# Data mining a Discourse site for links

**URL:** https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943
**Category:** Feature
**Created:** [26 באוקטובר,‏ 2015,‏ 10:59pm UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943 "2015-10-26T22:59:39Z")
**Posts on this page:** 11
**Page:** 1

<div class="post-metadata">

### Author: ![Jeffrey\_L\_Frusha](https://avatars.discourse-cdn.com/v4/letter/j/6f9a4e/32.png) [@Jeffrey\_L\_Frusha](https://meta.discourse.org/u/Jeffrey_L_Frusha)
#### Post date: [26 באוקטובר,‏ 2015,‏ 10:59pm UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/1 "2015-10-26T22:59:39Z")

</div>

Could a plug-in, like the feature where I can archive and download my posts, be adapted to archive and download the links from all posts in a Discourse forum, rather than reading all posts and digging each one out, manually?

This would allow someone to categorize and create a library of those links, for a ‘wiki’…

The advanced search ‘Google’ only brought up some 752 results (which seems low), with:

site:discuss.seasteading.org http

and 732 results with:

site:discuss.seasteading.org www

The question was brought up by another forum member. I’m just another forum user, not a mod, or anything.

---

<div class="post-metadata">

### Author: ![codinghorror](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/codinghorror/32/110067_2.png) [@codinghorror](https://meta.discourse.org/u/codinghorror)
#### Post date: [26 באוקטובר,‏ 2015,‏ 11:12pm UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/2 "2015-10-26T23:12:32Z")

</div>

Just download your posts from your user page and extract the links using a regex in an editor, or write a program to do it.

The latter is beyond the scope of what we do here at meta – we can’t teach you programming, or regular expressions.

---

<div class="post-metadata">

### Author: ![Mittineague](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/mittineague/32/114259_2.png) [@Mittineague](https://meta.discourse.org/u/Mittineague)
#### Post date: [26 באוקטובר,‏ 2015,‏ 11:20pm UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/3 "2015-10-26T23:20:41Z")

</div>

> [@codinghorror](#):
>
> Just download your posts from your user page …

Might it be possible to use the [Data Explorer](https://meta.discourse.org/t/32566?silent=true) plugin to query the Topic Links table?

---

<div class="post-metadata">

### Author: ![Jeffrey\_L\_Frusha](https://avatars.discourse-cdn.com/v4/letter/j/6f9a4e/32.png) [@Jeffrey\_L\_Frusha](https://meta.discourse.org/u/Jeffrey_L_Frusha)
#### Post date: [26 באוקטובר,‏ 2015,‏ 11:23pm UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/4 "2015-10-26T23:23:21Z")

</div>

> [@codinghorror](#):
>
> Just download your posts from your user page and extract the links using a regex in an editor, or write a program to do it.

If I wanted mine and mine alone, I could do that, however, the goal is to get all, or as many as possible into one download, to be categorized as a library of sorts.

---

<div class="post-metadata">

### Author: ![codinghorror](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/codinghorror/32/110067_2.png) [@codinghorror](https://meta.discourse.org/u/codinghorror)
#### Post date: [27 באוקטובר,‏ 2015,‏ 12:19am UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/5 "2015-10-27T00:19:44Z")

</div>

Perhaps @riking could weigh in on this?

---

<div class="post-metadata">

### Author: ![riking](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/riking/32/170938_2.png) [@riking](https://meta.discourse.org/u/riking)
#### Post date: [27 באוקטובר,‏ 2015,‏ 3:41am UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/6 "2015-10-27T03:41:39Z")

</div>

```plaintext
{"query":{"id":10,"sql":"SELECT url, count(*)\nFROM topic_links\nGROUP BY url\nORDER BY count DESC","name":"All outbound links","description":"Use the \"Download Results\" button to get it as a file.","param_info":[]}}

```

Import that, Run Query, then Click the “Download Results CSV” button.

---

<div class="post-metadata">

### Author: ![Jeffrey\_L\_Frusha](https://avatars.discourse-cdn.com/v4/letter/j/6f9a4e/32.png) [@Jeffrey\_L\_Frusha](https://meta.discourse.org/u/Jeffrey_L_Frusha)
#### Post date: [27 באוקטובר,‏ 2015,‏ 4:36am UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/7 "2015-10-27T04:36:10Z")

</div>

I assume that’s something only admin can do. Thank you.

@kensims Would you be so kind as to try this? KatOn Tri would certainly like a copy, as would I. As I’m up to it, I’ll try to go through and categorize, so the wiki can be updated.

---

<div class="post-metadata">

### Author: ![riking](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/riking/32/170938_2.png) [@riking](https://meta.discourse.org/u/riking)
#### Post date: [27 באוקטובר,‏ 2015,‏ 9:26am UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/8 "2015-10-27T09:26:19Z")

</div>

Yes, you might want to filter it to exclude links from PMs…

---

<div class="post-metadata">

### Author: ![Jeffrey\_L\_Frusha](https://avatars.discourse-cdn.com/v4/letter/j/6f9a4e/32.png) [@Jeffrey\_L\_Frusha](https://meta.discourse.org/u/Jeffrey_L_Frusha)
#### Post date: [27 באוקטובר,‏ 2015,‏ 10:55am UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/9 "2015-10-27T10:55:01Z")

</div>

Could this be done as a feature? Maybe a plug-in capable of archiving and downloading the links from a topic, or category, similar to the way I can archive and download my posts?

---

<div class="post-metadata">

### Author: ![codinghorror](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/codinghorror/32/110067_2.png) [@codinghorror](https://meta.discourse.org/u/codinghorror)
#### Post date: [27 באוקטובר,‏ 2015,‏ 6:49pm UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/10 "2015-10-27T18:49:43Z")

</div>

No, it cannot. Not by us anyway.

---

<div class="post-metadata">

### Author: ![chapoi](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/chapoi/32/537252_2.png) [@chapoi](https://meta.discourse.org/u/chapoi)
#### Post date: [4 בדצמבר,‏ 2025,‏ 11:37am UTC](https://meta.discourse.org/t/data-mining-a-discourse-site-for-links/34943/11 "2025-12-04T11:37:04Z")

</div>


