Full text searches (basic query)
The Full Text or Simple Search bar uses a "Google-like" toolbar to return full-text search results. FileHold has the ability to recognize, index text and allows users to search content of all your favourite document types including PDF, Microsoft Office, Zip and many more. When searching using the full text search, the system searches both the metadata associated and the content of the document for matches. Full text search is particularly useful for general searches of files stored in a records management system. For example, searching the system for all files that make reference to Sarbanes Oxley.
Types of simple searches:
- All words – By default an ‘and’ is placed between any of the words typed in the search bar. For example: banana pear apple pie returns any documents with the words banana, pear, apple and pie.
- Exact phrases – Enter exact words in quotes. For example: banana pear “apple pie” returns any documents with the words banana pear and the exact phrase “apple pie”.
- Any words – Type OR between words. For example: banana or pear “apple pie” returns any documents with the words banana or pear and the exact phrase “apple pie”.
- No words – Put a minus sign (-) before the words that you don’t want. For example: banana -pear returns any documents with the word banana but not pear.
There are three locations where you will find the simple search screen:
- Click Search in the Library tree. The simple search form that appears allows users to search the contents of the entire documents management system using the full text search engine.
- The Search field located in the top right-hand corner of the FileHold Desktop Application user interface:
- In a folder view or Search in the library tree of the Web Client:
To do a simple (full text) search
- In the simple search bar, enter the search term and click Search. You can use the various search request types in your search. The search results are displayed.
Search results relevance ranking
The relevance value (expressed as a percentage) is an approximation of how close the document is to the search criteria. The value is calculated by the full text search engine. It uses a complex algorithm to determine document version relevance that is guided by the search configuration prioritization of metadata versus document file contents. For example, words that appear near the beginning of a document are more relevant than the same words appearing towards the end of the document.
It is possible that the most relevant document does not have a value of 100%. This happens because any full text criteria are searched first. Database criteria are applied to the full text results to produce the final list. These database criteria can be explicitly provided by the user or implicitly provided by the search settings. For example, all document versions are searched when the full text criteria is applied. However, the user may have selected to exclude historical versions from the results. If the top result from the full text criteria is not a current version the highest relevance value will not be 100%.
Indexing files for full text search
Documents are indexed by the full text search engine on a schedule as set by the records management system’s library administrator. This schedule can be set frequently (every minute) or infrequently (once a week and anywhere in between) as required to support the search activities of users. Documents will not be able to be located by the search engine until the file is indexed. By default this is set to every minute via FileHold server scheduled task, and we do not recommend that this be changed.
The following user activities trigger the full text index to be updated for a document:
- New document(s) or a new version of a document is added to the records management system (both the documents Metadata values and the document content will be indexed).
- Metadata values that are associated with a document are edited (the new Metadata values are added, the old values are removed).
- A document is deleted from the document management system (both Metadata Values and content index entries associated with the document shall be removed)
System Administrators may customize the search engine settings.