset.set_metadata.  It always copies the rhs in order to
        # flag the object as modified for SQLAlchemy.
        if copy_from:
            dataset.metadata = copy_from.metadata

    def set_meta( self, dataset, overwrite=True, **kwd ):
        """Unimplemented method, allows guessing of metadata from contents of file"""
        return True

    def missing_meta( self, dataset, check=[], skip=[] ):
        """
        Checks for empty metadata values, Returns True if non-optional metadata is missing
        Specifying a list of 'check' values will only check those names provided; when used, optionality is ignored
        Specifying a list of 'skip' items will return True even when a named metadata value is missing
        """
        if check:
            to_check = ( ( to_check, dataset.metadata.get( to_check ) ) for to_check in check )
        else:
            to_check = dataset.metadata.items()
        for key, value in to_check:
            if key in skip or ( not check and dataset.metadata.spec[key].get( "optional" ) ):
                continue  # we skip check for optional and nonrequested values here
            if not value:
                return True
        return False

    def set_max_optional_metadata_filesize( self, max_value ):
        try:
            max_value = int( max_value )
        except:
            return
        self.__class__._max_optional_metadata_filesize = max_value

    def get_max_optional_metadata_filesize( self ):
        rval = self.__class__._max_optional_metadata_filesize
        if rval is None:
            return -1
        return rval

    max_optional_metadata_filesize = property( get_max_optional_metadata_filesize, set_max_optional_metadata_filesize )

    def set_peek( self, dataset, is_multi_byte=False ):
        """Set the peek and blurb text"""
        if not dataset.dataset.purged:
            dataset.peek = ''
            dataset.blurb = 'data'
        else:
            dataset.peek = 'file does not exist'
            dataset.blurb = 'file purged from disk'

    def display_peek(self, dataset ):
        """Create HTML table, used for displaying peek"""
        out = ['<table cellspacing="0" cellpadding="3">']
        try:
            if not dataset.peek:
                dataset.set_peek()
            data = dataset.peek
            lines = data.splitlines()
            for line in lines:
                line = line.strip()
                if not line:
                    continue
                out.append( '<tr><td>%s</td></tr>' % escape( unicodify( line, 'utf-8' ) ) )
            out.append( '</table>' )
            out = "".join( out )
        except Exception as exc:
            out = "Can't create peek %s" % str( exc )
        return out

    def _archive_main_file(self, archive, display_name, data_filename):
        """Called from _archive_composite_dataset to add central file to archive.

        Unless subclassed, this will add the main dataset file (argument data_filename)
        to the archive, as an HTML file with its filename derived from the dataset name
        (argument outfname).

        Returns a tuple of boolean, string, string: (error, msg, messagetype)
        """
        error, msg, messagetype = False, "", ""
        archname = '%s.html' % display_name  # fake the real nature of the html file
        try:
            archive.add(data_filename, archname)
        except IOError:
            error = True
            log.exception("Unable to add composite parent %s to temporary library download archive", data_filename)
            msg = "Unable to create archive for download, please report this error"
            messagetype = "error"
        return error, msg, messagetype

    def _archive_composite_dataset( self, trans, data=None, **kwd ):
        # save a composite object into a compressed archive for downloading
        params = util.Params( kwd )
        outfname = data.name[0:150]
        outfname = ''.join(c in FILENAME_VALID_CHARS and c or '_' for c in outfname)
        if params.do_action is None:
            params.do_action = 'zip'  # default
        msg = util.restore_text( params.get( 'msg', ''  ) )
        if not data:
            msg = "You must select at least one dataset"
        else:
            error = False
            try:
                if params.do_action == 'zip':
                    # Can't use mkstemp - the file must not exist first
                    tmpd = tempfile.mkdtemp()
                    util.umask_fix_perms( tmpd, trans.app.config.umask, 0o777, trans.app.config.gid )
                    tmpf = os.path.join( tmpd, 'library_download.' + params.do_action )
                    archive = zipfile.ZipFile( tmpf, 'w', zipfile.ZIP_DEFLATED, True )
                    archive.add = lambda x, y: archive.write( x, y.encode('CP437') )
                elif params.do_action == 'tgz':
                    archive = util.streamball.StreamBall( 'w|gz' )
                elif params.do_action == 'tbz':
                    archive = util.streamball.StreamBall( 'w|bz2' )
            except (OSError, zipfile.BadZipFile):
                error = True
                log.exception( "Unable to create archive for download" )
                msg = "Unable to create archive for %s for download, please report this error" % outfname
            if not error:
                ext = data.extension
                path = data.file_name
                efp = data.extra_files_path
                # Add any central file to the archive,

                display_name = os.path.splitext(outfname)[0]
                if not display_name.endswith(ext):
                    display_name = '%s_%s' % (display_name, ext)

                error, msg = self._archive_main_file(archive, display_name, path)[:2]
                if not error:
                    # Add any child files to the archive,
                    for fpath, rpath in self.__archive_extra_files_path(extra_files_path=efp):
                        try:
                            archive.add(fpath, rpath)
                        except IOError:
                            error = True
                            log.exception("Unable to add %s to temporary library download archive", rpath)
                            msg = "Unable to create archive for download, please report this error"
                            continue
                if not error:
                    if params.do_action == 'zip':
                        archive.close()
                        tmpfh = open( tmpf )
                        # CANNOT clean up - unlink/rmdir was always failing because file handle retained to return - must rely on a cron job to clean up tmp
                        trans.response.set_content_type( "application/x-zip-compressed" )
                        trans.response.headers[ "Content-Disposition" ] = 'attachment; filename="%s.zip"' % outfname
                        return tmpfh
                    else:
                        trans.response.set_content_type( "application/x-tar" )
                        outext = 'tgz'
                        if params.do_action == 'tbz':
                            outext = 'tbz'
                        trans.response.headers[ "Content-Disposition" ] = 'attachment; filename="%s.%s"' % (outfname, outext)
                        archive.wsgi_status = trans.response.wsgi_status()
                        archive.wsgi_headeritems = trans.response.wsgi_headeritems()
                        return archive.stream
        return trans.show_error_message( msg )

    def __archive_extra_files_path(self, extra_files_path):
        """Yield filepaths and relative filepaths for files in extra_files_path"""
        for root, dirs, files in os.walk(extra_files_path):
            for fname in files:
                fpath = os.path.join(root, fname)
                rpath = os.path.relpath(fpath, extra_files_path)
                yield fpath, rpath

    def _serve_raw(self, trans, dataset, to_ext, **kwd):
        trans.response.headers['Content-Length'] = int( os.stat( dataset.file_name ).st_size )
        trans.response.set_content_type( "application/octet-stream" )  # force octet-stream so Safari doesn't append mime extensions to filename
        filename = self._download_filename(dataset, to_ext, hdca=kwd.get("hdca", None), element_identifier=kwd.get("element_identifier", None))
        trans.response.headers["Content-Disposition"] = 'attachment; filename="%s"' % filename
        return open( dataset.file_name )

    def to_archive(self, trans, dataset, name=""):
        """
        Collect archive paths and file handles that need to be exported when archiving `dataset`.

        :param dataset: HistoryDatasetAssociation
        :param name: archive name, in collection context corresponds to collection name(s) and element_identifier,
                     joined by '/', e.g 'fastq_collection/sample1/forward'
        """
        composite_extensions = trans.app.datatypes_registry.get_composite_extensions( )
        composite_extensions.append('html')  # for archiving composite datatypes
        rel_paths = []
        file_paths = []
        if dataset.extension in composite_extensions:
            main_file = "%s.%s" % (name, 'html')
            rel_paths.append(main_file)
            file_paths.append(dataset.file_name)
            for fpath, rpath in self.__archive_extra_files_path(dataset.extra_files_path):
                rel_paths.append(os.path.join(name, rpath))
                file_paths.append(fpath)
        else:
            rel_paths.append("%s.%s" % (name or dataset.file_name, dataset.extension))
            file_paths.append(dataset.file_name)
        return zip(file_paths, rel_paths)

    def display_data(self, trans, data, preview=False, filename=None, to_ext=None, **kwd):
        """ Old display method, for transition - though still used by API and
        test framework. Datatypes should be very careful if overridding this
        method and this interface between datatypes and Galaxy will likely
        change.

        TOOD: Document alternatives to overridding this method (data
        providers?).
        """
        # Relocate all composite datatype display to a common location.
        composite_extensions = trans.app.datatypes_registry.get_composite_extensions( )
        composite_extensions.append('html')  # for archiving composite datatypes
        # Prevent IE8 from sniffing content type since we're explicit about it.  This prevents intentionally text/plain
        # content from being rendered in the browser
        trans.response.headers['X-Content-Type-Options'] = 'nosniff'
        if isinstance( data, six.string_types ):
            return data
        if filename and filename != "index":
            # For files in extra_files_path
            file_path = trans.app.object_store.get_filename(data.dataset, extra_dir='dataset_%s_files' % data.dataset.id, alt_name=filename)
            if os.path.exists( file_path ):
                if os.path.isdir( file_path ):
                    tmp_fh = tempfile.NamedTemporaryFile(delete=False)
                    tmp_file_name = tmp_fh.name
                    dir_items = sorted(os.listdir(file_path))
                    base_path, item_name = os.path.split(file_path)
                    tmp_fh.write('<html><head><h3>Directory %s contents: %d items</h3></head>\n' % (item_name, len(dir_items)))
                    tmp_fh.write('<body><p/><table cellpadding="2">\n')
                    for index, fname in enumerate(dir_items):
                        if index % 2 == 0:
                            bgcolor = '#D8D8D8'
                        else:
                            bgcolor = '#FFFFFF'
                        # Can't have an href link here because there is no route
                        # defined for files contained within multiple subdirectory
                        # levels of the primary dataset.  Something like this is
                        # close, but not quite correct:
                        # href = url_for(controller='dataset', action='display',
                        # dataset_id=trans.security.encode_id(data.dataset.id),
                        # preview=preview, filename=fname, to_ext=to_ext)
                        tmp_fh.write('<tr bgcolor="%s"><td>%s</td></tr>\n' % (bgcolor, fname))
                    tmp_fh.write('</table></body></html>\n')
                    tmp_fh.close()
                    return open(tmp_file_name)
                mime = mimetypes.guess_type( file_path )[0]
                if not mime:
                    try:
                        mime = trans.app.datatypes_registry.get_mimetype_by_extension( ".".split( file_path )[-1] )
                    except:
                        mime = "text/plain"
                self._clean_and_set_mime_type( trans, mime )
                return open( file_path )
            else:
                return paste.httpexceptions.HTTPNotFound( "Could not find '%s' on the extra files path %s." % ( filename, file_path ) )
        self._clean_and_set_mime_type( trans, data.get_mime() )

        trans.log_event( "Display dataset id: %s" % str( data.id ) )
        from galaxy import datatypes  # DBTODO REMOVE THIS AT REFACTOR
        if to_ext or isinstance(data.datatype, datatypes.binary.Binary):  # Saving the file, or binary file
            if data.extension in composite_extensions:
                return self._archive_composite_dataset( trans, data, **kwd )
            else:
                trans.response.headers['Content-Length'] = int( os.stat( data.file_name ).st_size )
                filename = self._download_filename(data, to_ext, hdca=kwd.get("hdca", None), element_identifier=kwd.get("element_identifier", None))
                trans.response.set_content_type( "application/octet-stream" )  # force octet-stream so Safari doesn't append mime extensions to filename
                trans.response.headers["Content-Disposition"] = 'attachment; filename="%s"' % filename
                return open( data.file_name )
        if not os.path.exists( data.file_name ):
            raise paste.httpexceptions.HTTPNotFound( "File Not Found (%s)." % data.file_name )
        max_peek_size = 1000000  # 1 MB
        if isinstance(data.datatype, datatypes.text.Html):
            max_peek_size = 10000000  # 10 MB for html
        preview = util.string_as_bool( preview )
        if not preview or isinstance(data.datatype, datatypes.images.Image) or os.stat( data.file_name ).st_size < max_peek_size:
            if trans.app.config.sanitize_all_html and trans.response.get_content_type() == "text/html":
                # Sanitize anytime we respond with plain text/html content.
                # Check to see if this dataset's parent job is whitelisted
                # We cannot currently trust imported datasets for rendering.
                if not data.creating_job.imported and data.creating_job.tool_id in trans.app.config.sanitize_whitelist:
                    return open(data.file_name).read()
                # This is returning to the browser, it needs to be encoded.
                # TODO Ideally this happens a layer higher, but this is a bad
                # issue affecting many tools
                return sanitize_html(open( data.file_name ).read()).encode('utf-8')
            return open( data.file_name )
        else:
            trans.response.set_content_type( "text/html" )
            return trans.stream_template_mako( "/dataset/large_file.mako",
                                               truncated_data=open( data.file_name ).read(max_peek_size),
                                               data=data)

    def _download_filename(self, dataset, to_ext, hdca=None, element_identifier=None):
        def escape(raw_identifier):
            return ''.join(c in FILENAME_VALID_CHARS and c or '_' for c in raw_identifier)[0:150]

        if not to_ext:
            to_ext = dataset.extension

        template_values = {
            "name": escape(dataset.name),
            "ext": to_ext,
            "hid": dataset.hid,
        }

        filename_pattern = DOWNLOAD_FILENAME_PATTERN_DATASET

        if hdca is not None:
            # Use collection context to build up filename.
            template_values["element_identifier"] = element_identifier
            template_values["hdca_name"] = escape(hdca.name)
            template_values["hdca_hid"] = hdca.hid

            filename_pattern = DOWNLOAD_FILENAME_PATTERN_COLLECTION_ELEMENT

        return string.Template(filename_pattern).substitute(**template_values)

    def display_name(self, dataset):
        """Returns formatted html of dataset name"""
        try:
            return escape( unicodify( dataset.name, 'utf-8' ) )
        except Exception:
            return "name unavailable"

    def display_info(self, dataset):
        """Returns formatted html of dataset info"""
        try:
            # Change new line chars to html
            info = escape( dataset.info )
            if info.find( '\r\n' ) >= 0:
                info = info.replace( '\r\n', '<br/>' )
            if info.find( '\r' ) >= 0:
                info = info.replace( '\r', '<br/>' )
            if info.find( '\n' ) >= 0:
                info = info.replace( '\n', '<br/>' )

            info = unicodify( info, 'utf-8' )

            return info
        except:
            return "info unavailable"

    def validate(self, dataset):
        """Unimplemented validate, return no exceptions"""
        return list()

    def repair_methods(self, dataset):
        """Unimplemented method, returns dict with method/option for repairing errors"""
        return None

    def get_mime(self):
        """Returns the mime type of the datatype"""
        return 'application/octet-stream'

    def add_display_app( self, app_id, label, file_function, links_function ):
        """
        Adds a display app to the datatype.
        app_id is a unique id
        label is the primary display label, e.g., display at 'UCSC'
        file_function is a string containing the name of the function that returns a properly formatted display
        links_function is a string containing the name of the function that returns a list of (link_name,link)
        """
        self.supported_display_apps = self.supported_display_apps.copy()
        self.supported_display_apps[app_id] = {'label': label, 'file_function': file_function, 'links_function': links_function}

    def remove_display_app(self, app_id):
        """Removes a display app from the datatype"""
        self.supported_display_apps = self.supported_display_apps.copy()
        try:
            del self.supported_display_apps[app_id]
        except:
            log.exception('Tried to remove display app %s from datatype %s, but this display app is not declared.', type, self.__class__.__name__ )

    def clear_display_apps( self ):
        self.supported_display_apps = {}

    def add_display_application( self, display_application ):
        """New style display applications"""
        assert display_application.id not in self.display_applications, 'Attempted to add a display application twice'
        self.display_applications[ display_application.id ] = display_application

    def get_display_application( self, key, default=None ):
        return self.display_applications.get( key, default )

    def get_display_applications_by_dataset( self, dataset, trans ):
        rval = odict()
        for key, value in self.display_applications.items():
            value = value.filter_by_dataset( dataset, trans )
            if value.links:
                rval[key] = value
        return rval

    def get_display_types(self):
        """Returns display types available"""
        return list(self.supported_display_apps.keys())

    def get_display_label(self, type):
        """Returns primary label for display app"""
        try:
            return self.supported_display_apps[type]['label']
        except:
            return 'unknown'

    def as_display_type(self, dataset, type, **kwd):
        """Returns modified file contents for a particular display type """
        try:
            if type in self.get_display_types():
                return getattr(self, self.supported_display_apps[type]['file_function'])(dataset, **kwd)
        except:
            log.exception('Function %s is referred to in datatype %s for displaying as type %s, but is not accessible', self.supported_display_apps[type]['file_function'], self.__class__.__name__, type )
        return "This display type (%s) is not implemented for this datatype (%s)." % ( type, dataset.ext)

    def get_display_links( self, dataset, type, app, base_url, target_frame='_blank', **kwd ):
        """
        Returns a list of tuples of (name, link) for a particular display type.  No check on
        'access' permissions is done here - if you can view the dataset, you can also save it
        or send it to a destination outside of Galaxy, so Galaxy security restrictions do not
        apply anyway.
        """
        try:
            if app.config.enable_old_display_applications and type in self.get_display_types():
                return target_frame, getattr( self, self.supported_display_apps[type]['links_function'] )( dataset, type, app, base_url, **kwd )
        except:
            log.exception( 'Function %s is referred to in datatype %s for generating links for type %s, but is not accessible',
                           self.supported_display_apps[type]['links_function'], self.__class__.__name__, type )
        return target_frame, []

    def get_converter_types(self, original_dataset, datatypes_registry):
        """Returns available converters by type for this dataset"""
        return datatypes_registry.get_converters_by_datatype(original_dataset.ext)

    def find_conversion_destination( self, dataset, accepted_formats, datatypes_registry, **kwd ):
        """Returns ( target_ext, existing converted dataset )"""
        return datatypes_registry.find_conversion_destination_for_dataset_by_extensions( dataset, accepted_formats, **kwd )

    def convert_dataset(self, trans, original_dataset, target_type, return_output=False, visible=True, deps=None, target_context=None, history=None):
        """This function adds a job to the queue to convert a dataset to another type. Returns a message about success/failure."""
        converter = trans.app.datatypes_registry.get_converter_by_target_type( original_dataset.ext, target_type )

        if converter is None:
            raise Exception( "A converter does not exist for %s to %s." % ( original_dataset.ext, target_type ) )
        # Generate parameter dictionary
        params = {}
        # determine input parameter name and add to params
        input_name = 'input1'
        for key, value in converter.inputs.items():
            if deps and value.name in deps:
                params[value.name] = deps[value.name]
            elif value.type == 'data':
                input_name = key
        # add potentially required/common internal tool parameters e.g. '__job_resource'
        if target_context:
            for key, value in target_context.items():
                if key.startswith( '__' ):
                    params[ key ] = value
        params[input_name] = original_dataset

        # Run converter, job is dispatched through Queue
        converted_dataset = converter.execute( trans, incoming=params, set_output_hid=visible, history=history )[1]
        if len(params) > 0:
            trans.log_event( "Converter params: %s" % (str(params)), tool_id=converter.id )
        if not visible:
            for value in converted_dataset.values():
                value.visible = False
        if return_output:
            return converted_dataset
        return "The file conversion of %s on data %s has been added to the Queue." % (converter.name, original_dataset.hid)

    # We need to clear associated files before we set metadata
    # so that as soon as metadata starts to be set, e.g. implicitly converted datasets are deleted and no longer available 'while' metadata is being set, not just after
    # We'll also clear after setting metadata, for backwards compatibility
    def after_setting_metadata( self, dataset ):
        """This function is called on the dataset after metadata is set."""
        dataset.clear_associated_files( metadata_safe=True )

    def before_setting_metadata( self, dataset ):
        """This function is called on the dataset before metadata is set."""
        dataset.clear_associated_files( metadata_safe=True )

    def __new_composite_file( self, name, optional=False, mimetype=None, description=None, substitute_name_with_metadata=None, is_binary=False, to_posix_lines=True, space_to_tab=False, **kwds ):
        kwds[ 'name' ] = name
        kwds[ 'optional' ] = optional
        kwds[ 'mimetype' ] = mimetype
        kwds[ 'description' ] = description
        kwds[ 'substitute_name_with_metadata' ] = substitute_name_with_metadata
        kwds[ 'is_binary' ] = is_binary
        kwds[ 'to_posix_lines' ] = to_posix_lines
        kwds[ 'space_to_tab' ] = space_to_tab
        return Bunch( **kwds )

    def add_composite_file( self, name, **kwds ):
        # self.composite_files = self.composite_files.copy()
        self.composite_files[ name ] = self.__new_composite_file( name, **kwds )

    def __substitute_composite_key( self, key, composite_file, dataset=None ):
        if composite_file.substitute_name_with_metadata:
            if dataset:
                meta_value = str( dataset.metadata.get( composite_file.substitute_name_with_metadata ) )
            else:
                meta_value = self.spec[composite_file.substitute_name_with_metadata].default
            return key % meta_value
        return key

    @property
    def writable_files( self, dataset=None ):
        files = odict()
        if self.composite_type != 'auto_primary_file':
            files[ self.primary_file_name ] = self.__new_composite_file( self.primary_file_name )
        for key, value in self.get_composite_files( dataset=dataset ).items():
            files[ key ] = value
        return files

    def get_composite_files( self, dataset=None ):
        def substitute_composite_key( key, composite_file ):
            if composite_file.substitute_name_with_metadata:
                if dataset:
                    meta_value = str( dataset.metadata.get( composite_file.substitute_name_with_metadata ) )
                else:
                    meta_value = self.metadata_spec[ composite_file.substitute_name_with_metadata ].default
                return key % meta_value
            return key
        files = odict()
        for key, value in self.composite_files.items():
            files[ substitute_composite_key( key, value ) ] = value
        return files

    def generate_primary_file( self, dataset=None ):
        raise Exception( "generate_primary_file is not implemented for this datatype." )

    @property
    def has_resolution(self):
        return False

    def matches_any( self, target_datatypes ):
        """
        Check if this datatype is of any of the target_datatypes or is
        a subtype thereof.
        """
        datatype_classes = tuple( [ datatype if isclass( datatype ) else datatype.__class__ for datatype in target_datatypes ] )
        return isinstance( self, datatype_classes )

    def merge( split_files, output_file):
        """
            Merge files with copy.copyfileobj() will not hit the
            max argument limitation of cat. gz and bz2 files are also working.
        """
        if not split_files:
            raise ValueError('Asked to merge zero files as %s' % output_file)
        elif len(split_files) == 1:
            shutil.copyfileobj(open(split_files[0], 'rb'), open(output_file, 'wb'))
        else:
            fdst = open(output_file, 'wb')
            for fsrc in split_files:
                shutil.copyfileobj(open(fsrc, 'rb'), fdst)
            fdst.close()

    merge = staticmethod(merge)

    def get_visualizations( self, dataset ):
        """
        Returns a list of visualizations for datatype.
        """

        if self.track_type:
            return [ 'trackster', 'circster' ]
        return []

    # ------------- Dataproviders
    def has_dataprovider( self, data_format ):
        """
        Returns True if `data_format` is available in `dataproviders`.
        """
        return data_format in self.dataproviders

    def dataprovider( self, dataset, data_format, **settings ):
        """
        Base dataprovider factory for all datatypes that returns the proper provider
        for the given `data_format` or raises a `NoProviderAvailable`.
        """
        if self.has_dataprovider( data_format ):
            return self.dataproviders[ data_format ]( self, dataset, **settings )
        raise dataproviders.exceptions.NoProviderAvailable( self, data_format )

    @dataproviders.decorators.dataprovider_factory( 'base' )
    def base_dataprovider( self, dataset, **settings ):
        dataset_source = dataproviders.dataset.DatasetDataProvider( dataset )
        return dataproviders.base.DataProvider( dataset_source, **settings )

    @dataproviders.decorators.dataprovider_factory( 'chunk', dataproviders.chunk.ChunkDataProvider.settings )
    def chunk_dataprovider( self, dataset, **settings ):
        dataset_source = dataproviders.dataset.DatasetDataProvider( dataset )
        return dataproviders.chunk.ChunkDataProvider( dataset_source, **settings )

    @dataproviders.decorators.dataprovider_factory( 'chunk64', dataproviders.chunk.Base64ChunkDataProvider.settings )
    def chunk64_dataprovider( self, dataset, **settings ):
        dataset_source = dataproviders.dataset.DatasetDataProvider( dataset )
        return dataproviders.chunk.Base64ChunkDataProvider( dataset_source, **settings )

    def _clean_and_set_mime_type(self, trans, mime):
        if mime.lower() in XSS_VULNERABLE_MIME_TYPES:
            if not getattr( trans.app.config, "serve_xss_vulnerable_mimetypes", True ):
                mime = DEFAULT_MIME_TYPE
        trans.response.set_content_type( mime )


@dataproviders.decorators.has_dataproviders
class Text( Data ):
    edam_format = "format_2330"
    file_ext = 'txt'
    line_class = 'line'

    # Add metadata elements
    MetadataElement( name="data_lines", default=0, desc="Number of data lines", readonly=True, optional=True, visible=False, no_value=0 )

    def write_from_stream(self, dataset, stream):
        """Writes data from a stream"""
        # write it twice for now
        fd, temp_name = tempfile.mkstemp()
        while True:
            chunk = stream.read(1048576)
            if not chunk:
                break
            os.write(fd, chunk)
        os.close(fd)
        # rewrite the file with unix newlines
        fp = open(dataset.file_name, 'w')
        for line in open(temp_name, "U"):
            line = line.strip() + '\n'
            fp.write(line)
        fp.close()

    def set_raw_data(self, dataset, data):
        """Saves the data on the disc"""
        fd, temp_name = tempfile.mkstemp()
        os.write(fd, data)
        os.close(fd)
        # rewrite the file with unix newlines
        fp = open(dataset.file_name, 'w')
        for line in open(temp_name, "U"):
            line = line.strip() + '\n'
            fp.write(line)
        fp.close()
        os.remove( temp_name )

    def get_mime(self):
        """Returns the mime type of the datatype"""
        return 'text/plain'

    def set_meta( self, dataset, **kwd ):
        """
        Set the number of lines of data in dataset.
        """
        dataset.metadata.data_lines = self.count_data_lines(dataset)

    def estimate_file_lines( self, dataset ):
        """
        Perform a rough estimate by extrapolating number of lines from a small read.
        """
        sample_size = 1048576
        dataset_fh = open( dataset.file_name )
        dataset_read = dataset_fh.read(sample_size)
        dataset_fh.close()
        sample_lines = dataset_read.count('\n')
        est_lines = int(sample_lines * (float(dataset.get_size()) / float(sample_size)))
        return est_lines

    def count_data_lines(self, dataset):
        """
        Count the number of lines of data in dataset,
        skipping all blank lines and comments.
        """
        data_lines = 0
        for line in open( dataset.file_name ):
            line = line.strip()
            if line and not line.startswith( '#' ):
                data_lines += 1
        return data_lines

    def set_peek( self, dataset, line_count=None, is_multi_byte=False, WIDTH=256, skipchars=None, line_wrap=True ):
        """
        Set the peek.  This method is used by various subclasses of Text.
        """
        if not dataset.dataset.purged:
            # The file must exist on disk for the get_file_peek() method
            dataset.peek = get_file_peek( dataset.file_name, is_multi_byte=is_multi_byte, WIDTH=WIDTH, skipchars=skipchars, line_wrap=line_wrap )
            if line_count is None:
                # See if line_count is stored in the metadata
                if dataset.metadata.data_lines:
                    dataset.blurb = "%s %s" % ( util.commaify( str(dataset.metadata.data_lines) ), inflector.cond_plural(dataset.metadata.data_lines, self.line_class) )
                else:
                    # Number of lines is not known ( this should not happen ), and auto-detect is
                    # needed to set metadata
                    # This can happen when the file is larger than max_optional_metadata_filesize.
                    if int(dataset.get_size()) <= 1048576:
                        # Small dataset, recount all lines and reset peek afterward.
                        lc = self.count_data_lines(dataset)
                        dataset.metadata.data_lines = lc
                        dataset.blurb = "%s %s" % ( util.commaify( str(lc) ), inflector.cond_plural(lc, self.line_class) )
                    else:
                        est_lines = self.estimate_file_lines(dataset)
                        dataset.blurb = "~%s %s" % ( util.commaify(util.roundify(str(est_lines))), inflector.cond_plural(est_lines, self.line_class) )
            else:
                dataset.blurb = "%s %s" % ( util.commaify( str(line_count) ), inflector.cond_plural(line_count, self.line_class) )
        else:
            dataset.peek = 'file does not exist'
            dataset.blurb = 'file purged from disk'

    def split( cls, input_datasets, subdir_generator_function, split_params):
        """
        Split the input files by line.
        """
        if split_params is None:
            return

        if len(input_datasets) > 1:
            raise Exception("Text file splitting does not support multiple files")
        input_files = [ds.file_name for ds in input_datasets]

        lines_per_file = None
        chunk_size = None
        if split_params['split_mode'] == 'number_of_parts':
            lines_per_file = []

            # Computing the length is expensive!
            def _file_len(fname):
                i = 0
                f = open(fname)
                for i, _ in enumerate(f):
                    pass
                f.close()
                return i + 1
            length = _file_len(input_files[0])
            parts = int(split_params['split_size'])
            if length < parts:
                parts = length
            len_each, remainder = divmod(length, parts)
            while length > 0:
                chunk = len_each
                if remainder > 0:
                    chunk += 1
                lines_per_file.append(chunk)
                remainder -= 1
                length -= chunk
        elif split_params['split_mode'] == 'to_size':
            chunk_size = int(split_params['split_size'])
        else:
            raise Exception('Unsupported split mode %s' % split_params['split_mode'])

        f = open(input_files[0], 'r')
        try:
            chunk_idx = 0
            file_done = False
            part_file = None
            while not file_done:
                if lines_per_file is None:
                    this_chunk_size = chunk_size
                elif chunk_idx < len(lines_per_file):
                    this_chunk_size = lines_per_file[chunk_idx]
                    chunk_idx += 1
                lines_remaining = this_chunk_size
                part_file = None
                while lines_remaining > 0:
                    a_line = f.readline()
                    if a_line == '':
                        file_done = True
                        break
                    if part_file is None:
                        part_dir = subdir_generator_function()
                        part_path = os.path.join(part_dir, os.path.basename(input_files[0]))
                        part_file = open(part_path, 'w')
                    part_file.write(a_line)
                    lines_remaining -= 1
                if part_file is not None:
                    part_file.close()
        except Exception as e:
            log.error('Unable to split files: %s' % str(e))
            f.close()
            if part_file is not None:
                part_file.close()
            raise
        f.close()
    split = classmethod(split)

    # ------------- Dataproviders
    @dataproviders.decorators.dataprovider_factory( 'line', dataproviders.line.FilteredLineDataProvider.settings )
    def line_dataprovider( self, dataset, **settings ):
        """
        Returns an iterator over the dataset's lines (that have been stripped)
        optionally excluding blank lines and lines that start with a comment character.
        """
        dataset_source = dataproviders.dataset.DatasetDataProvider( dataset )
        return dataproviders.line.FilteredLineDataProvider( dataset_source, **settings )

    @dataproviders.decorators.dataprovider_factory( 'regex-line', dataproviders.line.RegexLineDataProvider.settings )
    def regex_line_dataprovider( self, dataset, **settings ):
        """
        Returns an iterator over the dataset's lines
        optionally including/excluding lines that match one or more regex filters.
        """
        dataset_source = dataproviders.dataset.DatasetDataProvider( dataset )
        return dataproviders.line.RegexLineDataProvider( dataset_source, **settings )


class GenericAsn1( Text ):
    """Class for generic ASN.1 text format"""
    edam_data = "data_0849"
    edam_format = "format_1966"
    file_ext = 'asn1'


class LineCount( Text ):
    """
    Dataset contains a single line with a single integer that denotes the
    line count for a related dataset. Used for custom builds.
    """
    pass


class Newick( Text ):
    """New Hampshire/Newick Format"""
    edam_data = "data_0872"
    edam_format = "format_1910"
    file_ext = "nhx"

    def __init__(self, **kwd):
        """Initialize foobar datatype"""
        Text.__init__( self, **kwd )

    def init_meta( self, dataset, copy_from=None ):
        Text.init_meta( self, dataset, copy_from=copy_from )

    def sniff( self, filename ):
        """ Returning false as the newick format is too general and cannot be sniffed."""
        return False

    def get_visualizations( self, dataset ):
        """
        Returns a list of visualizations for datatype.
        """

        return [ 'phyloviz' ]


class Nexus( Text ):
    """Nexus format as used By Paup, Mr Bayes, etc"""
    edam_data = "data_0872"
    edam_format = "format_1912"
    file_ext = "nex"

    def __init__(self, **kwd):
        """Initialize foobar datatype"""
        Text.__init__( self, **kwd )

    def init_meta( self, dataset, copy_from=None ):
        Text.init_meta( self, dataset, copy_from=copy_from )

    def sniff( self, filename ):
        """All Nexus Files Simply puts a '#NEXUS' in its first line"""
        f = open( filename, "r" )
        firstline = f.readline().upper()
        f.close()

        if "#NEXUS" in firstline:
            return True
        else:
            return False

    def get_visualizations( self, dataset ):
        """
        Returns a list of visualizations for datatype.
        """

        return [ 'phyloviz' ]


# ------------- Utility methods --------------

# nice_size used to be here, but to resolve cyclical dependencies it's been
# moved to galaxy.util.  It belongs there anyway since it's used outside
# datatypes.
nice_size = util.nice_size


def get_test_fname( fname ):
    """Returns test data filename"""
    path = os.path.dirname(__file__)
    full_path = os.path.join( path, 'test', fname )
    return full_path


def get_file_peek( file_name, is_multi_byte=False, WIDTH=256, LINE_COUNT=5, skipchars=None, line_wrap=True ):
    """
    Returns the first LINE_COUNT lines wrapped to WIDTH

    >>> fname = get_test_fname('4.bed')
    >>> get_file_peek(fname, LINE_COUNT=1)
    u'chr22\\t30128507\\t31828507\\tuc003bnx.1_cds_2_0_chr22_29227_f\\t0\\t+\\n'
    """
    # Set size for file.readline() to a negative number to force it to
    # read until either a newline or EOF.  Needed for datasets with very
    # long lines.
    if WIDTH == 'unlimited':
        WIDTH = -1
    if skipchars is None:
        skipchars = []
    lines = []
    count = 0
    file_type = None
    data_checked = False
    temp = compression_utils.get_fileobj( file_name, "U" )
    try:
        while count < LINE_COUNT:
            line = temp.readline( WIDTH )
            if line and not is_multi_byte and not data_checked:
                # See if we have a compressed or binary file
                for char in line:
                    if ord( char ) > 128:
                        file_type = 'binary'
                        break
                data_checked = True
                if file_type == 'binary':
                    break
            if not line_wrap:
                if line.endswith('\n'):
                    line = line[:-1]
                else:
                    while True:
                        i = temp.read(1)
                        if not i or i == '\n':
                            break
            skip_line = False
            for skipchar in skipchars:
                if line.startswith( skipchar ):
                    skip_line = True
                    break
            if not skip_line:
                lines.append( line )
                count += 1
    finally:
        temp.close()
    if file_type == 'binary':
        text = "%s file" % file_type
    else:
        try:
            text = util.unicodify( '\n'.join( lines ) )
        except UnicodeDecodeError:
            text = "binary/unknown file"
    return text