From 7a6a911d30ee410ea79b1bca8dabdfb697c1de8a Mon Sep 17 00:00:00 2001
From: gwen <gwenaelremond@free.fr>
Date: Tue, 3 Oct 2023 15:12:16 +0200
Subject: [PATCH] add word extracts in the search results

---
 app/helper.py                  | 24 ++++++++++++++++++++----
 app/templates/plainsearch.html |  4 ++--
 requirements.txt               |  2 ++
 3 files changed, 24 insertions(+), 6 deletions(-)

diff --git a/app/helper.py b/app/helper.py
index 90ac0a8..205c241 100644
--- a/app/helper.py
+++ b/app/helper.py
@@ -1,10 +1,11 @@
-"""helper functions (is it really usefull?)
+"""helper functions
 
-TODO: maybe all these calculations are to be put in the db storage
 """
+from bs4 import BeautifulSoup
 
 from .dbinit import inverted_prince_bigram
 
+
 def make_timeitem_from_filename(filename):
     """
     "anj_isa_i_1441_08_05a" -> "1441_08_05a"
@@ -35,21 +36,36 @@ def plaintext_response(search, actecol, prince_bigram):
                             'filename': '$filename',
                             "place": "$place.name",
                             "diplo_state": "$diplo_state",
-                            "diplo_type": "$diplo_type"
+                            "diplo_type": "$diplo_type",
+                            "xmlcontent": "$xmlcontent"
                             }
                     }
         }]
+    
     results = list(actecol.aggregate(query))
     transformed_query = [pr['_id'] for pr in results]
     invert_prince_bigram = {val: key for key, val in prince_bigram.items()}
 
-    # constructing the dateitem
     for trs in transformed_query:
         trs['house'] = trs['house'].capitalize()
         trs['dateitem'] = make_timeitem_from_filename(trs['filename'])
         bigram, number = trs['prince_code'].split('_')
         long_prince_bigram = inverted_prince_bigram(bigram) + '_' + number
         trs['prince_url'] = prince_name=long_prince_bigram.capitalize()
+        
+        # find the word in the text
+        cleantext = BeautifulSoup(trs["xmlcontent"], "lxml").text
+        cleantext = cleantext.lower()
+        searchtext = search.lower()
+        begin = cleantext.find(searchtext)
+        #end = cleantext.find(searchtext, begin, len(cleantext))
+        strframe = 35
+        begin_mark = begin - strframe
+        if begin_mark < 0:
+            begin_mark = 0
+        end_mark = begin + len(searchtext)  + strframe
+        trs['show_text'] = cleantext[begin_mark:end_mark]
+        
     return transformed_query
 
 
diff --git a/app/templates/plainsearch.html b/app/templates/plainsearch.html
index cd7e709..43195bd 100644
--- a/app/templates/plainsearch.html
+++ b/app/templates/plainsearch.html
@@ -6,7 +6,7 @@
 
 <h1>Résulat de la recherche</h1>
 
-<h2>Résulat de la recherche sur "<b>{{ search }}</b>" : </h2>
+<h2>{{ actes | count }} résulats pour la recherche sur "<b>{{ search }}</b>" :</h2>
 
         {% for acte in actes %}
             <div>
@@ -14,7 +14,7 @@
                     <h3></h3>
                 </div>
                 <div class="pboly">
-                    <p class="chrono-p font-weight-bold"><a href="{{url_for('main.acte', house=acte['house'], prince=acte['prince_url'], dateitem=acte['dateitem'])}}">{{ acte['prince_name'] }}, {{ acte['date'] }}.</a></p>
+                    <p class="chrono-p font-weight-bold"><a href="{{url_for('main.acte', house=acte['house'], prince=acte['prince_url'], dateitem=acte['dateitem'])}}">{{ acte['prince_name'] }}, {{ acte['date'] }}, extrait : "... {{ acte['show_text'] }} ..."</a></p>
                     <p style="margin-bottom: 5%">
                         {% if acte['place'] != 'NS' %}
                             <span class="badge badge-pill badge" style="background-color: #284AA7; color: white; font-size: small;">{{ acte['place'] }}</span>
diff --git a/requirements.txt b/requirements.txt
index 903d0e1..e141833 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -22,3 +22,5 @@ zipp==3.9.0
 pyyaml==6.0.1
 Unidecode==1.3.6
 folium==0.14.0
+beautifulsoup4==4.12.2
+lxml==4.9.3