Searching for symbols and emojis

This topic describes how to make symbols and emojis searchable in a dtSearch index and how to use regular expressions to search for certain symbols reserved as search operators, such as the % sign.

Searching for symbols

There are two ways to make a character searchable, depending on whether the hexadecimal character code is less than 007F or greater than 0080.

Also refer to:

• Symbols quick reference guide
• dtSearch Unicode values for Special Characters article on Community

For hexadecimal characters less than 007F

To search for symbols with hexadecimal characters less than 007F, perform the following steps:

1. Create a dtSearch index.
2. Update the Alphabet file to include the % sign as an indexed character. For more information on the Alphabet list, see Alphabet list.
   1. Enter the following under [Letters] // Original letter, lower case, upper case, unaccented:
      [Space] [%] [Space] [%] [Space] [%] [Space] [%]
      You must have a leading space. You cannot have a trailing space.
      
   2. Delete the % sign from the [Ignore] section. Take care not to remove the gray boxes, which represent non-printable characters. Removing them may break the index.
      Before removing the %:
      
      After removing the %:
      
   3. Repeat these steps for any other symbols that you need to include in an index. Delete the appropriate symbols from the [Spaces], [Ignore], or [Hyphens] sections. Please note, if you edit the these sections, you must keep the leading space. Removal of leading spaces will cause errors in Relativity.
3. Perform a full build of the dtSearch index.

Example: Searching for percentage (%) terms in documents

An example would be searching for documents that contain the term 75%. Since the % symbol is a reserved dtSearch operator used for fuzzy searching, dtSearch interprets % as a search operator rather than a literal character, even after you have indexed it in the Alphabet file. This is why you must use the regular expression syntax (e.g., "##75\u0025") to search for the literal % symbol. Typing 75% directly into the search box will trigger a fuzzy search instead of looking for the literal text "75%," which may return false hits such as $75. See Fuzzy searching for more information on using the percentage sign for fuzzy searches. To search for documents that contain the term 75%, do the following:

1. Update the Alphabet file to include the % symbol as an indexed character. See Alphabet list for more information.
2. Delete the % sign from the [Ignore] section.
3. Perform a build of the dtSearch index. Enter the following regular expression in your search box, ensuring you select the proper dtSearch index:

1
"##75\u0025"

To break down this regular expression into its parts:

• ## signals to Relativity to treat the string as a regular expression.
• 75 indicates the percentage amount for the search.
• \u indicates a search using a character's Unicode value which follows in the next four characters.
  
• 0025 indicates the hexadecimal Unicode value for the % symbol.

For hexadecimal characters greater than 0080

For characters outside of the ASCII code range, characters greater than 0080, such as § and £, you cannot make them searchable by adding them to the [Letters] section. Instead, create an AdditionalLetters section at the bottom (very last line) of the Alphabet file and insert the characters' Unicode value. For example: AdditionalLetters = 20AC 00A3 00A5. This would make the Euro symbol (20AC), British Pound (00A3), and Japanese Yen (00A5) searchable.

Be sure to remove the "\u" before the four character value in the Unicode Entity field when adding to your Alphabet file.

Commonly requested searchable symbols

Below is a list of Unicode hex values necessary to make some commonly requested symbols searchable:

• 00A3 — £ (British Pound)
• 00A5 — ¥ (Yen half-width)
• FFE5 — ￥ (Yen full-width)
• 00A7 — § (Section break)
• 00A9 — © (Copyright)
• 00AE — ® (Registered Trademark)
• 00B6 — ¶ (Paragraph marker)
• 20AC — € (Euro)

Searching for emojis

Relativity supports searching for emojis with dtSearch indexes.

Note the following about emojis:

• Emoji search works with file formats that store emojis as Unicode characters, such as Word documents and emails.
  • File formats that store emojis in other formats, such as storing the emoji name, are currently not supported.
  • Some examples include messages from Slack and Teams.

To run emoji searches:

1. Enter an emoji in the search box.
   You can copy and paste an existing emoji, or use the appropriate shortcut keys for your operating system. For example, in Windows, click the Windows Key + period (.) to open the emoji sampler.
   
2. Click Enter.

Your search results appear in the document list.

Additional emojis

If you need to search for emojis specific to your organization, you can add them to the bottom of your Alphabet file using "AdditionalLetters32." After adding them to your Alphabet file, you must run a full index build.

If you currently have codes listed for AdditionalLetters32, add any new codes to the end of the existing listing. For example, if you currently have "AdditionalLetters32 = 231A 231B" and want to add "23EA" and "23EB," then add them after 231B like this: "AdditionalLetters32 = 231A 231B 23EA 23EB". Do not add another "AdditionalLetters 32" line.

In the Alphabet file, add the unicode for those emojis.

To get started, use the following unicode list, which covers a large variety of emojis you may encounter:

AdditionalLetters32 = 231A 231B 23E9 23EA 23EB 23EC 23F0 23F3 25FD 25FE 2614 2615 2648 2649 264A 264B 264C 264D 264E 264F 2650 2651 2652 2653 267F 2693 26A1 26AA 26AB 26BD 26BE 26C4 26C5 26CE 26D4 26EA 26F2 26F3 26F5 26FA 26FD 2705 270A 270B 2728 274C 274E 2753 2754 2755 2757 2795 2796 2797 27B0 27BF 2B1B 2B1C 2B50 2B55 0023 FE0F 20E3 00A9 00AE 203C 2049 2122 2194 2195 2196 2197 2198 2199 21A9 21AA 2328 23CF 23ED 23EE 23EF 23F1 23F2 23F8 23F9 23FA 24C2 25AA 25AB 25B6 25C0 25FB 25FC 2600 2601 2602 2603 2604 260E 2611 2618 261D 2620 2622 2623 2626 262A 262E 262F 2638 2639 263A 2640 2642 265F 2660 2663 2665 2666 2668 267B 267E 2692 2694 2695 2696 2697 2699 269B 269C 26A0 26A7 26B0 26B1 26C8 26CF 26D1 26D3 26E9 26F0 26F1 26F4 26F7 26F8 26F9 200D 2702 2708 2709 270C 270D 270F 2712 2714 2716 271D 2721 2733 2734 2744 2747 2763 2764 27A1 2934 2935 2B05 2B06 2B07 3030 3297 3299

For additional Unicode values, use the following Unicode converter: https://www.unicode.org/charts/

Be sure to remove the "U" before the four character value of the Unicode when adding it to the Alphabet file.