Fixed H1 parsing for None types. Should resolve further crashes.

2 files changed, 11 insertions, 5 deletions

diff --git a/parser.py b/parser.py
index 671e2e5..19333e8 100755
--- a/parser.py
+++ b/parser.py
@@ -208,7 +208,6 @@ def buildGuardian():
         h1=h1.split('<a href="', 1)[1]

         h1=h1.split('"', 1)[0]

 

-        print(h1)

         if h1!='https://www.theguardian.com/us':

             break

         else:

diff --git a/unbiasedFunctions.py b/unbiasedFunctions.py
index 950e16d..fca2f2d 100644
--- a/unbiasedFunctions.py
+++ b/unbiasedFunctions.py
@@ -96,9 +96,9 @@ def buildArticle(url, sourceName, encoding=None):#, titleDelStart, titleDelEnd,
                 print("SHOULDN'T GET HERE")

 

         #strip out self-references

-        description=description.replace(sourceName+"'s", 'our')

-        description=description.replace(sourceName+"'", 'our')

-        description=description.replace(sourceName, 'our')

+        description=description.replace(sourceName+"'s", '***')

+        description=description.replace(sourceName+"'", '***')

+        description=description.replace(sourceName, '***')

 

         if debugging:

             print(description)

@@ -123,7 +123,14 @@ def buildOutput(newsSourceArr):
     f.close()

     

     #set the random order for sources

-    h1RandomSources=random.sample(range(len(newsSourceArr)), 4)

+    h1RandomSources=[]

+    while len(h1RandomSources)<4:

+        x=random.sample(range(len(newsSourceArr)), 1)[0]

+        if len(newsSourceArr[x].h1Arr)>0:

+            if x not in h1RandomSources:

+                h1RandomSources.append(x)

+        else:

+            print('\n\n@@@@\nNo H1 stories in '+newsSourceArr[x].name+'\n@@@@\n\n')

     

     #For h2s and h3s, select N random sources (can repeat), then

     #a non-repetitive random article from within