Question on RSS Aggregator plugin

[faldo29]

Greetings.

I have a question about the Aggregator--does it require MySQL?

I'm using PGSql actually. Every time I enter a Feed Name and the URI of the feed, nothing seems to happen. I just get taken back to the RSS Aggregator page. Did I miss a step somewhere?

Running s9y 1.04 and RSS Aggregator 0.14.

And when I run http://myblog.com/index.php?/plugin/aggregator, I get a message saying Finish Planetarium. Please explain what that is.

Your help is greatly appreciated.

[garvinhicking]

Hi!

I believe there were some issues in the aggregator plugin with pgsql, but I think the important ones should have been fixed.

Can you check if you have tables called serendipity_aggregator_feeds, serendipity_aggregator_md5, serendipity_aggregator_feedcat and if they have content?

And when I run http://myblog.com/index.php?/plugin/aggregator, I get a message saying Finish Planetarium. Please explain what that is.

That URL should tell you which feeds the aggregator fetches. Since an aggregator is commonly also called "Planet" it tells you that the "Planetarium" finishes.

HTH,
Garvin

[faldo29]

Hi!
Can you check if you have tables called serendipity_aggregator_feeds, serendipity_aggregator_md5, serendipity_aggregator_feedcat and if they have content?

Hi there,

There is no serendipity_aggregator_md5, but there are entries for the other two. In my case I decided to use my domain as the prefix so it's myprefix_aggregator_feedcat, etc. That shouldn't matter though, correct?

Thank you for your attention.

[faldo29]

Hi,

Should I insert the aggregator_md5 into the table? Can you tell me how many columns, and what other values should be created?

[garvinhicking]

Hi!

Yes, creating that table should proove helpful. If you like, maybe you could inspect the aggregator.php file and search for "CREATE TABLE", which should show you the columns of all required tables?

If that is too complicated, tell me and I'll look it up for you.

Regards,
Garvin

[faldo29]

Hi!

If that is too complicated, tell me and I'll look it up for you.

Hi,

I suppose these are the columns that should be created:

entryid {AUTOINCREMENT} {PRIMARY},
md5 varchar(32) NOT NULL default '',
timestamp int(10) {UNSIGNED} default null,
key md5_idx (md5),
key timestamp_idx (timestamp)

The first three are okay. But I'm a little confused with the last two columns, in fact I'm not sure if they are columns at all. Can you set me straight on the last two?

Thank you.

[garvinhicking]

Hi!

The last two ("key") are simply indexes on the "md5" colum and the "timestamp" column. They just speed up things, but they are not vital to make things work.

Best regards,
Garvin

[faldo29]

The last two ("key") are simply indexes on the "md5" colum and the "timestamp" column. They just speed up things, but they are not vital to make things work.

Hi,

I'll keep that in mind and see what I can do about the index thing. Speeding things up is never a bad thing.

Anyway I got it working now. After creating the md5 tables, it still didn't work actually. So I traced it to the aggregator_feeds table, which didn't create the categoryid column. After I created that, it worked.

The problem now is that the articles being imported are not being put into the categories I defined in the RSS Aggregator backend setup. If you have any info on what I should do to remedy this, just send it my way please.

In summary, it seems this plugin still has issues with Postgres and maybe due for some fixing.

[garvinhicking]

Hi!

Sadly I currently have no pgsql installation available to test it with. But maybe it's related to not finding/setting the autoincrement keys automatically through pgsql sequences?

Since there is another bug in the aggregator with mysql, I'll try to take care of that next week!

Best regards,
Garvin

[faldo29]

Sadly I currently have no pgsql installation available to test it with. But maybe it's related to not finding/setting the autoincrement keys automatically through pgsql sequences?

Hi Garv,

My webhosts's Postgres is 7.3.4 so I don't know if this is a bug-free version or not. I only take what they give me.

Anyway, a couple other issues with the Aggregator--I have to say that the output is somewhat "aesthetically challenged" but I could be wrong. Let me explain.

By using the patched tpl, even the original blog entries have that little "ORIGINAL" link appended to it and it basically links back to the blog's home page. Also, the articles that are imported only have a few sentences in them and are cut in mid sentence. So here are my questions:

1. Is it possible to configure the entries.tpl so that it recognizes original blog entries from RSS entries? This is so that original entries do not get the "ORIGINAL" link appended to it, while RSS entries do. Also, it would be nice to append three periods (...) to the end of the last word of the RSS entry so that even if it's cut in mid sentence, it gets a semblance of logic with the three periods. In fact it would be nicer if the "ORIGINAL" link is appended after the three periods. One could change the ORIGINAL to something like "Read full article here" or something like that.

2. If the answer to #1 is yes, are there variables that differentiate between blog entries from RSS entries, eg, {$entry.properties.ep_aggregator_feedname} etc. If there is a previous thread that has already dealt with this, please point me in that direction.

Thanks again!

[garvinhicking]

Hi!

Usually the plugin was meant for blogs that will have no "own" content. But you can adapt the entries.tpl easily, and wrap a IF-clause around the originallink:

Code: Select all

{if $entry.properties.ep_aggregator_feedname}
Originallink: Blabla
[/if}

HTH,
Garvin

[faldo29]

Code: Select all

{if $entry.properties.ep_aggregator_feedname}
Originallink: Blabla
[/if}

HTH,
Garvin[/quote]

Hi Garv,

So far I've been able to make the entries.tpl differentiate an original blog entry from a RSS aggregated one. But I've run into a wall on differentiating if the {$entry.body} is original or RSS. I guess I'd have to define something like a {if $entry.is_RSSaggregated} but I don't know the exact variable that defines the RSS aggregated entry. My uneducated guess was {if $entry.is_ep_aggregator} but that didn't work. Can you give me an assist on this one? :-)

Thanks!

[garvinhicking]

Hi!

Simply use the IF-query from above. Only for RSS-Feeds this variable is set!

Best regards,
Garvin

[faldo29]

Simply use the IF-query from above. Only for RSS-Feeds this variable is set!

Typical of me to complicate matters when the simple solution was right in front of me

Many thanks.

[faldo29]

Greetings.

I have two questions regarding the aggregator:

1. When it imports articles that include characters like a quotation mark ("), it replaces with '. Is this my browser's problem or is there something that needs to be changed in the script?

2. Is it possible to increase the length of the body of the feed. I notice that the aggregated article will show only the first 20 or so words before it gets cut and I'd like to increase this a bit if possible so that the "preview" can be more coherent.

Thank you.