Apache2::Upload - Methods for dealing with file uploads.

    use Apache2::Upload;

    $req = Apache2::Request->new($r);
    $upload = $req->upload("foo");
    $size = $upload->size;

    # three methods to get at the upload's contents ... slurp, fh, io

    $upload->slurp($slurp_data);

    read $upload->fh, $fh_data, $size;
    ok $slurp_data eq $fh_data;

    my $io = $upload->io;
    print while <$io>;

Apache2::Upload is a new module based on the original package included in Apache2::Request 1.X. Users requiring the upload API must now CWuse Apache2::Upload, which adds the CWupload method to Apache2::Request. Apache2::Upload is largely backwards-compatible with the original 1.X API. See the PORTING from 1.X section below for a list of known issues.

This manpage documents the Apache2::Upload package.

    $upload->name()

The name of the HTML form element which generated the upload.

    $upload->filename()

The (client-side) filename as submitted in the HTML form. Note: some agents will submit the file's full pathname, while others may submit just the basename.

    $upload->fh()

Creates filehandle reference to the upload's spooled tempfile, which contains the full contents of the upload.

    $upload->io()

Creates a tied IO handle. This method is a more efficient version of CWfh, but with CWio the handle ref returned is not seekable. It is tied to an APR::Request::Brigade object, so you may use the brigade API on the tied object if you want to manipulate the IO stream (beyond simply reading from it).

The returned reference is actually an object which has CWread and CWreadline methods available. However these methods are just syntactic sugar for the underlying CWREAD and CWREADLINE methods from APR::Request::Brigade.

    $io = $upload->io;
    print while $io->read($_); # equivalent to: tied(*$io)->READ($_)

See READ and READLINE below for additional notes on their usage.

    $upload->bb()
    $upload->bb($set)

Get/set the APR::Brigade which represents the upload's contents.

    $upload->size()

Returns the size of the upload in bytes.

    $upload->info()
    $upload->info($set)

Get/set the additional header information table for the uploaded file. Returns a hash reference tied to the APR::Table class. An optional CW$table argument can be passed to reassign the upload's internal (apr_table_t) info table to the one CW$table represents.

    my $info = $upload->info;
    while (my($hdr_name, $hdr_value) = each %$info) {
        # ...
    }

    # fetch upload's Content-Type header
    my $type = $upload->info->{"Content-type"};

    $upload->type()

Returns the MIME type of the given Apache2::Upload object.

    my $type = $upload->type;

    #same as
    my $content_type = $upload->info->{"Content-Type"};
    $content_type =~ s/;.*$//ms;

    $upload->link()

To avoid recopying the upload's internal tempfile brigade on a *nix-like system, link will create a hard link to it:

  my $upload = $req->upload('foo');
  $upload->link("/path/to/newfile") or
      die sprintf "link from '%s' failed: $!", $upload->tempname;

Typically the new name must lie on the same device and partition as the brigade's tempfile. If this or any other reason prevents the OS from linking the files, CWlink() will instead copy the temporary file to the specified location.

    $upload->slurp($contents)

Reads the full contents of a file upload into the scalar argument. The return value is the length of the file.

    my $size = $upload->slurp($contents);

    $upload->tempname()

Provides the name of the spool file.

    my $tempname = $upload->tempname;

This class is derived from APR::Brigade, providing additional methods for TIEHANDLE, READ and READLINE. To be memory efficient, these methods delete buckets from the brigade as soon as their data is actually read, so you cannot CWseek on handles tied to this class. Such handles have semantics similar to that of a read-only socket.

    APR::Request::Brigade->TIEHANDLE($bb)

Creates a copy of the bucket brigade represented by CW$bb, and blesses that copy into the APR::Request::Brigade class. This provides syntactic sugar for using perl's builtin CWread, CWreadline, and CW<> operations on handles tied to this package:

    use Symbol;
    $fh = gensym;
    tie *$fh, "APR::Request:Brigade", $bb;
    print while <$fh>;

    $bb->READ($contents)
    $bb->READ($contents, $length)
    $bb->READ($contents, $length, $offset)

Reads data from the brigade CW$bb into CW$contents. When omitted CW$length defaults to CW-1, which reads the first bucket into CW$contents. A positive CW$length will read in CW$length bytes, or the remainder of the brigade, whichever is greater. CW$offset represents the index in CW$context to read the new data.

    $bb->READLINE()

Returns the first line of data from the bride. Lines are terminated by linefeeds (the '\012' character), but we may eventually use CW$/ instead.

APR::Request::Param::Table, APR::Request::Error, Apache2::Request, APR::Table(3)

  Copyright 2003-2006  The Apache Software Foundation

  Licensed under the Apache License, Version 2.0 (the "License");
  you may not use this file except in compliance with the License.
  You may obtain a copy of the License at

      http://www.apache.org/licenses/LICENSE-2.0

  Unless required by applicable law or agreed to in writing, software
  distributed under the License is distributed on an "AS IS" BASIS,
  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  See the License for the specific language governing permissions and
  limitations under the License.