s/max_hops_off_surt/max_hops_off/

brozzler/frontier.py

@@ -338,7 +338,7 @@ class RethinkDbFrontier:
                 hops_off = 0
             elif decision is None:
                 decision = parent_page.hops_off < site.scope.get(
-                        'max_hops_off_surt', 0)
+                        'max_hops_off', 0)
                 hops_off = parent_page.hops_off + 1
             if decision is True:
                 if brozzler.is_permitted_by_robots(site, str(url_for_crawling)):

brozzler/job_schema.yaml

@@ -65,7 +65,7 @@ id:
       max_hops:
         type: integer
 
-      max_hops_off_surt:
+      max_hops_off:
         type: integer
 
   metadata:

brozzler/model.py

@@ -183,6 +183,11 @@ class Site(doublethink.Document, ElapsedMixIn):
             self.last_claimed = brozzler.EPOCH_UTC
         if not "scope" in self:
             self.scope = {}
+        if ("max_hops_off_surt" in self.scope
+                and not "max_hops_off" in self.scope):
+            self.scope["max_hops_off"] = self.scope["max_hops_off_surt"]
+        if "max_hops_off_surt" in self.scope:
+            del self.scope["max_hops_off_surt"]
         if self.seed:
             self._accept_ssurt_if_not_redundant(
                     brozzler.site_surt_canon(self.seed).ssurt())

job-conf.rst

@@ -271,7 +271,7 @@ are not. Example::
         substring: wp-login.php?action=logout
       - domain: malware.us
       max_hops: 20
-      max_hops_off_surt: 0
+      max_hops_off: 0
 
 Toward the end of the process of brozzling a page, brozzler obtains a list of
 navigational links (``<a href="...">`` and similar) on the page, and evaluates
@@ -356,7 +356,7 @@ List of scope rules.
 | number | no       | *none*  |
 +--------+----------+---------+
 
-``max_hops_off_surt``
+``max_hops_off``
 ~~~~~~~~~~~~~~~~~~~~~
 +--------+----------+---------+
 | type   | required | default |

tests/test_frontier.py

@@ -989,6 +989,10 @@ def test_max_hops_off():
     brozzler.new_site(frontier, site)
     site.refresh()  # get it back from the db
 
+    # renamed this param
+    assert not 'max_hops_off_surt' in site.scope
+    assert site.scope['max_hops_off'] == 1
+
     seed_page = frontier.seed_page(site.id)
 
     assert site.accept_reject_or_neither('http://foo.org/', seed_page) is None