From d51791fb6ae69e3b426f7d598920b4349858908f Mon Sep 17 00:00:00 2001 From: Artur Malabarba Date: Mon, 22 Dec 2014 11:37:35 -0200 Subject: Refactor .site to .site_par It turns out some api objects do have a site property, except its value is not a string, it is another object. The actual string we've been referring to as .site is the .site.api_site_parameter To avoid conflicts, I've renamed all our uses of .site to .site_par, and sx-assoc-let now makes sure the object has a .site_par value, instead of a .site value (which it may or may not have now, and is the same object that the api refers to as site). --- sx.el | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'sx.el') diff --git a/sx.el b/sx.el index cda1acd..78995a8 100644 --- a/sx.el +++ b/sx.el @@ -303,9 +303,12 @@ DATA can also be the link itself." DATA can be a question, answer, comment, or user (or any object with a `link' property)." (when data - (unless (assq 'site data) - (setcdr data (cons (cons 'site (sx--site data)) - (cdr data)))) + (let-alist data + (unless .site_par + (setcdr data (cons (cons 'site_par + (or .site.api_site_parameter + (sx--site data))) + (cdr data))))) data)) (defmacro sx-assoc-let (alist &rest body) -- cgit v1.2.3