html-template-users — Discussion of HTML::Template and related modules

Re: [htmltmpl] H::T::Expr and escape

[Michael P. <mp...@pl...>]

Sam Tregar wrote:
> On Thu, 13 Apr 2006, Michael Peters wrote:
> 
>> Is it possible to combine the 'escape' attribute of H::T with
>> H::T::Expr?
> 
> No, I don't think so.  I don't think it would be too hard to add
> support for it though.  HTML::Template supports combining escape and
> code-ref params, so it'd just be a matter of making the
> HTML::Template::Expr parser smart enough to pass it through.

It looks like it also doesn't support 'default'. So by making the
"parser smart enough" do you mean copying that huge regex from H::T and
modifying it with the expr regex?

A very simple approach would be to take any tmpl_vars/loops, etc that
use expr and capture anything before the expr="..." and after it and
then just add them to the generated tags. This however doesn't solve the
sub-ref problem.

-- 
Michael Peters
Developer
Plus Three, LP

Re: [htmltmpl] H::T::Expr and escape

[Sam T. <sa...@tr...>]

On Thu, 13 Apr 2006, Michael Peters wrote:

> Is it possible to combine the 'escape' attribute of H::T with
> H::T::Expr?

No, I don't think so.  I don't think it would be too hard to add
support for it though.  HTML::Template supports combining escape and
code-ref params, so it'd just be a matter of making the
HTML::Template::Expr parser smart enough to pass it through.

-sam

[htmltmpl] H::T::Expr and escape

[Michael P. <mp...@pl...>]

Is it possible to combine the 'escape' attribute of H::T with
H::T::Expr? I'm guessing not since any combination I've tried fails. If
it's too hard to do, maybe a new function could be added so that you can
do somthing like

  <tmpl_var expr="escape(uc(title), 'HTML')">

-- 
Michael Peters
Developer
Plus Three, LP

[htmltmpl] ANNOUNCE: Krang 2.004

[Aaron D. <aa...@da...>]

We're pleased to announce that version 2.004 of Krang is now available.  This release
 is comprised of several enhancements and bug fixes.  Of note are improvements
to the permissioning system, an improved version of HTML::Template::Expr and
Xinha enhancements.


Notable changes in this release:

    * Added enforcement of permissions at the application and run-mode level rather
      than just at the UI link level.

    * Added support for multiple word action names in admin scheduler.

    * Added new Krang list system primitive, Krang::ElementClass::CheckBoxGroup.

    * Decoupled processing of assets from publish_story() and preview_story()
      allowing for future subclassing into Krang::Publisher::Parallel simply
      redefine _process_assets() and _process_preview_assets()

    * Added a new configuration parameter, Charset.  This may be used to
      control the character set for the Krang user-interface

    * Upgraded HTML::Template::Expr to version 0.06 - now supports chained binary
      operators!

    * Fixed bug in Krang::ElementClass::ListGroup and Krang::ElementClass::RadioGroup       such that associated ListItems will now be automatically added to the DataSet.

    * Fixed bug in Krang::ElementClass::Date where date would change if KDS
      was exported from a machine in one time zone and imported on a machine
      in an earlier time zone.

For more information about Krang, visit the Krang website:

  http://krang.sourceforge.net/

There you can download Krang, view screenshots, read documentation,
join our mailing-lists and access the CVS tree.

Detailed change-log here: http://krang.sf.net/docs/changelog.html

Krang is an Open Source web-publisher / content-management system
designed for large-scale magazine-style websites.  It is a 100% Perl
application using Apache/mod_perl and MySQL, as well as numerous CPAN
modules.

Krang provides a powerful and easy to use story and media editing
environment for magazine editors, as well as a complete template
development environment for web designers. On the back-end, Perl
programmers can customize Krang to control the data entered in the
story editor and add code to drive the templates to build
output. Krang can be enhanced with add-ons containing new skins and
other new features.  Krang easily handles large data sets and can
manage multiple websites in a single installation.

  - the Krang team

----
Aaron Dancygier

[htmltmpl] The HTML::Template::Filter::Dreamweaver code

[Paulsen, B. <BPa...@le...>]

As mentioned previously, I dug up the code and found that I already
submitted it to CPAN a while ago.

=46eel free to download it from there and let me know if you have
questions.

Brian


---------------------------------------------------------------------------=
---
This message is intended only for the personal and confidential use of the =
designated recipient(s) named above.  If you are not the intended recipient=
 =
of this message you are hereby notified that any review, dissemination, =
distribution or copying of this message is strictly prohibited.  This =
communication is for information purposes only and should not be regarded a=
s=
 an offer to sell or as a solicitation of an offer to buy any financial =
product, an official confirmation of any transaction, or as an official =
statement of Lehman Brothers.  Email transmission cannot be guaranteed to b=
e=
 secure or error-free.  Therefore, we do not represent that this informatio=
n=
 is complete or accurate and it should not be relied upon as such.  All =
information is subject to change without notice.

Re: [htmltmpl] Request/question

[Eric F. <er...@dm...>]

Hi,

So I can do that from within the app were I am using HTML::Template.. Ahh, 
I was wondering why everyone always uses croak by default in all of the 
modules I look at..

Thanks!

Eric

At 12:11 PM 3/21/2006, Sam Tregar wrote:
>On Tue, 21 Mar 2006, Eric Frazier wrote:
>
>>Is there any way you could change this line in Template.pm from croak to 
>>confess?
>
>No need to change Template.pm.  Just tell Carp you want verbose
>reports:
>
>   use Carp qw(verbose);
>
>Now all croak()s will have stack traces.
>
>-sam

Re: [htmltmpl] Request/question

[Sam T. <sa...@tr...>]

On Tue, 21 Mar 2006, Eric Frazier wrote:

> Is there any way you could change this line in Template.pm from croak to 
> confess?

No need to change Template.pm.  Just tell Carp you want verbose
reports:

   use Carp qw(verbose);

Now all croak()s will have stack traces.

-sam

[htmltmpl] Request/question

[Eric F. <er...@dm...>]

Hi,

Is there any way you could change this line in Template.pm from croak to 
confess?
I know I might be missing something about croak vs confess, but it is 
always a pain to get this error and then have to try to figure out where it 
came from, whereas with confess it is easy..

  # load in options supplied to new()
   for (my $x = 0; $x <= $#_; $x += 2) {
     defined($_[($x + 1)]) or croak("HTML::Template->new() called with odd 
number of option parameters - should be of the form option => value");
     $options->{lc($_[$x])} = $_[($x + 1)];
   }

Thanks,

Eric

[htmltmpl] ANN: Bricolage 1.10.1 Released

[David W. <da...@ki...>]

Fellow HTML::Templaters,

    Apologies for the previous email, which contained the announcement for
    Bricolage 1.10.0, which was released in January. This is the correct
    announcement for Bricolage 1.10.1, which was released today.

    The Bricolage development team is pleased to announce the release of
    Bricolage 1.10.1. This maintenance release adds a few new features, a
    number of improvements, and many bug fixes. Highlights include new SOAP
    modules, improved packaging support, and smother upgrades from 1.8.x
    versions. The most important changes are:

 New Features

      * Added ContribType, Destination, and Preference SOAP modules.
      [Scott]

      * Added upload and download feature to the template interface, so
      that templates can optionally be uploaded or downloaded. Suggested by
      Chris Sutton. [David]

 Improvements

      * Added installation defaults for FreeBSD that will be used when the
      USE_DEFAULTS environment variable is set to "freebsd" during
      installation. [Rod Taylor]

      * Created two new make targets, "install_files" and "install_db", to
      allow the files to be installed without doing anything with the
      database. Useful for packagers who need to let users manually create
      the database. [Marshall]

      * Updated Bric::ElementAdmin to be conversant in the merging of
      element type sets and elements. That is, it now makes sense in the
      context of Bricolage 1.10. Reported by Ryan O'Toole. [David]

      * The View/Edit link has been restored to stories, media, and
      templates on desks, and the title has been returned to its status as
      a preview link. The URI is no longer a link. [David]

 Bug Fixes

      * The database upgrade script now updates table permissions after
      executing all of the scripts for each version to be upgraded, rather
      than after all of the upgrade scripts have run, thus allowing later
      scripts to properly access tables created by earlier scripts.
      Reported by Nate Perry-Thistle. [David]

      * Fixed some of the 1.8.9 upgrade scripts that run after the 1.9.x
      upgrade scripts to that they are properly compatible. [Nate
      Perry-Thistle, Rod Taylor, and David]

      * Reverting stories and media no longer leads to occasional errors
      such as, "Too many Bric::Biz::Element::Container objects found".
      Reported by Simon Wilcox. [David]

      * Disabled the "sticky" scroll bar in the story profile. It just
      didn't work very well. [David]

      * The "Allow Multiple" checkbox and "Size" field in the "Add New
      Field" section of the element type profile work again. Reported by
      Rod Taylor and Paul Orrock. [David]

      * Multiple select lists no longer break the story and element
      profiles, and can once again have their values properly changed from
      multiple to single to none. Spotted by Rod Taylor. [David]

      * Assets can once again be properly deleted from desks. [David]

      * Cancelling the checkout of an media document now deletes any new
      file that has been uploaded and, if the AUTO_PREVIEW_MEDIA
      bricolage.conf directive is enabled, re-previews the previous version
      of the media file. [David]

      * The autopopulated fields for image element types are now created
      when a new image element type is created. This was made possible by
      moving the selection of media type (Image, Audio, Video, or Other
      Media) to the New Element Type screen, where it is now part of the
      "Content Type" select list, which also includes the Story and
      Subelement options. As a result, one can no longer change the media
      type in an existing element type (not that it ever made sense to do
      so). Reported by Paul Orrock. [David]

      * Failed jobs can once again be cancelled. Reported by Rod Taylor.
      [David]

      * Publishing multiple versions of a document before any of the
      publish jobs for those versions are executed no longer leads to
      publishing errors when the older version is published. Thanks to Rod
      Taylor for the spot! [David]

    For a complete list of the changes in Bricolage 1.10.1, see the changes
    list at
    http://www.bricolage.cc/news/announce/changes/bricolage-1.10.1/. For
    the complete history of ongoing changes in Bricolage, see Bric::Changes
    at http://www.bricolage.cc/docs/current/api/Bric::Changes.

    Download Bricolage 1.10.1 now from the Bricolage Website at
    http://www.bricolage.cc/downloads/, from the SourceForge download page
    at http://sourceforge.net/project/showfiles.php?group_id=34789, or from
    the Kineticode download page at
    http://www.kineticode.com/bricolage/downloads/.

About Bricolage

    Bricolage is a full-featured, enterprise-class content management and
    publishing system. It offers a browser-based interface for ease-of use,
    a full-fledged templating system with complete HTML::Mason,
    HTML::Template, PHP5, and Template Toolkit support for flexibility, and
    many other features. It operates in an Apache/mod_perl environment and
    uses the PostgreSQL RDBMS for its repository. A comprehensive,
    actively-developed open source CMS, Bricolage has been hailed by eWEEK
    as "quite possibly the most capable enterprise-class open-source
    application available."

    Enjoy!

    --The Bricolage Team

[htmltmpl] ANN: Bricolage 1.10.1 Released

[David W. <da...@ki...>]

Fellow HTML::Templaters,

    It is with great pleasure that the Bricolage development team announces
    the release of Bricolage 1.10. The culmination of over 19 months of
    development, version 1.10 represents a significant advance for the
    celebrated open-source content management and publishing system. Here
    are some of the highlights:

 PHP Templating

    Bricolage is the first content management system to support three
    different Perl-based templating architectures (Mason, Template Toolkit,
    and HTML::Template) as well as one in a completely different
    programming language: PHP 5. Bricolage 1.10 adds PHP templating
    support, allowing template developers to use the popular Web
    programming language to formatting their documents for output. This
    functionality is thanks to a killer new technology, known as
    PHP::Interpreter, that loads the PHP 5 interpreter into a Perl 5
    interpreter, and affords transparent access between PHP and Perl code.
    The upshot is that PHP templaters get full access to the entire
    Bricolage API, as well as the ability to use whatever other PHP or Perl
    libraries they wish.

    Our expect is that this development will push Bricolage into new
    environments where PHP developers can make use of the powerful content
    management and publishing system without having to learn a new
    programming language. Furthermore, we hope that PHP::Interpreter will
    act as a bridge between the Perl and PHP communities, such that there
    is a greater exchange of ideas and a greater ability to use each
    other's libraries.

    PHP::Interpreter was developed by OmniTI. PHP::Interpreter and the PHP
    templating support in Bricolage were sponsored by SAPO--Portugal
    Online.

 LDAP Authentication

    Bricolage 1.10 includes support for a pluggable authentication
    architecture, and in addition to its built-in authentication has added
    a module for authentication against an LDAP directory server. This new
    feature is sure to be welcome in busy enterprises that rely on a
    directory server, such as Windows Active Directory
    http://www.microsoft.com/windowsserver2003/technologies/directory/activ
    edirectory/default.mspx, Novel eDirectory
    http://www.novell.com/products/edirectory/, or OpenLDAP
    http://www.openldap.org/. Authentication can be limited to members of a
    directory group, and supports LDAP v.3 and TLS connectivity.
    Contributed by Kineticode.

 Revamped Interface

    Bricolage 1.10 sports a completely revamped browser interface that is
    XHTML compliant and handles all styling via CSS. Yes, our 1999-era
    table-driven interface is officially a thing of the past. The upshot is
    that the interface is much more elegant, easier to skin with your own
    look (by overriding its CSS files), allows search results and editing
    fields to expand and contract with the browser window size, and
    delivers pages as much as 70% smaller than they were before. The new
    interface was Contributed by Marshall Roch.

    A second major new UI feature is the revamped "Bulk Edit" interface.
    Gone is the old "Super Bulk Edit" interface, with the Bulk Edit
    revisions overtaking its functionality. Now you can edit the entire
    contents of a story document, from the top-most element to the
    bottom-most field, in a single textarea field with no reloads.

    The secret to allowing the full-text editing of Bricolage's unique
    hierarchical element structures is Plain Old Documentation, or "POD".
    Subelements are denoted by a new =begin POD tag, and end with a
    matching =end tag. The result is a much more natural editing interface.
    Even related stories and media are supported by new POD tags. We
    believe that this improvement will greatly facilitate the editing
    process, making Bricolage a much more enjoyable product for content
    editors to work with.

    The Bulk Edit revision is complemented by two new additions: diff
    support and a JavaScript-powered "Find and Replace" dialog box. Users
    can now see at a glance the changes between one version of a document
    and another. The changes are shown on a word-by-word basis, with
    additions in green with an underline and deletions in red with a
    strikeout. A similar interface is used to show the differences between
    versions of templates using the traditional "unified diff" format
    rather than word-by word.

    The JavaScript-powered "Find and Replace" dialog box can be used to
    search by strings or regular expressions in a Bulk Edit or Template
    editing environment. Found bits of text can also be replaced or even
    globally replaced. We believe that this powerful new feature, combined
    with the new Bulk Edit interface, makes Bricolage a compelling content
    editing environment.

    The Bulk Edit, diff, and Find and Replace features were contributed by
    Kineticode.

 What's in a Name?

    A somewhat less apparent but no less massive change in Bricolage 1.10
    is a system-wide naming normalization. Now all objects in Bricolage are
    known by the same names, from the UI to the class to the database to
    the SOAP server. Most noticeable in the UI will be the elimination of
    the old "Element Type" object, and the renaming of "Element" objects to
    "Element Types." This change has the benefit of disambiguating element
    types, which define the structure of documents, and elements, which are
    the document parts that contain content. Gone is the confusion between
    element administration and content elements; there are now only element
    types and elements.

    Another example is the renaming of "Data Elements" to "Field Types" and
    "Fields". And in tandem with this change, the storage of field values
    in the database has been denormalized, so that every field value does
    not also store the name and key name of the field. This greatly reduces
    the size of the database, and should make field lookups much faster,
    particularly in formatting templates.

    And while we were going about denormalizing field storage, the data
    types of the database columns were also normalized. Old-style,
    inefficient column types have been dumped in favor of more efficient,
    precise column types. For example, all "NUMERIC" columns, which
    everywhere only contained integers or booleans, have been converted the
    "INTEGER" and "BOOLEAN" data types, as appropriate. This change will
    also be invisible to the everyday Bricolage user, but should enhance
    database performance by optimizing the storage of object attributes.

    And finally, a more visible change: Bricolage 1.10 introduces much more
    flexible URI formats. You can now use many more parts of the cover date
    in the URI, and in whatever format you like. So you could have a format
    of "/%{categories}/%Y-%m-%d/" and end up with the URI
    "/foo/bar/2004-09-22/" if you wanted. Or even "/%{categories}/%Y/%V/"
    to get the week number as part of the URI. You can also include
    document UUIDs, and even your own text, (e.g. foobar in
    /%{categories}/%Y/%m/foobar/%{uuid}/". This enhancement finally allows
    users to almost always be able to replicate legacy URI formats in
    Bricolage, for a seamless upgrade from an older CMS.

What are You Waiting For?

    There are many, many more changes in Bricolage 1.10 that, overall, make
    using it a joy. For a complete list of the changes, see the changes
    list at
    http://www.bricolage.cc/news/announce/changes/bricolage-1.10.0/. For
    the complete history of ongoing changes in Bricolage, see Bric::Changes
    at http://www.bricolage.cc/docs/current/api/Bric::Changes.

    Download Bricolage 1.10.0 now from the Bricolage Website at
    http://www.bricolage.cc/downloads/, from the SourceForge download page
    at http://sourceforge.net/project/showfiles.php?group_id=34789, or from
    the Kineticode download page at
    http://www.kineticode.com/bricolage/downloads/.

About Bricolage

    Bricolage is a full-featured, enterprise-class content management and
    publishing system. It offers a browser-based interface for ease-of use,
    a full-fledged templating system with complete HTML::Mason,
    HTML::Template, PHP5, and Template Toolkit support for flexibility, and
    many other features. It operates in an Apache/mod_perl environment and
    uses the PostgreSQL RDBMS for its repository. A comprehensive,
    actively-developed open source CMS, Bricolage has been hailed by eWEEK
    as "quite possibly the most capable enterprise-class open-source
    application available."

    Enjoy!

    --The Bricolage Team

[htmltmpl] Re: bug fixes/patches/tests

[Mark S. <ma...@su...>]

On 2006-03-16, Tom Heady <tom...@pu...> wrote:
>
> Uh, but which one?
>
> Sourceforge: 
> http://sourceforge.net/tracker/?atid=101075&group_id=1075&func=browse
>
> or cpan:
> http://rt.cpan.org/Public/Dist/Display.html?Name=HTML-Template

Good point. I believe the rt.cpan.org is actively being maintained, and
is the one I would recommend. 

The "BUGS" section of the docs should updated to clarify this.

    Mark

-- 
http://mark.stosberg.com/ 

[htmltmpl] default_escape patches

[Tom H. <tom...@pu...>]

Attached are patches/tests for the following bugs:

* when specifying a default_escape, you can't turn it off

I have fixed ESCAPE=0 and added ESCAPE=NONE to turn off escaping 
altogether for a TMPL_VAR.  The pod was also updated to reflect this change.

Usage:
  <TMPL_VAR foo ESCAPE=NONE>
  <TMPL_VAR foo ESCAPE=0>

* ESCAPE='JS' did not work

I fixed a regex so that single quotes around JS now works.

I have also added a new test file: 04default_escape.t

Tom

Re: [htmltmpl] Re: bug fixes/patches/tests

[Tom H. <tom...@pu...>]

Mark Stosberg wrote:
> On 2006-03-15, Tom Heady <tom...@pu...> wrote:
> 
>>Hello,
>>
>>What would be the most appropriate way to contribute bug 
>>fixes/patches/tests?
>>
...
> 
> Tom,
> 
> I believe using the bug tracking system is the preferred route. I know
> Sam has mentioned that in the past. It's also a visible place for other
> users to find the contribution in case they want to use it before an
> official release happens.
> 
>     Mark
> 

Uh, but which one?

Sourceforge: 
http://sourceforge.net/tracker/?atid=101075&group_id=1075&func=browse

or cpan:
http://rt.cpan.org/Public/Dist/Display.html?Name=HTML-Template

Tom

[htmltmpl] Re: bug fixes/patches/tests

[Mark S. <ma...@su...>]

On 2006-03-15, Tom Heady <tom...@pu...> wrote:
> Hello,
>
> What would be the most appropriate way to contribute bug 
> fixes/patches/tests?
>
> I have a patch and tests that will fix the following:
>
> ------------------------------------
> use HTML::Template;
>
> my $text = q{<TMPL_VAR foo ESCAPE=0>};
> my $template = HTML::Template->new(
>      scalarref => \$text,
>      default_escape => "HTML" );
> $template->param(foo => '<b>this is bold\n');
> print $template->output;
>
> __DATA__
> &lt;b&gt;this is bold\n
>
> ------------------------------------

Tom,

I believe using the bug tracking system is the preferred route. I know
Sam has mentioned that in the past. It's also a visible place for other
users to find the contribution in case they want to use it before an
official release happens.

    Mark

-- 
http://mark.stosberg.com/ 

RE: [htmltmpl] How to manipulate rows in returned resultset

[Kapoor, N. <nis...@xc...>]

> #BUT, if I need to bold the 'search text', I need to iterate through =
the loop
> my @loopData=3D();
> foreach my $rec (@{$data}) {
>  my %loopRow;
>  $rec->{title} =3D~ s/(\b$searchText\b)/<b>$1<\/b>/ig;
>  $loopRow{title} =3D $rec->{title};
>  push @loopData, \%loopRow;
> }
> $template->param(myLoopVar =3D> \@loopData);

If you're dying to do it in one statement I suppose you could do:

   $template->param(myLoopVar =3D>
     [ map { $_->{title} =3D~ s!\b($searchText)\b!<b>$1</b>!ig;
             $_ } @$data ]);

[Nishi]=20
The 'single' statement lets me assign any and all fields in one go - =
without having to worry about individual fields. I was hoping if somehow =
I could populate @{$data} with the 'replaced' text, that would do it. =
Also, there are cases when the resultset contains different fields based =
on different conditions it was called with. So, access to individual =
fields is not always available. Not a big issue, but just wanted to run =
it by the group.

Thanks for the prompt response.
Nishi

[htmltmpl] bug fixes/patches/tests

[Tom H. <tom...@pu...>]

Hello,

What would be the most appropriate way to contribute bug 
fixes/patches/tests?

I have a patch and tests that will fix the following:

------------------------------------
use HTML::Template;

my $text = q{<TMPL_VAR foo ESCAPE=0>};
my $template = HTML::Template->new(
     scalarref => \$text,
     default_escape => "HTML" );
$template->param(foo => '<b>this is bold\n');
print $template->output;

__DATA__
&lt;b&gt;this is bold\n

------------------------------------


Thanks,

Tom

Re: [htmltmpl] How to manipulate rows in returned resultset

[Sam T. <sa...@tr...>]

On Mon, 13 Mar 2006, Kapoor, Nishikant wrote:

> Is there a way I can continue to use the 'single' assignment, and
> still achieve this objective?

You seem to have the right idea here:

> #BUT, if I need to bold the 'search text', I need to iterate through the loop
> my @loopData=();
> foreach my $rec (@{$data}) {
>  my %loopRow;
>  $rec->{title} =~ s/(\b$searchText\b)/<b>$1<\/b>/ig;
>  $loopRow{title} = $rec->{title};
>  push @loopData, \%loopRow;
> }
> $template->param(myLoopVar => \@loopData);

If you're dying to do it in one statement I suppose you could do:

   $template->param(myLoopVar =>
     [ map { $_->{title} =~ s!\b($searchText)\b!<b>$1</b>!ig;
             $_ } @$data ]);

I don't exactly consider that an improvement, personally.

-sam

Re: [htmltmpl] How to manipulate rows in returned resultset

[Roger B. W. <ro...@fi...>]

On Mon, Mar 13, 2006 at 04:41:15PM -0600, Kapoor, Nishikant wrote:
>For populating loops, I get the resultset from db, and assign it directly to the loop_var. This works wonderfully for most cases, except where there is a need to process each row before assigning it to the template. One particular case is where I am retrieving a resultset after a search query, and I need to highlight the text that was searched for.
>
>Is there a way I can continue to use the 'single' assignment, and still achieve this objective?

Well, you could cheat and roll the loop into a map statement...

my $data = &getDataFromDB;
map {$data[$_]{bold}=1} 0..$#{$data};
$template->param(myLoopVar => $data);

and have a

<tmpl_if name=bold><b></tmpl_if>
<tmpl_var name=title escape=html>
<tmpl_if name=bold></b></tmpl_if>

You really shouldn't be putting html markup into template parameters
anyway, without a very good reason.

Roger

[htmltmpl] How to manipulate rows in returned resultset

[Kapoor, N. <nis...@xc...>]

For populating loops, I get the resultset from db, and assign it =
directly to the loop_var. This works wonderfully for most cases, except =
where there is a need to process each row before assigning it to the =
template. One particular case is where I am retrieving a resultset after =
a search query, and I need to highlight the text that was searched for.

Is there a way I can continue to use the 'single' assignment, and still =
achieve this objective?

Thanks,
Nishi

Here is some code snippet to explain it more:

my $data =3D &getDataFromDB;
$template->param(myLoopVar =3D> $data);  #preferred way

#BUT, if I need to bold the 'search text', I need to iterate through the =
loop
my @loopData=3D();
foreach my $rec (@{$data}) {
  my %loopRow;
  $rec->{title} =3D~ s/(\b$searchText\b)/<b>$1<\/b>/ig;
  $loopRow{title} =3D $rec->{title};
  push @loopData, \%loopRow;
}
$template->param(myLoopVar =3D> \@loopData);

[htmltmpl] HTMLTemplate::Expr consumes huge amounts of memory

[Dan H. <dan...@re...>]

Hi

I'm using CGI::Application::Plugin::Anytemplate with =
HTML::Template::Expr
for an application, and it uses a heck of a lot of memory. Because my =
code
is spread over a number of packages, I'll summarise my logic in =
pseudo-code:

sub publish_all {
    foreach my $content_id ($self->get_content_ids()) {
       $self->publish_article($content_id)
    }
}

sub publish_article {
	my ($self, $content_id) =3D @_;
	my %content =3D $self->get_content($content_id);

      my $template =3D  $self->template->load(
         file =3D> $self->get_template($content{template})
      );

	$template->param(%content);
      write_to_file_system(${$template->output});

	1;
}

As I loop through "publish_article", the memory goes up the more times =
it is
called. No globals are defined in my modules.=20

Trying to track down memory usage in Perl can be hard. Running
Devel::FindGlobals, I see that the reported size for
$HTML::Template::Expr::PARSER is 100,039,491 bytes after publishing 500
articles! Assuming that the stats are correct, how can I keep $PARSER =
down
to a minimal size?

BTW. For my test, I invoked the app via the OS command line not through =
the
web server

Dan Horne

Re: [htmltmpl] Validating documents from HTML::Template

[Dan A. <dan...@sn...>]

> H::T won't change anything about your template that isn't a <tmpl_*> tag. So
> something else is changing your <br />. What other tools are you passing 
your
> output through?
> 
> -- 
Aargh!  It's a Firefox thing; view source shows it right, but when I save the 
page to a file for validation, the trailing /'s go away.  Never mind.

Re: [htmltmpl] Validating documents from HTML::Template

[Michael P. <mp...@pl...>]

Dan Axtell wrote:
> Hi,
> 
> I have a report that aims to be and XHTML 1.0 compliant document.  In my 
> template I use <br /> for my linebreaks; but after I fill in various template 
> tags via HTML::Template, they all become <br>, which fails to pass the 
> validator at w3c.org.    Am I missing something obvious here, or is 
> HTML::Template not a good idea for producing XHTML 1.0 compliant documents?

H::T won't change anything about your template that isn't a <tmpl_*> tag. So
something else is changing your <br />. What other tools are you passing your
output through?

-- 
Michael Peters
Developer
Plus Three, LP

[htmltmpl] Validating documents from HTML::Template

[Dan A. <dan...@sn...>]

Hi,

I have a report that aims to be and XHTML 1.0 compliant document.  In my 
template I use <br /> for my linebreaks; but after I fill in various template 
tags via HTML::Template, they all become <br>, which fails to pass the 
validator at w3c.org.    Am I missing something obvious here, or is 
HTML::Template not a good idea for producing XHTML 1.0 compliant documents?

Thanks,
Dan

Re: [htmltmpl] verified something

[Michael P. <mp...@pl...>]

Paulsen, Brian wrote:
> Try using double_file_cache?  I'm thinking this may be more of a
> HTML::Template issue than a mod_perl issue.

Just to let you know. H::T works just fine under mod_perl. If you have any
caching going on, it's probably because you're leaving some globals or package
level vars hanging around.

When H::T caches the template, it doesn't cache the output or the params you
pass in, it just caches the compiled state of the template.

-- 
Michael Peters
Developer
Plus Three, LP

RE: [htmltmpl] verified something

[Paulsen, B. <BPa...@le...>]

Try using double_file_cache=3F  I'm thinking this may be more of a
HTML::Template issue than a mod_perl issue.

What happens if you print the companyID value to STDERR =3F

=20

-----Original Message-----
=46rom: htm...@li...
[mailto:htm...@li...] On Behalf Of
Matthew Boehm
Sent: Tuesday, March 07, 2006 11:19 PM
To: htm...@li...
Subject: [htmltmpl] verified something

ok so I used telnet to grab a few instances of our CGI using
HTML::Template and the caching is still present.

drmac-mac:~ drmac$ telnet www.ourdomain.com 80 Trying 170.86.88.385...
Connected to www.ourdomain.com.
Escape character is '^]'.
GET /pages/sctest/roomlogin_new.cgi=3FcompanyID=3D2

<snip>
<p>Welcome to the <strong>omNovia</strong> Conference Center </snip>


drmac-mac:~ drmac$ telnet www.ourdomain.com 80 Trying 170.86.88.385...
Connected to www.ourdomain.com.
Escape character is '^]'.
GET /pages/sctest/roomlogin_new.cgi=3FcompanyID=3D1

<snip>
<p>Welcome to the <strong>Traders International</strong> Conference
Center </snip>

drmac-mac:~ drmac$ telnet www.ourdomain.com 80 Trying 170.86.88.385...
Connected to www.ourdomain.com.
Escape character is '^]'.
GET /pages/sctest/roomlogin_new.cgi=3FcompanyID=3D45

<snip>
<p>Welcome to the <strong>omNovia</strong> Conference Center </snip>


Any ideas=3F I should mention that roomlogin_new.cgi does use strict while
our old CGI does not. And the old CGI doesn't have this problem even
though both run under the same mod-perl enabled dir.

Thanks,
Matthew


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting
language that extends applications into web and mobile media. Attend the
live webcast and join the prime developer group breaking into this new
coding territory!
http://sel.as-us.falkag.net/sel=3Fcmd=3Dlnk&kid=3D110944&bid=3D241720&dat=
=3D121642
_______________________________________________
Html-template-users mailing list
Htm...@li...
https://lists.sourceforge.net/lists/listinfo/html-template-users



---------------------------------------------------------------------------=
---
This message is intended only for the personal and confidential use of the =
designated recipient(s) named above.  If you are not the intended recipient=
 =
of this message you are hereby notified that any review, dissemination, =
distribution or copying of this message is strictly prohibited.  This =
communication is for information purposes only and should not be regarded a=
s=
 an offer to sell or as a solicitation of an offer to buy any financial =
product, an official confirmation of any transaction, or as an official =
statement of Lehman Brothers.  Email transmission cannot be guaranteed to b=
e=
 secure or error-free.  Therefore, we do not represent that this informatio=
n=
 is complete or accurate and it should not be relied upon as such.  All =
information is subject to change without notice.