• Kirill Smelkov's avatar
    BigFile: Factor out code to append data chunk to ._appendData() · f126d0bf
    Kirill Smelkov authored
    So that data could be appended on server code via direct calls too.
    
    NOTE previously ._read_data() accepted data=None argument and callers
    were either providing it with current .data to append or None to forget
    old content and just add new fresh one.
    
    We could drop data=None from _read_data() signature, but leave it as is
    for compatibility with outside code (e.g. zope's
    OFS.Image.File.manage_upload() calls ._read_data(file) without any data
    argument and in that case file content should be recreated, not
    appended).
    
    On the other hand we rework our code in .PUT() so for both "new content"
    and "append range" in the end it always does "append" operation. For it
    to work this way "new content" simply truncates the file before
    proceeding to "append".
    Reviewed-by: Romain Courteaud's avatarRomain Courteaud <romain@nexedi.com>
    f126d0bf
BigFile.py 14 KB
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409
# -*- coding: utf-8 -*-
##############################################################################
#
# Copyright (c) 2002 Zope Foundation and Contributors.
#               2012 Nexedi SA and Contributors. All Rights Reserved.
#                    Romain Courteaud <romain@nexedi.com>
#
# This software is subject to the provisions of the Zope Public License,
# Version 2.1 (ZPL).  A copy of the ZPL should accompany this distribution.
# THIS SOFTWARE IS PROVIDED "AS IS" AND ANY AND ALL EXPRESS OR IMPLIED
# WARRANTIES ARE DISCLAIMED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF TITLE, MERCHANTABILITY, AGAINST INFRINGEMENT, AND FITNESS
# FOR A PARTICULAR PURPOSE
#
##############################################################################

from cStringIO import StringIO
from AccessControl import ClassSecurityInfo
from Products.ERP5Type import Permissions, PropertySheet
from Products.ERP5Type.Base import removeIContentishInterface
from Products.ERP5.Document.File import File, _MARKER
from Products.ERP5Type.BTreeData import BTreeData
from ZPublisher.HTTPRequest import FileUpload
from ZPublisher import HTTPRangeSupport
from webdav.common import rfc1123_date
from mimetools import choose_boundary
from Products.CMFCore.utils import getToolByName, _setCacheHeaders,\
    _ViewEmulator
import re

class BigFile(File):
  """
  Support storing huge file.
  No convertion is allowed for now.


  NOTE BigFile maintains the following invariant:

    data property is either

      - BTreeData instance,  or
      - str(*),  or
      - None.

    (*) str has to be supported because '' is a default value for `data` field
        from Data property sheet.

        Even more - for

            a) compatibility reasons, and
            b) desire to support automatic migration of File-based documents
               from document_module to BigFiles

        non-empty str for data also have to be supported.

        XXX(kirr) I'm not sure supporting non-empty str is a good idea (it
            would be simpler if .data could be either BTreeData or "empty"),
            but neither I'm experienced enough in erp5 nor know what are
            appropriate compatibility requirements.

            We discussed with Romain and settled on "None or str or BTreeData"
            invariant for now.
  """

  meta_type = 'ERP5 Big File'
  portal_type = 'Big File'

  # Declarative security
  security = ClassSecurityInfo()
  security.declareObjectProtected(Permissions.AccessContentsInformation)

  # Default Properties
  property_sheets = ( PropertySheet.Base
                    , PropertySheet.XMLObject
                    , PropertySheet.CategoryCore
                    , PropertySheet.DublinCore
                    , PropertySheet.Version
                    , PropertySheet.Reference
                    , PropertySheet.Document
                    , PropertySheet.Data
                    , PropertySheet.ExternalDocument
                    , PropertySheet.Url
                    , PropertySheet.Periodicity
    )

  # OFS.File has an overloaded __str__ that returns the file content
  __str__ = object.__str__

  security.declareProtected(Permissions.AccessContentsInformation,
                            'getData')
  def getData(self):
    """Read the full btree
    """
    btree = self._baseGetData()
    if isinstance(btree, BTreeData):
      return btree.read(0, len(btree))
    else:
      return btree

  security.declareProtected(Permissions.ModifyPortalContent, 'updateContentMd5')
  def updateContentMd5(self):
    """Update md5 checksum from the original file
    """
    self._setContentMd5(None)

  def _read_data(self, file, data=None, serialize=True):

    # We might need to make this value configurable. It is important to
    # consider the max quantity of object used in the cache. With a default
    # cache of 5000 objects, and with n = 64KB, this makes using about 330 MB
    # of memory.
    n=1 << 16

    if isinstance(file, str):
      # Big string: cut it into smaller chunks
      file = StringIO(file)

    if isinstance(file, FileUpload) and not file:
      raise ValueError, 'File not specified'

    seek=file.seek
    read=file.read

    seek(0,2)
    end=file.tell()

    if data is None:
      btree = BTreeData()
    elif isinstance(data, str):
      # we'll want to append content to this file -
      # - automatically convert str (empty or not) to BTreeData
      btree = BTreeData()
      btree.write(data, 0)
    else:
      btree = data
    seek(0)
    pos = file.tell()
    offset = len(btree)

    while pos < end:
      next = pos + n
      if next > end:
        next = end

      btree.write(read(next-pos), offset+pos)
      pos = file.tell()

    if serialize:
      self.serialize()
    return btree, len(btree)

  def _data_mtime(self):
    """get .data mtime if present and fallback to self._p_mtime"""
    # there is no data._p_mtime when data is None or str.
    # so try and fallback to self._p_mtime
    data = self._baseGetData()
    mtime = getattr(data, '_p_mtime', self._p_mtime)
    return mtime

  def _range_request_handler(self, REQUEST, RESPONSE):
    # HTTP Range header handling: return True if we've served a range
    # chunk out of our data.
    range = REQUEST.get_header('Range', None)
    request_range = REQUEST.get_header('Request-Range', None)
    if request_range is not None:
      # Netscape 2 through 4 and MSIE 3 implement a draft version
      # Later on, we need to serve a different mime-type as well.
      range = request_range
    if_range = REQUEST.get_header('If-Range', None)
    if range is not None:
      ranges = HTTPRangeSupport.parseRange(range)

      data = self._baseGetData()

      if if_range is not None:
        # Only send ranges if the data isn't modified, otherwise send
        # the whole object. Support both ETags and Last-Modified dates!
        if len(if_range) > 1 and if_range[:2] == 'ts':
          # ETag:
          if if_range != self.http__etag():
            # Modified, so send a normal response. We delete
            # the ranges, which causes us to skip to the 200
            # response.
            ranges = None
        else:
          # Date
          date = if_range.split( ';')[0]
          try: mod_since=long(DateTime(date).timeTime())
          except: mod_since=None
          if mod_since is not None:
            last_mod = self._data_mtime()
            if last_mod is None:
                last_mod = 0
            last_mod = long(last_mod)
            if last_mod > mod_since:
              # Modified, so send a normal response. We delete
              # the ranges, which causes us to skip to the 200
              # response.
              ranges = None

      if ranges:
        # Search for satisfiable ranges.
        satisfiable = 0
        for start, end in ranges:
          if start < self.getSize():
            satisfiable = 1
            break

        if not satisfiable:
          RESPONSE.setHeader('Content-Range',
              'bytes */%d' % self.getSize())
          RESPONSE.setHeader('Accept-Ranges', 'bytes')
          RESPONSE.setHeader('Last-Modified', rfc1123_date(self._data_mtime()))
          RESPONSE.setHeader('Content-Type', self.content_type)
          RESPONSE.setHeader('Content-Length', self.getSize())
          RESPONSE.setStatus(416)
          return True

        ranges = HTTPRangeSupport.expandRanges(ranges, self.getSize())

        if len(ranges) == 1:
          # Easy case, set extra header and return partial set.
          start, end = ranges[0]
          size = end - start

          RESPONSE.setHeader('Last-Modified', rfc1123_date(self._data_mtime()))
          RESPONSE.setHeader('Content-Type', self.content_type)
          RESPONSE.setHeader('Content-Length', size)
          RESPONSE.setHeader('Accept-Ranges', 'bytes')
          RESPONSE.setHeader('Content-Range',
              'bytes %d-%d/%d' % (start, end - 1, self.getSize()))
          RESPONSE.setStatus(206) # Partial content

          # NOTE data cannot be None here (if it is - ranges are not satisfiable)
          if isinstance(data, str):
            RESPONSE.write(data[start:end])
            return True
          for chunk in data.iterate(start, end-start):
            RESPONSE.write(chunk)
          return True

        else:
          boundary = choose_boundary()

          # Calculate the content length
          size = (8 + len(boundary) + # End marker length
              len(ranges) * (         # Constant lenght per set
                  49 + len(boundary) + len(self.content_type) +
                  len('%d' % self.getSize())))
          for start, end in ranges:
            # Variable length per set
            size = (size + len('%d%d' % (start, end - 1)) +
                end - start)

          data = self._baseGetData()

          # Some clients implement an earlier draft of the spec, they
          # will only accept x-byteranges.
          draftprefix = (request_range is not None) and 'x-' or ''

          RESPONSE.setHeader('Content-Length', size)
          RESPONSE.setHeader('Accept-Ranges', 'bytes')
          RESPONSE.setHeader('Last-Modified', rfc1123_date(self._data_mtime()))
          RESPONSE.setHeader('Content-Type',
              'multipart/%sbyteranges; boundary=%s' % (
                  draftprefix, boundary))
          RESPONSE.setStatus(206) # Partial content

          for start, end in ranges:
            RESPONSE.write('\r\n--%s\r\n' % boundary)
            RESPONSE.write('Content-Type: %s\r\n' %
                self.content_type)
            RESPONSE.write(
                'Content-Range: bytes %d-%d/%d\r\n\r\n' % (
                    start, end - 1, self.getSize()))

            # NOTE data cannot be None here (if it is - ranges are not satisfiable)
            if isinstance(data, str):
              RESPONSE.write(data[start:end])

            else:
              for chunk in data.iterate(start, end-start):
                RESPONSE.write(chunk)

          RESPONSE.write('\r\n--%s--\r\n' % boundary)
          return True

  security.declareProtected(Permissions.View, 'index_html')
  def index_html(self, REQUEST, RESPONSE, format=_MARKER, inline=_MARKER, **kw):
    """
      Support streaming
    """
    if self._range_request_handler(REQUEST, RESPONSE):
      # we served a chunk of content in response to a range request.
      return ''

    web_cache_kw = kw.copy()
    if format is not _MARKER:
      web_cache_kw['format'] = format
    _setCacheHeaders(_ViewEmulator().__of__(self), web_cache_kw)

    if format is _MARKER and not kw:
      # conversion parameters is mandatory to download the converted content.
      # By default allways return view action.
      # for all WevDAV access return raw content.
      return self.view()

    if format is _MARKER:
      format = None

    data = self._baseGetData()
    mime = self.getContentType()

    RESPONSE.setHeader('Content-Length', data is not None and  len(data)  or  0)
    RESPONSE.setHeader('Content-Type', mime)
    if inline is _MARKER:
      # by default, use inline for text and image formats
      inline = False
    if not inline:
      # need to return it as attachment
      filename = self.getStandardFilename(format=format)
      RESPONSE.setHeader('Cache-Control', 'Private') # workaround for Internet Explorer's bug
      RESPONSE.setHeader('Accept-Ranges', 'bytes')


    if data is None:
      return ''
    if isinstance(data, str):
      RESPONSE.setBase(None)
      return data
    for chunk in data.iterate():
      RESPONSE.write(chunk)
    return ''

  security.declareProtected(Permissions.ModifyPortalContent,'PUT')
  def PUT(self, REQUEST, RESPONSE):
    """Handle HTTP PUT requests"""
    self.dav__init(REQUEST, RESPONSE)
    self.dav__simpleifhandler(REQUEST, RESPONSE, refresh=1)

    type=REQUEST.get_header('content-type', None)

    file=REQUEST['BODYFILE']

    content_range = REQUEST.get_header('Content-Range', None)
    if content_range is None:
      # truncate the file
      self._baseSetData(None)
    else:
      current_size = int(self.getSize())
      query_range = re.compile('bytes \*/\*')
      append_range = re.compile('bytes (?P<first_byte>[0-9]+)-' \
                                '(?P<last_byte>[0-9]+)/' \
                                '(?P<total_content_length>[0-9]+)')
      if query_range.match(content_range):
        data = self._baseGetData()

        RESPONSE.setHeader('X-Explanation', 'Resume incomplete')
        RESPONSE.setHeader('Range', 'bytes 0-%s' % (current_size-1))
        RESPONSE.setStatus(308)
        return RESPONSE

      if append_range.match(content_range):

        result_dict = append_range.search(content_range).groupdict()
        first_byte = int(result_dict['first_byte'])
        last_byte = int(result_dict['last_byte'])
        total_content_length = int(result_dict['total_content_length'])
        content_length= int(REQUEST.get_header('Content-Length', '0'))

        if (first_byte != current_size):
          RESPONSE.setHeader('X-Explanation', 'Can only append data')
          RESPONSE.setStatus(400)
          return RESPONSE
        elif (last_byte+1 != total_content_length):
          RESPONSE.setHeader('X-Explanation', 'Total size unexpected')
          RESPONSE.setStatus(400)
          return RESPONSE
        elif (last_byte+1-first_byte != content_length):
          RESPONSE.setHeader('X-Explanation', 'Content length unexpected')
          RESPONSE.setStatus(400)
          return RESPONSE

      else:
        RESPONSE.setHeader('X-Explanation', 'Can not parse range')
        RESPONSE.setStatus(400) # Partial content
        return RESPONSE

    self._appendData(file, content_type=type)

    RESPONSE.setStatus(204)
    return RESPONSE


  def _appendData(self, data_chunk, content_type=None):
    """append data chunk to the end of the file

       NOTE if content_type is specified, it will change content_type for the
            whole file.
    """
    data, size = self._read_data(data_chunk, data=self._baseGetData())
    content_type=self._get_content_type(data_chunk, data, self.__name__,
                                        content_type or self.content_type)
    self.update_data(data, content_type, size)


# CMFFile also brings the IContentishInterface on CMF 2.2, remove it.
removeIContentishInterface(BigFile)