Why do some searches on Chinese characters match '#'?

Question

Help in Agent Ransack says: "Agent Ransack is built with the Windows Unicode libraries and therefore supports searching of the file system with full Unicode support."

Now, I need to search for some chinese characters, like: ??

Unfortunately, it matches with "##" in any kind of document. How can I solve that?

Answer 1

Agent Ransack creates two string parsers for a search: a Unicode one for Unicode files and an ASCII one for files whose type can't be determined. Your search term is converted to ASCII using the current code page and if a matching code point can't be found a '#' is used.

In your case your search term is being evaluated to '##' for ASCII searches. Instead it probably should have been evaluted to a non-searchable character, e.g. '\1'. It'll be fixed in the next version of Agent Ransack.

Answer 2

Agent Ransack creates two string parsers for a search: a Unicode one for Unicode files and an ASCII one for files whose type can't be determined. Your search term is converted to ASCII using the current code page and if a matching code point can't be found a '#' is used.

In your case your search term is being evaluated to '##' for ASCII searches. Instead it probably should have been evaluted to a non-searchable character. It'll be fixed in the next version of Agent Ransack.

Answer 3

Agent Ransack creates two string parsers for a search: a Unicode one for Unicode files and an ASCII one for files whose type can't be determined. Your search term is converted to ASCII using the current code page and if a matching code point can't be found a '#' is used.

In your case your search term is being evaluated to '##' for ASCII searches. Instead it probably should have been evaluted to a non-searchable character. It'll be fixed in the next version of Agent Ransack.