Symphony.

Hello,

I’m hitting a strange issue (or some massive overlook?).

I have DS1 (pulling form a section that uses Subsection Manager in one its fields) and DS2 (a normal section and its entries, which I am trying to filter).

DS1 Parameter Output is set to a field called “pages”, which is a Subsection Manager field, linking to a section called Pages. The debugger shows the Parameter Output correctly returning the linked page IDs (i.e: 115, 116, 121, 124).

When using this $ds-page-groups to filter the results of DS2, which pulls from the section called “Pages”, I only get the first (i.e: 115) in the XML output.

The only mention this issue I found is this

Any ideas?

What does your datasource page-groups look like? Is there a filter on that datasource? Would you mind taking screenshots of the datasources editor editor page on this particular datasource?

Sure, here’s a combo screenshot: Higher Res version

A print_r($this->$dsParamFILTERS) gives me this:

Array ( [id] => 111\, 113\, 114\, 116,115\, 117\, 122\, 123,124\, 125 )

You’ll see that’s some irregularity in that pattern: the last item for each group doesn’t have a trailing \.

Groups:
111, 113, 114, 116
115, 117, 122, 123
124, 125

What I am getting in the chained DS results is only the first item in each group.

This is a known Subsection Manager issue which I was not able to fix yet. Looking at your example it seems that it is related to Symphony’s comma escaping.

What kind of server are you using? Localhost or external? Linux or Windows?

External, Dreamhost.

Here’s a phpinfo() for you.

Can a possible solution be to use stripslashes() in the datasource php?

UPDATE

datasource hack seems to be doing its job fine:

public function grab(&$param_pool=NULL){

  function wrkStriplashes(&$element){ 
    $element = stripslashes($element); 
  }

  array_walk($this->dsParamFILTERS, 'wrkStripslashes');

I’m not sure. I’ll check the Symphony code to see if this is really a Subsection Manager issue or a Symphony bug.

FYI, my UPDATE came after Nils’ above post.

@Nils, thanks for getting to look at this. It’d still be nice to get this sorted at the root of the problem.

tmsinz, does the change mentioned in the first post of the bug report solve your issue as well?

Just tried it: Yes it does.