West Midlands | 26 March SDC | Iswat Bello | Sprint 1 | Purple Forest/bug report/Hashtag link doesn't work

[Iswanna]

Learners, PR Template

Self checklist

• I have titled my PR with Region | Cohort | FirstName LastName | Sprint | Assignment Title
• My changes meet the requirements of the task
• I have tested my changes
• My changes follow the style guide

Changelist

This PR fixes a bug where hashtags located in the middle of a sentence were incorrectly linked because the application captured trailing spaces and punctuation as part of the hashtag name.

1. Reproduction Steps

To confirm the bug, I performed the following:

1. Loaded the database with sample data.
2. From the UI, I located the bloom from user Swiz: "Let's get some #SwizBiz love!!"
3. Clicked on the #SwizBiz link.
4. Observation: The page loaded but was completely empty.
5. Network Trace: I checked the Chrome DevTools Network tab and found that the request URL was: http://localhost:3000/hashtag/SwizBiz%2520love!!.
6. Diagnosis: The application was trying to search for a hashtag that included the trailing space and the next word ("love!!").

2. The Problem

Hashtags located in the middle of a sentence were being "over-captured." Because the link included spaces and punctuation, the backend could not find a match in the database, resulting in an empty results page.

3. The Discovery

I investigated frontend/components/bloom.mjs and found the _formatHashtags function.

• The Bug: It used a Regex pattern: /\B#[^#]+/g.
• The Logic Error: The [^#]+ part tells the browser to "capture every character that is NOT a hash." Since spaces and exclamation marks are not hashes, the browser kept grabbing them until the end of the string.

4. The Fix

I updated the Regex pattern in frontend/components/bloom.mjs to:
/\B#\w+/g

5. The Logic

By switching to \w+, the Regex is now restricted to "Word Characters" (letters, numbers, and underscores).

• Result: As soon as the browser hits a space, a period, or an exclamation mark, it realizes that character is not a "word character" and stops the hashtag capture immediately.

6. Verification

After the fix, I returned to the timeline and clicked the same link.

• Observation: The search now correctly displays all blooms containing #SwizBiz.
• Network Trace: Confirmed the request is now clean: /hashtag/SwizBiz.

[nedssoft]

This is a beautiful piece of debugging, Iswat. 👏 Tracing the empty page back to the over-encoded URL (%2520), realising [^#]+ was greedily swallowing the trailing space and punctuation, then switching to \w+ so it stops at the first non-word character — that's a really clear grasp of how regex character classes work. Verified cleanly, too. Lovely work — nothing to change here.

[Iswanna]

This is a beautiful piece of debugging, Iswat. 👏 Tracing the empty page back to the over-encoded URL (%2520), realising [^#]+ was greedily swallowing the trailing space and punctuation, then switching to \w+ so it stops at the first non-word character — that's a really clear grasp of how regex character classes work. Verified cleanly, too. Lovely work — nothing to change here.

Thank you so much @nedssoft, for the feedback.