From ca753097b0734f6b328afa7edf88c5f8b2256fcd Mon Sep 17 00:00:00 2001
From: www <bofh@fet.at>
Date: Sat, 12 Dec 2020 10:15:20 +0000
Subject: [PATCH] start indexing members

---
 solrfet2020/__init__.py | 20 +++++++++++++-------
 solrfet2020/convert.py  |  4 +++-
 2 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/solrfet2020/__init__.py b/solrfet2020/__init__.py
index b914136..72c85e7 100644
--- a/solrfet2020/__init__.py
+++ b/solrfet2020/__init__.py
@@ -3,7 +3,7 @@ import environ
 import pysolr
 from .convert import post_to_solr
 from urllib.parse import urljoin,urlparse
-from fet2020api import fet2020postapi
+from fet2020api import fet2020postapi, fet2020memberapi
 import yaml
 env=environ.Env(
     SOLR_HOST=(str,"http://localhost:8980"),
@@ -12,6 +12,7 @@ env=environ.Env(
 
 
 fet=fet2020postapi(urljoin(env('TARGET'),"api/posts/"))
+fetmember=fet2020memberapi(urljoin(env('TARGET'),"api/members/"))
 def search_post(text=""):
     pass
 def reindex():
@@ -21,6 +22,7 @@ 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'])
@@ -35,23 +37,24 @@ def strip_html(text):
 def result_to_posts(result):
     docs = result.docs
     highlights =result.highlighting
-    posts = [fet.find_one({"slug": rr["id"]}) for rr in docs ]
+    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[p["slug"]]["text_txt"]))
+        str(strip_html(highlights["/posts/"+p["slug"]].get("text_txt","")))
     def create_highlights(p):
-        return  str(strip_html(highlights[p["slug"]]["text_txt"]))
+        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["image"]:post["image"]=urljoin(env('TARGET'),urlparse(post["image"]).path)
+            if post["imageurl"]:post["image"]=urljoin(env('TARGET'),urlparse(post["imageurl"]).path)
     return posts
     
 class SolrFet2020():
@@ -64,9 +67,12 @@ class SolrFet2020():
         self.solr.delete(q='*:*')
         p=post_to_solr(fet.find({"slug":""}))
         self.solr.add(p)
-
+        m=fetmember.find({"nickname":""})
+        print(m)
     def search(self,query):
-        r=self.solr.search("text_txt:*%s*" % query,sort="date_dt desc",**{
+        querystring="text_txt:*%s* title_txt:*%s*^2 tags_tkn:*%s*^2" % (query,query,query)
+
+        r=self.solr.search(querystring,sort="score desc, date_dt desc",**{
                 'hl':'true',
                 'hl.fragsize': 100,
                 'hl.fl': '*', 
diff --git a/solrfet2020/convert.py b/solrfet2020/convert.py
index 9458a64..314268c 100644
--- a/solrfet2020/convert.py
+++ b/solrfet2020/convert.py
@@ -28,7 +28,9 @@ def post_to_solr(p):
             raise PostKeyError("Post needs to have key '%s'" % k)
     # Return the solr structure
     return {
-        "id": p["slug"],
+        "id": p["url"],
         "date_dt": p["public_date"],
+        "tags_tkn": p["tag_string"],
+        "title_txt": p["title"],
         "text_txt": (get_text2(p.get("body","")) or "")+" "+get_text2(p.get("agenda_html","") or "")
     }