https://bugs.documentfoundation.org/show_bug.cgi?id=96499
Kevin Suo <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|FILEOPEN: HTML format .xls |FILEOPEN: HTML format .xls |file shows NUMERIC cell |file shows NUMERIC cell |value while TEXT type is |value while TEXT type is |expected (because |expected (because |ScHTMLQueryParser::ParseSty |ScHTMLQueryParser::ParseSty |le has failed) |le has failed) (related to | |orcus) CC| |[email protected] --- Comment #12 from Kevin Suo <[email protected]> --- I am adding Kohei Yoshida to cc: could you please take a look? This seems to be related to orcus. There are several issues: 1. If the stylesheet value contains Chinese characters which is not quoted, then the orcus::css_parser.parse() will raise an error. For instance, the html document in attachment 121310 contains the following: .style18 {mso-pattern:auto none; background:#FFCC99; color:#3F3F76; font-size:11.0pt; font-weight:400; font-style:normal; text-decoration:none; font-family:宋体; mso-generic-font-family:auto; mso-font-charset:0; border:.5pt solid #7F7F7F; mso-style-name:"输入";} in which the value for entry "font-family:宋体;" contains non-ascii chars and is not quoted, thus it raises an error in function css_parser<_Handler>::value() in css_parser.hpp: css::parse_error::throw_with("value:: illegal first character of a value '", c, "'"); 2. If I have all the above un-quoted Chinese char values quoted (see attachment attached in the next reply), there seems to be no errors raised during parsing, but the returned CSSHandler is empty in ScHTMLQueryParser::ParseStyle: -- You are receiving this mail because: You are the assignee for the bug.
