index member

solrfet2020/__init__.py

@@ -1,74 +1,41 @@
 from lxml.html.clean import clean_html, Cleaner
 import environ
-import pysolr
-from .convert import post_to_solr
-from urllib.parse import urljoin,urlparse
-from fet2020api import fet2020postapi, fet2020memberapi
-import yaml
+
 env=environ.Env(
     SOLR_HOST=(str,"http://localhost:8980"),
     TARGET=(str,"https://alpha.2020.fet.at")
     )
 
+import pysolr
+from .convert import post_to_solr, member_to_solr
+from urllib.parse import urljoin
+from fet2020api import fet2020postapi, fet2020memberapi
+import yaml
+from .solr_to_objects import result_to_object
 
 fet=fet2020postapi(urljoin(env('TARGET'),"api/posts/"))
 fetmember=fet2020memberapi(urljoin(env('TARGET'),"api/members/"))
-def search_post(text=""):
-    pass
-def reindex():
-    pass
-
-def replace_special(t):
-    return t.replace("\n","").replace("\r","").replace("\t","").replace("\\","")
-
-def strip_html(text):
-    if text=="": return ""
-    c=Cleaner(
-        allow_tags=['i','em','p'], 
-        remove_tags=['p','div'])
-    if type(text) is list:
-        h=""
-        for item in text:
-            h=h+" "+strip_html(item)+";"
-        return h
-    return c.clean_html(replace_special(text))[5:-6]
 
 
-def result_to_posts(result):
-    docs = result.docs
-    highlights =result.highlighting
-    posts = [fet.find_one({"slug": rr["id"].split("/")[2]}) for rr in docs ]
-    def create_text(p):
-        return "<b>" + \
-        p["title"]+ "</b>: "+ \
-        "(%s) " % p["public_date"] + \
-        urljoin(env('TARGET'),
-        p["url"]).rstrip("/")+"  "+\
-        str(strip_html(highlights["/posts/"+p["slug"]].get("text_txt","")))
-    def create_highlights(p):
-        print(highlights["/posts/"+p["slug"]])
-        return  str(strip_html(highlights["/posts/"+p["slug"]].get("text_txt","")))
-
-    for post in posts:
-        if post: 
-            post["text"] = create_text(post)
-            post["highlights"]=create_highlights(post)
-            if post["url"]: post["url"]= urljoin(env('TARGET'),post["url"]).rstrip("/")
-            if post["imageurl"]:post["image"]=urljoin(env('TARGET'),urlparse(post["imageurl"]).path)
-    return posts
-    
 class SolrFet2020():
     def __init__(self):
         self.solr=pysolr.Solr(
             urljoin(env('SOLR_HOST'),'/solr/core'), 
             always_commit=True
             )
+
     def reindex(self):
         self.solr.delete(q='*:*')
         p=post_to_solr(fet.find({"slug":""}))
         self.solr.add(p)
+        m=member_to_solr(fetmember.find({"nickname":""})) # search all members
+        self.solr.add(m)
+        #print(m)
+
+    def reindextest(self):
         m=fetmember.find({"nickname":""})
         print(m)
+
     def search(self,query):
         querystring="text_txt:*%s* title_txt:*%s*^2 tags_tkn:*%s*^2" % (query,query,query)
 
@@ -78,6 +45,7 @@ class SolrFet2020():
                 'hl.fl': '*', 
                 'hl.maxAnalyzedChars': -1,
                 'hl.snippets': 100, })
-        links = result_to_posts(r)
+
+        links = result_to_object(r)
         #print(yaml.dump(r))
         return links, r.hits