Module development and code questions

Hi all

I have developed a new module in drupal for my site .It works fine untill any user login or logout of the site.
On any login or logout the screen went blank n jus shows this warning:

warning: Cannot add header information - headers already sent by (output started at /var/www/html/drupal/modules/user_statistics.module:206) in /var/www/html/drupal/includes/common.inc on line 155.

by the way user_statistics is the name of my new module.
I think there is some problem in redirection .I dunno how to solve it Please help me.I need urgetnt help.

Hi all

I have developed a new module in drupal for my site .It works fine untill any user login or logout of the site.
On any login or logout the screen went blank n jus shows this warning:

warning: Cannot add header information - headers already sent by (output started at /var/www/html/drupal/modules/user_statistics.module:206) in /var/www/html/drupal/includes/common.inc on line 155.

I think there is some problem in redirection .I dunno how to solve it Please help me.I need urgetnt help.

Thanx in advance :)

I'm having several problems with the image module from head.

1. Going to admin/settings/image gives a blank page
2. When I create an image node, there's no place to enter a title or description.
3. The actual image never appears unless I have it listed and click on the name in the listing when I view the node.

Does anyone have a straight answer for why the 'taxonomy_select_nodes' function is limited to using a persistent variable as its means for deciding how many nodes to return? I have a cron process that needs to examine the nodes specific term, and while I don't need to examine ALL the nodes each time, I do need to examine a good chunk of them. If I let cron set the persistent 'feed_default_items' variable (in Drupal HEAD) or if I modify the persistent 'default_nodes_main' variable (in Drupal HEAD/4.6), then there's the potential for my users to occasionally wind up getting hit with a page that has WAY too many nodes on it.

So, since modifying the persistent variables even temporarily is a REALLY BAD solution, I only have two other options. I either have to manually query against the taxonomy tables, or I have to patch the 'taxonomy_select_nodes' function. I really don't like the former option. As for patching, I don't want to apply a patch unless the patch is planned for inclusion in the next official release.

Could we possibly see the addition of "from" and "count" parameters to this function? I really don't want to have to re-patch this function again and again anytime I update my site to a newer version of Drupal. Unless someone has a good reason NOT to put "from" and "count" parameters into the function, this is a change I would really like to see happen in the

Hello all,

A topic that makes people cry, especially if they have used Taxonomy_Access.module previously.

I've just been looking over the various functions that the taxonomy_access.module does and wondered if it would be possible to simplify this. rather than jumping in and developing a new module to test this I wondered if people could point out problems beforehand.

Here's the idea.

Everytime a node is about to be rendered for viewing, the new module would check the associated taxonomy term ids via the hook_nodeapi() function.

If the current viewing users role doesn't match the permission table for any of the taxonomies associated with the node in question, redirect the user to the Drupal 403 error page.

Special case

If the node path matches that of the 404 or 403 error nodes defined in the main admin->settings then always return true to prevent server 500 errors due to 403 caused by ErrorDocument not being available to Apache.

Patches

I think the above idea would work, and be much lighter to implement than the current taxonomy_access.module
It would still need the taxonomy.module patch in order to not return taxonomies that a user role should not be seeing.

I don't see the need to store view/create/update permissions for every node. Which is what it looks like taxonomy_access.module is doing everytime a node is touched (through the taxonomy_access_nodeapi() call).