change to CSV instead of fortune-style database

[phrases.git] / extract.py

diff --git a/extract.py b/extract.py
index 65f99143e2fca0bd24dc65ea8927633451540413..322a3020c3eccb18c9ad5c29b8ff9845f81771b8 100755 (executable)
--- a/extract.py
+++ b/extract.py
@@ -4,20 +4,53 @@
 
 from bs4 import BeautifulSoup
 import requests
+import csv
 
-url = 'https://en.wikipedia.org/wiki/List_of_Latin_phrases_(full)'
-response = requests.get(url)
-html = response.content
-
-soup = BeautifulSoup(html, "html.parser")
-list_table = soup.find_all("table", attrs={"class":"wikitable"})
-f = open("phrases", "w")
-
-for table in list_table:
-    for row in table.find_all("tr")[1:]:
-        f.write("%" )
-        cell = row.find_all("td")
-        for content in cell:
-            text = content.get_text()
-            f.write("\n" + text)
-f.close()
+def main():
+    url = 'https://en.wikipedia.org/wiki/List_of_Latin_phrases_(full)'
+    response = requests.get(url)
+    html = response.content
+
+    soup = BeautifulSoup(html, "html.parser")
+    list_table = soup.find_all("table", attrs={"class":"wikitable"})
+    with open('phrases.csv', 'w') as f:
+        writer = csv.writer(f)
+
+        i = 0 # For the phrase id
+        # iterate through the tables in the page
+        for table in list_table:
+            for row in table.find_all("tr")[1:]:
+                cell = row.find_all("td")
+                rowc = []
+
+                # append phrase id
+                rowc.append(i)
+
+                # avoid out of bounds errors
+                if len(cell) == 2:
+                    lan = 2
+                else:
+                    lan = 3
+
+                # add cell content
+                for j in range (0, lan):
+                    content = cell[j]
+                    text=(content.get_text()).rstrip()
+                    rowc.append(text)
+
+                # append length of phrase
+                rowc.append(len(rowc[1]))
+                writer.writerow(rowc)
+                i = i + 1
+    f.close()
+
+    # Strip empty lines
+    with open('phrases.csv', 'r+') as f:
+        lines = f.readlines()
+        f.seek(0)
+        f.writelines(line for line in lines if line.strip())
+        f.truncate()
+    f.close()
+
+if __name__ == "__main__":
+    main()