PHP Developer Needed for Fixing Arabic Search Issue
Budget: $8 – $15 USD
Problem Description:
We have a search functionality in our Arabic poetry database that currently does not handle searches consistently with respect to Arabic diacritical marks (Tashkeel). Specifically, the issue arises with the way search results are returned based on whether the input includes diacritical marks or not.
Current Behavior:
When diacritics are included in the search input: If a user searches using a word with diacritical marks, such as "شِيمَتُكَ", the search results correctly return entries containing the word with and without diacritical marks (e.g., both "شِيمَتُكَ" and "شيمتك").
When diacritics are not included in the search input: Conversely, if the search input is without diacritical marks, like "شيمتك", the results only include entries that match exactly, ignoring any entries where the word appears with diacritical marks (e.g., it misses "شِيمَتُكَ").
Desired Behavior:
The search functionality should return all relevant entries containing the searched word, regardless of whether the input or the database entries include diacritical marks. For instance, searching for "شيمتك" should return entries for both "شيمتك" and "شِيمَتُكَ", and vice versa.
----------------
Examples to Illustrate the Issue:
1-
Input: "شيمتك" (without diacritics)
Expected Results: Entries containing "شيمتك", "شِيمَتُكَ"
Actual Results: Only entries containing "شيمتك"
---------------
2-
Input: "شِيمَتُكَ" (with diacritics)
Expected Results: Entries containing "شيمتك", "شِيمَتُكَ"
Actual Results: Entries containing both "شيمتك" and "شِيمَتُكَ"
The issue lies in the example No. 1
==================
Objective:
Adjust the search mechanism to normalize both the input and the stored data, stripping diacritical marks before the search operation. This normalization should ensure that the search is primarily based on the root characters of the words, improving consistency and inclusivity of search results.
We have a search functionality in our Arabic poetry database that currently does not handle searches consistently with respect to Arabic diacritical marks (Tashkeel). Specifically, the issue arises with the way search results are returned based on whether the input includes diacritical marks or not.
Current Behavior:
When diacritics are included in the search input: If a user searches using a word with diacritical marks, such as "شِيمَتُكَ", the search results correctly return entries containing the word with and without diacritical marks (e.g., both "شِيمَتُكَ" and "شيمتك").
When diacritics are not included in the search input: Conversely, if the search input is without diacritical marks, like "شيمتك", the results only include entries that match exactly, ignoring any entries where the word appears with diacritical marks (e.g., it misses "شِيمَتُكَ").
Desired Behavior:
The search functionality should return all relevant entries containing the searched word, regardless of whether the input or the database entries include diacritical marks. For instance, searching for "شيمتك" should return entries for both "شيمتك" and "شِيمَتُكَ", and vice versa.
----------------
Examples to Illustrate the Issue:
1-
Input: "شيمتك" (without diacritics)
Expected Results: Entries containing "شيمتك", "شِيمَتُكَ"
Actual Results: Only entries containing "شيمتك"
---------------
2-
Input: "شِيمَتُكَ" (with diacritics)
Expected Results: Entries containing "شيمتك", "شِيمَتُكَ"
Actual Results: Entries containing both "شيمتك" and "شِيمَتُكَ"
The issue lies in the example No. 1
==================
Objective:
Adjust the search mechanism to normalize both the input and the stored data, stripping diacritical marks before the search operation. This normalization should ensure that the search is primarily based on the root characters of the words, improving consistency and inclusivity of search results.