duire

commit 25a879c0f2170ac80c0da862a0a5146b4ed1efa4
parent d698e2137fa0ba51ffd8b54dac0252110794b20c
Author: Antoine Amarilli <a3nm@a3nm.net>
Date:   Mon, 23 Mar 2015 01:41:21 +0100

continue

Diffstat:
calcpref.py
 | 
33
+++++++++++++++++++++++++++++++++
prefix.py
 | 
49
++++++++++++++++++++++---------------------------

2 files changed, 55 insertions(+), 27 deletions(-)
diff --git a/calcpref.py b/calcpref.py
@@ -0,0 +1,33 @@
+#!/usr/bin/python3 -O
+# -*- encoding: utf-8
+
+import operator
+import sys
+
+words = set()
+
+for l in sys.stdin.readlines():
+    words.add(l.rstrip())
+
+myprefs = {}
+done = 0
+every = 200
+
+for w in words:
+    if done % every == 0:
+        print("done %d" % done, file=sys.stderr)
+    done += 1
+    for i in range(len(words)):
+        w2 = w[i:]
+        if i == 0 or i == len(words):
+            continue
+        if w2 in words:
+            p = w[:i]
+            if p not in myprefs.keys():
+                myprefs[p] = 0
+            myprefs[p] += 1
+
+sorted_x = sorted(myprefs.items(), key=operator.itemgetter(1))
+for (t, v) in sorted_x:
+    print("%s %s" % (t, v))
+
diff --git a/prefix.py b/prefix.py
@@ -1,45 +1,40 @@
-#!/usr/bin/python3
+#!/usr/bin/python3 -O
 # -*- encoding: utf-8
 
+import operator
 import sys
 
-prefixes = ["dé", "re", "par", "ex", "sous", "sur", "in", "as", "bi", "em",
-"ac", "rec", "di", "su", "en"]
+prefixes = {}
 words = set()
 interesting = {}
 threshold = 2
+maxlen = 2
+keep = 30
+exp = 0.2
 
-for l in sys.stdin.readlines():
+fwords = open(sys.argv[1])
+for l in fwords.readlines():
     words.add(l.rstrip())
+fwords.close()
+fpref = open(sys.argv[2])
+for l in fpref.readlines():
+    t = l.rstrip().split(' ')
+    prefixes[t[0]] = int(t[1])
+fpref.close()
 
 for w in words:
-    for p in prefixes:
+    for p, v in prefixes.items():
         if w.startswith(p):
             w2 = w[len(p):]
+            if w2 in words:
+                continue
             if w2 not in interesting.keys():
                 interesting[w2] = set()
-            interesting[w2].add(w)
+            interesting[w2].add((w, v))
 
-for wi, ws in interesting.items():
-    if wi in words:
-        continue
-    if len(ws) >= threshold:
-        print(wi, ws)
+sortint = sorted(interesting.items(), key=lambda x: -sum((t[1]**exp for t in x[1]), 0))
 
-#myprefs = {}
-#pthresh = 4
-#cons = "bcçdfghjklmnpqrstvwxz"
-# for wi, ws in interesting.items():
-#     if wi in words:
-#         continue
-#     for w2 in words:
-#         if w2.endswith(wi):
-#             mypref = w2[:len(wi)]
-#             if mypref not in myprefs.keys():
-#                 myprefs[mypref] = 0
-#             myprefs[mypref] += 1
-# 
-# for mypref, v in myprefs.items():
-#     if v >= pthresh:
-#         print (mypref, v)
+for (wi, ws) in sortint:
+    ws = sorted(ws, key=operator.itemgetter(1), reverse=True)
+    print(wi+":", ' '.join(t[0] for t in ws))