fix bugs

.github/workflows/bayes.yml

@@ -13,13 +13,13 @@ on:
 #    weekly #m h d/m m d/w 
 #    - cron:  '30 1 * * 1'
 #    every month 3.rd day
-    - cron:  '40 8 3 * *' 
+    - cron:  '41 8 3 * *' 
 
 jobs:
   docker-run-action:
     runs-on: ubuntu-latest
     container:
-      image: tamboraorg/crecoding:2020.0
+      image: credocker/crecoding:2024.0
       volumes:
         - ${{ github.workspace }}:/cre/python
     steps:
@@ -33,7 +33,7 @@ jobs:
       - name: cd /cre/python/
         run: (cd /cre/python/)
       - name: Install HanTa textblob_de
-        run: (pip3 install HanTa textblob_de)
+        run: (pip3 install HanTa textblob==0.15.3 textblob_de==0.4.3 scikit-learn) 
       - name: Run bayes
         run: (python3 bayes.py)
       - name: Submit changes       

.github/workflows/diagrams.yml

@@ -13,13 +13,13 @@ on:
 #    weekly #m h d/m m d/w 
 #    - cron:  '30 1 * * 1'
 #    daily #m h d/m m d/w 
-    - cron:  '30 1 * * *'
+    - cron:  '50 1 * * *'
 
 jobs:
   docker-run-action:
     runs-on: ubuntu-latest
     container:
-      image: tamboraorg/crecoding:2020.0
+      image: credocker/crecoding:2024.0
       volumes:
         - ${{ github.workspace }}:/cre/python
     steps:
@@ -32,6 +32,8 @@ jobs:
 #          ref: main
       - name: cd /cre/python/
         run: (cd /cre/python/)
+      - name: Install nltk
+        run: (pip3 install nltk scikit-learn)
       - name: Run diagrams
         run: (python3 diagrams.py)
       - name: Submit changes       

.github/workflows/entities.yml

@@ -11,15 +11,15 @@ on:
   schedule:
     # * is a special character in YAML so you have to quote this string : 
 #    weekly #m h d/m m d/w 
-    - cron:  '10 1 * * 1'
+    - cron:  '20 5 * * 1'
 #    daily #m h d/m m d/w 
 #    - cron:  '30 1 * * *'
 
 jobs:
   docker-run-action:
     runs-on: ubuntu-latest
     container:
-      image: tamboraorg/crecoding:2020.0
+      image: credocker/crecoding:2024.0
       volumes:
         - ${{ github.workspace }}:/cre/python
     steps:
@@ -33,7 +33,7 @@ jobs:
       - name: cd /cre/python/
         run: (cd /cre/python/)
       - name: Install spacy textblob_de
-        run: (pip3 install spacy textblob_de)
+        run: (pip3 install spacy textblob==0.15.3 textblob_de==0.4.3 nltk)
       - name: Install download de_core_news_md
         run: (python3 -m spacy download de_core_news_md)
       - name: Run entities

.github/workflows/harvest.yml

@@ -10,7 +10,7 @@ on:
   schedule:
     # * is a special character in YAML so you have to quote this string : 
 #    daily,every 5 hours #m h d/m m d/w 
-    - cron:  '00 */5 * * *'
+    - cron:  '30 */5 * * *'
 #    every 2 hour
 #    - cron:  '10 */2 * * *'
 #    every month 2.nd day
@@ -20,7 +20,7 @@ jobs:
   docker-run-action:
     runs-on: ubuntu-latest
     container:
-      image: tamboraorg/crecoding:2020.0
+      image: credocker/crecoding:2020.0
       env:
         NEWSAPI_KEY: ${{ secrets.NEWSAPI_KEY }}
       volumes:

HOWTO.md

@@ -28,7 +28,7 @@ The harvester works best with about 20 search-terms (called keyword here) organi
 
 ![image of organisations](images/gh_organisation.png)
 
-* Create a new organization with unique name - i.e. choose your username and append '_news', like 'kmicha_news'
+* Create a new organization with unique name - i.e. choose your username and append '-news', like 'kmicha-news'
 
 ![image of organizations](images/gh_new_org.png)
 
@@ -84,5 +84,25 @@ Attention: You can adapt the keywords later, but for now you should not remove a
 
 ### 7. Remove existing news_20yy_mm.csv files
 
+Inside your repository, goto code, then inside csv folder, select news_2022_mm.csv file (one after the other) and delete it.
+After each deletion,a commit must be done. 
+
+![goto csv](images/gh_csv_folder.png)
+
+![select csv](images/gh_select_news_csv.png)
+
+![delete csv](images/gh_delete_news_csv.png)
+
+![submit csv](images/gh_submit_delete.png)
+
+
 ### 8.) Add the newsapi API-key to your organization
 
+Switch to your organization (the one ending with "-news") and goto the settings tab.
+Scroll down  untill you find the Security section at the left side panel; there open the Secrets-Range and click on Actions.
+
+![new secret](images/gh_secrets_new.png)
+
+Press the "New organization secret" button and add the newsapi-key inside the Value field. For the name use "NEWSAPI_KEY".
+
+![add secret](images/gh_secrets_add.png)

bayes.py

@@ -8,6 +8,7 @@
 
 
 import nltk
+import sklearn
 from nltk.corpus import stopwords
 from HanTa import HanoverTagger as ht
 from textblob_de import TextBlobDE
@@ -57,6 +58,7 @@ def getNewsDF():
 print(newsDf)   
 
 language = 'ger'
+nltk.download('punkt_tab')
 nltk.download('punkt')
 nltk.download('stopwords')
 tagger = ht.HanoverTagger('morphmodel_'+language+'.pgz')