Adding debian version 2.4.56-1~deb11u2.debian/2.4.56-1_deb11u2debian

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

1 files changed, 661 insertions, 0 deletions

diff --git a/debian/perl-framework/t/modules/include.t b/debian/perl-framework/t/modules/include.t
new file mode 100644
index 0000000..9ff2411
--- /dev/null
+++ b/debian/perl-framework/t/modules/include.t
@@ -0,0 +1,661 @@
+use strict;
+use warnings FATAL => 'all';
+
+use Apache::Test;
+use Apache::TestRequest;
+use Apache::TestUtil;
+
+use File::Spec::Functions qw(catfile splitpath);
+
+Apache::TestRequest::scheme('http'); #ssl not listening on this vhost
+Apache::TestRequest::module('mod_include'); #use this module's port
+
+use constant WINFU => Apache::TestConfig::WINFU;
+
+## mod_include tests
+my($res, $str, $doc);
+my $dir = "/modules/include/";
+my $have_apache_1 = have_apache 1;
+my $have_apache_2 = have_apache 2;
+my $have_apache_21 = have_min_apache_version "2.1.0";
+my $have_apache_20 = $have_apache_2 && ! $have_apache_21;
+my $htdocs = Apache::Test::vars('documentroot');
+
+# these match the SSI files with their expected results.
+# the expectations are set by the current 2.1 mod_include
+# implementation.
+
+my %test = (
+"echo.shtml"            =>    "echo.shtml",
+"set.shtml"             =>    "set works",
+"comment.shtml"         =>    "No  comment  here",
+"include1.shtml"        =>    "inc-two.shtml body  include.shtml body",
+"include2.shtml"        =>    "inc-two.shtml body  include.shtml body",
+"include3.shtml"        =>    "inc-two.shtml body  inc-one.shtml body  ".
+                              "include.shtml body",
+"include4.shtml"        =>    "inc-two.shtml body  inc-one.shtml body  ".
+                              "include.shtml body",
+"include5.shtml"        =>    "inc-two.shtml body  inc-one.shtml body  ".
+                              "inc-three.shtml body  include.shtml body",
+"include6.shtml"        =>    "inc-two.shtml body  inc-one.shtml body  ".
+                              "inc-three.shtml body  include.shtml body",
+"foo.shtml"             =>    "[an error occurred while processing this ".
+                              "directive] foo.shtml body",
+"foo1.shtml"            =>    "[an error occurred while processing this ".
+                              "directive] foo.shtml body",
+"foo2.shtml"            =>    "[an error occurred while processing this ".
+                              "directive] foo.shtml body",
+"encode.shtml"          =>    "\# \%\^ \%23\%20\%25\%5e",
+"errmsg1.shtml"         =>    "errmsg",
+"errmsg2.shtml"         =>    "errmsg",
+"errmsg3.shtml"         =>    "errmsg",
+"errmsg4.shtml"         =>    "pass errmsg",
+"errmsg5.shtml"         =>    "<!-- pass -->",
+"if1.shtml"             =>    "pass",
+"if2.shtml"             =>    "pass   pass",
+"if3.shtml"             =>    "pass   pass   pass",
+"if4.shtml"             =>    "pass   pass",
+"if5.shtml"             =>    "pass  pass  pass",
+"if6.shtml"             =>    "[an error occurred while processing this ".
+                              "directive]",
+"if7.shtml"             =>    "[an error occurred while processing this ".
+                              "directive]",
+"if8.shtml"             =>    "pass",
+"if9.shtml"             =>    "pass   pass",
+"if10.shtml"            =>    "pass",
+"if11.shtml"            =>    "pass",
+"big.shtml"             =>    "hello   pass  pass   pass     hello",
+"newline.shtml"         =>    "inc-two.shtml body",
+"inc-rfile.shtml"       =>    "inc-extra2.shtml body  inc-extra1.shtml body  ".
+                              "inc-rfile.shtml body",
+"inc-rvirtual.shtml"    =>    "inc-extra2.shtml body  inc-extra1.shtml body  ".
+                              "inc-rvirtual.shtml body",
+"extra/inc-bogus.shtml" =>    "[an error occurred while processing this ".
+                              "directive] inc-bogus.shtml body",
+"abs-path.shtml"        =>    "inc-extra2.shtml body  inc-extra1.shtml body  ".
+                              "abs-path.shtml body",
+"parse1.shtml"          =>    "-->",
+"parse2.shtml"          =>    '"',
+"regex.shtml"           =>    "(none)  1 (none)",
+"retagged1.shtml"       =>    ["retagged1.shtml",                   "retagged1"],
+"retagged2.shtml"       =>    ["----retagged2.shtml",               "retagged1"],
+"echo1.shtml"           =>    ["<!-- pass undefined echo -->",      "echo1"    ],
+"echo2.shtml"           =>    ["<!-- pass undefined echo -->  pass  config ".
+                              " echomsg  pass", "echo1"],
+"echo3.shtml"           =>    ['<!--#echo var="DOCUMENT_NAME" -->', "retagged1"], 
+"notreal.shtml"         =>    "pass <!--",
+"malformed.shtml"       =>    "[an error occurred while processing this ".
+                              "directive] malformed.shtml",
+"exec/off/cmd.shtml"    =>    "[an error occurred while processing this ".
+                              "directive]",
+"exec/on/cmd.shtml"     =>    "pass",
+"exec/off/cgi.shtml"    =>    "[an error occurred while processing this ".
+                              "directive]",
+"exec/on/cgi.shtml"     =>    "perl cgi",
+"ranged-virtual.shtml"  =>    "x"x32768,
+"var128.shtml"          =>    "x"x126 . "yz",  # PR#32985
+"virtualq.shtml?foo=bar" =>   "foo=bar  pass    inc-two.shtml body  foo=bar", # PR#12655
+
+"inc-nego.shtml"        =>    "index.html.en", # requires mod_negotiation
+"mod_request/echo.shtml"=>    "echo.shtml",
+"mod_request/post.shtml?foo=bar&foo2=bar2"
+                        =>    "GET foo: bar foo2: bar2",
+"mod_request/post.shtml"=>    "POST foo: bar foo2: bar2",   # will be twice, only the first one succeed
+);
+
+my %ap_expr_test = (
+"apexpr/if1.shtml"      =>    "pass",
+"apexpr/err.shtml"      =>    "[an error occurred while processing this ".
+                              "directive] err.shtml",
+"apexpr/restrict.shtml" =>    "[an error occurred while processing this ".
+                              "directive] restrict.shtml",
+"apexpr/var.shtml"      =>    "pass   pass   pass",
+"apexpr/lazyvar.shtml"  =>    "pass",
+);
+
+if (have_min_apache_version "2.3.13") {
+    %test = (%test, %ap_expr_test);
+}
+
+# now, assuming 2.1 has the proper behavior across the board,
+# let's adjust our expectations for other versions
+
+# these tests are known to be broken in 2.0
+# we'll mark them as TODO tests in the hopes
+# that the 2.1 fixes will be backported
+
+my %todo = (
+);
+
+# some behaviors will never be backported, for various
+# reasons.  these are the 1.3 legacy tests and expectations
+my %legacy_1_3 = (
+"errmsg4.shtml"         =>    "pass",
+"malformed.shtml"       =>    "",
+"if6.shtml"             =>    "",
+"if7.shtml"             =>    "",
+);
+
+# 2.0 has no legacy tests at the moment
+# but when it does, they will go here
+my %legacy_2_0 = ();
+
+# ok, now that we have our hashes established, here are
+# the manual tweaks
+if ($have_apache_1) {
+    # apache 1.3 uses different semantics for some
+    # of the if.*shtml tests to achieve the same results
+    $test{"if8a.shtml"}  = delete $test{"if8.shtml"};
+    $test{"if9a.shtml"}  = delete $test{"if9.shtml"};
+    $test{"if10a.shtml"} = delete $test{"if10.shtml"};
+
+    # while other tests are for entirely new behaviors
+    # and don't make sense to test at all in 1.3
+    delete $test{"echo1.shtml"};
+    delete $test{"echo2.shtml"};
+    delete $test{"echo3.shtml"};
+    delete $test{"retagged1.shtml"};
+    delete $test{"retagged2.shtml"};
+    delete $test{"regex.shtml"};
+
+    # finally, these tests are only broken in 1.3
+    $todo{"notreal.shtml"} = delete $test{"notreal.shtml"};
+}
+
+unless ($have_apache_20) {
+    # these tests are broken only in 2.0 - 
+    # in 1.3 they work fine so shift them from %todo to %test
+
+    # none at the moment, but the syntax here would be
+    # $test{"errmsg5.shtml"} = delete $todo{"errmsg5.shtml"};
+}
+
+unless (have_min_apache_version "2.0.53") {
+    # this test doesn't work in 2.0 yet but should work in 1.3 and 2.1
+    delete $test{"ranged-virtual.shtml"};
+}
+
+unless ($have_apache_21) {
+    # apache 1.3 and 2.0 do not support these tests
+    delete $test{"echo2.shtml"};
+}
+
+unless (have_module 'mod_negotiation') {
+    delete $test{"inc-nego.shtml"};
+}
+
+# this test does not work on win32 (<!--#exec cmd="echo pass"-->)
+if (WINFU) {
+    delete $test{'exec/on/cmd.shtml'};
+}
+
+my @patterns = (
+    'mod_include test',
+    'Hello World',
+    'footer',
+);
+
+# with the tweaks out of the way, we can get on
+# with planning the tests
+
+# first, total the number of hashed tests
+# note that some legacy tests will redefine the main
+# %test hash, so the total is not necessarily the sum
+# of all the keys 
+my %tests = ();
+
+if ($have_apache_21) {
+    %tests = (%test, %todo);
+}
+elsif ($have_apache_2) {
+    %tests = (%test, %todo, %legacy_2_0);
+}
+else {
+    %tests = (%test, %todo, %legacy_1_3);
+}
+
+# now for the TODO tests
+my @todo = ();
+unless ($have_apache_21) {
+    # if 1.3 or 2.0, dynamically determine which of %test
+    # will end up being TODO tests.  
+
+    my $counter = 0;
+    foreach my $test (sort keys %tests) {
+      $counter++;
+      push @todo, $counter if $todo{$test};
+    }
+}
+
+unless ($have_apache_2) {
+    # fsize comes immediately after the hashed tests
+    push @todo, (scalar keys %tests) + 1;
+}
+
+# in addition to %tests, there are 1 mod_request expected failure,
+# 1 fsize and 1 flastmod test,
+# 1 GET test, 2 query string tests, 14 XBitHack tests and 14 
+# tests that use mod_bucketeer to construct brigades for mod_include
+
+my $tests = (scalar keys %tests) + 1 + @patterns + 1 + 1 + 1 + 2 + 14 + 14;
+
+plan tests => $tests,
+              todo => \@todo,
+              need 'DateTime', need_lwp, need_module 'include';
+
+foreach $doc (sort keys %tests) {
+    # do as much from %test as we can
+    if (ref $tests{$doc}) {
+        ok t_cmp(super_chomp(GET_BODY "$dir$doc", Host => $tests{$doc}[1]),
+                 $tests{$doc}[0],
+                 "GET $dir$doc"
+                );
+    }
+    elsif ($doc =~ m/ranged/) {
+        if (have_cgi) {
+            ok t_cmp(GET_BODY("$dir$doc", Range => "bytes=0-"),
+                     $tests{$doc},
+                     "GET $dir$doc with Range"
+                     );
+        }
+        else {
+            skip "Skipping virtual-range test; no cgi module", 1;
+        }
+    }
+    elsif ($doc =~ m/cgi/) {
+        if (have_cgi) {
+            ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+                     $tests{$doc},
+                     "GET $dir$doc"
+                    );
+        }
+        else {
+            skip "Skipping 'exec cgi' test; no cgi module.", 1;
+        }
+    }
+    elsif ($doc =~ m/mod_request.*\?/) {
+        # param in the url ==> use GET
+        if (have_cgi) {
+            ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+                     $tests{$doc},
+                     "GET $dir$doc"
+                    );
+        }
+        else {
+            skip "Skipping 'exec cgi' test; no cgi module.", 1;
+        }
+    }
+    elsif ($doc =~ m/mod_request/) {
+        # no param in the url ==> use POST with a content
+        if (have_cgi) {
+            ok t_cmp(super_chomp(POST_BODY "$dir$doc", content => "foo=bar&foo2=bar2"),
+                     $tests{$doc},
+                     "POST $dir$doc"
+                    );
+            if ($doc =~ m/mod_request.*post/) {
+                # KeptBodySize is 32
+                my $r = POST("$dir$doc", content => "foo=bar&foo2=bar2&foo3=bar3&foo4=bar4");
+                ok t_cmp($r->code, 413, "sizeof(body) > KeptBodySize");
+            }
+        }
+        else {
+            skip "Skipping 'exec cgi' test; no cgi module.", 2;
+        }
+    }
+    else {
+        ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+                 $tests{$doc},
+                 "GET $dir$doc"
+                );
+    }
+}
+
+### FLASTMOD/FSIZE TESTS
+
+# marked as TODO in 1.3 - hoping for a format backport
+{
+    my $file = catfile($htdocs, splitpath($dir), "size.shtml");
+    my $size = (stat $file)[7];
+
+    # round perl's stat size for <!--#config sizefmt="abbrev"-->
+    # this assumes the size of size.shtml is such that it is
+    # rendered in K (which it is).  if size.shtml is made much
+    # larger or smaller this formatting will need to change too
+    my $abbrev = sprintf("%.1fK", $size/1024);
+
+    # and commify for <!--#config sizefmt="bytes"-->
+    my $bytes = commify($size);
+
+    my $expected = join ' ', $bytes, $bytes, $abbrev, $abbrev;
+
+    my $result = super_chomp(GET_BODY "${dir}size.shtml");
+
+    # trim output
+    $result =~ s/X//g;   # the Xs were there just to pad the filesiez
+    $result = single_space($result);
+
+    ok t_cmp("$result",
+             "$expected",
+             "GET ${dir}size.shtml"
+            );
+}
+
+unless(eval "require POSIX") {
+    skip "POSIX module not found", 1;
+}
+else {
+    # use DateTime and avoid the system locale messing things up
+    use DateTime;
+    # Only for checking, whether system strftime supports %s
+    use POSIX;
+    my $strftime_gnu = (POSIX::strftime("%s", gmtime()) eq '%s' ? 0 : 1);
+
+    my $result = super_chomp(GET_BODY "${dir}file.shtml");
+    $result = single_space($result);
+
+    my $httpdtz = $1 if $result =~ /\w+, \d+-\w+-\d+ \d+:\d+:\d+ (\w+) /;
+
+    my $file = catfile($htdocs, splitpath($dir), "file.shtml");
+    my $mtime = (stat $file)[9];
+
+    my $dt = DateTime->from_epoch( epoch => $mtime,
+                locale => 'en_US', time_zone => $httpdtz||'UTC' );
+    
+    my $expected = join ' ' =>
+        $dt->strftime("%A, %B %e, %G"),
+        $dt->strftime("%A, %B %e, %G"),
+        $strftime_gnu ? $dt->strftime("%s") : '%s',
+        $strftime_gnu ? $dt->strftime("%s") : '%s';
+
+    # trim output
+    $expected = single_space($expected);
+
+    ok t_cmp("$result",
+             "$expected",
+             "GET ${dir}file.shtml"
+            );
+}
+
+# some tests that can't be easily assimilated
+
+$doc = "printenv.shtml";
+ok t_cmp(GET("$dir$doc")->code,
+         "200",
+         "GET $dir$doc"
+        );
+
+### test include + query string
+$res = GET "${dir}virtual.shtml";
+
+ok $res->is_success;
+
+$str = $res->content;
+
+ok $str;
+
+for my $pat (@patterns) {
+    ok t_cmp($str, qr/$pat/, "/$pat/");
+}
+
+### MOD_BUCKETEER+MOD_INCLUDE TESTS
+if (WINFU) {
+    for (1..13) {
+        skip "Skipping XBitHack tests on this platform", 1;
+    }
+}
+else {
+    ### XBITHACK TESTS
+    # test xbithack off
+    $doc = "xbithack/off/test.html";
+    foreach ("0444", "0544", "0554") {
+        chmod oct($_), "$htdocs/$dir$doc";
+        ok t_cmp(super_chomp(GET_BODY "$dir$doc"),,
+                 "<BODY> <!--#include virtual=\"../../inc-two.shtml\"--> </BODY>",
+                 "XBitHack off [$_]"
+                );
+    }
+
+    # test xbithack on
+    $doc = "xbithack/on/test.html";
+    chmod 0444, "$htdocs$dir$doc";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             "<BODY> <!--#include virtual=\"../../inc-two.shtml\"--> </BODY>",
+             "XBitHack on [0444]"
+            );
+
+    foreach ("0544", "0554") {
+        chmod oct($_), "$htdocs/$dir$doc";
+        ok t_cmp(check_xbithack(GET "$dir$doc"),
+                 "No Last-modified date ; <BODY> inc-two.shtml body  </BODY>",
+                 "XBitHack on [$_]"
+                );
+    }
+
+    # test timefmt - make sure filter only inserted once
+    # if Option Include and xbithack both say to process
+    $doc = "xbithack/both/timefmt.shtml";
+    my @now = localtime();
+    my $year = $now[5] + 1900;
+    chmod 0555, "$htdocs/$dir$doc";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             "xx${year}xx",
+             "XBitHack both [timefmt]"
+             );
+
+    # test xbithack full
+    $doc = "xbithack/full/test.html";
+    chmod 0444, "$htdocs/$dir$doc";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             "<BODY> <!--#include virtual=\"../../inc-two.shtml\"--> </BODY>",
+             "XBitHack full [0444]"
+            );
+    chmod 0544, "$htdocs/$dir$doc";
+    ok t_cmp(check_xbithack(GET "$dir$doc"),
+             "No Last-modified date ; <BODY> inc-two.shtml body  </BODY>",
+             "XBitHack full [0544]"
+            );
+
+    my $lm;
+
+    chmod 0554, "$htdocs/$dir$doc";
+    ok t_cmp(check_xbithack(GET("$dir$doc"), \$lm),
+             "Has Last-modified date ; <BODY> inc-two.shtml body  </BODY>",
+             "XBitHack full [0554]"
+            );
+
+    ok t_cmp(check_xbithack_etag(GET("$dir$doc", 'If-Modified-Since' => $lm)),
+             "No ETag ; ",
+             "XBitHack full [0554] / ETag"
+            );
+
+    ok t_cmp(GET("$dir$doc", 'If-Modified-Since' => $lm)->code, 304,
+             "XBitHack full [0554] / If-Modified-Since"
+            );
+
+    chmod 0544, "$htdocs/$dir$doc";
+    ok t_cmp(GET("$dir$doc", 'If-Modified-Since' => $lm)->code, 200,
+             "XBitHack full [0544] / If-Modified-Since"
+            );
+
+    ok t_cmp(check_xbithack_etag(GET("$dir$doc", 'If-Modified-Since' => $lm)),
+             "No ETag ; <BODY> inc-two.shtml body  </BODY>",
+             "XBitHack full [0544] / ETag"
+            );
+}
+
+# we can use mod_bucketeer to create edge conditions for mod_include, since
+# it allows us to create bucket and brigade boundaries wherever we want
+if (have_module 'mod_bucketeer') {
+
+    my $expected = "____ _____ _____ ___________________ </table>  ".
+                   "##################################1/8</tr> ".
+                   "##################################2/8</tr> ".
+                   "##################################3/8</tr> ".
+                   "##################################4/8</tr> ".
+                   "##################################5/8</tr> ".
+                   "##################################6/8$htdocs</tr> ".
+                   "##################################7/8</tr> ".
+                   "##################################8/8</tr> ".
+                   "@@@@@@@@ @@@@@@@@@@@@@@@@@@@@@@@@";
+
+    $doc = "bucketeer/y.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             $expected,
+             "GET $dir$doc"
+            );
+
+    $expected = "____ ___________________________________".
+                "________________________________________".
+                "___ ____________________________________".
+                "________________________________________".
+                "__________ ___________________ </table>  ".
+                "#####################################</tr> ".
+                "#####################################</tr> ".
+                "#####################################</tr> ".
+                "#####################################</tr> ".
+                "#####################################</tr> ".
+                "#####################################</tr> ".
+                "#####################################</tr> ".
+                "#####################################</tr> ".
+                "@@@@@@@@ @@@@@@@@@@@@@@@@@@@@@@@@";
+
+    for (0..3) {
+        $doc = "bucketeer/y$_.shtml";
+        my ($body) = super_chomp(GET_BODY "$dir$doc");
+        $body =~ s/\002/^B/g;
+        $body =~ s/\006/^F/g;
+        $body =~ s/\020/^P/g;
+        ok t_cmp($body,
+                 $expected,
+                 "GET $dir$doc"
+                );
+    }
+
+    $expected = "[an error occurred while processing this directive]";
+    $doc = "bucketeer/y4.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             $expected,
+             "GET $dir$doc"
+            );
+
+
+    $expected= "pass [an error occurred while processing this directive]  ".
+               "pass pass1";
+    $doc = "bucketeer/y5.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             $expected,
+             "GET $dir$doc"
+            );
+
+    $expected= "BeforeIfElseBlockAfterIf";
+    $doc = "bucketeer/y6.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             $expected,
+             "GET $dir$doc"
+            );
+
+    $expected= "Before If <!-- comment -->SomethingElse".
+               "<!-- right after if -->After if";
+    $doc = "bucketeer/y7.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             $expected,
+             "GET $dir$doc"
+            );
+
+    $expected= "FalseSetDone";
+    $doc = "bucketeer/y8.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             $expected,
+             "GET $dir$doc"
+            );
+
+    $expected= "FalseSetDone";
+    $doc = "bucketeer/y9.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             $expected,
+             "GET $dir$doc"
+            );
+
+    $expected= "\"pass\"";
+    $doc = "bucketeer/y10.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc"),
+             $expected,
+             "GET $dir$doc"
+            );
+
+    ### exotic SSI(Start|End)Tags
+
+    $expected= "----retagged3.shtml";
+    $doc = "bucketeer/retagged3.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc", Host => 'retagged1'),
+             $expected,
+             "GET $dir$doc"
+            );
+
+    $expected= "---pass";
+    $doc = "bucketeer/retagged4.shtml";
+    ok t_cmp(super_chomp(GET_BODY "$dir$doc", Host => 'retagged2'),
+             $expected,
+             "GET $dir$doc"
+            );
+}
+else {
+    for (1..14) {
+        skip "Skipping bucket boundary tests, no mod_bucketeer", 1;
+    }
+}
+
+sub super_chomp {
+    my ($body) = shift;
+
+    ## super chomp - all leading and trailing \n (and \r for win32)
+    $body =~ s/^[\n\r]*//;
+    $body =~ s/[\n\r]*$//;
+    ## and all the rest change to spaces
+    $body =~ s/\n/ /g;
+    $body =~ s/\r//g; #rip out all remaining \r's
+
+    $body;
+}
+
+sub check_xbithack {
+    my ($resp) = shift;
+    my ($body) = super_chomp($resp->content);
+    my ($lastmod) = ($resp->last_modified)
+                      ? "Has Last-modified date" : "No Last-modified date";
+
+    my $data = shift;
+    $$data = $resp->header('Last-Modified') if $data;
+
+    "$lastmod ; $body";
+}
+
+sub check_xbithack_etag {
+    my ($resp) = shift;
+    my ($body) = super_chomp($resp->content);
+    my ($etag) = ($resp->header('ETag'))
+                   ? "Has ETag" : "No ETag";
+
+    my $data = shift;
+    $$data = $etag if $data;
+
+    "$etag ; $body";
+}
+
+sub commify {
+    # add standard commas to numbers.  from perlfaq5
+
+    local $_  = shift;
+    1 while s/^([-+]?\d+)(\d{3})/$1,$2/;
+    return $_;
+}
+
+sub single_space {
+    # condense multiple spaces between values to a single
+    # space.  also trim initial and trailing whitespace
+
+    local $_ = shift; 
+    s/\s+/ /g;
+    s/(^ )|( $)//;
+    return $_;
+}