diff options
Diffstat (limited to 'src/object/sp-object.h')
| -rw-r--r-- | src/object/sp-object.h | 30 |
1 files changed, 0 insertions, 30 deletions
diff --git a/src/object/sp-object.h b/src/object/sp-object.h index b5e59cec5..ca3ddaec5 100644 --- a/src/object/sp-object.h +++ b/src/object/sp-object.h @@ -720,36 +720,6 @@ public: void removeAttribute(char const *key, SPException *ex=nullptr); - /** - * Returns an object style property. - * - * \todo - * fixme: Use proper CSS parsing. The current version is buggy - * in a number of situations where key is a substring of the - * style string other than as a property name (including - * where key is a substring of a property name), and is also - * buggy in its handling of inheritance for properties that - * aren't inherited by default. It also doesn't allow for - * the case where the property is specified but with an invalid - * value (in which case I believe the CSS2 error-handling - * behaviour applies, viz. behave as if the property hadn't - * been specified). Also, the current code doesn't use CRSelEng - * stuff to take a value from stylesheets. Also, we aren't - * setting any hooks to force an update for changes in any of - * the inputs (i.e., in any of the elements that this function - * queries). - * - * \par - * Given that the default value for a property depends on what - * property it is (e.g., whether to inherit or not), and given - * the above comment about ignoring invalid values, and that the - * repr parent isn't necessarily the right element to inherit - * from (e.g., maybe we need to inherit from the referencing - * <use> element instead), we should probably make the caller - * responsible for ascending the repr tree as necessary. - */ - char const *getStyleProperty(char const *key, char const *def) const; - void setCSS(SPCSSAttr *css, char const *attr); void changeCSS(SPCSSAttr *css, char const *attr); |