This patch should be enough to make it work.

0001-Correct-the-usage-of-the-GObject-boilerplate-macros.patch

From 710b85a3ebabc98127c4d7ce7345caec05e26e20 Mon Sep 17 00:00:00 2001
From: Felipe Borges <felipeborges@gnome.org>
Date: Mon, 19 Jun 2017 14:42:42 +0200
Subject: [PATCH] Correct the usage of the GObject boilerplate macros

---
 backend/gcp.h | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/backend/gcp.h b/backend/gcp.h
index 7c00b76..ef903ca 100644
--- a/backend/gcp.h
+++ b/backend/gcp.h
@@ -6,14 +6,13 @@
 
 G_BEGIN_DECLS
 
-typedef struct _GCPObject GCPObject;
-typedef struct _GCPObjectClass GCPObjectClass;
-
 #define GCP_TYPE_OBJECT gcp_object_get_type()
 G_DECLARE_DERIVABLE_TYPE (GCPObject, gcp_object, GCP, OBJECT, GObject)
 
 struct _GCPObjectClass
 {
+  GObjectClass parent;
+
   gchar * (*get_printers) (GCPObject *, const gchar *);
 
   gchar * (*get_printer_options) (GCPObject *,
-- 
2.13.0

shouldn't it be
G_DEFINE_TYPE (GCPObject, gcp_object, G_TYPE_OBJECT)
instead of
G_DEFINE_TYPE (GCPObject, gcp_object, GCP_TYPE_OBJECT)

as the third param is supposed to be the GType of the parent type
reference : https://developer.gnome.org/gobject/stable/gobject-Type-Information.html#G-DEFINE-TYPE:CAPS

so if I keep it as it is... then after running main it is calling gcp_object_new... after that it is calling gcp_object_get_printers()... here it is getting stuck in second line when it calls g_return_val_if_fail ()
the output in my terminal is

./main
inside main()
calling gcp_object_new()
inside gcp_object_new()

(process:3290): GLib-GObject-WARNING **: specified class size for type 'GCPObject' is smaller than the parent type's 'GObject' class size

(process:3290): GLib-CRITICAL **: g_once_init_leave: assertion 'result != 0' failed

(process:3290): GLib-GObject-CRITICAL **: g_object_new: assertion 'G_TYPE_IS_OBJECT (object_type)' failed
outside gcp_object_new()
calling gcp_object_get_printers()
inside gcp_object_get_printers()