From ed27e549b73b7befc0d3f532ad20fff9fffc6638 Mon Sep 17 00:00:00 2001
From: Ozbolt Menegatti <ozbolt.menegatti@gmail.com>
Date: Mon, 3 Jun 2019 09:37:48 +0200
Subject: [PATCH] Adding slimming script

---
 slim_sskj.py | 45 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 45 insertions(+)
 create mode 100644 slim_sskj.py

diff --git a/slim_sskj.py b/slim_sskj.py
new file mode 100644
index 0000000..50c75a0
--- /dev/null
+++ b/slim_sskj.py
@@ -0,0 +1,45 @@
+# import xml.etree.ElementTree as ElementTree
+import lxml.etree as ElementTree
+import sys
+import re
+
+filein = sys.argv[1]
+fileout = sys.argv[2]
+
+lemmas = sys.argv[3:]
+
+with open(filein, 'r') as fp:
+    xmlstring = fp.read()[len('<?xml version="1.0" encoding="UTF-8"?>'):]
+    xmlstring = re.sub(' xmlns="[^"]+"', '', xmlstring, count=1)
+    xmlstring = xmlstring.replace(' xml:', ' ')
+    root = ElementTree.XML(xmlstring)
+
+good_s = set()
+
+for word in root.iterfind('.//w'):
+    if word.attrib['lemma'] in lemmas:
+        good_s.add(".".join(word.attrib['id'].split('.')[:-1]))
+
+for div in root.iterfind('.//div'):
+    print("div  ", div.attrib['id'])
+    good_d = False
+
+    for par in div.iterfind('./p'):
+        good_p = False
+        for s in par.iterfind('./s'):
+            if s.attrib['id'] not in good_s:
+                par.remove(s)
+                print("remove", s.attrib)
+            else:
+                print("keep", s.attrib)
+                good_p = True
+        
+        if not good_p:
+            div.remove(par)
+        else:
+            good_d = True
+    
+    if not good_d:
+        root.remove(div)
+
+ElementTree.ElementTree(root).write(fileout, encoding='utf-8')